├── beats ├── __init__.py ├── README.md ├── Tokenizers.py ├── BEATs.py ├── modules.py └── quantizer.py ├── .gitignore ├── dcase2023_task2_evaluator ├── teams │ └── .gitignore ├── teams_result │ └── .gitignore ├── teams_additional_result │ └── .gitignore ├── .gitignore ├── LISENCE.pdf ├── requirements.txt ├── dockerfile ├── 03_evaluation_eval_data.sh ├── tools │ ├── plot_anm_score.py │ └── test_plots.py ├── ground_truth_data │ ├── ground_truth_ToyDrone_section_00_test.csv │ ├── ground_truth_ToyNscale_section_00_test.csv │ ├── ground_truth_ToyTank_section_00_test.csv │ ├── ground_truth_Vacuum_section_00_test.csv │ ├── ground_truth_bandsaw_section_00_test.csv │ ├── ground_truth_grinder_section_00_test.csv │ └── ground_truth_shaker_section_00_test.csv ├── ground_truth_domain │ ├── ground_truth_ToyTank_section_00_test.csv │ ├── ground_truth_Vacuum_section_00_test.csv │ ├── ground_truth_bandsaw_section_00_test.csv │ ├── ground_truth_grinder_section_00_test.csv │ ├── ground_truth_shaker_section_00_test.csv │ ├── ground_truth_ToyDrone_section_00_test.csv │ └── ground_truth_ToyNscale_section_00_test.csv ├── README.md └── ground_truth_attributes │ ├── ground_truth_bandsaw_section_00_test.csv │ ├── ground_truth_shaker_section_00_test.csv │ ├── ground_truth_ToyDrone_section_00_test.csv │ └── ground_truth_ToyTank_section_00_test.csv ├── requirements.txt ├── LICENSE ├── src └── datasets │ ├── utils.py │ ├── prepare_dcase2020.py │ ├── prepare_dcase2023.py │ └── audio_dataset.py ├── README.md └── data_config.yaml /beats/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | __pycache__ 2 | out/ 3 | train_features* -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/teams/.gitignore: -------------------------------------------------------------------------------- 1 | * 2 | !.gitignore -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/teams_result/.gitignore: -------------------------------------------------------------------------------- 1 | * 2 | !.gitignore -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/teams_additional_result/.gitignore: -------------------------------------------------------------------------------- 1 | * 2 | !.gitignore -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/.gitignore: -------------------------------------------------------------------------------- 1 | __pycache__/ 2 | build 3 | .devcontainer 4 | .vscode -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/LISENCE.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/Phuriches/GenRepASD/HEAD/dcase2023_task2_evaluator/LISENCE.pdf -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | librosa==0.10.2.post1 2 | numpy==1.24.3 3 | pandas==2.0.2 4 | PyYAML==6.0.1 5 | scikit_learn==1.5.1 6 | scipy==1.14.1 7 | tqdm==4.66.4 8 | torch==2.1.2 9 | torchaudio==2.1.2 10 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/requirements.txt: -------------------------------------------------------------------------------- 1 | scipy==1.10.1 2 | librosa==0.9.2 3 | matplotlib==3.7.0 4 | tqdm==4.63 5 | seaborn==0.12.2 6 | 7 | # Included in docker image 8 | torch==1.13.1 9 | torchvision==0.14.1 10 | pyYAML==6.0 -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/dockerfile: -------------------------------------------------------------------------------- 1 | FROM pytorch/pytorch:1.13.1-cuda11.6-cudnn8-runtime 2 | 3 | ARG DEBIAN_FRONTEND=noninteractive 4 | 5 | RUN apt update && apt-get --yes install libsndfile1 6 | 7 | ADD requirements.txt ./ 8 | RUN pip install -U pip 9 | RUN pip install -r requirements.txt 10 | 11 | ARG USERNAME 12 | ARG GROUPNAME 13 | ARG UID 14 | ARG GID 15 | RUN groupadd -g $GID $GROUPNAME && \ 16 | useradd -m -s /bin/bash -u $UID -g $GID $USERNAME 17 | USER $USERNAME 18 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/03_evaluation_eval_data.sh: -------------------------------------------------------------------------------- 1 | # basic parameters 2 | ## directory containing team results. 3 | cd dcase2023_task2_evaluator 4 | teams_root_dir=./teams 5 | result_dir=${teams_root_dir}_result 6 | ## what depth to search '--teams_root_dir' using glob. 7 | dir_depth=2 8 | 9 | # Parameters for exporting additional results. 10 | out_all=False 11 | additional_result_dir=${teams_root_dir}_additional_result 12 | 13 | # if using filename is DCASE2023 baseline style, change parameters as necessary. 14 | # example filename: 'anomaly_score_DCASE2023T2_
_test_seed_Eval.csv' 15 | seed=13711 16 | tag="_id(0_)" 17 | 18 | echo "python dcase2023_task2_evaluator.py" 19 | python dcase2023_task2_evaluator.py \ 20 | --dir_depth=$dir_depth \ 21 | --teams_root_dir=$teams_root_dir \ 22 | --result_dir=$result_dir \ 23 | --additional_result_dir=$additional_result_dir \ 24 | --out_all=$out_all \ 25 | --seed=$seed \ 26 | -tag=$tag \ 27 | 28 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/tools/plot_anm_score.py: -------------------------------------------------------------------------------- 1 | from tools.test_plots import Figdata, show_figs 2 | 3 | class AnmScoreFigData(): 4 | def __init__(self): 5 | self.figdatas = [] 6 | 7 | def anm_score_to_figdata(self, scores, title=""): 8 | nml_scores = [x[1] for x in scores if x[0]==0] 9 | anm_scores = [x[1] for x in scores if x[0]==1] 10 | 11 | figdata = Figdata( 12 | data=nml_scores, 13 | data2=anm_scores, 14 | type="boxplot", 15 | labels=["nml","anm"], 16 | ylabel="score", 17 | title=title 18 | ) 19 | 20 | return figdata 21 | 22 | 23 | def append_figdata(self, figdata): 24 | self.figdatas.append(figdata) 25 | 26 | 27 | def show_fig(self, title="anm_score", export_dir="results", is_display_console=False): 28 | show_figs( 29 | *self.figdatas, 30 | fold_interval=len(self.figdatas), 31 | sup_title=title, 32 | export_path=f"{export_dir}/{title}.png", 33 | is_display_console=is_display_console 34 | ) 35 | -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2024 phurich 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /src/datasets/utils.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | 3 | 4 | def trainDataPct(dataset, pct=1, num_samples=None, machine_names:list=None, seed=42): 5 | # sample from each machine name 6 | np.random.seed(seed) 7 | 8 | if machine_names is not None and type(machine_names) == str: 9 | machine_names = [machine_names] 10 | print(machine_names) 11 | 12 | if machine_names is None or machine_names[0] == "all": 13 | machine_names = np.unique(dataset['machine_names']).tolist() 14 | 15 | file_list = [] 16 | label_list = [] 17 | attr_list = [] 18 | machine_name_list = [] 19 | source_list = None 20 | if "source_list" in dataset.keys(): 21 | source_list = [] 22 | machine_list = dataset['machine_names'] 23 | for mn in machine_names: 24 | machine_audio_files = np.array(dataset["file_list"])[machine_list == mn] 25 | machine_label_files = np.array(dataset["label_list"])[machine_list == mn] 26 | machine_file_attrs = np.array(dataset["file_attrs"])[machine_list == mn] 27 | if source_list is not None: 28 | machine_source_files = np.array(dataset["source_list"])[machine_list == mn] 29 | 30 | num_all = len(machine_audio_files) 31 | num_train = int(pct * num_all) if num_samples is None else num_samples 32 | id_list = np.random.choice(num_all, num_train, replace=False) 33 | 34 | sample_file_list = [machine_audio_files[i] for i in id_list] 35 | sample_label_list = [machine_label_files[i] for i in id_list] 36 | sample_attr_list = [machine_file_attrs[i] for i in id_list] 37 | if source_list is not None: 38 | sample_source_list = [machine_source_files[i] for i in id_list] 39 | # print(f'machine: {mn}, num_all: {num_all}, num_samples: {num_samples}, len(file_list): {len(sample_file_list)}') 40 | 41 | file_list.extend(sample_file_list) 42 | label_list.extend(sample_label_list) 43 | attr_list.extend(sample_attr_list) 44 | if source_list is not None: 45 | source_list.extend(sample_source_list) 46 | machine_name_list.extend([mn] * num_train) 47 | print(f"machine_names: {machine_names}, num_samples: {len(file_list)}") 48 | 49 | 50 | # print(f"total samples: {len(file_list)}") 51 | return { 52 | 'file_list': np.array(file_list), 53 | 'label_list': np.array(label_list), 54 | 'file_attrs': np.array(attr_list), 55 | 'source_list': np.array(source_list) if source_list is not None else None, 56 | 'machine_names': np.array(machine_name_list) 57 | } -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # GenRepASD 2 | Pytorch implementation of Deep Generic Representations for Domain-Generalized Anomalous Sound Detection, accepted at ICASSP 2025: https://arxiv.org/abs/2409.05035 3 | 4 | ## Setting up 5 | 1. Install the requirements `pip install -r requirements.txt` 6 | 7 | 2. Download the [DCASE2020T2](https://dcase.community/challenge2020/task-unsupervised-detection-of-anomalous-sounds#download) and [DCASE2023T2](https://dcase.community/challenge2023/task-first-shot-unsupervised-anomalous-sound-detection-for-machine-condition-monitoring#download) datasets and place them according to the given directory structure specified in `data_config.yaml`. 8 | 9 | 3. Download pre-trained weights of BEATs from https://github.com/microsoft/unilm/tree/master/beats and place them in a pre-trained directory. 10 | 11 | ## Domain-shift experiment on DCASE2023T2 Eval set. 12 | Run GenRep using MemMixup with Ks=990 (best performance). 13 | ``` 14 | python run_genrep_dcase2023.py \ 15 | --dataset_name dcase2023 \ 16 | --model_name beats_ft1 \ 17 | --pretrained_model_dir \ 18 | --temporal_pooling \ 19 | --n_mix_support 990 \ 20 | --alpha 0.9 \ 21 | --save_official 22 | ``` 23 | 24 | After run the above script, you can also evaluate the performance using DCASE2023T2 official evaluator (you should get the same result as above): 25 | ``` 26 | bash dcase2023_task2_evaluator/03_evaluation_eval_data.sh 27 | ``` 28 | 29 | ## Low-shot experiment on DCASE2020T2 Dev set. 30 | with 200-shot 31 | ``` 32 | python run_genrep_dcase2020.py \ 33 | --dataset_name dcase2020 \ 34 | --model_name beats_ft1 \ 35 | --pretrained_model_dir \ 36 | --temporal_pooling \ 37 | --num_samples 200 38 | ``` 39 | 40 | ## Acknowledgement 41 | - We thanks the authors of [BEATs](https://arxiv.org/abs/2212.09058) for providing the pre-trained weights. 42 | - We thanks [SPADE](https://github.com/byungjae89/SPADE-pytorch), [ssl4asd](https://github.com/wilkinghoff/ssl4asd), [STgram-MFN](https://github.com/liuyoude/STgram-MFN/tree/main), [DCASE2023](https://github.com/nttcslab/dcase2023_task2_evaluator) for providing the codebase and evaluation scripts. 43 | 44 | ## Citation 45 | If you find this work useful, please consider citing: 46 | ``` 47 | @inproceedings{saengthong2025deep, 48 | title={Deep Generic Representations for Domain-Generalized Anomalous Sound Detection}, 49 | author={Phurich Saengthong and Takahiro Shinozaki}, 50 | booktitle={Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)}, 51 | year={2025}, 52 | publisher={IEEE} 53 | } 54 | ``` -------------------------------------------------------------------------------- /data_config.yaml: -------------------------------------------------------------------------------- 1 | # dcase 2020 datasets 2 | dcase2020_train_dirs: 3 | - ../database/dcase2020t2/train_data/fan/train 4 | - ../database/dcase2020t2/train_data/pump/train 5 | - ../database/dcase2020t2/train_data/slider/train 6 | - ../database/dcase2020t2/train_data/ToyCar/train 7 | - ../database/dcase2020t2/train_data/ToyConveyor/train 8 | - ../database/dcase2020t2/train_data/valve/train 9 | dcase2020_add_dirs: 10 | - ../database/dcase2020t2/add_data/fan/train 11 | - ../database/dcase2020t2/add_data/pump/train 12 | - ../database/dcase2020t2/add_data/slider/train 13 | - ../database/dcase2020t2/add_data/ToyCar/train 14 | - ../database/dcase2020t2/add_data/ToyConveyor/train 15 | - ../database/dcase2020t2/add_data/valve/train 16 | dcase2020_valid_dirs: 17 | - ../database/dcase2020t2/train_data/fan/test 18 | - ../database/dcase2020t2/train_data/pump/test 19 | - ../database/dcase2020t2/train_data/slider/test 20 | - ../database/dcase2020t2/train_data/ToyCar/test 21 | - ../database/dcase2020t2/train_data/ToyConveyor/test 22 | - ../database/dcase2020t2/train_data/valve/test 23 | dcase2020_test_dirs: 24 | - ../database/dcase2020t2/eval_data/fan/test 25 | - ../database/dcase2020t2/eval_data/pump/test 26 | - ../database/dcase2020t2/eval_data/slider/test 27 | - ../database/dcase2020t2/eval_data/ToyCar/test 28 | - ../database/dcase2020t2/eval_data/ToyConveyor/test 29 | - ../database/dcase2020t2/eval_data/valve/test 30 | 31 | # dcase2023 datasets 32 | dcase2023_train_dirs: 33 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/bearing/train 34 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/fan/train 35 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/gearbox/train 36 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/slider/train 37 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/ToyCar/train 38 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/ToyTrain/train 39 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/valve/train 40 | dcase2023_add_dirs: 41 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/bandsaw/train 42 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/grinder/train 43 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/shaker/train 44 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/ToyDrone/train 45 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/ToyNscale/train 46 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/ToyTank/train 47 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/Vacuum/train 48 | dcase2023_valid_dirs: 49 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/bearing/test 50 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/fan/test 51 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/gearbox/test 52 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/slider/test 53 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/ToyCar/test 54 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/ToyTrain/test 55 | - ../database/dcase2023t2/data/dcase2023t2/dev_data/raw/valve/test 56 | dcase2023_test_dirs: 57 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/bandsaw/test 58 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/grinder/test 59 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/shaker/test 60 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/ToyDrone/test 61 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/ToyNscale/test 62 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/ToyTank/test 63 | - ../database/dcase2023t2/data/dcase2023t2/eval_data/raw/Vacuum/test -------------------------------------------------------------------------------- /src/datasets/prepare_dcase2020.py: -------------------------------------------------------------------------------- 1 | import os 2 | import re 3 | import glob 4 | import numpy as np 5 | import pandas as pd 6 | 7 | 8 | def get_filename_list(dir_path, pattern='*', ext='*'): 9 | """ 10 | find all extention files under directory 11 | :param dir_path: directory path 12 | :param ext: extention name, like wav, png... 13 | :param pattern: filename pattern for searching 14 | :return: files path list 15 | """ 16 | filename_list = [] 17 | for root, _, _ in os.walk(dir_path): 18 | file_path_pattern = os.path.join(root, f'{pattern}.{ext}') 19 | files = sorted(glob.glob(file_path_pattern)) 20 | filename_list += files 21 | return filename_list 22 | 23 | def get_meta_list(file_list): 24 | 25 | label_list = [] 26 | for file in file_list: 27 | label_list.append(0 if 'normal' in file else 1) 28 | return label_list 29 | 30 | def get_data_list(file_dirs): 31 | file_list = [] 32 | for fd in file_dirs: 33 | files = get_filename_list(fd, ext="wav") 34 | file_list.extend(files) 35 | # print(f'len of files: {fd.split("/")[-2]} | {len(files)}') 36 | label_list = get_meta_list(file_list) 37 | return file_list, label_list 38 | 39 | def get_attributes(file_list): 40 | attrs = [] 41 | machines = [] 42 | for idx, ext_id in enumerate(file_list): 43 | machine = ext_id.split('/')[-3] 44 | file = ext_id.split('/')[-1] 45 | section_id = re.findall('id_[0-9][0-9]', file)[0].split('_')[-1] 46 | machine_id = machine + '_' + section_id 47 | attrs.append(machine_id) 48 | machines.append(machine) 49 | return np.array(attrs), np.array(machines) 50 | 51 | def prepare_data(split_dirs): 52 | file_list, label_list = get_data_list(split_dirs) 53 | file_attrs, machine_names = get_attributes(file_list) 54 | return file_list, label_list, file_attrs, machine_names 55 | 56 | def prepare_test_data(): 57 | df_test = pd.read_csv("./src/datasets/preprocessed_eval_data_list.csv") 58 | root_path = "../database/dcase2020t2/eval_data/{machine}/test" 59 | file_list = df_test['file_name'].tolist() 60 | machine_list = df_test['machine'].tolist() 61 | label_list = df_test['label'].tolist() 62 | def get_attributes(): 63 | files = [] 64 | attrs = [] 65 | for file_name, machine_name in zip(file_list, machine_list): 66 | file_name = os.path.join(root_path.format(machine=machine_name), file_name) 67 | section_id = re.findall('id_[0-9][0-9]', file_name)[0].split('_')[-1] 68 | machine_id = machine_name + '_' + section_id 69 | files.append(file_name) 70 | attrs.append(machine_id) 71 | return files, np.array(attrs) 72 | file_list, file_attrs = get_attributes() 73 | machine_names = np.array(machine_list) 74 | return file_list, label_list, file_attrs, machine_names 75 | 76 | def get_dcase2020(train_split="train+add"): 77 | """ Args: 78 | dcase: can be dcase2020, 2021, 2022, and 2023 79 | train_split: train or train+add 80 | """ 81 | import yaml 82 | with open("./data_config.yaml", "r") as ymlfile: 83 | data_config = yaml.load(ymlfile, Loader=yaml.FullLoader) 84 | 85 | if train_split == "train+add": 86 | train_dirs = data_config["dcase2020_train_dirs"] + data_config["dcase2020_add_dirs"] 87 | train_file_list, train_label_list, train_file_attrs, train_machine_names = prepare_data(train_dirs) 88 | elif train_split == "train": 89 | train_dirs = data_config["dcase2020_train_dirs"] 90 | train_file_list, train_label_list, train_file_attrs, train_machine_names = prepare_data(train_dirs) 91 | elif train_split == "add": 92 | train_dirs = data_config["dcase2020_add_dirs"] 93 | train_file_list, train_label_list, train_file_attrs, train_machine_names = prepare_data(train_dirs) 94 | valid_dirs = data_config["dcase2020_valid_dirs"] 95 | # test_dirs = data_config["dcase2020_test_dirs"] 96 | valid_file_list, valid_label_list, valid_file_attrs, valid_machine_names = prepare_data(valid_dirs) 97 | 98 | test_file_list, test_label_list, test_file_attrs, test_machine_names = prepare_test_data() 99 | 100 | datasets = { 101 | "train": { 102 | "file_list": train_file_list, 103 | "label_list": train_label_list, 104 | "file_attrs": train_file_attrs, 105 | "machine_names": train_machine_names 106 | }, 107 | "valid": { 108 | "file_list": valid_file_list, 109 | "label_list": valid_label_list, 110 | "file_attrs": valid_file_attrs, 111 | "machine_names": valid_machine_names 112 | }, 113 | "test": { 114 | "file_list": test_file_list, 115 | "label_list": test_label_list, 116 | "file_attrs": test_file_attrs, 117 | "machine_names": test_machine_names 118 | } 119 | } 120 | return datasets -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_data/ground_truth_ToyDrone_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,1 2 | section_00_0001.wav,0 3 | section_00_0002.wav,0 4 | section_00_0003.wav,1 5 | section_00_0004.wav,1 6 | section_00_0005.wav,0 7 | section_00_0006.wav,0 8 | section_00_0007.wav,1 9 | section_00_0008.wav,0 10 | section_00_0009.wav,1 11 | section_00_0010.wav,1 12 | section_00_0011.wav,0 13 | section_00_0012.wav,0 14 | section_00_0013.wav,0 15 | section_00_0014.wav,0 16 | section_00_0015.wav,1 17 | section_00_0016.wav,1 18 | section_00_0017.wav,0 19 | section_00_0018.wav,1 20 | section_00_0019.wav,1 21 | section_00_0020.wav,0 22 | section_00_0021.wav,0 23 | section_00_0022.wav,0 24 | section_00_0023.wav,1 25 | section_00_0024.wav,1 26 | section_00_0025.wav,1 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,1 30 | section_00_0029.wav,1 31 | section_00_0030.wav,1 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,0 35 | section_00_0034.wav,1 36 | section_00_0035.wav,0 37 | section_00_0036.wav,0 38 | section_00_0037.wav,0 39 | section_00_0038.wav,1 40 | section_00_0039.wav,1 41 | section_00_0040.wav,1 42 | section_00_0041.wav,0 43 | section_00_0042.wav,0 44 | section_00_0043.wav,1 45 | section_00_0044.wav,1 46 | section_00_0045.wav,1 47 | section_00_0046.wav,0 48 | section_00_0047.wav,1 49 | section_00_0048.wav,0 50 | section_00_0049.wav,1 51 | section_00_0050.wav,0 52 | section_00_0051.wav,0 53 | section_00_0052.wav,0 54 | section_00_0053.wav,1 55 | section_00_0054.wav,0 56 | section_00_0055.wav,1 57 | section_00_0056.wav,1 58 | section_00_0057.wav,0 59 | section_00_0058.wav,0 60 | section_00_0059.wav,0 61 | section_00_0060.wav,1 62 | section_00_0061.wav,0 63 | section_00_0062.wav,1 64 | section_00_0063.wav,0 65 | section_00_0064.wav,1 66 | section_00_0065.wav,1 67 | section_00_0066.wav,1 68 | section_00_0067.wav,1 69 | section_00_0068.wav,1 70 | section_00_0069.wav,1 71 | section_00_0070.wav,1 72 | section_00_0071.wav,0 73 | section_00_0072.wav,0 74 | section_00_0073.wav,0 75 | section_00_0074.wav,0 76 | section_00_0075.wav,1 77 | section_00_0076.wav,1 78 | section_00_0077.wav,0 79 | section_00_0078.wav,1 80 | section_00_0079.wav,1 81 | section_00_0080.wav,1 82 | section_00_0081.wav,1 83 | section_00_0082.wav,1 84 | section_00_0083.wav,1 85 | section_00_0084.wav,1 86 | section_00_0085.wav,0 87 | section_00_0086.wav,1 88 | section_00_0087.wav,0 89 | section_00_0088.wav,0 90 | section_00_0089.wav,0 91 | section_00_0090.wav,0 92 | section_00_0091.wav,1 93 | section_00_0092.wav,0 94 | section_00_0093.wav,1 95 | section_00_0094.wav,1 96 | section_00_0095.wav,1 97 | section_00_0096.wav,0 98 | section_00_0097.wav,0 99 | section_00_0098.wav,0 100 | section_00_0099.wav,0 101 | section_00_0100.wav,0 102 | section_00_0101.wav,0 103 | section_00_0102.wav,0 104 | section_00_0103.wav,0 105 | section_00_0104.wav,1 106 | section_00_0105.wav,1 107 | section_00_0106.wav,0 108 | section_00_0107.wav,0 109 | section_00_0108.wav,1 110 | section_00_0109.wav,1 111 | section_00_0110.wav,1 112 | section_00_0111.wav,0 113 | section_00_0112.wav,1 114 | section_00_0113.wav,0 115 | section_00_0114.wav,0 116 | section_00_0115.wav,1 117 | section_00_0116.wav,0 118 | section_00_0117.wav,0 119 | section_00_0118.wav,0 120 | section_00_0119.wav,0 121 | section_00_0120.wav,0 122 | section_00_0121.wav,0 123 | section_00_0122.wav,1 124 | section_00_0123.wav,1 125 | section_00_0124.wav,1 126 | section_00_0125.wav,1 127 | section_00_0126.wav,0 128 | section_00_0127.wav,0 129 | section_00_0128.wav,1 130 | section_00_0129.wav,0 131 | section_00_0130.wav,0 132 | section_00_0131.wav,0 133 | section_00_0132.wav,1 134 | section_00_0133.wav,1 135 | section_00_0134.wav,0 136 | section_00_0135.wav,1 137 | section_00_0136.wav,0 138 | section_00_0137.wav,1 139 | section_00_0138.wav,1 140 | section_00_0139.wav,0 141 | section_00_0140.wav,0 142 | section_00_0141.wav,0 143 | section_00_0142.wav,1 144 | section_00_0143.wav,1 145 | section_00_0144.wav,1 146 | section_00_0145.wav,1 147 | section_00_0146.wav,1 148 | section_00_0147.wav,1 149 | section_00_0148.wav,1 150 | section_00_0149.wav,0 151 | section_00_0150.wav,1 152 | section_00_0151.wav,0 153 | section_00_0152.wav,1 154 | section_00_0153.wav,1 155 | section_00_0154.wav,1 156 | section_00_0155.wav,1 157 | section_00_0156.wav,1 158 | section_00_0157.wav,0 159 | section_00_0158.wav,1 160 | section_00_0159.wav,0 161 | section_00_0160.wav,0 162 | section_00_0161.wav,0 163 | section_00_0162.wav,0 164 | section_00_0163.wav,1 165 | section_00_0164.wav,1 166 | section_00_0165.wav,1 167 | section_00_0166.wav,1 168 | section_00_0167.wav,0 169 | section_00_0168.wav,1 170 | section_00_0169.wav,0 171 | section_00_0170.wav,1 172 | section_00_0171.wav,0 173 | section_00_0172.wav,0 174 | section_00_0173.wav,0 175 | section_00_0174.wav,1 176 | section_00_0175.wav,0 177 | section_00_0176.wav,0 178 | section_00_0177.wav,1 179 | section_00_0178.wav,1 180 | section_00_0179.wav,0 181 | section_00_0180.wav,1 182 | section_00_0181.wav,0 183 | section_00_0182.wav,1 184 | section_00_0183.wav,0 185 | section_00_0184.wav,1 186 | section_00_0185.wav,1 187 | section_00_0186.wav,1 188 | section_00_0187.wav,0 189 | section_00_0188.wav,0 190 | section_00_0189.wav,0 191 | section_00_0190.wav,0 192 | section_00_0191.wav,0 193 | section_00_0192.wav,0 194 | section_00_0193.wav,0 195 | section_00_0194.wav,0 196 | section_00_0195.wav,1 197 | section_00_0196.wav,0 198 | section_00_0197.wav,1 199 | section_00_0198.wav,1 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_data/ground_truth_ToyNscale_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,1 2 | section_00_0001.wav,0 3 | section_00_0002.wav,0 4 | section_00_0003.wav,0 5 | section_00_0004.wav,1 6 | section_00_0005.wav,0 7 | section_00_0006.wav,0 8 | section_00_0007.wav,1 9 | section_00_0008.wav,1 10 | section_00_0009.wav,1 11 | section_00_0010.wav,0 12 | section_00_0011.wav,1 13 | section_00_0012.wav,1 14 | section_00_0013.wav,1 15 | section_00_0014.wav,1 16 | section_00_0015.wav,1 17 | section_00_0016.wav,1 18 | section_00_0017.wav,0 19 | section_00_0018.wav,0 20 | section_00_0019.wav,1 21 | section_00_0020.wav,1 22 | section_00_0021.wav,1 23 | section_00_0022.wav,0 24 | section_00_0023.wav,0 25 | section_00_0024.wav,0 26 | section_00_0025.wav,0 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,1 30 | section_00_0029.wav,1 31 | section_00_0030.wav,1 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,1 35 | section_00_0034.wav,1 36 | section_00_0035.wav,1 37 | section_00_0036.wav,1 38 | section_00_0037.wav,0 39 | section_00_0038.wav,0 40 | section_00_0039.wav,0 41 | section_00_0040.wav,1 42 | section_00_0041.wav,0 43 | section_00_0042.wav,1 44 | section_00_0043.wav,0 45 | section_00_0044.wav,1 46 | section_00_0045.wav,0 47 | section_00_0046.wav,0 48 | section_00_0047.wav,1 49 | section_00_0048.wav,0 50 | section_00_0049.wav,0 51 | section_00_0050.wav,1 52 | section_00_0051.wav,0 53 | section_00_0052.wav,0 54 | section_00_0053.wav,1 55 | section_00_0054.wav,1 56 | section_00_0055.wav,1 57 | section_00_0056.wav,1 58 | section_00_0057.wav,0 59 | section_00_0058.wav,1 60 | section_00_0059.wav,0 61 | section_00_0060.wav,1 62 | section_00_0061.wav,1 63 | section_00_0062.wav,0 64 | section_00_0063.wav,0 65 | section_00_0064.wav,1 66 | section_00_0065.wav,1 67 | section_00_0066.wav,1 68 | section_00_0067.wav,0 69 | section_00_0068.wav,0 70 | section_00_0069.wav,0 71 | section_00_0070.wav,1 72 | section_00_0071.wav,0 73 | section_00_0072.wav,0 74 | section_00_0073.wav,0 75 | section_00_0074.wav,1 76 | section_00_0075.wav,1 77 | section_00_0076.wav,1 78 | section_00_0077.wav,0 79 | section_00_0078.wav,0 80 | section_00_0079.wav,1 81 | section_00_0080.wav,1 82 | section_00_0081.wav,0 83 | section_00_0082.wav,1 84 | section_00_0083.wav,1 85 | section_00_0084.wav,0 86 | section_00_0085.wav,1 87 | section_00_0086.wav,1 88 | section_00_0087.wav,1 89 | section_00_0088.wav,1 90 | section_00_0089.wav,0 91 | section_00_0090.wav,1 92 | section_00_0091.wav,1 93 | section_00_0092.wav,0 94 | section_00_0093.wav,0 95 | section_00_0094.wav,1 96 | section_00_0095.wav,0 97 | section_00_0096.wav,0 98 | section_00_0097.wav,0 99 | section_00_0098.wav,0 100 | section_00_0099.wav,1 101 | section_00_0100.wav,1 102 | section_00_0101.wav,1 103 | section_00_0102.wav,1 104 | section_00_0103.wav,0 105 | section_00_0104.wav,1 106 | section_00_0105.wav,0 107 | section_00_0106.wav,0 108 | section_00_0107.wav,1 109 | section_00_0108.wav,0 110 | section_00_0109.wav,1 111 | section_00_0110.wav,0 112 | section_00_0111.wav,0 113 | section_00_0112.wav,0 114 | section_00_0113.wav,0 115 | section_00_0114.wav,1 116 | section_00_0115.wav,0 117 | section_00_0116.wav,1 118 | section_00_0117.wav,1 119 | section_00_0118.wav,0 120 | section_00_0119.wav,1 121 | section_00_0120.wav,1 122 | section_00_0121.wav,1 123 | section_00_0122.wav,0 124 | section_00_0123.wav,0 125 | section_00_0124.wav,1 126 | section_00_0125.wav,0 127 | section_00_0126.wav,1 128 | section_00_0127.wav,0 129 | section_00_0128.wav,0 130 | section_00_0129.wav,0 131 | section_00_0130.wav,1 132 | section_00_0131.wav,0 133 | section_00_0132.wav,1 134 | section_00_0133.wav,0 135 | section_00_0134.wav,0 136 | section_00_0135.wav,1 137 | section_00_0136.wav,0 138 | section_00_0137.wav,0 139 | section_00_0138.wav,0 140 | section_00_0139.wav,1 141 | section_00_0140.wav,1 142 | section_00_0141.wav,1 143 | section_00_0142.wav,1 144 | section_00_0143.wav,0 145 | section_00_0144.wav,0 146 | section_00_0145.wav,0 147 | section_00_0146.wav,0 148 | section_00_0147.wav,0 149 | section_00_0148.wav,1 150 | section_00_0149.wav,1 151 | section_00_0150.wav,0 152 | section_00_0151.wav,1 153 | section_00_0152.wav,0 154 | section_00_0153.wav,1 155 | section_00_0154.wav,1 156 | section_00_0155.wav,1 157 | section_00_0156.wav,1 158 | section_00_0157.wav,1 159 | section_00_0158.wav,0 160 | section_00_0159.wav,0 161 | section_00_0160.wav,0 162 | section_00_0161.wav,1 163 | section_00_0162.wav,1 164 | section_00_0163.wav,0 165 | section_00_0164.wav,1 166 | section_00_0165.wav,1 167 | section_00_0166.wav,1 168 | section_00_0167.wav,0 169 | section_00_0168.wav,1 170 | section_00_0169.wav,0 171 | section_00_0170.wav,1 172 | section_00_0171.wav,0 173 | section_00_0172.wav,0 174 | section_00_0173.wav,0 175 | section_00_0174.wav,1 176 | section_00_0175.wav,0 177 | section_00_0176.wav,1 178 | section_00_0177.wav,1 179 | section_00_0178.wav,1 180 | section_00_0179.wav,0 181 | section_00_0180.wav,1 182 | section_00_0181.wav,0 183 | section_00_0182.wav,0 184 | section_00_0183.wav,0 185 | section_00_0184.wav,1 186 | section_00_0185.wav,0 187 | section_00_0186.wav,0 188 | section_00_0187.wav,1 189 | section_00_0188.wav,1 190 | section_00_0189.wav,1 191 | section_00_0190.wav,0 192 | section_00_0191.wav,1 193 | section_00_0192.wav,1 194 | section_00_0193.wav,0 195 | section_00_0194.wav,0 196 | section_00_0195.wav,0 197 | section_00_0196.wav,0 198 | section_00_0197.wav,0 199 | section_00_0198.wav,0 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_data/ground_truth_ToyTank_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,0 2 | section_00_0001.wav,1 3 | section_00_0002.wav,1 4 | section_00_0003.wav,1 5 | section_00_0004.wav,0 6 | section_00_0005.wav,1 7 | section_00_0006.wav,0 8 | section_00_0007.wav,1 9 | section_00_0008.wav,0 10 | section_00_0009.wav,1 11 | section_00_0010.wav,1 12 | section_00_0011.wav,1 13 | section_00_0012.wav,0 14 | section_00_0013.wav,0 15 | section_00_0014.wav,1 16 | section_00_0015.wav,0 17 | section_00_0016.wav,0 18 | section_00_0017.wav,1 19 | section_00_0018.wav,0 20 | section_00_0019.wav,1 21 | section_00_0020.wav,0 22 | section_00_0021.wav,0 23 | section_00_0022.wav,0 24 | section_00_0023.wav,1 25 | section_00_0024.wav,1 26 | section_00_0025.wav,1 27 | section_00_0026.wav,1 28 | section_00_0027.wav,0 29 | section_00_0028.wav,0 30 | section_00_0029.wav,0 31 | section_00_0030.wav,0 32 | section_00_0031.wav,1 33 | section_00_0032.wav,1 34 | section_00_0033.wav,0 35 | section_00_0034.wav,1 36 | section_00_0035.wav,0 37 | section_00_0036.wav,0 38 | section_00_0037.wav,1 39 | section_00_0038.wav,1 40 | section_00_0039.wav,1 41 | section_00_0040.wav,1 42 | section_00_0041.wav,1 43 | section_00_0042.wav,1 44 | section_00_0043.wav,1 45 | section_00_0044.wav,1 46 | section_00_0045.wav,1 47 | section_00_0046.wav,0 48 | section_00_0047.wav,0 49 | section_00_0048.wav,0 50 | section_00_0049.wav,1 51 | section_00_0050.wav,1 52 | section_00_0051.wav,0 53 | section_00_0052.wav,0 54 | section_00_0053.wav,1 55 | section_00_0054.wav,0 56 | section_00_0055.wav,1 57 | section_00_0056.wav,1 58 | section_00_0057.wav,0 59 | section_00_0058.wav,0 60 | section_00_0059.wav,1 61 | section_00_0060.wav,1 62 | section_00_0061.wav,0 63 | section_00_0062.wav,0 64 | section_00_0063.wav,0 65 | section_00_0064.wav,1 66 | section_00_0065.wav,1 67 | section_00_0066.wav,1 68 | section_00_0067.wav,0 69 | section_00_0068.wav,0 70 | section_00_0069.wav,1 71 | section_00_0070.wav,1 72 | section_00_0071.wav,1 73 | section_00_0072.wav,0 74 | section_00_0073.wav,1 75 | section_00_0074.wav,0 76 | section_00_0075.wav,0 77 | section_00_0076.wav,1 78 | section_00_0077.wav,0 79 | section_00_0078.wav,1 80 | section_00_0079.wav,1 81 | section_00_0080.wav,1 82 | section_00_0081.wav,0 83 | section_00_0082.wav,1 84 | section_00_0083.wav,1 85 | section_00_0084.wav,1 86 | section_00_0085.wav,0 87 | section_00_0086.wav,1 88 | section_00_0087.wav,1 89 | section_00_0088.wav,0 90 | section_00_0089.wav,0 91 | section_00_0090.wav,1 92 | section_00_0091.wav,0 93 | section_00_0092.wav,0 94 | section_00_0093.wav,0 95 | section_00_0094.wav,1 96 | section_00_0095.wav,0 97 | section_00_0096.wav,1 98 | section_00_0097.wav,0 99 | section_00_0098.wav,0 100 | section_00_0099.wav,1 101 | section_00_0100.wav,1 102 | section_00_0101.wav,1 103 | section_00_0102.wav,0 104 | section_00_0103.wav,1 105 | section_00_0104.wav,0 106 | section_00_0105.wav,0 107 | section_00_0106.wav,0 108 | section_00_0107.wav,0 109 | section_00_0108.wav,1 110 | section_00_0109.wav,1 111 | section_00_0110.wav,1 112 | section_00_0111.wav,0 113 | section_00_0112.wav,0 114 | section_00_0113.wav,0 115 | section_00_0114.wav,1 116 | section_00_0115.wav,1 117 | section_00_0116.wav,0 118 | section_00_0117.wav,0 119 | section_00_0118.wav,1 120 | section_00_0119.wav,0 121 | section_00_0120.wav,0 122 | section_00_0121.wav,0 123 | section_00_0122.wav,1 124 | section_00_0123.wav,0 125 | section_00_0124.wav,1 126 | section_00_0125.wav,0 127 | section_00_0126.wav,0 128 | section_00_0127.wav,1 129 | section_00_0128.wav,0 130 | section_00_0129.wav,1 131 | section_00_0130.wav,1 132 | section_00_0131.wav,1 133 | section_00_0132.wav,0 134 | section_00_0133.wav,0 135 | section_00_0134.wav,0 136 | section_00_0135.wav,1 137 | section_00_0136.wav,0 138 | section_00_0137.wav,1 139 | section_00_0138.wav,1 140 | section_00_0139.wav,0 141 | section_00_0140.wav,0 142 | section_00_0141.wav,0 143 | section_00_0142.wav,1 144 | section_00_0143.wav,0 145 | section_00_0144.wav,0 146 | section_00_0145.wav,1 147 | section_00_0146.wav,0 148 | section_00_0147.wav,1 149 | section_00_0148.wav,1 150 | section_00_0149.wav,0 151 | section_00_0150.wav,0 152 | section_00_0151.wav,1 153 | section_00_0152.wav,0 154 | section_00_0153.wav,0 155 | section_00_0154.wav,1 156 | section_00_0155.wav,0 157 | section_00_0156.wav,0 158 | section_00_0157.wav,0 159 | section_00_0158.wav,0 160 | section_00_0159.wav,1 161 | section_00_0160.wav,1 162 | section_00_0161.wav,0 163 | section_00_0162.wav,1 164 | section_00_0163.wav,1 165 | section_00_0164.wav,1 166 | section_00_0165.wav,0 167 | section_00_0166.wav,1 168 | section_00_0167.wav,0 169 | section_00_0168.wav,0 170 | section_00_0169.wav,1 171 | section_00_0170.wav,1 172 | section_00_0171.wav,1 173 | section_00_0172.wav,1 174 | section_00_0173.wav,0 175 | section_00_0174.wav,0 176 | section_00_0175.wav,1 177 | section_00_0176.wav,0 178 | section_00_0177.wav,1 179 | section_00_0178.wav,1 180 | section_00_0179.wav,0 181 | section_00_0180.wav,0 182 | section_00_0181.wav,1 183 | section_00_0182.wav,0 184 | section_00_0183.wav,0 185 | section_00_0184.wav,1 186 | section_00_0185.wav,0 187 | section_00_0186.wav,1 188 | section_00_0187.wav,0 189 | section_00_0188.wav,0 190 | section_00_0189.wav,1 191 | section_00_0190.wav,1 192 | section_00_0191.wav,0 193 | section_00_0192.wav,1 194 | section_00_0193.wav,0 195 | section_00_0194.wav,1 196 | section_00_0195.wav,0 197 | section_00_0196.wav,0 198 | section_00_0197.wav,1 199 | section_00_0198.wav,0 200 | section_00_0199.wav,1 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_data/ground_truth_Vacuum_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,0 2 | section_00_0001.wav,1 3 | section_00_0002.wav,0 4 | section_00_0003.wav,0 5 | section_00_0004.wav,1 6 | section_00_0005.wav,0 7 | section_00_0006.wav,1 8 | section_00_0007.wav,1 9 | section_00_0008.wav,1 10 | section_00_0009.wav,1 11 | section_00_0010.wav,1 12 | section_00_0011.wav,1 13 | section_00_0012.wav,1 14 | section_00_0013.wav,0 15 | section_00_0014.wav,0 16 | section_00_0015.wav,1 17 | section_00_0016.wav,0 18 | section_00_0017.wav,1 19 | section_00_0018.wav,0 20 | section_00_0019.wav,0 21 | section_00_0020.wav,0 22 | section_00_0021.wav,1 23 | section_00_0022.wav,0 24 | section_00_0023.wav,1 25 | section_00_0024.wav,1 26 | section_00_0025.wav,1 27 | section_00_0026.wav,1 28 | section_00_0027.wav,1 29 | section_00_0028.wav,1 30 | section_00_0029.wav,1 31 | section_00_0030.wav,1 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,0 35 | section_00_0034.wav,1 36 | section_00_0035.wav,1 37 | section_00_0036.wav,0 38 | section_00_0037.wav,1 39 | section_00_0038.wav,0 40 | section_00_0039.wav,1 41 | section_00_0040.wav,1 42 | section_00_0041.wav,1 43 | section_00_0042.wav,1 44 | section_00_0043.wav,0 45 | section_00_0044.wav,0 46 | section_00_0045.wav,1 47 | section_00_0046.wav,0 48 | section_00_0047.wav,0 49 | section_00_0048.wav,0 50 | section_00_0049.wav,0 51 | section_00_0050.wav,0 52 | section_00_0051.wav,1 53 | section_00_0052.wav,0 54 | section_00_0053.wav,0 55 | section_00_0054.wav,0 56 | section_00_0055.wav,0 57 | section_00_0056.wav,0 58 | section_00_0057.wav,0 59 | section_00_0058.wav,1 60 | section_00_0059.wav,1 61 | section_00_0060.wav,0 62 | section_00_0061.wav,0 63 | section_00_0062.wav,1 64 | section_00_0063.wav,1 65 | section_00_0064.wav,1 66 | section_00_0065.wav,1 67 | section_00_0066.wav,1 68 | section_00_0067.wav,1 69 | section_00_0068.wav,0 70 | section_00_0069.wav,1 71 | section_00_0070.wav,1 72 | section_00_0071.wav,0 73 | section_00_0072.wav,0 74 | section_00_0073.wav,1 75 | section_00_0074.wav,1 76 | section_00_0075.wav,0 77 | section_00_0076.wav,1 78 | section_00_0077.wav,0 79 | section_00_0078.wav,1 80 | section_00_0079.wav,0 81 | section_00_0080.wav,0 82 | section_00_0081.wav,1 83 | section_00_0082.wav,0 84 | section_00_0083.wav,1 85 | section_00_0084.wav,1 86 | section_00_0085.wav,1 87 | section_00_0086.wav,0 88 | section_00_0087.wav,0 89 | section_00_0088.wav,1 90 | section_00_0089.wav,0 91 | section_00_0090.wav,0 92 | section_00_0091.wav,0 93 | section_00_0092.wav,0 94 | section_00_0093.wav,0 95 | section_00_0094.wav,0 96 | section_00_0095.wav,1 97 | section_00_0096.wav,0 98 | section_00_0097.wav,1 99 | section_00_0098.wav,1 100 | section_00_0099.wav,0 101 | section_00_0100.wav,1 102 | section_00_0101.wav,1 103 | section_00_0102.wav,1 104 | section_00_0103.wav,1 105 | section_00_0104.wav,0 106 | section_00_0105.wav,1 107 | section_00_0106.wav,1 108 | section_00_0107.wav,0 109 | section_00_0108.wav,1 110 | section_00_0109.wav,0 111 | section_00_0110.wav,1 112 | section_00_0111.wav,0 113 | section_00_0112.wav,0 114 | section_00_0113.wav,1 115 | section_00_0114.wav,1 116 | section_00_0115.wav,1 117 | section_00_0116.wav,0 118 | section_00_0117.wav,0 119 | section_00_0118.wav,1 120 | section_00_0119.wav,1 121 | section_00_0120.wav,1 122 | section_00_0121.wav,1 123 | section_00_0122.wav,1 124 | section_00_0123.wav,0 125 | section_00_0124.wav,1 126 | section_00_0125.wav,0 127 | section_00_0126.wav,0 128 | section_00_0127.wav,0 129 | section_00_0128.wav,1 130 | section_00_0129.wav,1 131 | section_00_0130.wav,0 132 | section_00_0131.wav,1 133 | section_00_0132.wav,1 134 | section_00_0133.wav,0 135 | section_00_0134.wav,1 136 | section_00_0135.wav,1 137 | section_00_0136.wav,1 138 | section_00_0137.wav,0 139 | section_00_0138.wav,1 140 | section_00_0139.wav,1 141 | section_00_0140.wav,0 142 | section_00_0141.wav,1 143 | section_00_0142.wav,0 144 | section_00_0143.wav,0 145 | section_00_0144.wav,1 146 | section_00_0145.wav,0 147 | section_00_0146.wav,0 148 | section_00_0147.wav,0 149 | section_00_0148.wav,0 150 | section_00_0149.wav,0 151 | section_00_0150.wav,1 152 | section_00_0151.wav,0 153 | section_00_0152.wav,0 154 | section_00_0153.wav,0 155 | section_00_0154.wav,1 156 | section_00_0155.wav,1 157 | section_00_0156.wav,0 158 | section_00_0157.wav,1 159 | section_00_0158.wav,1 160 | section_00_0159.wav,0 161 | section_00_0160.wav,0 162 | section_00_0161.wav,1 163 | section_00_0162.wav,1 164 | section_00_0163.wav,0 165 | section_00_0164.wav,0 166 | section_00_0165.wav,0 167 | section_00_0166.wav,0 168 | section_00_0167.wav,0 169 | section_00_0168.wav,0 170 | section_00_0169.wav,1 171 | section_00_0170.wav,0 172 | section_00_0171.wav,1 173 | section_00_0172.wav,0 174 | section_00_0173.wav,0 175 | section_00_0174.wav,1 176 | section_00_0175.wav,1 177 | section_00_0176.wav,1 178 | section_00_0177.wav,1 179 | section_00_0178.wav,0 180 | section_00_0179.wav,1 181 | section_00_0180.wav,0 182 | section_00_0181.wav,0 183 | section_00_0182.wav,0 184 | section_00_0183.wav,0 185 | section_00_0184.wav,1 186 | section_00_0185.wav,0 187 | section_00_0186.wav,1 188 | section_00_0187.wav,0 189 | section_00_0188.wav,0 190 | section_00_0189.wav,1 191 | section_00_0190.wav,1 192 | section_00_0191.wav,0 193 | section_00_0192.wav,0 194 | section_00_0193.wav,1 195 | section_00_0194.wav,1 196 | section_00_0195.wav,0 197 | section_00_0196.wav,0 198 | section_00_0197.wav,0 199 | section_00_0198.wav,0 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_data/ground_truth_bandsaw_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,0 2 | section_00_0001.wav,1 3 | section_00_0002.wav,0 4 | section_00_0003.wav,1 5 | section_00_0004.wav,0 6 | section_00_0005.wav,1 7 | section_00_0006.wav,0 8 | section_00_0007.wav,0 9 | section_00_0008.wav,0 10 | section_00_0009.wav,1 11 | section_00_0010.wav,0 12 | section_00_0011.wav,0 13 | section_00_0012.wav,0 14 | section_00_0013.wav,0 15 | section_00_0014.wav,0 16 | section_00_0015.wav,0 17 | section_00_0016.wav,1 18 | section_00_0017.wav,0 19 | section_00_0018.wav,1 20 | section_00_0019.wav,0 21 | section_00_0020.wav,0 22 | section_00_0021.wav,1 23 | section_00_0022.wav,1 24 | section_00_0023.wav,0 25 | section_00_0024.wav,0 26 | section_00_0025.wav,0 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,0 30 | section_00_0029.wav,1 31 | section_00_0030.wav,0 32 | section_00_0031.wav,0 33 | section_00_0032.wav,0 34 | section_00_0033.wav,0 35 | section_00_0034.wav,0 36 | section_00_0035.wav,1 37 | section_00_0036.wav,0 38 | section_00_0037.wav,0 39 | section_00_0038.wav,0 40 | section_00_0039.wav,0 41 | section_00_0040.wav,1 42 | section_00_0041.wav,0 43 | section_00_0042.wav,0 44 | section_00_0043.wav,0 45 | section_00_0044.wav,1 46 | section_00_0045.wav,1 47 | section_00_0046.wav,1 48 | section_00_0047.wav,0 49 | section_00_0048.wav,0 50 | section_00_0049.wav,1 51 | section_00_0050.wav,0 52 | section_00_0051.wav,1 53 | section_00_0052.wav,0 54 | section_00_0053.wav,0 55 | section_00_0054.wav,0 56 | section_00_0055.wav,1 57 | section_00_0056.wav,0 58 | section_00_0057.wav,0 59 | section_00_0058.wav,0 60 | section_00_0059.wav,0 61 | section_00_0060.wav,0 62 | section_00_0061.wav,1 63 | section_00_0062.wav,1 64 | section_00_0063.wav,1 65 | section_00_0064.wav,0 66 | section_00_0065.wav,0 67 | section_00_0066.wav,0 68 | section_00_0067.wav,0 69 | section_00_0068.wav,0 70 | section_00_0069.wav,1 71 | section_00_0070.wav,0 72 | section_00_0071.wav,1 73 | section_00_0072.wav,0 74 | section_00_0073.wav,0 75 | section_00_0074.wav,0 76 | section_00_0075.wav,1 77 | section_00_0076.wav,0 78 | section_00_0077.wav,1 79 | section_00_0078.wav,0 80 | section_00_0079.wav,1 81 | section_00_0080.wav,0 82 | section_00_0081.wav,0 83 | section_00_0082.wav,1 84 | section_00_0083.wav,1 85 | section_00_0084.wav,0 86 | section_00_0085.wav,1 87 | section_00_0086.wav,1 88 | section_00_0087.wav,0 89 | section_00_0088.wav,0 90 | section_00_0089.wav,1 91 | section_00_0090.wav,0 92 | section_00_0091.wav,1 93 | section_00_0092.wav,1 94 | section_00_0093.wav,1 95 | section_00_0094.wav,0 96 | section_00_0095.wav,1 97 | section_00_0096.wav,1 98 | section_00_0097.wav,0 99 | section_00_0098.wav,1 100 | section_00_0099.wav,0 101 | section_00_0100.wav,1 102 | section_00_0101.wav,0 103 | section_00_0102.wav,0 104 | section_00_0103.wav,0 105 | section_00_0104.wav,1 106 | section_00_0105.wav,0 107 | section_00_0106.wav,1 108 | section_00_0107.wav,0 109 | section_00_0108.wav,0 110 | section_00_0109.wav,1 111 | section_00_0110.wav,1 112 | section_00_0111.wav,1 113 | section_00_0112.wav,1 114 | section_00_0113.wav,1 115 | section_00_0114.wav,0 116 | section_00_0115.wav,1 117 | section_00_0116.wav,0 118 | section_00_0117.wav,1 119 | section_00_0118.wav,1 120 | section_00_0119.wav,0 121 | section_00_0120.wav,1 122 | section_00_0121.wav,0 123 | section_00_0122.wav,1 124 | section_00_0123.wav,0 125 | section_00_0124.wav,0 126 | section_00_0125.wav,0 127 | section_00_0126.wav,0 128 | section_00_0127.wav,0 129 | section_00_0128.wav,0 130 | section_00_0129.wav,0 131 | section_00_0130.wav,0 132 | section_00_0131.wav,0 133 | section_00_0132.wav,1 134 | section_00_0133.wav,0 135 | section_00_0134.wav,1 136 | section_00_0135.wav,0 137 | section_00_0136.wav,1 138 | section_00_0137.wav,0 139 | section_00_0138.wav,0 140 | section_00_0139.wav,0 141 | section_00_0140.wav,0 142 | section_00_0141.wav,0 143 | section_00_0142.wav,0 144 | section_00_0143.wav,1 145 | section_00_0144.wav,0 146 | section_00_0145.wav,1 147 | section_00_0146.wav,0 148 | section_00_0147.wav,0 149 | section_00_0148.wav,0 150 | section_00_0149.wav,1 151 | section_00_0150.wav,0 152 | section_00_0151.wav,0 153 | section_00_0152.wav,0 154 | section_00_0153.wav,0 155 | section_00_0154.wav,1 156 | section_00_0155.wav,0 157 | section_00_0156.wav,1 158 | section_00_0157.wav,1 159 | section_00_0158.wav,1 160 | section_00_0159.wav,0 161 | section_00_0160.wav,0 162 | section_00_0161.wav,1 163 | section_00_0162.wav,1 164 | section_00_0163.wav,0 165 | section_00_0164.wav,0 166 | section_00_0165.wav,0 167 | section_00_0166.wav,0 168 | section_00_0167.wav,0 169 | section_00_0168.wav,1 170 | section_00_0169.wav,0 171 | section_00_0170.wav,0 172 | section_00_0171.wav,1 173 | section_00_0172.wav,1 174 | section_00_0173.wav,0 175 | section_00_0174.wav,1 176 | section_00_0175.wav,1 177 | section_00_0176.wav,1 178 | section_00_0177.wav,0 179 | section_00_0178.wav,0 180 | section_00_0179.wav,0 181 | section_00_0180.wav,0 182 | section_00_0181.wav,0 183 | section_00_0182.wav,1 184 | section_00_0183.wav,0 185 | section_00_0184.wav,0 186 | section_00_0185.wav,1 187 | section_00_0186.wav,0 188 | section_00_0187.wav,0 189 | section_00_0188.wav,1 190 | section_00_0189.wav,0 191 | section_00_0190.wav,0 192 | section_00_0191.wav,1 193 | section_00_0192.wav,0 194 | section_00_0193.wav,0 195 | section_00_0194.wav,1 196 | section_00_0195.wav,0 197 | section_00_0196.wav,0 198 | section_00_0197.wav,0 199 | section_00_0198.wav,1 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_data/ground_truth_grinder_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,0 2 | section_00_0001.wav,0 3 | section_00_0002.wav,0 4 | section_00_0003.wav,0 5 | section_00_0004.wav,0 6 | section_00_0005.wav,0 7 | section_00_0006.wav,0 8 | section_00_0007.wav,0 9 | section_00_0008.wav,0 10 | section_00_0009.wav,0 11 | section_00_0010.wav,1 12 | section_00_0011.wav,0 13 | section_00_0012.wav,1 14 | section_00_0013.wav,0 15 | section_00_0014.wav,0 16 | section_00_0015.wav,0 17 | section_00_0016.wav,0 18 | section_00_0017.wav,0 19 | section_00_0018.wav,1 20 | section_00_0019.wav,1 21 | section_00_0020.wav,0 22 | section_00_0021.wav,0 23 | section_00_0022.wav,0 24 | section_00_0023.wav,0 25 | section_00_0024.wav,1 26 | section_00_0025.wav,1 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,0 30 | section_00_0029.wav,0 31 | section_00_0030.wav,1 32 | section_00_0031.wav,1 33 | section_00_0032.wav,0 34 | section_00_0033.wav,0 35 | section_00_0034.wav,0 36 | section_00_0035.wav,0 37 | section_00_0036.wav,0 38 | section_00_0037.wav,0 39 | section_00_0038.wav,0 40 | section_00_0039.wav,1 41 | section_00_0040.wav,0 42 | section_00_0041.wav,1 43 | section_00_0042.wav,0 44 | section_00_0043.wav,1 45 | section_00_0044.wav,0 46 | section_00_0045.wav,0 47 | section_00_0046.wav,1 48 | section_00_0047.wav,0 49 | section_00_0048.wav,1 50 | section_00_0049.wav,0 51 | section_00_0050.wav,0 52 | section_00_0051.wav,0 53 | section_00_0052.wav,0 54 | section_00_0053.wav,0 55 | section_00_0054.wav,0 56 | section_00_0055.wav,0 57 | section_00_0056.wav,1 58 | section_00_0057.wav,1 59 | section_00_0058.wav,0 60 | section_00_0059.wav,0 61 | section_00_0060.wav,0 62 | section_00_0061.wav,0 63 | section_00_0062.wav,0 64 | section_00_0063.wav,0 65 | section_00_0064.wav,0 66 | section_00_0065.wav,1 67 | section_00_0066.wav,0 68 | section_00_0067.wav,0 69 | section_00_0068.wav,0 70 | section_00_0069.wav,0 71 | section_00_0070.wav,0 72 | section_00_0071.wav,0 73 | section_00_0072.wav,0 74 | section_00_0073.wav,1 75 | section_00_0074.wav,0 76 | section_00_0075.wav,0 77 | section_00_0076.wav,1 78 | section_00_0077.wav,0 79 | section_00_0078.wav,0 80 | section_00_0079.wav,0 81 | section_00_0080.wav,0 82 | section_00_0081.wav,0 83 | section_00_0082.wav,0 84 | section_00_0083.wav,0 85 | section_00_0084.wav,0 86 | section_00_0085.wav,0 87 | section_00_0086.wav,0 88 | section_00_0087.wav,0 89 | section_00_0088.wav,0 90 | section_00_0089.wav,0 91 | section_00_0090.wav,1 92 | section_00_0091.wav,1 93 | section_00_0092.wav,0 94 | section_00_0093.wav,0 95 | section_00_0094.wav,1 96 | section_00_0095.wav,1 97 | section_00_0096.wav,0 98 | section_00_0097.wav,1 99 | section_00_0098.wav,0 100 | section_00_0099.wav,0 101 | section_00_0100.wav,1 102 | section_00_0101.wav,0 103 | section_00_0102.wav,1 104 | section_00_0103.wav,1 105 | section_00_0104.wav,0 106 | section_00_0105.wav,0 107 | section_00_0106.wav,1 108 | section_00_0107.wav,1 109 | section_00_0108.wav,0 110 | section_00_0109.wav,1 111 | section_00_0110.wav,0 112 | section_00_0111.wav,0 113 | section_00_0112.wav,1 114 | section_00_0113.wav,1 115 | section_00_0114.wav,0 116 | section_00_0115.wav,1 117 | section_00_0116.wav,0 118 | section_00_0117.wav,0 119 | section_00_0118.wav,0 120 | section_00_0119.wav,0 121 | section_00_0120.wav,0 122 | section_00_0121.wav,0 123 | section_00_0122.wav,0 124 | section_00_0123.wav,1 125 | section_00_0124.wav,0 126 | section_00_0125.wav,0 127 | section_00_0126.wav,0 128 | section_00_0127.wav,0 129 | section_00_0128.wav,1 130 | section_00_0129.wav,1 131 | section_00_0130.wav,0 132 | section_00_0131.wav,0 133 | section_00_0132.wav,0 134 | section_00_0133.wav,0 135 | section_00_0134.wav,1 136 | section_00_0135.wav,0 137 | section_00_0136.wav,1 138 | section_00_0137.wav,0 139 | section_00_0138.wav,0 140 | section_00_0139.wav,1 141 | section_00_0140.wav,1 142 | section_00_0141.wav,0 143 | section_00_0142.wav,1 144 | section_00_0143.wav,1 145 | section_00_0144.wav,1 146 | section_00_0145.wav,0 147 | section_00_0146.wav,0 148 | section_00_0147.wav,0 149 | section_00_0148.wav,0 150 | section_00_0149.wav,1 151 | section_00_0150.wav,0 152 | section_00_0151.wav,0 153 | section_00_0152.wav,1 154 | section_00_0153.wav,1 155 | section_00_0154.wav,0 156 | section_00_0155.wav,0 157 | section_00_0156.wav,1 158 | section_00_0157.wav,1 159 | section_00_0158.wav,0 160 | section_00_0159.wav,0 161 | section_00_0160.wav,0 162 | section_00_0161.wav,0 163 | section_00_0162.wav,0 164 | section_00_0163.wav,0 165 | section_00_0164.wav,0 166 | section_00_0165.wav,1 167 | section_00_0166.wav,0 168 | section_00_0167.wav,0 169 | section_00_0168.wav,0 170 | section_00_0169.wav,0 171 | section_00_0170.wav,0 172 | section_00_0171.wav,0 173 | section_00_0172.wav,0 174 | section_00_0173.wav,0 175 | section_00_0174.wav,0 176 | section_00_0175.wav,0 177 | section_00_0176.wav,0 178 | section_00_0177.wav,0 179 | section_00_0178.wav,0 180 | section_00_0179.wav,1 181 | section_00_0180.wav,1 182 | section_00_0181.wav,1 183 | section_00_0182.wav,0 184 | section_00_0183.wav,0 185 | section_00_0184.wav,0 186 | section_00_0185.wav,1 187 | section_00_0186.wav,0 188 | section_00_0187.wav,0 189 | section_00_0188.wav,1 190 | section_00_0189.wav,0 191 | section_00_0190.wav,0 192 | section_00_0191.wav,0 193 | section_00_0192.wav,1 194 | section_00_0193.wav,1 195 | section_00_0194.wav,0 196 | section_00_0195.wav,0 197 | section_00_0196.wav,1 198 | section_00_0197.wav,1 199 | section_00_0198.wav,0 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_data/ground_truth_shaker_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,1 2 | section_00_0001.wav,1 3 | section_00_0002.wav,1 4 | section_00_0003.wav,0 5 | section_00_0004.wav,1 6 | section_00_0005.wav,0 7 | section_00_0006.wav,1 8 | section_00_0007.wav,0 9 | section_00_0008.wav,1 10 | section_00_0009.wav,0 11 | section_00_0010.wav,1 12 | section_00_0011.wav,0 13 | section_00_0012.wav,0 14 | section_00_0013.wav,1 15 | section_00_0014.wav,0 16 | section_00_0015.wav,0 17 | section_00_0016.wav,1 18 | section_00_0017.wav,1 19 | section_00_0018.wav,1 20 | section_00_0019.wav,1 21 | section_00_0020.wav,0 22 | section_00_0021.wav,0 23 | section_00_0022.wav,0 24 | section_00_0023.wav,1 25 | section_00_0024.wav,0 26 | section_00_0025.wav,0 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,0 30 | section_00_0029.wav,1 31 | section_00_0030.wav,0 32 | section_00_0031.wav,1 33 | section_00_0032.wav,0 34 | section_00_0033.wav,1 35 | section_00_0034.wav,0 36 | section_00_0035.wav,1 37 | section_00_0036.wav,1 38 | section_00_0037.wav,0 39 | section_00_0038.wav,0 40 | section_00_0039.wav,0 41 | section_00_0040.wav,0 42 | section_00_0041.wav,1 43 | section_00_0042.wav,1 44 | section_00_0043.wav,1 45 | section_00_0044.wav,1 46 | section_00_0045.wav,0 47 | section_00_0046.wav,0 48 | section_00_0047.wav,1 49 | section_00_0048.wav,0 50 | section_00_0049.wav,0 51 | section_00_0050.wav,0 52 | section_00_0051.wav,1 53 | section_00_0052.wav,0 54 | section_00_0053.wav,0 55 | section_00_0054.wav,1 56 | section_00_0055.wav,0 57 | section_00_0056.wav,0 58 | section_00_0057.wav,1 59 | section_00_0058.wav,1 60 | section_00_0059.wav,0 61 | section_00_0060.wav,0 62 | section_00_0061.wav,0 63 | section_00_0062.wav,0 64 | section_00_0063.wav,1 65 | section_00_0064.wav,0 66 | section_00_0065.wav,0 67 | section_00_0066.wav,1 68 | section_00_0067.wav,1 69 | section_00_0068.wav,1 70 | section_00_0069.wav,0 71 | section_00_0070.wav,0 72 | section_00_0071.wav,0 73 | section_00_0072.wav,1 74 | section_00_0073.wav,1 75 | section_00_0074.wav,0 76 | section_00_0075.wav,1 77 | section_00_0076.wav,1 78 | section_00_0077.wav,0 79 | section_00_0078.wav,0 80 | section_00_0079.wav,0 81 | section_00_0080.wav,1 82 | section_00_0081.wav,1 83 | section_00_0082.wav,0 84 | section_00_0083.wav,1 85 | section_00_0084.wav,0 86 | section_00_0085.wav,1 87 | section_00_0086.wav,1 88 | section_00_0087.wav,0 89 | section_00_0088.wav,1 90 | section_00_0089.wav,1 91 | section_00_0090.wav,0 92 | section_00_0091.wav,1 93 | section_00_0092.wav,0 94 | section_00_0093.wav,0 95 | section_00_0094.wav,1 96 | section_00_0095.wav,0 97 | section_00_0096.wav,1 98 | section_00_0097.wav,1 99 | section_00_0098.wav,1 100 | section_00_0099.wav,0 101 | section_00_0100.wav,1 102 | section_00_0101.wav,1 103 | section_00_0102.wav,1 104 | section_00_0103.wav,1 105 | section_00_0104.wav,1 106 | section_00_0105.wav,1 107 | section_00_0106.wav,1 108 | section_00_0107.wav,0 109 | section_00_0108.wav,1 110 | section_00_0109.wav,0 111 | section_00_0110.wav,0 112 | section_00_0111.wav,1 113 | section_00_0112.wav,1 114 | section_00_0113.wav,1 115 | section_00_0114.wav,1 116 | section_00_0115.wav,0 117 | section_00_0116.wav,0 118 | section_00_0117.wav,1 119 | section_00_0118.wav,0 120 | section_00_0119.wav,1 121 | section_00_0120.wav,0 122 | section_00_0121.wav,1 123 | section_00_0122.wav,1 124 | section_00_0123.wav,0 125 | section_00_0124.wav,1 126 | section_00_0125.wav,0 127 | section_00_0126.wav,0 128 | section_00_0127.wav,1 129 | section_00_0128.wav,0 130 | section_00_0129.wav,0 131 | section_00_0130.wav,0 132 | section_00_0131.wav,0 133 | section_00_0132.wav,0 134 | section_00_0133.wav,1 135 | section_00_0134.wav,1 136 | section_00_0135.wav,1 137 | section_00_0136.wav,1 138 | section_00_0137.wav,0 139 | section_00_0138.wav,1 140 | section_00_0139.wav,0 141 | section_00_0140.wav,1 142 | section_00_0141.wav,1 143 | section_00_0142.wav,1 144 | section_00_0143.wav,0 145 | section_00_0144.wav,0 146 | section_00_0145.wav,0 147 | section_00_0146.wav,1 148 | section_00_0147.wav,1 149 | section_00_0148.wav,0 150 | section_00_0149.wav,1 151 | section_00_0150.wav,1 152 | section_00_0151.wav,0 153 | section_00_0152.wav,1 154 | section_00_0153.wav,1 155 | section_00_0154.wav,0 156 | section_00_0155.wav,1 157 | section_00_0156.wav,1 158 | section_00_0157.wav,0 159 | section_00_0158.wav,1 160 | section_00_0159.wav,1 161 | section_00_0160.wav,0 162 | section_00_0161.wav,0 163 | section_00_0162.wav,1 164 | section_00_0163.wav,0 165 | section_00_0164.wav,0 166 | section_00_0165.wav,0 167 | section_00_0166.wav,0 168 | section_00_0167.wav,0 169 | section_00_0168.wav,0 170 | section_00_0169.wav,1 171 | section_00_0170.wav,1 172 | section_00_0171.wav,0 173 | section_00_0172.wav,0 174 | section_00_0173.wav,1 175 | section_00_0174.wav,1 176 | section_00_0175.wav,0 177 | section_00_0176.wav,0 178 | section_00_0177.wav,1 179 | section_00_0178.wav,0 180 | section_00_0179.wav,1 181 | section_00_0180.wav,0 182 | section_00_0181.wav,1 183 | section_00_0182.wav,0 184 | section_00_0183.wav,1 185 | section_00_0184.wav,1 186 | section_00_0185.wav,1 187 | section_00_0186.wav,1 188 | section_00_0187.wav,0 189 | section_00_0188.wav,0 190 | section_00_0189.wav,1 191 | section_00_0190.wav,0 192 | section_00_0191.wav,1 193 | section_00_0192.wav,0 194 | section_00_0193.wav,0 195 | section_00_0194.wav,1 196 | section_00_0195.wav,0 197 | section_00_0196.wav,0 198 | section_00_0197.wav,0 199 | section_00_0198.wav,1 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_domain/ground_truth_ToyTank_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,1 2 | section_00_0001.wav,1 3 | section_00_0002.wav,0 4 | section_00_0003.wav,1 5 | section_00_0004.wav,0 6 | section_00_0005.wav,0 7 | section_00_0006.wav,0 8 | section_00_0007.wav,1 9 | section_00_0008.wav,0 10 | section_00_0009.wav,1 11 | section_00_0010.wav,0 12 | section_00_0011.wav,1 13 | section_00_0012.wav,1 14 | section_00_0013.wav,0 15 | section_00_0014.wav,0 16 | section_00_0015.wav,1 17 | section_00_0016.wav,0 18 | section_00_0017.wav,0 19 | section_00_0018.wav,1 20 | section_00_0019.wav,0 21 | section_00_0020.wav,0 22 | section_00_0021.wav,1 23 | section_00_0022.wav,1 24 | section_00_0023.wav,1 25 | section_00_0024.wav,0 26 | section_00_0025.wav,1 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,1 30 | section_00_0029.wav,0 31 | section_00_0030.wav,0 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,1 35 | section_00_0034.wav,1 36 | section_00_0035.wav,0 37 | section_00_0036.wav,0 38 | section_00_0037.wav,0 39 | section_00_0038.wav,1 40 | section_00_0039.wav,1 41 | section_00_0040.wav,0 42 | section_00_0041.wav,0 43 | section_00_0042.wav,1 44 | section_00_0043.wav,1 45 | section_00_0044.wav,0 46 | section_00_0045.wav,1 47 | section_00_0046.wav,1 48 | section_00_0047.wav,1 49 | section_00_0048.wav,0 50 | section_00_0049.wav,0 51 | section_00_0050.wav,1 52 | section_00_0051.wav,1 53 | section_00_0052.wav,0 54 | section_00_0053.wav,0 55 | section_00_0054.wav,1 56 | section_00_0055.wav,0 57 | section_00_0056.wav,0 58 | section_00_0057.wav,1 59 | section_00_0058.wav,0 60 | section_00_0059.wav,0 61 | section_00_0060.wav,1 62 | section_00_0061.wav,1 63 | section_00_0062.wav,1 64 | section_00_0063.wav,0 65 | section_00_0064.wav,1 66 | section_00_0065.wav,0 67 | section_00_0066.wav,1 68 | section_00_0067.wav,0 69 | section_00_0068.wav,1 70 | section_00_0069.wav,0 71 | section_00_0070.wav,0 72 | section_00_0071.wav,0 73 | section_00_0072.wav,0 74 | section_00_0073.wav,1 75 | section_00_0074.wav,0 76 | section_00_0075.wav,1 77 | section_00_0076.wav,0 78 | section_00_0077.wav,0 79 | section_00_0078.wav,0 80 | section_00_0079.wav,1 81 | section_00_0080.wav,1 82 | section_00_0081.wav,0 83 | section_00_0082.wav,0 84 | section_00_0083.wav,1 85 | section_00_0084.wav,1 86 | section_00_0085.wav,1 87 | section_00_0086.wav,1 88 | section_00_0087.wav,0 89 | section_00_0088.wav,1 90 | section_00_0089.wav,1 91 | section_00_0090.wav,0 92 | section_00_0091.wav,1 93 | section_00_0092.wav,1 94 | section_00_0093.wav,1 95 | section_00_0094.wav,1 96 | section_00_0095.wav,1 97 | section_00_0096.wav,0 98 | section_00_0097.wav,0 99 | section_00_0098.wav,1 100 | section_00_0099.wav,1 101 | section_00_0100.wav,1 102 | section_00_0101.wav,0 103 | section_00_0102.wav,0 104 | section_00_0103.wav,0 105 | section_00_0104.wav,0 106 | section_00_0105.wav,1 107 | section_00_0106.wav,0 108 | section_00_0107.wav,0 109 | section_00_0108.wav,0 110 | section_00_0109.wav,0 111 | section_00_0110.wav,0 112 | section_00_0111.wav,0 113 | section_00_0112.wav,0 114 | section_00_0113.wav,1 115 | section_00_0114.wav,0 116 | section_00_0115.wav,0 117 | section_00_0116.wav,1 118 | section_00_0117.wav,1 119 | section_00_0118.wav,0 120 | section_00_0119.wav,1 121 | section_00_0120.wav,0 122 | section_00_0121.wav,1 123 | section_00_0122.wav,0 124 | section_00_0123.wav,1 125 | section_00_0124.wav,0 126 | section_00_0125.wav,1 127 | section_00_0126.wav,1 128 | section_00_0127.wav,1 129 | section_00_0128.wav,1 130 | section_00_0129.wav,0 131 | section_00_0130.wav,1 132 | section_00_0131.wav,1 133 | section_00_0132.wav,0 134 | section_00_0133.wav,0 135 | section_00_0134.wav,0 136 | section_00_0135.wav,0 137 | section_00_0136.wav,1 138 | section_00_0137.wav,1 139 | section_00_0138.wav,0 140 | section_00_0139.wav,1 141 | section_00_0140.wav,0 142 | section_00_0141.wav,0 143 | section_00_0142.wav,1 144 | section_00_0143.wav,1 145 | section_00_0144.wav,1 146 | section_00_0145.wav,1 147 | section_00_0146.wav,1 148 | section_00_0147.wav,1 149 | section_00_0148.wav,0 150 | section_00_0149.wav,0 151 | section_00_0150.wav,0 152 | section_00_0151.wav,0 153 | section_00_0152.wav,0 154 | section_00_0153.wav,1 155 | section_00_0154.wav,1 156 | section_00_0155.wav,0 157 | section_00_0156.wav,1 158 | section_00_0157.wav,1 159 | section_00_0158.wav,0 160 | section_00_0159.wav,1 161 | section_00_0160.wav,0 162 | section_00_0161.wav,1 163 | section_00_0162.wav,1 164 | section_00_0163.wav,1 165 | section_00_0164.wav,1 166 | section_00_0165.wav,0 167 | section_00_0166.wav,0 168 | section_00_0167.wav,1 169 | section_00_0168.wav,0 170 | section_00_0169.wav,1 171 | section_00_0170.wav,1 172 | section_00_0171.wav,1 173 | section_00_0172.wav,1 174 | section_00_0173.wav,0 175 | section_00_0174.wav,1 176 | section_00_0175.wav,0 177 | section_00_0176.wav,0 178 | section_00_0177.wav,1 179 | section_00_0178.wav,1 180 | section_00_0179.wav,0 181 | section_00_0180.wav,1 182 | section_00_0181.wav,1 183 | section_00_0182.wav,1 184 | section_00_0183.wav,1 185 | section_00_0184.wav,0 186 | section_00_0185.wav,0 187 | section_00_0186.wav,0 188 | section_00_0187.wav,0 189 | section_00_0188.wav,0 190 | section_00_0189.wav,1 191 | section_00_0190.wav,1 192 | section_00_0191.wav,0 193 | section_00_0192.wav,1 194 | section_00_0193.wav,0 195 | section_00_0194.wav,1 196 | section_00_0195.wav,1 197 | section_00_0196.wav,0 198 | section_00_0197.wav,0 199 | section_00_0198.wav,0 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_domain/ground_truth_Vacuum_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,1 2 | section_00_0001.wav,0 3 | section_00_0002.wav,1 4 | section_00_0003.wav,0 5 | section_00_0004.wav,1 6 | section_00_0005.wav,0 7 | section_00_0006.wav,0 8 | section_00_0007.wav,1 9 | section_00_0008.wav,0 10 | section_00_0009.wav,1 11 | section_00_0010.wav,1 12 | section_00_0011.wav,1 13 | section_00_0012.wav,0 14 | section_00_0013.wav,1 15 | section_00_0014.wav,0 16 | section_00_0015.wav,1 17 | section_00_0016.wav,0 18 | section_00_0017.wav,1 19 | section_00_0018.wav,1 20 | section_00_0019.wav,1 21 | section_00_0020.wav,0 22 | section_00_0021.wav,0 23 | section_00_0022.wav,0 24 | section_00_0023.wav,0 25 | section_00_0024.wav,0 26 | section_00_0025.wav,1 27 | section_00_0026.wav,1 28 | section_00_0027.wav,1 29 | section_00_0028.wav,0 30 | section_00_0029.wav,1 31 | section_00_0030.wav,0 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,0 35 | section_00_0034.wav,1 36 | section_00_0035.wav,0 37 | section_00_0036.wav,0 38 | section_00_0037.wav,0 39 | section_00_0038.wav,1 40 | section_00_0039.wav,0 41 | section_00_0040.wav,1 42 | section_00_0041.wav,0 43 | section_00_0042.wav,0 44 | section_00_0043.wav,1 45 | section_00_0044.wav,0 46 | section_00_0045.wav,1 47 | section_00_0046.wav,1 48 | section_00_0047.wav,0 49 | section_00_0048.wav,1 50 | section_00_0049.wav,0 51 | section_00_0050.wav,0 52 | section_00_0051.wav,0 53 | section_00_0052.wav,0 54 | section_00_0053.wav,1 55 | section_00_0054.wav,0 56 | section_00_0055.wav,1 57 | section_00_0056.wav,0 58 | section_00_0057.wav,1 59 | section_00_0058.wav,1 60 | section_00_0059.wav,1 61 | section_00_0060.wav,1 62 | section_00_0061.wav,0 63 | section_00_0062.wav,1 64 | section_00_0063.wav,0 65 | section_00_0064.wav,1 66 | section_00_0065.wav,0 67 | section_00_0066.wav,0 68 | section_00_0067.wav,0 69 | section_00_0068.wav,0 70 | section_00_0069.wav,0 71 | section_00_0070.wav,0 72 | section_00_0071.wav,0 73 | section_00_0072.wav,0 74 | section_00_0073.wav,1 75 | section_00_0074.wav,0 76 | section_00_0075.wav,0 77 | section_00_0076.wav,1 78 | section_00_0077.wav,0 79 | section_00_0078.wav,0 80 | section_00_0079.wav,1 81 | section_00_0080.wav,0 82 | section_00_0081.wav,1 83 | section_00_0082.wav,0 84 | section_00_0083.wav,0 85 | section_00_0084.wav,1 86 | section_00_0085.wav,1 87 | section_00_0086.wav,0 88 | section_00_0087.wav,0 89 | section_00_0088.wav,1 90 | section_00_0089.wav,1 91 | section_00_0090.wav,1 92 | section_00_0091.wav,1 93 | section_00_0092.wav,0 94 | section_00_0093.wav,0 95 | section_00_0094.wav,0 96 | section_00_0095.wav,1 97 | section_00_0096.wav,0 98 | section_00_0097.wav,1 99 | section_00_0098.wav,1 100 | section_00_0099.wav,0 101 | section_00_0100.wav,1 102 | section_00_0101.wav,1 103 | section_00_0102.wav,0 104 | section_00_0103.wav,0 105 | section_00_0104.wav,1 106 | section_00_0105.wav,0 107 | section_00_0106.wav,0 108 | section_00_0107.wav,1 109 | section_00_0108.wav,1 110 | section_00_0109.wav,0 111 | section_00_0110.wav,1 112 | section_00_0111.wav,0 113 | section_00_0112.wav,1 114 | section_00_0113.wav,1 115 | section_00_0114.wav,1 116 | section_00_0115.wav,0 117 | section_00_0116.wav,1 118 | section_00_0117.wav,1 119 | section_00_0118.wav,1 120 | section_00_0119.wav,1 121 | section_00_0120.wav,1 122 | section_00_0121.wav,1 123 | section_00_0122.wav,1 124 | section_00_0123.wav,0 125 | section_00_0124.wav,0 126 | section_00_0125.wav,1 127 | section_00_0126.wav,1 128 | section_00_0127.wav,0 129 | section_00_0128.wav,0 130 | section_00_0129.wav,0 131 | section_00_0130.wav,1 132 | section_00_0131.wav,1 133 | section_00_0132.wav,0 134 | section_00_0133.wav,1 135 | section_00_0134.wav,0 136 | section_00_0135.wav,1 137 | section_00_0136.wav,0 138 | section_00_0137.wav,0 139 | section_00_0138.wav,0 140 | section_00_0139.wav,0 141 | section_00_0140.wav,1 142 | section_00_0141.wav,0 143 | section_00_0142.wav,1 144 | section_00_0143.wav,1 145 | section_00_0144.wav,0 146 | section_00_0145.wav,1 147 | section_00_0146.wav,1 148 | section_00_0147.wav,1 149 | section_00_0148.wav,1 150 | section_00_0149.wav,0 151 | section_00_0150.wav,0 152 | section_00_0151.wav,0 153 | section_00_0152.wav,1 154 | section_00_0153.wav,0 155 | section_00_0154.wav,1 156 | section_00_0155.wav,1 157 | section_00_0156.wav,0 158 | section_00_0157.wav,0 159 | section_00_0158.wav,1 160 | section_00_0159.wav,1 161 | section_00_0160.wav,1 162 | section_00_0161.wav,0 163 | section_00_0162.wav,0 164 | section_00_0163.wav,1 165 | section_00_0164.wav,1 166 | section_00_0165.wav,1 167 | section_00_0166.wav,1 168 | section_00_0167.wav,1 169 | section_00_0168.wav,0 170 | section_00_0169.wav,1 171 | section_00_0170.wav,0 172 | section_00_0171.wav,0 173 | section_00_0172.wav,1 174 | section_00_0173.wav,1 175 | section_00_0174.wav,0 176 | section_00_0175.wav,0 177 | section_00_0176.wav,1 178 | section_00_0177.wav,1 179 | section_00_0178.wav,0 180 | section_00_0179.wav,1 181 | section_00_0180.wav,0 182 | section_00_0181.wav,1 183 | section_00_0182.wav,0 184 | section_00_0183.wav,0 185 | section_00_0184.wav,0 186 | section_00_0185.wav,1 187 | section_00_0186.wav,0 188 | section_00_0187.wav,1 189 | section_00_0188.wav,1 190 | section_00_0189.wav,0 191 | section_00_0190.wav,1 192 | section_00_0191.wav,1 193 | section_00_0192.wav,0 194 | section_00_0193.wav,0 195 | section_00_0194.wav,1 196 | section_00_0195.wav,0 197 | section_00_0196.wav,1 198 | section_00_0197.wav,1 199 | section_00_0198.wav,0 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_domain/ground_truth_bandsaw_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,1 2 | section_00_0001.wav,1 3 | section_00_0002.wav,1 4 | section_00_0003.wav,1 5 | section_00_0004.wav,1 6 | section_00_0005.wav,0 7 | section_00_0006.wav,1 8 | section_00_0007.wav,1 9 | section_00_0008.wav,1 10 | section_00_0009.wav,1 11 | section_00_0010.wav,0 12 | section_00_0011.wav,0 13 | section_00_0012.wav,0 14 | section_00_0013.wav,1 15 | section_00_0014.wav,0 16 | section_00_0015.wav,0 17 | section_00_0016.wav,0 18 | section_00_0017.wav,0 19 | section_00_0018.wav,1 20 | section_00_0019.wav,0 21 | section_00_0020.wav,1 22 | section_00_0021.wav,0 23 | section_00_0022.wav,1 24 | section_00_0023.wav,1 25 | section_00_0024.wav,1 26 | section_00_0025.wav,0 27 | section_00_0026.wav,0 28 | section_00_0027.wav,1 29 | section_00_0028.wav,1 30 | section_00_0029.wav,1 31 | section_00_0030.wav,1 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,0 35 | section_00_0034.wav,0 36 | section_00_0035.wav,0 37 | section_00_0036.wav,1 38 | section_00_0037.wav,0 39 | section_00_0038.wav,0 40 | section_00_0039.wav,0 41 | section_00_0040.wav,1 42 | section_00_0041.wav,1 43 | section_00_0042.wav,1 44 | section_00_0043.wav,0 45 | section_00_0044.wav,0 46 | section_00_0045.wav,0 47 | section_00_0046.wav,0 48 | section_00_0047.wav,0 49 | section_00_0048.wav,0 50 | section_00_0049.wav,1 51 | section_00_0050.wav,0 52 | section_00_0051.wav,1 53 | section_00_0052.wav,0 54 | section_00_0053.wav,1 55 | section_00_0054.wav,1 56 | section_00_0055.wav,0 57 | section_00_0056.wav,1 58 | section_00_0057.wav,1 59 | section_00_0058.wav,0 60 | section_00_0059.wav,1 61 | section_00_0060.wav,1 62 | section_00_0061.wav,0 63 | section_00_0062.wav,1 64 | section_00_0063.wav,0 65 | section_00_0064.wav,0 66 | section_00_0065.wav,0 67 | section_00_0066.wav,1 68 | section_00_0067.wav,0 69 | section_00_0068.wav,0 70 | section_00_0069.wav,1 71 | section_00_0070.wav,1 72 | section_00_0071.wav,0 73 | section_00_0072.wav,0 74 | section_00_0073.wav,0 75 | section_00_0074.wav,0 76 | section_00_0075.wav,0 77 | section_00_0076.wav,0 78 | section_00_0077.wav,0 79 | section_00_0078.wav,1 80 | section_00_0079.wav,0 81 | section_00_0080.wav,1 82 | section_00_0081.wav,1 83 | section_00_0082.wav,0 84 | section_00_0083.wav,0 85 | section_00_0084.wav,0 86 | section_00_0085.wav,0 87 | section_00_0086.wav,0 88 | section_00_0087.wav,1 89 | section_00_0088.wav,1 90 | section_00_0089.wav,0 91 | section_00_0090.wav,1 92 | section_00_0091.wav,1 93 | section_00_0092.wav,1 94 | section_00_0093.wav,0 95 | section_00_0094.wav,0 96 | section_00_0095.wav,0 97 | section_00_0096.wav,0 98 | section_00_0097.wav,1 99 | section_00_0098.wav,0 100 | section_00_0099.wav,0 101 | section_00_0100.wav,1 102 | section_00_0101.wav,0 103 | section_00_0102.wav,0 104 | section_00_0103.wav,1 105 | section_00_0104.wav,0 106 | section_00_0105.wav,1 107 | section_00_0106.wav,1 108 | section_00_0107.wav,0 109 | section_00_0108.wav,0 110 | section_00_0109.wav,0 111 | section_00_0110.wav,1 112 | section_00_0111.wav,1 113 | section_00_0112.wav,1 114 | section_00_0113.wav,1 115 | section_00_0114.wav,0 116 | section_00_0115.wav,1 117 | section_00_0116.wav,0 118 | section_00_0117.wav,0 119 | section_00_0118.wav,1 120 | section_00_0119.wav,1 121 | section_00_0120.wav,0 122 | section_00_0121.wav,1 123 | section_00_0122.wav,0 124 | section_00_0123.wav,1 125 | section_00_0124.wav,0 126 | section_00_0125.wav,1 127 | section_00_0126.wav,0 128 | section_00_0127.wav,1 129 | section_00_0128.wav,1 130 | section_00_0129.wav,1 131 | section_00_0130.wav,0 132 | section_00_0131.wav,1 133 | section_00_0132.wav,1 134 | section_00_0133.wav,1 135 | section_00_0134.wav,1 136 | section_00_0135.wav,0 137 | section_00_0136.wav,0 138 | section_00_0137.wav,0 139 | section_00_0138.wav,0 140 | section_00_0139.wav,0 141 | section_00_0140.wav,0 142 | section_00_0141.wav,1 143 | section_00_0142.wav,1 144 | section_00_0143.wav,0 145 | section_00_0144.wav,0 146 | section_00_0145.wav,1 147 | section_00_0146.wav,1 148 | section_00_0147.wav,0 149 | section_00_0148.wav,0 150 | section_00_0149.wav,0 151 | section_00_0150.wav,1 152 | section_00_0151.wav,0 153 | section_00_0152.wav,0 154 | section_00_0153.wav,0 155 | section_00_0154.wav,1 156 | section_00_0155.wav,1 157 | section_00_0156.wav,0 158 | section_00_0157.wav,1 159 | section_00_0158.wav,1 160 | section_00_0159.wav,0 161 | section_00_0160.wav,1 162 | section_00_0161.wav,1 163 | section_00_0162.wav,0 164 | section_00_0163.wav,0 165 | section_00_0164.wav,0 166 | section_00_0165.wav,0 167 | section_00_0166.wav,1 168 | section_00_0167.wav,0 169 | section_00_0168.wav,1 170 | section_00_0169.wav,0 171 | section_00_0170.wav,1 172 | section_00_0171.wav,0 173 | section_00_0172.wav,0 174 | section_00_0173.wav,1 175 | section_00_0174.wav,0 176 | section_00_0175.wav,0 177 | section_00_0176.wav,1 178 | section_00_0177.wav,1 179 | section_00_0178.wav,1 180 | section_00_0179.wav,0 181 | section_00_0180.wav,0 182 | section_00_0181.wav,0 183 | section_00_0182.wav,1 184 | section_00_0183.wav,1 185 | section_00_0184.wav,0 186 | section_00_0185.wav,1 187 | section_00_0186.wav,1 188 | section_00_0187.wav,1 189 | section_00_0188.wav,0 190 | section_00_0189.wav,0 191 | section_00_0190.wav,0 192 | section_00_0191.wav,1 193 | section_00_0192.wav,1 194 | section_00_0193.wav,0 195 | section_00_0194.wav,0 196 | section_00_0195.wav,1 197 | section_00_0196.wav,1 198 | section_00_0197.wav,0 199 | section_00_0198.wav,0 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_domain/ground_truth_grinder_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,0 2 | section_00_0001.wav,0 3 | section_00_0002.wav,0 4 | section_00_0003.wav,0 5 | section_00_0004.wav,1 6 | section_00_0005.wav,0 7 | section_00_0006.wav,1 8 | section_00_0007.wav,1 9 | section_00_0008.wav,0 10 | section_00_0009.wav,0 11 | section_00_0010.wav,0 12 | section_00_0011.wav,0 13 | section_00_0012.wav,1 14 | section_00_0013.wav,1 15 | section_00_0014.wav,0 16 | section_00_0015.wav,0 17 | section_00_0016.wav,0 18 | section_00_0017.wav,0 19 | section_00_0018.wav,1 20 | section_00_0019.wav,0 21 | section_00_0020.wav,0 22 | section_00_0021.wav,0 23 | section_00_0022.wav,0 24 | section_00_0023.wav,1 25 | section_00_0024.wav,1 26 | section_00_0025.wav,0 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,0 30 | section_00_0029.wav,0 31 | section_00_0030.wav,1 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,0 35 | section_00_0034.wav,0 36 | section_00_0035.wav,0 37 | section_00_0036.wav,0 38 | section_00_0037.wav,0 39 | section_00_0038.wav,0 40 | section_00_0039.wav,0 41 | section_00_0040.wav,0 42 | section_00_0041.wav,1 43 | section_00_0042.wav,1 44 | section_00_0043.wav,1 45 | section_00_0044.wav,0 46 | section_00_0045.wav,1 47 | section_00_0046.wav,1 48 | section_00_0047.wav,0 49 | section_00_0048.wav,0 50 | section_00_0049.wav,1 51 | section_00_0050.wav,0 52 | section_00_0051.wav,0 53 | section_00_0052.wav,0 54 | section_00_0053.wav,1 55 | section_00_0054.wav,0 56 | section_00_0055.wav,1 57 | section_00_0056.wav,0 58 | section_00_0057.wav,0 59 | section_00_0058.wav,0 60 | section_00_0059.wav,0 61 | section_00_0060.wav,1 62 | section_00_0061.wav,1 63 | section_00_0062.wav,0 64 | section_00_0063.wav,1 65 | section_00_0064.wav,1 66 | section_00_0065.wav,1 67 | section_00_0066.wav,0 68 | section_00_0067.wav,0 69 | section_00_0068.wav,0 70 | section_00_0069.wav,1 71 | section_00_0070.wav,0 72 | section_00_0071.wav,0 73 | section_00_0072.wav,1 74 | section_00_0073.wav,1 75 | section_00_0074.wav,0 76 | section_00_0075.wav,0 77 | section_00_0076.wav,1 78 | section_00_0077.wav,0 79 | section_00_0078.wav,1 80 | section_00_0079.wav,0 81 | section_00_0080.wav,0 82 | section_00_0081.wav,0 83 | section_00_0082.wav,1 84 | section_00_0083.wav,0 85 | section_00_0084.wav,0 86 | section_00_0085.wav,0 87 | section_00_0086.wav,0 88 | section_00_0087.wav,0 89 | section_00_0088.wav,0 90 | section_00_0089.wav,0 91 | section_00_0090.wav,1 92 | section_00_0091.wav,1 93 | section_00_0092.wav,0 94 | section_00_0093.wav,0 95 | section_00_0094.wav,0 96 | section_00_0095.wav,1 97 | section_00_0096.wav,0 98 | section_00_0097.wav,0 99 | section_00_0098.wav,0 100 | section_00_0099.wav,0 101 | section_00_0100.wav,1 102 | section_00_0101.wav,0 103 | section_00_0102.wav,1 104 | section_00_0103.wav,1 105 | section_00_0104.wav,0 106 | section_00_0105.wav,1 107 | section_00_0106.wav,1 108 | section_00_0107.wav,1 109 | section_00_0108.wav,0 110 | section_00_0109.wav,0 111 | section_00_0110.wav,0 112 | section_00_0111.wav,1 113 | section_00_0112.wav,0 114 | section_00_0113.wav,0 115 | section_00_0114.wav,0 116 | section_00_0115.wav,1 117 | section_00_0116.wav,1 118 | section_00_0117.wav,0 119 | section_00_0118.wav,1 120 | section_00_0119.wav,1 121 | section_00_0120.wav,0 122 | section_00_0121.wav,0 123 | section_00_0122.wav,0 124 | section_00_0123.wav,1 125 | section_00_0124.wav,0 126 | section_00_0125.wav,0 127 | section_00_0126.wav,0 128 | section_00_0127.wav,0 129 | section_00_0128.wav,0 130 | section_00_0129.wav,1 131 | section_00_0130.wav,0 132 | section_00_0131.wav,0 133 | section_00_0132.wav,0 134 | section_00_0133.wav,0 135 | section_00_0134.wav,0 136 | section_00_0135.wav,1 137 | section_00_0136.wav,0 138 | section_00_0137.wav,0 139 | section_00_0138.wav,1 140 | section_00_0139.wav,1 141 | section_00_0140.wav,1 142 | section_00_0141.wav,0 143 | section_00_0142.wav,0 144 | section_00_0143.wav,1 145 | section_00_0144.wav,0 146 | section_00_0145.wav,0 147 | section_00_0146.wav,0 148 | section_00_0147.wav,0 149 | section_00_0148.wav,0 150 | section_00_0149.wav,0 151 | section_00_0150.wav,0 152 | section_00_0151.wav,0 153 | section_00_0152.wav,1 154 | section_00_0153.wav,1 155 | section_00_0154.wav,0 156 | section_00_0155.wav,1 157 | section_00_0156.wav,1 158 | section_00_0157.wav,0 159 | section_00_0158.wav,1 160 | section_00_0159.wav,0 161 | section_00_0160.wav,0 162 | section_00_0161.wav,1 163 | section_00_0162.wav,0 164 | section_00_0163.wav,1 165 | section_00_0164.wav,1 166 | section_00_0165.wav,0 167 | section_00_0166.wav,0 168 | section_00_0167.wav,0 169 | section_00_0168.wav,0 170 | section_00_0169.wav,0 171 | section_00_0170.wav,0 172 | section_00_0171.wav,0 173 | section_00_0172.wav,0 174 | section_00_0173.wav,0 175 | section_00_0174.wav,0 176 | section_00_0175.wav,0 177 | section_00_0176.wav,1 178 | section_00_0177.wav,1 179 | section_00_0178.wav,1 180 | section_00_0179.wav,0 181 | section_00_0180.wav,0 182 | section_00_0181.wav,0 183 | section_00_0182.wav,0 184 | section_00_0183.wav,0 185 | section_00_0184.wav,0 186 | section_00_0185.wav,0 187 | section_00_0186.wav,0 188 | section_00_0187.wav,0 189 | section_00_0188.wav,1 190 | section_00_0189.wav,1 191 | section_00_0190.wav,0 192 | section_00_0191.wav,0 193 | section_00_0192.wav,1 194 | section_00_0193.wav,1 195 | section_00_0194.wav,1 196 | section_00_0195.wav,0 197 | section_00_0196.wav,0 198 | section_00_0197.wav,0 199 | section_00_0198.wav,0 200 | section_00_0199.wav,0 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_domain/ground_truth_shaker_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,0 2 | section_00_0001.wav,1 3 | section_00_0002.wav,0 4 | section_00_0003.wav,1 5 | section_00_0004.wav,0 6 | section_00_0005.wav,1 7 | section_00_0006.wav,0 8 | section_00_0007.wav,0 9 | section_00_0008.wav,0 10 | section_00_0009.wav,0 11 | section_00_0010.wav,0 12 | section_00_0011.wav,1 13 | section_00_0012.wav,1 14 | section_00_0013.wav,1 15 | section_00_0014.wav,1 16 | section_00_0015.wav,1 17 | section_00_0016.wav,1 18 | section_00_0017.wav,0 19 | section_00_0018.wav,1 20 | section_00_0019.wav,0 21 | section_00_0020.wav,1 22 | section_00_0021.wav,0 23 | section_00_0022.wav,1 24 | section_00_0023.wav,1 25 | section_00_0024.wav,1 26 | section_00_0025.wav,1 27 | section_00_0026.wav,0 28 | section_00_0027.wav,1 29 | section_00_0028.wav,1 30 | section_00_0029.wav,0 31 | section_00_0030.wav,1 32 | section_00_0031.wav,0 33 | section_00_0032.wav,0 34 | section_00_0033.wav,0 35 | section_00_0034.wav,0 36 | section_00_0035.wav,0 37 | section_00_0036.wav,0 38 | section_00_0037.wav,1 39 | section_00_0038.wav,0 40 | section_00_0039.wav,0 41 | section_00_0040.wav,0 42 | section_00_0041.wav,0 43 | section_00_0042.wav,0 44 | section_00_0043.wav,0 45 | section_00_0044.wav,0 46 | section_00_0045.wav,0 47 | section_00_0046.wav,0 48 | section_00_0047.wav,1 49 | section_00_0048.wav,1 50 | section_00_0049.wav,1 51 | section_00_0050.wav,1 52 | section_00_0051.wav,0 53 | section_00_0052.wav,0 54 | section_00_0053.wav,1 55 | section_00_0054.wav,1 56 | section_00_0055.wav,0 57 | section_00_0056.wav,0 58 | section_00_0057.wav,1 59 | section_00_0058.wav,1 60 | section_00_0059.wav,1 61 | section_00_0060.wav,0 62 | section_00_0061.wav,0 63 | section_00_0062.wav,0 64 | section_00_0063.wav,1 65 | section_00_0064.wav,1 66 | section_00_0065.wav,0 67 | section_00_0066.wav,0 68 | section_00_0067.wav,1 69 | section_00_0068.wav,0 70 | section_00_0069.wav,0 71 | section_00_0070.wav,1 72 | section_00_0071.wav,1 73 | section_00_0072.wav,1 74 | section_00_0073.wav,1 75 | section_00_0074.wav,0 76 | section_00_0075.wav,0 77 | section_00_0076.wav,1 78 | section_00_0077.wav,1 79 | section_00_0078.wav,1 80 | section_00_0079.wav,1 81 | section_00_0080.wav,1 82 | section_00_0081.wav,0 83 | section_00_0082.wav,0 84 | section_00_0083.wav,1 85 | section_00_0084.wav,0 86 | section_00_0085.wav,1 87 | section_00_0086.wav,0 88 | section_00_0087.wav,0 89 | section_00_0088.wav,1 90 | section_00_0089.wav,1 91 | section_00_0090.wav,0 92 | section_00_0091.wav,0 93 | section_00_0092.wav,0 94 | section_00_0093.wav,1 95 | section_00_0094.wav,1 96 | section_00_0095.wav,1 97 | section_00_0096.wav,1 98 | section_00_0097.wav,0 99 | section_00_0098.wav,0 100 | section_00_0099.wav,1 101 | section_00_0100.wav,0 102 | section_00_0101.wav,1 103 | section_00_0102.wav,1 104 | section_00_0103.wav,1 105 | section_00_0104.wav,1 106 | section_00_0105.wav,0 107 | section_00_0106.wav,0 108 | section_00_0107.wav,1 109 | section_00_0108.wav,1 110 | section_00_0109.wav,1 111 | section_00_0110.wav,0 112 | section_00_0111.wav,0 113 | section_00_0112.wav,0 114 | section_00_0113.wav,1 115 | section_00_0114.wav,1 116 | section_00_0115.wav,0 117 | section_00_0116.wav,1 118 | section_00_0117.wav,0 119 | section_00_0118.wav,0 120 | section_00_0119.wav,0 121 | section_00_0120.wav,0 122 | section_00_0121.wav,1 123 | section_00_0122.wav,1 124 | section_00_0123.wav,0 125 | section_00_0124.wav,1 126 | section_00_0125.wav,0 127 | section_00_0126.wav,1 128 | section_00_0127.wav,0 129 | section_00_0128.wav,0 130 | section_00_0129.wav,0 131 | section_00_0130.wav,0 132 | section_00_0131.wav,0 133 | section_00_0132.wav,0 134 | section_00_0133.wav,0 135 | section_00_0134.wav,0 136 | section_00_0135.wav,1 137 | section_00_0136.wav,0 138 | section_00_0137.wav,1 139 | section_00_0138.wav,1 140 | section_00_0139.wav,1 141 | section_00_0140.wav,0 142 | section_00_0141.wav,0 143 | section_00_0142.wav,0 144 | section_00_0143.wav,1 145 | section_00_0144.wav,1 146 | section_00_0145.wav,1 147 | section_00_0146.wav,1 148 | section_00_0147.wav,1 149 | section_00_0148.wav,1 150 | section_00_0149.wav,1 151 | section_00_0150.wav,1 152 | section_00_0151.wav,0 153 | section_00_0152.wav,1 154 | section_00_0153.wav,1 155 | section_00_0154.wav,0 156 | section_00_0155.wav,0 157 | section_00_0156.wav,1 158 | section_00_0157.wav,0 159 | section_00_0158.wav,1 160 | section_00_0159.wav,0 161 | section_00_0160.wav,1 162 | section_00_0161.wav,1 163 | section_00_0162.wav,0 164 | section_00_0163.wav,0 165 | section_00_0164.wav,0 166 | section_00_0165.wav,1 167 | section_00_0166.wav,1 168 | section_00_0167.wav,0 169 | section_00_0168.wav,1 170 | section_00_0169.wav,1 171 | section_00_0170.wav,1 172 | section_00_0171.wav,1 173 | section_00_0172.wav,0 174 | section_00_0173.wav,0 175 | section_00_0174.wav,0 176 | section_00_0175.wav,1 177 | section_00_0176.wav,0 178 | section_00_0177.wav,0 179 | section_00_0178.wav,1 180 | section_00_0179.wav,1 181 | section_00_0180.wav,0 182 | section_00_0181.wav,0 183 | section_00_0182.wav,1 184 | section_00_0183.wav,0 185 | section_00_0184.wav,1 186 | section_00_0185.wav,0 187 | section_00_0186.wav,0 188 | section_00_0187.wav,0 189 | section_00_0188.wav,0 190 | section_00_0189.wav,1 191 | section_00_0190.wav,1 192 | section_00_0191.wav,0 193 | section_00_0192.wav,0 194 | section_00_0193.wav,1 195 | section_00_0194.wav,0 196 | section_00_0195.wav,1 197 | section_00_0196.wav,1 198 | section_00_0197.wav,0 199 | section_00_0198.wav,0 200 | section_00_0199.wav,1 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_domain/ground_truth_ToyDrone_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,1 2 | section_00_0001.wav,1 3 | section_00_0002.wav,0 4 | section_00_0003.wav,1 5 | section_00_0004.wav,1 6 | section_00_0005.wav,0 7 | section_00_0006.wav,0 8 | section_00_0007.wav,1 9 | section_00_0008.wav,1 10 | section_00_0009.wav,0 11 | section_00_0010.wav,1 12 | section_00_0011.wav,0 13 | section_00_0012.wav,0 14 | section_00_0013.wav,1 15 | section_00_0014.wav,1 16 | section_00_0015.wav,0 17 | section_00_0016.wav,0 18 | section_00_0017.wav,1 19 | section_00_0018.wav,0 20 | section_00_0019.wav,0 21 | section_00_0020.wav,1 22 | section_00_0021.wav,1 23 | section_00_0022.wav,0 24 | section_00_0023.wav,1 25 | section_00_0024.wav,0 26 | section_00_0025.wav,1 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,1 30 | section_00_0029.wav,0 31 | section_00_0030.wav,0 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,0 35 | section_00_0034.wav,1 36 | section_00_0035.wav,0 37 | section_00_0036.wav,0 38 | section_00_0037.wav,1 39 | section_00_0038.wav,0 40 | section_00_0039.wav,1 41 | section_00_0040.wav,1 42 | section_00_0041.wav,0 43 | section_00_0042.wav,0 44 | section_00_0043.wav,1 45 | section_00_0044.wav,1 46 | section_00_0045.wav,0 47 | section_00_0046.wav,0 48 | section_00_0047.wav,1 49 | section_00_0048.wav,1 50 | section_00_0049.wav,1 51 | section_00_0050.wav,1 52 | section_00_0051.wav,0 53 | section_00_0052.wav,1 54 | section_00_0053.wav,1 55 | section_00_0054.wav,1 56 | section_00_0055.wav,0 57 | section_00_0056.wav,0 58 | section_00_0057.wav,1 59 | section_00_0058.wav,1 60 | section_00_0059.wav,1 61 | section_00_0060.wav,0 62 | section_00_0061.wav,0 63 | section_00_0062.wav,1 64 | section_00_0063.wav,1 65 | section_00_0064.wav,1 66 | section_00_0065.wav,0 67 | section_00_0066.wav,0 68 | section_00_0067.wav,0 69 | section_00_0068.wav,0 70 | section_00_0069.wav,0 71 | section_00_0070.wav,1 72 | section_00_0071.wav,1 73 | section_00_0072.wav,1 74 | section_00_0073.wav,0 75 | section_00_0074.wav,1 76 | section_00_0075.wav,0 77 | section_00_0076.wav,0 78 | section_00_0077.wav,0 79 | section_00_0078.wav,0 80 | section_00_0079.wav,0 81 | section_00_0080.wav,1 82 | section_00_0081.wav,1 83 | section_00_0082.wav,0 84 | section_00_0083.wav,1 85 | section_00_0084.wav,0 86 | section_00_0085.wav,0 87 | section_00_0086.wav,0 88 | section_00_0087.wav,1 89 | section_00_0088.wav,1 90 | section_00_0089.wav,1 91 | section_00_0090.wav,0 92 | section_00_0091.wav,1 93 | section_00_0092.wav,1 94 | section_00_0093.wav,0 95 | section_00_0094.wav,1 96 | section_00_0095.wav,0 97 | section_00_0096.wav,0 98 | section_00_0097.wav,0 99 | section_00_0098.wav,0 100 | section_00_0099.wav,1 101 | section_00_0100.wav,0 102 | section_00_0101.wav,0 103 | section_00_0102.wav,1 104 | section_00_0103.wav,1 105 | section_00_0104.wav,0 106 | section_00_0105.wav,1 107 | section_00_0106.wav,1 108 | section_00_0107.wav,1 109 | section_00_0108.wav,1 110 | section_00_0109.wav,1 111 | section_00_0110.wav,1 112 | section_00_0111.wav,0 113 | section_00_0112.wav,1 114 | section_00_0113.wav,0 115 | section_00_0114.wav,0 116 | section_00_0115.wav,0 117 | section_00_0116.wav,1 118 | section_00_0117.wav,0 119 | section_00_0118.wav,0 120 | section_00_0119.wav,0 121 | section_00_0120.wav,0 122 | section_00_0121.wav,1 123 | section_00_0122.wav,0 124 | section_00_0123.wav,1 125 | section_00_0124.wav,0 126 | section_00_0125.wav,0 127 | section_00_0126.wav,0 128 | section_00_0127.wav,0 129 | section_00_0128.wav,0 130 | section_00_0129.wav,1 131 | section_00_0130.wav,1 132 | section_00_0131.wav,1 133 | section_00_0132.wav,0 134 | section_00_0133.wav,1 135 | section_00_0134.wav,1 136 | section_00_0135.wav,0 137 | section_00_0136.wav,0 138 | section_00_0137.wav,0 139 | section_00_0138.wav,1 140 | section_00_0139.wav,0 141 | section_00_0140.wav,0 142 | section_00_0141.wav,0 143 | section_00_0142.wav,1 144 | section_00_0143.wav,0 145 | section_00_0144.wav,1 146 | section_00_0145.wav,1 147 | section_00_0146.wav,1 148 | section_00_0147.wav,1 149 | section_00_0148.wav,0 150 | section_00_0149.wav,1 151 | section_00_0150.wav,0 152 | section_00_0151.wav,1 153 | section_00_0152.wav,0 154 | section_00_0153.wav,1 155 | section_00_0154.wav,1 156 | section_00_0155.wav,1 157 | section_00_0156.wav,1 158 | section_00_0157.wav,1 159 | section_00_0158.wav,0 160 | section_00_0159.wav,0 161 | section_00_0160.wav,1 162 | section_00_0161.wav,0 163 | section_00_0162.wav,0 164 | section_00_0163.wav,1 165 | section_00_0164.wav,0 166 | section_00_0165.wav,0 167 | section_00_0166.wav,1 168 | section_00_0167.wav,1 169 | section_00_0168.wav,1 170 | section_00_0169.wav,0 171 | section_00_0170.wav,0 172 | section_00_0171.wav,1 173 | section_00_0172.wav,0 174 | section_00_0173.wav,0 175 | section_00_0174.wav,0 176 | section_00_0175.wav,1 177 | section_00_0176.wav,1 178 | section_00_0177.wav,0 179 | section_00_0178.wav,1 180 | section_00_0179.wav,1 181 | section_00_0180.wav,1 182 | section_00_0181.wav,1 183 | section_00_0182.wav,0 184 | section_00_0183.wav,1 185 | section_00_0184.wav,1 186 | section_00_0185.wav,1 187 | section_00_0186.wav,0 188 | section_00_0187.wav,0 189 | section_00_0188.wav,1 190 | section_00_0189.wav,0 191 | section_00_0190.wav,0 192 | section_00_0191.wav,0 193 | section_00_0192.wav,1 194 | section_00_0193.wav,1 195 | section_00_0194.wav,0 196 | section_00_0195.wav,0 197 | section_00_0196.wav,1 198 | section_00_0197.wav,0 199 | section_00_0198.wav,1 200 | section_00_0199.wav,1 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_domain/ground_truth_ToyNscale_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,0 2 | section_00_0001.wav,1 3 | section_00_0002.wav,1 4 | section_00_0003.wav,0 5 | section_00_0004.wav,0 6 | section_00_0005.wav,0 7 | section_00_0006.wav,1 8 | section_00_0007.wav,1 9 | section_00_0008.wav,1 10 | section_00_0009.wav,0 11 | section_00_0010.wav,1 12 | section_00_0011.wav,0 13 | section_00_0012.wav,0 14 | section_00_0013.wav,1 15 | section_00_0014.wav,0 16 | section_00_0015.wav,0 17 | section_00_0016.wav,1 18 | section_00_0017.wav,0 19 | section_00_0018.wav,0 20 | section_00_0019.wav,1 21 | section_00_0020.wav,1 22 | section_00_0021.wav,1 23 | section_00_0022.wav,0 24 | section_00_0023.wav,1 25 | section_00_0024.wav,0 26 | section_00_0025.wav,0 27 | section_00_0026.wav,0 28 | section_00_0027.wav,0 29 | section_00_0028.wav,0 30 | section_00_0029.wav,1 31 | section_00_0030.wav,0 32 | section_00_0031.wav,0 33 | section_00_0032.wav,1 34 | section_00_0033.wav,0 35 | section_00_0034.wav,1 36 | section_00_0035.wav,0 37 | section_00_0036.wav,1 38 | section_00_0037.wav,1 39 | section_00_0038.wav,1 40 | section_00_0039.wav,0 41 | section_00_0040.wav,1 42 | section_00_0041.wav,1 43 | section_00_0042.wav,0 44 | section_00_0043.wav,0 45 | section_00_0044.wav,0 46 | section_00_0045.wav,0 47 | section_00_0046.wav,0 48 | section_00_0047.wav,1 49 | section_00_0048.wav,0 50 | section_00_0049.wav,0 51 | section_00_0050.wav,1 52 | section_00_0051.wav,1 53 | section_00_0052.wav,1 54 | section_00_0053.wav,1 55 | section_00_0054.wav,0 56 | section_00_0055.wav,0 57 | section_00_0056.wav,0 58 | section_00_0057.wav,0 59 | section_00_0058.wav,1 60 | section_00_0059.wav,1 61 | section_00_0060.wav,1 62 | section_00_0061.wav,0 63 | section_00_0062.wav,0 64 | section_00_0063.wav,0 65 | section_00_0064.wav,0 66 | section_00_0065.wav,0 67 | section_00_0066.wav,1 68 | section_00_0067.wav,1 69 | section_00_0068.wav,1 70 | section_00_0069.wav,1 71 | section_00_0070.wav,0 72 | section_00_0071.wav,0 73 | section_00_0072.wav,0 74 | section_00_0073.wav,0 75 | section_00_0074.wav,1 76 | section_00_0075.wav,1 77 | section_00_0076.wav,0 78 | section_00_0077.wav,0 79 | section_00_0078.wav,0 80 | section_00_0079.wav,1 81 | section_00_0080.wav,0 82 | section_00_0081.wav,1 83 | section_00_0082.wav,0 84 | section_00_0083.wav,1 85 | section_00_0084.wav,0 86 | section_00_0085.wav,1 87 | section_00_0086.wav,0 88 | section_00_0087.wav,0 89 | section_00_0088.wav,0 90 | section_00_0089.wav,1 91 | section_00_0090.wav,0 92 | section_00_0091.wav,0 93 | section_00_0092.wav,1 94 | section_00_0093.wav,0 95 | section_00_0094.wav,0 96 | section_00_0095.wav,1 97 | section_00_0096.wav,1 98 | section_00_0097.wav,1 99 | section_00_0098.wav,1 100 | section_00_0099.wav,1 101 | section_00_0100.wav,0 102 | section_00_0101.wav,1 103 | section_00_0102.wav,0 104 | section_00_0103.wav,1 105 | section_00_0104.wav,0 106 | section_00_0105.wav,0 107 | section_00_0106.wav,0 108 | section_00_0107.wav,0 109 | section_00_0108.wav,0 110 | section_00_0109.wav,1 111 | section_00_0110.wav,0 112 | section_00_0111.wav,1 113 | section_00_0112.wav,1 114 | section_00_0113.wav,0 115 | section_00_0114.wav,0 116 | section_00_0115.wav,1 117 | section_00_0116.wav,0 118 | section_00_0117.wav,1 119 | section_00_0118.wav,0 120 | section_00_0119.wav,1 121 | section_00_0120.wav,1 122 | section_00_0121.wav,0 123 | section_00_0122.wav,1 124 | section_00_0123.wav,1 125 | section_00_0124.wav,0 126 | section_00_0125.wav,1 127 | section_00_0126.wav,0 128 | section_00_0127.wav,1 129 | section_00_0128.wav,1 130 | section_00_0129.wav,1 131 | section_00_0130.wav,1 132 | section_00_0131.wav,0 133 | section_00_0132.wav,0 134 | section_00_0133.wav,0 135 | section_00_0134.wav,0 136 | section_00_0135.wav,0 137 | section_00_0136.wav,0 138 | section_00_0137.wav,0 139 | section_00_0138.wav,1 140 | section_00_0139.wav,1 141 | section_00_0140.wav,0 142 | section_00_0141.wav,1 143 | section_00_0142.wav,0 144 | section_00_0143.wav,1 145 | section_00_0144.wav,1 146 | section_00_0145.wav,0 147 | section_00_0146.wav,1 148 | section_00_0147.wav,1 149 | section_00_0148.wav,1 150 | section_00_0149.wav,1 151 | section_00_0150.wav,1 152 | section_00_0151.wav,1 153 | section_00_0152.wav,0 154 | section_00_0153.wav,1 155 | section_00_0154.wav,0 156 | section_00_0155.wav,1 157 | section_00_0156.wav,1 158 | section_00_0157.wav,1 159 | section_00_0158.wav,1 160 | section_00_0159.wav,1 161 | section_00_0160.wav,1 162 | section_00_0161.wav,0 163 | section_00_0162.wav,1 164 | section_00_0163.wav,1 165 | section_00_0164.wav,1 166 | section_00_0165.wav,0 167 | section_00_0166.wav,1 168 | section_00_0167.wav,0 169 | section_00_0168.wav,0 170 | section_00_0169.wav,0 171 | section_00_0170.wav,1 172 | section_00_0171.wav,1 173 | section_00_0172.wav,1 174 | section_00_0173.wav,1 175 | section_00_0174.wav,1 176 | section_00_0175.wav,1 177 | section_00_0176.wav,0 178 | section_00_0177.wav,1 179 | section_00_0178.wav,0 180 | section_00_0179.wav,0 181 | section_00_0180.wav,0 182 | section_00_0181.wav,1 183 | section_00_0182.wav,1 184 | section_00_0183.wav,1 185 | section_00_0184.wav,1 186 | section_00_0185.wav,0 187 | section_00_0186.wav,0 188 | section_00_0187.wav,1 189 | section_00_0188.wav,1 190 | section_00_0189.wav,1 191 | section_00_0190.wav,0 192 | section_00_0191.wav,0 193 | section_00_0192.wav,1 194 | section_00_0193.wav,0 195 | section_00_0194.wav,1 196 | section_00_0195.wav,0 197 | section_00_0196.wav,0 198 | section_00_0197.wav,0 199 | section_00_0198.wav,0 200 | section_00_0199.wav,1 201 | -------------------------------------------------------------------------------- /src/datasets/prepare_dcase2023.py: -------------------------------------------------------------------------------- 1 | import os 2 | import re 3 | import glob 4 | import numpy as np 5 | import pandas as pd 6 | 7 | 8 | def get_filename_list(dir_path, pattern='*', ext='*'): 9 | """ 10 | find all extention files under directory 11 | :param dir_path: directory path 12 | :param ext: extention name, like wav, png... 13 | :param pattern: filename pattern for searching 14 | :return: files path list 15 | """ 16 | filename_list = [] 17 | for root, _, _ in os.walk(dir_path): 18 | file_path_pattern = os.path.join(root, f'{pattern}.{ext}') 19 | files = sorted(glob.glob(file_path_pattern)) 20 | filename_list += files 21 | return filename_list 22 | 23 | def get_meta_list(file_list): 24 | 25 | label_list = [] 26 | source_list = [] 27 | for file in file_list: 28 | label_list.append(0 if 'normal' in file else 1) 29 | source_list.append('source' if 'source' in file else 'target') 30 | return label_list, source_list 31 | 32 | def get_data_list(file_dirs): 33 | file_list = [] 34 | for fd in file_dirs: 35 | files = get_filename_list(fd, ext="wav") 36 | file_list.extend(files) 37 | # print(f'len of files: {fd.split("/")[-2]} | {len(files)}') 38 | label_list, source_list = get_meta_list(file_list) 39 | return file_list, label_list, source_list 40 | 41 | def get_test_data_list(test_dirs): 42 | root_path = "./" 43 | test_file_list = [] 44 | test_label_list = [] 45 | test_machine_list = [] 46 | test_source_list = [] 47 | for td in test_dirs: 48 | machine = td.split("/")[-2] 49 | test_files = get_filename_list(td, ext="wav") 50 | gt_list = np.array(pd.read_csv(root_path + 'dcase2023_task2_evaluator/ground_truth_data/ground_truth_' + machine + '_section_00_test.csv', header=None).iloc[:, 1]).tolist() 51 | s_list = np.array(pd.read_csv(root_path + 'dcase2023_task2_evaluator/ground_truth_domain/ground_truth_' + machine + '_section_00_test.csv', header=None).iloc[:, 1] == 0).tolist() 52 | s_list = ['source' if s else 'target' for s in s_list] 53 | test_file_list.extend(test_files) 54 | test_label_list.extend(gt_list) 55 | test_source_list.extend(s_list) 56 | test_machine_list.extend([machine] * len(test_files)) 57 | # print(f'len of files: {td.split("/")[-2]} | {len(test_files)}') 58 | test_attr_list = [0] * len(test_file_list) 59 | return test_file_list, test_label_list, test_source_list, np.array(test_machine_list), test_attr_list 60 | 61 | def get_attributes(file_list, source_list): 62 | machine_list = [] 63 | attrs = [] 64 | for idx, ext_id in enumerate(file_list): 65 | source = str(source_list[idx]) 66 | machine = ext_id.split('/')[-3] 67 | file = ext_id.split('/')[-1] 68 | section_id = re.findall('section_[0-9][0-9]', file)[0].split('_')[-1] 69 | # if idx == 0: 70 | # print(f"att id: {file.split('.wav')[0].split('_')[6:]}") 71 | att_id = '_'.join(file.split('.wav')[0].split('_')[6:]) 72 | machine_id = machine + '_' + section_id 73 | machine_list.append(machine) 74 | attrs.append("###".join([machine_id, att_id, source])) 75 | # attrs.append("###".join([machine_id, att_id])) 76 | return np.array(machine_list), np.array(attrs) 77 | 78 | def prepare_data(split_dirs): 79 | file_list, label_list, source_list = get_data_list(split_dirs) 80 | machine_names, file_attrs = get_attributes(file_list, source_list) 81 | return file_list, label_list, source_list, machine_names, file_attrs 82 | 83 | def get_dcase2023(train_split="train+add"): 84 | """ Args: 85 | dcase: can be dcase2020, 2021, 2022, and 2023 86 | train_split: train or train+add 87 | """ 88 | import yaml 89 | with open("./data_config.yaml", "r") as ymlfile: 90 | data_config = yaml.load(ymlfile, Loader=yaml.FullLoader) 91 | 92 | if train_split == "train+add": 93 | train_dirs = sorted(data_config["dcase2023_train_dirs"]) + sorted(data_config["dcase2023_add_dirs"]) 94 | train_file_list, train_label_list, train_source_list, train_machine_names, train_file_attrs = prepare_data(train_dirs) 95 | elif train_split == "train": 96 | train_dirs = data_config["dcase2023_train_dirs"] 97 | train_file_list, train_label_list, train_source_list, train_machine_names, train_file_attrs = prepare_data(train_dirs) 98 | elif train_split == "add": 99 | train_dirs = data_config["dcase2023_add_dirs"] 100 | train_file_list, train_label_list, train_source_list, train_machine_names, train_file_attrs = prepare_data(train_dirs) 101 | valid_dirs = sorted(data_config["dcase2023_valid_dirs"]) 102 | test_dirs = sorted(data_config["dcase2023_test_dirs"]) 103 | valid_file_list, valid_label_list, valid_source_list, valid_machine_names, valid_file_attrs = prepare_data(valid_dirs) 104 | test_file_list, test_label_list, test_source_list, test_machine_names, test_file_attrs = get_test_data_list(test_dirs) 105 | 106 | datasets = { 107 | "train": { 108 | "file_list": train_file_list, 109 | "label_list": train_label_list, 110 | "source_list": train_source_list, 111 | "machine_names": train_machine_names, 112 | "file_attrs": train_file_attrs 113 | }, 114 | "valid": { 115 | "file_list": valid_file_list, 116 | "label_list": valid_label_list, 117 | "source_list": valid_source_list, 118 | "machine_names": valid_machine_names, 119 | "file_attrs": valid_file_attrs 120 | }, 121 | "test": { 122 | "file_list": test_file_list, 123 | "label_list": test_label_list, 124 | "source_list": test_source_list, 125 | "machine_names": test_machine_names, 126 | "file_attrs": test_file_attrs 127 | } 128 | } 129 | return datasets -------------------------------------------------------------------------------- /beats/README.md: -------------------------------------------------------------------------------- 1 | # BEATs 2 | 3 | [**BEATs**](https://arxiv.org/abs/2212.09058): **Audio Pre-Training with Acoustic Tokenizers** 4 | 5 | Official PyTorch implementation and pretrained models of BEATs 6 | 7 | ## Pre-Trained and Fine-Tuned Tokenizers and Models 8 | Iterations | Tokenizer | Pre-Trained Model | AudioSet Fine-Tuned Model 1 | AudioSet Fine-Tuned Model 2 9 | |---|---|---|---|--- 10 | Iter1 | Random Projection | [BEATs_iter1](https://1drv.ms/u/s!AqeByhGUtINrgcpmY7IHhgc9q0pT7Q?e=uQuisJ) | [Fine-tuned BEATs_iter1 (cpt1)](https://1drv.ms/u/s!AqeByhGUtINrgcpuRfRZmco2XulmFw?e=f2INHa) | [Fine-tuned BEATs_iter1 (cpt2)](https://1drv.ms/u/s!AqeByhGUtINrgcpyMlTmnRh0Wp_Qgg?e=sgzv8H) | 11 | Iter2 | [Tokenizer_iter2](https://1drv.ms/u/s!AqeByhGUtINrgcpnFGsfd_buKng5Pw?e=avWBJw)| [BEATs_iter2](https://1drv.ms/u/s!AqeByhGUtINrgcpwwEGgUyiI-jQyQw?e=1rP1RI) | [Fine-tuned BEATs_iter2 (cpt1)](https://1drv.ms/u/s!AqeByhGUtINrgcp4l547zKa7xPqy8w?e=rsLdPr) | [Fine-tuned BEATs_iter2 (cpt2)](https://1drv.ms/u/s!AqeByhGUtINrgcp5APbt_2bdIQvX0w?e=2cd2ry) | 12 | Iter3 | [Tokenizer_iter3](https://1drv.ms/u/s!AqeByhGUtINrgcp1DEzUBtzHapxcqw?e=JZI5Uf)| [BEATs_iter3](https://1drv.ms/u/s!AqeByhGUtINrgcpxJUNDxg4eU0r-vA?e=qezPJ5) | [Fine-tuned BEATs_iter3 (cpt1)](https://1drv.ms/u/s!AqeByhGUtINrgcplb48ll1zIt82eWQ?e=XyxrX7) | [Fine-tuned BEATs_iter3 (cpt2)](https://1drv.ms/u/s!AqeByhGUtINrgcptb4S-CeJnlJGtZA?e=2FyDy3) | 13 | Iter3+ | [Tokenizer_iter3+ (AS20K)](https://1drv.ms/u/s!AqeByhGUtINrgcpz_SnXxs0SrwHEwA?e=14nugm)| [BEATs_iter3+ (AS20K)](https://1drv.ms/u/s!AqeByhGUtINrgcpvdNz8-aYim60CIg?e=53V8pg) | [Fine-tuned BEATs_iter3+ (AS20K) (cpt1)](https://1drv.ms/u/s!AqeByhGUtINrgcp2YHUCT1uZx2Kysw?e=nvu1Dw) | [Fine-tuned BEATs_iter3+ (AS20K) (cpt2)](https://1drv.ms/u/s!AqeByhGUtINrgcp092af0h7P3kXKFA?e=kUkPhN) | 14 | Iter3+ | [Tokenizer_iter3+ (AS2M)](https://1drv.ms/u/s!AqeByhGUtINrgcppJUDx2TmXiIMFyQ?e=pJsOLl)| [BEATs_iter3+ (AS2M)](https://1drv.ms/u/s!AqeByhGUtINrgcpke6_lRSZEKD5j2Q?e=A3FpOf) | [Fine-tuned BEATs_iter3+ (AS2M) (cpt1)](https://1drv.ms/u/s!AqeByhGUtINrgcpoZecQbiXeaUjN8A?e=DasbeC) | [Fine-tuned BEATs_iter3+ (AS2M) (cpt2)](https://1drv.ms/u/s!AqeByhGUtINrgcpj8ujXH1YUtxooEg?e=E9Ncea) | 15 | 16 | 17 | ### Load Tokenizers 18 | 19 | ```python 20 | import torch 21 | from Tokenizers import TokenizersConfig, Tokenizers 22 | 23 | # load the pre-trained checkpoints 24 | checkpoint = torch.load('/path/to/tokenizer.pt') 25 | 26 | cfg = TokenizersConfig(checkpoint['cfg']) 27 | BEATs_tokenizer = Tokenizers(cfg) 28 | BEATs_tokenizer.load_state_dict(checkpoint['model']) 29 | BEATs_tokenizer.eval() 30 | 31 | # tokenize the audio and generate the labels 32 | audio_input_16khz = torch.randn(1, 10000) 33 | padding_mask = torch.zeros(1, 10000).bool() 34 | 35 | labels = BEATs_tokenizer.extract_labels(audio_input_16khz, padding_mask=padding_mask) 36 | ``` 37 | 38 | 39 | ### Load Pre-Trained Models 40 | 41 | ```python 42 | import torch 43 | from BEATs import BEATs, BEATsConfig 44 | 45 | # load the pre-trained checkpoints 46 | checkpoint = torch.load('/path/to/model.pt') 47 | 48 | cfg = BEATsConfig(checkpoint['cfg']) 49 | BEATs_model = BEATs(cfg) 50 | BEATs_model.load_state_dict(checkpoint['model']) 51 | BEATs_model.eval() 52 | 53 | # extract the the audio representation 54 | audio_input_16khz = torch.randn(1, 10000) 55 | padding_mask = torch.zeros(1, 10000).bool() 56 | 57 | representation = BEATs_model.extract_features(audio_input_16khz, padding_mask=padding_mask)[0] 58 | ``` 59 | 60 | 61 | ### Load Fine-tuned Models 62 | 63 | ```python 64 | import torch 65 | from BEATs import BEATs, BEATsConfig 66 | 67 | # load the fine-tuned checkpoints 68 | checkpoint = torch.load('/path/to/model.pt') 69 | 70 | cfg = BEATsConfig(checkpoint['cfg']) 71 | BEATs_model = BEATs(cfg) 72 | BEATs_model.load_state_dict(checkpoint['model']) 73 | BEATs_model.eval() 74 | 75 | # predict the classification probability of each class 76 | audio_input_16khz = torch.randn(3, 10000) 77 | padding_mask = torch.zeros(3, 10000).bool() 78 | 79 | probs = BEATs_model.extract_features(audio_input_16khz, padding_mask=padding_mask)[0] 80 | 81 | for i, (top5_label_prob, top5_label_idx) in enumerate(zip(*probs.topk(k=5))): 82 | top5_label = [checkpoint['label_dict'][label_idx.item()] for label_idx in top5_label_idx] 83 | print(f'Top 5 predicted labels of the {i}th audio are {top5_label} with probability of {top5_label_prob}') 84 | ``` 85 | 86 | ## Evaluation Results 87 | 88 | ### Comparing with the SOTA Single Models 89 | ![alt text](Evaluation_Results/Comparing_with_the_SOTA_Single_Models.png) 90 | 91 | 92 | ### Comparing with the SOTA Ensemble Models 93 | ![alt text](Evaluation_Results/Comparing_with_the_SOTA_Ensemble_Models.png) 94 | 95 | 96 | ### Comparing Different BEATS Tokenizers 97 | ![alt text](Evaluation_Results/Comparing_Different_BEATS_Tokenizers.png) 98 | 99 | 100 | ### Comparing Different Pre-Training Targets 101 | ![alt text](Evaluation_Results/Comparing_Different_Pre-Training_Targets.png) 102 | 103 | 104 | ## License 105 | This project is licensed under the license found in the LICENSE file in the root directory of this source tree. 106 | Portions of the source code are based on the [FAIRSEQ](https://github.com/pytorch/fairseq) and [VQGAN](https://github.com/CompVis/taming-transformers) project. 107 | 108 | [Microsoft Open Source Code of Conduct](https://opensource.microsoft.com/codeofconduct) 109 | 110 | 111 | ### Reference 112 | If you find our work is useful in your research, please cite the following paper: 113 | ``` latex 114 | @article{Chen2022beats, 115 | title = {BEATs: Audio Pre-Training with Acoustic Tokenizers}, 116 | author = {Sanyuan Chen and Yu Wu and Chengyi Wang and Shujie Liu and Daniel Tompkins and Zhuo Chen and Furu Wei}, 117 | eprint={2212.09058}, 118 | archivePrefix={arXiv}, 119 | year={2022} 120 | } 121 | ``` 122 | ### Contact Information 123 | 124 | For help or issues using BEATs models, please submit a GitHub issue. 125 | 126 | For other communications related to BEATs, please contact Yu Wu (`yuwu1@microsoft.com`). -------------------------------------------------------------------------------- /src/datasets/audio_dataset.py: -------------------------------------------------------------------------------- 1 | import numpy as np 2 | from tqdm import tqdm 3 | import torch 4 | import torchaudio 5 | from torch.utils.data import Dataset 6 | 7 | class AudioDataset(Dataset): 8 | def __init__( 9 | self, 10 | file_list, 11 | label_list, 12 | machine_list, 13 | source_list=None, 14 | class_ids=None, 15 | load_in_memory=False, 16 | machine_name=None, 17 | audio_length=None, 18 | extracted_features=None, 19 | domain=None 20 | ): 21 | self.machine_name = machine_name 22 | self.extracted_features = None 23 | self.domain = domain 24 | # Filter by machine name 25 | if self.machine_name is not None: 26 | self.file_list = [file for file, machine in zip(file_list, machine_list) if machine == self.machine_name] 27 | self.label_list = [label for label, machine in zip(label_list, machine_list) if machine == self.machine_name] 28 | self.machine_list = [machine for machine in machine_list if machine == self.machine_name] 29 | 30 | if source_list is not None: 31 | self.source_list = [source for source, machine in zip(source_list, machine_list) if machine == self.machine_name] 32 | 33 | if class_ids is not None: 34 | self.class_ids = [class_id for class_id, machine in zip(class_ids, machine_list) if machine == self.machine_name] 35 | 36 | if extracted_features is not None: 37 | self.extracted_features = [feature for feature, machine in zip(extracted_features, machine_list) if machine == self.machine_name] 38 | else: 39 | self.file_list = file_list 40 | self.label_list = label_list 41 | self.machine_list = machine_list 42 | 43 | if source_list is not None: 44 | self.source_list = source_list 45 | 46 | if class_ids is not None: 47 | self.class_ids = class_ids 48 | 49 | if extracted_features is not None: 50 | self.extracted_features = extracted_features 51 | 52 | if source_list is not None: 53 | if domain == "source": 54 | self.file_list = [file for file, source in zip(self.file_list, self.source_list) if source == "source"] 55 | self.label_list = [label for label, source in zip(self.label_list, self.source_list) if source == "source"] 56 | self.machine_list = [machine for machine, source in zip(self.machine_list, self.source_list) if source == "source"] 57 | self.source_list = [source for source in self.source_list if source == "source"] 58 | if class_ids is not None: 59 | self.class_ids = [class_id for class_id, source in zip(class_ids, self.source_list) if source == "source"] 60 | if extracted_features is not None: 61 | self.extracted_features = [feature for feature, source in zip(self.extracted_features, self.source_list) if source == "source"] 62 | 63 | elif domain == "target": 64 | self.file_list = [file for file, source in zip(self.file_list, self.source_list) if source == "target"] 65 | self.label_list = [label for label, source in zip(self.label_list, self.source_list) if source == "target"] 66 | self.machine_list = [machine for machine, source in zip(self.machine_list, self.source_list) if source == "target"] 67 | self.source_list = [source for source in self.source_list if source == "target"] 68 | if class_ids is not None: 69 | self.class_ids = [class_id for class_id, source in zip(class_ids, self.source_list) if source == "target"] 70 | if extracted_features is not None: 71 | self.extracted_features = [feature for feature, source in zip(self.extracted_features, self.source_list) if source == "target"] 72 | 73 | self.load_in_memory = load_in_memory 74 | self.audio_length = audio_length 75 | 76 | if self.extracted_features is not None: 77 | self.data_list = [self.get_extracted_data(feature, label, class_id) for feature, label, class_id in 78 | tqdm(zip(self.extracted_features, self.label_list, self.class_ids))] if load_in_memory else [] 79 | else: 80 | self.data_list = [self.get_data(file, label) for file, label in 81 | zip(self.file_list, self.label_list)] if load_in_memory else [] 82 | 83 | print(f"Number of samples: {len(self)}") 84 | 85 | def __len__(self): 86 | return len(self.file_list) 87 | 88 | def __getitem__(self, item): 89 | if self.extracted_features is not None: 90 | data_item = self.data_list[item] if self.load_in_memory else self.get_extracted_data( 91 | self.extracted_features[item], self.label_list[item], self.class_ids[item] 92 | ) 93 | 94 | else: 95 | data_item = self.data_list[item] if self.load_in_memory else self.get_data( 96 | self.file_list[item], self.label_list[item] 97 | ) 98 | return data_item 99 | 100 | def get_data(self, filename, label): 101 | wave, sr = torchaudio.load(filename) 102 | if sr != 16000: 103 | raise ValueError(f"Sample rate is not 16kHz: {sr}") 104 | 105 | if self.audio_length is not None: 106 | wave = self.adjust_size(wave.squeeze(), self.audio_length) 107 | 108 | return { 109 | "input": wave, # some are not 10 secs 110 | "label": torch.Tensor([label]).long(), 111 | } 112 | 113 | def get_extracted_data(self, feature, label, class_id): 114 | return { 115 | "input": feature, # some are not 10 secs 116 | "label": torch.Tensor([label]).long(), 117 | # "class_id": torch.Tensor([class_id]).long(), 118 | "class_id": class_id, 119 | } 120 | 121 | def adjust_size(self, wave, new_size): 122 | audio_length = new_size 123 | if wave.shape[0] < audio_length: 124 | wave = np.pad(wave, (0, audio_length-wave.shape[0]), 'constant') 125 | else: 126 | wave = wave[:audio_length] 127 | return wave -------------------------------------------------------------------------------- /beats/Tokenizers.py: -------------------------------------------------------------------------------- 1 | # -------------------------------------------------------- 2 | # BEATs: Audio Pre-Training with Acoustic Tokenizers (https://arxiv.org/abs/2212.09058) 3 | # Github source: https://github.com/microsoft/unilm/tree/master/beats 4 | # Copyright (c) 2022 Microsoft 5 | # Licensed under The MIT License [see LICENSE for details] 6 | # Based on fairseq code bases 7 | # https://github.com/pytorch/fairseq 8 | # -------------------------------------------------------- 9 | 10 | 11 | import torch 12 | import torch.nn as nn 13 | from torch.nn import LayerNorm 14 | import torchaudio.compliance.kaldi as ta_kaldi 15 | 16 | from beats.backbone import ( 17 | TransformerEncoder, 18 | ) 19 | from beats.quantizer import ( 20 | NormEMAVectorQuantizer, 21 | ) 22 | 23 | import logging 24 | from typing import Optional 25 | 26 | logger = logging.getLogger(__name__) 27 | 28 | 29 | class TokenizersConfig: 30 | def __init__(self, cfg=None): 31 | self.input_patch_size: int = -1 # path size of patch embedding 32 | self.embed_dim: int = 512 # patch embedding dimension 33 | self.conv_bias: bool = False # include bias in conv encoder 34 | 35 | self.encoder_layers: int = 12 # num encoder layers in the transformer 36 | self.encoder_embed_dim: int = 768 # encoder embedding dimension 37 | self.encoder_ffn_embed_dim: int = 3072 # encoder embedding dimension for FFN 38 | self.encoder_attention_heads: int = 12 # num encoder attention heads 39 | self.activation_fn: str = "gelu" # activation function to use 40 | 41 | self.layer_norm_first: bool = False # apply layernorm first in the transformer 42 | self.deep_norm: bool = False # apply deep_norm first in the transformer 43 | 44 | # dropouts 45 | self.dropout: float = 0.1 # dropout probability for the transformer 46 | self.attention_dropout: float = 0.1 # dropout probability for attention weights 47 | self.activation_dropout: float = 0.0 # dropout probability after activation in FFN 48 | self.encoder_layerdrop: float = 0.0 # probability of dropping a tarnsformer layer 49 | self.dropout_input: float = 0.0 # dropout to apply to the input (after feat extr) 50 | 51 | # positional embeddings 52 | self.conv_pos: int = 128 # number of filters for convolutional positional embeddings 53 | self.conv_pos_groups: int = 16 # number of groups for convolutional positional embedding 54 | 55 | # relative position embedding 56 | self.relative_position_embedding: bool = False # apply relative position embedding 57 | self.num_buckets: int = 320 # number of buckets for relative position embedding 58 | self.max_distance: int = 1280 # maximum distance for relative position embedding 59 | self.gru_rel_pos: bool = False # apply gated relative position embedding 60 | 61 | # quantizer 62 | self.quant_n: int = 1024 # codebook number in quantizer 63 | self.quant_dim: int = 256 # codebook dimension in quantizer 64 | 65 | if cfg is not None: 66 | self.update(cfg) 67 | 68 | def update(self, cfg: dict): 69 | self.__dict__.update(cfg) 70 | 71 | 72 | class Tokenizers(nn.Module): 73 | def __init__( 74 | self, 75 | cfg: TokenizersConfig, 76 | ) -> None: 77 | super().__init__() 78 | logger.info(f"Tokenizers Config: {cfg.__dict__}") 79 | 80 | self.cfg = cfg 81 | 82 | self.embed = cfg.embed_dim 83 | self.post_extract_proj = ( 84 | nn.Linear(self.embed, cfg.encoder_embed_dim) 85 | if self.embed != cfg.encoder_embed_dim 86 | else None 87 | ) 88 | 89 | self.input_patch_size = cfg.input_patch_size 90 | self.patch_embedding = nn.Conv2d(1, self.embed, kernel_size=self.input_patch_size, stride=self.input_patch_size, 91 | bias=cfg.conv_bias) 92 | 93 | self.dropout_input = nn.Dropout(cfg.dropout_input) 94 | 95 | assert not cfg.deep_norm or not cfg.layer_norm_first 96 | self.encoder = TransformerEncoder(cfg) 97 | self.layer_norm = LayerNorm(self.embed) 98 | 99 | self.quantize = NormEMAVectorQuantizer( 100 | n_embed=cfg.quant_n, embedding_dim=cfg.quant_dim, beta=1.0, kmeans_init=True, decay=0.99, 101 | ) 102 | self.quant_n = cfg.quant_n 103 | self.quantize_layer = nn.Sequential( 104 | nn.Linear(cfg.encoder_embed_dim, cfg.encoder_embed_dim), 105 | nn.Tanh(), 106 | nn.Linear(cfg.encoder_embed_dim, cfg.quant_dim) # for quantize 107 | ) 108 | 109 | def forward_padding_mask( 110 | self, 111 | features: torch.Tensor, 112 | padding_mask: torch.Tensor, 113 | ) -> torch.Tensor: 114 | extra = padding_mask.size(1) % features.size(1) 115 | if extra > 0: 116 | padding_mask = padding_mask[:, :-extra] 117 | padding_mask = padding_mask.view( 118 | padding_mask.size(0), features.size(1), -1 119 | ) 120 | padding_mask = padding_mask.all(-1) 121 | return padding_mask 122 | 123 | def preprocess( 124 | self, 125 | source: torch.Tensor, 126 | fbank_mean: float = 15.41663, 127 | fbank_std: float = 6.55582, 128 | ) -> torch.Tensor: 129 | fbanks = [] 130 | for waveform in source: 131 | waveform = waveform.unsqueeze(0) * 2 ** 15 132 | fbank = ta_kaldi.fbank(waveform, num_mel_bins=128, sample_frequency=16000, frame_length=25, frame_shift=10) 133 | fbanks.append(fbank) 134 | fbank = torch.stack(fbanks, dim=0) 135 | fbank = (fbank - fbank_mean) / (2 * fbank_std) 136 | return fbank 137 | 138 | def extract_labels( 139 | self, 140 | source: torch.Tensor, 141 | padding_mask: Optional[torch.Tensor] = None, 142 | fbank_mean: float = 15.41663, 143 | fbank_std: float = 6.55582, 144 | ): 145 | fbank = self.preprocess(source, fbank_mean=fbank_mean, fbank_std=fbank_std) 146 | 147 | if padding_mask is not None: 148 | padding_mask = self.forward_padding_mask(fbank, padding_mask) 149 | 150 | fbank = fbank.unsqueeze(1) 151 | features = self.patch_embedding(fbank) 152 | features = features.reshape(features.shape[0], features.shape[1], -1) 153 | features = features.transpose(1, 2) 154 | features = self.layer_norm(features) 155 | 156 | if padding_mask is not None: 157 | padding_mask = self.forward_padding_mask(features, padding_mask) 158 | 159 | if self.post_extract_proj is not None: 160 | features = self.post_extract_proj(features) 161 | 162 | x = self.dropout_input(features) 163 | 164 | x, layer_results = self.encoder( 165 | x, 166 | padding_mask=padding_mask, 167 | ) 168 | 169 | quantize_input = self.quantize_layer(x) 170 | quantize_feature, embed_loss, embed_ind = self.quantize(quantize_input) 171 | 172 | return x, quantize_input, quantize_feature, embed_loss, embed_ind -------------------------------------------------------------------------------- /beats/BEATs.py: -------------------------------------------------------------------------------- 1 | # -------------------------------------------------------- 2 | # BEATs: Audio Pre-Training with Acoustic Tokenizers (https://arxiv.org/abs/2212.09058) 3 | # Github source: https://github.com/microsoft/unilm/tree/master/beats 4 | # Copyright (c) 2022 Microsoft 5 | # Licensed under The MIT License [see LICENSE for details] 6 | # Based on fairseq code bases 7 | # https://github.com/pytorch/fairseq 8 | # -------------------------------------------------------- 9 | 10 | 11 | import torch 12 | import torch.nn as nn 13 | from torch.nn import LayerNorm 14 | import torchaudio.compliance.kaldi as ta_kaldi 15 | 16 | from beats.backbone import ( 17 | TransformerEncoder, 18 | ) 19 | 20 | import logging 21 | from typing import Optional 22 | 23 | logger = logging.getLogger(__name__) 24 | 25 | 26 | class BEATsConfig: 27 | def __init__(self, cfg=None): 28 | self.input_patch_size: int = -1 # path size of patch embedding 29 | self.embed_dim: int = 512 # patch embedding dimension 30 | self.conv_bias: bool = False # include bias in conv encoder 31 | 32 | self.encoder_layers: int = 12 # num encoder layers in the transformer 33 | self.encoder_embed_dim: int = 768 # encoder embedding dimension 34 | self.encoder_ffn_embed_dim: int = 3072 # encoder embedding dimension for FFN 35 | self.encoder_attention_heads: int = 12 # num encoder attention heads 36 | self.activation_fn: str = "gelu" # activation function to use 37 | 38 | self.layer_wise_gradient_decay_ratio: float = 1.0 # ratio for layer-wise gradient decay 39 | self.layer_norm_first: bool = False # apply layernorm first in the transformer 40 | self.deep_norm: bool = False # apply deep_norm first in the transformer 41 | 42 | # dropouts 43 | self.dropout: float = 0.1 # dropout probability for the transformer 44 | self.attention_dropout: float = 0.1 # dropout probability for attention weights 45 | self.activation_dropout: float = 0.0 # dropout probability after activation in FFN 46 | self.encoder_layerdrop: float = 0.0 # probability of dropping a tarnsformer layer 47 | self.dropout_input: float = 0.0 # dropout to apply to the input (after feat extr) 48 | 49 | # positional embeddings 50 | self.conv_pos: int = 128 # number of filters for convolutional positional embeddings 51 | self.conv_pos_groups: int = 16 # number of groups for convolutional positional embedding 52 | 53 | # relative position embedding 54 | self.relative_position_embedding: bool = False # apply relative position embedding 55 | self.num_buckets: int = 320 # number of buckets for relative position embedding 56 | self.max_distance: int = 1280 # maximum distance for relative position embedding 57 | self.gru_rel_pos: bool = False # apply gated relative position embedding 58 | 59 | # label predictor 60 | self.finetuned_model: bool = False # whether the model is a fine-tuned model. 61 | self.predictor_dropout: float = 0.1 # dropout probability for the predictor 62 | self.predictor_class: int = 527 # target class number for the predictor 63 | 64 | if cfg is not None: 65 | self.update(cfg) 66 | 67 | def update(self, cfg: dict): 68 | self.__dict__.update(cfg) 69 | 70 | 71 | class BEATs(nn.Module): 72 | def __init__( 73 | self, 74 | cfg: BEATsConfig, 75 | ) -> None: 76 | super().__init__() 77 | logger.info(f"BEATs Config: {cfg.__dict__}") 78 | 79 | self.cfg = cfg 80 | 81 | self.embed = cfg.embed_dim 82 | self.post_extract_proj = ( 83 | nn.Linear(self.embed, cfg.encoder_embed_dim) 84 | if self.embed != cfg.encoder_embed_dim 85 | else None 86 | ) 87 | 88 | self.input_patch_size = cfg.input_patch_size 89 | self.patch_embedding = nn.Conv2d(1, self.embed, kernel_size=self.input_patch_size, stride=self.input_patch_size, 90 | bias=cfg.conv_bias) 91 | 92 | self.dropout_input = nn.Dropout(cfg.dropout_input) 93 | 94 | assert not cfg.deep_norm or not cfg.layer_norm_first 95 | self.encoder = TransformerEncoder(cfg) 96 | self.layer_norm = LayerNorm(self.embed) 97 | 98 | if cfg.finetuned_model: 99 | self.predictor_dropout = nn.Dropout(cfg.predictor_dropout) 100 | self.predictor = nn.Linear(cfg.encoder_embed_dim, cfg.predictor_class) 101 | else: 102 | self.predictor = None 103 | 104 | def forward_padding_mask( 105 | self, 106 | features: torch.Tensor, 107 | padding_mask: torch.Tensor, 108 | ) -> torch.Tensor: 109 | extra = padding_mask.size(1) % features.size(1) 110 | if extra > 0: 111 | padding_mask = padding_mask[:, :-extra] 112 | padding_mask = padding_mask.view( 113 | padding_mask.size(0), features.size(1), -1 114 | ) 115 | padding_mask = padding_mask.all(-1) 116 | return padding_mask 117 | 118 | def preprocess( 119 | self, 120 | source: torch.Tensor, 121 | fbank_mean: float = 15.41663, 122 | fbank_std: float = 6.55582, 123 | ) -> torch.Tensor: 124 | fbanks = [] 125 | for waveform in source: 126 | waveform = waveform.unsqueeze(0) * 2 ** 15 127 | fbank = ta_kaldi.fbank(waveform, num_mel_bins=128, sample_frequency=16000, frame_length=25, frame_shift=10) 128 | fbanks.append(fbank) 129 | fbank = torch.stack(fbanks, dim=0) 130 | fbank = (fbank - fbank_mean) / (2 * fbank_std) 131 | return fbank 132 | 133 | def extract_features( 134 | self, 135 | source: torch.Tensor, 136 | padding_mask: Optional[torch.Tensor] = None, 137 | fbank_mean: float = 15.41663, 138 | fbank_std: float = 6.55582, 139 | need_weights: bool = False, 140 | layer: Optional[int] = None, 141 | ): 142 | fbank = self.preprocess(source, fbank_mean=fbank_mean, fbank_std=fbank_std) 143 | 144 | if padding_mask is not None: 145 | padding_mask = self.forward_padding_mask(fbank, padding_mask) 146 | 147 | fbank = fbank.unsqueeze(1) 148 | features = self.patch_embedding(fbank) 149 | features = features.reshape(features.shape[0], features.shape[1], -1) 150 | features = features.transpose(1, 2) 151 | features = self.layer_norm(features) 152 | 153 | if padding_mask is not None: 154 | padding_mask = self.forward_padding_mask(features, padding_mask) 155 | 156 | if self.post_extract_proj is not None: 157 | features = self.post_extract_proj(features) 158 | 159 | x = self.dropout_input(features) 160 | 161 | x, layer_results = self.encoder( 162 | x, 163 | padding_mask=padding_mask, 164 | need_weights=need_weights, 165 | layer=layer 166 | ) 167 | 168 | if self.predictor is not None: 169 | x = self.predictor_dropout(x) 170 | logits = self.predictor(x) 171 | 172 | if padding_mask is not None and padding_mask.any(): 173 | logits[padding_mask] = 0 174 | logits = logits.sum(dim=1) 175 | logits = logits / (~padding_mask).sum(dim=1).unsqueeze(-1).expand_as(logits) 176 | else: 177 | logits = logits.mean(dim=1) 178 | 179 | lprobs = torch.sigmoid(logits) 180 | 181 | # return lprobs, padding_mask # <- default 182 | return x, padding_mask, layer_results 183 | else: 184 | return x, padding_mask, layer_results -------------------------------------------------------------------------------- /beats/modules.py: -------------------------------------------------------------------------------- 1 | # -------------------------------------------------------- 2 | # BEATs: Audio Pre-Training with Acoustic Tokenizers (https://arxiv.org/abs/2212.09058) 3 | # Github source: https://github.com/microsoft/unilm/tree/master/beats 4 | # Copyright (c) 2022 Microsoft 5 | # Licensed under The MIT License [see LICENSE for details] 6 | # Based on fairseq code bases 7 | # https://github.com/pytorch/fairseq 8 | # -------------------------------------------------------- 9 | 10 | import math 11 | import warnings 12 | import torch 13 | from torch import Tensor, nn 14 | import torch.nn.functional as F 15 | 16 | 17 | class GradMultiply(torch.autograd.Function): 18 | @staticmethod 19 | def forward(ctx, x, scale): 20 | ctx.scale = scale 21 | res = x.new(x) 22 | return res 23 | 24 | @staticmethod 25 | def backward(ctx, grad): 26 | return grad * ctx.scale, None 27 | 28 | 29 | class SamePad(nn.Module): 30 | def __init__(self, kernel_size, causal=False): 31 | super().__init__() 32 | if causal: 33 | self.remove = kernel_size - 1 34 | else: 35 | self.remove = 1 if kernel_size % 2 == 0 else 0 36 | 37 | def forward(self, x): 38 | if self.remove > 0: 39 | x = x[:, :, : -self.remove] 40 | return x 41 | 42 | 43 | class Swish(nn.Module): 44 | def __init__(self): 45 | super(Swish, self).__init__() 46 | self.act = torch.nn.Sigmoid() 47 | 48 | def forward(self, x): 49 | return x * self.act(x) 50 | 51 | 52 | class GLU_Linear(nn.Module): 53 | def __init__(self, input_dim, output_dim, glu_type="sigmoid", bias_in_glu=True): 54 | super(GLU_Linear, self).__init__() 55 | 56 | self.glu_type = glu_type 57 | self.output_dim = output_dim 58 | 59 | if glu_type == "sigmoid": 60 | self.glu_act = torch.nn.Sigmoid() 61 | elif glu_type == "swish": 62 | self.glu_act = Swish() 63 | elif glu_type == "relu": 64 | self.glu_act = torch.nn.ReLU() 65 | elif glu_type == "gelu": 66 | self.glu_act = torch.nn.GELU() 67 | 68 | if bias_in_glu: 69 | self.linear = nn.Linear(input_dim, output_dim * 2, True) 70 | else: 71 | self.linear = nn.Linear(input_dim, output_dim * 2, False) 72 | 73 | def forward(self, x): 74 | # to be consistent with GLU_Linear, we assume the input always has the #channel (#dim) in the last dimension of the tensor, so need to switch the dimension first for 1D-Conv case 75 | x = self.linear(x) 76 | 77 | if self.glu_type == "bilinear": 78 | x = (x[:, :, 0:self.output_dim] * x[:, :, self.output_dim:self.output_dim * 2]) 79 | else: 80 | x = (x[:, :, 0:self.output_dim] * self.glu_act(x[:, :, self.output_dim:self.output_dim * 2])) 81 | 82 | return x 83 | 84 | 85 | def gelu_accurate(x): 86 | if not hasattr(gelu_accurate, "_a"): 87 | gelu_accurate._a = math.sqrt(2 / math.pi) 88 | return ( 89 | 0.5 * x * (1 + torch.tanh(gelu_accurate._a * (x + 0.044715 * torch.pow(x, 3)))) 90 | ) 91 | 92 | 93 | def gelu(x: torch.Tensor) -> torch.Tensor: 94 | return torch.nn.functional.gelu(x.float()).type_as(x) 95 | 96 | 97 | def get_activation_fn(activation: str): 98 | """Returns the activation function corresponding to `activation`""" 99 | 100 | if activation == "relu": 101 | return F.relu 102 | elif activation == "gelu": 103 | return gelu 104 | elif activation == "gelu_fast": 105 | warnings.warn( 106 | "--activation-fn=gelu_fast has been renamed to gelu_accurate" 107 | ) 108 | return gelu_accurate 109 | elif activation == "gelu_accurate": 110 | return gelu_accurate 111 | elif activation == "tanh": 112 | return torch.tanh 113 | elif activation == "linear": 114 | return lambda x: x 115 | elif activation == "glu": 116 | return lambda x: x 117 | else: 118 | raise RuntimeError("--activation-fn {} not supported".format(activation)) 119 | 120 | 121 | def quant_noise(module, p, block_size): 122 | """ 123 | Wraps modules and applies quantization noise to the weights for 124 | subsequent quantization with Iterative Product Quantization as 125 | described in "Training with Quantization Noise for Extreme Model Compression" 126 | 127 | Args: 128 | - module: nn.Module 129 | - p: amount of Quantization Noise 130 | - block_size: size of the blocks for subsequent quantization with iPQ 131 | 132 | Remarks: 133 | - Module weights must have the right sizes wrt the block size 134 | - Only Linear, Embedding and Conv2d modules are supported for the moment 135 | - For more detail on how to quantize by blocks with convolutional weights, 136 | see "And the Bit Goes Down: Revisiting the Quantization of Neural Networks" 137 | - We implement the simplest form of noise here as stated in the paper 138 | which consists in randomly dropping blocks 139 | """ 140 | 141 | # if no quantization noise, don't register hook 142 | if p <= 0: 143 | return module 144 | 145 | # supported modules 146 | assert isinstance(module, (nn.Linear, nn.Embedding, nn.Conv2d)) 147 | 148 | # test whether module.weight has the right sizes wrt block_size 149 | is_conv = module.weight.ndim == 4 150 | 151 | # 2D matrix 152 | if not is_conv: 153 | assert ( 154 | module.weight.size(1) % block_size == 0 155 | ), "Input features must be a multiple of block sizes" 156 | 157 | # 4D matrix 158 | else: 159 | # 1x1 convolutions 160 | if module.kernel_size == (1, 1): 161 | assert ( 162 | module.in_channels % block_size == 0 163 | ), "Input channels must be a multiple of block sizes" 164 | # regular convolutions 165 | else: 166 | k = module.kernel_size[0] * module.kernel_size[1] 167 | assert k % block_size == 0, "Kernel size must be a multiple of block size" 168 | 169 | def _forward_pre_hook(mod, input): 170 | # no noise for evaluation 171 | if mod.training: 172 | if not is_conv: 173 | # gather weight and sizes 174 | weight = mod.weight 175 | in_features = weight.size(1) 176 | out_features = weight.size(0) 177 | 178 | # split weight matrix into blocks and randomly drop selected blocks 179 | mask = torch.zeros( 180 | in_features // block_size * out_features, device=weight.device 181 | ) 182 | mask.bernoulli_(p) 183 | mask = mask.repeat_interleave(block_size, -1).view(-1, in_features) 184 | 185 | else: 186 | # gather weight and sizes 187 | weight = mod.weight 188 | in_channels = mod.in_channels 189 | out_channels = mod.out_channels 190 | 191 | # split weight matrix into blocks and randomly drop selected blocks 192 | if mod.kernel_size == (1, 1): 193 | mask = torch.zeros( 194 | int(in_channels // block_size * out_channels), 195 | device=weight.device, 196 | ) 197 | mask.bernoulli_(p) 198 | mask = mask.repeat_interleave(block_size, -1).view(-1, in_channels) 199 | else: 200 | mask = torch.zeros( 201 | weight.size(0), weight.size(1), device=weight.device 202 | ) 203 | mask.bernoulli_(p) 204 | mask = ( 205 | mask.unsqueeze(2) 206 | .unsqueeze(3) 207 | .repeat(1, 1, mod.kernel_size[0], mod.kernel_size[1]) 208 | ) 209 | 210 | # scale weights and apply mask 211 | mask = mask.to( 212 | torch.bool 213 | ) # x.bool() is not currently supported in TorchScript 214 | s = 1 / (1 - p) 215 | mod.weight.data = s * weight.masked_fill(mask, 0) 216 | 217 | module.register_forward_pre_hook(_forward_pre_hook) 218 | return module -------------------------------------------------------------------------------- /beats/quantizer.py: -------------------------------------------------------------------------------- 1 | # -------------------------------------------------------- 2 | # BEATs: Audio Pre-Training with Acoustic Tokenizers (https://arxiv.org/abs/2212.09058) 3 | # Github source: https://github.com/microsoft/unilm/tree/master/beats 4 | # Copyright (c) 2022 Microsoft 5 | # Licensed under The MIT License [see LICENSE for details] 6 | # Based on VQGAN code bases 7 | # https://github.com/CompVis/taming-transformers 8 | # --------------------------------------------------------' 9 | 10 | import torch 11 | import torch.nn as nn 12 | import torch.nn.functional as F 13 | import torch.distributed as distributed 14 | 15 | try: 16 | from einops import rearrange, repeat 17 | except ImportError: 18 | pass 19 | 20 | 21 | def l2norm(t): 22 | return F.normalize(t, p=2, dim=-1) 23 | 24 | 25 | def ema_inplace(moving_avg, new, decay): 26 | moving_avg.data.mul_(decay).add_(new, alpha=(1 - decay)) 27 | 28 | 29 | def sample_vectors(samples, num): 30 | num_samples, device = samples.shape[0], samples.device 31 | 32 | if num_samples >= num: 33 | indices = torch.randperm(num_samples, device=device)[:num] 34 | else: 35 | indices = torch.randint(0, num_samples, (num,), device=device) 36 | 37 | return samples[indices] 38 | 39 | 40 | def kmeans(samples, num_clusters, num_iters=10, use_cosine_sim=False): 41 | dim, dtype, device = samples.shape[-1], samples.dtype, samples.device 42 | 43 | means = sample_vectors(samples, num_clusters) 44 | 45 | for _ in range(num_iters): 46 | if use_cosine_sim: 47 | dists = samples @ means.t() 48 | else: 49 | diffs = rearrange(samples, 'n d -> n () d') \ 50 | - rearrange(means, 'c d -> () c d') 51 | dists = -(diffs ** 2).sum(dim=-1) 52 | 53 | buckets = dists.max(dim=-1).indices 54 | bins = torch.bincount(buckets, minlength=num_clusters) 55 | zero_mask = bins == 0 56 | bins_min_clamped = bins.masked_fill(zero_mask, 1) 57 | 58 | new_means = buckets.new_zeros(num_clusters, dim, dtype=dtype) 59 | new_means.scatter_add_(0, repeat(buckets, 'n -> n d', d=dim), samples) 60 | new_means = new_means / bins_min_clamped[..., None] 61 | 62 | if use_cosine_sim: 63 | new_means = l2norm(new_means) 64 | 65 | means = torch.where(zero_mask[..., None], means, new_means) 66 | 67 | return means, bins 68 | 69 | 70 | class EmbeddingEMA(nn.Module): 71 | def __init__(self, num_tokens, codebook_dim, decay=0.99, eps=1e-5, kmeans_init=True, codebook_init_path=''): 72 | super().__init__() 73 | self.num_tokens = num_tokens 74 | self.codebook_dim = codebook_dim 75 | self.decay = decay 76 | self.eps = eps 77 | if codebook_init_path == '': 78 | if not kmeans_init: 79 | weight = torch.randn(num_tokens, codebook_dim) 80 | weight = l2norm(weight) 81 | else: 82 | weight = torch.zeros(num_tokens, codebook_dim) 83 | self.register_buffer('initted', torch.Tensor([not kmeans_init])) 84 | else: 85 | print(f"load init codebook weight from {codebook_init_path}") 86 | codebook_ckpt_weight = torch.load(codebook_init_path, map_location='cpu') 87 | weight = codebook_ckpt_weight.clone() 88 | self.register_buffer('initted', torch.Tensor([True])) 89 | 90 | self.weight = nn.Parameter(weight, requires_grad=False) 91 | self.cluster_size = nn.Parameter(torch.zeros(num_tokens), requires_grad=False) 92 | self.embed_avg = nn.Parameter(weight.clone(), requires_grad=False) 93 | # self.register_buffer('initted', torch.Tensor([not kmeans_init])) 94 | self.update = True 95 | 96 | @torch.jit.ignore 97 | def init_embed_(self, data): 98 | if self.initted: 99 | return 100 | print("Performing Kemans init for codebook") 101 | embed, cluster_size = kmeans(data, self.num_tokens, 10, use_cosine_sim=True) 102 | self.weight.data.copy_(embed) 103 | self.cluster_size.data.copy_(cluster_size) 104 | self.initted.data.copy_(torch.Tensor([True])) 105 | 106 | def forward(self, embed_id): 107 | return F.embedding(embed_id, self.weight) 108 | 109 | def cluster_size_ema_update(self, new_cluster_size): 110 | self.cluster_size.data.mul_(self.decay).add_(new_cluster_size, alpha=1 - self.decay) 111 | 112 | def embed_avg_ema_update(self, new_embed_avg): 113 | self.embed_avg.data.mul_(self.decay).add_(new_embed_avg, alpha=1 - self.decay) 114 | 115 | def weight_update(self, num_tokens): 116 | n = self.cluster_size.sum() 117 | smoothed_cluster_size = ( 118 | (self.cluster_size + self.eps) / (n + num_tokens * self.eps) * n 119 | ) 120 | # normalize embedding average with smoothed cluster size 121 | embed_normalized = self.embed_avg / smoothed_cluster_size.unsqueeze(1) 122 | # embed_normalized = l2norm(self.embed_avg / smoothed_cluster_size.unsqueeze(1)) 123 | self.weight.data.copy_(embed_normalized) 124 | 125 | 126 | def norm_ema_inplace(moving_avg, new, decay): 127 | moving_avg.data.mul_(decay).add_(new, alpha=(1 - decay)) 128 | moving_avg.data.copy_(l2norm(moving_avg.data)) 129 | 130 | 131 | class NormEMAVectorQuantizer(nn.Module): 132 | def __init__(self, n_embed, embedding_dim, beta, decay=0.99, eps=1e-5, 133 | statistic_code_usage=True, kmeans_init=False, codebook_init_path=''): 134 | super().__init__() 135 | self.codebook_dim = embedding_dim 136 | self.num_tokens = n_embed 137 | self.beta = beta 138 | self.decay = decay 139 | 140 | # learnable = True if orthogonal_reg_weight > 0 else False 141 | self.embedding = EmbeddingEMA(self.num_tokens, self.codebook_dim, decay, eps, kmeans_init, codebook_init_path) 142 | 143 | self.statistic_code_usage = statistic_code_usage 144 | if statistic_code_usage: 145 | self.register_buffer('cluster_size', torch.zeros(n_embed)) 146 | if distributed.is_available() and distributed.is_initialized(): 147 | print("ddp is enable, so use ddp_reduce to sync the statistic_code_usage for each gpu!") 148 | self.all_reduce_fn = distributed.all_reduce 149 | else: 150 | self.all_reduce_fn = nn.Identity() 151 | 152 | def reset_cluster_size(self, device): 153 | if self.statistic_code_usage: 154 | self.register_buffer('cluster_size', torch.zeros(self.num_tokens)) 155 | self.cluster_size = self.cluster_size.to(device) 156 | 157 | def forward(self, z): 158 | # reshape z -> (batch, height, width, channel) and flatten 159 | # z, 'b c h w -> b h w c' 160 | # z = rearrange(z, 'b c h w -> b h w c') 161 | # z = z.transpose(1, 2) 162 | z = l2norm(z) 163 | z_flattened = z.reshape(-1, self.codebook_dim) 164 | 165 | self.embedding.init_embed_(z_flattened) 166 | 167 | d = z_flattened.pow(2).sum(dim=1, keepdim=True) + \ 168 | self.embedding.weight.pow(2).sum(dim=1) - 2 * \ 169 | torch.einsum('bd,nd->bn', z_flattened, self.embedding.weight) # 'n d -> d n' 170 | 171 | encoding_indices = torch.argmin(d, dim=1) 172 | 173 | z_q = self.embedding(encoding_indices).view(z.shape) 174 | 175 | encodings = F.one_hot(encoding_indices, self.num_tokens).type(z.dtype) 176 | 177 | if not self.training: 178 | with torch.no_grad(): 179 | cluster_size = encodings.sum(0) 180 | self.all_reduce_fn(cluster_size) 181 | ema_inplace(self.cluster_size, cluster_size, self.decay) 182 | 183 | if self.training and self.embedding.update: 184 | # EMA cluster size 185 | 186 | bins = encodings.sum(0) 187 | self.all_reduce_fn(bins) 188 | 189 | # self.embedding.cluster_size_ema_update(bins) 190 | ema_inplace(self.cluster_size, bins, self.decay) 191 | 192 | zero_mask = (bins == 0) 193 | bins = bins.masked_fill(zero_mask, 1.) 194 | 195 | embed_sum = z_flattened.t() @ encodings 196 | self.all_reduce_fn(embed_sum) 197 | 198 | embed_normalized = (embed_sum / bins.unsqueeze(0)).t() 199 | embed_normalized = l2norm(embed_normalized) 200 | 201 | embed_normalized = torch.where(zero_mask[..., None], self.embedding.weight, 202 | embed_normalized) 203 | norm_ema_inplace(self.embedding.weight, embed_normalized, self.decay) 204 | 205 | # compute loss for embedding 206 | loss = self.beta * F.mse_loss(z_q.detach(), z) 207 | 208 | # preserve gradients 209 | z_q = z + (z_q - z).detach() 210 | 211 | # reshape back to match original input shape 212 | # z_q, 'b h w c -> b c h w' 213 | # z_q = rearrange(z_q, 'b h w c -> b c h w') 214 | # z_q = z_q.transpose(1, 2) 215 | return z_q, loss, encoding_indices -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/tools/test_plots.py: -------------------------------------------------------------------------------- 1 | #!/usr/bin/env python3 2 | # -*- coding: utf-8 -*- 3 | 4 | import sys 5 | import os 6 | import matplotlib.pyplot as plt 7 | import matplotlib.ticker as ticker 8 | import matplotlib.patches as patches 9 | import numpy as np 10 | import librosa 11 | import math 12 | import seaborn as sns 13 | import pandas as pd 14 | #import soundfile as sf 15 | #import scipy.io.wavfile as wio 16 | 17 | 18 | font_scalings = { 19 | 'xx-small' : 0.579, 20 | 'x-small' : 0.694, 21 | 'small' : 0.833, 22 | 'medium' : 1.0, 23 | 'large' : 1.200, 24 | 'x-large' : 1.440, 25 | 'xx-large' : 1.728, 26 | 'larger' : 1.2, 27 | 'smaller' : 0.833, 28 | None : 1.0 29 | } 30 | 31 | plot_type = { 32 | 'plot', 33 | 'boxplot', 34 | 'scatter', 35 | 'bar', 36 | 'barth', 37 | 'hist', 38 | 'image', 39 | 'confusion_matrix', 40 | None 41 | } 42 | 43 | class Figdata: 44 | def __init__(self, data, data2=[], type=None, labels=None, title=None, xlabel=None, ylabel=None, xticks=None, yticks=None, xlim=None, ylim=None, color=None, color2=None, mask=None, highlight_label=[]): 45 | self.data = data 46 | self.data2 = data2 47 | self.type = type 48 | self.title = title 49 | self.xlabel = xlabel 50 | self.ylabel = ylabel 51 | self.xticks = xticks 52 | self.yticks = yticks 53 | self.xlim = xlim 54 | self.ylim = ylim 55 | self.color = color 56 | self.color2 = color2 57 | self.labels = labels 58 | self.highlight_label = highlight_label 59 | 60 | 61 | def show_figs(*args, sup_title=None, sup_titlesize=None, dpi=100, width_mm=120, height_mm=30, 62 | margin_top_mm=15, margin_bottom_mm=15, margin_left_mm=25, margin_right_mm=15, margin_middle_mm=15, 63 | fold_interval=0, export_path="fig.png", is_display_console=False): 64 | num_figs_x = math.ceil(len(args)/fold_interval) 65 | num_figs_y = fold_interval 66 | #print('In show_figure(*args, sup_title=None, dpi=100, width_mm=120, height_mm=30)') 67 | #print("args: %d" % (num_figs)) 68 | #print("figsize: ", (plt.rcParams["figure.figsize"])) 69 | #print("dpi: ", (plt.rcParams["figure.dpi"])) 70 | #print("fontsize:", plt.rcParams["font.size"]) 71 | #print("supsize:", plt.rcParams["figure.titlesize"]) 72 | 73 | font_size = plt.rcParams["font.size"] 74 | font_scale = font_scalings[plt.rcParams["axes.titlesize"]] 75 | title_size = font_size * font_scale 76 | if sup_titlesize: 77 | #print(sup_titlesize) 78 | if sup_titlesize in font_scalings: 79 | font_scale = font_scalings[sup_titlesize] 80 | title_size = font_size * font_scale 81 | else: 82 | title_size = sup_titlesize 83 | #print(title_size) 84 | 85 | #print("title fontsize: ", font_size) 86 | #print("title font scale: ", font_scale) 87 | #print("title size: ", title_size) 88 | 89 | plt.style.use('dark_background') 90 | #width_mm = 120 91 | #height_mm = 30 92 | #margin_top_mm = 15 93 | #margin_bottom_mm = 15 94 | #margin_left_mm = 25 95 | #margin_right_mm = 10 96 | #margin_middle_mm = 15 97 | mm_per_inch = 25.4 98 | total_height_mm = margin_top_mm + margin_bottom_mm + (margin_middle_mm + height_mm)*(num_figs_y-1) + height_mm 99 | total_width_mm = width_mm + margin_left_mm + margin_right_mm + (margin_middle_mm + width_mm)*(num_figs_x-1) 100 | 101 | width_inch = width_mm / mm_per_inch 102 | height_inch = height_mm / mm_per_inch 103 | margin_top_inch = margin_top_mm / mm_per_inch 104 | margin_bottom_inch = margin_bottom_mm / mm_per_inch 105 | margin_left_inch = margin_left_mm / mm_per_inch 106 | margin_middle_inch = margin_middle_mm / mm_per_inch 107 | total_height_inch = total_height_mm / mm_per_inch 108 | total_width_inch = total_width_mm / mm_per_inch 109 | 110 | fig = plt.figure(figsize=(total_width_inch, total_height_inch), dpi=dpi) 111 | ax = [] 112 | for idx in range(len(args)): 113 | height = height_inch / total_height_inch 114 | width = width_inch / total_width_inch 115 | x0 = (margin_left_inch + (width_inch + margin_middle_inch)*(num_figs_x - 1 - idx//fold_interval)) / total_width_inch 116 | y0 = (margin_bottom_inch + (height_inch + margin_middle_inch)*(num_figs_y - 1 - idx%fold_interval)) / total_height_inch 117 | ax.append(fig.add_axes((x0, y0, width, height))) 118 | if type(args[idx]) is Figdata: 119 | if args[idx].type == None or args[idx].type == 'plot': 120 | if args[idx].color: 121 | ax[idx].plot(args[idx].data, color=args[idx].color) 122 | else: 123 | ax[idx].plot(args[idx].data) 124 | data2 = np.array(args[idx].data2) 125 | if len(data2.shape) != 0: 126 | if len(data2.shape) == 2: 127 | for d in data2: 128 | if args[idx].color2: 129 | ax[idx].plot(d, color=args[idx].color2) 130 | else: 131 | ax[idx].plot(d) 132 | else: 133 | if args[idx].color2: 134 | ax[idx].plot(data2, color=args[idx].color2) 135 | else: 136 | ax[idx].plot(data2) 137 | if args[idx].labels: 138 | ax[idx].legend(args[idx].labels) 139 | elif args[idx].type == 'boxplot': 140 | if len(np.shape(args[idx].data2)) > 1: 141 | ax[idx].boxplot((args[idx].data, *args[idx].data2), vert=False, showmeans=True, widths=0.7, labels=args[idx].labels) 142 | else: 143 | if len(args[idx].data2) != 0: 144 | ax[idx].boxplot((args[idx].data, args[idx].data2), vert=False, showmeans=True, widths=0.7, labels=args[idx].labels) 145 | else: 146 | ax[idx].boxplot(args[idx].data, vert=False, showmeans=True, widths=0.7, labels=args[idx].labels) 147 | elif args[idx].type == 'image' : 148 | im = plt.imshow(args[idx].data) 149 | if args[idx].data.shape[2] == 1: 150 | plt.colorbar(im) 151 | elif args[idx].type == 'confusion_matrix' : 152 | heatmap = sns.heatmap( 153 | args[idx].data, 154 | annot=True, 155 | xticklabels=args[idx].xticks, 156 | yticklabels=args[idx].yticks, 157 | fmt="d" 158 | ) 159 | for i in range(len(args[idx].highlight_label)): 160 | if args[idx].highlight_label[i] >= 0: 161 | heatmap.get_xticklabels()[args[idx].highlight_label[i]].set_weight("bold") 162 | heatmap.get_xticklabels()[args[idx].highlight_label[i]].set_color("#00ff00") 163 | heatmap.get_yticklabels()[i].set_weight("bold") 164 | heatmap.get_yticklabels()[i].set_color("#00ff00") 165 | heatmap.add_patch(patches.Rectangle(xy=(args[idx].highlight_label[i], i), width=1, height=1, ec='#00ff00', fill=False, linewidth=3)) 166 | if args[idx].xlabel: 167 | ax[idx].set_xlabel(args[idx].xlabel) 168 | if args[idx].ylabel: 169 | ax[idx].set_ylabel(args[idx].ylabel) 170 | if args[idx].title: 171 | ax[idx].set_title(args[idx].title, loc='left') 172 | if args[idx].xlim: 173 | ax[idx].set_xlim(args[idx].xlim) 174 | if args[idx].ylim: 175 | ax[idx].set_ylim(args[idx].ylim) 176 | else: 177 | ax[idx].plot(args[idx]) 178 | if sup_title: 179 | fig.suptitle(sup_title, y=(1 - 0.3 * margin_top_inch / total_height_inch), fontsize=title_size) 180 | if is_display_console: 181 | plt.show() 182 | plt.savefig(export_path) 183 | print("export fig -> {}".format(export_path)) 184 | plt.close() 185 | return 186 | 187 | 188 | @ticker.FuncFormatter 189 | def major_formatter_khz(y, pos): 190 | return '{:.0f}'.format(y/1000) 191 | 192 | if __name__ == '__main__': 193 | 194 | #print("figsize: ", (plt.rcParams["figure.figsize"])) 195 | #print("dpi: ", (plt.rcParams["figure.dpi"])) 196 | 197 | n_fft:int = 2048 198 | n_shift:int = 1024 199 | n_overlap = n_fft // n_shift 200 | 201 | x = np.linspace(0, 2*np.pi, 2048) 202 | vector1 = np.cos(x) 203 | vector2 = np.sin(x) 204 | 205 | filename = "./avemaria.wav" 206 | # y, sr = librosa.load(filename, sr=None, mono=False) 207 | y1, sr = librosa.load(filename, sr=None, mono=False, offset=3.0, duration=1.0) 208 | y2, sr = librosa.load(filename, sr=None, mono=False, offset=6.5, duration=1.0) 209 | 210 | y1_l = y1[0, :] 211 | y1_r = y1[1, :] 212 | y2_l = y2[0, :] 213 | y2_r = y2[1, :] 214 | 215 | ymax = max(max(y1_l),max(y1_r),max(y2_l),max(y2_r)) 216 | ymin = min(min(y1_l),min(y1_r),min(y2_l),min(y2_l)) 217 | #ymax = max(abs(ymax),abs(ymin)) 218 | #ymin = -ymax 219 | #ymax = 1 220 | #ymin = -1 221 | 222 | S = librosa.stft(y2_l, n_fft=n_fft, window='hamm') 223 | f0y_0 = Figdata(np.abs(S)[200], data2=np.full(len(np.abs(S[200])), 0.1), xlabel="freq", ylabel="magnitude", color='g', title="FFT: abs") 224 | f0y_1 = Figdata(np.log(np.abs(S)[200]), xlabel="freq", ylabel="magnitude", color='r', title="FFT: log(abs)") 225 | f0y_2 = Figdata(np.angle(S)[200], data2='5', xlabel="freq", ylabel="magnitude", title="FFT: angle") 226 | f1y_l = Figdata(y1_l, xlabel="freq", ylabel="magnitude", title="Fig1", ylim=(ymin,ymax)) 227 | f1y_r = Figdata(y1_r, xlabel="freq", ylabel="magnitude", title="Fig2", ylim=(ymin,ymax)) 228 | f2y_l = Figdata(y2_l, data2=np.full(len(y2_l), 0.3), color2='yellow', xlabel="freq", ylabel="magnitude", title="Fig3", ylim=(ymin,ymax)) 229 | f2y_r = Figdata(y2_r, xlabel="freq", ylabel="magnitude", title="Fig4", ylim=(ymin,ymax)) 230 | s1 = Figdata(np.random.randn(100), data2=np.random.randn(200), type='boxplot', color='r', ylabel='condition', xlabel='anomaly score', title='baseline AE', labels=['anomaly', 'normal']) 231 | 232 | show_figs(vector1, f0y_0, f0y_1, f0y_2, f1y_l, s1, f1y_r, f2y_l, f2y_r, sup_title="test", sup_titlesize='xx-large', dpi=70) 233 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/README.md: -------------------------------------------------------------------------------- 1 | # dcase2023\_task2\_evaluator 2 | The **dcase2023\_task2\_evaluator** is a script for calculating the AUC, pAUC, precision, recall, and F1 scores from the anomaly score list for the [evaluation dataset](https://zenodo.org/record/7860847) in DCASE 2023 Challenge Task 2 "First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring." 3 | 4 | [https://dcase.community/challenge2023/task-first-shot-unsupervised-anomalous-sound-detection-for-machine-condition-monitoring](https://dcase.community/challenge2023/task-first-shot-unsupervised-anomalous-sound-detection-for-machine-condition-monitoring) 5 | 6 | ## Description 7 | 8 | The **dcase2023\_task2\_evaluator** consists of two scripts: 9 | 10 | - `dcase2023_task2_evaluator.py` 11 | - This script outputs the AUC and pAUC scores by using: 12 | - Ground truth of the normal and anomaly labels 13 | - Anomaly scores for each wave file listed in the csv file for each machine type, section, and domain 14 | - Detection results for each wave file listed in the csv file for each machine type, section, and domain 15 | - `03_evaluation_eval_data.sh` 16 | - This script execute `dcase2023_task2_evaluator.py`. 17 | 18 | ## Usage 19 | ### 1. Clone repository 20 | Clone this repository from Github. 21 | 22 | ### 2. Prepare data 23 | - Anomaly scores 24 | - Generate csv files `anomaly_score__section__test.csv` and `decision_result__section__test.csv` or `anomaly_score_DCASE2023T2_section_
_test_seed_Eval.csv` and `decision_result_DCASE2023T2_section_
_test_seed_Eval.csv` by using a system for the [evaluation dataset](https://zenodo.org/record/7860847). (The format information is described [here](https://dcase.community/challenge2023/task-first-shot-unsupervised-anomalous-sound-detection-for-machine-condition-monitoring#submission).) 25 | - Rename the directory containing the csv files to a team name 26 | - Move the directory into `./teams/` 27 | 28 | ### 3. Check directory structure 29 | - ./dcase2023\_task2\_evaluator 30 | - /dcase2023\_task2\_evaluator.py 31 | - /03\_evaluation\_eval\_data.sh 32 | - /ground\_truth\_attributes 33 | - ground\_truth\_bandsaw\_section\_00\_test.csv 34 | - ground\_truth\_grinder\_section\_00\_test.csv 35 | - ... 36 | - /ground\_truth\_data 37 | - ground\_truth\_bandsaw\_section\_00\_test.csv 38 | - ground\_truth\_grinder\section\_00\_test.csv 39 | - ... 40 | - /ground\_truth\_domain 41 | - ground\_truth\_bandsaw\_section\_00\_test.csv 42 | - ground\_truth\_grinder\_section\_00\_test.csv 43 | - ... 44 | - /teams 45 | - /\ 46 | - /\ 47 | - anomaly\_score\_bandsaw\_section\_00\_test.csv 48 | - anomaly\_score\_grinder\_section\_00\_test.csv 49 | - ... 50 | - decision\_result\_ToyTank\_section\_00\_test.csv 51 | - decision\_result\_Vacuum\_section\_00\_test.csv 52 | - /\ 53 | - anomaly\_score\_DCASE2023T2bandsaw\_section\_00\_test\_seed\<--seed\>\<--tag\>\_Eval.csv 54 | - anomaly\_score\_DCASE2023T2grinder\_section\_00\_test\_seed\<--seed\>\<--tag\>\_Eval.csv 55 | - ... 56 | - decision\_result\_DCASE2023T2ToyTank\_section\_00\_test\_seed\<--seed\>\<--tag\>\_Eval.csv 57 | - decision\_result\_DCASE2023T2Vacuum\_section\_00\_test\_seed\<--seed\>\<--tag\>\_Eval.csv 58 | - /\ 59 | - /\ 60 | - anomaly\_score\_bandsaw\_section\_00\_test.csv 61 | - anomaly\_score\_grinder\_section\_00\test.csv 62 | - ... 63 | - decision\_result\_ToyTank\_section\_00\_test.csv 64 | - decision\_result\_Vacuum\_section\_00\_test.csv 65 | - ... 66 | - /teams\_result 67 | - \\_result.csv 68 | - \\_result.csv 69 | - \_result.csv 70 | - ... 71 | - /teams\_additional\_result \*`out_all==True` 72 | - teams\_official\_score.csv 73 | - teams\_official\_score\_paper.csv 74 | - teams\_section\_00\_auc.csv 75 | - teams\_section\_00\_score.csv 76 | - /\ 77 | - official\_score.csv 78 | - \\_bandsaw\_section\_00\_anm\_score.png 79 | - ... 80 | - \\_Vacuum\_section\_00\_anm\_score.png 81 | - /\ 82 | - official\_score.csv 83 | - \\_bandsaw\_section\_00\_anm\_score.png 84 | - ... 85 | - \\_Vacuum\_section\_00\_anm\_score.png 86 | - /\ 87 | - official\_score.csv 88 | - \\_bandsaw\_section\_00\_anm\_score.png 89 | - ... 90 | - \\_Vacuum\_section\_00\_anm\_score.png 91 | - ... 92 | - /tools 93 | - plot\_anm\_score.py 94 | - test\_plots.py 95 | - /README.md 96 | 97 | 98 | ### 4. Change parameters 99 | The parameters are defined in the script `dcase2023_task2_evaluator.py` as follows. 100 | - **MAX\_FPR** 101 | - The FPR threshold for pAUC : default 0.1 102 | - **--result\_dir** 103 | - The output directory : default `./teams_result/` 104 | - **--teams\_root\_dir** 105 | - Directory containing team results. : default `./teams/` 106 | - **--dir\_depth** 107 | - What depth to search `--teams_root_dir` using glob. : default `2` 108 | - If --dir\_depth=2, then `glob.glob(/*/*)` 109 | - **--tag** 110 | - File name tag. : default `_id(0_)` 111 | - If using filename is DCASE2023 baseline style, change parameters as necessary. 112 | - **--seed** 113 | - Seed used during train. : default `13711` 114 | - If using filename is DCASE2023 baseline style, change parameters as necessary. 115 | - **--out\_all** 116 | - If this parameter is `True`, export supplemental data. : default `False` 117 | - **--additional\_result\_dir** 118 | - The output additional results directory. : default `./teams_additional_result/` 119 | - Used when `--out_all==True`. 120 | 121 | ### 5. Run script 122 | Run the script `dcase2023_task2_evaluator.py` 123 | ``` 124 | $ python dcase2023_task2_evaluator.py 125 | ``` 126 | or 127 | ``` 128 | $ bash 03_evaluation_eval_data.sh 129 | ``` 130 | The script `dcase2023_task2_evaluator.py` calculates the AUC, pAUC, precision, recall, and F1 scores for each machine type, section, and domain and output the calculated scores into the csv files (`_result.csv`, `_result.csv`, ...) in **--result\_dir** (default: `./teams_result/`). 131 | If **--out\_all=True**, each team results are then aggregated into a csv file (`teams_official_score.csv`, `teams_official_score_paper.csv`) in **--additional\_result\_dir** (default: `./teams_additional_result`). 132 | 133 | ### 6. Check results 134 | You can check the AUC, pAUC, precision, recall, and F1 scores in the `_result.csv` in **--result\_dir**. 135 | The AUC, pAUC, precision, recall, and F1 scores for each machine type, section, and domain are listed as follows: 136 | 137 | `_result.csv` 138 | ``` 139 | ToyDrone 140 | section,AUC (all),AUC (source),AUC (target),pAUC,precision (source),precision (target),recall (source),recall (target),F1 score (source),F1 score (target) 141 | 00,0.6789,0.7968000000000001,0.561,0.5368421052631579,0.7560975609756098,0.5079365079365079,0.62,0.64,0.6813186813186813,0.5663716814159292 142 | ,,AUC,pAUC,precision,recall,F1 score 143 | arithmetic mean,,0.6789000000000001,0.5368421052631579,0.6320170344560588,0.63,0.6238451813673053 144 | harmonic mean,,0.6584250994255415,0.5368421052631579,0.6076569678407351,0.6298412698412698,0.6185502727981294 145 | source harmonic mean,,0.7968000000000001,0.5368421052631579,0.7560975609756098,0.62,0.6813186813186813 146 | target harmonic mean,,0.561,0.5368421052631579,0.5079365079365079,0.64,0.5663716814159292 147 | 148 | ... 149 | 150 | shaker 151 | section,AUC (all),AUC (source),AUC (target),pAUC,precision (source),precision (target),recall (source),recall (target),F1 score (source),F1 score (target) 152 | 00,0.6253625362536254,0.69428983714698,0.5604118104118104,0.5491654428600755,0.578125,0.4936708860759494,0.6981132075471698,0.8478260869565217,0.6324786324786325,0.624 153 | ,,AUC,pAUC,precision,recall,F1 score 154 | arithmetic mean,,0.6273508237793952,0.5491654428600755,0.5358979430379747,0.7729696472518457,0.6282393162393163 155 | harmonic mean,,0.6202083584461523,0.5491654428600755,0.5325705849787784,0.765720350225524,0.628210709621245 156 | source harmonic mean,,0.69428983714698,0.5491654428600755,0.578125,0.6981132075471698,0.6324786324786325 157 | target harmonic mean,,0.5604118104118104,0.5491654428600755,0.4936708860759494,0.8478260869565217,0.624 158 | 159 | ... 160 | 161 | ,,AUC,pAUC,precision,recall,F1 score 162 | "arithmetic mean over all machine types, sections, and domains",,0.6403576632460674,0.5535708782745333,0.5364448553682957,0.7308992232966801,0.6006994950456381 163 | "harmonic mean over all machine types, sections, and domains",,0.6152996272906976,0.5517419647782388,0.5032829900980702,0.7137886024875123,0.590331192259057 164 | "source harmonic mean over all machine types, sections, and domains",,0.7423494890244248,0.5517419647782388,0.5356629533316296,0.660146438268587,0.5914253446046336 165 | "target harmonic mean over all machine types, sections, and domains",,0.5253826834789426,0.5517419647782388,0.4745945156180243,0.7769195103318602,0.5892410808585007 166 | 167 | official score,,0.5925469043549957 168 | official score ci95,,1.531898879903843e-05 169 | ``` 170 | 171 | Aggregated results for each baseline are listed as follows: 172 | 173 | ```_seed13711_official_score_paper.csv 174 | System,metric,h-mean,a-mean,bandsaw,grinder,shaker,ToyDrone,ToyNscale,ToyTank,Vacuum 175 | baseline_MAHALA,AUC (source),0.7871141310430937,0.7922050973063174,0.836434267021059,0.7409411378914575,0.8476602762317048,0.8130000000000001,0.6678000000000002,0.8011999999999999,0.8383999999999999 176 | baseline_MAHALA,AUC (target),0.53090862919051,0.5643372095607136,0.5885844748858448,0.50682261208577,0.6299533799533801,0.4622,0.409,0.4532,0.9006 177 | baseline_MAHALA,"pAUC (source, target)",0.5682103280865886,0.5727080310911363,0.5753594967896497,0.5955291246149972,0.6233307541280444,0.5142105263157895,0.5089473684210526,0.5384210526315789,0.653157894736842 178 | baseline_MAHALA,TOTAL score,0.6105082186925268,0.6430834459860557,,,,,,, 179 | baseline_MSE,AUC (source),0.7423494890244248,0.7482088346609131,0.6667348190554078,0.706837186424004,0.69428983714698,0.7968000000000001,0.77,0.7212000000000001,0.8816 180 | baseline_MSE,AUC (target),0.5253826834789426,0.5325064918312215,0.48287671232876717,0.5516569200779727,0.5604118104118104,0.561,0.4716,0.6468,0.45320000000000005 181 | baseline_MSE,"pAUC (source, target)",0.5517419647782388,0.5535708782745333,0.5091087658743451,0.5846166760294185,0.5491654428600755,0.5368421052631579,0.5178947368421053,0.5826315789473684,0.5947368421052631 182 | baseline_MSE,TOTAL score,0.5925469043549957,0.6114287349222227,,,,,,, 183 | 184 | ``` 185 | 186 | If you use this system, please cite all the following four papers: 187 | 188 | + Kota Dohi, Keisuke Imoto, Noboru Harada, Daisuke Niizumi, Yuma Koizumi, Tomoya Nishida, Harsh Purohit, Ryo Tanabe, Takashi Endo, and Yohei Kawaguchi, "Description and Discussion on DCASE 2023 Challenge Task 2: First-Shot Unsupervised Anomalous Sound Detection for Machine Condition Monitoring," in arXiv-eprints: 2305.07828, 2023. [URL](https://arxiv.org/abs/2305.07828) 189 | + Noboru Harada, Daisuke Niizumi, Daiki Takeuchi, Yasunori Ohishi, Masahiro Yasuda, Shoichiro Saito, "ToyADMOS2: Another Dataset of Miniature-Machine Operating Sounds for Anomalous Sound Detection under Domain Shift Conditions," in Proc. DCASE 2022 Workshop, 2022. [URL](https://dcase.community/documents/workshop2021/proceedings/DCASE2021Workshop_Harada_6.pdf) 190 | + Kota Dohi, Tomoya Nishida, Harsh Purohit, Ryo Tanabe, Takashi Endo, Masaaki Yamamoto, Yuki Nikaido, and Yohei Kawaguchi, "MIMII DG: sound dataset for malfunctioning industrial machine investigation and inspection for domain generalization task," in Proc. DCASE 2022 Workshop, 2022. [URL](https://dcase.community/documents/workshop2022/proceedings/DCASE2022Workshop_Dohi_62.pdf) 191 | + Noboru Harada, Daisuke Niizumi, Daiki Takeuchi, Yasunori Ohishi, Masahiro Yasuda, "First-Shot Anomaly Detection for Machine Condition Monitoring: A Domain Generalization Baseline," in arXiv e-prints: 2303.00455, 2023. [URL](https://arxiv.org/abs/2303.00455) -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_attributes/ground_truth_bandsaw_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,section_00_target_test_normal_0000_vel_12 2 | section_00_0001.wav,section_00_target_test_anomaly_0001_vel_12 3 | section_00_0002.wav,section_00_target_test_normal_0002_vel_12 4 | section_00_0003.wav,section_00_target_test_anomaly_0003_vel_12 5 | section_00_0004.wav,section_00_target_test_normal_0004_vel_12 6 | section_00_0005.wav,section_00_source_test_anomaly_0005_vel_15 7 | section_00_0006.wav,section_00_target_test_normal_0006_vel_14 8 | section_00_0007.wav,section_00_target_test_normal_0007_vel_14 9 | section_00_0008.wav,section_00_target_test_normal_0008_vel_14 10 | section_00_0009.wav,section_00_target_test_anomaly_0009_vel_14 11 | section_00_0010.wav,section_00_source_test_normal_0010_vel_15 12 | section_00_0011.wav,section_00_source_test_normal_0011_vel_15 13 | section_00_0012.wav,section_00_source_test_normal_0012_vel_15 14 | section_00_0013.wav,section_00_target_test_normal_0013_vel_12 15 | section_00_0014.wav,section_00_source_test_normal_0014_vel_13 16 | section_00_0015.wav,section_00_source_test_normal_0015_vel_15 17 | section_00_0016.wav,section_00_source_test_anomaly_0016_vel_15 18 | section_00_0017.wav,section_00_source_test_normal_0017_vel_13 19 | section_00_0018.wav,section_00_target_test_anomaly_0018_vel_12 20 | section_00_0019.wav,section_00_source_test_normal_0019_vel_15 21 | section_00_0020.wav,section_00_target_test_normal_0020_vel_14 22 | section_00_0021.wav,section_00_source_test_anomaly_0021_vel_15 23 | section_00_0022.wav,section_00_target_test_anomaly_0022_vel_14 24 | section_00_0023.wav,section_00_target_test_normal_0023_vel_12 25 | section_00_0024.wav,section_00_target_test_normal_0024_vel_14 26 | section_00_0025.wav,section_00_source_test_normal_0025_vel_13 27 | section_00_0026.wav,section_00_source_test_normal_0026_vel_13 28 | section_00_0027.wav,section_00_target_test_normal_0027_vel_12 29 | section_00_0028.wav,section_00_target_test_normal_0028_vel_12 30 | section_00_0029.wav,section_00_target_test_anomaly_0029_vel_14 31 | section_00_0030.wav,section_00_target_test_normal_0030_vel_12 32 | section_00_0031.wav,section_00_source_test_normal_0031_vel_15 33 | section_00_0032.wav,section_00_target_test_normal_0032_vel_12 34 | section_00_0033.wav,section_00_source_test_normal_0033_vel_13 35 | section_00_0034.wav,section_00_source_test_normal_0034_vel_15 36 | section_00_0035.wav,section_00_source_test_anomaly_0035_vel_15 37 | section_00_0036.wav,section_00_target_test_normal_0036_vel_12 38 | section_00_0037.wav,section_00_source_test_normal_0037_vel_15 39 | section_00_0038.wav,section_00_source_test_normal_0038_vel_13 40 | section_00_0039.wav,section_00_source_test_normal_0039_vel_15 41 | section_00_0040.wav,section_00_target_test_anomaly_0040_vel_14 42 | section_00_0041.wav,section_00_target_test_normal_0041_vel_14 43 | section_00_0042.wav,section_00_target_test_normal_0042_vel_14 44 | section_00_0043.wav,section_00_source_test_normal_0043_vel_15 45 | section_00_0044.wav,section_00_source_test_anomaly_0044_vel_13 46 | section_00_0045.wav,section_00_source_test_anomaly_0045_vel_15 47 | section_00_0046.wav,section_00_source_test_anomaly_0046_vel_15 48 | section_00_0047.wav,section_00_source_test_normal_0047_vel_13 49 | section_00_0048.wav,section_00_source_test_normal_0048_vel_13 50 | section_00_0049.wav,section_00_target_test_anomaly_0049_vel_12 51 | section_00_0050.wav,section_00_source_test_normal_0050_vel_13 52 | section_00_0051.wav,section_00_target_test_anomaly_0051_vel_14 53 | section_00_0052.wav,section_00_source_test_normal_0052_vel_15 54 | section_00_0053.wav,section_00_target_test_normal_0053_vel_12 55 | section_00_0054.wav,section_00_target_test_normal_0054_vel_12 56 | section_00_0055.wav,section_00_source_test_anomaly_0055_vel_13 57 | section_00_0056.wav,section_00_target_test_normal_0056_vel_14 58 | section_00_0057.wav,section_00_target_test_normal_0057_vel_12 59 | section_00_0058.wav,section_00_source_test_normal_0058_vel_15 60 | section_00_0059.wav,section_00_target_test_normal_0059_vel_12 61 | section_00_0060.wav,section_00_target_test_normal_0060_vel_12 62 | section_00_0061.wav,section_00_source_test_anomaly_0061_vel_15 63 | section_00_0062.wav,section_00_target_test_anomaly_0062_vel_12 64 | section_00_0063.wav,section_00_source_test_anomaly_0063_vel_15 65 | section_00_0064.wav,section_00_source_test_normal_0064_vel_13 66 | section_00_0065.wav,section_00_source_test_normal_0065_vel_13 67 | section_00_0066.wav,section_00_target_test_normal_0066_vel_14 68 | section_00_0067.wav,section_00_source_test_normal_0067_vel_15 69 | section_00_0068.wav,section_00_source_test_normal_0068_vel_13 70 | section_00_0069.wav,section_00_target_test_anomaly_0069_vel_12 71 | section_00_0070.wav,section_00_target_test_normal_0070_vel_14 72 | section_00_0071.wav,section_00_source_test_anomaly_0071_vel_15 73 | section_00_0072.wav,section_00_source_test_normal_0072_vel_13 74 | section_00_0073.wav,section_00_source_test_normal_0073_vel_15 75 | section_00_0074.wav,section_00_source_test_normal_0074_vel_13 76 | section_00_0075.wav,section_00_source_test_anomaly_0075_vel_13 77 | section_00_0076.wav,section_00_source_test_normal_0076_vel_15 78 | section_00_0077.wav,section_00_source_test_anomaly_0077_vel_15 79 | section_00_0078.wav,section_00_target_test_normal_0078_vel_14 80 | section_00_0079.wav,section_00_source_test_anomaly_0079_vel_15 81 | section_00_0080.wav,section_00_target_test_normal_0080_vel_14 82 | section_00_0081.wav,section_00_target_test_normal_0081_vel_12 83 | section_00_0082.wav,section_00_source_test_anomaly_0082_vel_15 84 | section_00_0083.wav,section_00_source_test_anomaly_0083_vel_15 85 | section_00_0084.wav,section_00_source_test_normal_0084_vel_13 86 | section_00_0085.wav,section_00_source_test_anomaly_0085_vel_13 87 | section_00_0086.wav,section_00_source_test_anomaly_0086_vel_13 88 | section_00_0087.wav,section_00_target_test_normal_0087_vel_14 89 | section_00_0088.wav,section_00_target_test_normal_0088_vel_12 90 | section_00_0089.wav,section_00_source_test_anomaly_0089_vel_15 91 | section_00_0090.wav,section_00_target_test_normal_0090_vel_12 92 | section_00_0091.wav,section_00_target_test_anomaly_0091_vel_14 93 | section_00_0092.wav,section_00_target_test_anomaly_0092_vel_12 94 | section_00_0093.wav,section_00_source_test_anomaly_0093_vel_13 95 | section_00_0094.wav,section_00_source_test_normal_0094_vel_15 96 | section_00_0095.wav,section_00_source_test_anomaly_0095_vel_15 97 | section_00_0096.wav,section_00_source_test_anomaly_0096_vel_15 98 | section_00_0097.wav,section_00_target_test_normal_0097_vel_12 99 | section_00_0098.wav,section_00_source_test_anomaly_0098_vel_13 100 | section_00_0099.wav,section_00_source_test_normal_0099_vel_13 101 | section_00_0100.wav,section_00_target_test_anomaly_0100_vel_14 102 | section_00_0101.wav,section_00_source_test_normal_0101_vel_15 103 | section_00_0102.wav,section_00_source_test_normal_0102_vel_15 104 | section_00_0103.wav,section_00_target_test_normal_0103_vel_12 105 | section_00_0104.wav,section_00_source_test_anomaly_0104_vel_15 106 | section_00_0105.wav,section_00_target_test_normal_0105_vel_14 107 | section_00_0106.wav,section_00_target_test_anomaly_0106_vel_12 108 | section_00_0107.wav,section_00_source_test_normal_0107_vel_15 109 | section_00_0108.wav,section_00_source_test_normal_0108_vel_15 110 | section_00_0109.wav,section_00_source_test_anomaly_0109_vel_15 111 | section_00_0110.wav,section_00_target_test_anomaly_0110_vel_12 112 | section_00_0111.wav,section_00_target_test_anomaly_0111_vel_12 113 | section_00_0112.wav,section_00_target_test_anomaly_0112_vel_12 114 | section_00_0113.wav,section_00_target_test_anomaly_0113_vel_12 115 | section_00_0114.wav,section_00_source_test_normal_0114_vel_15 116 | section_00_0115.wav,section_00_target_test_anomaly_0115_vel_12 117 | section_00_0116.wav,section_00_source_test_normal_0116_vel_15 118 | section_00_0117.wav,section_00_source_test_anomaly_0117_vel_13 119 | section_00_0118.wav,section_00_target_test_anomaly_0118_vel_12 120 | section_00_0119.wav,section_00_target_test_normal_0119_vel_12 121 | section_00_0120.wav,section_00_source_test_anomaly_0120_vel_13 122 | section_00_0121.wav,section_00_target_test_normal_0121_vel_14 123 | section_00_0122.wav,section_00_source_test_anomaly_0122_vel_13 124 | section_00_0123.wav,section_00_target_test_normal_0123_vel_12 125 | section_00_0124.wav,section_00_source_test_normal_0124_vel_15 126 | section_00_0125.wav,section_00_target_test_normal_0125_vel_14 127 | section_00_0126.wav,section_00_source_test_normal_0126_vel_15 128 | section_00_0127.wav,section_00_target_test_normal_0127_vel_12 129 | section_00_0128.wav,section_00_target_test_normal_0128_vel_14 130 | section_00_0129.wav,section_00_target_test_normal_0129_vel_12 131 | section_00_0130.wav,section_00_source_test_normal_0130_vel_13 132 | section_00_0131.wav,section_00_target_test_normal_0131_vel_14 133 | section_00_0132.wav,section_00_target_test_anomaly_0132_vel_12 134 | section_00_0133.wav,section_00_target_test_normal_0133_vel_12 135 | section_00_0134.wav,section_00_target_test_anomaly_0134_vel_12 136 | section_00_0135.wav,section_00_source_test_normal_0135_vel_13 137 | section_00_0136.wav,section_00_source_test_anomaly_0136_vel_15 138 | section_00_0137.wav,section_00_source_test_normal_0137_vel_13 139 | section_00_0138.wav,section_00_source_test_normal_0138_vel_13 140 | section_00_0139.wav,section_00_source_test_normal_0139_vel_15 141 | section_00_0140.wav,section_00_source_test_normal_0140_vel_13 142 | section_00_0141.wav,section_00_target_test_normal_0141_vel_12 143 | section_00_0142.wav,section_00_target_test_normal_0142_vel_12 144 | section_00_0143.wav,section_00_source_test_anomaly_0143_vel_15 145 | section_00_0144.wav,section_00_source_test_normal_0144_vel_13 146 | section_00_0145.wav,section_00_target_test_anomaly_0145_vel_14 147 | section_00_0146.wav,section_00_target_test_normal_0146_vel_12 148 | section_00_0147.wav,section_00_source_test_normal_0147_vel_15 149 | section_00_0148.wav,section_00_source_test_normal_0148_vel_13 150 | section_00_0149.wav,section_00_source_test_anomaly_0149_vel_15 151 | section_00_0150.wav,section_00_target_test_normal_0150_vel_12 152 | section_00_0151.wav,section_00_source_test_normal_0151_vel_15 153 | section_00_0152.wav,section_00_source_test_normal_0152_vel_15 154 | section_00_0153.wav,section_00_source_test_normal_0153_vel_13 155 | section_00_0154.wav,section_00_target_test_anomaly_0154_vel_14 156 | section_00_0155.wav,section_00_target_test_normal_0155_vel_12 157 | section_00_0156.wav,section_00_source_test_anomaly_0156_vel_15 158 | section_00_0157.wav,section_00_target_test_anomaly_0157_vel_14 159 | section_00_0158.wav,section_00_target_test_anomaly_0158_vel_12 160 | section_00_0159.wav,section_00_source_test_normal_0159_vel_15 161 | section_00_0160.wav,section_00_target_test_normal_0160_vel_12 162 | section_00_0161.wav,section_00_target_test_anomaly_0161_vel_14 163 | section_00_0162.wav,section_00_source_test_anomaly_0162_vel_15 164 | section_00_0163.wav,section_00_source_test_normal_0163_vel_13 165 | section_00_0164.wav,section_00_source_test_normal_0164_vel_15 166 | section_00_0165.wav,section_00_source_test_normal_0165_vel_15 167 | section_00_0166.wav,section_00_target_test_normal_0166_vel_14 168 | section_00_0167.wav,section_00_source_test_normal_0167_vel_15 169 | section_00_0168.wav,section_00_target_test_anomaly_0168_vel_12 170 | section_00_0169.wav,section_00_source_test_normal_0169_vel_15 171 | section_00_0170.wav,section_00_target_test_normal_0170_vel_12 172 | section_00_0171.wav,section_00_source_test_anomaly_0171_vel_15 173 | section_00_0172.wav,section_00_source_test_anomaly_0172_vel_15 174 | section_00_0173.wav,section_00_target_test_normal_0173_vel_12 175 | section_00_0174.wav,section_00_source_test_anomaly_0174_vel_13 176 | section_00_0175.wav,section_00_source_test_anomaly_0175_vel_15 177 | section_00_0176.wav,section_00_target_test_anomaly_0176_vel_14 178 | section_00_0177.wav,section_00_target_test_normal_0177_vel_14 179 | section_00_0178.wav,section_00_target_test_normal_0178_vel_12 180 | section_00_0179.wav,section_00_source_test_normal_0179_vel_15 181 | section_00_0180.wav,section_00_source_test_normal_0180_vel_15 182 | section_00_0181.wav,section_00_source_test_normal_0181_vel_15 183 | section_00_0182.wav,section_00_target_test_anomaly_0182_vel_12 184 | section_00_0183.wav,section_00_target_test_normal_0183_vel_12 185 | section_00_0184.wav,section_00_source_test_normal_0184_vel_15 186 | section_00_0185.wav,section_00_target_test_anomaly_0185_vel_14 187 | section_00_0186.wav,section_00_target_test_normal_0186_vel_14 188 | section_00_0187.wav,section_00_target_test_normal_0187_vel_14 189 | section_00_0188.wav,section_00_source_test_anomaly_0188_vel_13 190 | section_00_0189.wav,section_00_source_test_normal_0189_vel_13 191 | section_00_0190.wav,section_00_source_test_normal_0190_vel_15 192 | section_00_0191.wav,section_00_target_test_anomaly_0191_vel_12 193 | section_00_0192.wav,section_00_target_test_normal_0192_vel_14 194 | section_00_0193.wav,section_00_source_test_normal_0193_vel_13 195 | section_00_0194.wav,section_00_source_test_anomaly_0194_vel_15 196 | section_00_0195.wav,section_00_target_test_normal_0195_vel_12 197 | section_00_0196.wav,section_00_target_test_normal_0196_vel_14 198 | section_00_0197.wav,section_00_source_test_normal_0197_vel_13 199 | section_00_0198.wav,section_00_source_test_anomaly_0198_vel_15 200 | section_00_0199.wav,section_00_source_test_normal_0199_vel_13 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_attributes/ground_truth_shaker_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,section_00_source_test_anomaly_0000_speed_1000 2 | section_00_0001.wav,section_00_target_test_anomaly_0001_speed_1200 3 | section_00_0002.wav,section_00_source_test_anomaly_0002_speed_1500 4 | section_00_0003.wav,section_00_target_test_normal_0003_speed_1200 5 | section_00_0004.wav,section_00_source_test_anomaly_0004_speed_1000 6 | section_00_0005.wav,section_00_target_test_normal_0005_speed_1200 7 | section_00_0006.wav,section_00_source_test_anomaly_0006_speed_1500 8 | section_00_0007.wav,section_00_source_test_normal_0007_speed_1500 9 | section_00_0008.wav,section_00_source_test_anomaly_0008_speed_1000 10 | section_00_0009.wav,section_00_source_test_normal_0009_speed_1000 11 | section_00_0010.wav,section_00_source_test_anomaly_0010_speed_1500 12 | section_00_0011.wav,section_00_target_test_normal_0011_speed_1200 13 | section_00_0012.wav,section_00_target_test_normal_0012_speed_1200 14 | section_00_0013.wav,section_00_target_test_anomaly_0013_speed_1200 15 | section_00_0014.wav,section_00_target_test_normal_0014_speed_1200 16 | section_00_0015.wav,section_00_target_test_normal_0015_speed_1200 17 | section_00_0016.wav,section_00_target_test_anomaly_0016_speed_1200 18 | section_00_0017.wav,section_00_source_test_anomaly_0017_speed_1500 19 | section_00_0018.wav,section_00_target_test_anomaly_0018_speed_1200 20 | section_00_0019.wav,section_00_source_test_anomaly_0019_speed_1000 21 | section_00_0020.wav,section_00_target_test_normal_0020_speed_1200 22 | section_00_0021.wav,section_00_source_test_normal_0021_speed_1000 23 | section_00_0022.wav,section_00_target_test_normal_0022_speed_1200 24 | section_00_0023.wav,section_00_target_test_anomaly_0023_speed_1200 25 | section_00_0024.wav,section_00_target_test_normal_0024_speed_1200 26 | section_00_0025.wav,section_00_target_test_normal_0025_speed_1200 27 | section_00_0026.wav,section_00_source_test_normal_0026_speed_1000 28 | section_00_0027.wav,section_00_target_test_normal_0027_speed_1200 29 | section_00_0028.wav,section_00_target_test_normal_0028_speed_1200 30 | section_00_0029.wav,section_00_source_test_anomaly_0029_speed_1500 31 | section_00_0030.wav,section_00_target_test_normal_0030_speed_1200 32 | section_00_0031.wav,section_00_source_test_anomaly_0031_speed_1500 33 | section_00_0032.wav,section_00_source_test_normal_0032_speed_1500 34 | section_00_0033.wav,section_00_source_test_anomaly_0033_speed_1500 35 | section_00_0034.wav,section_00_source_test_normal_0034_speed_1000 36 | section_00_0035.wav,section_00_source_test_anomaly_0035_speed_1000 37 | section_00_0036.wav,section_00_source_test_anomaly_0036_speed_1500 38 | section_00_0037.wav,section_00_target_test_normal_0037_speed_1200 39 | section_00_0038.wav,section_00_source_test_normal_0038_speed_1500 40 | section_00_0039.wav,section_00_source_test_normal_0039_speed_1000 41 | section_00_0040.wav,section_00_source_test_normal_0040_speed_1000 42 | section_00_0041.wav,section_00_source_test_anomaly_0041_speed_1000 43 | section_00_0042.wav,section_00_source_test_anomaly_0042_speed_1000 44 | section_00_0043.wav,section_00_source_test_anomaly_0043_speed_1000 45 | section_00_0044.wav,section_00_source_test_anomaly_0044_speed_1500 46 | section_00_0045.wav,section_00_source_test_normal_0045_speed_1500 47 | section_00_0046.wav,section_00_source_test_normal_0046_speed_1500 48 | section_00_0047.wav,section_00_target_test_anomaly_0047_speed_1200 49 | section_00_0048.wav,section_00_target_test_normal_0048_speed_1200 50 | section_00_0049.wav,section_00_target_test_normal_0049_speed_1200 51 | section_00_0050.wav,section_00_target_test_normal_0050_speed_1200 52 | section_00_0051.wav,section_00_source_test_anomaly_0051_speed_1000 53 | section_00_0052.wav,section_00_source_test_normal_0052_speed_1500 54 | section_00_0053.wav,section_00_target_test_normal_0053_speed_1200 55 | section_00_0054.wav,section_00_target_test_anomaly_0054_speed_1200 56 | section_00_0055.wav,section_00_source_test_normal_0055_speed_1000 57 | section_00_0056.wav,section_00_source_test_normal_0056_speed_1500 58 | section_00_0057.wav,section_00_target_test_anomaly_0057_speed_1200 59 | section_00_0058.wav,section_00_target_test_anomaly_0058_speed_1200 60 | section_00_0059.wav,section_00_target_test_normal_0059_speed_1200 61 | section_00_0060.wav,section_00_source_test_normal_0060_speed_1000 62 | section_00_0061.wav,section_00_source_test_normal_0061_speed_1000 63 | section_00_0062.wav,section_00_source_test_normal_0062_speed_1000 64 | section_00_0063.wav,section_00_target_test_anomaly_0063_speed_1200 65 | section_00_0064.wav,section_00_target_test_normal_0064_speed_1200 66 | section_00_0065.wav,section_00_source_test_normal_0065_speed_1500 67 | section_00_0066.wav,section_00_source_test_anomaly_0066_speed_1000 68 | section_00_0067.wav,section_00_target_test_anomaly_0067_speed_1200 69 | section_00_0068.wav,section_00_source_test_anomaly_0068_speed_1000 70 | section_00_0069.wav,section_00_source_test_normal_0069_speed_1500 71 | section_00_0070.wav,section_00_target_test_normal_0070_speed_1200 72 | section_00_0071.wav,section_00_target_test_normal_0071_speed_1200 73 | section_00_0072.wav,section_00_target_test_anomaly_0072_speed_1200 74 | section_00_0073.wav,section_00_target_test_anomaly_0073_speed_1200 75 | section_00_0074.wav,section_00_source_test_normal_0074_speed_1500 76 | section_00_0075.wav,section_00_source_test_anomaly_0075_speed_1500 77 | section_00_0076.wav,section_00_target_test_anomaly_0076_speed_1200 78 | section_00_0077.wav,section_00_target_test_normal_0077_speed_1200 79 | section_00_0078.wav,section_00_target_test_normal_0078_speed_1200 80 | section_00_0079.wav,section_00_target_test_normal_0079_speed_1200 81 | section_00_0080.wav,section_00_target_test_anomaly_0080_speed_1200 82 | section_00_0081.wav,section_00_source_test_anomaly_0081_speed_1500 83 | section_00_0082.wav,section_00_source_test_normal_0082_speed_1000 84 | section_00_0083.wav,section_00_target_test_anomaly_0083_speed_1200 85 | section_00_0084.wav,section_00_source_test_normal_0084_speed_1500 86 | section_00_0085.wav,section_00_target_test_anomaly_0085_speed_1200 87 | section_00_0086.wav,section_00_source_test_anomaly_0086_speed_1500 88 | section_00_0087.wav,section_00_source_test_normal_0087_speed_1500 89 | section_00_0088.wav,section_00_target_test_anomaly_0088_speed_1200 90 | section_00_0089.wav,section_00_target_test_anomaly_0089_speed_1200 91 | section_00_0090.wav,section_00_source_test_normal_0090_speed_1000 92 | section_00_0091.wav,section_00_source_test_anomaly_0091_speed_1500 93 | section_00_0092.wav,section_00_source_test_normal_0092_speed_1000 94 | section_00_0093.wav,section_00_target_test_normal_0093_speed_1200 95 | section_00_0094.wav,section_00_target_test_anomaly_0094_speed_1200 96 | section_00_0095.wav,section_00_target_test_normal_0095_speed_1200 97 | section_00_0096.wav,section_00_target_test_anomaly_0096_speed_1200 98 | section_00_0097.wav,section_00_source_test_anomaly_0097_speed_1000 99 | section_00_0098.wav,section_00_source_test_anomaly_0098_speed_1000 100 | section_00_0099.wav,section_00_target_test_normal_0099_speed_1200 101 | section_00_0100.wav,section_00_source_test_anomaly_0100_speed_1000 102 | section_00_0101.wav,section_00_target_test_anomaly_0101_speed_1200 103 | section_00_0102.wav,section_00_target_test_anomaly_0102_speed_1200 104 | section_00_0103.wav,section_00_target_test_anomaly_0103_speed_1200 105 | section_00_0104.wav,section_00_target_test_anomaly_0104_speed_1200 106 | section_00_0105.wav,section_00_source_test_anomaly_0105_speed_1000 107 | section_00_0106.wav,section_00_source_test_anomaly_0106_speed_1000 108 | section_00_0107.wav,section_00_target_test_normal_0107_speed_1200 109 | section_00_0108.wav,section_00_target_test_anomaly_0108_speed_1200 110 | section_00_0109.wav,section_00_target_test_normal_0109_speed_1200 111 | section_00_0110.wav,section_00_source_test_normal_0110_speed_1000 112 | section_00_0111.wav,section_00_source_test_anomaly_0111_speed_1500 113 | section_00_0112.wav,section_00_source_test_anomaly_0112_speed_1000 114 | section_00_0113.wav,section_00_target_test_anomaly_0113_speed_1200 115 | section_00_0114.wav,section_00_target_test_anomaly_0114_speed_1200 116 | section_00_0115.wav,section_00_source_test_normal_0115_speed_1500 117 | section_00_0116.wav,section_00_target_test_normal_0116_speed_1200 118 | section_00_0117.wav,section_00_source_test_anomaly_0117_speed_1000 119 | section_00_0118.wav,section_00_source_test_normal_0118_speed_1000 120 | section_00_0119.wav,section_00_source_test_anomaly_0119_speed_1000 121 | section_00_0120.wav,section_00_source_test_normal_0120_speed_1500 122 | section_00_0121.wav,section_00_target_test_anomaly_0121_speed_1200 123 | section_00_0122.wav,section_00_target_test_anomaly_0122_speed_1200 124 | section_00_0123.wav,section_00_source_test_normal_0123_speed_1500 125 | section_00_0124.wav,section_00_target_test_anomaly_0124_speed_1200 126 | section_00_0125.wav,section_00_source_test_normal_0125_speed_1000 127 | section_00_0126.wav,section_00_target_test_normal_0126_speed_1200 128 | section_00_0127.wav,section_00_source_test_anomaly_0127_speed_1500 129 | section_00_0128.wav,section_00_source_test_normal_0128_speed_1500 130 | section_00_0129.wav,section_00_source_test_normal_0129_speed_1000 131 | section_00_0130.wav,section_00_source_test_normal_0130_speed_1500 132 | section_00_0131.wav,section_00_source_test_normal_0131_speed_1500 133 | section_00_0132.wav,section_00_source_test_normal_0132_speed_1000 134 | section_00_0133.wav,section_00_source_test_anomaly_0133_speed_1500 135 | section_00_0134.wav,section_00_source_test_anomaly_0134_speed_1000 136 | section_00_0135.wav,section_00_target_test_anomaly_0135_speed_1200 137 | section_00_0136.wav,section_00_source_test_anomaly_0136_speed_1500 138 | section_00_0137.wav,section_00_target_test_normal_0137_speed_1200 139 | section_00_0138.wav,section_00_target_test_anomaly_0138_speed_1200 140 | section_00_0139.wav,section_00_target_test_normal_0139_speed_1200 141 | section_00_0140.wav,section_00_source_test_anomaly_0140_speed_1000 142 | section_00_0141.wav,section_00_source_test_anomaly_0141_speed_1500 143 | section_00_0142.wav,section_00_source_test_anomaly_0142_speed_1500 144 | section_00_0143.wav,section_00_target_test_normal_0143_speed_1200 145 | section_00_0144.wav,section_00_target_test_normal_0144_speed_1200 146 | section_00_0145.wav,section_00_target_test_normal_0145_speed_1200 147 | section_00_0146.wav,section_00_target_test_anomaly_0146_speed_1200 148 | section_00_0147.wav,section_00_target_test_anomaly_0147_speed_1200 149 | section_00_0148.wav,section_00_target_test_normal_0148_speed_1200 150 | section_00_0149.wav,section_00_target_test_anomaly_0149_speed_1200 151 | section_00_0150.wav,section_00_target_test_anomaly_0150_speed_1200 152 | section_00_0151.wav,section_00_source_test_normal_0151_speed_1500 153 | section_00_0152.wav,section_00_target_test_anomaly_0152_speed_1200 154 | section_00_0153.wav,section_00_target_test_anomaly_0153_speed_1200 155 | section_00_0154.wav,section_00_source_test_normal_0154_speed_1000 156 | section_00_0155.wav,section_00_source_test_anomaly_0155_speed_1500 157 | section_00_0156.wav,section_00_target_test_anomaly_0156_speed_1200 158 | section_00_0157.wav,section_00_source_test_normal_0157_speed_1500 159 | section_00_0158.wav,section_00_target_test_anomaly_0158_speed_1200 160 | section_00_0159.wav,section_00_source_test_anomaly_0159_speed_1000 161 | section_00_0160.wav,section_00_target_test_normal_0160_speed_1200 162 | section_00_0161.wav,section_00_target_test_normal_0161_speed_1200 163 | section_00_0162.wav,section_00_source_test_anomaly_0162_speed_1500 164 | section_00_0163.wav,section_00_source_test_normal_0163_speed_1000 165 | section_00_0164.wav,section_00_source_test_normal_0164_speed_1500 166 | section_00_0165.wav,section_00_target_test_normal_0165_speed_1200 167 | section_00_0166.wav,section_00_target_test_normal_0166_speed_1200 168 | section_00_0167.wav,section_00_source_test_normal_0167_speed_1500 169 | section_00_0168.wav,section_00_target_test_normal_0168_speed_1200 170 | section_00_0169.wav,section_00_target_test_anomaly_0169_speed_1200 171 | section_00_0170.wav,section_00_target_test_anomaly_0170_speed_1200 172 | section_00_0171.wav,section_00_target_test_normal_0171_speed_1200 173 | section_00_0172.wav,section_00_source_test_normal_0172_speed_1000 174 | section_00_0173.wav,section_00_source_test_anomaly_0173_speed_1500 175 | section_00_0174.wav,section_00_source_test_anomaly_0174_speed_1500 176 | section_00_0175.wav,section_00_target_test_normal_0175_speed_1200 177 | section_00_0176.wav,section_00_source_test_normal_0176_speed_1000 178 | section_00_0177.wav,section_00_source_test_anomaly_0177_speed_1500 179 | section_00_0178.wav,section_00_target_test_normal_0178_speed_1200 180 | section_00_0179.wav,section_00_target_test_anomaly_0179_speed_1200 181 | section_00_0180.wav,section_00_source_test_normal_0180_speed_1000 182 | section_00_0181.wav,section_00_source_test_anomaly_0181_speed_1500 183 | section_00_0182.wav,section_00_target_test_normal_0182_speed_1200 184 | section_00_0183.wav,section_00_source_test_anomaly_0183_speed_1500 185 | section_00_0184.wav,section_00_target_test_anomaly_0184_speed_1200 186 | section_00_0185.wav,section_00_source_test_anomaly_0185_speed_1500 187 | section_00_0186.wav,section_00_source_test_anomaly_0186_speed_1500 188 | section_00_0187.wav,section_00_source_test_normal_0187_speed_1500 189 | section_00_0188.wav,section_00_source_test_normal_0188_speed_1500 190 | section_00_0189.wav,section_00_target_test_anomaly_0189_speed_1200 191 | section_00_0190.wav,section_00_target_test_normal_0190_speed_1200 192 | section_00_0191.wav,section_00_source_test_anomaly_0191_speed_1500 193 | section_00_0192.wav,section_00_source_test_normal_0192_speed_1000 194 | section_00_0193.wav,section_00_target_test_normal_0193_speed_1200 195 | section_00_0194.wav,section_00_source_test_anomaly_0194_speed_1000 196 | section_00_0195.wav,section_00_target_test_normal_0195_speed_1200 197 | section_00_0196.wav,section_00_target_test_normal_0196_speed_1200 198 | section_00_0197.wav,section_00_source_test_normal_0197_speed_1000 199 | section_00_0198.wav,section_00_source_test_anomaly_0198_speed_1000 200 | section_00_0199.wav,section_00_target_test_normal_0199_speed_1200 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_attributes/ground_truth_ToyDrone_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,section_00_target_test_anomaly_0012_car_C1_spd_1_mic_2 2 | section_00_0001.wav,section_00_target_test_normal_0031_car_C1_spd_1_mic_2 3 | section_00_0002.wav,section_00_source_test_normal_0031_car_B2_spd_2_mic_1 4 | section_00_0003.wav,section_00_target_test_anomaly_0037_car_C1_spd_1_mic_2 5 | section_00_0004.wav,section_00_target_test_anomaly_0001_car_C1_spd_1_mic_2 6 | section_00_0005.wav,section_00_source_test_normal_0003_car_A1_spd_3_mic_1 7 | section_00_0006.wav,section_00_source_test_normal_0048_car_A2_spd_3_mic_1 8 | section_00_0007.wav,section_00_target_test_anomaly_0047_car_C2_spd_1_mic_2 9 | section_00_0008.wav,section_00_target_test_normal_0004_car_C1_spd_1_mic_2 10 | section_00_0009.wav,section_00_source_test_anomaly_0031_car_B2_spd_2_mic_1 11 | section_00_0010.wav,section_00_target_test_anomaly_0007_car_C2_spd_1_mic_2 12 | section_00_0011.wav,section_00_source_test_normal_0045_car_B2_spd_2_mic_1 13 | section_00_0012.wav,section_00_source_test_normal_0014_car_B1_spd_2_mic_1 14 | section_00_0013.wav,section_00_target_test_normal_0005_car_C1_spd_1_mic_2 15 | section_00_0014.wav,section_00_target_test_normal_0047_car_C1_spd_1_mic_2 16 | section_00_0015.wav,section_00_source_test_anomaly_0002_car_A1_spd_3_mic_1 17 | section_00_0016.wav,section_00_source_test_anomaly_0022_car_A1_spd_3_mic_1 18 | section_00_0017.wav,section_00_target_test_normal_0009_car_C1_spd_1_mic_2 19 | section_00_0018.wav,section_00_source_test_anomaly_0032_car_B1_spd_2_mic_1 20 | section_00_0019.wav,section_00_source_test_anomaly_0048_car_B2_spd_2_mic_1 21 | section_00_0020.wav,section_00_target_test_normal_0043_car_C1_spd_1_mic_2 22 | section_00_0021.wav,section_00_target_test_normal_0023_car_C1_spd_1_mic_2 23 | section_00_0022.wav,section_00_source_test_normal_0041_car_B2_spd_2_mic_1 24 | section_00_0023.wav,section_00_target_test_anomaly_0026_car_C2_spd_1_mic_2 25 | section_00_0024.wav,section_00_source_test_anomaly_0044_car_B2_spd_2_mic_1 26 | section_00_0025.wav,section_00_target_test_anomaly_0025_car_C1_spd_1_mic_2 27 | section_00_0026.wav,section_00_source_test_normal_0027_car_B1_spd_2_mic_1 28 | section_00_0027.wav,section_00_source_test_normal_0049_car_A2_spd_3_mic_1 29 | section_00_0028.wav,section_00_target_test_anomaly_0020_car_C1_spd_1_mic_2 30 | section_00_0029.wav,section_00_source_test_anomaly_0047_car_B1_spd_2_mic_1 31 | section_00_0030.wav,section_00_source_test_anomaly_0003_car_A1_spd_3_mic_1 32 | section_00_0031.wav,section_00_source_test_normal_0013_car_B1_spd_2_mic_1 33 | section_00_0032.wav,section_00_target_test_anomaly_0002_car_C1_spd_1_mic_2 34 | section_00_0033.wav,section_00_source_test_normal_0044_car_B2_spd_2_mic_1 35 | section_00_0034.wav,section_00_target_test_anomaly_0046_car_C2_spd_1_mic_2 36 | section_00_0035.wav,section_00_source_test_normal_0015_car_B1_spd_2_mic_1 37 | section_00_0036.wav,section_00_source_test_normal_0038_car_A1_spd_3_mic_1 38 | section_00_0037.wav,section_00_target_test_normal_0028_car_C1_spd_1_mic_2 39 | section_00_0038.wav,section_00_source_test_anomaly_0043_car_B2_spd_2_mic_1 40 | section_00_0039.wav,section_00_target_test_anomaly_0030_car_C2_spd_1_mic_2 41 | section_00_0040.wav,section_00_target_test_anomaly_0021_car_C1_spd_1_mic_2 42 | section_00_0041.wav,section_00_source_test_normal_0024_car_A1_spd_3_mic_1 43 | section_00_0042.wav,section_00_source_test_normal_0017_car_B2_spd_2_mic_1 44 | section_00_0043.wav,section_00_target_test_anomaly_0035_car_C1_spd_1_mic_2 45 | section_00_0044.wav,section_00_target_test_anomaly_0018_car_C1_spd_1_mic_2 46 | section_00_0045.wav,section_00_source_test_anomaly_0010_car_A1_spd_3_mic_1 47 | section_00_0046.wav,section_00_source_test_normal_0033_car_B2_spd_2_mic_1 48 | section_00_0047.wav,section_00_target_test_anomaly_0032_car_C2_spd_1_mic_2 49 | section_00_0048.wav,section_00_target_test_normal_0021_car_C2_spd_1_mic_2 50 | section_00_0049.wav,section_00_target_test_anomaly_0044_car_C2_spd_1_mic_2 51 | section_00_0050.wav,section_00_target_test_normal_0018_car_C2_spd_1_mic_2 52 | section_00_0051.wav,section_00_source_test_normal_0021_car_A1_spd_3_mic_1 53 | section_00_0052.wav,section_00_target_test_normal_0032_car_C1_spd_1_mic_2 54 | section_00_0053.wav,section_00_target_test_anomaly_0039_car_C1_spd_1_mic_2 55 | section_00_0054.wav,section_00_target_test_normal_0016_car_C2_spd_1_mic_2 56 | section_00_0055.wav,section_00_source_test_anomaly_0033_car_B1_spd_2_mic_1 57 | section_00_0056.wav,section_00_source_test_anomaly_0017_car_A2_spd_3_mic_1 58 | section_00_0057.wav,section_00_target_test_normal_0024_car_C1_spd_1_mic_2 59 | section_00_0058.wav,section_00_target_test_normal_0020_car_C2_spd_1_mic_2 60 | section_00_0059.wav,section_00_target_test_normal_0006_car_C1_spd_1_mic_2 61 | section_00_0060.wav,section_00_source_test_anomaly_0026_car_B1_spd_2_mic_1 62 | section_00_0061.wav,section_00_source_test_normal_0040_car_A1_spd_3_mic_1 63 | section_00_0062.wav,section_00_target_test_anomaly_0008_car_C2_spd_1_mic_2 64 | section_00_0063.wav,section_00_target_test_normal_0015_car_C2_spd_1_mic_2 65 | section_00_0064.wav,section_00_target_test_anomaly_0017_car_C1_spd_1_mic_2 66 | section_00_0065.wav,section_00_source_test_anomaly_0025_car_A2_spd_3_mic_1 67 | section_00_0066.wav,section_00_source_test_anomaly_0011_car_A2_spd_3_mic_1 68 | section_00_0067.wav,section_00_source_test_anomaly_0028_car_B1_spd_2_mic_1 69 | section_00_0068.wav,section_00_source_test_anomaly_0020_car_A1_spd_3_mic_1 70 | section_00_0069.wav,section_00_source_test_anomaly_0007_car_A2_spd_3_mic_1 71 | section_00_0070.wav,section_00_target_test_anomaly_0015_car_C2_spd_1_mic_2 72 | section_00_0071.wav,section_00_target_test_normal_0045_car_C2_spd_1_mic_2 73 | section_00_0072.wav,section_00_target_test_normal_0010_car_C1_spd_1_mic_2 74 | section_00_0073.wav,section_00_source_test_normal_0006_car_A2_spd_3_mic_1 75 | section_00_0074.wav,section_00_target_test_normal_0040_car_C2_spd_1_mic_2 76 | section_00_0075.wav,section_00_source_test_anomaly_0034_car_B1_spd_2_mic_1 77 | section_00_0076.wav,section_00_source_test_anomaly_0040_car_B1_spd_2_mic_1 78 | section_00_0077.wav,section_00_source_test_normal_0032_car_B2_spd_2_mic_1 79 | section_00_0078.wav,section_00_source_test_anomaly_0014_car_A1_spd_3_mic_1 80 | section_00_0079.wav,section_00_source_test_anomaly_0041_car_B1_spd_2_mic_1 81 | section_00_0080.wav,section_00_target_test_anomaly_0045_car_C2_spd_1_mic_2 82 | section_00_0081.wav,section_00_target_test_anomaly_0029_car_C2_spd_1_mic_2 83 | section_00_0082.wav,section_00_source_test_anomaly_0019_car_A2_spd_3_mic_1 84 | section_00_0083.wav,section_00_target_test_anomaly_0050_car_C2_spd_1_mic_2 85 | section_00_0084.wav,section_00_source_test_anomaly_0027_car_B1_spd_2_mic_1 86 | section_00_0085.wav,section_00_source_test_normal_0036_car_A1_spd_3_mic_1 87 | section_00_0086.wav,section_00_source_test_anomaly_0050_car_B2_spd_2_mic_1 88 | section_00_0087.wav,section_00_target_test_normal_0039_car_C2_spd_1_mic_2 89 | section_00_0088.wav,section_00_target_test_normal_0022_car_C1_spd_1_mic_2 90 | section_00_0089.wav,section_00_target_test_normal_0003_car_C1_spd_1_mic_2 91 | section_00_0090.wav,section_00_source_test_normal_0019_car_B2_spd_2_mic_1 92 | section_00_0091.wav,section_00_target_test_anomaly_0049_car_C2_spd_1_mic_2 93 | section_00_0092.wav,section_00_target_test_normal_0048_car_C1_spd_1_mic_2 94 | section_00_0093.wav,section_00_source_test_anomaly_0024_car_A2_spd_3_mic_1 95 | section_00_0094.wav,section_00_target_test_anomaly_0033_car_C2_spd_1_mic_2 96 | section_00_0095.wav,section_00_source_test_anomaly_0001_car_A1_spd_3_mic_1 97 | section_00_0096.wav,section_00_source_test_normal_0010_car_A2_spd_3_mic_1 98 | section_00_0097.wav,section_00_source_test_normal_0012_car_B1_spd_2_mic_1 99 | section_00_0098.wav,section_00_source_test_normal_0008_car_A2_spd_3_mic_1 100 | section_00_0099.wav,section_00_target_test_normal_0037_car_C2_spd_1_mic_2 101 | section_00_0100.wav,section_00_source_test_normal_0029_car_B1_spd_2_mic_1 102 | section_00_0101.wav,section_00_source_test_normal_0043_car_B2_spd_2_mic_1 103 | section_00_0102.wav,section_00_target_test_normal_0050_car_C2_spd_1_mic_2 104 | section_00_0103.wav,section_00_target_test_normal_0036_car_C2_spd_1_mic_2 105 | section_00_0104.wav,section_00_source_test_anomaly_0036_car_B2_spd_2_mic_1 106 | section_00_0105.wav,section_00_target_test_anomaly_0038_car_C1_spd_1_mic_2 107 | section_00_0106.wav,section_00_target_test_normal_0046_car_C2_spd_1_mic_2 108 | section_00_0107.wav,section_00_target_test_normal_0017_car_C2_spd_1_mic_2 109 | section_00_0108.wav,section_00_target_test_anomaly_0022_car_C1_spd_1_mic_2 110 | section_00_0109.wav,section_00_target_test_anomaly_0005_car_C2_spd_1_mic_2 111 | section_00_0110.wav,section_00_target_test_anomaly_0048_car_C2_spd_1_mic_2 112 | section_00_0111.wav,section_00_source_test_normal_0030_car_B1_spd_2_mic_1 113 | section_00_0112.wav,section_00_target_test_anomaly_0009_car_C1_spd_1_mic_2 114 | section_00_0113.wav,section_00_source_test_normal_0026_car_B1_spd_2_mic_1 115 | section_00_0114.wav,section_00_source_test_normal_0001_car_A1_spd_3_mic_1 116 | section_00_0115.wav,section_00_source_test_anomaly_0006_car_A2_spd_3_mic_1 117 | section_00_0116.wav,section_00_target_test_normal_0038_car_C2_spd_1_mic_2 118 | section_00_0117.wav,section_00_source_test_normal_0005_car_A1_spd_3_mic_1 119 | section_00_0118.wav,section_00_source_test_normal_0025_car_A1_spd_3_mic_1 120 | section_00_0119.wav,section_00_source_test_normal_0016_car_B2_spd_2_mic_1 121 | section_00_0120.wav,section_00_source_test_normal_0050_car_A2_spd_3_mic_1 122 | section_00_0121.wav,section_00_target_test_normal_0041_car_C2_spd_1_mic_2 123 | section_00_0122.wav,section_00_source_test_anomaly_0045_car_B1_spd_2_mic_1 124 | section_00_0123.wav,section_00_target_test_anomaly_0016_car_C2_spd_1_mic_2 125 | section_00_0124.wav,section_00_source_test_anomaly_0021_car_A1_spd_3_mic_1 126 | section_00_0125.wav,section_00_source_test_anomaly_0029_car_B2_spd_2_mic_1 127 | section_00_0126.wav,section_00_source_test_normal_0004_car_A1_spd_3_mic_1 128 | section_00_0127.wav,section_00_source_test_normal_0022_car_A1_spd_3_mic_1 129 | section_00_0128.wav,section_00_source_test_anomaly_0005_car_A2_spd_3_mic_1 130 | section_00_0129.wav,section_00_target_test_normal_0027_car_C1_spd_1_mic_2 131 | section_00_0130.wav,section_00_target_test_normal_0034_car_C2_spd_1_mic_2 132 | section_00_0131.wav,section_00_target_test_normal_0008_car_C1_spd_1_mic_2 133 | section_00_0132.wav,section_00_source_test_anomaly_0030_car_B2_spd_2_mic_1 134 | section_00_0133.wav,section_00_target_test_anomaly_0040_car_C1_spd_1_mic_2 135 | section_00_0134.wav,section_00_target_test_normal_0012_car_C2_spd_1_mic_2 136 | section_00_0135.wav,section_00_source_test_anomaly_0037_car_B2_spd_2_mic_1 137 | section_00_0136.wav,section_00_source_test_normal_0020_car_B2_spd_2_mic_1 138 | section_00_0137.wav,section_00_source_test_anomaly_0038_car_B2_spd_2_mic_1 139 | section_00_0138.wav,section_00_target_test_anomaly_0036_car_C1_spd_1_mic_2 140 | section_00_0139.wav,section_00_source_test_normal_0028_car_B1_spd_2_mic_1 141 | section_00_0140.wav,section_00_source_test_normal_0007_car_A2_spd_3_mic_1 142 | section_00_0141.wav,section_00_source_test_normal_0042_car_B2_spd_2_mic_1 143 | section_00_0142.wav,section_00_target_test_anomaly_0003_car_C1_spd_1_mic_2 144 | section_00_0143.wav,section_00_source_test_anomaly_0049_car_B2_spd_2_mic_1 145 | section_00_0144.wav,section_00_target_test_anomaly_0006_car_C2_spd_1_mic_2 146 | section_00_0145.wav,section_00_target_test_anomaly_0004_car_C1_spd_1_mic_2 147 | section_00_0146.wav,section_00_target_test_anomaly_0010_car_C1_spd_1_mic_2 148 | section_00_0147.wav,section_00_target_test_anomaly_0028_car_C2_spd_1_mic_2 149 | section_00_0148.wav,section_00_source_test_anomaly_0023_car_A2_spd_3_mic_1 150 | section_00_0149.wav,section_00_target_test_normal_0033_car_C2_spd_1_mic_2 151 | section_00_0150.wav,section_00_source_test_anomaly_0035_car_B1_spd_2_mic_1 152 | section_00_0151.wav,section_00_target_test_normal_0035_car_C2_spd_1_mic_2 153 | section_00_0152.wav,section_00_source_test_anomaly_0018_car_A2_spd_3_mic_1 154 | section_00_0153.wav,section_00_target_test_anomaly_0014_car_C2_spd_1_mic_2 155 | section_00_0154.wav,section_00_target_test_anomaly_0034_car_C2_spd_1_mic_2 156 | section_00_0155.wav,section_00_target_test_anomaly_0023_car_C1_spd_1_mic_2 157 | section_00_0156.wav,section_00_target_test_anomaly_0024_car_C1_spd_1_mic_2 158 | section_00_0157.wav,section_00_target_test_normal_0029_car_C1_spd_1_mic_2 159 | section_00_0158.wav,section_00_source_test_anomaly_0004_car_A2_spd_3_mic_1 160 | section_00_0159.wav,section_00_source_test_normal_0047_car_A2_spd_3_mic_1 161 | section_00_0160.wav,section_00_target_test_normal_0025_car_C1_spd_1_mic_2 162 | section_00_0161.wav,section_00_source_test_normal_0037_car_A1_spd_3_mic_1 163 | section_00_0162.wav,section_00_source_test_normal_0023_car_A1_spd_3_mic_1 164 | section_00_0163.wav,section_00_target_test_anomaly_0019_car_C1_spd_1_mic_2 165 | section_00_0164.wav,section_00_source_test_anomaly_0039_car_B1_spd_2_mic_1 166 | section_00_0165.wav,section_00_source_test_anomaly_0013_car_A2_spd_3_mic_1 167 | section_00_0166.wav,section_00_target_test_anomaly_0011_car_C1_spd_1_mic_2 168 | section_00_0167.wav,section_00_target_test_normal_0013_car_C2_spd_1_mic_2 169 | section_00_0168.wav,section_00_target_test_anomaly_0031_car_C2_spd_1_mic_2 170 | section_00_0169.wav,section_00_source_test_normal_0011_car_B1_spd_2_mic_1 171 | section_00_0170.wav,section_00_source_test_anomaly_0008_car_A1_spd_3_mic_1 172 | section_00_0171.wav,section_00_target_test_normal_0007_car_C1_spd_1_mic_2 173 | section_00_0172.wav,section_00_source_test_normal_0009_car_A2_spd_3_mic_1 174 | section_00_0173.wav,section_00_source_test_normal_0002_car_A1_spd_3_mic_1 175 | section_00_0174.wav,section_00_source_test_anomaly_0009_car_A1_spd_3_mic_1 176 | section_00_0175.wav,section_00_target_test_normal_0011_car_C2_spd_1_mic_2 177 | section_00_0176.wav,section_00_target_test_normal_0001_car_C1_spd_1_mic_2 178 | section_00_0177.wav,section_00_source_test_anomaly_0016_car_A1_spd_3_mic_1 179 | section_00_0178.wav,section_00_target_test_anomaly_0042_car_C1_spd_1_mic_2 180 | section_00_0179.wav,section_00_target_test_normal_0049_car_C2_spd_1_mic_2 181 | section_00_0180.wav,section_00_target_test_anomaly_0041_car_C1_spd_1_mic_2 182 | section_00_0181.wav,section_00_target_test_normal_0002_car_C1_spd_1_mic_2 183 | section_00_0182.wav,section_00_source_test_anomaly_0012_car_A2_spd_3_mic_1 184 | section_00_0183.wav,section_00_target_test_normal_0014_car_C2_spd_1_mic_2 185 | section_00_0184.wav,section_00_target_test_anomaly_0027_car_C2_spd_1_mic_2 186 | section_00_0185.wav,section_00_target_test_anomaly_0043_car_C2_spd_1_mic_2 187 | section_00_0186.wav,section_00_source_test_anomaly_0015_car_A1_spd_3_mic_1 188 | section_00_0187.wav,section_00_source_test_normal_0034_car_B2_spd_2_mic_1 189 | section_00_0188.wav,section_00_target_test_normal_0044_car_C1_spd_1_mic_2 190 | section_00_0189.wav,section_00_source_test_normal_0018_car_B2_spd_2_mic_1 191 | section_00_0190.wav,section_00_source_test_normal_0039_car_A1_spd_3_mic_1 192 | section_00_0191.wav,section_00_source_test_normal_0035_car_B2_spd_2_mic_1 193 | section_00_0192.wav,section_00_target_test_normal_0026_car_C1_spd_1_mic_2 194 | section_00_0193.wav,section_00_target_test_normal_0030_car_C1_spd_1_mic_2 195 | section_00_0194.wav,section_00_source_test_normal_0046_car_A2_spd_3_mic_1 196 | section_00_0195.wav,section_00_source_test_anomaly_0046_car_B1_spd_2_mic_1 197 | section_00_0196.wav,section_00_target_test_normal_0019_car_C2_spd_1_mic_2 198 | section_00_0197.wav,section_00_source_test_anomaly_0042_car_B2_spd_2_mic_1 199 | section_00_0198.wav,section_00_target_test_anomaly_0013_car_C2_spd_1_mic_2 200 | section_00_0199.wav,section_00_target_test_normal_0042_car_C2_spd_1_mic_2 201 | -------------------------------------------------------------------------------- /dcase2023_task2_evaluator/ground_truth_attributes/ground_truth_ToyTank_section_00_test.csv: -------------------------------------------------------------------------------- 1 | section_00_0000.wav,section_00_target_test_normal_0022_car_C1_spd_1_mic_1 2 | section_00_0001.wav,section_00_target_test_anomaly_0032_car_C1_spd_5_mic_1 3 | section_00_0002.wav,section_00_source_test_anomaly_0016_car_A1_spd_4_mic_1 4 | section_00_0003.wav,section_00_target_test_anomaly_0036_car_C2_spd_5_mic_1 5 | section_00_0004.wav,section_00_source_test_normal_0003_car_A1_spd_4_mic_1 6 | section_00_0005.wav,section_00_source_test_anomaly_0028_car_B1_spd_3_mic_1 7 | section_00_0006.wav,section_00_source_test_normal_0048_car_A2_spd_3_mic_1 8 | section_00_0007.wav,section_00_target_test_anomaly_0004_car_C1_spd_1_mic_1 9 | section_00_0008.wav,section_00_source_test_normal_0013_car_B1_spd_3_mic_1 10 | section_00_0009.wav,section_00_target_test_anomaly_0044_car_C1_spd_5_mic_1 11 | section_00_0010.wav,section_00_source_test_anomaly_0036_car_B2_spd_3_mic_1 12 | section_00_0011.wav,section_00_target_test_anomaly_0022_car_C1_spd_1_mic_1 13 | section_00_0012.wav,section_00_target_test_normal_0045_car_C1_spd_5_mic_1 14 | section_00_0013.wav,section_00_source_test_normal_0010_car_A2_spd_4_mic_1 15 | section_00_0014.wav,section_00_source_test_anomaly_0002_car_A1_spd_4_mic_1 16 | section_00_0015.wav,section_00_target_test_normal_0013_car_C2_spd_1_mic_1 17 | section_00_0016.wav,section_00_source_test_normal_0023_car_A1_spd_3_mic_1 18 | section_00_0017.wav,section_00_source_test_anomaly_0035_car_B2_spd_3_mic_1 19 | section_00_0018.wav,section_00_target_test_normal_0036_car_C2_spd_1_mic_1 20 | section_00_0019.wav,section_00_source_test_anomaly_0004_car_A1_spd_4_mic_1 21 | section_00_0020.wav,section_00_source_test_normal_0041_car_B2_spd_3_mic_1 22 | section_00_0021.wav,section_00_target_test_normal_0011_car_C2_spd_1_mic_1 23 | section_00_0022.wav,section_00_target_test_normal_0029_car_C2_spd_1_mic_1 24 | section_00_0023.wav,section_00_target_test_anomaly_0034_car_C1_spd_5_mic_1 25 | section_00_0024.wav,section_00_source_test_anomaly_0050_car_B2_spd_3_mic_1 26 | section_00_0025.wav,section_00_target_test_anomaly_0002_car_C1_spd_1_mic_1 27 | section_00_0026.wav,section_00_source_test_anomaly_0009_car_A2_spd_4_mic_1 28 | section_00_0027.wav,section_00_source_test_normal_0011_car_B1_spd_3_mic_1 29 | section_00_0028.wav,section_00_target_test_normal_0005_car_C1_spd_1_mic_1 30 | section_00_0029.wav,section_00_source_test_normal_0024_car_A1_spd_3_mic_1 31 | section_00_0030.wav,section_00_source_test_normal_0009_car_A2_spd_4_mic_1 32 | section_00_0031.wav,section_00_source_test_anomaly_0023_car_A2_spd_4_mic_1 33 | section_00_0032.wav,section_00_target_test_anomaly_0047_car_C2_spd_5_mic_1 34 | section_00_0033.wav,section_00_target_test_normal_0021_car_C1_spd_1_mic_1 35 | section_00_0034.wav,section_00_target_test_anomaly_0008_car_C1_spd_1_mic_1 36 | section_00_0035.wav,section_00_source_test_normal_0043_car_B2_spd_3_mic_1 37 | section_00_0036.wav,section_00_source_test_normal_0029_car_B1_spd_4_mic_1 38 | section_00_0037.wav,section_00_source_test_anomaly_0008_car_A2_spd_4_mic_1 39 | section_00_0038.wav,section_00_target_test_anomaly_0049_car_C2_spd_5_mic_1 40 | section_00_0039.wav,section_00_target_test_anomaly_0018_car_C1_spd_1_mic_1 41 | section_00_0040.wav,section_00_source_test_anomaly_0010_car_A2_spd_4_mic_1 42 | section_00_0041.wav,section_00_source_test_anomaly_0044_car_B1_spd_3_mic_1 43 | section_00_0042.wav,section_00_target_test_anomaly_0024_car_C2_spd_1_mic_1 44 | section_00_0043.wav,section_00_target_test_anomaly_0015_car_C2_spd_1_mic_1 45 | section_00_0044.wav,section_00_source_test_anomaly_0043_car_B1_spd_3_mic_1 46 | section_00_0045.wav,section_00_target_test_anomaly_0010_car_C2_spd_1_mic_1 47 | section_00_0046.wav,section_00_target_test_normal_0003_car_C1_spd_1_mic_1 48 | section_00_0047.wav,section_00_target_test_normal_0017_car_C2_spd_1_mic_1 49 | section_00_0048.wav,section_00_source_test_normal_0035_car_B2_spd_4_mic_1 50 | section_00_0049.wav,section_00_source_test_anomaly_0014_car_A1_spd_4_mic_1 51 | section_00_0050.wav,section_00_target_test_anomaly_0025_car_C2_spd_1_mic_1 52 | section_00_0051.wav,section_00_target_test_normal_0047_car_C1_spd_5_mic_1 53 | section_00_0052.wav,section_00_source_test_normal_0004_car_A1_spd_4_mic_1 54 | section_00_0053.wav,section_00_source_test_anomaly_0040_car_B1_spd_3_mic_1 55 | section_00_0054.wav,section_00_target_test_normal_0049_car_C2_spd_5_mic_1 56 | section_00_0055.wav,section_00_source_test_anomaly_0038_car_B2_spd_3_mic_1 57 | section_00_0056.wav,section_00_source_test_anomaly_0018_car_A1_spd_4_mic_1 58 | section_00_0057.wav,section_00_target_test_normal_0034_car_C2_spd_1_mic_1 59 | section_00_0058.wav,section_00_source_test_normal_0020_car_B2_spd_3_mic_1 60 | section_00_0059.wav,section_00_source_test_anomaly_0042_car_B1_spd_3_mic_1 61 | section_00_0060.wav,section_00_target_test_anomaly_0026_car_C2_spd_1_mic_1 62 | section_00_0061.wav,section_00_target_test_normal_0030_car_C2_spd_1_mic_1 63 | section_00_0062.wav,section_00_target_test_normal_0037_car_C2_spd_1_mic_1 64 | section_00_0063.wav,section_00_source_test_normal_0036_car_A1_spd_4_mic_1 65 | section_00_0064.wav,section_00_target_test_anomaly_0020_car_C1_spd_1_mic_1 66 | section_00_0065.wav,section_00_source_test_anomaly_0017_car_A1_spd_4_mic_1 67 | section_00_0066.wav,section_00_target_test_anomaly_0021_car_C1_spd_1_mic_1 68 | section_00_0067.wav,section_00_source_test_normal_0047_car_A2_spd_3_mic_1 69 | section_00_0068.wav,section_00_target_test_normal_0012_car_C2_spd_1_mic_1 70 | section_00_0069.wav,section_00_source_test_anomaly_0041_car_B1_spd_3_mic_1 71 | section_00_0070.wav,section_00_source_test_anomaly_0025_car_A2_spd_4_mic_1 72 | section_00_0071.wav,section_00_source_test_anomaly_0027_car_B1_spd_3_mic_1 73 | section_00_0072.wav,section_00_source_test_normal_0038_car_A1_spd_4_mic_1 74 | section_00_0073.wav,section_00_target_test_anomaly_0017_car_C1_spd_1_mic_1 75 | section_00_0074.wav,section_00_source_test_normal_0045_car_B2_spd_3_mic_1 76 | section_00_0075.wav,section_00_target_test_normal_0009_car_C2_spd_1_mic_1 77 | section_00_0076.wav,section_00_source_test_anomaly_0046_car_B2_spd_3_mic_1 78 | section_00_0077.wav,section_00_source_test_normal_0017_car_B2_spd_3_mic_1 79 | section_00_0078.wav,section_00_source_test_anomaly_0012_car_A2_spd_4_mic_1 80 | section_00_0079.wav,section_00_target_test_anomaly_0033_car_C1_spd_5_mic_1 81 | section_00_0080.wav,section_00_target_test_anomaly_0006_car_C1_spd_1_mic_1 82 | section_00_0081.wav,section_00_source_test_normal_0033_car_B2_spd_4_mic_1 83 | section_00_0082.wav,section_00_source_test_anomaly_0015_car_A1_spd_4_mic_1 84 | section_00_0083.wav,section_00_target_test_anomaly_0030_car_C2_spd_1_mic_1 85 | section_00_0084.wav,section_00_target_test_anomaly_0050_car_C2_spd_5_mic_1 86 | section_00_0085.wav,section_00_target_test_normal_0006_car_C1_spd_1_mic_1 87 | section_00_0086.wav,section_00_target_test_anomaly_0007_car_C1_spd_1_mic_1 88 | section_00_0087.wav,section_00_source_test_anomaly_0029_car_B1_spd_3_mic_1 89 | section_00_0088.wav,section_00_target_test_normal_0016_car_C2_spd_1_mic_1 90 | section_00_0089.wav,section_00_target_test_normal_0032_car_C2_spd_1_mic_1 91 | section_00_0090.wav,section_00_source_test_anomaly_0049_car_B2_spd_3_mic_1 92 | section_00_0091.wav,section_00_target_test_normal_0035_car_C2_spd_1_mic_1 93 | section_00_0092.wav,section_00_target_test_normal_0010_car_C2_spd_1_mic_1 94 | section_00_0093.wav,section_00_target_test_normal_0002_car_C1_spd_1_mic_1 95 | section_00_0094.wav,section_00_target_test_anomaly_0011_car_C2_spd_1_mic_1 96 | section_00_0095.wav,section_00_target_test_normal_0020_car_C1_spd_1_mic_1 97 | section_00_0096.wav,section_00_source_test_anomaly_0007_car_A2_spd_4_mic_1 98 | section_00_0097.wav,section_00_source_test_normal_0012_car_B1_spd_3_mic_1 99 | section_00_0098.wav,section_00_target_test_normal_0050_car_C2_spd_5_mic_1 100 | section_00_0099.wav,section_00_target_test_anomaly_0048_car_C2_spd_5_mic_1 101 | section_00_0100.wav,section_00_target_test_anomaly_0046_car_C1_spd_5_mic_1 102 | section_00_0101.wav,section_00_source_test_anomaly_0031_car_B1_spd_3_mic_1 103 | section_00_0102.wav,section_00_source_test_normal_0040_car_A1_spd_4_mic_1 104 | section_00_0103.wav,section_00_source_test_anomaly_0024_car_A2_spd_4_mic_1 105 | section_00_0104.wav,section_00_source_test_normal_0050_car_A2_spd_3_mic_1 106 | section_00_0105.wav,section_00_target_test_normal_0042_car_C2_spd_5_mic_1 107 | section_00_0106.wav,section_00_source_test_normal_0032_car_B2_spd_4_mic_1 108 | section_00_0107.wav,section_00_source_test_normal_0002_car_A1_spd_4_mic_1 109 | section_00_0108.wav,section_00_source_test_anomaly_0020_car_A2_spd_4_mic_1 110 | section_00_0109.wav,section_00_source_test_anomaly_0003_car_A1_spd_4_mic_1 111 | section_00_0110.wav,section_00_source_test_anomaly_0047_car_B2_spd_3_mic_1 112 | section_00_0111.wav,section_00_source_test_normal_0028_car_B1_spd_4_mic_1 113 | section_00_0112.wav,section_00_source_test_normal_0026_car_B1_spd_4_mic_1 114 | section_00_0113.wav,section_00_target_test_normal_0044_car_C2_spd_5_mic_1 115 | section_00_0114.wav,section_00_source_test_anomaly_0026_car_B1_spd_3_mic_1 116 | section_00_0115.wav,section_00_source_test_anomaly_0021_car_A2_spd_4_mic_1 117 | section_00_0116.wav,section_00_target_test_normal_0031_car_C2_spd_1_mic_1 118 | section_00_0117.wav,section_00_target_test_normal_0018_car_C1_spd_1_mic_1 119 | section_00_0118.wav,section_00_source_test_anomaly_0022_car_A2_spd_4_mic_1 120 | section_00_0119.wav,section_00_target_test_normal_0038_car_C2_spd_1_mic_1 121 | section_00_0120.wav,section_00_source_test_normal_0014_car_B1_spd_3_mic_1 122 | section_00_0121.wav,section_00_target_test_normal_0023_car_C1_spd_1_mic_1 123 | section_00_0122.wav,section_00_source_test_anomaly_0005_car_A1_spd_4_mic_1 124 | section_00_0123.wav,section_00_target_test_normal_0043_car_C2_spd_5_mic_1 125 | section_00_0124.wav,section_00_source_test_anomaly_0034_car_B2_spd_3_mic_1 126 | section_00_0125.wav,section_00_target_test_normal_0019_car_C1_spd_1_mic_1 127 | section_00_0126.wav,section_00_target_test_normal_0028_car_C2_spd_1_mic_1 128 | section_00_0127.wav,section_00_target_test_anomaly_0031_car_C2_spd_1_mic_1 129 | section_00_0128.wav,section_00_target_test_normal_0025_car_C1_spd_1_mic_1 130 | section_00_0129.wav,section_00_source_test_anomaly_0048_car_B2_spd_3_mic_1 131 | section_00_0130.wav,section_00_target_test_anomaly_0027_car_C2_spd_1_mic_1 132 | section_00_0131.wav,section_00_target_test_anomaly_0009_car_C2_spd_1_mic_1 133 | section_00_0132.wav,section_00_source_test_normal_0022_car_A1_spd_3_mic_1 134 | section_00_0133.wav,section_00_source_test_normal_0037_car_A1_spd_4_mic_1 135 | section_00_0134.wav,section_00_source_test_normal_0005_car_A1_spd_4_mic_1 136 | section_00_0135.wav,section_00_source_test_anomaly_0011_car_A2_spd_4_mic_1 137 | section_00_0136.wav,section_00_target_test_normal_0027_car_C2_spd_1_mic_1 138 | section_00_0137.wav,section_00_target_test_anomaly_0040_car_C2_spd_5_mic_1 139 | section_00_0138.wav,section_00_source_test_anomaly_0039_car_B1_spd_3_mic_1 140 | section_00_0139.wav,section_00_target_test_normal_0026_car_C1_spd_1_mic_1 141 | section_00_0140.wav,section_00_source_test_normal_0049_car_A2_spd_3_mic_1 142 | section_00_0141.wav,section_00_source_test_normal_0034_car_B2_spd_4_mic_1 143 | section_00_0142.wav,section_00_target_test_anomaly_0037_car_C2_spd_5_mic_1 144 | section_00_0143.wav,section_00_target_test_normal_0046_car_C1_spd_5_mic_1 145 | section_00_0144.wav,section_00_target_test_normal_0039_car_C1_spd_5_mic_1 146 | section_00_0145.wav,section_00_target_test_anomaly_0023_car_C1_spd_1_mic_1 147 | section_00_0146.wav,section_00_target_test_normal_0040_car_C1_spd_5_mic_1 148 | section_00_0147.wav,section_00_target_test_anomaly_0042_car_C2_spd_5_mic_1 149 | section_00_0148.wav,section_00_source_test_anomaly_0045_car_B2_spd_3_mic_1 150 | section_00_0149.wav,section_00_source_test_normal_0018_car_B2_spd_3_mic_1 151 | section_00_0150.wav,section_00_source_test_normal_0039_car_A1_spd_4_mic_1 152 | section_00_0151.wav,section_00_source_test_anomaly_0037_car_B2_spd_3_mic_1 153 | section_00_0152.wav,section_00_source_test_normal_0006_car_A2_spd_4_mic_1 154 | section_00_0153.wav,section_00_target_test_normal_0008_car_C2_spd_1_mic_1 155 | section_00_0154.wav,section_00_target_test_anomaly_0014_car_C2_spd_1_mic_1 156 | section_00_0155.wav,section_00_source_test_normal_0001_car_A1_spd_4_mic_1 157 | section_00_0156.wav,section_00_target_test_normal_0001_car_C1_spd_1_mic_1 158 | section_00_0157.wav,section_00_target_test_normal_0041_car_C1_spd_5_mic_1 159 | section_00_0158.wav,section_00_source_test_normal_0021_car_A1_spd_3_mic_1 160 | section_00_0159.wav,section_00_target_test_anomaly_0003_car_C1_spd_1_mic_1 161 | section_00_0160.wav,section_00_source_test_anomaly_0001_car_A1_spd_4_mic_1 162 | section_00_0161.wav,section_00_target_test_normal_0004_car_C1_spd_1_mic_1 163 | section_00_0162.wav,section_00_target_test_anomaly_0035_car_C1_spd_5_mic_1 164 | section_00_0163.wav,section_00_target_test_anomaly_0016_car_C1_spd_1_mic_1 165 | section_00_0164.wav,section_00_target_test_anomaly_0001_car_C1_spd_1_mic_1 166 | section_00_0165.wav,section_00_source_test_normal_0042_car_B2_spd_3_mic_1 167 | section_00_0166.wav,section_00_source_test_anomaly_0006_car_A1_spd_4_mic_1 168 | section_00_0167.wav,section_00_target_test_normal_0048_car_C2_spd_5_mic_1 169 | section_00_0168.wav,section_00_source_test_normal_0016_car_B2_spd_3_mic_1 170 | section_00_0169.wav,section_00_target_test_anomaly_0039_car_C2_spd_5_mic_1 171 | section_00_0170.wav,section_00_target_test_anomaly_0041_car_C2_spd_5_mic_1 172 | section_00_0171.wav,section_00_target_test_anomaly_0038_car_C2_spd_5_mic_1 173 | section_00_0172.wav,section_00_target_test_anomaly_0012_car_C2_spd_1_mic_1 174 | section_00_0173.wav,section_00_source_test_normal_0046_car_A2_spd_3_mic_1 175 | section_00_0174.wav,section_00_target_test_normal_0014_car_C2_spd_1_mic_1 176 | section_00_0175.wav,section_00_source_test_anomaly_0013_car_A2_spd_4_mic_1 177 | section_00_0176.wav,section_00_source_test_normal_0015_car_B1_spd_3_mic_1 178 | section_00_0177.wav,section_00_target_test_anomaly_0019_car_C1_spd_1_mic_1 179 | section_00_0178.wav,section_00_target_test_anomaly_0045_car_C1_spd_5_mic_1 180 | section_00_0179.wav,section_00_source_test_normal_0030_car_B1_spd_4_mic_1 181 | section_00_0180.wav,section_00_target_test_normal_0015_car_C2_spd_1_mic_1 182 | section_00_0181.wav,section_00_target_test_anomaly_0005_car_C1_spd_1_mic_1 183 | section_00_0182.wav,section_00_target_test_normal_0033_car_C2_spd_1_mic_1 184 | section_00_0183.wav,section_00_target_test_normal_0024_car_C1_spd_1_mic_1 185 | section_00_0184.wav,section_00_source_test_anomaly_0032_car_B2_spd_3_mic_1 186 | section_00_0185.wav,section_00_source_test_normal_0019_car_B2_spd_3_mic_1 187 | section_00_0186.wav,section_00_source_test_anomaly_0033_car_B2_spd_3_mic_1 188 | section_00_0187.wav,section_00_source_test_normal_0007_car_A2_spd_4_mic_1 189 | section_00_0188.wav,section_00_source_test_normal_0031_car_B2_spd_4_mic_1 190 | section_00_0189.wav,section_00_target_test_anomaly_0028_car_C2_spd_1_mic_1 191 | section_00_0190.wav,section_00_target_test_anomaly_0043_car_C1_spd_5_mic_1 192 | section_00_0191.wav,section_00_source_test_normal_0025_car_A1_spd_3_mic_1 193 | section_00_0192.wav,section_00_target_test_anomaly_0029_car_C2_spd_1_mic_1 194 | section_00_0193.wav,section_00_source_test_normal_0044_car_B2_spd_3_mic_1 195 | section_00_0194.wav,section_00_target_test_anomaly_0013_car_C2_spd_1_mic_1 196 | section_00_0195.wav,section_00_target_test_normal_0007_car_C1_spd_1_mic_1 197 | section_00_0196.wav,section_00_source_test_normal_0008_car_A2_spd_4_mic_1 198 | section_00_0197.wav,section_00_source_test_anomaly_0019_car_A1_spd_4_mic_1 199 | section_00_0198.wav,section_00_source_test_normal_0027_car_B1_spd_4_mic_1 200 | section_00_0199.wav,section_00_source_test_anomaly_0030_car_B1_spd_3_mic_1 201 | --------------------------------------------------------------------------------