├── .github └── workflows │ ├── lint.yml │ ├── publish-pypi.yml │ └── test.yml ├── .gitignore ├── .pre-commit-config.yaml ├── DEVELOPMENT.md ├── LICENSE ├── MANIFEST.in ├── README.md ├── apps ├── realtime-asr │ ├── audio_capture.py │ ├── config.py │ ├── requirements.txt │ ├── stream_asr.py │ └── test_with_file.py └── streamlit │ ├── __init__.py │ ├── app.py │ ├── audio_processing.py │ ├── config.py │ ├── requirements.txt │ ├── transcription.py │ ├── ui_components.py │ └── utils.py ├── chunkformer ├── __init__.py ├── bin │ ├── __init__.py │ ├── alignment.py │ ├── average_model.py │ ├── classify.py │ ├── recognize.py │ └── train.py ├── chunkformer_model.py ├── dataset │ ├── __init__.py │ ├── datapipes.py │ ├── dataset.py │ ├── kaldi_io.py │ ├── processor.py │ └── wav_distortion.py ├── modules │ ├── __init__.py │ ├── asr_model.py │ ├── attention.py │ ├── classification_model.py │ ├── cmvn.py │ ├── convolution.py │ ├── ctc.py │ ├── decoder.py │ ├── decoder_layer.py │ ├── embedding.py │ ├── encoder.py │ ├── encoder_layer.py │ ├── label_smoothing_loss.py │ ├── norm.py │ ├── positionwise_feed_forward.py │ ├── search.py │ ├── subsampling.py │ └── swish.py ├── text │ ├── __init__.py │ ├── base_tokenizer.py │ ├── bpe_tokenizer.py │ ├── char_tokenizer.py │ └── tokenize_utils.py ├── transducer │ ├── __init__.py │ ├── joint.py │ ├── predictor.py │ ├── search │ │ ├── __init__.py │ │ ├── greedy_search.py │ │ └── prefix_beam_search.py │ └── transducer.py └── utils │ ├── __init__.py │ ├── checkpoint.py │ ├── class_utils.py │ ├── cmvn.py │ ├── common.py │ ├── config.py │ ├── context_graph.py │ ├── executor.py │ ├── file_utils.py │ ├── fsdp_utils.py │ ├── init_dataset.py │ ├── init_model.py │ ├── init_tokenizer.py │ ├── mask.py │ ├── model_utils.py │ ├── scheduler.py │ └── train_utils.py ├── docs ├── chunkformer_architecture.png └── paper.pdf ├── examples ├── README.md ├── asr │ ├── README.md │ ├── ctc │ │ ├── chunkformer │ │ ├── conf │ │ │ ├── chunkformer-ctc-small-libri-100h.yaml │ │ │ ├── chunkformer-ctc-small-libri-960h-stream-dct.yaml │ │ │ └── chunkformer-ctc-small-libri-960h.yaml │ │ ├── path.sh │ │ ├── run.sh │ │ └── tools │ └── rnnt │ │ ├── chunkformer │ │ ├── conf │ │ ├── chunkformer-rnnt-large-vie.yaml │ │ └── chunkformer-rnnt-small-libri-100h.yaml │ │ ├── path.sh │ │ ├── run.sh │ │ └── tools └── classification │ ├── README.md │ ├── RESULTS.md │ ├── conf │ ├── multi_task.yaml │ └── single_task.yaml │ ├── path.sh │ └── run.sh ├── pyproject.toml ├── samples ├── audios │ ├── audio_1.wav │ ├── audio_2.wav │ ├── audio_3.wav │ ├── audio_4.wav │ └── audio_5.wav └── data.tsv ├── setup.py ├── tests ├── test_classification_output.py ├── test_wer_ctc_performance.py └── test_wer_rnnt_performance.py └── tools ├── compute-wer.py ├── compute_classification_metrics.py ├── compute_cmvn_stats.py ├── compute_label_stats.py ├── convert_text_labels_to_int.py ├── fst ├── add_lex_disambig.pl ├── compile_lexicon_token_fst.sh ├── ctc_token_fst.py ├── ctc_token_fst_compact.py ├── ctc_token_fst_corrected.py ├── eps2disambig.pl ├── make_lexicon_fst.pl ├── make_tlg.sh ├── prepare_dict.py ├── remove_oovs.pl ├── rnnt_token_fst.py └── s2eps.pl ├── parse_options.sh ├── push_model_hf.py ├── split_train_test.py ├── spm_encode.py ├── spm_train.py └── tsv_to_list.py /.github/workflows/lint.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/.github/workflows/lint.yml -------------------------------------------------------------------------------- /.github/workflows/publish-pypi.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/.github/workflows/publish-pypi.yml -------------------------------------------------------------------------------- /.github/workflows/test.yml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/.github/workflows/test.yml -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/.gitignore -------------------------------------------------------------------------------- /.pre-commit-config.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/.pre-commit-config.yaml -------------------------------------------------------------------------------- /DEVELOPMENT.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/DEVELOPMENT.md -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/LICENSE -------------------------------------------------------------------------------- /MANIFEST.in: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/MANIFEST.in -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/README.md -------------------------------------------------------------------------------- /apps/realtime-asr/audio_capture.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/realtime-asr/audio_capture.py -------------------------------------------------------------------------------- /apps/realtime-asr/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/realtime-asr/config.py -------------------------------------------------------------------------------- /apps/realtime-asr/requirements.txt: -------------------------------------------------------------------------------- 1 | pyaudio>=0.2.14 2 | sounddevice>=0.5.3 3 | -------------------------------------------------------------------------------- /apps/realtime-asr/stream_asr.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/realtime-asr/stream_asr.py -------------------------------------------------------------------------------- /apps/realtime-asr/test_with_file.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/realtime-asr/test_with_file.py -------------------------------------------------------------------------------- /apps/streamlit/__init__.py: -------------------------------------------------------------------------------- 1 | """ 2 | Chunkformer Streamlit App Package 3 | """ 4 | 5 | __version__ = "0.1.0" 6 | -------------------------------------------------------------------------------- /apps/streamlit/app.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/streamlit/app.py -------------------------------------------------------------------------------- /apps/streamlit/audio_processing.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/streamlit/audio_processing.py -------------------------------------------------------------------------------- /apps/streamlit/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/streamlit/config.py -------------------------------------------------------------------------------- /apps/streamlit/requirements.txt: -------------------------------------------------------------------------------- 1 | streamlit>=1.32.0 2 | pydub>=0.25.0 3 | -------------------------------------------------------------------------------- /apps/streamlit/transcription.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/streamlit/transcription.py -------------------------------------------------------------------------------- /apps/streamlit/ui_components.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/streamlit/ui_components.py -------------------------------------------------------------------------------- /apps/streamlit/utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/apps/streamlit/utils.py -------------------------------------------------------------------------------- /chunkformer/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/__init__.py -------------------------------------------------------------------------------- /chunkformer/bin/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /chunkformer/bin/alignment.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/bin/alignment.py -------------------------------------------------------------------------------- /chunkformer/bin/average_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/bin/average_model.py -------------------------------------------------------------------------------- /chunkformer/bin/classify.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/bin/classify.py -------------------------------------------------------------------------------- /chunkformer/bin/recognize.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/bin/recognize.py -------------------------------------------------------------------------------- /chunkformer/bin/train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/bin/train.py -------------------------------------------------------------------------------- /chunkformer/chunkformer_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/chunkformer_model.py -------------------------------------------------------------------------------- /chunkformer/dataset/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /chunkformer/dataset/datapipes.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/dataset/datapipes.py -------------------------------------------------------------------------------- /chunkformer/dataset/dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/dataset/dataset.py -------------------------------------------------------------------------------- /chunkformer/dataset/kaldi_io.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/dataset/kaldi_io.py -------------------------------------------------------------------------------- /chunkformer/dataset/processor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/dataset/processor.py -------------------------------------------------------------------------------- /chunkformer/dataset/wav_distortion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/dataset/wav_distortion.py -------------------------------------------------------------------------------- /chunkformer/modules/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /chunkformer/modules/asr_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/asr_model.py -------------------------------------------------------------------------------- /chunkformer/modules/attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/attention.py -------------------------------------------------------------------------------- /chunkformer/modules/classification_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/classification_model.py -------------------------------------------------------------------------------- /chunkformer/modules/cmvn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/cmvn.py -------------------------------------------------------------------------------- /chunkformer/modules/convolution.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/convolution.py -------------------------------------------------------------------------------- /chunkformer/modules/ctc.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/ctc.py -------------------------------------------------------------------------------- /chunkformer/modules/decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/decoder.py -------------------------------------------------------------------------------- /chunkformer/modules/decoder_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/decoder_layer.py -------------------------------------------------------------------------------- /chunkformer/modules/embedding.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/embedding.py -------------------------------------------------------------------------------- /chunkformer/modules/encoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/encoder.py -------------------------------------------------------------------------------- /chunkformer/modules/encoder_layer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/encoder_layer.py -------------------------------------------------------------------------------- /chunkformer/modules/label_smoothing_loss.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/label_smoothing_loss.py -------------------------------------------------------------------------------- /chunkformer/modules/norm.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/norm.py -------------------------------------------------------------------------------- /chunkformer/modules/positionwise_feed_forward.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/positionwise_feed_forward.py -------------------------------------------------------------------------------- /chunkformer/modules/search.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/search.py -------------------------------------------------------------------------------- /chunkformer/modules/subsampling.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/subsampling.py -------------------------------------------------------------------------------- /chunkformer/modules/swish.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/modules/swish.py -------------------------------------------------------------------------------- /chunkformer/text/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /chunkformer/text/base_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/text/base_tokenizer.py -------------------------------------------------------------------------------- /chunkformer/text/bpe_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/text/bpe_tokenizer.py -------------------------------------------------------------------------------- /chunkformer/text/char_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/text/char_tokenizer.py -------------------------------------------------------------------------------- /chunkformer/text/tokenize_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/text/tokenize_utils.py -------------------------------------------------------------------------------- /chunkformer/transducer/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /chunkformer/transducer/joint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/transducer/joint.py -------------------------------------------------------------------------------- /chunkformer/transducer/predictor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/transducer/predictor.py -------------------------------------------------------------------------------- /chunkformer/transducer/search/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /chunkformer/transducer/search/greedy_search.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/transducer/search/greedy_search.py -------------------------------------------------------------------------------- /chunkformer/transducer/search/prefix_beam_search.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/transducer/search/prefix_beam_search.py -------------------------------------------------------------------------------- /chunkformer/transducer/transducer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/transducer/transducer.py -------------------------------------------------------------------------------- /chunkformer/utils/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /chunkformer/utils/checkpoint.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/checkpoint.py -------------------------------------------------------------------------------- /chunkformer/utils/class_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/class_utils.py -------------------------------------------------------------------------------- /chunkformer/utils/cmvn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/cmvn.py -------------------------------------------------------------------------------- /chunkformer/utils/common.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/common.py -------------------------------------------------------------------------------- /chunkformer/utils/config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/config.py -------------------------------------------------------------------------------- /chunkformer/utils/context_graph.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/context_graph.py -------------------------------------------------------------------------------- /chunkformer/utils/executor.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/executor.py -------------------------------------------------------------------------------- /chunkformer/utils/file_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/file_utils.py -------------------------------------------------------------------------------- /chunkformer/utils/fsdp_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/fsdp_utils.py -------------------------------------------------------------------------------- /chunkformer/utils/init_dataset.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/init_dataset.py -------------------------------------------------------------------------------- /chunkformer/utils/init_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/init_model.py -------------------------------------------------------------------------------- /chunkformer/utils/init_tokenizer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/init_tokenizer.py -------------------------------------------------------------------------------- /chunkformer/utils/mask.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/mask.py -------------------------------------------------------------------------------- /chunkformer/utils/model_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/model_utils.py -------------------------------------------------------------------------------- /chunkformer/utils/scheduler.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/scheduler.py -------------------------------------------------------------------------------- /chunkformer/utils/train_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/chunkformer/utils/train_utils.py -------------------------------------------------------------------------------- /docs/chunkformer_architecture.png: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/docs/chunkformer_architecture.png -------------------------------------------------------------------------------- /docs/paper.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/docs/paper.pdf -------------------------------------------------------------------------------- /examples/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/README.md -------------------------------------------------------------------------------- /examples/asr/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/README.md -------------------------------------------------------------------------------- /examples/asr/ctc/chunkformer: -------------------------------------------------------------------------------- 1 | ../../../chunkformer -------------------------------------------------------------------------------- /examples/asr/ctc/conf/chunkformer-ctc-small-libri-100h.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/ctc/conf/chunkformer-ctc-small-libri-100h.yaml -------------------------------------------------------------------------------- /examples/asr/ctc/conf/chunkformer-ctc-small-libri-960h-stream-dct.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/ctc/conf/chunkformer-ctc-small-libri-960h-stream-dct.yaml -------------------------------------------------------------------------------- /examples/asr/ctc/conf/chunkformer-ctc-small-libri-960h.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/ctc/conf/chunkformer-ctc-small-libri-960h.yaml -------------------------------------------------------------------------------- /examples/asr/ctc/path.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/ctc/path.sh -------------------------------------------------------------------------------- /examples/asr/ctc/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/ctc/run.sh -------------------------------------------------------------------------------- /examples/asr/ctc/tools: -------------------------------------------------------------------------------- 1 | ../../../tools -------------------------------------------------------------------------------- /examples/asr/rnnt/chunkformer: -------------------------------------------------------------------------------- 1 | ../../../chunkformer -------------------------------------------------------------------------------- /examples/asr/rnnt/conf/chunkformer-rnnt-large-vie.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/rnnt/conf/chunkformer-rnnt-large-vie.yaml -------------------------------------------------------------------------------- /examples/asr/rnnt/conf/chunkformer-rnnt-small-libri-100h.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/rnnt/conf/chunkformer-rnnt-small-libri-100h.yaml -------------------------------------------------------------------------------- /examples/asr/rnnt/path.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/rnnt/path.sh -------------------------------------------------------------------------------- /examples/asr/rnnt/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/asr/rnnt/run.sh -------------------------------------------------------------------------------- /examples/asr/rnnt/tools: -------------------------------------------------------------------------------- 1 | ../../../tools -------------------------------------------------------------------------------- /examples/classification/README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/classification/README.md -------------------------------------------------------------------------------- /examples/classification/RESULTS.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/classification/RESULTS.md -------------------------------------------------------------------------------- /examples/classification/conf/multi_task.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/classification/conf/multi_task.yaml -------------------------------------------------------------------------------- /examples/classification/conf/single_task.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/classification/conf/single_task.yaml -------------------------------------------------------------------------------- /examples/classification/path.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/classification/path.sh -------------------------------------------------------------------------------- /examples/classification/run.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/examples/classification/run.sh -------------------------------------------------------------------------------- /pyproject.toml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/pyproject.toml -------------------------------------------------------------------------------- /samples/audios/audio_1.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/samples/audios/audio_1.wav -------------------------------------------------------------------------------- /samples/audios/audio_2.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/samples/audios/audio_2.wav -------------------------------------------------------------------------------- /samples/audios/audio_3.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/samples/audios/audio_3.wav -------------------------------------------------------------------------------- /samples/audios/audio_4.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/samples/audios/audio_4.wav -------------------------------------------------------------------------------- /samples/audios/audio_5.wav: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/samples/audios/audio_5.wav -------------------------------------------------------------------------------- /samples/data.tsv: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/samples/data.tsv -------------------------------------------------------------------------------- /setup.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/setup.py -------------------------------------------------------------------------------- /tests/test_classification_output.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tests/test_classification_output.py -------------------------------------------------------------------------------- /tests/test_wer_ctc_performance.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tests/test_wer_ctc_performance.py -------------------------------------------------------------------------------- /tests/test_wer_rnnt_performance.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tests/test_wer_rnnt_performance.py -------------------------------------------------------------------------------- /tools/compute-wer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/compute-wer.py -------------------------------------------------------------------------------- /tools/compute_classification_metrics.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/compute_classification_metrics.py -------------------------------------------------------------------------------- /tools/compute_cmvn_stats.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/compute_cmvn_stats.py -------------------------------------------------------------------------------- /tools/compute_label_stats.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/compute_label_stats.py -------------------------------------------------------------------------------- /tools/convert_text_labels_to_int.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/convert_text_labels_to_int.py -------------------------------------------------------------------------------- /tools/fst/add_lex_disambig.pl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/add_lex_disambig.pl -------------------------------------------------------------------------------- /tools/fst/compile_lexicon_token_fst.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/compile_lexicon_token_fst.sh -------------------------------------------------------------------------------- /tools/fst/ctc_token_fst.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/ctc_token_fst.py -------------------------------------------------------------------------------- /tools/fst/ctc_token_fst_compact.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/ctc_token_fst_compact.py -------------------------------------------------------------------------------- /tools/fst/ctc_token_fst_corrected.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/ctc_token_fst_corrected.py -------------------------------------------------------------------------------- /tools/fst/eps2disambig.pl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/eps2disambig.pl -------------------------------------------------------------------------------- /tools/fst/make_lexicon_fst.pl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/make_lexicon_fst.pl -------------------------------------------------------------------------------- /tools/fst/make_tlg.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/make_tlg.sh -------------------------------------------------------------------------------- /tools/fst/prepare_dict.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/prepare_dict.py -------------------------------------------------------------------------------- /tools/fst/remove_oovs.pl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/remove_oovs.pl -------------------------------------------------------------------------------- /tools/fst/rnnt_token_fst.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/rnnt_token_fst.py -------------------------------------------------------------------------------- /tools/fst/s2eps.pl: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/fst/s2eps.pl -------------------------------------------------------------------------------- /tools/parse_options.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/parse_options.sh -------------------------------------------------------------------------------- /tools/push_model_hf.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/push_model_hf.py -------------------------------------------------------------------------------- /tools/split_train_test.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/split_train_test.py -------------------------------------------------------------------------------- /tools/spm_encode.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/spm_encode.py -------------------------------------------------------------------------------- /tools/spm_train.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/spm_train.py -------------------------------------------------------------------------------- /tools/tsv_to_list.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/khanld/chunkformer/HEAD/tools/tsv_to_list.py --------------------------------------------------------------------------------