├── .gitignore ├── .gitmodules ├── DATA ├── __init__.py ├── create_distillation_tsv.py ├── data_utils.py └── mustc │ ├── __init__.py │ ├── download.sh │ ├── get_data_mt.sh │ ├── get_mustc.sh │ ├── mustc_noise.list │ ├── prep_mustc_data.py │ └── seg_mustc_data.py ├── README.md ├── codebase ├── __init__.py ├── agents │ ├── cif_agent.py │ └── default_agent.py ├── criterion │ ├── __init__.py │ ├── best_alignment │ │ ├── .gitignore │ │ ├── LICENSE │ │ ├── __init__.py │ │ ├── best_alignment.cpp │ │ └── best_alignment.cu │ ├── cif_criterion.py │ ├── joint_ctc_criterion.py │ ├── mma_criterion.py │ ├── rnnt_criterion.py │ └── ssnt_criterion.py ├── models │ ├── __init__.py │ ├── cif_transformer.py │ ├── mma_model.py │ ├── mt_transformer.py │ ├── s2t_emformer.py │ ├── s2t_transformer.py │ ├── torchaudio_models │ │ ├── __init__.py │ │ ├── conformer.py │ │ ├── conv_tasnet.py │ │ ├── deepspeech.py │ │ ├── emformer.py │ │ ├── rnnt.py │ │ ├── rnnt_decoder.py │ │ ├── tacotron2.py │ │ ├── wav2letter.py │ │ ├── wav2vec2 │ │ │ ├── __init__.py │ │ │ ├── components.py │ │ │ ├── model.py │ │ │ └── utils │ │ │ │ ├── __init__.py │ │ │ │ ├── import_fairseq.py │ │ │ │ └── import_huggingface.py │ │ └── wavernn.py │ └── transducer_model.py ├── modules │ ├── __init__.py │ ├── causal_conv.py │ ├── fixed_pre_decision.py │ └── monotonic_multihead_attention.py ├── tasks │ ├── __init__.py │ ├── inference_config.py │ ├── speech_to_text_infer.py │ └── translation_infer.py └── utils │ ├── __init__.py │ ├── functions.py │ ├── monotonic_attention.py │ └── p_choose_strategy.py ├── docs ├── apex_installation.md ├── cif.md ├── data_preparation.md ├── mma.md ├── simuleval_instruction.md └── waitk.md ├── eval ├── 0-gen_simul_list.sh ├── 1-simuleval.sh ├── eval_asr.sh ├── eval_cif.sh ├── eval_mt.sh ├── eval_st.sh ├── eval_transducer.sh ├── generate.py ├── interactive.py ├── latency-bleu.ipynb ├── run_all_simuleval.sh └── visualize.ipynb ├── exp ├── 0-mt.sh ├── 0a-decode_distillation.sh ├── 0b-create_distillation_set.sh ├── 1a-pretrain_asr.sh ├── 1b-fintune_asr_on_es.sh ├── 2-mma.sh ├── 2b-mma_finetune.sh ├── 3-cif.sh ├── 3b-cif_finetune.sh ├── data_path.sh ├── infer_asr.yaml ├── infer_mt.yaml └── infer_st.yaml ├── requirements.txt └── scripts ├── average_checkpoints.py ├── migrate_data_path.sh └── update_config.py /.gitignore: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/.gitignore -------------------------------------------------------------------------------- /.gitmodules: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/.gitmodules -------------------------------------------------------------------------------- /DATA/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /DATA/create_distillation_tsv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/DATA/create_distillation_tsv.py -------------------------------------------------------------------------------- /DATA/data_utils.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/DATA/data_utils.py -------------------------------------------------------------------------------- /DATA/mustc/__init__.py: -------------------------------------------------------------------------------- 1 | -------------------------------------------------------------------------------- /DATA/mustc/download.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/DATA/mustc/download.sh -------------------------------------------------------------------------------- /DATA/mustc/get_data_mt.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/DATA/mustc/get_data_mt.sh -------------------------------------------------------------------------------- /DATA/mustc/get_mustc.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/DATA/mustc/get_mustc.sh -------------------------------------------------------------------------------- /DATA/mustc/mustc_noise.list: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/DATA/mustc/mustc_noise.list -------------------------------------------------------------------------------- /DATA/mustc/prep_mustc_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/DATA/mustc/prep_mustc_data.py -------------------------------------------------------------------------------- /DATA/mustc/seg_mustc_data.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/DATA/mustc/seg_mustc_data.py -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/README.md -------------------------------------------------------------------------------- /codebase/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/__init__.py -------------------------------------------------------------------------------- /codebase/agents/cif_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/agents/cif_agent.py -------------------------------------------------------------------------------- /codebase/agents/default_agent.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/agents/default_agent.py -------------------------------------------------------------------------------- /codebase/criterion/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/__init__.py -------------------------------------------------------------------------------- /codebase/criterion/best_alignment/.gitignore: -------------------------------------------------------------------------------- 1 | build/ -------------------------------------------------------------------------------- /codebase/criterion/best_alignment/LICENSE: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/best_alignment/LICENSE -------------------------------------------------------------------------------- /codebase/criterion/best_alignment/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/best_alignment/__init__.py -------------------------------------------------------------------------------- /codebase/criterion/best_alignment/best_alignment.cpp: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/best_alignment/best_alignment.cpp -------------------------------------------------------------------------------- /codebase/criterion/best_alignment/best_alignment.cu: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/best_alignment/best_alignment.cu -------------------------------------------------------------------------------- /codebase/criterion/cif_criterion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/cif_criterion.py -------------------------------------------------------------------------------- /codebase/criterion/joint_ctc_criterion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/joint_ctc_criterion.py -------------------------------------------------------------------------------- /codebase/criterion/mma_criterion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/mma_criterion.py -------------------------------------------------------------------------------- /codebase/criterion/rnnt_criterion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/rnnt_criterion.py -------------------------------------------------------------------------------- /codebase/criterion/ssnt_criterion.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/criterion/ssnt_criterion.py -------------------------------------------------------------------------------- /codebase/models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/__init__.py -------------------------------------------------------------------------------- /codebase/models/cif_transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/cif_transformer.py -------------------------------------------------------------------------------- /codebase/models/mma_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/mma_model.py -------------------------------------------------------------------------------- /codebase/models/mt_transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/mt_transformer.py -------------------------------------------------------------------------------- /codebase/models/s2t_emformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/s2t_emformer.py -------------------------------------------------------------------------------- /codebase/models/s2t_transformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/s2t_transformer.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/__init__.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/conformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/conformer.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/conv_tasnet.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/conv_tasnet.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/deepspeech.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/deepspeech.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/emformer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/emformer.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/rnnt.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/rnnt.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/rnnt_decoder.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/rnnt_decoder.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/tacotron2.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/tacotron2.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/wav2letter.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/wav2letter.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/wav2vec2/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/wav2vec2/__init__.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/wav2vec2/components.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/wav2vec2/components.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/wav2vec2/model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/wav2vec2/model.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/wav2vec2/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/wav2vec2/utils/__init__.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/wav2vec2/utils/import_fairseq.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/wav2vec2/utils/import_fairseq.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/wav2vec2/utils/import_huggingface.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/wav2vec2/utils/import_huggingface.py -------------------------------------------------------------------------------- /codebase/models/torchaudio_models/wavernn.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/torchaudio_models/wavernn.py -------------------------------------------------------------------------------- /codebase/models/transducer_model.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/models/transducer_model.py -------------------------------------------------------------------------------- /codebase/modules/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/modules/__init__.py -------------------------------------------------------------------------------- /codebase/modules/causal_conv.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/modules/causal_conv.py -------------------------------------------------------------------------------- /codebase/modules/fixed_pre_decision.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/modules/fixed_pre_decision.py -------------------------------------------------------------------------------- /codebase/modules/monotonic_multihead_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/modules/monotonic_multihead_attention.py -------------------------------------------------------------------------------- /codebase/tasks/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/tasks/__init__.py -------------------------------------------------------------------------------- /codebase/tasks/inference_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/tasks/inference_config.py -------------------------------------------------------------------------------- /codebase/tasks/speech_to_text_infer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/tasks/speech_to_text_infer.py -------------------------------------------------------------------------------- /codebase/tasks/translation_infer.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/tasks/translation_infer.py -------------------------------------------------------------------------------- /codebase/utils/__init__.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/utils/__init__.py -------------------------------------------------------------------------------- /codebase/utils/functions.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/utils/functions.py -------------------------------------------------------------------------------- /codebase/utils/monotonic_attention.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/utils/monotonic_attention.py -------------------------------------------------------------------------------- /codebase/utils/p_choose_strategy.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/codebase/utils/p_choose_strategy.py -------------------------------------------------------------------------------- /docs/apex_installation.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/docs/apex_installation.md -------------------------------------------------------------------------------- /docs/cif.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/docs/cif.md -------------------------------------------------------------------------------- /docs/data_preparation.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/docs/data_preparation.md -------------------------------------------------------------------------------- /docs/mma.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/docs/mma.md -------------------------------------------------------------------------------- /docs/simuleval_instruction.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/docs/simuleval_instruction.md -------------------------------------------------------------------------------- /docs/waitk.md: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/docs/waitk.md -------------------------------------------------------------------------------- /eval/0-gen_simul_list.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/0-gen_simul_list.sh -------------------------------------------------------------------------------- /eval/1-simuleval.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/1-simuleval.sh -------------------------------------------------------------------------------- /eval/eval_asr.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/eval_asr.sh -------------------------------------------------------------------------------- /eval/eval_cif.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/eval_cif.sh -------------------------------------------------------------------------------- /eval/eval_mt.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/eval_mt.sh -------------------------------------------------------------------------------- /eval/eval_st.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/eval_st.sh -------------------------------------------------------------------------------- /eval/eval_transducer.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/eval_transducer.sh -------------------------------------------------------------------------------- /eval/generate.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/generate.py -------------------------------------------------------------------------------- /eval/interactive.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/interactive.py -------------------------------------------------------------------------------- /eval/latency-bleu.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/latency-bleu.ipynb -------------------------------------------------------------------------------- /eval/run_all_simuleval.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/run_all_simuleval.sh -------------------------------------------------------------------------------- /eval/visualize.ipynb: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/eval/visualize.ipynb -------------------------------------------------------------------------------- /exp/0-mt.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/0-mt.sh -------------------------------------------------------------------------------- /exp/0a-decode_distillation.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/0a-decode_distillation.sh -------------------------------------------------------------------------------- /exp/0b-create_distillation_set.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/0b-create_distillation_set.sh -------------------------------------------------------------------------------- /exp/1a-pretrain_asr.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/1a-pretrain_asr.sh -------------------------------------------------------------------------------- /exp/1b-fintune_asr_on_es.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/1b-fintune_asr_on_es.sh -------------------------------------------------------------------------------- /exp/2-mma.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/2-mma.sh -------------------------------------------------------------------------------- /exp/2b-mma_finetune.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/2b-mma_finetune.sh -------------------------------------------------------------------------------- /exp/3-cif.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/3-cif.sh -------------------------------------------------------------------------------- /exp/3b-cif_finetune.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/3b-cif_finetune.sh -------------------------------------------------------------------------------- /exp/data_path.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/data_path.sh -------------------------------------------------------------------------------- /exp/infer_asr.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/infer_asr.yaml -------------------------------------------------------------------------------- /exp/infer_mt.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/infer_mt.yaml -------------------------------------------------------------------------------- /exp/infer_st.yaml: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/exp/infer_st.yaml -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/requirements.txt -------------------------------------------------------------------------------- /scripts/average_checkpoints.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/scripts/average_checkpoints.py -------------------------------------------------------------------------------- /scripts/migrate_data_path.sh: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/scripts/migrate_data_path.sh -------------------------------------------------------------------------------- /scripts/update_config.py: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/George0828Zhang/simulst/HEAD/scripts/update_config.py --------------------------------------------------------------------------------