--- tags: - espnet - audio - speech-enhancement-recognition language: en datasets: - wsj0_2mix_spatialized license: cc-by-4.0 --- ## ESPnet2 EnhS2T model ### `espnet/yoshiki_wsj0_2mix_spatialized_enh_asr_tfgridnet_waspaa2023_raw_en_char` This model was trained by Yoshiki using wsj0_2mix_spatialized recipe in [espnet](https://github.com/espnet/espnet/). ### Demo: How to use in ESPnet2 Follow the [ESPnet installation instructions](https://espnet.github.io/espnet/installation.html) if you haven't done that already. ```bash cd espnet pip install -e . cd egs2/wsj0_2mix_spatialized/enh_asr1 ./run.sh --skip_data_prep false --skip_train true --download_model espnet/yoshiki_wsj0_2mix_spatialized_enh_asr_tfgridnet_waspaa2023_raw_en_char ``` # RESULTS ## Environments - date: `Sun Aug 13 19:05:53 UTC 2023` - python version: `3.7.4 (default, Aug 13 2019, 20:35:49) [GCC 7.3.0]` - espnet version: `espnet 202304` - pytorch version: `pytorch 1.10.1+cu111` - Git hash: `` - Commit date: `` ## exp/enh_asr_train_enh_asr_tfgridnet_waspaa2023_raw_en_char ### WER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |---|---|---|---|---|---|---|---|---| |decode_asr_transformer_normalize_output_wavtrue_lm_lm_train_lm_transformer_en_char_valid.loss.ave_enh_asr_model_valid.acc.best/tt_spatialized_anechoic_multich_max_16k|6000|98613|98.7|1.2|0.1|0.5|1.7|16.5| |decode_asr_transformer_normalize_output_wavtrue_lm_lm_train_lm_transformer_en_char_valid.loss.ave_enh_asr_model_valid.acc.best/tt_spatialized_reverb_multich_max_16k|6000|98613|98.7|1.3|0.1|0.4|1.7|17.8| ### CER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |---|---|---|---|---|---|---|---|---| |decode_asr_transformer_normalize_output_wavtrue_lm_lm_train_lm_transformer_en_char_valid.loss.ave_enh_asr_model_valid.acc.best/tt_spatialized_anechoic_multich_max_16k|6000|598296|99.6|0.2|0.2|0.3|0.7|21.6| |decode_asr_transformer_normalize_output_wavtrue_lm_lm_train_lm_transformer_en_char_valid.loss.ave_enh_asr_model_valid.acc.best/tt_spatialized_reverb_multich_max_16k|6000|598296|99.6|0.2|0.3|0.3|0.7|23.0| ### TER |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |---|---|---|---|---|---|---|---|---| ## EnhS2T config
expand ``` config: conf/tuning/train_enh_asr_tfgridnet_waspaa2023.yaml print_config: false log_level: INFO dry_run: false iterator_type: sequence output_dir: exp/enh_asr_train_enh_asr_tfgridnet_waspaa2023_raw_en_char ngpu: 1 seed: 0 num_workers: 0 num_att_plot: 0 dist_backend: nccl dist_init_method: env:// dist_world_size: null dist_rank: null local_rank: 0 dist_master_addr: null dist_master_port: null dist_launcher: null multiprocessing_distributed: false unused_parameters: true sharded_ddp: false cudnn_enabled: true cudnn_benchmark: false cudnn_deterministic: true collect_stats: false write_collected_feats: false max_epoch: 11 patience: 10 val_scheduler_criterion: - valid - loss early_stopping_criterion: - valid - loss - min best_model_criterion: - - valid - acc - max - - train - loss - min keep_nbest_models: 10 nbest_averaging_interval: 0 grad_clip: 5 grad_clip_type: 2.0 grad_noise: false accum_grad: 4 no_forward_run: false resume: true train_dtype: float32 use_amp: false log_interval: null use_matplotlib: true use_tensorboard: true create_graph_in_tensorboard: false use_wandb: false wandb_project: null wandb_id: null wandb_entity: null wandb_name: null wandb_model_log_interval: -1 detect_anomaly: false pretrain_path: null init_param: - ../enh1/exp/enh_train_enh_tfgridnet_waspaa2023_raw/valid.loss.best.pth:separator:enh_model.separator - ../../wsj/asr1/exp/asr_train_asr_conformer_s3prlfrontend_wavlm_raw_en_char/valid.acc.best.pth:frontend:s2t_model.frontend - ../../wsj/asr1/exp/asr_train_asr_conformer_s3prlfrontend_wavlm_raw_en_char/valid.acc.best.pth:preencoder:s2t_model.preencoder - ../../wsj/asr1/exp/asr_train_asr_conformer_s3prlfrontend_wavlm_raw_en_char/valid.acc.best.pth:encoder:s2t_model.encoder - ../../wsj/asr1/exp/asr_train_asr_conformer_s3prlfrontend_wavlm_raw_en_char/valid.acc.best.pth:ctc:s2t_model.ctc - ../../wsj/asr1/exp/asr_train_asr_conformer_s3prlfrontend_wavlm_raw_en_char/valid.acc.best.pth:decoder:s2t_model.decoder ignore_init_mismatch: false freeze_param: - s2t_model.frontend.upstream num_iters_per_epoch: 20000 batch_size: 2 valid_batch_size: null batch_bins: 1000000 valid_batch_bins: null train_shape_file: - exp/enh_asr_stats_raw_en_char/train/speech_shape - exp/enh_asr_stats_raw_en_char/train/speech_ref1_shape - exp/enh_asr_stats_raw_en_char/train/text_spk1_shape.char - exp/enh_asr_stats_raw_en_char/train/speech_ref2_shape - exp/enh_asr_stats_raw_en_char/train/text_spk2_shape.char valid_shape_file: - exp/enh_asr_stats_raw_en_char/valid/speech_shape - exp/enh_asr_stats_raw_en_char/valid/speech_ref1_shape - exp/enh_asr_stats_raw_en_char/valid/text_spk1_shape.char - exp/enh_asr_stats_raw_en_char/valid/speech_ref2_shape - exp/enh_asr_stats_raw_en_char/valid/text_spk2_shape.char batch_type: folded valid_batch_type: null fold_length: - 80000 - 80000 - 150 - 80000 - 150 sort_in_batch: descending sort_batch: descending multiple_iterator: false chunk_length: 500 chunk_shift_ratio: 0.5 num_cache_chunks: 1024 chunk_excluded_key_prefixes: [] train_data_path_and_name_and_type: - - dump/raw/tr_spatialized_multi_multich_max_16k/wav.scp - speech - sound - - dump/raw/tr_spatialized_multi_multich_max_16k/spk1.scp - speech_ref1 - sound - - dump/raw/tr_spatialized_multi_multich_max_16k/text_spk1 - text_spk1 - text - - dump/raw/tr_spatialized_multi_multich_max_16k/spk2.scp - speech_ref2 - sound - - dump/raw/tr_spatialized_multi_multich_max_16k/text_spk2 - text_spk2 - text valid_data_path_and_name_and_type: - - dump/raw/cv_spatialized_multi_multich_max_16k/wav.scp - speech - sound - - dump/raw/cv_spatialized_multi_multich_max_16k/spk1.scp - speech_ref1 - sound - - dump/raw/cv_spatialized_multi_multich_max_16k/text_spk1 - text_spk1 - text - - dump/raw/cv_spatialized_multi_multich_max_16k/spk2.scp - speech_ref2 - sound - - dump/raw/cv_spatialized_multi_multich_max_16k/text_spk2 - text_spk2 - text allow_variable_data_keys: false max_cache_size: 0.0 max_cache_fd: 32 valid_max_cache_size: null exclude_weight_decay: false exclude_weight_decay_conf: {} optim: sgd optim_conf: lr: 0.001 momentum: 0.9 scheduler: null scheduler_conf: {} token_list: data/en_token_list/char/tokens.txt src_token_list: null init: xavier_uniform input_size: null ctc_conf: dropout_rate: 0.0 ctc_type: builtin reduce: true ignore_nan_grad: null zero_infinity: true enh_criterions: - name: mr_l1_tfd conf: window_sz: - 512 time_domain_weight: 1.0 wrapper: pit wrapper_conf: weight: 1.0 diar_num_spk: null diar_input_size: null enh_model_conf: stft_consistency: false loss_type: mask_mse mask_type: null asr_model_conf: ctc_weight: 0.3 lsm_weight: 0.1 length_normalized_loss: false extract_feats_in_collect_stats: false st_model_conf: stft_consistency: false loss_type: mask_mse mask_type: null diar_model_conf: diar_weight: 1.0 attractor_weight: 1.0 subtask_series: - enh - asr model_conf: bypass_enh_prob: 0.0 calc_enh_loss: false use_preprocessor: true token_type: char bpemodel: null src_token_type: bpe src_bpemodel: null non_linguistic_symbols: data/nlsyms.txt cleaner: null g2p: null text_name: - text_spk1 - text_spk2 enh_encoder: same enh_encoder_conf: {} enh_separator: tfgridnet enh_separator_conf: n_srcs: 2 n_fft: 512 stride: 256 window: hann n_imics: 8 n_layers: 6 lstm_hidden_units: 192 attn_n_head: 4 attn_approx_qk_dim: 512 emb_dim: 48 emb_ks: 4 emb_hs: 2 activation: gelu eps: 1.0e-05 ref_channel: 0 enh_decoder: same enh_decoder_conf: {} enh_mask_module: multi_mask enh_mask_module_conf: {} frontend: s3prl frontend_conf: frontend_conf: upstream: wavlm_large download_dir: ./hub multilayer_feature: true fs: 16k specaug: specaug specaug_conf: apply_time_warp: true time_warp_window: 5 time_warp_mode: bicubic apply_freq_mask: true freq_mask_width_range: - 0 - 100 num_freq_mask: 4 apply_time_mask: true time_mask_width_range: - 0 - 40 num_time_mask: 2 normalize: utterance_mvn normalize_conf: {} asr_preencoder: linear asr_preencoder_conf: input_size: 1024 output_size: 80 asr_encoder: conformer asr_encoder_conf: output_size: 256 attention_heads: 4 linear_units: 2048 num_blocks: 12 dropout_rate: 0.1 positional_dropout_rate: 0.1 attention_dropout_rate: 0.0 input_layer: conv2d2 normalize_before: true macaron_style: true rel_pos_type: latest pos_enc_layer_type: rel_pos selfattention_layer_type: rel_selfattn activation_type: swish use_cnn_module: true cnn_module_kernel: 15 asr_postencoder: null asr_postencoder_conf: {} asr_decoder: transformer asr_decoder_conf: input_layer: embed attention_heads: 4 linear_units: 2048 num_blocks: 6 dropout_rate: 0.1 positional_dropout_rate: 0.1 self_attention_dropout_rate: 0.0 src_attention_dropout_rate: 0.0 st_preencoder: null st_preencoder_conf: {} st_encoder: rnn st_encoder_conf: {} st_postencoder: null st_postencoder_conf: {} st_decoder: rnn st_decoder_conf: {} st_extra_asr_decoder: rnn st_extra_asr_decoder_conf: {} st_extra_mt_decoder: rnn st_extra_mt_decoder_conf: {} diar_frontend: default diar_frontend_conf: {} diar_specaug: null diar_specaug_conf: {} diar_normalize: utterance_mvn diar_normalize_conf: {} diar_encoder: transformer diar_encoder_conf: {} diar_decoder: linear diar_decoder_conf: {} label_aggregator: label_aggregator label_aggregator_conf: {} diar_attractor: null diar_attractor_conf: {} required: - output_dir version: '202304' distributed: false ```
### Citing ESPnet ```BibTex @inproceedings{watanabe2018espnet, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, title={{ESPnet}: End-to-End Speech Processing Toolkit}, year={2018}, booktitle={Proceedings of Interspeech}, pages={2207--2211}, doi={10.21437/Interspeech.2018-1456}, url={http://dx.doi.org/10.21437/Interspeech.2018-1456} } ``` or arXiv: ```bibtex @misc{watanabe2018espnet, title={ESPnet: End-to-End Speech Processing Toolkit}, author={Shinji Watanabe and Takaaki Hori and Shigeki Karita and Tomoki Hayashi and Jiro Nishitoba and Yuya Unno and Nelson Yalta and Jahn Heymann and Matthew Wiesner and Nanxin Chen and Adithya Renduchintala and Tsubasa Ochiai}, year={2018}, eprint={1804.00015}, archivePrefix={arXiv}, primaryClass={cs.CL} } ```