--- tags: - espnet - audio - audio-to-audio language: en datasets: - librimix license: cc-by-4.0 --- ## ESPnet2 ENH model ### `espnet/Wangyou_Zhang_librimix_train_enh_tse_td_speakerbeam_raw` This model was trained by Wangyou Zhang using librimix 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/librimix/tse1 ./run.sh --skip_data_prep false --skip_train true --is_tse_task true --download_model espnet/Wangyou_Zhang_librimix_train_enh_tse_td_speakerbeam_raw ``` # RESULTS ## Environments - date: `Mon Jun 5 22:42:07 CST 2023` - python version: `3.8.16 (default, Mar 2 2023, 03:21:46) [GCC 11.2.0]` - espnet version: `espnet 202301` - pytorch version: `pytorch 2.0.1` - Git hash: `` - Commit date: `` ## enh_train_raw config: ./conf/train.yaml |dataset|PESQ_WB|STOI|SAR|SDR|SIR|SI_SNR|OVRL|SIG|BAK|P808_MOS| |---|---|---|---|---|---|---|---|---|---|---| |dev|1.08|64.43|7.18|-1.71|0.08|-1.81|1.60|2.26|1.62|2.68| |test|1.08|64.56|6.90|-1.83|0.09|-1.93|1.63|2.33|1.66|2.71| |enhanced_dev|1.73|86.50|12.50|11.40|24.83|10.58|2.95|3.24|3.92|3.23| |enhanced_test|1.73|87.36|12.34|11.47|24.51|10.74|2.99|3.29|3.91|3.25| ## ENH config
expand ``` config: ./conf/train.yaml print_config: false log_level: INFO dry_run: false iterator_type: chunk output_dir: exp/enh_train_raw ngpu: 1 seed: 0 num_workers: 2 num_att_plot: 3 dist_backend: nccl dist_init_method: env:// dist_world_size: 4 dist_rank: 0 local_rank: 0 dist_master_addr: localhost dist_master_port: 43837 dist_launcher: null multiprocessing_distributed: true unused_parameters: true sharded_ddp: false cudnn_enabled: true cudnn_benchmark: false cudnn_deterministic: true collect_stats: false write_collected_feats: false skip_stats_npz: false max_epoch: 100 patience: 20 val_scheduler_criterion: - valid - loss early_stopping_criterion: - valid - loss - min best_model_criterion: - - valid - snr - max - - valid - loss - min keep_nbest_models: 1 nbest_averaging_interval: 0 grad_clip: 5.0 grad_clip_type: 2.0 grad_noise: false accum_grad: 1 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: [] ignore_init_mismatch: false freeze_param: [] num_iters_per_epoch: null batch_size: 16 valid_batch_size: null batch_bins: 1000000 valid_batch_bins: null train_shape_file: - exp/enh_stats_train_dev_16k/train/speech_mix_shape - exp/enh_stats_train_dev_16k/train/speech_ref1_shape - exp/enh_stats_train_dev_16k/train/enroll_ref1_shape - exp/enh_stats_train_dev_16k/train/speech_ref2_shape - exp/enh_stats_train_dev_16k/train/enroll_ref2_shape valid_shape_file: - exp/enh_stats_train_dev_16k/valid/speech_mix_shape - exp/enh_stats_train_dev_16k/valid/speech_ref1_shape - exp/enh_stats_train_dev_16k/valid/enroll_ref1_shape - exp/enh_stats_train_dev_16k/valid/speech_ref2_shape - exp/enh_stats_train_dev_16k/valid/enroll_ref2_shape batch_type: folded valid_batch_type: null fold_length: - 80000 - 80000 - 80000 - 80000 - 80000 sort_in_batch: descending sort_batch: descending multiple_iterator: false chunk_length: 48000 chunk_shift_ratio: 0.5 num_cache_chunks: 1024 chunk_excluded_key_prefixes: - enroll_ref train_data_path_and_name_and_type: - - dump/raw/train/wav.scp - speech_mix - sound - - dump/raw/train/spk1.scp - speech_ref1 - sound - - dump/raw/train/enroll_spk1.scp - enroll_ref1 - text - - dump/raw/train/spk2.scp - speech_ref2 - sound - - dump/raw/train/enroll_spk2.scp - enroll_ref2 - text valid_data_path_and_name_and_type: - - dump/raw/dev/wav.scp - speech_mix - sound - - dump/raw/dev/spk1.scp - speech_ref1 - sound - - dump/raw/dev/enroll_spk1.scp - enroll_ref1 - text - - dump/raw/dev/spk2.scp - speech_ref2 - sound - - dump/raw/dev/enroll_spk2.scp - enroll_ref2 - 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: adam optim_conf: lr: 0.001 eps: 1.0e-08 weight_decay: 0 scheduler: reducelronplateau scheduler_conf: mode: min factor: 0.7 patience: 3 init: null model_conf: num_spk: 2 share_encoder: true criterions: - name: snr conf: eps: 1.0e-07 wrapper: fixed_order wrapper_conf: weight: 1.0 - name: l1_fd conf: only_for_test: true wrapper: fixed_order wrapper_conf: weight: 0.0 - name: l1_td conf: only_for_test: true wrapper: fixed_order wrapper_conf: weight: 0.0 - name: mse_fd conf: only_for_test: true wrapper: fixed_order wrapper_conf: weight: 0.0 - name: mse_td conf: only_for_test: true wrapper: fixed_order wrapper_conf: weight: 0.0 train_spk2enroll: data/train-100/spk2enroll.json enroll_segment: 48000 load_spk_embedding: false load_all_speakers: false rir_scp: null rir_apply_prob: 1.0 noise_scp: null noise_apply_prob: 1.0 noise_db_range: '13_15' short_noise_thres: 0.5 speech_volume_normalize: null use_reverberant_ref: false num_spk: 1 num_noise_type: 1 sample_rate: 8000 force_single_channel: false channel_reordering: false categories: [] encoder: conv encoder_conf: channel: 256 kernel_size: 32 stride: 16 extractor: td_speakerbeam extractor_conf: layer: 8 stack: 4 bottleneck_dim: 256 hidden_dim: 512 skip_dim: 256 kernel: 3 causal: false norm_type: gLN pre_nonlinear: prelu nonlinear: relu i_adapt_layer: 7 adapt_layer_type: mul adapt_enroll_dim: 256 use_spk_emb: false decoder: conv decoder_conf: channel: 256 kernel_size: 32 stride: 16 preprocessor: tse preprocessor_conf: {} required: - output_dir version: '202301' distributed: true ```
### 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} } @inproceedings{ESPnet-SE, author = {Chenda Li and Jing Shi and Wangyou Zhang and Aswin Shanmugam Subramanian and Xuankai Chang and Naoyuki Kamo and Moto Hira and Tomoki Hayashi and Christoph B{"{o}}ddeker and Zhuo Chen and Shinji Watanabe}, title = {ESPnet-SE: End-To-End Speech Enhancement and Separation Toolkit Designed for {ASR} Integration}, booktitle = {{IEEE} Spoken Language Technology Workshop, {SLT} 2021, Shenzhen, China, January 19-22, 2021}, pages = {785--792}, publisher = {{IEEE}}, year = {2021}, url = {https://doi.org/10.1109/SLT48900.2021.9383615}, doi = {10.1109/SLT48900.2021.9383615}, timestamp = {Mon, 12 Apr 2021 17:08:59 +0200}, biburl = {https://dblp.org/rec/conf/slt/Li0ZSCKHHBC021.bib}, bibsource = {dblp computer science bibliography, https://dblp.org} } ``` 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} } ```