Hugging Face's logo

atharva253
/
tfgridnetv2_wsj_kinect

Audio-to-Audio
ESPnet
English
audio
Model card Files Community
Edit model card

ESPnet2 ENH model

atharva253/tfgridnetv2_wsj_kinect

This model was trained by Atharva Anand Joshi using wsj_kinect recipe in espnet.

Demo: How to use in ESPnet2

Follow the ESPnet installation instructions if you haven't done that already.

cd espnet
git checkout 37828ea9708cd2f541220fdfe180457c7f7d67f1
pip install -e .
cd egs2/wsj_kinect/enh1
./run.sh --skip_data_prep false --skip_train true --download_model atharva253/tfgridnetv2_wsj_kinect

RESULTS

Environments

  • date: Mon Apr 22 17:21:05 EDT 2024
  • python version: 3.9.18 (main, Sep 11 2023, 13:41:44) [GCC 11.2.0]
  • espnet version: espnet 202402
  • pytorch version: pytorch 2.1.0
  • Git hash: 37828ea9708cd2f541220fdfe180457c7f7d67f1
    • Commit date: Thu Mar 21 22:52:57 2024 -0400

enh_train_enh_tfgridnetv2_tf_lr-patience3_patience5_I_1_J_1_D_128_batch_8_raw

config: conf/tuning/train_enh_tfgridnetv2_tf_lr-patience3_patience5_I_1_J_1_D_128_batch_8.yaml

dataset STOI SAR SDR SIR SI_SNR
enhanced_cv 85.97 10.51 10.07 21.63 9.61
enhanced_tt 88.76 11.22 10.69 21.36 10.26

ENH config

expand 
config: conf/tuning/train_enh_tfgridnetv2_tf_lr-patience3_patience5_I_1_J_1_D_128_batch_8.yaml
print_config: false
log_level: INFO
drop_last_iter: false
dry_run: false
iterator_type: chunk
valid_iterator_type: null
output_dir: exp/enh_train_enh_tfgridnetv2_tf_lr-patience3_patience5_I_1_J_1_D_128_batch_8_raw
ngpu: 1
seed: 0
num_workers: 4
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: 45443
dist_launcher: null
multiprocessing_distributed: true
unused_parameters: false
sharded_ddp: false
cudnn_enabled: true
cudnn_benchmark: false
cudnn_deterministic: true
collect_stats: false
write_collected_feats: false
max_epoch: 150
patience: 5
val_scheduler_criterion:
- valid
- loss
early_stopping_criterion:
- valid
- loss
- min
best_model_criterion:
-   - valid
    - si_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
use_adapter: false
adapter: lora
save_strategy: all
adapter_conf: {}
pretrain_path: null
init_param: []
ignore_init_mismatch: false
freeze_param: []
num_iters_per_epoch: null
batch_size: 8
valid_batch_size: null
batch_bins: 1000000
valid_batch_bins: null
train_shape_file:
- exp/enh_stats_16k/train/speech_mix_shape
- exp/enh_stats_16k/train/speech_ref1_shape
- exp/enh_stats_16k/train/speech_ref2_shape
valid_shape_file:
- exp/enh_stats_16k/valid/speech_mix_shape
- exp/enh_stats_16k/valid/speech_ref1_shape
- exp/enh_stats_16k/valid/speech_ref2_shape
batch_type: folded
valid_batch_type: null
fold_length:
- 80000
- 80000
- 80000
sort_in_batch: descending
shuffle_within_batch: false
sort_batch: descending
multiple_iterator: false
chunk_length: 32000
chunk_shift_ratio: 0.5
num_cache_chunks: 1024
chunk_excluded_key_prefixes: []
chunk_default_fs: null
train_data_path_and_name_and_type:
-   - dump/raw/tr/wav.scp
    - speech_mix
    - sound
-   - dump/raw/tr/spk1.scp
    - speech_ref1
    - sound
-   - dump/raw/tr/spk2.scp
    - speech_ref2
    - sound
valid_data_path_and_name_and_type:
-   - dump/raw/cv/wav.scp
    - speech_mix
    - sound
-   - dump/raw/cv/spk1.scp
    - speech_ref1
    - sound
-   - dump/raw/cv/spk2.scp
    - speech_ref2
    - sound
allow_variable_data_keys: false
max_cache_size: 0.0
max_cache_fd: 32
allow_multi_rates: false
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: 0.5
    patience: 3
init: xavier_uniform
model_conf:
    stft_consistency: false
    loss_type: mask_mse
    mask_type: null
    flexible_numspk: false
    extract_feats_in_collect_stats: false
    normalize_variance: false
    normalize_variance_per_ch: false
    categories: []
    category_weights: []
criterions:
-   name: si_snr
    conf:
        eps: 1.0e-07
    wrapper: pit
    wrapper_conf:
        weight: 1.0
        independent_perm: true
speech_volume_normalize: null
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
use_reverberant_ref: false
num_spk: 1
num_noise_type: 1
sample_rate: 8000
force_single_channel: false
channel_reordering: false
categories: []
speech_segment: null
avoid_allzero_segment: true
flexible_numspk: false
dynamic_mixing: false
utt2spk: null
dynamic_mixing_gain_db: 0.0
encoder: same
encoder_conf: {}
separator: tfgridnetv2
separator_conf:
    n_srcs: 2
    n_fft: 128
    stride: 64
    window: hann
    n_imics: 4
    n_layers: 6
    lstm_hidden_units: 192
    attn_n_head: 4
    attn_approx_qk_dim: 512
    emb_dim: 128
    emb_ks: 1
    emb_hs: 1
    activation: prelu
    eps: 1.0e-05
decoder: same
decoder_conf: {}
mask_module: multi_mask
mask_module_conf: {}
preprocessor: null
preprocessor_conf: {}
diffusion_model: null
diffusion_model_conf: {}
required:
- output_dir
version: '202402'
distributed: true

Citing ESPnet 

@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:

@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}
}
Downloads last month
2
Inference API
Audio-to-Audio
or
Inference API (serverless) does not yet support espnet models for this pipeline type.