espnet/yoshiki_wsj_asr_conformer_s3prlfrontend_wavlm_raw_en_char

ESPnet2 ASR model

espnet/yoshiki_wsj_asr_conformer_s3prlfrontend_wavlm_raw_en_char

This model was trained by Yoshiki using wsj recipe in espnet.

Demo: How to use in ESPnet2

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

cd espnet

pip install -e .
cd egs2/wsj/asr1
./run.sh --skip_data_prep false --skip_train true --download_model espnet/yoshiki_wsj_asr_conformer_s3prlfrontend_wavlm_raw_en_char

RESULTS

Environments

• date: Tue Jul 11 05:34:44 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: d7172fcb7181ffdcca9c0061400254b63e37bf21
  • Commit date: Sat Jul 15 15:01:30 2023 +0900

exp/asr_train_asr_conformer_s3prlfrontend_wavlm_raw_en_char

WER

dataset	Snt	Wrd	Corr	Sub	Del	Ins	Err	S.Err
decode_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/test_eval92	333	5643	98.6	1.3	0.1	0.1	1.5	19.5

CER

dataset	Snt	Wrd	Corr	Sub	Del	Ins	Err	S.Err
decode_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave/test_eval92	333	33341	99.4	0.2	0.3	0.1	0.7	33.0

TER

dataset	Snt	Wrd	Corr	Sub	Del	Ins	Err	S.Err

exp/asr_train_asr_conformer_s3prlfrontend_wavlm_raw_en_char/decode_lm_lm_train_lm_transformer_en_char_valid.loss.ave_asr_model_valid.acc.ave

WER

dataset	Snt	Wrd	Corr	Sub	Del	Ins	Err	S.Err
org/test_dev93	503	8234	97.7	2.1	0.3	0.2	2.6	28.0

CER

dataset	Snt	Wrd	Corr	Sub	Del	Ins	Err	S.Err
org/test_dev93	503	48634	99.1	0.3	0.5	0.2	1.0	37.8

TER

dataset	Snt	Wrd	Corr	Sub	Del	Ins	Err	S.Err

ASR config

expand

config: conf/tuning/train_asr_conformer_s3prlfrontend_wavlm.yaml
print_config: false
log_level: INFO
dry_run: false
iterator_type: sequence
output_dir: exp/asr_train_asr_conformer_s3prlfrontend_wavlm_raw_en_char
ngpu: 1
seed: 0
num_workers: 8
num_att_plot: 3
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: 100
patience: null
val_scheduler_criterion:
- valid
- loss
early_stopping_criterion:
- valid
- loss
- min
best_model_criterion:
-   - valid
    - acc
    - max
keep_nbest_models: 10
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:
- frontend.upstream
num_iters_per_epoch: null
batch_size: 20
valid_batch_size: null
batch_bins: 4000000
valid_batch_bins: null
train_shape_file:
- exp/asr_stats_raw_en_char/train/speech_shape
- exp/asr_stats_raw_en_char/train/text_shape.char
valid_shape_file:
- exp/asr_stats_raw_en_char/valid/speech_shape
- exp/asr_stats_raw_en_char/valid/text_shape.char
batch_type: numel
valid_batch_type: null
fold_length:
- 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/train_si284/wav.scp
    - speech
    - sound
-   - dump/raw/train_si284/text
    - text
    - text
valid_data_path_and_name_and_type:
-   - dump/raw/test_dev93/wav.scp
    - speech
    - sound
-   - dump/raw/test_dev93/text
    - text
    - 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
scheduler: warmuplr
scheduler_conf:
    warmup_steps: 20000
token_list:
- <blank>
- <unk>
- <space>
- E
- T
- A
- N
- I
- O
- S
- R
- H
- L
- D
- C
- U
- M
- P
- F
- G
- Y
- W
- B
- V
- K
- .
- X
- ''''
- J
- Q
- Z
- <NOISE>
- ','
- '-'
- '"'
- '*'
- ':'
- (
- )
- '?'
- '!'
- '&'
- ;
- '1'
- '2'
- '0'
- /
- $
- '{'
- '}'
- '8'
- '9'
- '6'
- '3'
- '5'
- '7'
- '4'
- '~'
- '`'
- _
- <*IN*>
- <*MR.*>
- \
- ^
- <sos/eos>
init: xavier_uniform
input_size: null
ctc_conf:
    ignore_nan_grad: true
joint_net_conf: null
use_preprocessor: true
token_type: char
bpemodel: null
non_linguistic_symbols: data/nlsyms.txt
cleaner: null
g2p: null
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
aux_ctc_tasks: []
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: {}
model: espnet
model_conf:
    ctc_weight: 0.3
    lsm_weight: 0.1
    length_normalized_loss: false
    extract_feats_in_collect_stats: false
preencoder: linear
preencoder_conf:
    input_size: 1024
    output_size: 80
encoder: conformer
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
postencoder: null
postencoder_conf: {}
decoder: transformer
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
preprocessor: default
preprocessor_conf: {}
required:
- output_dir
- token_list
version: '202304'
distributed: false

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}
}

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}
}