|
|
--- |
|
|
tags: |
|
|
- espnet |
|
|
- audio |
|
|
- automatic-speech-recognition |
|
|
language: noinfo |
|
|
datasets: |
|
|
- tamil |
|
|
license: cc-by-4.0 |
|
|
--- |
|
|
|
|
|
## ESPnet2 ASR model |
|
|
|
|
|
### `espnet/tamil_slu` |
|
|
|
|
|
This model was trained by Sujay S Kumar using tamil recipe in [espnet](https://github.com/espnet/espnet/). |
|
|
|
|
|
### Demo: How to use in ESPnet2 |
|
|
|
|
|
```bash |
|
|
cd espnet |
|
|
git checkout 395bda6123ae268f991e5ef1dab887b6e677974a |
|
|
pip install -e . |
|
|
cd egs2/tamil/asr1 |
|
|
./run.sh --skip_data_prep false --skip_train true --download_model espnet/tamil_slu |
|
|
``` |
|
|
|
|
|
<!-- Generated by scripts/utils/show_asr_result.sh --> |
|
|
# RESULTS |
|
|
## Environments |
|
|
- date: `Sun Oct 3 20:59:46 EDT 2021` |
|
|
- python version: `3.9.5 (default, Jun 4 2021, 12:28:51) [GCC 7.5.0]` |
|
|
- espnet version: `espnet 0.10.3a3` |
|
|
- pytorch version: `pytorch 1.8.1+cu102` |
|
|
- Git hash: `b41391336042a4876e30d9fe5c66afb4e4be404c` |
|
|
- Commit date: `Wed Sep 22 10:02:03 2021 -0400` |
|
|
|
|
|
## asr_train_asr_wav2vec2_xlsr_raw_word |
|
|
### WER |
|
|
|
|
|
|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |
|
|
|---|---|---|---|---|---|---|---|---| |
|
|
|inference_asr_model_valid.acc.ave_5best/test|80|372|70.4|22.6|7.0|3.2|32.8|56.3| |
|
|
|inference_asr_model_valid.acc.ave_5best/valid|80|372|70.4|22.6|7.0|3.2|32.8|56.3| |
|
|
|
|
|
### CER |
|
|
|
|
|
|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |
|
|
|---|---|---|---|---|---|---|---|---| |
|
|
|inference_asr_model_valid.acc.ave_5best/test|80|3234|85.9|8.2|5.9|5.5|19.6|56.3| |
|
|
|inference_asr_model_valid.acc.ave_5best/valid|80|3234|85.9|8.2|5.9|5.5|19.6|56.3| |
|
|
|
|
|
### TER |
|
|
|
|
|
|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err| |
|
|
|---|---|---|---|---|---|---|---|---| |
|
|
|
|
|
## ASR config |
|
|
|
|
|
<details><summary>expand</summary> |
|
|
|
|
|
``` |
|
|
config: conf/train_asr_wav2vec2_xlsr.yaml |
|
|
print_config: false |
|
|
log_level: INFO |
|
|
dry_run: false |
|
|
iterator_type: sequence |
|
|
output_dir: exp_train_asr_wav2vec2_xlsr/asr_train_asr_wav2vec2_xlsr_raw_word |
|
|
ngpu: 1 |
|
|
seed: 0 |
|
|
num_workers: 1 |
|
|
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: false |
|
|
sharded_ddp: false |
|
|
cudnn_enabled: true |
|
|
cudnn_benchmark: false |
|
|
cudnn_deterministic: true |
|
|
collect_stats: false |
|
|
write_collected_feats: false |
|
|
max_epoch: 250 |
|
|
patience: null |
|
|
val_scheduler_criterion: |
|
|
- valid |
|
|
- loss |
|
|
early_stopping_criterion: |
|
|
- valid |
|
|
- loss |
|
|
- min |
|
|
best_model_criterion: |
|
|
- - train |
|
|
- loss |
|
|
- min |
|
|
- - valid |
|
|
- loss |
|
|
- min |
|
|
- - train |
|
|
- acc |
|
|
- max |
|
|
- - valid |
|
|
- acc |
|
|
- max |
|
|
keep_nbest_models: 5 |
|
|
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_tensorboard: true |
|
|
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: 1000000 |
|
|
valid_batch_bins: null |
|
|
train_shape_file: |
|
|
- exp_train_asr_wav2vec2_xlsr/asr_stats_raw_word/train/speech_shape |
|
|
- exp_train_asr_wav2vec2_xlsr/asr_stats_raw_word/train/text_shape.word |
|
|
valid_shape_file: |
|
|
- exp_train_asr_wav2vec2_xlsr/asr_stats_raw_word/valid/speech_shape |
|
|
- exp_train_asr_wav2vec2_xlsr/asr_stats_raw_word/valid/text_shape.word |
|
|
batch_type: folded |
|
|
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 |
|
|
train_data_path_and_name_and_type: |
|
|
- - dump/raw/train/wav.scp |
|
|
- speech |
|
|
- sound |
|
|
- - dump/raw/train/text |
|
|
- text |
|
|
- text |
|
|
valid_data_path_and_name_and_type: |
|
|
- - dump/raw/valid/wav.scp |
|
|
- speech |
|
|
- sound |
|
|
- - dump/raw/valid/text |
|
|
- text |
|
|
- text |
|
|
allow_variable_data_keys: false |
|
|
max_cache_size: 0.0 |
|
|
max_cache_fd: 32 |
|
|
valid_max_cache_size: null |
|
|
optim: adam |
|
|
optim_conf: |
|
|
lr: 0.0001 |
|
|
scheduler: warmuplr |
|
|
scheduler_conf: |
|
|
warmup_steps: 5000 |
|
|
token_list: |
|
|
- <blank> |
|
|
- <unk> |
|
|
- காசு |
|
|
- வேணும் |
|
|
- Request_Acc_balance |
|
|
- Account |
|
|
- Money_deposit |
|
|
- Money_withdraw |
|
|
- Credit_card_payments |
|
|
- card |
|
|
- மீதி |
|
|
- Money_transfer |
|
|
- எவ்வளோ |
|
|
- Bill_payments |
|
|
- Credit |
|
|
- கட்ட |
|
|
- எவ்வளவு |
|
|
- காச |
|
|
- கட்டவேணும் |
|
|
- இந்த |
|
|
- Balance |
|
|
- வேண்டும் |
|
|
- போடோணும் |
|
|
- கணக்கு |
|
|
- செய்ய |
|
|
- Bill |
|
|
- போட |
|
|
- account |
|
|
- மாத்த |
|
|
- credit |
|
|
- pay |
|
|
- பண்ணோணும் |
|
|
- Deposit |
|
|
- மீளெடுக்க |
|
|
- வைப்பு |
|
|
- எடுக்கவேணும் |
|
|
- ல |
|
|
- இருக்கிற |
|
|
- எடுக்கணும் |
|
|
- இல |
|
|
- இருந்து |
|
|
- மற்ற |
|
|
- accountக்கு |
|
|
- balance |
|
|
- என்ன |
|
|
- bill |
|
|
- அ |
|
|
- ஒருக்கா |
|
|
- ஏலுமோ |
|
|
- deposit |
|
|
- பண்ண |
|
|
- payment |
|
|
- Account-la |
|
|
- காசெடுக்கோணும் |
|
|
- அனுப்பவேணும் |
|
|
- காசெடுக்க |
|
|
- இன்னொரு |
|
|
- கு |
|
|
- Cash |
|
|
- <sos/eos> |
|
|
init: null |
|
|
input_size: null |
|
|
ctc_conf: |
|
|
dropout_rate: 0.0 |
|
|
ctc_type: builtin |
|
|
reduce: true |
|
|
ignore_nan_grad: true |
|
|
model_conf: |
|
|
ctc_weight: 0.3 |
|
|
lsm_weight: 0.1 |
|
|
length_normalized_loss: false |
|
|
extract_feats_in_collect_stats: false |
|
|
use_preprocessor: true |
|
|
token_type: word |
|
|
bpemodel: null |
|
|
non_linguistic_symbols: null |
|
|
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' |
|
|
frontend: s3prl |
|
|
frontend_conf: |
|
|
frontend_conf: |
|
|
upstream: wav2vec2_xlsr |
|
|
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 |
|
|
- 30 |
|
|
num_freq_mask: 2 |
|
|
apply_time_mask: true |
|
|
time_mask_width_range: |
|
|
- 0 |
|
|
- 40 |
|
|
num_time_mask: 2 |
|
|
normalize: utterance_mvn |
|
|
normalize_conf: {} |
|
|
preencoder: linear |
|
|
preencoder_conf: |
|
|
input_size: 1024 |
|
|
output_size: 80 |
|
|
encoder: conformer |
|
|
encoder_conf: |
|
|
output_size: 512 |
|
|
attention_heads: 8 |
|
|
linear_units: 2048 |
|
|
num_blocks: 12 |
|
|
dropout_rate: 0.1 |
|
|
positional_dropout_rate: 0.1 |
|
|
attention_dropout_rate: 0.1 |
|
|
input_layer: conv2d |
|
|
normalize_before: true |
|
|
macaron_style: true |
|
|
pos_enc_layer_type: rel_pos |
|
|
selfattention_layer_type: rel_selfattn |
|
|
activation_type: swish |
|
|
use_cnn_module: true |
|
|
cnn_module_kernel: 31 |
|
|
postencoder: null |
|
|
postencoder_conf: {} |
|
|
decoder: transformer |
|
|
decoder_conf: |
|
|
attention_heads: 8 |
|
|
linear_units: 2048 |
|
|
num_blocks: 4 |
|
|
dropout_rate: 0.1 |
|
|
positional_dropout_rate: 0.1 |
|
|
self_attention_dropout_rate: 0.1 |
|
|
src_attention_dropout_rate: 0.1 |
|
|
required: |
|
|
- output_dir |
|
|
- token_list |
|
|
version: 0.10.3a3 |
|
|
distributed: false |
|
|
``` |
|
|
|
|
|
</details> |
|
|
|
|
|
|
|
|
|
|
|
### 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} |
|
|
} |
|
|
``` |
|
|
|
