Update model

README.md

@@ -0,0 +1,436 @@
+---
+tags:
+- espnet
+- audio
+- automatic-speech-recognition
+language: be
+datasets:
+- commonvoice
+license: cc-by-4.0
+---
+
+## ESPnet2 ASR model 
+
+### `espnet/belarusian_commonvoice_blstm`
+
+This model was trained by dzeinali using commonvoice recipe in [espnet](https://github.com/espnet/espnet/).
+
+### Demo: How to use in ESPnet2
+
+```bash
+cd espnet
+git checkout 716eb8f92e19708acfd08ba3bd39d40890d3a84b
+pip install -e .
+cd egs2/commonvoice/asr1
+./run.sh --skip_data_prep false --skip_train true --download_model espnet/belarusian_commonvoice_blstm
+```
+
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Thu May 19 18:39:24 EDT 2022`
+- python version: `3.9.5 (default, Jun  4 2021, 12:28:51)  [GCC 7.5.0]`
+- espnet version: `espnet 0.10.6a1`
+- pytorch version: `pytorch 1.8.1+cu102`
+- Git hash: `716eb8f92e19708acfd08ba3bd39d40890d3a84b`
+  - Commit date: `Thu Apr 28 19:50:59 2022 -0400`
+
+## asr_train_asr_rnn_raw_be_bpe150_sp
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_rnn_asr_model_valid.acc.best/test_be|15801|128674|84.2|14.4|1.5|1.5|17.3|62.1|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_rnn_asr_model_valid.acc.best/test_be|15801|906360|97.1|1.8|1.2|0.7|3.6|62.1|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_rnn_asr_model_valid.acc.best/test_be|15801|639798|95.5|2.9|1.6|0.9|5.3|62.1|
+
+## ASR config
+
+<details><summary>expand</summary>
+
+```
+config: conf/tuning/train_asr_rnn.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/asr_train_asr_rnn_raw_be_bpe150_sp
+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: 15
+patience: 3
+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:
+- 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
+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: 30
+valid_batch_size: null
+batch_bins: 1000000
+valid_batch_bins: null
+train_shape_file:
+- exp/asr_stats_raw_be_bpe150_sp/train/speech_shape
+- exp/asr_stats_raw_be_bpe150_sp/train/text_shape.bpe
+valid_shape_file:
+- exp/asr_stats_raw_be_bpe150_sp/valid/speech_shape
+- exp/asr_stats_raw_be_bpe150_sp/valid/text_shape.bpe
+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_be_sp/wav.scp
+    - speech
+    - sound
+-   - dump/raw/train_be_sp/text
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev_be/wav.scp
+    - speech
+    - sound
+-   - dump/raw/dev_be/text
+    - text
+    - text
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adadelta
+optim_conf:
+    lr: 0.1
+scheduler: null
+scheduler_conf: {}
+token_list:
+- <blank>
+- <unk>
+- ▁
+- я
+- с
+- а
+- .
+- н
+- о
+- м
+- е
+- у
+- ','
+- на
+- ра
+- ў
+- д
+- р
+- т
+- і
+- ва
+- х
+- к
+- ка
+- й
+- б
+- з
+- ць
+- п
+- ц
+- лі
+- ны
+- ▁па
+- ма
+- г
+- ▁с
+- ла
+- ▁і
+- ю
+- ш
+- в
+- та
+- ы
+- ры
+- э
+- га
+- кі
+- ▁з
+- ні
+- ▁на
+- л
+- ча
+- да
+- ё
+- ле
+- ▁не
+- ль
+- ч
+- ▁ў
+- ж
+- не
+- ▁а
+- чы
+- ты
+- цца
+- ля
+- ▁за
+- ▁пра
+- мі
+- ве
+- бы
+- рэ
+- ▁А
+- ста
+- ку
+- го
+- цы
+- пе
+- ст
+- ▁вы
+- па
+- ▁ад
+- ▁я
+- ці
+- ба
+- дзе
+- ▁да
+- вы
+- оў
+- ту
+- ко
+- ды
+- ▁гэта
+- ло
+- ▁што
+- ▁ка
+- ф
+- жа
+- П
+- ві
+- дзі
+- Я
+- Т
+- '"'
+- ь
+- '?'
+- І
+- М
+- З
+- Д
+- В
+- Н
+- С
+- '!'
+- К
+- '-'
+- ''''
+- Б
+- Ц
+- Г
+- Р
+- Ш
+- Х
+- Л
+- А
+- Ж
+- Ф
+- У
+- Э
+- О
+- Е
+- I
+- —
+- Ю
+- ‒
+- ':'
+- Ў
+- ��
+- »
+- Ы
+- ;
+- ’
+- Ь
+- Й
+- ґ
+- «
+- Ч
+- N
+- Ё
+- <sos/eos>
+init: null
+input_size: null
+ctc_conf:
+    dropout_rate: 0.0
+    ctc_type: builtin
+    reduce: true
+    ignore_nan_grad: true
+joint_net_conf: null
+model_conf:
+    ctc_weight: 0.5
+use_preprocessor: true
+token_type: bpe
+bpemodel: data/be_token_list/bpe_unigram150/bpe.model
+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: default
+frontend_conf:
+    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
+    - 27
+    num_freq_mask: 2
+    apply_time_mask: true
+    time_mask_width_ratio_range:
+    - 0.0
+    - 0.05
+    num_time_mask: 2
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/asr_stats_raw_be_bpe150_sp/train/feats_stats.npz
+preencoder: null
+preencoder_conf: {}
+encoder: vgg_rnn
+encoder_conf:
+    rnn_type: lstm
+    bidirectional: true
+    use_projection: true
+    num_layers: 4
+    hidden_size: 1024
+    output_size: 1024
+postencoder: null
+postencoder_conf: {}
+decoder: rnn
+decoder_conf:
+    num_layers: 2
+    hidden_size: 1024
+    sampling_probability: 0
+    att_conf:
+        atype: location
+        adim: 1024
+        aconv_chans: 10
+        aconv_filts: 100
+required:
+- output_dir
+- token_list
+version: 0.10.6a1
+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}
+}
+```

data/be_token_list/bpe_unigram150/bpe.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a210bcfd345eebc065197c321dbe2d2b0b0d1b68b93b6660b489db0a7dc674b9
+size 239520

exp/asr_train_asr_rnn_raw_be_bpe150_sp/2epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd76e80633fa03880778f0e61412a1addadd494acc0ff4f0ee6942efc53ed951
+size 447981362

exp/asr_train_asr_rnn_raw_be_bpe150_sp/RESULTS.md

@@ -0,0 +1,29 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Thu May 19 18:39:24 EDT 2022`
+- python version: `3.9.5 (default, Jun  4 2021, 12:28:51)  [GCC 7.5.0]`
+- espnet version: `espnet 0.10.6a1`
+- pytorch version: `pytorch 1.8.1+cu102`
+- Git hash: `716eb8f92e19708acfd08ba3bd39d40890d3a84b`
+  - Commit date: `Thu Apr 28 19:50:59 2022 -0400`
+
+## asr_train_asr_rnn_raw_be_bpe150_sp
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_rnn_asr_model_valid.acc.best/test_be|15801|128674|84.2|14.4|1.5|1.5|17.3|62.1|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_rnn_asr_model_valid.acc.best/test_be|15801|906360|97.1|1.8|1.2|0.7|3.6|62.1|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_rnn_asr_model_valid.acc.best/test_be|15801|639798|95.5|2.9|1.6|0.9|5.3|62.1|
+

exp/asr_train_asr_rnn_raw_be_bpe150_sp/config.yaml

@@ -0,0 +1,339 @@
+config: conf/tuning/train_asr_rnn.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/asr_train_asr_rnn_raw_be_bpe150_sp
+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: 15
+patience: 3
+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:
+- 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
+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: 30
+valid_batch_size: null
+batch_bins: 1000000
+valid_batch_bins: null
+train_shape_file:
+- exp/asr_stats_raw_be_bpe150_sp/train/speech_shape
+- exp/asr_stats_raw_be_bpe150_sp/train/text_shape.bpe
+valid_shape_file:
+- exp/asr_stats_raw_be_bpe150_sp/valid/speech_shape
+- exp/asr_stats_raw_be_bpe150_sp/valid/text_shape.bpe
+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_be_sp/wav.scp
+    - speech
+    - sound
+-   - dump/raw/train_be_sp/text
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev_be/wav.scp
+    - speech
+    - sound
+-   - dump/raw/dev_be/text
+    - text
+    - text
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: adadelta
+optim_conf:
+    lr: 0.1
+scheduler: null
+scheduler_conf: {}
+token_list:
+- <blank>
+- <unk>
+- ▁
+- я
+- с
+- а
+- .
+- н
+- о
+- м
+- е
+- у
+- ','
+- на
+- ра
+- ў
+- д
+- р
+- т
+- і
+- ва
+- х
+- к
+- ка
+- й
+- б
+- з
+- ць
+- п
+- ц
+- лі
+- ны
+- ▁па
+- ма
+- г
+- ▁с
+- ла
+- ▁і
+- ю
+- ш
+- в
+- та
+- ы
+- ры
+- э
+- га
+- кі
+- ▁з
+- ні
+- ▁на
+- л
+- ча
+- да
+- ё
+- ле
+- ▁не
+- ль
+- ч
+- ▁ў
+- ж
+- не
+- ▁а
+- чы
+- ты
+- цца
+- ля
+- ▁за
+- ▁пра
+- мі
+- ве
+- бы
+- рэ
+- ▁А
+- ста
+- ку
+- го
+- цы
+- пе
+- ст
+- ▁вы
+- па
+- ▁ад
+- ▁я
+- ці
+- ба
+- дзе
+- ▁да
+- вы
+- оў
+- ту
+- ко
+- ды
+- ▁гэта
+- ло
+- ▁што
+- ▁ка
+- ф
+- жа
+- П
+- ві
+- дзі
+- Я
+- Т
+- '"'
+- ь
+- '?'
+- І
+- М
+- З
+- Д
+- В
+- Н
+- С
+- '!'
+- К
+- '-'
+- ''''
+- Б
+- Ц
+- Г
+- Р
+- Ш
+- Х
+- Л
+- А
+- Ж
+- Ф
+- У
+- Э
+- О
+- Е
+- I
+- —
+- Ю
+- ‒
+- ':'
+- Ў
+- –
+- »
+- Ы
+- ;
+- ’
+- Ь
+- Й
+- ґ
+- «
+- Ч
+- N
+- Ё
+- <sos/eos>
+init: null
+input_size: null
+ctc_conf:
+    dropout_rate: 0.0
+    ctc_type: builtin
+    reduce: true
+    ignore_nan_grad: true
+joint_net_conf: null
+model_conf:
+    ctc_weight: 0.5
+use_preprocessor: true
+token_type: bpe
+bpemodel: data/be_token_list/bpe_unigram150/bpe.model
+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: default
+frontend_conf:
+    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
+    - 27
+    num_freq_mask: 2
+    apply_time_mask: true
+    time_mask_width_ratio_range:
+    - 0.0
+    - 0.05
+    num_time_mask: 2
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/asr_stats_raw_be_bpe150_sp/train/feats_stats.npz
+preencoder: null
+preencoder_conf: {}
+encoder: vgg_rnn
+encoder_conf:
+    rnn_type: lstm
+    bidirectional: true
+    use_projection: true
+    num_layers: 4
+    hidden_size: 1024
+    output_size: 1024
+postencoder: null
+postencoder_conf: {}
+decoder: rnn
+decoder_conf:
+    num_layers: 2
+    hidden_size: 1024
+    sampling_probability: 0
+    att_conf:
+        atype: location
+        adim: 1024
+        aconv_chans: 10
+        aconv_filts: 100
+required:
+- output_dir
+- token_list
+version: 0.10.6a1
+distributed: false

meta.yaml

@@ -0,0 +1,8 @@
+espnet: 0.10.6a1
+files:
+  asr_model_file: exp/asr_train_asr_rnn_raw_be_bpe150_sp/2epoch.pth
+python: "3.9.5 (default, Jun  4 2021, 12:28:51) \n[GCC 7.5.0]"
+timestamp: 1653267030.739202
+torch: 1.8.1+cu102
+yaml_files:
+  asr_train_config: exp/asr_train_asr_rnn_raw_be_bpe150_sp/config.yaml