ouktlab
/

espnet_streaming_katakana_csj_asr_train_asr_transformer_lm_rnn

Model card Files Files and versions Community

rtakeda commited on Aug 8, 2024

Commit

d6b2b3e

verified ·

1 Parent(s): 8cc0c23

Upload 7 files

Browse files

Files changed (7) hide show

README.md +70 -3
exp/asr_stats_raw_jp_char_sp/train/feats_stats.npz +3 -0
exp/asr_train_asr_streaming_transformer_ja_raw_jp_char_sp/config.yaml +279 -0
exp/asr_train_asr_streaming_transformer_ja_raw_jp_char_sp/valid.acc.ave_10best.pth +3 -0
exp/lm_train_lm_ja_char/config.yaml +203 -0
exp/lm_train_lm_ja_char/valid.loss.ave.pth +3 -0
meta.yaml +6 -0

README.md CHANGED Viewed

@@ -1,3 +1,70 @@
----
-license: cc-by-nc-4.0
----

+---
+{}
+---
+## ESPnet2 Streaming Katakana ASR model
+### `ouktlab/espnet_streaming_csj_asr_train_asr_transformer_lm_rnn`
+This model was trained using csj recipe in [espnet](https://github.com/espnet/espnet/).
+### 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}
+}
+```
+### Training data recipe of this model (rev.+bgn.)
+```BibTex
+@inproceedings{rtakeda2023:apsipa,
+  author={Ryu Takeda and Yui Sudo and Kazunori Komatani},
+  title={Flexible Evidence Model to Reduce Uncertainty Mismatch Between Speech Enhancement and ASR Based on Encoder-Decoder Architecture},
+  year={2023},
+  booktitle={Proceedings of Asia Pacific Signal and Information Processing Association (APSIPA)},
+  pages={1830-1837}
+}
+### Katakana model
+```BibTex
+@inproceedings{rtakeda2024:iwsds,
+  author={Ryu Takeda and Kazunori Komatani},
+  title={Toward OOV-word Acquisition during Spoken Dialogue using Syllable-based ASR and Word Segmentation},
+  year={2024},
+  booktitle={Proceedings of International Workshop on Spoken Dialogue Systems Technology (IWSDS)},
+}
+@inproceedings{oshio2023:apsipa,
+  author={Miki Oshio, Hokuto Munakata, Ryu Takeda and Kazunori Komatani},
+  title={Out-Of-Vocabulary Word Detection in Spoken Dialogues Based on Joint Decoding with User Response Patterns},
+  year={2023},
+  booktitle={Proceedings of Asia Pacific Signal and Information Processing Association (APSIPA)},
+  pages={1753-1759}
+}
+```
+license: cc-by-nc-4.0

exp/asr_stats_raw_jp_char_sp/train/feats_stats.npz ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:621b82abd629a39b062a42ff8bcd32be2286b8f70b7a992254dfcb129a3d9de8
+size 1402

exp/asr_train_asr_streaming_transformer_ja_raw_jp_char_sp/config.yaml ADDED Viewed

	@@ -0,0 +1,279 @@

+config: conf/train_asr_streaming_transformer_ja.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/asr_train_asr_streaming_transformer_ja_raw_jp_char_sp
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 3
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 49623
+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: 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: 6
+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: 20
+valid_batch_size: null
+batch_bins: 15000000
+valid_batch_bins: null
+train_shape_file:
+- exp/asr_stats_raw_jp_char_sp/train/speech_shape
+- exp/asr_stats_raw_jp_char_sp/train/text_shape.char
+valid_shape_file:
+- exp/asr_stats_raw_jp_char_sp/valid/speech_shape
+- exp/asr_stats_raw_jp_char_sp/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
+train_data_path_and_name_and_type:
+-   - dump/raw/train_nodup_sp/wav.scp
+    - speech
+    - sound
+-   - dump/raw/train_nodup_sp/text
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/train_dev/wav.scp
+    - speech
+    - sound
+-   - dump/raw/train_dev/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: 0.002
+scheduler: warmuplr
+scheduler_conf:
+    warmup_steps: 25000
+token_list:
+- <blank>
+- <unk>
+- ー
+- ン
+- イ
+- ト
+- ノ
+- シ
+- カ
+- テ
+- デ
+- タ
+- ス
+- マ
+- ッ
+- ナ
+- コ
+- オ
+- ニ
+- エ
+- ワ
+- ク
+- ア
+- ガ
+- モ
+- キ
+- ル
+- ジ
+- レ
+- リ
+- ョ
+- ソ
+- ラ
+- ツ
+- チ
+- ケ
+- ユ
+- ド
+- サ
+- セ
+- ヨ
+- ダ
+- ュ
+- ヒ
+- ウ
+- ネ
+- ホ
+- ハ
+- ミ
+- ゴ
+- ロ
+- ブ
+- バ
+- ヤ
+- メ
+- ャ
+- フ
+- ズ
+- ゲ
+- ム
+- ギ
+- グ
+- パ
+- ベ
+- ゼ
+- ビ
+- ザ
+- ヘ
+- ボ
+- ィ
+- ゾ
+- プ
+- ピ
+- ヌ
+- ポ
+- ペ
+- ェ
+- ァ
+- ゥ
+- ォ
+- ヴ
+- ヮ
+- ヅ
+- <sos/eos>
+init: xavier_uniform
+input_size: null
+ctc_conf:
+    dropout_rate: 0.0
+    ctc_type: builtin
+    reduce: true
+    ignore_nan_grad: null
+    zero_infinity: true
+joint_net_conf: null
+use_preprocessor: true
+token_type: char
+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'
+short_noise_thres: 0.5
+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
+    - 30
+    num_freq_mask: 2
+    apply_time_mask: true
+    time_mask_width_range:
+    - 0
+    - 40
+    num_time_mask: 2
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/asr_stats_raw_jp_char_sp/train/feats_stats.npz
+model: espnet
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1
+    length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: contextual_block_transformer
+encoder_conf:
+    output_size: 512
+    attention_heads: 8
+    linear_units: 2048
+    num_blocks: 18
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    attention_dropout_rate: 0.1
+    input_layer: conv2d6
+    normalize_before: true
+    block_size: 40
+    hop_size: 16
+    look_ahead: 16
+    init_average: true
+    ctx_pos_enc: true
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+    attention_heads: 8
+    linear_units: 2048
+    num_blocks: 6
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.1
+    src_attention_dropout_rate: 0.1
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202211'
+distributed: true

exp/asr_train_asr_streaming_transformer_ja_raw_jp_char_sp/valid.acc.ave_10best.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:9ffa823a00bc1b5e8b1161e327f3306ff8a68ba2c5a18a389d6d028eab1fc89a
+size 367467429

exp/lm_train_lm_ja_char/config.yaml ADDED Viewed

	@@ -0,0 +1,203 @@

+config: conf/train_lm.yaml
+print_config: false
+log_level: INFO
+dry_run: false
+iterator_type: sequence
+output_dir: exp/lm_train_lm_jp_char
+ngpu: 1
+seed: 0
+num_workers: 1
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 3
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 54257
+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: 40
+patience: 3
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+-   - 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: 256
+valid_batch_size: null
+batch_bins: 1000000
+valid_batch_bins: null
+train_shape_file:
+- exp/lm_stats_jp_char/train/text_shape.char
+valid_shape_file:
+- exp/lm_stats_jp_char/valid/text_shape.char
+batch_type: folded
+valid_batch_type: null
+fold_length:
+- 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/lm_train.txt
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/train_dev/text
+    - text
+    - text
+allow_variable_data_keys: false
+max_cache_size: 0.0
+max_cache_fd: 32
+valid_max_cache_size: null
+optim: sgd
+optim_conf: {}
+scheduler: null
+scheduler_conf: {}
+token_list:
+- <blank>
+- <unk>
+- ー
+- ン
+- イ
+- ト
+- ノ
+- シ
+- カ
+- テ
+- デ
+- タ
+- ス
+- マ
+- ッ
+- ナ
+- コ
+- オ
+- ニ
+- エ
+- ワ
+- ク
+- ア
+- ガ
+- モ
+- キ
+- ル
+- ジ
+- レ
+- リ
+- ョ
+- ソ
+- ラ
+- ツ
+- チ
+- ケ
+- ユ
+- ド
+- サ
+- セ
+- ヨ
+- ダ
+- ュ
+- ヒ
+- ウ
+- ネ
+- ホ
+- ハ
+- ミ
+- ゴ
+- ロ
+- ブ
+- バ
+- ヤ
+- メ
+- ャ
+- フ
+- ズ
+- ゲ
+- ム
+- ギ
+- グ
+- パ
+- ベ
+- ゼ
+- ビ
+- ザ
+- ヘ
+- ボ
+- ィ
+- ゾ
+- プ
+- ピ
+- ヌ
+- ポ
+- ペ
+- ェ
+- ァ
+- ゥ
+- ォ
+- ヴ
+- ヮ
+- ヅ
+- <sos/eos>
+init: null
+model_conf:
+    ignore_id: 0
+use_preprocessor: true
+token_type: char
+bpemodel: null
+non_linguistic_symbols: null
+cleaner: null
+g2p: null
+lm: seq_rnn
+lm_conf:
+    rnn_type: lstm
+    nlayers: 2
+    unit: 650
+required:
+- output_dir
+- token_list
+version: '202211'
+distributed: true

exp/lm_train_lm_ja_char/valid.loss.ave.pth ADDED Viewed

	@@ -0,0 +1,3 @@

+version https://git-lfs.github.com/spec/v1
+oid sha256:f6ded470555649ab4dec4c8403de325e81743c443006a0dadf18399836145df3
+size 27520991

meta.yaml ADDED Viewed

	@@ -0,0 +1,6 @@

+files:
+  asr_model_file: exp/asr_train_asr_streaming_transformer_ja_raw_jp_char_sp/valid.acc.ave_10best.pth
+  lm_file: exp/lm_train_lm_ja_char/valid.loss.ave.pth
+yaml_files:
+  asr_train_config: exp/asr_train_asr_streaming_transformer_ja_raw_jp_char_sp/config.yaml
+  lm_train_config: exp/lm_train_lm_ja_char/config.yaml