diff --git a/data_respin/bh/nlsyms.txt b/data_respin/bh/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..01e79c32a8c99c557f0757da7cb6d65b3414466d
--- /dev/null
+++ b/data_respin/bh/nlsyms.txt
@@ -0,0 +1,3 @@
+1
+2
+3
diff --git a/data_respin/bn/nlsyms.txt b/data_respin/bn/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..8a1218a1024a212bb3db30becd860315f9f3ac52
--- /dev/null
+++ b/data_respin/bn/nlsyms.txt
@@ -0,0 +1,5 @@
+1
+2
+3
+4
+5
diff --git a/data_respin/ch/nlsyms.txt b/data_respin/ch/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..94ebaf900161394059478fd88aec30e59092a1d7
--- /dev/null
+++ b/data_respin/ch/nlsyms.txt
@@ -0,0 +1,4 @@
+1
+2
+3
+4
diff --git a/data_respin/hi/nlsyms.txt b/data_respin/hi/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..8a1218a1024a212bb3db30becd860315f9f3ac52
--- /dev/null
+++ b/data_respin/hi/nlsyms.txt
@@ -0,0 +1,5 @@
+1
+2
+3
+4
+5
diff --git a/data_respin/kn/nlsyms.txt b/data_respin/kn/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..8a1218a1024a212bb3db30becd860315f9f3ac52
--- /dev/null
+++ b/data_respin/kn/nlsyms.txt
@@ -0,0 +1,5 @@
+1
+2
+3
+4
+5
diff --git a/data_respin/mg/nlsyms.txt b/data_respin/mg/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..94ebaf900161394059478fd88aec30e59092a1d7
--- /dev/null
+++ b/data_respin/mg/nlsyms.txt
@@ -0,0 +1,4 @@
+1
+2
+3
+4
diff --git a/data_respin/mr/nlsyms.txt b/data_respin/mr/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..94ebaf900161394059478fd88aec30e59092a1d7
--- /dev/null
+++ b/data_respin/mr/nlsyms.txt
@@ -0,0 +1,4 @@
+1
+2
+3
+4
diff --git a/data_respin/mt/nlsyms.txt b/data_respin/mt/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..94ebaf900161394059478fd88aec30e59092a1d7
--- /dev/null
+++ b/data_respin/mt/nlsyms.txt
@@ -0,0 +1,4 @@
+1
+2
+3
+4
diff --git a/data_respin/te/nlsyms.txt b/data_respin/te/nlsyms.txt
new file mode 100644
index 0000000000000000000000000000000000000000..94ebaf900161394059478fd88aec30e59092a1d7
--- /dev/null
+++ b/data_respin/te/nlsyms.txt
@@ -0,0 +1,4 @@
+1
+2
+3
+4
diff --git a/exp_small/exp_bh/README.md b/exp_small/exp_bh/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..4a7f1ba5d38b7bed4e13f2e678052afe235051bc
--- /dev/null
+++ b/exp_small/exp_bh/README.md
@@ -0,0 +1,405 @@
+---
+tags:
+- espnet
+- audio
+- automatic-speech-recognition
+language: bh
+datasets:
+- respin_small
+license: cc-by-4.0
+---
+
+## ESPnet2 ASR model
+
+### `SpireLab/spire_respin_baselines_espnet`
+
+This model was trained by wtc7 using respin_small recipe in [espnet](https://github.com/espnet/espnet/).
+
+### Demo: How to use in ESPnet2
+
+Follow the [ESPnet installation instructions](https://espnet.github.io/espnet/installation.html)
+if you haven't done that already.
+
+```bash
+cd espnet
+
+pip install -e .
+cd egs2/respin_small/asr1
+./run.sh --skip_data_prep false --skip_train true --download_model SpireLab/spire_respin_baselines_espnet
+```
+
+
+# RESULTS
+## Environments
+- date: `Sun May 25 02:31:02 IST 2025`
+- python version: `3.8.10 (default, Mar 18 2025, 20:04:55) [GCC 9.4.0]`
+- espnet version: `espnet 202412`
+- pytorch version: `pytorch 2.3.0+cu121`
+- Git hash: `0fe7b8581fbc68841eb48776f052aa9a5989108c`
+ - Commit date: `Tue Jan 14 20:06:15 2025 -0500`
+
+## exp_small/exp_bh/asr_bh_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_lid_asr_model_valid.acc.ave/test_bh|2220|22453|85.9|13.4|0.7|1.1|15.2|73.2|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_lid_asr_model_valid.acc.ave/test_bh|2220|104745|96.8|2.1|1.1|1.2|4.4|73.2|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+
+## ASR config
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e8_mactrue_bs6M_gacc1.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_bh/asr_bh_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+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: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 5
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+- - valid
+ - acc
+ - max
+keep_nbest_models: 5
+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: true
+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: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp_small/exp_bh/asr_stats_raw_bh_char_sp/train/speech_shape
+- exp_small/exp_bh/asr_stats_raw_bh_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_bh/asr_stats_raw_bh_char_sp/valid/speech_shape
+- exp_small/exp_bh/asr_stats_raw_bh_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/bh/raw/train_bh_sp/wav.scp
+ - speech
+ - sound
+- - dump/bh/raw/train_bh_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/bh/raw/dev_bh/wav.scp
+ - speech
+ - sound
+- - dump/bh/raw/dev_bh/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- ा
+- े
+- क
+- र
+- ल
+- स
+- न
+- म
+- त
+- ी
+- ि
+- ह
+- ब
+- ्
+- प
+- व
+- ज
+- ं
+- ो
+- द
+- ख
+- य
+- ग
+- ट
+- ु
+- अ
+- ई
+- इ
+- च
+- भ
+- आ
+- ू
+- उ
+- ए
+- श
+- ै
+- ध
+- ड
+- फ
+- ड़
+- ौ
+- .
+- छ
+- ण
+- ष
+- थ
+- ओ
+- ढ़
+- घ
+- ठ
+- ॉ
+- ृ
+- ढ
+- ऑ
+- ँ
+- ऊ
+- ऋ
+- औ
+- झ
+- ज़
+- फ़
+- ऐ
+- ञ
+- ऽ
+- ख़
+- क़
+- ़
+- ः
+- ॅ
+- ऱ
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 8
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d2
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202409'
+distributed: false
+```
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e8_mactrue_bs6M_gacc1.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_bn/asr_bn_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+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: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 5
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+- - valid
+ - acc
+ - max
+keep_nbest_models: 5
+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: true
+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: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp_small/exp_bn/asr_stats_raw_bn_char_sp/train/speech_shape
+- exp_small/exp_bn/asr_stats_raw_bn_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_bn/asr_stats_raw_bn_char_sp/valid/speech_shape
+- exp_small/exp_bn/asr_stats_raw_bn_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/bn/raw/train_bn_sp/wav.scp
+ - speech
+ - sound
+- - dump/bn/raw/train_bn_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/bn/raw/dev_bn/wav.scp
+ - speech
+ - sound
+- - dump/bn/raw/dev_bn/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- া
+- ে
+- র
+- ক
+- ্
+- ি
+- ন
+- ব
+- ল
+- য
+- ম
+- স
+- ত
+- প
+- ট
+- য়
+- হ
+- ু
+- দ
+- ো
+- জ
+- ই
+- গ
+- চ
+- ছ
+- শ
+- আ
+- থ
+- ভ
+- এ
+- ষ
+- ধ
+- ী
+- উ
+- ফ
+- খ
+- ড
+- অ
+- ং
+- ও
+- ড়
+- ণ
+- ঙ
+- ঁ
+- ৃ
+- .
+- ঠ
+- ৈ
+- ূ
+- ৎ
+- ঞ
+- ঘ
+- ঋ
+- ঝ
+- ৌ
+- ঢ
+- ়
+- ঢ়
+- ঃ
+- ঊ
+- ঐ
+- ঔ
+- ঈ
+- ৠ
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 8
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d2
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202409'
+distributed: false
+```
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e8_mactrue_bs6M_gacc1.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_ch/asr_ch_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+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: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 5
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+- - valid
+ - acc
+ - max
+keep_nbest_models: 5
+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: true
+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: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp_small/exp_ch/asr_stats_raw_ch_char_sp/train/speech_shape
+- exp_small/exp_ch/asr_stats_raw_ch_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_ch/asr_stats_raw_ch_char_sp/valid/speech_shape
+- exp_small/exp_ch/asr_stats_raw_ch_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/ch/raw/train_ch_sp/wav.scp
+ - speech
+ - sound
+- - dump/ch/raw/train_ch_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/ch/raw/dev_ch/wav.scp
+ - speech
+ - sound
+- - dump/ch/raw/dev_ch/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- े
+- ा
+- क
+- र
+- न
+- ल
+- म
+- ह
+- स
+- ब
+- ो
+- ी
+- ि
+- त
+- ज
+- प
+- ं
+- थ
+- य
+- ग
+- द
+- व
+- ख
+- इ
+- ्
+- ु
+- अ
+- ट
+- च
+- ू
+- ए
+- उ
+- भ
+- घ
+- फ
+- आ
+- ड़
+- ध
+- ओ
+- ई
+- ड
+- छ
+- .
+- ँ
+- ै
+- ठ
+- ौ
+- झ
+- ढ़
+- श
+- ढ
+- ण
+- ऊ
+- ॉ
+- ऑ
+- ष
+- ऋ
+- ृ
+- ऐ
+- औ
+- फ़
+- ज़
+- ॅ
+- ः
+- क़
+- ख़
+- ञ
+- ़
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 8
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d2
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202409'
+distributed: false
+```
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e8_mactrue_bs6M_gacc1.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_hi/asr_hi_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+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: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 5
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+- - valid
+ - acc
+ - max
+keep_nbest_models: 5
+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: true
+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: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp_small/exp_hi/asr_stats_raw_hi_char_sp/train/speech_shape
+- exp_small/exp_hi/asr_stats_raw_hi_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_hi/asr_stats_raw_hi_char_sp/valid/speech_shape
+- exp_small/exp_hi/asr_stats_raw_hi_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/hi/raw/train_hi_sp/wav.scp
+ - speech
+ - sound
+- - dump/hi/raw/train_hi_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/hi/raw/dev_hi/wav.scp
+ - speech
+ - sound
+- - dump/hi/raw/dev_hi/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- ा
+- क
+- े
+- र
+- ्
+- त
+- स
+- ी
+- न
+- ह
+- ं
+- ि
+- म
+- ो
+- प
+- ै
+- ल
+- य
+- ज
+- ब
+- व
+- द
+- ग
+- ु
+- ट
+- ए
+- ू
+- श
+- च
+- भ
+- अ
+- ख
+- आ
+- ध
+- ड
+- फ
+- उ
+- ण
+- ई
+- ष
+- इ
+- थ
+- ौ
+- ड़
+- .
+- छ
+- औ
+- ॉ
+- ृ
+- ँ
+- झ
+- ऋ
+- घ
+- ओ
+- ढ़
+- ठ
+- ज़
+- ऑ
+- ऊ
+- ऐ
+- ञ
+- ़
+- फ़
+- ढ
+- ः
+- ख़
+- क़
+- ग़
+- ङ
+- ॅ
+- ऍ
+- ॠ
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 8
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d2
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202409'
+distributed: false
+```
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e8_mactrue_bs6M_gacc1.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_kn/asr_kn_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+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: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 5
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+- - valid
+ - acc
+ - max
+keep_nbest_models: 5
+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: true
+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: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp_small/exp_kn/asr_stats_raw_kn_char_sp/train/speech_shape
+- exp_small/exp_kn/asr_stats_raw_kn_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_kn/asr_stats_raw_kn_char_sp/valid/speech_shape
+- exp_small/exp_kn/asr_stats_raw_kn_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/kn/raw/train_kn_sp/wav.scp
+ - speech
+ - sound
+- - dump/kn/raw/train_kn_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/kn/raw/dev_kn/wav.scp
+ - speech
+ - sound
+- - dump/kn/raw/dev_kn/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- ್
+- ಿ
+- ಾ
+- ರ
+- ು
+- ನ
+- ಕ
+- ತ
+- ದ
+- ೆ
+- ಗ
+- ಸ
+- ಲ
+- ವ
+- ಯ
+- ಂ
+- ಮ
+- ಬ
+- ಳ
+- ಡ
+- ಟ
+- ಹ
+- ಪ
+- ೇ
+- ಅ
+- ೊ
+- ಣ
+- ೋ
+- ಜ
+- ಇ
+- ೂ
+- ಷ
+- ಚ
+- ೀ
+- ಎ
+- ಆ
+- ಶ
+- ೈ
+- ಧ
+- ಒ
+- ಭ
+- .
+- ಉ
+- ಫ
+- ಥ
+- ಖ
+- ೃ
+- ೌ
+- ಏ
+- ಐ
+- ಈ
+- ಠ
+- ಘ
+- ಛ
+- ಓ
+- ಔ
+- ಞ
+- ಊ
+- ಋ
+- ೕ
+- ಢ
+- ಃ
+- ಝ
+- ೖ
+- ೯
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 8
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d2
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202409'
+distributed: false
+```
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e8_mactrue_bs6M_gacc1.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_mg/asr_mg_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+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: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 5
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+- - valid
+ - acc
+ - max
+keep_nbest_models: 5
+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: true
+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: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp_small/exp_mg/asr_stats_raw_mg_char_sp/train/speech_shape
+- exp_small/exp_mg/asr_stats_raw_mg_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_mg/asr_stats_raw_mg_char_sp/valid/speech_shape
+- exp_small/exp_mg/asr_stats_raw_mg_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/mg/raw/train_mg_sp/wav.scp
+ - speech
+ - sound
+- - dump/mg/raw/train_mg_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/mg/raw/dev_mg/wav.scp
+ - speech
+ - sound
+- - dump/mg/raw/dev_mg/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- ा
+- क
+- े
+- र
+- ्
+- स
+- ल
+- न
+- ह
+- त
+- म
+- ि
+- ी
+- प
+- ो
+- ब
+- य
+- ं
+- व
+- ज
+- द
+- ग
+- इ
+- ट
+- ु
+- ई
+- ै
+- ख
+- च
+- छ
+- ू
+- श
+- भ
+- अ
+- आ
+- ध
+- ए
+- ड
+- उ
+- फ
+- ष
+- ण
+- थ
+- ड़
+- ौ
+- .
+- ऽ
+- ृ
+- ॉ
+- औ
+- ढ़
+- घ
+- ठ
+- ँ
+- ओ
+- ऋ
+- ऑ
+- ऊ
+- झ
+- ज़
+- ढ
+- ऐ
+- फ़
+- ञ
+- ः
+- ख़
+- क़
+- ङ
+- ग़
+- ऍ
+- ॅ
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 8
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d2
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202409'
+distributed: false
+```
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e8_mactrue_bs6M_gacc1.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_mr/asr_mr_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+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: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 5
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+- - valid
+ - acc
+ - max
+keep_nbest_models: 5
+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: true
+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: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp_small/exp_mr/asr_stats_raw_mr_char_sp/train/speech_shape
+- exp_small/exp_mr/asr_stats_raw_mr_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_mr/asr_stats_raw_mr_char_sp/valid/speech_shape
+- exp_small/exp_mr/asr_stats_raw_mr_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/mr/raw/train_mr_sp/wav.scp
+ - speech
+ - sound
+- - dump/mr/raw/train_mr_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/mr/raw/dev_mr/wav.scp
+ - speech
+ - sound
+- - dump/mr/raw/dev_mr/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- ा
+- ्
+- र
+- त
+- क
+- े
+- य
+- ी
+- स
+- न
+- व
+- ल
+- म
+- ि
+- प
+- च
+- ं
+- ह
+- ो
+- ज
+- ण
+- द
+- आ
+- ग
+- श
+- ब
+- ु
+- ट
+- ू
+- ड
+- ध
+- अ
+- ख
+- ठ
+- ळ
+- भ
+- ष
+- फ
+- उ
+- ए
+- थ
+- .
+- घ
+- झ
+- ँ
+- ै
+- ई
+- ढ
+- इ
+- ॉ
+- ऊ
+- ॅ
+- ृ
+- ऑ
+- ऱ
+- ओ
+- ौ
+- छ
+- ञ
+- औ
+- ॲ
+- ः
+- ऐ
+- ऍ
+- ऋ
+- ़
+- ':'
+- ड़
+- फ़
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 8
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d2
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202409'
+distributed: false
+```
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e12_mactrue_edrop0.0_ddrop0.0.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_mt/asr_mt_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+num_workers: 4
+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
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 10
+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: 4
+no_forward_run: false
+resume: true
+train_dtype: float32
+use_amp: true
+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: 20
+valid_batch_size: null
+batch_bins: 12000000
+valid_batch_bins: null
+train_shape_file:
+- exp_small/exp_mt/asr_stats_raw_mt_char_sp/train/speech_shape
+- exp_small/exp_mt/asr_stats_raw_mt_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_mt/asr_stats_raw_mt_char_sp/valid/speech_shape
+- exp_small/exp_mt/asr_stats_raw_mt_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/mt/raw/train_mt_sp/wav.scp
+ - speech
+ - sound
+- - dump/mt/raw/train_mt_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/mt/raw/dev_mt/wav.scp
+ - speech
+ - sound
+- - dump/mt/raw/dev_mt/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- ा
+- क
+- र
+- े
+- ्
+- ल
+- स
+- न
+- ि
+- त
+- म
+- ै
+- ी
+- य
+- प
+- ो
+- ब
+- छ
+- ह
+- ज
+- ं
+- व
+- द
+- ग
+- ु
+- ट
+- इ
+- अ
+- भ
+- ख
+- आ
+- श
+- च
+- ए
+- ू
+- ध
+- उ
+- ण
+- ँ
+- ष
+- फ
+- ड
+- थ
+- ड़
+- .
+- ई
+- ृ
+- ौ
+- ॅ
+- ओ
+- ऋ
+- घ
+- ढ़
+- ठ
+- ॉ
+- ऽ
+- ऑ
+- झ
+- ऊ
+- औ
+- ञ
+- ढ
+- ः
+- ऐ
+- फ़
+- ज़
+- ॠ
+- ख़
+- क़
+- ङ
+- ग़
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 12
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202402'
+distributed: false
+```
+
+expand
+
+```
+config: conf/tuning/train_asr_e_branchformer_size256_mlp1024_linear1024_e8_mactrue_bs6M_gacc1.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp_small/exp_te/asr_te_ebf_size256_mlp1024_lin1024_e8_mactrue_bs6M_gacc1
+ngpu: 1
+seed: 2022
+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: false
+sharded_ddp: false
+use_deepspeed: false
+deepspeed_config: null
+cudnn_enabled: true
+cudnn_benchmark: false
+cudnn_deterministic: true
+use_tf32: false
+collect_stats: false
+write_collected_feats: false
+max_epoch: 70
+patience: 5
+val_scheduler_criterion:
+- valid
+- loss
+early_stopping_criterion:
+- valid
+- loss
+- min
+best_model_criterion:
+- - valid
+ - acc
+ - max
+keep_nbest_models: 5
+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: true
+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: 20
+valid_batch_size: null
+batch_bins: 6000000
+valid_batch_bins: null
+category_sample_size: 10
+train_shape_file:
+- exp_small/exp_te/asr_stats_raw_te_char_sp/train/speech_shape
+- exp_small/exp_te/asr_stats_raw_te_char_sp/train/text_shape.char
+valid_shape_file:
+- exp_small/exp_te/asr_stats_raw_te_char_sp/valid/speech_shape
+- exp_small/exp_te/asr_stats_raw_te_char_sp/valid/text_shape.char
+batch_type: numel
+valid_batch_type: null
+fold_length:
+- 80000
+- 150
+sort_in_batch: descending
+shuffle_within_batch: false
+sort_batch: descending
+multiple_iterator: false
+chunk_length: 500
+chunk_shift_ratio: 0.5
+num_cache_chunks: 1024
+chunk_excluded_key_prefixes: []
+chunk_default_fs: null
+chunk_max_abs_length: null
+chunk_discard_short_samples: true
+train_data_path_and_name_and_type:
+- - dump/te/raw/train_te_sp/wav.scp
+ - speech
+ - sound
+- - dump/te/raw/train_te_sp/text
+ - text
+ - text
+valid_data_path_and_name_and_type:
+- - dump/te/raw/dev_te/wav.scp
+ - speech
+ - sound
+- - dump/te/raw/dev_te/text
+ - text
+ - text
+multi_task_dataset: false
+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: 0.002
+ weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+ warmup_steps: 15000
+token_list:
+-
+-
+-
+- ్
+- ు
+- ా
+- ి
+- న
+- ల
+- ం
+- క
+- ర
+- ప
+- త
+- వ
+- ట
+- స
+- ే
+- య
+- డ
+- ద
+- మ
+- చ
+- ో
+- గ
+- ె
+- బ
+- ీ
+- అ
+- ొ
+- ఎ
+- ూ
+- జ
+- ై
+- ఉ
+- ధ
+- ఇ
+- ఆ
+- ష
+- భ
+- శ
+- ఏ
+- ళ
+- ఫ
+- ణ
+- .
+- హ
+- థ
+- ఒ
+- ఖ
+- ఈ
+- ౌ
+- ఐ
+- ృ
+- ఓ
+- ఊ
+- ఋ
+- ఛ
+- ఘ
+- ఠ
+- ఔ
+- ఱ
+- ఢ
+- ఞ
+- ః
+- ౖ
+- ౦
+-
+init: null
+input_size: null
+ctc_conf:
+ dropout_rate: 0.0
+ ctc_type: builtin
+ reduce: true
+ ignore_nan_grad: null
+ zero_infinity: true
+ brctc_risk_strategy: exp
+ brctc_group_strategy: end
+ brctc_risk_factor: 0.0
+joint_net_conf: null
+use_preprocessor: true
+use_lang_prompt: false
+use_nlp_prompt: false
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+ n_fft: 512
+ win_length: 400
+ hop_length: 160
+ 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: 5
+normalize: utterance_mvn
+normalize_conf: {}
+model: espnet
+model_conf:
+ ctc_weight: 0.3
+ lsm_weight: 0.1
+ length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+encoder: e_branchformer
+encoder_conf:
+ output_size: 256
+ attention_heads: 4
+ attention_layer_type: rel_selfattn
+ pos_enc_layer_type: rel_pos
+ rel_pos_type: latest
+ cgmlp_linear_units: 1024
+ cgmlp_conv_kernel: 31
+ use_linear_after_conv: false
+ gate_activation: identity
+ num_blocks: 8
+ dropout_rate: 0.1
+ positional_dropout_rate: 0.1
+ attention_dropout_rate: 0.1
+ input_layer: conv2d2
+ layer_drop_rate: 0.0
+ linear_units: 1024
+ positionwise_layer_type: linear
+ use_ffn: true
+ macaron_ffn: true
+ merge_conv_kernel: 31
+postencoder: null
+postencoder_conf: {}
+decoder: transformer
+decoder_conf:
+ attention_heads: 4
+ 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
+ layer_drop_rate: 0.0
+preprocessor: default
+preprocessor_conf: {}
+required:
+- output_dir
+- token_list
+version: '202409'
+distributed: false
+```
+
+