Upload 72 files

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/36epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f4217df828df017fa016adc7deafb581f60786e5797d8ff74c9442245fdabad
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/37epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bdfa48edb0f3e665835b6f180c134e9e01217e2e766e3fee9280cc444bb4878c
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/38epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f827f2ea8142d01b3cd1ded431be08fd28bfa161a9d397d570110671ba946f5b
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/39epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdfc45a3ce59665376017b5b466a4a46a1f6831f39a50e2980223224c47070bf
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/40epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:37e9df078558505ce825e097b66cd056b82bd8b0953f24d94bf05c3826f1a9a8
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/41epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f659297813d1c971f32058b51ce2937d08d326cedbb8472cada849c123b2432
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/42epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:350674327ad9832917df12de6fc98438484a80da168e5c5be95de56a39b7733e
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/43epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c1d8f733e6ea4a25a9bae32e3d0b816b883f21f98d324e2d688c57ec1474e716
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/44epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41f6f2f1ebc97903480a52f418e71aec4e71861db9aca9bcaa67112687533d68
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/45epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b27ef31ab1e2e5974a590d9f464fc233cb8c99c2e55bf7708f403dd84b2add7b
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/RESULTS.md

@@ -0,0 +1,107 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Mon Mar  4 23:35:01 CST 2024`
+- python version: `3.9.18 (main, Sep 11 2023, 13:41:44)  [GCC 11.2.0]`
+- espnet version: `espnet 202308`
+- pytorch version: `pytorch 1.12.1+cu116`
+- Git hash: `884659f9ee95374811015381c976fa3b4f6e01db`
+  - Commit date: `Thu Nov 23 00:23:29 2023 +0800`
+
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_sot_asr_model_25epoch/test_clean_kaldi_fmt|961|64007|44.2|47.9|7.9|17.8|73.5|99.6|
+|decode_sot_asr_model_25epoch/test_other_kaldi_fmt|992|80370|40.2|50.5|9.3|17.3|77.2|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk|3315|226216|48.8|35.5|15.7|13.1|64.3|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk_kaldi_fmt_mtt|2282|142104|45.8|35.1|19.1|13.1|67.3|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk|2059|209679|34.6|45.3|20.2|9.6|75.1|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk_kaldi_fmt|1004|124462|37.8|43.5|18.7|9.9|72.1|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk|1467|200029|28.3|46.5|25.2|6.4|78.0|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk_kaldi_fmt|721|119166|31.3|45.7|23.0|9.4|78.1|100.0|
+|decode_sot_asr_model_valid.acc.best/sot_sdm1_dev|2382|35243|26.8|64.8|8.4|81.8|155.0|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_2spk|4570|301042|47.8|36.4|15.8|13.7|65.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_3spk|2072|212871|33.0|47.3|19.8|10.4|77.4|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_4spk|1326|185394|27.7|47.0|25.3|6.9|79.1|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_2spk|4663|336490|45.8|38.8|15.4|13.0|67.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_3spk|2453|266074|30.2|50.2|19.6|9.5|79.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_4spk|1795|259138|25.1|49.3|25.6|5.9|80.8|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_kaldi_fmt|2180|178761|49.8|34.6|15.6|13.0|63.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_3spk_kaldi_fmt|977|124741|34.3|47.1|18.5|12.0|77.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_4spk_kaldi_fmt|632|109072|30.3|46.8|22.9|9.8|79.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_kaldi_fmt|2363|205496|47.8|37.3|14.9|11.8|63.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_3spk_kaldi_fmt|1246|162996|31.6|50.6|17.8|11.8|80.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_4spk_kaldi_fmt|901|157123|27.6|49.4|23.0|9.3|81.8|100.0|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_sot_asr_model_25epoch/test_clean_kaldi_fmt|961|329390|67.6|20.4|12.0|18.5|50.9|99.6|
+|decode_sot_asr_model_25epoch/test_other_kaldi_fmt|992|416899|64.3|21.7|14.0|18.2|53.8|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk|3315|1230801|66.7|16.6|16.7|11.9|45.2|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk_kaldi_fmt_mtt|2282|770287|63.3|17.0|19.7|12.2|48.9|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk|2059|1140428|55.3|20.3|24.4|10.3|54.9|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk_kaldi_fmt|1004|677017|58.2|19.1|22.7|10.3|52.1|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk|1467|1087409|49.2|20.3|30.5|7.8|58.6|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk_kaldi_fmt|721|647884|52.2|19.5|28.3|9.9|57.7|100.0|
+|decode_sot_asr_model_valid.acc.best/sot_sdm1_dev|2382|169857|60.5|29.1|10.4|82.0|121.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_2spk|4570|1550429|66.5|17.2|16.3|12.9|46.4|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_3spk|2072|1084475|54.9|22.0|23.1|11.9|56.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_4spk|1326|938467|49.6|21.4|29.1|8.8|59.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_2spk|4663|1742136|65.5|17.9|16.5|12.4|46.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_3spk|2453|1381987|53.3|22.5|24.2|11.1|57.8|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_4spk|1795|1346646|47.7|21.5|30.8|7.9|60.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_kaldi_fmt|2180|921344|67.7|16.3|16.0|11.8|44.1|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_3spk_kaldi_fmt|977|635802|56.4|21.4|22.2|13.2|56.8|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_4spk_kaldi_fmt|632|552325|52.4|20.9|26.7|11.6|59.1|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_kaldi_fmt|2363|1064868|67.0|16.9|16.1|11.0|44.0|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_3spk_kaldi_fmt|1246|847159|55.0|22.6|22.4|13.0|58.0|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_4spk_kaldi_fmt|901|817228|50.5|21.4|28.1|10.4|59.9|100.0|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/decode_sot_asr_model_25epoch
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|47659|45.3|44.2|10.5|15.3|70.0|99.5|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_kaldi_fmt|605|258151|66.1|19.0|14.9|15.2|49.0|99.5|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/decode_sot_asr_model_valid.acc.best
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_2spk_kaldi_fmt|1606|135101|50.7|33.4|15.9|12.0|61.3|100.0|
+|org/sot_sdm1_dev|4142|96496|25.8|65.4|8.8|98.2|172.4|100.0|
+|org/sot_sdm1_eval|2385|37529|22.6|70.1|7.3|78.7|156.1|100.0|
+|org/tt_mix_clean_reverb_max_16k|3000|3000|0.0|100.0|0.0|3936.7|4036.7|100.0|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_2spk_kaldi_fmt|1606|735694|67.8|15.4|16.8|10.4|42.6|100.0|
+|org/sot_sdm1_dev|4142|469537|61.1|28.8|10.2|100.6|139.5|100.0|
+|org/sot_sdm1_eval|2385|183036|57.6|32.1|10.3|76.3|118.7|100.0|
+|org/tt_mix_clean_reverb_max_16k|3000|143026|16.9|83.0|0.1|376.6|459.6|100.0|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/checkpoint.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97b9144b902ec389690dc5de336cd2f7e72606f66a6fabd8d0be4f5bef3d32f2
+size 516804611

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/config.yaml

@@ -0,0 +1,227 @@
+config: conf/tuning/train_sot_asr_conformer.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new
+ngpu: 1
+seed: 0
+num_workers: 16
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 2
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 43895
+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: 60
+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: 4
+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:
+- /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 20
+valid_batch_size: null
+batch_bins: 10000000
+valid_batch_bins: null
+train_shape_file:
+- exp/asr_stats_raw_en_char_sp/train/speech_shape
+- exp/asr_stats_raw_en_char_sp/train/text_shape.char
+valid_shape_file:
+- exp/asr_stats_raw_en_char_sp/valid/speech_shape
+- exp/asr_stats_raw_en_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: []
+train_data_path_and_name_and_type:
+-   - dump/raw/train_small_kaldi_fmt_sp/wav.scp
+    - speech
+    - kaldi_ark
+-   - dump/raw/train_small_kaldi_fmt_sp/text
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/dev_kaldi_fmt/wav.scp
+    - speech
+    - kaldi_ark
+-   - dump/raw/dev_kaldi_fmt/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: 0.002
+    weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+    warmup_steps: 20000
+token_list:
+- <blank>
+- <unk>
+- <sc>
+- <space>
+- E
+- T
+- A
+- O
+- N
+- I
+- H
+- S
+- R
+- D
+- L
+- U
+- M
+- C
+- W
+- F
+- G
+- Y
+- P
+- B
+- V
+- K
+- ''''
+- X
+- J
+- Q
+- Z
+- <sos/eos>
+init: null
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+    fs: 16k
+specaug: null
+specaug_conf: {}
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/asr_stats_raw_en_char_sp/train/feats_stats.npz
+model: espnet
+model_conf:
+    ctc_weight: 0.0
+    lsm_weight: 0.1
+    length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+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: 0.1
+    input_layer: conv2d
+    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: 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
+preprocessor: multi
+preprocessor_conf:
+    speaker_change_symbol:
+    - <sc>
+required:
+- output_dir
+- token_list
+version: '202308'
+distributed: true

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/latest.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b27ef31ab1e2e5974a590d9f464fc233cb8c99c2e55bf7708f403dd84b2add7b
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/run.sh

@@ -0,0 +1 @@
+./asr.sh --lang en --audio_format flac.ark --stage 11 --feats_type raw --token_type char --sot_asr true --max_wav_duration 50 --speed_perturb_factors '0.9 1.0 1.1' --feats_normalize global_mvn --use_lm false --pretrained_model /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth --asr_config conf/tuning/train_sot_asr_conformer.yaml --lm_config conf/tuning/train_lm_transformer.yaml --inference_config conf/tuning/decode_sot.yaml --train_set train_small_kaldi_fmt --valid_set dev_kaldi_fmt --test_sets 'dev_kaldi_fmt test_clean_kaldi_fmt test_other_kaldi_fmt' --ngpu 2 --asr_tag train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new --lm_train_text data/local/other_text/text --bpe_train_text data/train_small_kaldi_fmt/text --stage 11 "$@"; exit $?

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/train/events.out.tfevents.1700139527.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.2101480.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0377748d19696a7f0633ffa2127d6092745fac79a961243e1f853c951e6ba8b2
+size 625891718

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/tensorboard/valid/events.out.tfevents.1700139527.de-74279-k2-train-9-0208143539-7dbf569d4f-r7nrb.2101480.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d54a186f38796fe2f11fb9d18374dd5c6dcbd2f0af0363dd45c6720138bb2cd0
+size 12730

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new/valid.acc.best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b27ef31ab1e2e5974a590d9f464fc233cb8c99c2e55bf7708f403dd84b2add7b
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/51epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f1b493b1613f5da3d77832f43f9c7fe0f50d9f1f200669037d51bb6c3a9b8bd
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/52epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7726bb1977608f1088f61ed42209b7a9bb1836944e59c6baf0fcd2c639cc658c
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/53epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:db4f175139f8a7a3f0165567e214db2edceef3982fd4e20898afe7ef6f5e3a8a
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/54epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9ff56cf1f853ba75bf1849975c7648bf947424186ca3958b03973977ef57e8cc
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/55epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6466de0d2c33ccc74f6ac3dfe5a9dfedcce6534767381c4b367ab5984ed94cbb
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/56epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92f209ba881aa29c698f7272a5e2166646b374b167941865153a6bad1f2a4cff
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/57epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bb9da49349d1027c17302dfdaa9dc307147e68c306723d447097f57d11cadbb
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/58epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:262dc26dbdb3d0433518cb3bb919765f939209630860e0da68a6b4a8024606ca
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/59epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93a466e2bdee6f717582e8f55b8cf7d2d189c6a38d7268196dcebd12b7b0ec2b
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/60epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fc0710e1565fce6d3ce275190a1335a03d912f728a67066516df16d751f26514
+size 172367337

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/RESULTS.md

@@ -0,0 +1,51 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Tue Mar  5 10:57:33 CST 2024`
+- python version: `3.9.18 (main, Sep 11 2023, 13:41:44)  [GCC 11.2.0]`
+- espnet version: `espnet 202308`
+- pytorch version: `pytorch 1.12.1+cu116`
+- Git hash: `884659f9ee95374811015381c976fa3b4f6e01db`
+  - Commit date: `Thu Nov 23 00:23:29 2023 +0800`
+
+## exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_sot_asr_model_valid.acc.best/dev_2spk|3315|226216|58.5|26.0|15.5|12.5|54.0|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk_kaldi_fmt|1606|135101|62.0|23.4|14.6|10.4|48.3|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk|2059|209679|41.4|38.0|20.5|9.3|67.8|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk|1467|200029|33.3|41.0|25.7|6.5|73.3|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_2spk|4570|301042|58.6|26.3|15.1|12.4|53.8|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_3spk|2072|212871|40.4|38.8|20.8|10.8|70.4|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_4spk|1326|185394|32.5|41.2|26.3|7.1|74.6|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_2spk|4663|336490|54.9|30.5|14.6|12.4|57.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_3spk|2453|266074|37.0|43.0|20.0|10.0|73.1|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_4spk|1795|259138|29.2|44.9|25.9|6.4|77.3|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_kaldi_fmt|2180|178761|62.5|24.0|13.5|10.1|47.6|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_kaldi_fmt|2363|205496|58.1|28.9|13.1|10.1|52.1|100.0|
+|decode_sot_asr_model_valid.acc.best/tt_mix_clean_reverb_max_16k|3000|3000|0.0|100.0|0.0|3115.7|3215.7|100.0|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_sot_asr_model_valid.acc.best/dev_2spk|3315|1230801|72.5|12.8|14.6|10.4|37.9|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_2spk_kaldi_fmt|1606|735694|75.0|11.2|13.8|8.2|33.2|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk|2059|1140428|59.7|17.4|22.9|9.2|49.5|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk|1467|1087409|52.4|18.0|29.6|7.4|54.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_2spk|4570|1550429|73.0|13.0|14.0|10.5|37.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_3spk|2072|1084475|59.1|18.6|22.3|10.8|51.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_4spk|1326|938467|52.2|18.9|28.9|8.2|56.0|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_2spk|4663|1742136|71.1|14.7|14.1|10.8|39.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_3spk|2453|1381987|57.5|19.5|23.0|10.7|53.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_4spk|1795|1346646|50.2|19.8|30.1|7.8|57.6|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_kaldi_fmt|2180|921344|75.9|11.5|12.6|8.3|32.3|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_kaldi_fmt|2363|1064868|73.7|13.3|13.0|8.8|35.1|100.0|
+|decode_sot_asr_model_valid.acc.best/tt_mix_clean_reverb_max_16k|3000|143026|16.3|83.6|0.2|298.3|382.0|100.0|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/checkpoint.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a89e7a0173da35e614c843035bef69d1f419c73a219c247975e38b71dd606062
+size 516972446

small/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr/config.yaml

@@ -0,0 +1,227 @@
+config: conf/tuning/train_sot_asr_conformer.yaml
+print_config: false
+log_level: INFO
+drop_last_iter: false
+dry_run: false
+iterator_type: sequence
+valid_iterator_type: null
+output_dir: exp/asr_train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr
+ngpu: 1
+seed: 0
+num_workers: 16
+num_att_plot: 3
+dist_backend: nccl
+dist_init_method: env://
+dist_world_size: 4
+dist_rank: 0
+local_rank: 0
+dist_master_addr: localhost
+dist_master_port: 53697
+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: 60
+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: 4
+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:
+- /star-home/jinzengrui/dev/espnet/egs2/librimix/sot_asr1_pretrain/exp/asr_train_sot_asr_conformer_raw_en_char_sp/45epoch.pth
+ignore_init_mismatch: false
+freeze_param: []
+num_iters_per_epoch: null
+batch_size: 20
+valid_batch_size: null
+batch_bins: 10000000
+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
+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: []
+train_data_path_and_name_and_type:
+-   - dump/raw/small_w_whamr_sp/wav.scp
+    - speech
+    - kaldi_ark
+-   - dump/raw/small_w_whamr_sp/text
+    - text
+    - text
+valid_data_path_and_name_and_type:
+-   - dump/raw/cv_mix_clean_reverb_max_16k/wav.scp
+    - speech
+    - kaldi_ark
+-   - dump/raw/cv_mix_clean_reverb_max_16k/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: 0.002
+    weight_decay: 1.0e-06
+scheduler: warmuplr
+scheduler_conf:
+    warmup_steps: 20000
+token_list:
+- <blank>
+- <unk>
+- <sc>
+- <space>
+- E
+- T
+- A
+- O
+- N
+- I
+- H
+- S
+- R
+- D
+- L
+- U
+- M
+- C
+- W
+- F
+- G
+- Y
+- P
+- B
+- V
+- K
+- ''''
+- X
+- J
+- Q
+- Z
+- <sos/eos>
+init: null
+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
+aux_ctc_tasks: []
+frontend: default
+frontend_conf:
+    fs: 16k
+specaug: null
+specaug_conf: {}
+normalize: global_mvn
+normalize_conf:
+    stats_file: exp/asr_stats_raw_en_char/train/feats_stats.npz
+model: espnet
+model_conf:
+    ctc_weight: 0.0
+    lsm_weight: 0.1
+    length_normalized_loss: false
+preencoder: null
+preencoder_conf: {}
+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: 0.1
+    input_layer: conv2d
+    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: 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
+preprocessor: multi
+preprocessor_conf:
+    speaker_change_symbol:
+    - <sc>
+required:
+- output_dir
+- token_list
+version: '202308'
+distributed: true