Upload 36 files

whamr_only/51epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5acd362a3cb79f46930aee1277b508d21c3827dbac6439018690f6c03676e3e4
+size 172367337

whamr_only/52epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d2bb10a024f9bf6b8ce278ec89504f31077831f8131552a050863d3e41a2c12
+size 172367337

whamr_only/53epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:22bb11d3c7e6407b83bab483bab65d33f726e90ec3054f72e42d9ca5e6d1d05b
+size 172367337

whamr_only/54epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:55751dfbafe484a89b4932998a8a562f9a38652f98df6ea1d1ad4ff8f10a4d92
+size 172367337

whamr_only/55epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:80ff99937fd1827a3d4477f8a3416ccbe8956f6ef8e1e694c68d1f2f7402bd0d
+size 172367337

whamr_only/56epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2193e1c399a35bba42860e80432ca78426092d4ca9a3312098db7193dd730252
+size 172367337

whamr_only/57epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1175ff22c0a9f3523af30e2dbe35ccfeefa07a709170089059898c6389b083ff
+size 172367337

whamr_only/58epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d528953b7b2b1c8791279d243744450414be15e61123fd2b9216083c2bbb8213
+size 172367337

whamr_only/59epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1fcfc4f27cac05ed4176327b86ae49d5c65096332481a8fc18fee7051a1d5858
+size 172367337

whamr_only/60epoch.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4f732edcb7d3085b74db6df2cc49208ca2424933e4c06f5f2823b737fbf4964
+size 172367337

whamr_only/RESULTS.md

@@ -0,0 +1,66 @@
+<!-- Generated by scripts/utils/show_asr_result.sh -->
+# RESULTS
+## Environments
+- date: `Mon Mar  4 20:54:29 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_only
+### WER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_sot_asr_model_valid.acc.best/dev_2spk|3315|226216|14.7|50.1|35.2|3.9|89.2|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk|2059|209679|11.3|38.3|50.4|1.2|89.8|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk|1467|200029|9.7|30.3|60.0|0.6|90.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_2spk|4570|301042|13.6|51.0|35.4|3.8|90.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_3spk|2072|212871|10.7|38.7|50.6|1.1|90.4|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_4spk|1326|185394|9.3|29.4|61.3|0.6|91.3|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_2spk|4663|336490|12.4|50.7|36.9|3.0|90.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_3spk|2453|266074|9.6|38.5|51.8|0.9|91.2|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_4spk|1795|259138|8.4|29.5|62.1|0.5|92.1|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_kaldi_fmt|2180|178761|13.8|44.8|41.4|2.3|88.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_kaldi_fmt|2363|205496|12.6|44.9|42.5|1.9|89.3|100.0|
+|decode_sot_asr_model_valid.acc.best/tt_mix_clean_reverb_max_16k|3000|3000|0.0|100.0|0.0|3139.9|3239.9|100.0|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|decode_sot_asr_model_valid.acc.best/dev_2spk|3315|1230801|38.3|22.6|39.0|7.4|69.1|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_3spk|2059|1140428|31.0|16.9|52.1|2.9|71.8|100.0|
+|decode_sot_asr_model_valid.acc.best/dev_4spk|1467|1087409|26.8|12.5|60.7|1.5|74.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_2spk|4570|1550429|38.0|23.4|38.6|7.8|69.8|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_3spk|2072|1084475|31.3|17.8|50.9|3.2|71.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test-clean_4spk|1326|938467|26.8|12.7|60.5|1.5|74.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_2spk|4663|1742136|37.1|23.4|39.5|7.0|69.9|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_3spk|2453|1381987|30.2|17.3|52.4|2.8|72.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test-other_4spk|1795|1346646|25.9|12.4|61.7|1.3|75.5|100.0|
+|decode_sot_asr_model_valid.acc.best/test_clean_2spk_kaldi_fmt|2180|921344|36.4|20.0|43.7|5.1|68.7|100.0|
+|decode_sot_asr_model_valid.acc.best/test_other_2spk_kaldi_fmt|2363|1064868|35.5|20.1|44.3|4.6|69.1|100.0|
+|decode_sot_asr_model_valid.acc.best/tt_mix_clean_reverb_max_16k|3000|143026|16.2|83.6|0.1|294.7|378.4|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_whamr_only/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|15.0|43.3|41.8|2.5|87.5|100.0|
+
+### CER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|
+|org/dev_2spk_kaldi_fmt|1606|735694|36.4|19.0|44.5|4.7|68.2|100.0|
+
+### TER
+
+|dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
+|---|---|---|---|---|---|---|---|---|

whamr_only/checkpoint.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:164e2e87550cbb3ed02f9b49dcd7ff27a0c5af8590e73476b1bd69587e13c487
+size 516972510

whamr_only/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_only
+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: 54011
+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/tr_mix_clean_reverb_max_16k_sp/wav.scp
+    - speech
+    - kaldi_ark
+-   - dump/raw/tr_mix_clean_reverb_max_16k_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

whamr_only/latest.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4f732edcb7d3085b74db6df2cc49208ca2424933e4c06f5f2823b737fbf4964
+size 172367337

whamr_only/run.sh

@@ -0,0 +1 @@
+./asr.sh --lang en --audio_format flac.ark --stage 6 --stop_stage 15 --speed_perturb_factors '0.9 1.0 1.1' --feats_type raw --token_type char --sot_asr true --max_wav_duration 50 --speed_perturb_factors '' --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 tr_mix_clean_reverb_max_16k --valid_set cv_mix_clean_reverb_max_16k --test_sets tt_mix_clean_reverb_max_16k --ngpu 2 --asr_tag train_sot_asr_conformer_raw_en_char_sp_finetune_ls100_45epoch_new_whamr_only --lm_train_text data/local/other_text/text --bpe_train_text data/tr_mix_clean_reverb_max_16k/text --stage 11 "$@"; exit $?

whamr_only/tensorboard/train/events.out.tfevents.1708415307.de-74279-k2-train-7-1218101249-5bcbfb5567-jsftr.946846.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc726117ceea4f047b43a54ccb4085a699b3c7c06e08da30e5e4c718f8b77098
+size 810259841

whamr_only/tensorboard/valid/events.out.tfevents.1708415308.de-74279-k2-train-7-1218101249-5bcbfb5567-jsftr.946846.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36576199cf0a9a358be6ff8cbeb97275d530ce3896f3bb091ea7398f1faa14b9
+size 16918

whamr_only/valid.acc.ave.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b28c5f6d1c3ff5b939be7b4169b60eeb2f33e37c2ed4190b9c632f2d75e300cf
+size 172358249

whamr_only/valid.acc.ave_10best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b28c5f6d1c3ff5b939be7b4169b60eeb2f33e37c2ed4190b9c632f2d75e300cf
+size 172358249

whamr_only/valid.acc.best.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d4f732edcb7d3085b74db6df2cc49208ca2424933e4c06f5f2823b737fbf4964
+size 172367337