Yoshiki commited on
Commit
1f58bde
1 Parent(s): 6a61ad5

Update model

Browse files
Files changed (20) hide show
  1. README.md +71 -0
  2. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/RESULTS.md +44 -0
  3. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/config.yaml +0 -0
  4. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/acc.png +0 -0
  5. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/backward_time.png +0 -0
  6. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/cer.png +0 -0
  7. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/clip.png +0 -0
  8. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/forward_time.png +0 -0
  9. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/gpu_max_cached_mem_GB.png +0 -0
  10. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/grad_norm.png +0 -0
  11. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/iter_time.png +0 -0
  12. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/loss.png +0 -0
  13. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/loss_att.png +0 -0
  14. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/loss_scale.png +0 -0
  15. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/optim0_lr0.png +0 -0
  16. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/optim_step_time.png +0 -0
  17. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/train_time.png +0 -0
  18. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/wer.png +0 -0
  19. exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/valid.acc.ave_3best.pth +3 -0
  20. meta.yaml +8 -0
README.md CHANGED
@@ -1,3 +1,74 @@
1
  ---
 
 
 
 
 
 
 
2
  license: cc-by-4.0
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ tags:
3
+ - espnet
4
+ - audio
5
+ - automatic-speech-recognition
6
+ language: en
7
+ datasets:
8
+ - wsj
9
  license: cc-by-4.0
10
  ---
11
+
12
+ ## ESPnet2 ASR model
13
+
14
+ ### `espnet/yoshiki_wsj_whisper_medium_finetuning`
15
+
16
+ This model was trained by Yoshiki using wsj recipe in [espnet](https://github.com/espnet/espnet/).
17
+
18
+ ### Demo: How to use in ESPnet2
19
+
20
+ Follow the [ESPnet installation instructions](https://espnet.github.io/espnet/installation.html)
21
+ if you haven't done that already.
22
+
23
+ ```bash
24
+ cd espnet
25
+ git checkout fe00740b80cd26fad7c550cd9e975609deb664db
26
+ pip install -e .
27
+ cd egs2/wsj/asr1
28
+ ./run.sh --skip_data_prep false --skip_train true --download_model espnet/yoshiki_wsj_whisper_medium_finetuning
29
+ ```
30
+
31
+ <!-- Generated by scripts/utils/show_asr_result.sh -->
32
+ # RESULTS
33
+ ## Environments
34
+ - date: `Fri Jul 21 12:47:17 JST 2023`
35
+ - python version: `3.10.10 (main, Mar 21 2023, 18:45:11) [GCC 11.2.0]`
36
+ - espnet version: `espnet 202304`
37
+ - pytorch version: `pytorch 1.13.1`
38
+ - Git hash: `d7172fcb7181ffdcca9c0061400254b63e37bf21`
39
+ - Commit date: `Sat Jul 15 15:01:30 2023 +0900`
40
+
41
+ ## /scratch/espnet-hackathon/egs2/wsj/asr1/exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual
42
+ ### WER
43
+
44
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
45
+ |---|---|---|---|---|---|---|---|---|
46
+ |decode_asr_whisper_noctc_greedy_asr_model_valid.acc.ave/test_eval92|333|5564|98.5|1.4|0.1|0.2|1.7|19.5|
47
+
48
+ ### CER
49
+
50
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
51
+ |---|---|---|---|---|---|---|---|---|
52
+ |decode_asr_whisper_noctc_greedy_asr_model_valid.acc.ave/test_eval92|333|32096|99.6|0.2|0.2|0.2|0.6|19.5|
53
+
54
+ ### TER
55
+
56
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
57
+ |---|---|---|---|---|---|---|---|---|
58
+ ## /scratch/espnet-hackathon/egs2/wsj/asr1/exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/decode_asr_whisper_noctc_greedy_asr_model_valid.acc.ave
59
+ ### WER
60
+
61
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
62
+ |---|---|---|---|---|---|---|---|---|
63
+ |org/test_dev93|503|8132|97.7|2.0|0.4|0.3|2.6|24.9|
64
+
65
+ ### CER
66
+
67
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
68
+ |---|---|---|---|---|---|---|---|---|
69
+ |org/test_dev93|503|46683|99.4|0.3|0.3|0.2|0.8|24.9|
70
+
71
+ ### TER
72
+
73
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
74
+ |---|---|---|---|---|---|---|---|---|
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/RESULTS.md ADDED
@@ -0,0 +1,44 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ <!-- Generated by scripts/utils/show_asr_result.sh -->
2
+ # RESULTS
3
+ ## Environments
4
+ - date: `Fri Jul 21 12:47:17 JST 2023`
5
+ - python version: `3.10.10 (main, Mar 21 2023, 18:45:11) [GCC 11.2.0]`
6
+ - espnet version: `espnet 202304`
7
+ - pytorch version: `pytorch 1.13.1`
8
+ - Git hash: `d7172fcb7181ffdcca9c0061400254b63e37bf21`
9
+ - Commit date: `Sat Jul 15 15:01:30 2023 +0900`
10
+
11
+ ## /scratch/espnet-hackathon/egs2/wsj/asr1/exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual
12
+ ### WER
13
+
14
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
15
+ |---|---|---|---|---|---|---|---|---|
16
+ |decode_asr_whisper_noctc_greedy_asr_model_valid.acc.ave/test_eval92|333|5564|98.5|1.4|0.1|0.2|1.7|19.5|
17
+
18
+ ### CER
19
+
20
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
21
+ |---|---|---|---|---|---|---|---|---|
22
+ |decode_asr_whisper_noctc_greedy_asr_model_valid.acc.ave/test_eval92|333|32096|99.6|0.2|0.2|0.2|0.6|19.5|
23
+
24
+ ### TER
25
+
26
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
27
+ |---|---|---|---|---|---|---|---|---|
28
+ ## /scratch/espnet-hackathon/egs2/wsj/asr1/exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/decode_asr_whisper_noctc_greedy_asr_model_valid.acc.ave
29
+ ### WER
30
+
31
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
32
+ |---|---|---|---|---|---|---|---|---|
33
+ |org/test_dev93|503|8132|97.7|2.0|0.4|0.3|2.6|24.9|
34
+
35
+ ### CER
36
+
37
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
38
+ |---|---|---|---|---|---|---|---|---|
39
+ |org/test_dev93|503|46683|99.4|0.3|0.3|0.2|0.8|24.9|
40
+
41
+ ### TER
42
+
43
+ |dataset|Snt|Wrd|Corr|Sub|Del|Ins|Err|S.Err|
44
+ |---|---|---|---|---|---|---|---|---|
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/config.yaml ADDED
The diff for this file is too large to render. See raw diff
 
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/acc.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/backward_time.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/cer.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/clip.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/forward_time.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/gpu_max_cached_mem_GB.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/grad_norm.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/iter_time.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/loss.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/loss_att.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/loss_scale.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/optim0_lr0.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/optim_step_time.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/train_time.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/images/wer.png ADDED
exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/valid.acc.ave_3best.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8124a87993b59790a0cadb4c106d4d05f9b98f575f9a0c2c6885d30ee284cec7
3
+ size 3055787223
meta.yaml ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ espnet: '202304'
2
+ files:
3
+ asr_model_file: exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/valid.acc.ave_3best.pth
4
+ python: 3.10.10 (main, Mar 21 2023, 18:45:11) [GCC 11.2.0]
5
+ timestamp: 1690041531.683268
6
+ torch: 1.13.1
7
+ yaml_files:
8
+ asr_train_config: exp/asr_train_asr_whisper_full_warmup3000_raw_en_whisper_multilingual/config.yaml