Hugging Face's logo

Marcusxx
/
chungnam_firestation_model

TensorBoard
Safetensors
Korean
whisper
hf-asr-leaderboard
Generated from Trainer
Model card Files Metrics Community
Marcusxx commited on
Commit
d448fa6
·
verified ·
1 Parent(s): 285ee77

Model save

Browse files
Files changed (4) hide show
  1. README.md +24 -29
  2. generation_config.json +12 -12
  3. model.safetensors +1 -1
  4. runs/Jan07_16-52-02_localhost.localdomain/events.out.tfevents.1736236325.localhost.localdomain.9186.0 +2 -2
README.md CHANGED
@@ -1,13 +1,8 @@
1
  ---
2
- language:
3
- - ko
4
  license: apache-2.0
5
- base_model: openai/whisper-tiny
6
  tags:
7
- - hf-asr-leaderboard
8
  - generated_from_trainer
9
- datasets:
10
- - Marcusxx/chungnam_firestation
11
  model-index:
12
  - name: chungnam_firestation_model
13
  results: []
@@ -18,10 +13,10 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  # chungnam_firestation_model
20
 
21
- This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the Marcusxx/chungnam_firestation dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.0629
24
- - Cer: 63.1942
25
 
26
  ## Model description
27
 
@@ -54,26 +49,26 @@ The following hyperparameters were used during training:
54
 
55
  | Training Loss | Epoch | Step | Validation Loss | Cer |
56
  |:-------------:|:-------:|:----:|:---------------:|:--------:|
57
- | 1.9426 | 0.6623 | 100 | 1.5319 | 162.0432 |
58
- | 0.9551 | 1.3245 | 200 | 0.9569 | 145.6691 |
59
- | 0.6638 | 1.9868 | 300 | 0.6358 | 120.2302 |
60
- | 0.3381 | 2.6490 | 400 | 0.4435 | 117.1511 |
61
- | 0.1798 | 3.3113 | 500 | 0.2969 | 95.3669 |
62
- | 0.132 | 3.9735 | 600 | 0.2135 | 171.2518 |
63
- | 0.0656 | 4.6358 | 700 | 0.1496 | 119.7986 |
64
- | 0.0348 | 5.2980 | 800 | 0.1106 | 69.1511 |
65
- | 0.0273 | 5.9603 | 900 | 0.0907 | 77.3525 |
66
- | 0.0149 | 6.6225 | 1000 | 0.0739 | 58.1295 |
67
- | 0.0103 | 7.2848 | 1100 | 0.0705 | 70.2446 |
68
- | 0.0065 | 7.9470 | 1200 | 0.0675 | 71.0504 |
69
- | 0.0072 | 8.6093 | 1300 | 0.0622 | 52.3741 |
70
- | 0.0036 | 9.2715 | 1400 | 0.0625 | 61.9856 |
71
- | 0.0029 | 9.9338 | 1500 | 0.0620 | 61.8129 |
72
- | 0.0027 | 10.5960 | 1600 | 0.0629 | 67.5396 |
73
- | 0.002 | 11.2583 | 1700 | 0.0629 | 73.0935 |
74
- | 0.0019 | 11.9205 | 1800 | 0.0627 | 52.2878 |
75
- | 0.0018 | 12.5828 | 1900 | 0.0629 | 58.4173 |
76
- | 0.0018 | 13.2450 | 2000 | 0.0629 | 63.1942 |
77
 
78
 
79
  ### Framework versions
 
1
  ---
 
 
2
  license: apache-2.0
3
+ base_model: openai/whisper-medium
4
  tags:
 
5
  - generated_from_trainer
 
 
6
  model-index:
7
  - name: chungnam_firestation_model
8
  results: []
 
13
 
14
  # chungnam_firestation_model
15
 
16
+ This model is a fine-tuned version of [openai/whisper-medium](https://huggingface.co/openai/whisper-medium) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Loss: 0.0727
19
+ - Cer: 36.2014
20
 
21
  ## Model description
22
 
 
49
 
50
  | Training Loss | Epoch | Step | Validation Loss | Cer |
51
  |:-------------:|:-------:|:----:|:---------------:|:--------:|
52
+ | 2.1673 | 0.6623 | 100 | 1.7541 | 180.4317 |
53
+ | 0.2872 | 1.3245 | 200 | 0.3407 | 306.7626 |
54
+ | 0.1676 | 1.9868 | 300 | 0.1805 | 171.6259 |
55
+ | 0.0508 | 2.6490 | 400 | 0.1241 | 80.5180 |
56
+ | 0.0288 | 3.3113 | 500 | 0.1000 | 63.5108 |
57
+ | 0.0388 | 3.9735 | 600 | 0.0704 | 88.1151 |
58
+ | 0.022 | 4.6358 | 700 | 0.0763 | 54.9928 |
59
+ | 0.015 | 5.2980 | 800 | 0.0862 | 31.1655 |
60
+ | 0.0016 | 5.9603 | 900 | 0.0650 | 32.4892 |
61
+ | 0.0053 | 6.6225 | 1000 | 0.0706 | 31.5396 |
62
+ | 0.0006 | 7.2848 | 1100 | 0.0715 | 39.0791 |
63
+ | 0.0004 | 7.9470 | 1200 | 0.0691 | 17.2374 |
64
+ | 0.0003 | 8.6093 | 1300 | 0.0695 | 20.2590 |
65
+ | 0.0003 | 9.2715 | 1400 | 0.0708 | 20.3165 |
66
+ | 0.0003 | 9.9338 | 1500 | 0.0714 | 31.7986 |
67
+ | 0.0003 | 10.5960 | 1600 | 0.0719 | 32.2014 |
68
+ | 0.0003 | 11.2583 | 1700 | 0.0723 | 38.1007 |
69
+ | 0.0003 | 11.9205 | 1800 | 0.0725 | 37.9856 |
70
+ | 0.0003 | 12.5828 | 1900 | 0.0727 | 37.8417 |
71
+ | 0.0003 | 13.2450 | 2000 | 0.0727 | 36.2014 |
72
 
73
 
74
  ### Framework versions
generation_config.json CHANGED
@@ -1,28 +1,28 @@
1
  {
2
  "alignment_heads": [
3
  [
4
- 2,
5
- 2
6
  ],
7
  [
8
- 3,
9
- 0
10
  ],
11
  [
12
- 3,
13
- 2
14
  ],
15
  [
16
- 3,
17
- 3
18
  ],
19
  [
20
- 3,
21
- 4
22
  ],
23
  [
24
- 3,
25
- 5
26
  ]
27
  ],
28
  "begin_suppress_tokens": [
 
1
  {
2
  "alignment_heads": [
3
  [
4
+ 13,
5
+ 15
6
  ],
7
  [
8
+ 15,
9
+ 4
10
  ],
11
  [
12
+ 15,
13
+ 15
14
  ],
15
  [
16
+ 16,
17
+ 1
18
  ],
19
  [
20
+ 20,
21
+ 0
22
  ],
23
  [
24
+ 23,
25
+ 4
26
  ]
27
  ],
28
  "begin_suppress_tokens": [
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f1fff3b6e74ac162ec71e511703ac11f80c7632397ed54e77e010895431f451f
3
  size 3055544304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1c63d5a8531b66fdf3d797d13abdceb564a01abbecc40993856bec7e10c65695
3
  size 3055544304
runs/Jan07_16-52-02_localhost.localdomain/events.out.tfevents.1736236325.localhost.localdomain.9186.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:867b2ff83d50e06b03dbfe87eec96f3adbf089bfb12f87651478ec06c87a7ab5
3
- size 28893
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:176f5a5c3a387e4e274e66ceb9486760738095e3b08f1758fa40edd1ab71a4c6
3
+ size 29247