Hugging Face's logo

bigmorning
/
whisper_new_split_0020

Automatic Speech Recognition
Transformers
TensorFlow
whisper
generated_from_keras_callback
Model card Files
xet
 Community
bigmorning commited on
Commit
e4cff75
·
1 Parent(s): 2d2566e

Upload TFWhisperForConditionalGeneration

Browse files
Files changed (3) hide show
  1. README.md +25 -25
  2. config.json +14 -2
  3. tf_model.h5 +1 -1
README.md CHANGED
@@ -14,12 +14,12 @@ probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
- - Train Loss: 0.0862
18
- - Train Accuracy: 0.0331
19
- - Train Wermet: 10.7710
20
- - Validation Loss: 0.4677
21
  - Validation Accuracy: 0.0313
22
- - Validation Wermet: 6.6251
23
  - Epoch: 19
24
 
25
  ## Model description
@@ -46,26 +46,26 @@ The following hyperparameters were used during training:
46
 
47
  | Train Loss | Train Accuracy | Train Wermet | Validation Loss | Validation Accuracy | Validation Wermet | Epoch |
48
  |:----------:|:--------------:|:------------:|:---------------:|:-------------------:|:-----------------:|:-----:|
49
- | 5.0901 | 0.0113 | 53.3790 | 4.4090 | 0.0122 | 42.3548 | 0 |
50
- | 4.3135 | 0.0127 | 42.3551 | 3.9430 | 0.0149 | 37.1045 | 1 |
51
- | 3.3458 | 0.0173 | 31.6069 | 2.3945 | 0.0222 | 25.5461 | 2 |
52
- | 1.9669 | 0.0232 | 13.7935 | 1.4966 | 0.0261 | 6.9562 | 3 |
53
- | 1.2830 | 0.0262 | 10.0196 | 1.1100 | 0.0279 | 9.5683 | 4 |
54
- | 0.9517 | 0.0278 | 8.1513 | 0.9065 | 0.0289 | 7.8180 | 5 |
55
- | 0.7555 | 0.0287 | 7.5457 | 0.7892 | 0.0295 | 5.1479 | 6 |
56
- | 0.6204 | 0.0295 | 7.0748 | 0.7025 | 0.0299 | 6.9938 | 7 |
57
- | 0.5202 | 0.0300 | 7.2085 | 0.6409 | 0.0303 | 7.6979 | 8 |
58
- | 0.4418 | 0.0305 | 6.6665 | 0.5963 | 0.0305 | 4.9877 | 9 |
59
- | 0.3773 | 0.0309 | 6.3833 | 0.5633 | 0.0307 | 5.6072 | 10 |
60
- | 0.3239 | 0.0313 | 6.3658 | 0.5361 | 0.0308 | 9.7748 | 11 |
61
- | 0.2784 | 0.0316 | 7.6413 | 0.5146 | 0.0310 | 8.5224 | 12 |
62
- | 0.2390 | 0.0319 | 8.3862 | 0.5053 | 0.0310 | 8.1694 | 13 |
63
- | 0.2049 | 0.0321 | 8.4188 | 0.4899 | 0.0311 | 9.4708 | 14 |
64
- | 0.1749 | 0.0323 | 8.7733 | 0.4805 | 0.0312 | 8.5083 | 15 |
65
- | 0.1480 | 0.0326 | 8.1859 | 0.4735 | 0.0312 | 16.2408 | 16 |
66
- | 0.1242 | 0.0328 | 10.7089 | 0.4745 | 0.0312 | 6.8974 | 17 |
67
- | 0.1042 | 0.0329 | 10.2003 | 0.4675 | 0.0313 | 9.7003 | 18 |
68
- | 0.0862 | 0.0331 | 10.7710 | 0.4677 | 0.0313 | 6.6251 | 19 |
69
 
70
 
71
  ### Framework versions
 
14
 
15
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Train Loss: 0.0899
18
+ - Train Accuracy: 0.0330
19
+ - Train Wermet: 24.8354
20
+ - Validation Loss: 0.4715
21
  - Validation Accuracy: 0.0313
22
+ - Validation Wermet: 21.1618
23
  - Epoch: 19
24
 
25
  ## Model description
 
46
 
47
  | Train Loss | Train Accuracy | Train Wermet | Validation Loss | Validation Accuracy | Validation Wermet | Epoch |
48
  |:----------:|:--------------:|:------------:|:---------------:|:-------------------:|:-----------------:|:-----:|
49
+ | 5.1027 | 0.0113 | 52.5530 | 4.4267 | 0.0121 | 41.4796 | 0 |
50
+ | 4.3285 | 0.0126 | 38.6893 | 3.9835 | 0.0145 | 33.6050 | 1 |
51
+ | 3.4573 | 0.0168 | 30.7714 | 2.5568 | 0.0215 | 31.7559 | 2 |
52
+ | 2.0878 | 0.0226 | 20.5131 | 1.5738 | 0.0257 | 21.2159 | 3 |
53
+ | 1.3529 | 0.0258 | 17.4367 | 1.1712 | 0.0276 | 17.7695 | 4 |
54
+ | 0.9953 | 0.0275 | 18.7308 | 0.9389 | 0.0287 | 20.5259 | 5 |
55
+ | 0.7852 | 0.0286 | 18.5731 | 0.8074 | 0.0294 | 17.6576 | 6 |
56
+ | 0.6428 | 0.0293 | 18.2945 | 0.7219 | 0.0298 | 19.9850 | 7 |
57
+ | 0.5384 | 0.0299 | 18.9258 | 0.6610 | 0.0301 | 18.9327 | 8 |
58
+ | 0.4565 | 0.0304 | 19.0749 | 0.6117 | 0.0304 | 21.9796 | 9 |
59
+ | 0.3901 | 0.0308 | 19.2099 | 0.5693 | 0.0306 | 18.0965 | 10 |
60
+ | 0.3348 | 0.0312 | 20.4777 | 0.5449 | 0.0307 | 19.9518 | 11 |
61
+ | 0.2877 | 0.0315 | 20.3181 | 0.5232 | 0.0309 | 20.4017 | 12 |
62
+ | 0.2471 | 0.0318 | 19.2073 | 0.5057 | 0.0310 | 18.7612 | 13 |
63
+ | 0.2120 | 0.0320 | 19.0961 | 0.4925 | 0.0311 | 22.3187 | 14 |
64
+ | 0.1809 | 0.0323 | 20.7944 | 0.4849 | 0.0311 | 27.2314 | 15 |
65
+ | 0.1539 | 0.0325 | 22.0951 | 0.4787 | 0.0312 | 25.2171 | 16 |
66
+ | 0.1299 | 0.0327 | 22.7652 | 0.4733 | 0.0312 | 22.7492 | 17 |
67
+ | 0.1087 | 0.0329 | 25.2223 | 0.4701 | 0.0312 | 28.9044 | 18 |
68
+ | 0.0899 | 0.0330 | 24.8354 | 0.4715 | 0.0313 | 21.1618 | 19 |
69
 
70
 
71
  ### Framework versions
config.json CHANGED
@@ -8,7 +8,6 @@
8
  ],
9
  "attention_dropout": 0.0,
10
  "begin_suppress_tokens": [
11
- 220,
12
  50257
13
  ],
14
  "bos_token_id": 50257,
@@ -24,7 +23,20 @@
24
  "encoder_layerdrop": 0.0,
25
  "encoder_layers": 4,
26
  "eos_token_id": 50257,
27
- "forced_decoder_ids": null,
 
 
 
 
 
 
 
 
 
 
 
 
 
28
  "init_std": 0.02,
29
  "is_encoder_decoder": true,
30
  "mask_feature_length": 10,
 
8
  ],
9
  "attention_dropout": 0.0,
10
  "begin_suppress_tokens": [
 
11
  50257
12
  ],
13
  "bos_token_id": 50257,
 
23
  "encoder_layerdrop": 0.0,
24
  "encoder_layers": 4,
25
  "eos_token_id": 50257,
26
+ "forced_decoder_ids": [
27
+ [
28
+ 1,
29
+ 50289
30
+ ],
31
+ [
32
+ 2,
33
+ 50359
34
+ ],
35
+ [
36
+ 3,
37
+ 50363
38
+ ]
39
+ ],
40
  "init_std": 0.02,
41
  "is_encoder_decoder": true,
42
  "mask_feature_length": 10,
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:39c6968d9778a568fbc4368e2ab827ba8fe96ee22b273116fa7a730187afa795
3
  size 151253984
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7f82a30961ebcec0bad19ab68e1915b4c2f9aa0c58127b39e272f40cf0d28335
3
  size 151253984