Hugging Face's logo

bigmorning
/
whisper_new_split_ch_08__0010

Automatic Speech Recognition
Transformers
TensorFlow
whisper
generated_from_keras_callback
Inference Endpoints
Model card Files Community
bigmorning commited on
Commit
989e0e0
•
1 Parent(s): 9de1963

Upload TFWhisperForConditionalGeneration

Browse files
Files changed (3) hide show
  1. README.md +15 -18
  2. config.json +1 -0
  3. tf_model.h5 +1 -1
README.md CHANGED
@@ -15,12 +15,9 @@ probably proofread and complete it, then remove this comment. -->
15
 
16
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
- - Train Loss: 2.8290
19
- - Train Accuracy: 0.0380
20
- - Train Wermet: 0.1529
21
- - Validation Loss: 2.8448
22
- - Validation Accuracy: 0.0369
23
- - Validation Wermet: 0.0515
24
  - Epoch: 9
25
 
26
  ## Model description
@@ -45,18 +42,18 @@ The following hyperparameters were used during training:
45
 
46
  ### Training results
47
 
48
- | Train Loss | Train Accuracy | Train Wermet | Validation Loss | Validation Accuracy | Validation Wermet | Epoch |
49
- |:----------:|:--------------:|:------------:|:---------------:|:-------------------:|:-----------------:|:-----:|
50
- | 5.1494 | 0.0196 | 3.7403 | 4.3409 | 0.0256 | 0.7553 | 0 |
51
- | 4.0236 | 0.0274 | 0.6316 | 3.6960 | 0.0302 | 0.1658 | 1 |
52
- | 3.5216 | 0.0312 | 0.1467 | 3.3497 | 0.0311 | 0.0012 | 2 |
53
- | 3.2259 | 0.0330 | 0.0850 | 3.1532 | 0.0337 | 0.0023 | 3 |
54
- | 3.0948 | 0.0343 | 0.0859 | 3.0780 | 0.0341 | 0.0021 | 4 |
55
- | 3.0263 | 0.0346 | 0.0848 | 3.0221 | 0.0343 | 0.0019 | 5 |
56
- | 2.9743 | 0.0354 | 0.0865 | 2.9779 | 0.0347 | 0.0045 | 6 |
57
- | 2.9255 | 0.0361 | 0.0990 | 2.9337 | 0.0352 | 0.0174 | 7 |
58
- | 2.8788 | 0.0367 | 0.1068 | 2.8909 | 0.0365 | 0.0365 | 8 |
59
- | 2.8290 | 0.0380 | 0.1529 | 2.8448 | 0.0369 | 0.0515 | 9 |
60
 
61
 
62
  ### Framework versions
 
15
 
16
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on an unknown dataset.
17
  It achieves the following results on the evaluation set:
18
+ - Train Loss: 3.5190
19
+ - Train Accuracy: 0.0142
20
+ - Train Wermet: 6.2910
 
 
 
21
  - Epoch: 9
22
 
23
  ## Model description
 
42
 
43
  ### Training results
44
 
45
+ | Train Loss | Train Accuracy | Train Wermet | Epoch |
46
+ |:----------:|:--------------:|:------------:|:-----:|
47
+ | 7.0502 | 0.0020 | 5.6150 | 0 |
48
+ | 5.6527 | 0.0046 | 11.3162 | 1 |
49
+ | 4.8654 | 0.0085 | 4.8190 | 2 |
50
+ | 4.4295 | 0.0107 | 5.4574 | 3 |
51
+ | 4.1588 | 0.0111 | 6.9059 | 4 |
52
+ | 3.9861 | 0.0112 | 6.9104 | 5 |
53
+ | 3.8010 | 0.0121 | 6.8845 | 6 |
54
+ | 3.6776 | 0.0124 | 6.9533 | 7 |
55
+ | 3.5939 | 0.0135 | 6.6166 | 8 |
56
+ | 3.5190 | 0.0142 | 6.2910 | 9 |
57
 
58
 
59
  ### Framework versions
config.json CHANGED
@@ -8,6 +8,7 @@
8
  ],
9
  "attention_dropout": 0.0,
10
  "begin_suppress_tokens": [
 
11
  50257
12
  ],
13
  "bos_token_id": 50257,
 
8
  ],
9
  "attention_dropout": 0.0,
10
  "begin_suppress_tokens": [
11
+ 220,
12
  50257
13
  ],
14
  "bos_token_id": 50257,
tf_model.h5 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:178fbd0531ac11052d90f5a05b271ef7d25811f934cb2e28954fabd05f8badc5
3
  size 151253960
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:724251136a72b5964b9e5c918cd31b237966633542fce8b02b2a8e6a3b13e7f9
3
  size 151253960