Hugging Face's logo

Baselhany
/
fine_tune_Whisper_tiny2

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Arabic
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Baselhany commited on
Commit
bf3ec06
·
verified ·
1 Parent(s): 0865151

Training finished

Browse files
Files changed (3) hide show
  1. README.md +16 -29
  2. model.safetensors +1 -1
  3. runs/Jan17_04-55-18_15a6ae774958/events.out.tfevents.1737103467.15a6ae774958.18.1 +3 -0
README.md CHANGED
@@ -20,9 +20,9 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the quran-ayat-speech-to-text dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.0304
24
- - Wer: 0.1538
25
- - Cer: 0.0502
26
 
27
  ## Model description
28
 
@@ -50,38 +50,25 @@ The following hyperparameters were used during training:
50
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: linear
52
  - lr_scheduler_warmup_steps: 500
53
- - num_epochs: 25
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
- | 0.0214 | 1.0 | 219 | 0.0198 | 0.2299 | 0.0689 |
61
- | 0.0165 | 2.0 | 438 | 0.0198 | 0.2304 | 0.0779 |
62
- | 0.012 | 3.0 | 657 | 0.0195 | 0.2326 | 0.0765 |
63
- | 0.0103 | 4.0 | 876 | 0.0188 | 0.2205 | 0.0719 |
64
- | 0.0062 | 5.0 | 1095 | 0.0191 | 0.1992 | 0.0635 |
65
- | 0.0049 | 6.0 | 1314 | 0.0198 | 0.2170 | 0.0737 |
66
- | 0.0034 | 7.0 | 1533 | 0.0209 | 0.2 | 0.0669 |
67
- | 0.0021 | 8.0 | 1752 | 0.0213 | 0.1937 | 0.0646 |
68
- | 0.0021 | 9.0 | 1971 | 0.0218 | 0.1998 | 0.0617 |
69
- | 0.0013 | 10.0 | 2190 | 0.0225 | 0.1891 | 0.0595 |
70
- | 0.0011 | 11.0 | 2409 | 0.0223 | 0.2074 | 0.0682 |
71
- | 0.0006 | 12.0 | 2628 | 0.0227 | 0.1884 | 0.0631 |
72
- | 0.0005 | 13.0 | 2847 | 0.0238 | 0.1817 | 0.0583 |
73
- | 0.0003 | 14.0 | 3066 | 0.0238 | 0.1763 | 0.0540 |
74
- | 0.0002 | 15.0 | 3285 | 0.0239 | 0.1744 | 0.0583 |
75
- | 0.0001 | 16.0 | 3504 | 0.0248 | 0.1795 | 0.0567 |
76
- | 0.0001 | 17.0 | 3723 | 0.0255 | 0.1727 | 0.0547 |
77
- | 0.0 | 18.0 | 3942 | 0.0254 | 0.1700 | 0.0538 |
78
- | 0.0 | 19.0 | 4161 | 0.0264 | 0.1723 | 0.0525 |
79
- | 0.0 | 20.0 | 4380 | 0.0266 | 0.1599 | 0.0536 |
80
- | 0.0 | 21.0 | 4599 | 0.0273 | 0.1625 | 0.0522 |
81
- | 0.0 | 22.0 | 4818 | 0.0286 | 0.1662 | 0.0545 |
82
- | 0.0 | 23.0 | 5037 | 0.0296 | 0.1560 | 0.0488 |
83
- | 0.0 | 24.0 | 5256 | 0.0309 | 0.1540 | 0.0473 |
84
- | 0.0 | 25.0 | 5475 | 0.0310 | 0.1543 | 0.0501 |
85
 
86
 
87
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the quran-ayat-speech-to-text dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.0191
24
+ - Wer: 0.1462
25
+ - Cer: 0.0494
26
 
27
  ## Model description
28
 
 
50
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
51
  - lr_scheduler_type: linear
52
  - lr_scheduler_warmup_steps: 500
53
+ - num_epochs: 20
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
+ | 0.0148 | 1.0 | 250 | 0.0146 | 0.1685 | 0.0520 |
61
+ | 0.0109 | 2.0 | 500 | 0.0144 | 0.1765 | 0.0595 |
62
+ | 0.0082 | 3.0 | 750 | 0.0152 | 0.1634 | 0.0542 |
63
+ | 0.0065 | 4.0 | 1000 | 0.0155 | 0.1593 | 0.0505 |
64
+ | 0.0033 | 5.0 | 1250 | 0.0181 | 0.1644 | 0.0571 |
65
+ | 0.0025 | 6.0 | 1500 | 0.0182 | 0.1558 | 0.0533 |
66
+ | 0.0024 | 7.0 | 1750 | 0.0180 | 0.1544 | 0.0492 |
67
+ | 0.0014 | 8.0 | 2000 | 0.0190 | 0.1461 | 0.0508 |
68
+ | 0.001 | 9.0 | 2250 | 0.0196 | 0.1430 | 0.0479 |
69
+ | 0.0006 | 10.0 | 2500 | 0.0198 | 0.1467 | 0.0497 |
70
+ | 0.0005 | 11.0 | 2750 | 0.0198 | 0.1531 | 0.0503 |
71
+ | 0.0004 | 12.0 | 3000 | 0.0197 | 0.1446 | 0.0473 |
 
 
 
 
 
 
 
 
 
 
 
 
 
72
 
73
 
74
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:1db03329e6fed72dc835743ea663f1ff7855ab62081e24cd194397637ea93841
3
  size 151061672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:59696a9986b04d5be7e3c8d542646248679995d2958e77813db26f71a585b6b0
3
  size 151061672
runs/Jan17_04-55-18_15a6ae774958/events.out.tfevents.1737103467.15a6ae774958.18.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d4d655ce0b089993e985e176d072236348cf1f4044c42eeb82251b1fcecf783b
3
+ size 453