Hugging Face's logo

Baselhany
/
Graduation_Project_Whisper_tiny

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Arabic
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Baselhany commited on
Commit
160b819
·
verified ·
1 Parent(s): fc0c0d0

Training finished

Browse files
Files changed (3) hide show
  1. README.md +18 -18
  2. model.safetensors +1 -1
  3. runs/Jan31_07-30-19_5d1321445a4c/events.out.tfevents.1738332810.5d1321445a4c.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: 0.0079
24
- - Wer: 0.0858
25
- - Cer: 0.0336
26
 
27
  ## Model description
28
 
@@ -57,21 +57,21 @@ The following hyperparameters were used during training:
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
- | 0.006 | 1.0 | 469 | 0.0063 | 0.0878 | 0.0332 |
61
- | 0.0045 | 2.0 | 938 | 0.0061 | 0.0831 | 0.0303 |
62
- | 0.0028 | 3.0 | 1407 | 0.0063 | 0.0811 | 0.0285 |
63
- | 0.0033 | 4.0 | 1876 | 0.0066 | 0.0813 | 0.0294 |
64
- | 0.0019 | 5.0 | 2345 | 0.0070 | 0.0789 | 0.0288 |
65
- | 0.0018 | 6.0 | 2814 | 0.0075 | 0.0805 | 0.0303 |
66
- | 0.0011 | 7.0 | 3283 | 0.0081 | 0.0813 | 0.0303 |
67
- | 0.001 | 8.0 | 3752 | 0.0083 | 0.1604 | 0.0670 |
68
- | 0.0006 | 9.0 | 4221 | 0.0088 | 0.1605 | 0.0677 |
69
- | 0.0006 | 10.0 | 4690 | 0.0092 | 0.1584 | 0.0663 |
70
- | 0.0005 | 11.0 | 5159 | 0.0094 | 0.1609 | 0.0677 |
71
- | 0.0003 | 12.0 | 5628 | 0.0096 | 0.1598 | 0.0672 |
72
- | 0.0005 | 13.0 | 6097 | 0.0098 | 0.1613 | 0.0681 |
73
- | 0.0005 | 14.0 | 6566 | 0.0112 | 0.1111 | 0.0460 |
74
- | 0.0003 | 15.0 | 7035 | 0.0099 | 0.1602 | 0.0681 |
75
 
76
 
77
  ### 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: 0.0098
24
+ - Wer: 0.0953
25
+ - Cer: 0.0388
26
 
27
  ## Model description
28
 
 
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
+ | 0.0067 | 1.0 | 469 | 0.0060 | 0.0820 | 0.0308 |
61
+ | 0.0074 | 2.0 | 938 | 0.0059 | 0.0822 | 0.0318 |
62
+ | 0.0058 | 3.0 | 1407 | 0.0059 | 0.0782 | 0.0291 |
63
+ | 0.0033 | 4.0 | 1876 | 0.0063 | 0.0793 | 0.0304 |
64
+ | 0.0019 | 5.0 | 2345 | 0.0066 | 0.0787 | 0.0306 |
65
+ | 0.0035 | 6.0 | 2814 | 0.0071 | 0.0767 | 0.0289 |
66
+ | 0.0024 | 7.0 | 3283 | 0.0075 | 0.0771 | 0.0288 |
67
+ | 0.0009 | 8.0 | 3752 | 0.0079 | 0.0767 | 0.0381 |
68
+ | 0.0016 | 9.0 | 4221 | 0.0083 | 0.0764 | 0.0384 |
69
+ | 0.0009 | 10.0 | 4690 | 0.0086 | 0.0760 | 0.0281 |
70
+ | 0.0004 | 11.0 | 5159 | 0.0088 | 0.0744 | 0.0271 |
71
+ | 0.0008 | 12.0 | 5628 | 0.0091 | 0.0753 | 0.0365 |
72
+ | 0.0009 | 13.0 | 6097 | 0.0092 | 0.0762 | 0.0369 |
73
+ | 0.0004 | 14.0 | 6566 | 0.0103 | 0.0960 | 0.0407 |
74
+ | 0.0001 | 15.0 | 7035 | 0.0094 | 0.0769 | 0.0377 |
75
 
76
 
77
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2fb77079bef3950893a65a3d1895d12c296bd281fff882d606aac1672e54c511
3
  size 151061672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b6aa572146b04ac1a0dc45513c65caf79808481e8b3a4fc182b8128d2d05e58c
3
  size 151061672
runs/Jan31_07-30-19_5d1321445a4c/events.out.tfevents.1738332810.5d1321445a4c.18.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b66e248019a382890cdbb1b14ea818257b98921d66bcb8b8ae478b70134e2a55
3
+ size 453