Hugging Face's logo

YoussefAshmawy
/
Graduation_Project_Whisper_base

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Arabic
whisper
Generated from Trainer
arabic
quran
Eval Results
Model card Files
xet
Metrics Community
YoussefAshmawy commited on
Commit
65eb061
·
verified ·
1 Parent(s): 732d26e

Training finished

Browse files
Files changed (3) hide show
  1. README.md +29 -29
  2. model.safetensors +1 -1
  3. runs/Feb24_06-59-30_12d9d8bca2af/events.out.tfevents.1740419339.12d9d8bca2af.23.1 +3 -0
README.md CHANGED
@@ -21,8 +21,8 @@ should probably proofread and complete it, then remove this comment. -->
21
  This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the quran-ayat-speech-to-text dataset.
22
  It achieves the following results on the evaluation set:
23
  - Loss: 0.0088
24
- - Wer: 0.0676
25
- - Cer: 0.0268
26
 
27
  ## Model description
28
 
@@ -55,33 +55,33 @@ The following hyperparameters were used during training:
55
 
56
  ### Training results
57
 
58
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
- |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
60
- | 0.0015 | 1.0 | 500 | 0.0068 | 0.0731 | 0.0283 |
61
- | 0.0011 | 2.0 | 1000 | 0.0081 | 0.0753 | 0.0294 |
62
- | 0.0004 | 3.0 | 1500 | 0.0075 | 0.0796 | 0.0322 |
63
- | 0.0008 | 4.0 | 2000 | 0.0085 | 0.0764 | 0.0281 |
64
- | 0.0007 | 5.0 | 2500 | 0.0089 | 0.0807 | 0.0311 |
65
- | 0.0006 | 6.0 | 3000 | 0.0083 | 0.0860 | 0.0310 |
66
- | 0.0006 | 7.0 | 3500 | 0.0081 | 0.0784 | 0.0289 |
67
- | 0.0006 | 8.0 | 4000 | 0.0088 | 0.0773 | 0.0270 |
68
- | 0.0003 | 9.0 | 4500 | 0.0079 | 0.0820 | 0.0344 |
69
- | 0.0003 | 10.0 | 5000 | 0.0088 | 0.0813 | 0.0285 |
70
- | 0.0003 | 11.0 | 5500 | 0.0083 | 0.0786 | 0.0282 |
71
- | 0.0001 | 12.0 | 6000 | 0.0083 | 0.0760 | 0.0295 |
72
- | 0.0001 | 13.0 | 6500 | 0.0083 | 0.0691 | 0.0257 |
73
- | 0.0001 | 14.0 | 7000 | 0.0089 | 0.0767 | 0.0298 |
74
- | 0.0002 | 15.0 | 7500 | 0.0085 | 0.0733 | 0.0277 |
75
- | 0.0 | 16.0 | 8000 | 0.0085 | 0.0726 | 0.0268 |
76
- | 0.0 | 17.0 | 8500 | 0.0088 | 0.0681 | 0.0252 |
77
- | 0.0001 | 18.0 | 9000 | 0.0093 | 0.0751 | 0.0309 |
78
- | 0.0003 | 19.0 | 9500 | 0.0087 | 0.0771 | 0.0287 |
79
- | 0.0 | 20.0 | 10000 | 0.0081 | 0.0728 | 0.0255 |
80
- | 0.0 | 21.0 | 10500 | 0.0084 | 0.0737 | 0.0272 |
81
- | 0.0 | 22.0 | 11000 | 0.0086 | 0.0715 | 0.0268 |
82
- | 0.0 | 23.0 | 11500 | 0.0087 | 0.0717 | 0.0268 |
83
- | 0.0 | 24.0 | 12000 | 0.0088 | 0.0676 | 0.0268 |
84
- | 0.0 | 25.0 | 12500 | 0.0088 | 0.0711 | 0.0267 |
85
 
86
 
87
  ### Framework versions
 
21
  This model is a fine-tuned version of [openai/whisper-base](https://huggingface.co/openai/whisper-base) on the quran-ayat-speech-to-text dataset.
22
  It achieves the following results on the evaluation set:
23
  - Loss: 0.0088
24
+ - Wer: 0.0638
25
+ - Cer: 0.0245
26
 
27
  ## Model description
28
 
 
55
 
56
  ### Training results
57
 
58
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
+ |:-------------:|:-------:|:-----:|:---------------:|:------:|:------:|
60
+ | 0.0045 | 0.9991 | 562 | 0.0055 | 0.0662 | 0.0270 |
61
+ | 0.0013 | 2.0 | 1125 | 0.0060 | 0.0691 | 0.0249 |
62
+ | 0.0012 | 2.9991 | 1687 | 0.0076 | 0.0820 | 0.0308 |
63
+ | 0.0011 | 4.0 | 2250 | 0.0074 | 0.0729 | 0.0283 |
64
+ | 0.0005 | 4.9991 | 2812 | 0.0074 | 0.0702 | 0.0259 |
65
+ | 0.0007 | 6.0 | 3375 | 0.0081 | 0.0735 | 0.0268 |
66
+ | 0.0007 | 6.9991 | 3937 | 0.0083 | 0.0742 | 0.0274 |
67
+ | 0.0006 | 8.0 | 4500 | 0.0086 | 0.0771 | 0.0292 |
68
+ | 0.0003 | 8.9991 | 5062 | 0.0084 | 0.0740 | 0.0284 |
69
+ | 0.0003 | 10.0 | 5625 | 0.0083 | 0.0677 | 0.0241 |
70
+ | 0.0006 | 10.9991 | 6187 | 0.0083 | 0.0769 | 0.0290 |
71
+ | 0.0002 | 12.0 | 6750 | 0.0091 | 0.0695 | 0.0276 |
72
+ | 0.0002 | 12.9991 | 7312 | 0.0086 | 0.0710 | 0.0265 |
73
+ | 0.0004 | 14.0 | 7875 | 0.0087 | 0.0700 | 0.0267 |
74
+ | 0.0 | 14.9991 | 8437 | 0.0090 | 0.0682 | 0.0244 |
75
+ | 0.0 | 16.0 | 9000 | 0.0089 | 0.0681 | 0.0260 |
76
+ | 0.0 | 16.9991 | 9562 | 0.0094 | 0.0695 | 0.0244 |
77
+ | 0.0001 | 18.0 | 10125 | 0.0097 | 0.0688 | 0.0256 |
78
+ | 0.0001 | 18.9991 | 10687 | 0.0096 | 0.0690 | 0.0251 |
79
+ | 0.0 | 20.0 | 11250 | 0.0100 | 0.0662 | 0.0234 |
80
+ | 0.0 | 20.9991 | 11812 | 0.0100 | 0.0670 | 0.0245 |
81
+ | 0.0 | 22.0 | 12375 | 0.0101 | 0.0671 | 0.0239 |
82
+ | 0.0 | 22.9991 | 12937 | 0.0102 | 0.0664 | 0.0234 |
83
+ | 0.0 | 24.0 | 13500 | 0.0088 | 0.0638 | 0.0245 |
84
+ | 0.0 | 24.9778 | 14050 | 0.0103 | 0.0657 | 0.0231 |
85
 
86
 
87
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e80594d4da32237c3e37e71186c2f6d420da572b581d96d170ac095e4c3a6426
3
  size 290403936
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:71bf56c6da3a71c14f1c2856eaa2ea57c133b87c92696d4e6f462ba8966299cd
3
  size 290403936
runs/Feb24_06-59-30_12d9d8bca2af/events.out.tfevents.1740419339.12d9d8bca2af.23.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c81171ba65342b3701abb671f6c5d6056d2fcbd7fc2491e835db61a66bb51de5
3
+ size 453