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
e201350
·
verified ·
1 Parent(s): 4c4de12

Training finished

Browse files
Files changed (4) hide show
  1. README.md +35 -35
  2. model.safetensors +1 -1
  3. runs/Apr28_08-30-10_b0e4abf6d388/events.out.tfevents.1745829014.b0e4abf6d388.19.0 +2 -2
  4. runs/Apr28_08-30-10_b0e4abf6d388/events.out.tfevents.1745870856.b0e4abf6d388.19.1 +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-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: 0.0029
24
- - Wer: 0.0484
25
- - Cer: 0.0204
26
 
27
  ## Model description
28
 
@@ -55,38 +55,38 @@ The following hyperparameters were used during training:
55
 
56
  ### Training results
57
 
58
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
- |:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
60
- | 0.0032 | 1.0 | 313 | 0.0029 | 0.0463 | 0.0195 |
61
- | 0.0009 | 2.0 | 626 | 0.0041 | 0.0547 | 0.0233 |
62
- | 0.0013 | 3.0 | 939 | 0.0048 | 0.0624 | 0.0229 |
63
- | 0.001 | 4.0 | 1252 | 0.0059 | 0.0614 | 0.0230 |
64
- | 0.0007 | 5.0 | 1565 | 0.0057 | 0.0677 | 0.0270 |
65
- | 0.0006 | 6.0 | 1878 | 0.0066 | 0.0708 | 0.0265 |
66
- | 0.0003 | 7.0 | 2191 | 0.0061 | 0.0688 | 0.0261 |
67
- | 0.0003 | 8.0 | 2504 | 0.0067 | 0.0717 | 0.0285 |
68
- | 0.0002 | 9.0 | 2817 | 0.0062 | 0.0684 | 0.0249 |
69
- | 0.0005 | 10.0 | 3130 | 0.0066 | 0.0686 | 0.0265 |
70
- | 0.0001 | 11.0 | 3443 | 0.0069 | 0.0720 | 0.0283 |
71
- | 0.0006 | 12.0 | 3756 | 0.0062 | 0.0758 | 0.0269 |
72
- | 0.0 | 13.0 | 4069 | 0.0064 | 0.0675 | 0.0276 |
73
- | 0.0 | 14.0 | 4382 | 0.0071 | 0.0664 | 0.0258 |
74
- | 0.0 | 15.0 | 4695 | 0.0068 | 0.0671 | 0.0260 |
75
- | 0.0 | 16.0 | 5008 | 0.0072 | 0.0657 | 0.0252 |
76
- | 0.0 | 17.0 | 5321 | 0.0074 | 0.0650 | 0.0262 |
77
- | 0.0 | 18.0 | 5634 | 0.0075 | 0.0657 | 0.0270 |
78
- | 0.0 | 19.0 | 5947 | 0.0076 | 0.0659 | 0.0271 |
79
- | 0.0 | 20.0 | 6260 | 0.0076 | 0.0644 | 0.0258 |
80
- | 0.0 | 21.0 | 6573 | 0.0077 | 0.0643 | 0.0259 |
81
- | 0.0 | 22.0 | 6886 | 0.0077 | 0.0635 | 0.0255 |
82
- | 0.0 | 23.0 | 7199 | 0.0078 | 0.0635 | 0.0255 |
83
- | 0.0 | 24.0 | 7512 | 0.0078 | 0.0633 | 0.0255 |
84
- | 0.0 | 25.0 | 7825 | 0.0079 | 0.0635 | 0.0255 |
85
- | 0.0 | 26.0 | 8138 | 0.0079 | 0.0637 | 0.0256 |
86
- | 0.0 | 27.0 | 8451 | 0.0079 | 0.0635 | 0.0255 |
87
- | 0.0 | 28.0 | 8764 | 0.0079 | 0.0632 | 0.0256 |
88
- | 0.0 | 29.0 | 9077 | 0.0075 | 0.0617 | 0.0251 |
89
- | 0.0 | 29.9056 | 9360 | 0.0080 | 0.0633 | 0.0256 |
90
 
91
 
92
  ### Framework versions
 
20
 
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: 0.0034
24
+ - Wer: 0.0540
25
+ - Cer: 0.0220
26
 
27
  ## Model description
28
 
 
55
 
56
  ### Training results
57
 
58
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
60
+ | 0.0037 | 1.0 | 500 | 0.0035 | 0.0478 | 0.0178 |
61
+ | 0.0011 | 2.0 | 1000 | 0.0050 | 0.0682 | 0.0270 |
62
+ | 0.0007 | 3.0 | 1500 | 0.0050 | 0.0655 | 0.0250 |
63
+ | 0.0007 | 4.0 | 2000 | 0.0054 | 0.0646 | 0.0231 |
64
+ | 0.0004 | 5.0 | 2500 | 0.0055 | 0.0588 | 0.0222 |
65
+ | 0.0006 | 6.0 | 3000 | 0.0055 | 0.0644 | 0.0255 |
66
+ | 0.0004 | 7.0 | 3500 | 0.0059 | 0.0695 | 0.0282 |
67
+ | 0.0008 | 8.0 | 4000 | 0.0059 | 0.0657 | 0.0252 |
68
+ | 0.0007 | 9.0 | 4500 | 0.0056 | 0.0653 | 0.0249 |
69
+ | 0.0002 | 10.0 | 5000 | 0.0053 | 0.0644 | 0.0232 |
70
+ | 0.0002 | 11.0 | 5500 | 0.0063 | 0.0619 | 0.0238 |
71
+ | 0.0002 | 12.0 | 6000 | 0.0057 | 0.0610 | 0.0243 |
72
+ | 0.0002 | 13.0 | 6500 | 0.0059 | 0.0666 | 0.0264 |
73
+ | 0.0003 | 14.0 | 7000 | 0.0061 | 0.0606 | 0.0225 |
74
+ | 0.0001 | 15.0 | 7500 | 0.0063 | 0.0583 | 0.0209 |
75
+ | 0.0001 | 16.0 | 8000 | 0.0060 | 0.0605 | 0.0206 |
76
+ | 0.0 | 17.0 | 8500 | 0.0060 | 0.0626 | 0.0227 |
77
+ | 0.0 | 18.0 | 9000 | 0.0063 | 0.0637 | 0.0235 |
78
+ | 0.0 | 19.0 | 9500 | 0.0066 | 0.0563 | 0.0227 |
79
+ | 0.0 | 20.0 | 10000 | 0.0068 | 0.0567 | 0.0211 |
80
+ | 0.0001 | 21.0 | 10500 | 0.0070 | 0.0646 | 0.0258 |
81
+ | 0.0001 | 22.0 | 11000 | 0.0069 | 0.0633 | 0.0262 |
82
+ | 0.0001 | 23.0 | 11500 | 0.0064 | 0.0550 | 0.0226 |
83
+ | 0.0 | 24.0 | 12000 | 0.0065 | 0.0534 | 0.0207 |
84
+ | 0.0001 | 25.0 | 12500 | 0.0066 | 0.0536 | 0.0203 |
85
+ | 0.0 | 26.0 | 13000 | 0.0067 | 0.0539 | 0.0210 |
86
+ | 0.0 | 27.0 | 13500 | 0.0068 | 0.0527 | 0.0197 |
87
+ | 0.0 | 28.0 | 14000 | 0.0069 | 0.0523 | 0.0200 |
88
+ | 0.0 | 29.0 | 14500 | 0.0069 | 0.0582 | 0.0228 |
89
+ | 0.0 | 30.0 | 15000 | 0.0070 | 0.0521 | 0.0199 |
90
 
91
 
92
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:fb34a1a91f6a42c99ee6b5bf1bc3d6b146ac2466e0ac2307877522da51c20247
3
  size 290403936
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:70fbdc760b0e50fda922f2576182b8f922e0d0fc831a5c0b6ad0a6eefaac50d4
3
  size 290403936
runs/Apr28_08-30-10_b0e4abf6d388/events.out.tfevents.1745829014.b0e4abf6d388.19.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0e19a978e946b27ce4ba0ef3407aa078fc05b82f6c04cd6b180641392187836b
3
- size 138839
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1b28be98a7b63701852211ff194325e8a159eb5cd658afff7158a55d97d4e26
3
+ size 143778
runs/Apr28_08-30-10_b0e4abf6d388/events.out.tfevents.1745870856.b0e4abf6d388.19.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4f3f14d4cb34b733f062b80d86a3f7c39b5700bae30b66d82fbefc041c453ac6
3
+ size 453