Hugging Face's logo

YoussefAshmawy
/
Graduation_Project_Whisper_base

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Arabic
whisper
Generated from Trainer
arabic
quran
Eval Results (legacy)
Model card Files
xet
 Metrics Community
YoussefAshmawy commited on
Commit
a13c936
·
verified ·
1 Parent(s): 08157db

Training finished

Browse files
Files changed (4) hide show
  1. README.md +34 -34
  2. model.safetensors +1 -1
  3. runs/Jun08_14-53-50_4d65d228fdc0/events.out.tfevents.1749394433.4d65d228fdc0.19.0 +2 -2
  4. runs/Jun08_14-53-50_4d65d228fdc0/events.out.tfevents.1749435213.4d65d228fdc0.19.1 +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: 0.0023
24
- - Wer: 0.0430
25
- - Cer: 0.0191
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
- | 0.0056 | 1.0 | 513 | 0.0024 | 0.0434 | 0.0207 |
61
- | 0.0034 | 2.0 | 1026 | 0.0031 | 0.0442 | 0.0200 |
62
- | 0.0016 | 3.0 | 1539 | 0.0036 | 0.0496 | 0.0226 |
63
- | 0.0011 | 4.0 | 2052 | 0.0038 | 0.0486 | 0.0230 |
64
- | 0.0006 | 5.0 | 2565 | 0.0041 | 0.0478 | 0.0220 |
65
- | 0.001 | 6.0 | 3078 | 0.0045 | 0.0542 | 0.0243 |
66
- | 0.0009 | 7.0 | 3591 | 0.0043 | 0.0488 | 0.0202 |
67
- | 0.0006 | 8.0 | 4104 | 0.0046 | 0.0496 | 0.0227 |
68
- | 0.0005 | 9.0 | 4617 | 0.0047 | 0.0507 | 0.0232 |
69
- | 0.0005 | 10.0 | 5130 | 0.0052 | 0.0482 | 0.0210 |
70
- | 0.0005 | 11.0 | 5643 | 0.0050 | 0.0507 | 0.0236 |
71
- | 0.0003 | 12.0 | 6156 | 0.0047 | 0.0511 | 0.0237 |
72
- | 0.0005 | 13.0 | 6669 | 0.0048 | 0.0478 | 0.0209 |
73
- | 0.0002 | 14.0 | 7182 | 0.0046 | 0.0521 | 0.0243 |
74
- | 0.0001 | 15.0 | 7695 | 0.0053 | 0.0544 | 0.0265 |
75
- | 0.0002 | 16.0 | 8208 | 0.0051 | 0.0503 | 0.0236 |
76
- | 0.0001 | 17.0 | 8721 | 0.0053 | 0.0523 | 0.0242 |
77
- | 0.0001 | 18.0 | 9234 | 0.0048 | 0.0476 | 0.0209 |
78
- | 0.0 | 19.0 | 9747 | 0.0048 | 0.0480 | 0.0222 |
79
- | 0.0001 | 20.0 | 10260 | 0.0051 | 0.0486 | 0.0210 |
80
- | 0.0001 | 21.0 | 10773 | 0.0052 | 0.0488 | 0.0207 |
81
- | 0.0 | 22.0 | 11286 | 0.0052 | 0.0478 | 0.0207 |
82
- | 0.0 | 23.0 | 11799 | 0.0051 | 0.0488 | 0.0222 |
83
- | 0.0 | 24.0 | 12312 | 0.0051 | 0.0478 | 0.0215 |
84
- | 0.0 | 25.0 | 12825 | 0.0052 | 0.0469 | 0.0210 |
85
- | 0.0 | 26.0 | 13338 | 0.0052 | 0.0463 | 0.0211 |
86
- | 0.0 | 27.0 | 13851 | 0.0053 | 0.0465 | 0.0213 |
87
- | 0.0 | 28.0 | 14364 | 0.0053 | 0.0465 | 0.0212 |
88
- | 0.0 | 29.0 | 14877 | 0.0057 | 0.0503 | 0.0230 |
89
- | 0.0 | 29.9424 | 15360 | 0.0053 | 0.0465 | 0.0211 |
90
 
91
 
92
  ### 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: 0.0023
24
+ - Wer: 0.0405
25
+ - Cer: 0.0195
26
 
27
  ## Model description
28
 
 
55
 
56
  ### Training results
57
 
58
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
60
+ | 0.0058 | 1.0 | 525 | 0.0025 | 0.0353 | 0.0177 |
61
+ | 0.0018 | 2.0 | 1050 | 0.0031 | 0.0428 | 0.0197 |
62
+ | 0.0017 | 3.0 | 1575 | 0.0040 | 0.0511 | 0.0246 |
63
+ | 0.001 | 4.0 | 2100 | 0.0039 | 0.0469 | 0.0212 |
64
+ | 0.0013 | 5.0 | 2625 | 0.0043 | 0.0505 | 0.0240 |
65
+ | 0.0006 | 6.0 | 3150 | 0.0042 | 0.0478 | 0.0223 |
66
+ | 0.0007 | 7.0 | 3675 | 0.0049 | 0.0534 | 0.0227 |
67
+ | 0.0007 | 8.0 | 4200 | 0.0048 | 0.0552 | 0.0235 |
68
+ | 0.0005 | 9.0 | 4725 | 0.0048 | 0.0501 | 0.0218 |
69
+ | 0.0005 | 10.0 | 5250 | 0.0048 | 0.0513 | 0.0215 |
70
+ | 0.0006 | 11.0 | 5775 | 0.0055 | 0.0528 | 0.0217 |
71
+ | 0.0002 | 12.0 | 6300 | 0.0055 | 0.0542 | 0.0232 |
72
+ | 0.0003 | 13.0 | 6825 | 0.0056 | 0.0530 | 0.0238 |
73
+ | 0.0002 | 14.0 | 7350 | 0.0057 | 0.0498 | 0.0237 |
74
+ | 0.0001 | 15.0 | 7875 | 0.0057 | 0.0446 | 0.0189 |
75
+ | 0.0003 | 16.0 | 8400 | 0.0054 | 0.0567 | 0.0254 |
76
+ | 0.0002 | 17.0 | 8925 | 0.0057 | 0.0540 | 0.0256 |
77
+ | 0.0002 | 18.0 | 9450 | 0.0057 | 0.0530 | 0.0239 |
78
+ | 0.0 | 19.0 | 9975 | 0.0056 | 0.0478 | 0.0228 |
79
+ | 0.0 | 20.0 | 10500 | 0.0055 | 0.0473 | 0.0223 |
80
+ | 0.0 | 21.0 | 11025 | 0.0056 | 0.0449 | 0.0202 |
81
+ | 0.0 | 22.0 | 11550 | 0.0056 | 0.0461 | 0.0213 |
82
+ | 0.0 | 23.0 | 12075 | 0.0057 | 0.0461 | 0.0213 |
83
+ | 0.0 | 24.0 | 12600 | 0.0058 | 0.0465 | 0.0218 |
84
+ | 0.0 | 25.0 | 13125 | 0.0058 | 0.0474 | 0.0224 |
85
+ | 0.0 | 26.0 | 13650 | 0.0059 | 0.0465 | 0.0218 |
86
+ | 0.0 | 27.0 | 14175 | 0.0059 | 0.0469 | 0.0219 |
87
+ | 0.0 | 28.0 | 14700 | 0.0059 | 0.0461 | 0.0218 |
88
+ | 0.0 | 29.0 | 15225 | 0.0054 | 0.0513 | 0.0229 |
89
+ | 0.0 | 30.0 | 15750 | 0.0060 | 0.0463 | 0.0217 |
90
 
91
 
92
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:6d19eff42afac8b0a767710ca5367483c313deff6d48e11f66309e25e6130db3
3
  size 290403936
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:31201b42f6b4c9d3dda41d48160e69c214969f162e97cd7be5efa9f4f93e7fc5
3
  size 290403936
runs/Jun08_14-53-50_4d65d228fdc0/events.out.tfevents.1749394433.4d65d228fdc0.19.0 CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:efff508ff1e3ef9a9f9594a409705319f61a086b73954a7723813525d5393a82
3
- size 144958
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1728a41ca52a242377cce13b1a5e9262e4bc5938f4e5cafd12c16e69dcf44cab
3
+ size 150108
runs/Jun08_14-53-50_4d65d228fdc0/events.out.tfevents.1749435213.4d65d228fdc0.19.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0b599eb6e830807861dbd8f46de5685144492d8f97553c757c7337b0c6091f1
3
+ size 453