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
7ceafc1
·
verified ·
1 Parent(s): fbfe4be

Training finished

Browse files
Files changed (4) hide show
  1. README.md +24 -24
  2. generation_config.json +1 -1
  3. model.safetensors +1 -1
  4. runs/Jan25_08-21-52_5eba7b6ea6b5/events.out.tfevents.1737815076.5eba7b6ea6b5.18.1 +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: 0.0155
24
- - Wer: 0.1082
25
- - Cer: 0.0393
26
 
27
  ## Model description
28
 
@@ -47,7 +47,7 @@ The following hyperparameters were used during training:
47
  - seed: 42
48
  - gradient_accumulation_steps: 2
49
  - total_train_batch_size: 32
50
- - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
  - lr_scheduler_type: linear
52
  - lr_scheduler_warmup_steps: 500
53
  - num_epochs: 15
@@ -55,28 +55,28 @@ The following hyperparameters were used during training:
55
 
56
  ### Training results
57
 
58
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
- |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
- | 0.0148 | 1.0 | 219 | 0.0134 | 0.1157 | 0.0395 |
61
- | 0.0101 | 2.0 | 438 | 0.0105 | 0.1175 | 0.0406 |
62
- | 0.0076 | 3.0 | 657 | 0.0099 | 0.1086 | 0.0379 |
63
- | 0.0071 | 4.0 | 876 | 0.0100 | 0.1066 | 0.0370 |
64
- | 0.0046 | 5.0 | 1095 | 0.0104 | 0.1081 | 0.0366 |
65
- | 0.0026 | 6.0 | 1314 | 0.0108 | 0.1137 | 0.0421 |
66
- | 0.003 | 7.0 | 1533 | 0.0116 | 0.1086 | 0.0391 |
67
- | 0.0026 | 8.0 | 1752 | 0.0123 | 0.1090 | 0.0392 |
68
- | 0.001 | 9.0 | 1971 | 0.0128 | 0.1081 | 0.0377 |
69
- | 0.0012 | 10.0 | 2190 | 0.0133 | 0.1097 | 0.0391 |
70
- | 0.001 | 11.0 | 2409 | 0.0137 | 0.1091 | 0.0373 |
71
- | 0.0004 | 12.0 | 2628 | 0.0141 | 0.1081 | 0.0379 |
72
- | 0.0009 | 13.0 | 2847 | 0.0143 | 0.1064 | 0.0363 |
73
- | 0.0005 | 14.0 | 3066 | 0.0157 | 0.1086 | 0.0397 |
74
- | 0.0009 | 15.0 | 3285 | 0.0145 | 0.1071 | 0.0370 |
75
 
76
 
77
  ### Framework versions
78
 
79
- - Transformers 4.44.2
80
- - Pytorch 2.4.1+cu121
81
  - Datasets 3.2.0
82
- - Tokenizers 0.19.1
 
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: 0.0095
24
+ - Wer: 0.1037
25
+ - Cer: 0.0382
26
 
27
  ## Model description
28
 
 
47
  - seed: 42
48
  - gradient_accumulation_steps: 2
49
  - total_train_batch_size: 32
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: 15
 
55
 
56
  ### Training results
57
 
58
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
+ |:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
60
+ | 0.0104 | 1.0 | 407 | 0.0098 | 0.1182 | 0.0449 |
61
+ | 0.0068 | 2.0 | 814 | 0.0088 | 0.1055 | 0.0373 |
62
+ | 0.0075 | 3.0 | 1221 | 0.0088 | 0.1008 | 0.0356 |
63
+ | 0.0057 | 4.0 | 1628 | 0.0091 | 0.0992 | 0.0345 |
64
+ | 0.0047 | 5.0 | 2035 | 0.0097 | 0.0997 | 0.0349 |
65
+ | 0.0038 | 6.0 | 2442 | 0.0103 | 0.0994 | 0.0340 |
66
+ | 0.0024 | 7.0 | 2849 | 0.0109 | 0.1033 | 0.0357 |
67
+ | 0.0031 | 8.0 | 3256 | 0.0113 | 0.1015 | 0.0351 |
68
+ | 0.0014 | 9.0 | 3663 | 0.0118 | 0.1003 | 0.0350 |
69
+ | 0.0018 | 10.0 | 4070 | 0.0123 | 0.1014 | 0.0349 |
70
+ | 0.0013 | 11.0 | 4477 | 0.0128 | 0.1122 | 0.0405 |
71
+ | 0.0011 | 12.0 | 4884 | 0.0130 | 0.1037 | 0.0379 |
72
+ | 0.0004 | 13.0 | 5291 | 0.0132 | 0.1032 | 0.0379 |
73
+ | 0.0019 | 14.0 | 5698 | 0.0141 | 0.1055 | 0.0397 |
74
+ | 0.001 | 14.9643 | 6090 | 0.0135 | 0.1017 | 0.0371 |
75
 
76
 
77
  ### Framework versions
78
 
79
+ - Transformers 4.47.0
80
+ - Pytorch 2.5.1+cu121
81
  - Datasets 3.2.0
82
+ - Tokenizers 0.21.0
generation_config.json CHANGED
@@ -244,5 +244,5 @@
244
  "transcribe": 50359,
245
  "translate": 50358
246
  },
247
- "transformers_version": "4.44.2"
248
  }
 
244
  "transcribe": 50359,
245
  "translate": 50358
246
  },
247
+ "transformers_version": "4.47.0"
248
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f514563560b166bd83ccab9853ff7ffcf8b63af35161b6589669298517e02874
3
  size 151061672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d27b4cd16fa73c8157bdf850c84d341f3f3982e7927ebb8eabd3695d1493da9
3
  size 151061672
runs/Jan25_08-21-52_5eba7b6ea6b5/events.out.tfevents.1737815076.5eba7b6ea6b5.18.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f4c4d791b4ca74201e5f810909f4363cc291c5f6a59623796b1d060deee4a7cf
3
+ size 453