Hugging Face's logo

Baselhany
/
Graduation_Project_Whisper_base_with_segments

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Arabic
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Baselhany commited on
Commit
7729ec0
•
1 Parent(s): 9af2a9f

Training finished

Browse files
Files changed (3) hide show
  1. README.md +12 -20
  2. model.safetensors +1 -1
  3. runs/Dec12_19-59-27_5b6e68c20e69/events.out.tfevents.1734037440.5b6e68c20e69.23.1 +3 -0
README.md CHANGED
@@ -20,8 +20,8 @@ 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-segments dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.0259
24
- - Wer: 0.2831
25
 
26
  ## Model description
27
 
@@ -41,11 +41,11 @@ More information needed
41
 
42
  The following hyperparameters were used during training:
43
  - learning_rate: 0.0001
44
- - train_batch_size: 8
45
- - eval_batch_size: 4
46
  - seed: 42
47
  - gradient_accumulation_steps: 2
48
- - total_train_batch_size: 16
49
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
50
  - lr_scheduler_type: linear
51
  - lr_scheduler_warmup_steps: 500
@@ -56,21 +56,13 @@ The following hyperparameters were used during training:
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Wer |
58
  |:-------------:|:-------:|:----:|:---------------:|:------:|
59
- | 0.0529 | 1.3333 | 100 | 0.0475 | 0.6531 |
60
- | 0.0214 | 2.6667 | 200 | 0.0303 | 0.4790 |
61
- | 0.015 | 4.0 | 300 | 0.0302 | 0.4221 |
62
- | 0.0098 | 5.3333 | 400 | 0.0304 | 0.4160 |
63
- | 0.0105 | 6.6667 | 500 | 0.0313 | 0.4063 |
64
- | 0.0074 | 8.0 | 600 | 0.0286 | 0.3631 |
65
- | 0.0028 | 9.3333 | 700 | 0.0304 | 0.3639 |
66
- | 0.0013 | 10.6667 | 800 | 0.0264 | 0.3263 |
67
- | 0.0011 | 12.0 | 900 | 0.0257 | 0.3053 |
68
- | 0.0004 | 13.3333 | 1000 | 0.0263 | 0.2993 |
69
- | 0.0002 | 14.6667 | 1100 | 0.0256 | 0.2896 |
70
- | 0.0 | 16.0 | 1200 | 0.0256 | 0.2855 |
71
- | 0.0 | 17.3333 | 1300 | 0.0258 | 0.2843 |
72
- | 0.0 | 18.6667 | 1400 | 0.0259 | 0.2831 |
73
- | 0.0 | 20.0 | 1500 | 0.0259 | 0.2831 |
74
 
75
 
76
  ### 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-segments dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.0171
24
+ - Wer: 0.2407
25
 
26
  ## Model description
27
 
 
41
 
42
  The following hyperparameters were used during training:
43
  - learning_rate: 0.0001
44
+ - train_batch_size: 16
45
+ - eval_batch_size: 8
46
  - seed: 42
47
  - gradient_accumulation_steps: 2
48
+ - total_train_batch_size: 32
49
  - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
50
  - lr_scheduler_type: linear
51
  - lr_scheduler_warmup_steps: 500
 
56
 
57
  | Training Loss | Epoch | Step | Validation Loss | Wer |
58
  |:-------------:|:-------:|:----:|:---------------:|:------:|
59
+ | 0.0105 | 2.6667 | 100 | 0.0154 | 0.2514 |
60
+ | 0.0021 | 5.3333 | 200 | 0.0171 | 0.2407 |
61
+ | 0.0022 | 8.0 | 300 | 0.0200 | 0.2475 |
62
+ | 0.0026 | 10.6667 | 400 | 0.0215 | 0.2847 |
63
+ | 0.0027 | 13.3333 | 500 | 0.0237 | 0.2929 |
64
+ | 0.0014 | 16.0 | 600 | 0.0216 | 0.2548 |
65
+ | 0.0001 | 18.6667 | 700 | 0.0210 | 0.2386 |
 
 
 
 
 
 
 
 
66
 
67
 
68
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d7cc76653241dec11c8eac9f0750e4133b674aa3394b16239ec43cd69ffb6fe1
3
  size 290403936
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7ca08f5dd44d9c6222470a0c54333086082e0176081f4cf3f2688d685e82b5b4
3
  size 290403936
runs/Dec12_19-59-27_5b6e68c20e69/events.out.tfevents.1734037440.5b6e68c20e69.23.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:105c95d932c1e5c60d38fe5ea265144821fba904b00082be84fc84f875298eb6
3
+ size 406