Hugging Face's logo

Baselhany
/
Graduation_Project_Whisper_tiny2

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Arabic
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Baselhany commited on
Commit
84a20fd
·
verified ·
1 Parent(s): d0dc48f

Training finished

Browse files
Files changed (3) hide show
  1. README.md +12 -17
  2. model.safetensors +1 -1
  3. runs/Jan02_23-29-37_b6b339af972b/events.out.tfevents.1735866286.b6b339af972b.40.2 +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: 0.0163
24
- - Wer: 0.0952
25
- - Cer: 0.0350
26
 
27
  ## Model description
28
 
@@ -41,7 +41,7 @@ More information needed
41
  ### Training hyperparameters
42
 
43
  The following hyperparameters were used during training:
44
- - learning_rate: 0.0001
45
  - train_batch_size: 16
46
  - eval_batch_size: 16
47
  - seed: 42
@@ -50,23 +50,18 @@ The following hyperparameters were used during training:
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: 10
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
- |:-------------:|:------:|:----:|:---------------:|:------:|:------:|
60
- | 0.0011 | 0.9988 | 406 | 0.0191 | 0.1350 | 0.0462 |
61
- | 0.0026 | 2.0 | 813 | 0.0172 | 0.1482 | 0.0518 |
62
- | 0.0028 | 2.9988 | 1219 | 0.0162 | 0.1390 | 0.0470 |
63
- | 0.0019 | 4.0 | 1626 | 0.0165 | 0.1356 | 0.0467 |
64
- | 0.0008 | 4.9988 | 2032 | 0.0169 | 0.1225 | 0.0399 |
65
- | 0.0007 | 6.0 | 2439 | 0.0164 | 0.1214 | 0.0401 |
66
- | 0.0003 | 6.9988 | 2845 | 0.0168 | 0.1091 | 0.0371 |
67
- | 0.0001 | 8.0 | 3252 | 0.0164 | 0.1024 | 0.0350 |
68
- | 0.0 | 8.9988 | 3658 | 0.0162 | 0.0936 | 0.0324 |
69
- | 0.0 | 9.9877 | 4060 | 0.0164 | 0.0921 | 0.0335 |
70
 
71
 
72
  ### Framework versions
 
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: 0.0145
24
+ - Wer: 0.0963
25
+ - Cer: 0.0363
26
 
27
  ## Model description
28
 
 
41
  ### Training hyperparameters
42
 
43
  The following hyperparameters were used during training:
44
+ - learning_rate: 1e-05
45
  - train_batch_size: 16
46
  - eval_batch_size: 16
47
  - seed: 42
 
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: 5
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
+ |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
+ | 0.0011 | 1.0 | 219 | 0.0134 | 0.1178 | 0.0421 |
61
+ | 0.0004 | 2.0 | 438 | 0.0136 | 0.1097 | 0.0399 |
62
+ | 0.0001 | 3.0 | 657 | 0.0141 | 0.1062 | 0.0390 |
63
+ | 0.0 | 4.0 | 876 | 0.0145 | 0.0963 | 0.0363 |
64
+ | 0.0 | 5.0 | 1095 | 0.0146 | 0.0976 | 0.0346 |
 
 
 
 
 
65
 
66
 
67
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:501fbe78e993fa1fc0ec41b83479cdffea462e4d27693aa5a3c65d43474d918e
3
  size 151061672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d13ed13f84638615e42d730538e226821267dd5e40e5c890f0d1b721eadfad4
3
  size 151061672
runs/Jan02_23-29-37_b6b339af972b/events.out.tfevents.1735866286.b6b339af972b.40.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8c004cf78a6bf8cee63c4dc2e00d0ea85d53b900d0a3d1ed974e875abe693c47
3
+ size 453