Hugging Face's logo

Baselhany
/
fine_tune_Whisper_tiny2

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Arabic
whisper
Generated from Trainer
Model card Files Metrics Community
Baselhany commited on
Commit
1639a6d
·
verified ·
1 Parent(s): b2f0730

Training finished

Browse files
Files changed (3) hide show
  1. README.md +21 -12
  2. model.safetensors +1 -1
  3. runs/Jan19_19-52-00_379765d20ece/events.out.tfevents.1737327065.379765d20ece.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: 0.0074
24
- - Wer: 0.0857
25
- - Cer: 0.0352
26
 
27
  ## Model description
28
 
@@ -50,19 +50,28 @@ The following hyperparameters were used during training:
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: 20
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
- | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
- |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
- | 0.0067 | 1.0 | 79 | 0.0067 | 0.0769 | 0.0297 |
61
- | 0.0067 | 2.0 | 158 | 0.0065 | 0.0807 | 0.0306 |
62
- | 0.0067 | 3.0 | 237 | 0.0064 | 0.0778 | 0.0293 |
63
- | 0.0047 | 4.0 | 316 | 0.0064 | 0.0786 | 0.0300 |
64
- | 0.0059 | 5.0 | 395 | 0.0066 | 0.0780 | 0.0292 |
65
- | 0.0038 | 6.0 | 474 | 0.0068 | 0.0796 | 0.0311 |
 
 
 
 
 
 
 
 
 
66
 
67
 
68
  ### 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: 0.0067
24
+ - Wer: 0.0879
25
+ - Cer: 0.0338
26
 
27
  ## Model description
28
 
 
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
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
+ | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
+ |:-------------:|:-------:|:----:|:---------------:|:------:|:------:|
60
+ | 0.006 | 1.0 | 79 | 0.0066 | 0.0798 | 0.0302 |
61
+ | 0.0081 | 2.0 | 158 | 0.0063 | 0.0793 | 0.0289 |
62
+ | 0.0062 | 3.0 | 237 | 0.0062 | 0.0791 | 0.0294 |
63
+ | 0.0047 | 4.0 | 316 | 0.0062 | 0.0824 | 0.0317 |
64
+ | 0.0048 | 5.0 | 395 | 0.0063 | 0.0831 | 0.0312 |
65
+ | 0.0036 | 6.0 | 474 | 0.0065 | 0.0824 | 0.0316 |
66
+ | 0.0032 | 7.0 | 553 | 0.0067 | 0.0833 | 0.0324 |
67
+ | 0.0022 | 8.0 | 632 | 0.0071 | 0.0825 | 0.0330 |
68
+ | 0.0018 | 9.0 | 711 | 0.0076 | 0.0818 | 0.0318 |
69
+ | 0.0007 | 10.0 | 790 | 0.0079 | 0.0827 | 0.0310 |
70
+ | 0.001 | 11.0 | 869 | 0.0082 | 0.0860 | 0.0337 |
71
+ | 0.0008 | 12.0 | 948 | 0.0084 | 0.0878 | 0.0330 |
72
+ | 0.0006 | 13.0 | 1027 | 0.0085 | 0.0867 | 0.0336 |
73
+ | 0.0003 | 14.0 | 1106 | 0.0093 | 0.0941 | 0.0400 |
74
+ | 0.0005 | 14.8179 | 1170 | 0.0087 | 0.0894 | 0.0345 |
75
 
76
 
77
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:659fc015d6fbfba22b54b4d2daf7bdf5fce0b6b5ff249fb4a80d9673723bd379
3
  size 151061672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6622e6b018b7feaa466fdefdddee4ba47b3e1d66855ad971097976f113c41fee
3
  size 151061672
runs/Jan19_19-52-00_379765d20ece/events.out.tfevents.1737327065.379765d20ece.18.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c29813f9ca016bebcdda23345dd9c9761ca4dfec7d339c65f53b78a1ec343869
3
+ size 453