Hugging Face's logo

Baselhany
/
Whisper_tiny_tring_large_sample_with_early_stop2

Automatic Speech Recognition
Transformers
TensorBoard
Safetensors
Arabic
whisper
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
Baselhany commited on
Commit
32fb16a
·
verified ·
1 Parent(s): e8edbbe

Training finished

Browse files
Files changed (3) hide show
  1. README.md +9 -13
  2. model.safetensors +1 -1
  3. runs/Jan27_08-36-51_da7ce78b1ba8/events.out.tfevents.1737982261.da7ce78b1ba8.18.1 +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: 0.0128
24
- - Wer: 0.2558
25
- - Cer: 0.1079
26
 
27
  ## Model description
28
 
@@ -50,22 +50,18 @@ 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: 15
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
- | 0.0122 | 1.0 | 469 | 0.0144 | 0.6112 | 0.2654 |
61
- | 0.0095 | 2.0 | 938 | 0.0138 | 0.3718 | 0.1563 |
62
- | 0.0105 | 3.0 | 1407 | 0.0136 | 0.2865 | 0.1192 |
63
- | 0.0069 | 4.0 | 1876 | 0.0135 | 0.3586 | 0.1424 |
64
- | 0.0062 | 5.0 | 2345 | 0.0132 | 0.3569 | 0.1506 |
65
- | 0.0075 | 6.0 | 2814 | 0.0132 | 0.2715 | 0.1097 |
66
- | 0.0058 | 7.0 | 3283 | 0.0130 | 0.3638 | 0.1479 |
67
- | 0.0061 | 8.0 | 3752 | 0.0131 | 0.3557 | 0.1408 |
68
- | 0.0049 | 9.0 | 4221 | 0.0132 | 0.3437 | 0.1350 |
69
 
70
 
71
  ### 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: 0.0111
24
+ - Wer: 0.1829
25
+ - Cer: 0.0676
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: 10
54
  - mixed_precision_training: Native AMP
55
 
56
  ### Training results
57
 
58
  | Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
59
  |:-------------:|:-----:|:----:|:---------------:|:------:|:------:|
60
+ | 0.0037 | 1.0 | 469 | 0.0114 | 0.1557 | 0.0492 |
61
+ | 0.0045 | 2.0 | 938 | 0.0117 | 0.1553 | 0.0499 |
62
+ | 0.0033 | 3.0 | 1407 | 0.0116 | 0.1580 | 0.0513 |
63
+ | 0.0024 | 4.0 | 1876 | 0.0117 | 0.1558 | 0.0499 |
64
+ | 0.0029 | 5.0 | 2345 | 0.0119 | 0.1569 | 0.0525 |
 
 
 
 
65
 
66
 
67
  ### Framework versions
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0daa4daa700824e082056a0b786d6468d6198c0539696f4120a8830fdcec87ed
3
  size 151061672
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2bff3fe12dc28a47f60ef3746b4050620b24070e4ea1f3161adc5206ef349d61
3
  size 151061672
runs/Jan27_08-36-51_da7ce78b1ba8/events.out.tfevents.1737982261.da7ce78b1ba8.18.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f18383bdfe596c2610c266f0539c7b01c345a9c1f43d2935c7236da73fa3e926
3
+ size 453