Hugging Face's logo

jindaznb
/
torgo_tiny_finetune_F01

Automatic Speech Recognition
Transformers
PyTorch
whisper
Generated from Trainer
Model card Files
xet
 Community
1
jindaznb commited on
Commit
deeff7c
·
verified ·
1 Parent(s): 52e7041

End of training

Browse files
Files changed (3) hide show
  1. README.md +28 -16
  2. pytorch_model.bin +1 -1
  3. training_args.bin +2 -2
README.md CHANGED
@@ -17,8 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
17
 
18
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 3.4295
21
- - Wer: 96.0
22
 
23
  ## Model description
24
 
@@ -48,20 +48,32 @@ The following hyperparameters were used during training:
48
 
49
  ### Training results
50
 
51
- | Training Loss | Epoch | Step | Validation Loss | Wer |
52
- |:-------------:|:-----:|:-----:|:---------------:|:--------:|
53
- | 0.1254 | 1.66 | 1000 | 1.8418 | 92.7273 |
54
- | 0.0684 | 3.32 | 2000 | 2.5681 | 96.7273 |
55
- | 0.0381 | 4.98 | 3000 | 2.7969 | 177.8182 |
56
- | 0.0223 | 6.63 | 4000 | 2.9649 | 174.9091 |
57
- | 0.0139 | 8.29 | 5000 | 3.3330 | 99.2727 |
58
- | 0.0095 | 9.95 | 6000 | 3.4180 | 97.8182 |
59
- | 0.0049 | 11.61 | 7000 | 3.5189 | 98.5455 |
60
- | 0.0017 | 13.27 | 8000 | 3.4448 | 97.0909 |
61
- | 0.001 | 14.93 | 9000 | 3.4596 | 98.5455 |
62
- | 0.0004 | 16.58 | 10000 | 3.4080 | 96.3636 |
63
- | 0.0003 | 18.24 | 11000 | 3.4101 | 96.3636 |
64
- | 0.0 | 19.9 | 12000 | 3.4295 | 96.0 |
 
 
 
 
 
 
 
 
 
 
 
 
65
 
66
 
67
  ### Framework versions
 
17
 
18
  This model is a fine-tuned version of [openai/whisper-tiny](https://huggingface.co/openai/whisper-tiny) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 0.2909
21
+ - Wer: 24.6180
22
 
23
  ## Model description
24
 
 
48
 
49
  ### Training results
50
 
51
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
52
+ |:-------------:|:-----:|:-----:|:---------------:|:-------:|
53
+ | 0.6321 | 0.83 | 500 | 0.3242 | 51.0187 |
54
+ | 0.1046 | 1.66 | 1000 | 0.4511 | 55.3480 |
55
+ | 0.099 | 2.49 | 1500 | 0.3225 | 31.5789 |
56
+ | 0.0656 | 3.32 | 2000 | 0.3007 | 50.6791 |
57
+ | 0.0506 | 4.15 | 2500 | 0.2984 | 27.6740 |
58
+ | 0.0383 | 4.98 | 3000 | 0.2853 | 23.6842 |
59
+ | 0.0296 | 5.8 | 3500 | 0.3449 | 32.3430 |
60
+ | 0.0198 | 6.63 | 4000 | 0.2730 | 26.6553 |
61
+ | 0.0192 | 7.46 | 4500 | 0.3049 | 49.2360 |
62
+ | 0.0136 | 8.29 | 5000 | 0.3279 | 25.8065 |
63
+ | 0.0121 | 9.12 | 5500 | 0.3082 | 23.8540 |
64
+ | 0.0101 | 9.95 | 6000 | 0.2722 | 25.5518 |
65
+ | 0.0065 | 10.78 | 6500 | 0.3414 | 32.0883 |
66
+ | 0.0062 | 11.61 | 7000 | 0.3140 | 22.9202 |
67
+ | 0.0053 | 12.44 | 7500 | 0.2601 | 24.7029 |
68
+ | 0.002 | 13.27 | 8000 | 0.2978 | 33.8710 |
69
+ | 0.0021 | 14.1 | 8500 | 0.2798 | 31.1545 |
70
+ | 0.0011 | 14.93 | 9000 | 0.3137 | 25.1273 |
71
+ | 0.0006 | 15.75 | 9500 | 0.2926 | 22.2411 |
72
+ | 0.0003 | 16.58 | 10000 | 0.2891 | 23.4295 |
73
+ | 0.0001 | 17.41 | 10500 | 0.2930 | 25.2122 |
74
+ | 0.0001 | 18.24 | 11000 | 0.2906 | 24.7878 |
75
+ | 0.0001 | 19.07 | 11500 | 0.2906 | 24.6180 |
76
+ | 0.0 | 19.9 | 12000 | 0.2909 | 24.6180 |
77
 
78
 
79
  ### Framework versions
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c7e873246e74ca2b9c86200b86108e2ce161d71bb74966ed985ae080ee098b16
3
  size 151099494
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a3ad96089ff5b316483d65cdb2c6f5f4c143c5ce8c9ac9bbc876ea5afb86fa37
3
  size 151099494
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:bbf713631beeae7404f504156ff2e27a8a88341b8c24e19a377e62e56b018cdd
3
- size 4664
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:794bf0e45994d167e3f0016c78b60f32cb05786d8443d21785d8ba7e297138a4
3
+ size 4728