Hugging Face's logo

dmatekenya
/
whisper-large-v3-chichewa

Automatic Speech Recognition
Transformers
Safetensors
whisper
Generated from Trainer
Inference Endpoints
Model card Files Community
dmatekenya commited on
Commit
c2d1aa7
•
1 Parent(s): b0b9b13

End of training

Browse files
Files changed (4) hide show
  1. README.md +18 -13
  2. generation_config.json +1 -1
  3. model-00001-of-00002.safetensors +1 -1
  4. model-00002-of-00002.safetensors +1 -1
README.md CHANGED
@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
18
 
19
  This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 2.7213
22
- - Wer: 77.8505
23
 
24
  ## Model description
25
 
@@ -45,23 +45,28 @@ The following hyperparameters were used during training:
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_steps: 500
48
- - training_steps: 5000
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
- | Training Loss | Epoch | Step | Validation Loss | Wer |
54
- |:-------------:|:-------:|:----:|:---------------:|:-------:|
55
- | 0.6163 | 7.4627 | 1000 | 1.8803 | 79.6019 |
56
- | 0.1102 | 14.9254 | 2000 | 2.2954 | 81.6099 |
57
- | 0.0654 | 22.3881 | 3000 | 2.5129 | 87.0942 |
58
- | 0.0502 | 29.8507 | 4000 | 2.6569 | 77.7621 |
59
- | 0.0482 | 37.3134 | 5000 | 2.7213 | 77.8505 |
 
 
 
 
 
60
 
61
 
62
  ### Framework versions
63
 
64
- - Transformers 4.44.2
65
  - Pytorch 2.0.1
66
- - Datasets 3.0.0
67
- - Tokenizers 0.19.1
 
18
 
19
  This model is a fine-tuned version of [openai/whisper-large-v3](https://huggingface.co/openai/whisper-large-v3) on the None dataset.
20
  It achieves the following results on the evaluation set:
21
+ - Loss: 2.9810
22
+ - Wer: 101.7738
23
 
24
  ## Model description
25
 
 
45
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
  - lr_scheduler_type: linear
47
  - lr_scheduler_warmup_steps: 500
48
+ - training_steps: 10000
49
  - mixed_precision_training: Native AMP
50
 
51
  ### Training results
52
 
53
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
54
+ |:-------------:|:-------:|:-----:|:---------------:|:--------:|
55
+ | 0.9765 | 5.2910 | 1000 | 1.8781 | 198.6546 |
56
+ | 0.2576 | 10.5820 | 2000 | 2.2548 | 81.4057 |
57
+ | 0.1191 | 15.8730 | 3000 | 2.4730 | 80.9351 |
58
+ | 0.0755 | 21.1640 | 4000 | 2.6557 | 79.4872 |
59
+ | 0.0726 | 26.4550 | 5000 | 2.6869 | 78.3228 |
60
+ | 0.0722 | 31.7460 | 6000 | 2.7548 | 93.2428 |
61
+ | 0.0624 | 37.0370 | 7000 | 2.8531 | 79.5958 |
62
+ | 0.0605 | 42.3280 | 8000 | 2.8869 | 77.6893 |
63
+ | 0.062 | 47.6190 | 9000 | 2.9388 | 86.6606 |
64
+ | 0.0515 | 52.9101 | 10000 | 2.9810 | 101.7738 |
65
 
66
 
67
  ### Framework versions
68
 
69
+ - Transformers 4.45.1
70
  - Pytorch 2.0.1
71
+ - Datasets 3.0.1
72
+ - Tokenizers 0.20.0
generation_config.json CHANGED
@@ -253,5 +253,5 @@
253
  "transcribe": 50360,
254
  "translate": 50359
255
  },
256
- "transformers_version": "4.44.2"
257
  }
 
253
  "transcribe": 50360,
254
  "translate": 50359
255
  },
256
+ "transformers_version": "4.45.1"
257
  }
model-00001-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ac2845a05cc20e5c919e164acf0c45faccabc19442357bea35ba626c1305403c
3
  size 4993448880
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f20fe1e327718b0a7fc8bd41ffd91087172b3781c6f1307344f4f6fcf7fdd6f9
3
  size 4993448880
model-00002-of-00002.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:815f09081a28a51eb46c59c981e79bb3fab958ae9b5e31375c6609b3545dbdba
3
  size 1180663192
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:aa52bcacc8c27812020e72c41820294a2035274127772a1eaa7ab9ea96df24bd
3
  size 1180663192