Hugging Face's logo

TheAIchemist13
/
hindi_beekeeping_wav2vec2-2

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Community
TheAIchemist13 commited on
Commit
c01d984
•
1 Parent(s): 18bb068

End of training

Browse files
Files changed (2) hide show
  1. README.md +24 -20
  2. pytorch_model.bin +1 -1
README.md CHANGED
@@ -15,10 +15,10 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # hindi_beekeeping_wav2vec2-2
17
 
18
- This model is a fine-tuned version of [Harveenchadha/hindi_base_wav2vec2](https://huggingface.co/Harveenchadha/hindi_base_wav2vec2) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.7999
21
- - Wer: 0.3017
22
 
23
  ## Model description
24
 
@@ -52,27 +52,31 @@ The following hyperparameters were used during training:
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Wer |
54
  |:-------------:|:-----:|:----:|:---------------:|:------:|
55
- | 0.3186 | 5.56 | 25 | 0.5615 | 0.2890 |
56
- | 0.0831 | 11.11 | 50 | 1.0522 | 0.4262 |
57
- | 0.0499 | 16.67 | 75 | 0.7605 | 0.3354 |
58
- | 0.0518 | 22.22 | 100 | 0.6797 | 0.3713 |
59
- | 0.0355 | 27.78 | 125 | 0.8345 | 0.3333 |
60
- | 0.0284 | 33.33 | 150 | 0.8702 | 0.3608 |
61
- | 0.0248 | 38.89 | 175 | 0.7246 | 0.3734 |
62
- | 0.0229 | 44.44 | 200 | 0.7885 | 0.3291 |
63
- | 0.0193 | 50.0 | 225 | 0.8082 | 0.3312 |
64
- | 0.0164 | 55.56 | 250 | 0.7141 | 0.3186 |
65
- | 0.0109 | 61.11 | 275 | 0.9168 | 0.3333 |
66
- | 0.0082 | 66.67 | 300 | 0.9048 | 0.3418 |
67
- | 0.007 | 72.22 | 325 | 0.9089 | 0.3080 |
68
- | 0.0074 | 77.78 | 350 | 0.8113 | 0.2911 |
69
- | 0.0053 | 83.33 | 375 | 0.8197 | 0.3186 |
70
- | 0.0065 | 88.89 | 400 | 0.7999 | 0.3017 |
 
 
 
 
71
 
72
 
73
  ### Framework versions
74
 
75
- - Transformers 4.34.0
76
  - Pytorch 2.0.1+cu118
77
  - Datasets 2.14.5
78
  - Tokenizers 0.14.1
 
15
 
16
  # hindi_beekeeping_wav2vec2-2
17
 
18
+ This model is a fine-tuned version of [Harveenchadha/hindi_base_wav2vec2](https://huggingface.co/Harveenchadha/hindi_base_wav2vec2) on the None dataset.
19
  It achieves the following results on the evaluation set:
20
+ - Loss: 1.5244
21
+ - Wer: 0.3984
22
 
23
  ## Model description
24
 
 
52
 
53
  | Training Loss | Epoch | Step | Validation Loss | Wer |
54
  |:-------------:|:-----:|:----:|:---------------:|:------:|
55
+ | 0.3815 | 4.55 | 25 | 1.0849 | 0.3984 |
56
+ | 0.1143 | 9.09 | 50 | 1.0752 | 0.3659 |
57
+ | 0.0827 | 13.64 | 75 | 1.4514 | 0.4390 |
58
+ | 0.0546 | 18.18 | 100 | 1.6510 | 0.4146 |
59
+ | 0.0506 | 22.73 | 125 | 1.5120 | 0.4593 |
60
+ | 0.0394 | 27.27 | 150 | 1.5948 | 0.4797 |
61
+ | 0.0356 | 31.82 | 175 | 1.4317 | 0.4228 |
62
+ | 0.0314 | 36.36 | 200 | 1.7475 | 0.4553 |
63
+ | 0.0278 | 40.91 | 225 | 1.6500 | 0.4634 |
64
+ | 0.0335 | 45.45 | 250 | 1.5310 | 0.4756 |
65
+ | 0.0213 | 50.0 | 275 | 1.4817 | 0.4593 |
66
+ | 0.0177 | 54.55 | 300 | 1.6921 | 0.4228 |
67
+ | 0.0115 | 59.09 | 325 | 1.6241 | 0.4106 |
68
+ | 0.0152 | 63.64 | 350 | 1.5167 | 0.4106 |
69
+ | 0.0109 | 68.18 | 375 | 1.7510 | 0.4553 |
70
+ | 0.011 | 72.73 | 400 | 1.6425 | 0.4146 |
71
+ | 0.0101 | 77.27 | 425 | 1.6253 | 0.4146 |
72
+ | 0.0091 | 81.82 | 450 | 1.6383 | 0.3984 |
73
+ | 0.008 | 86.36 | 475 | 1.5251 | 0.3984 |
74
+ | 0.0069 | 90.91 | 500 | 1.5244 | 0.3984 |
75
 
76
 
77
  ### Framework versions
78
 
79
+ - Transformers 4.34.1
80
  - Pytorch 2.0.1+cu118
81
  - Datasets 2.14.5
82
  - Tokenizers 0.14.1
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:11ae6979831cdb066b37a94897d1aafc42d583dc060c66705cc70ab3b825fb71
3
  size 377818721
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5d8970100b8a53fdb43424c739d0586079b92c0cfba9fbe5cd22e9de2352002c
3
  size 377818721