Hugging Face's logo

Rishavnine
/
wav2vecresults

Automatic Speech Recognition
Transformers
Safetensors
wav2vec2
Generated from Trainer
Model card Files
xet
 Community
Rishavnine commited on
Commit
080e2de
·
verified ·
1 Parent(s): b8ecab7

End of training

Browse files
Files changed (1) hide show
  1. README.md +67 -37
README.md CHANGED
@@ -1,6 +1,5 @@
1
  ---
2
  library_name: transformers
3
- base_model: anish-shilpakar/wav2vec2-nepali
4
  tags:
5
  - generated_from_trainer
6
  metrics:
@@ -15,10 +14,10 @@ should probably proofread and complete it, then remove this comment. -->
15
 
16
  # wav2vecresults
17
 
18
- This model is a fine-tuned version of [anish-shilpakar/wav2vec2-nepali](https://huggingface.co/anish-shilpakar/wav2vec2-nepali) on an unknown dataset.
19
  It achieves the following results on the evaluation set:
20
- - Loss: 0.3861
21
- - Wer: 0.3466
22
 
23
  ## Model description
24
 
@@ -43,43 +42,74 @@ The following hyperparameters were used during training:
43
  - seed: 42
44
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
45
  - lr_scheduler_type: linear
46
- - num_epochs: 8
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
- | Training Loss | Epoch | Step | Validation Loss | Wer |
52
- |:-------------:|:------:|:----:|:---------------:|:------:|
53
- | 3.3595 | 0.2584 | 100 | 3.0147 | 1.0 |
54
- | 2.7713 | 0.5168 | 200 | 2.6457 | 0.9987 |
55
- | 2.3829 | 0.7752 | 300 | 2.1221 | 0.9779 |
56
- | 1.6705 | 1.0336 | 400 | 1.4914 | 0.8859 |
57
- | 1.4199 | 1.2920 | 500 | 1.0897 | 0.8255 |
58
- | 1.0916 | 1.5504 | 600 | 0.8375 | 0.7083 |
59
- | 0.9454 | 1.8088 | 700 | 0.7106 | 0.5955 |
60
- | 1.052 | 2.0672 | 800 | 0.6276 | 0.5671 |
61
- | 0.8545 | 2.3256 | 900 | 0.5644 | 0.5180 |
62
- | 0.7128 | 2.5840 | 1000 | 0.5431 | 0.4991 |
63
- | 0.6581 | 2.8424 | 1100 | 0.5072 | 0.4675 |
64
- | 0.6191 | 3.1008 | 1200 | 0.4774 | 0.4222 |
65
- | 0.7273 | 3.3592 | 1300 | 0.4636 | 0.4184 |
66
- | 0.653 | 3.6176 | 1400 | 0.4441 | 0.4083 |
67
- | 0.5957 | 3.8760 | 1500 | 0.4392 | 0.4178 |
68
- | 0.67 | 4.1344 | 1600 | 0.4311 | 0.4071 |
69
- | 0.5323 | 4.3928 | 1700 | 0.4392 | 0.3995 |
70
- | 0.4944 | 4.6512 | 1800 | 0.4168 | 0.3642 |
71
- | 0.5163 | 4.9096 | 1900 | 0.4111 | 0.3812 |
72
- | 0.4862 | 5.1680 | 2000 | 0.4182 | 0.3724 |
73
- | 0.4327 | 5.4264 | 2100 | 0.4103 | 0.3642 |
74
- | 0.5428 | 5.6848 | 2200 | 0.4110 | 0.3629 |
75
- | 0.5885 | 5.9432 | 2300 | 0.3904 | 0.3617 |
76
- | 0.4507 | 6.2016 | 2400 | 0.3933 | 0.3459 |
77
- | 0.4821 | 6.4599 | 2500 | 0.3999 | 0.3459 |
78
- | 0.3707 | 6.7183 | 2600 | 0.3910 | 0.3516 |
79
- | 0.4995 | 6.9767 | 2700 | 0.3839 | 0.3522 |
80
- | 0.4108 | 7.2351 | 2800 | 0.3882 | 0.3478 |
81
- | 0.4508 | 7.4935 | 2900 | 0.3888 | 0.3447 |
82
- | 0.4551 | 7.7519 | 3000 | 0.3861 | 0.3466 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
83
 
84
 
85
  ### Framework versions
 
1
  ---
2
  library_name: transformers
 
3
  tags:
4
  - generated_from_trainer
5
  metrics:
 
14
 
15
  # wav2vecresults
16
 
17
+ This model was trained from scratch on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.3445
20
+ - Wer: 0.2804
21
 
22
  ## Model description
23
 
 
42
  - seed: 42
43
  - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
44
  - lr_scheduler_type: linear
45
+ - num_epochs: 16
46
  - mixed_precision_training: Native AMP
47
 
48
  ### Training results
49
 
50
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
51
+ |:-------------:|:-------:|:----:|:---------------:|:------:|
52
+ | 3.3595 | 0.2584 | 100 | 3.0147 | 1.0 |
53
+ | 2.7713 | 0.5168 | 200 | 2.6457 | 0.9987 |
54
+ | 2.3829 | 0.7752 | 300 | 2.1221 | 0.9779 |
55
+ | 1.6705 | 1.0336 | 400 | 1.4914 | 0.8859 |
56
+ | 1.4199 | 1.2920 | 500 | 1.0897 | 0.8255 |
57
+ | 1.0916 | 1.5504 | 600 | 0.8375 | 0.7083 |
58
+ | 0.9454 | 1.8088 | 700 | 0.7106 | 0.5955 |
59
+ | 1.052 | 2.0672 | 800 | 0.6276 | 0.5671 |
60
+ | 0.8545 | 2.3256 | 900 | 0.5644 | 0.5180 |
61
+ | 0.7128 | 2.5840 | 1000 | 0.5431 | 0.4991 |
62
+ | 0.6581 | 2.8424 | 1100 | 0.5072 | 0.4675 |
63
+ | 0.6191 | 3.1008 | 1200 | 0.4774 | 0.4222 |
64
+ | 0.7273 | 3.3592 | 1300 | 0.4636 | 0.4184 |
65
+ | 0.653 | 3.6176 | 1400 | 0.4441 | 0.4083 |
66
+ | 0.5957 | 3.8760 | 1500 | 0.4392 | 0.4178 |
67
+ | 0.67 | 4.1344 | 1600 | 0.4311 | 0.4071 |
68
+ | 0.5323 | 4.3928 | 1700 | 0.4392 | 0.3995 |
69
+ | 0.4944 | 4.6512 | 1800 | 0.4168 | 0.3642 |
70
+ | 0.5163 | 4.9096 | 1900 | 0.4111 | 0.3812 |
71
+ | 0.4862 | 5.1680 | 2000 | 0.4182 | 0.3724 |
72
+ | 0.4327 | 5.4264 | 2100 | 0.4103 | 0.3642 |
73
+ | 0.5428 | 5.6848 | 2200 | 0.4110 | 0.3629 |
74
+ | 0.5885 | 5.9432 | 2300 | 0.3904 | 0.3617 |
75
+ | 0.4507 | 6.2016 | 2400 | 0.3933 | 0.3459 |
76
+ | 0.4821 | 6.4599 | 2500 | 0.3999 | 0.3459 |
77
+ | 0.3707 | 6.7183 | 2600 | 0.3910 | 0.3516 |
78
+ | 0.4995 | 6.9767 | 2700 | 0.3839 | 0.3522 |
79
+ | 0.4108 | 7.2351 | 2800 | 0.3882 | 0.3478 |
80
+ | 0.4508 | 7.4935 | 2900 | 0.3888 | 0.3447 |
81
+ | 0.4551 | 7.7519 | 3000 | 0.3861 | 0.3466 |
82
+ | 0.2842 | 8.0103 | 3100 | 0.4146 | 0.3415 |
83
+ | 0.4566 | 8.2687 | 3200 | 0.4189 | 0.3592 |
84
+ | 0.4027 | 8.5271 | 3300 | 0.3945 | 0.3497 |
85
+ | 0.4543 | 8.7855 | 3400 | 0.4095 | 0.3503 |
86
+ | 0.325 | 9.0439 | 3500 | 0.3837 | 0.3308 |
87
+ | 0.4744 | 9.3023 | 3600 | 0.3791 | 0.3579 |
88
+ | 0.3841 | 9.5607 | 3700 | 0.3815 | 0.3144 |
89
+ | 0.4444 | 9.8191 | 3800 | 0.3729 | 0.3182 |
90
+ | 0.3398 | 10.0775 | 3900 | 0.3670 | 0.3239 |
91
+ | 0.3695 | 10.3359 | 4000 | 0.3637 | 0.3157 |
92
+ | 0.3802 | 10.5943 | 4100 | 0.3536 | 0.3081 |
93
+ | 0.3958 | 10.8527 | 4200 | 0.3440 | 0.2974 |
94
+ | 0.3055 | 11.1111 | 4300 | 0.3593 | 0.3031 |
95
+ | 0.3651 | 11.3695 | 4400 | 0.3543 | 0.2892 |
96
+ | 0.3045 | 11.6279 | 4500 | 0.3526 | 0.2936 |
97
+ | 0.3498 | 11.8863 | 4600 | 0.3503 | 0.2861 |
98
+ | 0.3108 | 12.1447 | 4700 | 0.3564 | 0.2905 |
99
+ | 0.3688 | 12.4031 | 4800 | 0.3502 | 0.2968 |
100
+ | 0.3453 | 12.6615 | 4900 | 0.3558 | 0.3050 |
101
+ | 0.2544 | 12.9199 | 5000 | 0.3508 | 0.2892 |
102
+ | 0.3211 | 13.1783 | 5100 | 0.3615 | 0.2924 |
103
+ | 0.2879 | 13.4367 | 5200 | 0.3497 | 0.2886 |
104
+ | 0.2663 | 13.6951 | 5300 | 0.3582 | 0.3050 |
105
+ | 0.3259 | 13.9535 | 5400 | 0.3532 | 0.2911 |
106
+ | 0.2812 | 14.2119 | 5500 | 0.3555 | 0.2886 |
107
+ | 0.2298 | 14.4703 | 5600 | 0.3407 | 0.2728 |
108
+ | 0.3442 | 14.7287 | 5700 | 0.3527 | 0.2804 |
109
+ | 0.2619 | 14.9871 | 5800 | 0.3434 | 0.2716 |
110
+ | 0.2779 | 15.2455 | 5900 | 0.3424 | 0.2873 |
111
+ | 0.2758 | 15.5039 | 6000 | 0.3448 | 0.2829 |
112
+ | 0.2339 | 15.7623 | 6100 | 0.3445 | 0.2804 |
113
 
114
 
115
  ### Framework versions