Hugging Face's logo

greenw0lf
/
wav2vec2-large-xls-r-1b-frisian-cv-8

Automatic Speech Recognition
Transformers
PyTorch
wav2vec2
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Community
greenw0lf commited on
Commit
9429cf2
·
1 Parent(s): e4df855

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +51 -50
README.md CHANGED
@@ -21,20 +21,7 @@ model-index:
21
  metrics:
22
  - name: Wer
23
  type: wer
24
- value: 0.14290815597771747
25
- - task:
26
- name: Automatic Speech Recognition
27
- type: automatic-speech-recognition
28
- dataset:
29
- name: common_voice_8_0
30
- type: common_voice_8_0
31
- config: fy-NL
32
- split: test
33
- args: fy-NL
34
- metrics:
35
- - name: Wer
36
- type: wer
37
- value: 0.1413499060557884
38
  ---
39
 
40
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -44,11 +31,8 @@ should probably proofread and complete it, then remove this comment. -->
44
 
45
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the common_voice_8_0 dataset.
46
  It achieves the following results on the evaluation set:
47
- - Loss: 0.2131
48
- - Wer: 0.1429
49
-
50
- And on the test set:
51
- - Wer: 0.1413
52
 
53
  ## Model description
54
 
@@ -67,49 +51,66 @@ More information needed
67
  ### Training hyperparameters
68
 
69
  The following hyperparameters were used during training:
70
- - learning_rate: 5e-05
71
  - train_batch_size: 32
72
  - eval_batch_size: 8
73
  - seed: 42
74
  - optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
75
  - lr_scheduler_type: linear
76
  - lr_scheduler_warmup_ratio: 0.1
77
- - num_epochs: 50
78
  - mixed_precision_training: Native AMP
79
 
80
  ### Training results
81
 
82
  | Training Loss | Epoch | Step | Validation Loss | Wer |
83
  |:-------------:|:-----:|:----:|:---------------:|:------:|
84
- | 6.0565 | 1.72 | 200 | 3.1053 | 1.0 |
85
- | 2.7675 | 3.45 | 400 | 1.1551 | 0.8611 |
86
- | 1.3474 | 5.17 | 600 | 0.4770 | 0.4397 |
87
- | 0.9617 | 6.9 | 800 | 0.3218 | 0.3343 |
88
- | 0.9058 | 8.62 | 1000 | 0.2741 | 0.2768 |
89
- | 0.9712 | 10.34 | 1200 | 0.2619 | 0.2505 |
90
- | 0.6908 | 12.07 | 1400 | 0.2288 | 0.2243 |
91
- | 0.745 | 13.79 | 1600 | 0.2288 | 0.2095 |
92
- | 0.7742 | 15.52 | 1800 | 0.2289 | 0.1979 |
93
- | 0.7231 | 17.24 | 2000 | 0.2198 | 0.1940 |
94
- | 0.6475 | 18.97 | 2200 | 0.2180 | 0.1992 |
95
- | 0.6421 | 20.69 | 2400 | 0.2133 | 0.1741 |
96
- | 0.5925 | 22.41 | 2600 | 0.1998 | 0.1747 |
97
- | 0.5608 | 24.14 | 2800 | 0.2212 | 0.1950 |
98
- | 0.5315 | 25.86 | 3000 | 0.2187 | 0.1624 |
99
- | 0.5362 | 27.59 | 3200 | 0.2057 | 0.1718 |
100
- | 0.563 | 29.31 | 3400 | 0.2090 | 0.1613 |
101
- | 0.4218 | 31.03 | 3600 | 0.2126 | 0.1531 |
102
- | 0.3826 | 32.76 | 3800 | 0.2084 | 0.1538 |
103
- | 0.356 | 34.48 | 4000 | 0.2115 | 0.1612 |
104
- | 0.2966 | 36.21 | 4200 | 0.2093 | 0.1536 |
105
- | 0.3377 | 37.93 | 4400 | 0.2061 | 0.1527 |
106
- | 0.321 | 39.66 | 4600 | 0.2121 | 0.1463 |
107
- | 0.2942 | 41.38 | 4800 | 0.2158 | 0.1441 |
108
- | 0.2931 | 43.1 | 5000 | 0.2173 | 0.1446 |
109
- | 0.2346 | 44.83 | 5200 | 0.2152 | 0.1436 |
110
- | 0.2543 | 46.55 | 5400 | 0.2066 | 0.1445 |
111
- | 0.2385 | 48.28 | 5600 | 0.2108 | 0.1432 |
112
- | 0.2726 | 50.0 | 5800 | 0.2131 | 0.1429 |
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
113
 
114
 
115
  ### Framework versions
 
21
  metrics:
22
  - name: Wer
23
  type: wer
24
+ value: 0.1339808598771604
 
 
 
 
 
 
 
 
 
 
 
 
 
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
31
 
32
  This model is a fine-tuned version of [facebook/wav2vec2-xls-r-1b](https://huggingface.co/facebook/wav2vec2-xls-r-1b) on the common_voice_8_0 dataset.
33
  It achieves the following results on the evaluation set:
34
+ - Loss: 0.2054
35
+ - Wer: 0.1340
 
 
 
36
 
37
  ## Model description
38
 
 
51
  ### Training hyperparameters
52
 
53
  The following hyperparameters were used during training:
54
+ - learning_rate: 3e-05
55
  - train_batch_size: 32
56
  - eval_batch_size: 8
57
  - seed: 42
58
  - optimizer: Adam with betas=(0.9,0.98) and epsilon=1e-08
59
  - lr_scheduler_type: linear
60
  - lr_scheduler_warmup_ratio: 0.1
61
+ - num_epochs: 80
62
  - mixed_precision_training: Native AMP
63
 
64
  ### Training results
65
 
66
  | Training Loss | Epoch | Step | Validation Loss | Wer |
67
  |:-------------:|:-----:|:----:|:---------------:|:------:|
68
+ | 3.0483 | 1.72 | 200 | 3.0438 | 1.0 |
69
+ | 2.6284 | 3.45 | 400 | 1.1501 | 0.9229 |
70
+ | 1.4359 | 5.17 | 600 | 0.5618 | 0.5329 |
71
+ | 1.1366 | 6.9 | 800 | 0.3899 | 0.3845 |
72
+ | 0.988 | 8.62 | 1000 | 0.3370 | 0.3302 |
73
+ | 0.8377 | 10.34 | 1200 | 0.2765 | 0.2834 |
74
+ | 0.8001 | 12.07 | 1400 | 0.2750 | 0.2438 |
75
+ | 0.8678 | 13.79 | 1600 | 0.2258 | 0.2160 |
76
+ | 0.7023 | 15.52 | 1800 | 0.2260 | 0.2072 |
77
+ | 0.8111 | 17.24 | 2000 | 0.2223 | 0.2070 |
78
+ | 0.658 | 18.97 | 2200 | 0.2121 | 0.1834 |
79
+ | 0.6574 | 20.69 | 2400 | 0.2136 | 0.1812 |
80
+ | 0.7521 | 22.41 | 2600 | 0.2175 | 0.1775 |
81
+ | 0.6515 | 24.14 | 2800 | 0.2018 | 0.1718 |
82
+ | 0.6955 | 25.86 | 3000 | 0.2121 | 0.1863 |
83
+ | 0.6605 | 27.59 | 3200 | 0.2003 | 0.1607 |
84
+ | 0.5403 | 29.31 | 3400 | 0.2042 | 0.1668 |
85
+ | 0.5064 | 31.03 | 3600 | 0.2021 | 0.1616 |
86
+ | 0.6811 | 32.76 | 3800 | 0.2026 | 0.1668 |
87
+ | 0.6787 | 34.48 | 4000 | 0.2122 | 0.1613 |
88
+ | 0.5595 | 36.21 | 4200 | 0.2001 | 0.1547 |
89
+ | 0.5225 | 37.93 | 4400 | 0.1992 | 0.1615 |
90
+ | 0.5522 | 39.66 | 4600 | 0.2023 | 0.1603 |
91
+ | 0.5364 | 41.38 | 4800 | 0.1992 | 0.1531 |
92
+ | 0.5157 | 43.1 | 5000 | 0.2060 | 0.1550 |
93
+ | 0.4382 | 44.83 | 5200 | 0.1985 | 0.1427 |
94
+ | 0.3658 | 46.55 | 5400 | 0.1964 | 0.1427 |
95
+ | 0.5336 | 48.28 | 5600 | 0.2143 | 0.1471 |
96
+ | 0.5479 | 50.0 | 5800 | 0.1962 | 0.1402 |
97
+ | 0.5203 | 51.72 | 6000 | 0.2022 | 0.1418 |
98
+ | 0.363 | 53.45 | 6200 | 0.2103 | 0.1429 |
99
+ | 0.3828 | 55.17 | 6400 | 0.2070 | 0.1417 |
100
+ | 0.3875 | 56.9 | 6600 | 0.2070 | 0.1411 |
101
+ | 0.3433 | 58.62 | 6800 | 0.2049 | 0.1418 |
102
+ | 0.2826 | 60.34 | 7000 | 0.2047 | 0.1417 |
103
+ | 0.294 | 62.07 | 7200 | 0.2022 | 0.1369 |
104
+ | 0.2776 | 63.79 | 7400 | 0.2115 | 0.1365 |
105
+ | 0.3178 | 65.52 | 7600 | 0.2005 | 0.1377 |
106
+ | 0.2913 | 67.24 | 7800 | 0.2047 | 0.1355 |
107
+ | 0.2642 | 68.97 | 8000 | 0.2069 | 0.1338 |
108
+ | 0.255 | 70.69 | 8200 | 0.2041 | 0.1336 |
109
+ | 0.2746 | 72.41 | 8400 | 0.2064 | 0.1331 |
110
+ | 0.2485 | 74.14 | 8600 | 0.2068 | 0.1327 |
111
+ | 0.2741 | 75.86 | 8800 | 0.2073 | 0.1331 |
112
+ | 0.2223 | 77.59 | 9000 | 0.2055 | 0.1338 |
113
+ | 0.2327 | 79.31 | 9200 | 0.2054 | 0.1340 |
114
 
115
 
116
  ### Framework versions