Hugging Face's logo

lgris
/
wav2vec2-large-xlsr-coraa-portuguese-cv7

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
pt
robust-speech-event
Inference Endpoints
Model card Files Metrics Community
lgris commited on
Commit
685c851
•
1 Parent(s): 5210222

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +111 -0
README.md ADDED
@@ -0,0 +1,111 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ datasets:
6
+ - common_voice
7
+ model-index:
8
+ - name: wav2vec2-large-xlsr-coraa-portuguese-cv7
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # wav2vec2-large-xlsr-coraa-portuguese-cv7
16
+
17
+ This model is a fine-tuned version of [Edresson/wav2vec2-large-xlsr-coraa-portuguese](https://huggingface.co/Edresson/wav2vec2-large-xlsr-coraa-portuguese) on the common_voice dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.1777
20
+ - Wer: 0.1339
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - learning_rate: 0.0001
40
+ - train_batch_size: 8
41
+ - eval_batch_size: 8
42
+ - seed: 42
43
+ - gradient_accumulation_steps: 2
44
+ - total_train_batch_size: 16
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 100
48
+ - training_steps: 5000
49
+
50
+ ### Training results
51
+
52
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
53
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
54
+ | 0.4779 | 0.13 | 100 | 0.2620 | 0.2020 |
55
+ | 0.4505 | 0.26 | 200 | 0.2339 | 0.1998 |
56
+ | 0.4285 | 0.39 | 300 | 0.2507 | 0.2109 |
57
+ | 0.4148 | 0.52 | 400 | 0.2311 | 0.2101 |
58
+ | 0.4072 | 0.65 | 500 | 0.2278 | 0.1899 |
59
+ | 0.388 | 0.78 | 600 | 0.2193 | 0.1898 |
60
+ | 0.3952 | 0.91 | 700 | 0.2108 | 0.1901 |
61
+ | 0.3851 | 1.04 | 800 | 0.2121 | 0.1788 |
62
+ | 0.3496 | 1.17 | 900 | 0.2154 | 0.1776 |
63
+ | 0.3063 | 1.3 | 1000 | 0.2095 | 0.1730 |
64
+ | 0.3376 | 1.43 | 1100 | 0.2129 | 0.1801 |
65
+ | 0.3273 | 1.56 | 1200 | 0.2132 | 0.1776 |
66
+ | 0.3347 | 1.69 | 1300 | 0.2054 | 0.1698 |
67
+ | 0.323 | 1.82 | 1400 | 0.1986 | 0.1724 |
68
+ | 0.3079 | 1.95 | 1500 | 0.2005 | 0.1701 |
69
+ | 0.3029 | 2.08 | 1600 | 0.2159 | 0.1644 |
70
+ | 0.2694 | 2.21 | 1700 | 0.1992 | 0.1678 |
71
+ | 0.2733 | 2.34 | 1800 | 0.2032 | 0.1657 |
72
+ | 0.269 | 2.47 | 1900 | 0.2056 | 0.1592 |
73
+ | 0.2869 | 2.6 | 2000 | 0.2058 | 0.1616 |
74
+ | 0.2813 | 2.73 | 2100 | 0.1868 | 0.1584 |
75
+ | 0.2616 | 2.86 | 2200 | 0.1841 | 0.1550 |
76
+ | 0.2809 | 2.99 | 2300 | 0.1902 | 0.1577 |
77
+ | 0.2598 | 3.12 | 2400 | 0.1910 | 0.1514 |
78
+ | 0.24 | 3.25 | 2500 | 0.1971 | 0.1555 |
79
+ | 0.2481 | 3.38 | 2600 | 0.1853 | 0.1537 |
80
+ | 0.2437 | 3.51 | 2700 | 0.1897 | 0.1496 |
81
+ | 0.2384 | 3.64 | 2800 | 0.1842 | 0.1495 |
82
+ | 0.2405 | 3.77 | 2900 | 0.1884 | 0.1500 |
83
+ | 0.2372 | 3.9 | 3000 | 0.1950 | 0.1548 |
84
+ | 0.229 | 4.03 | 3100 | 0.1928 | 0.1477 |
85
+ | 0.2047 | 4.16 | 3200 | 0.1891 | 0.1472 |
86
+ | 0.2102 | 4.29 | 3300 | 0.1930 | 0.1473 |
87
+ | 0.199 | 4.42 | 3400 | 0.1914 | 0.1456 |
88
+ | 0.2121 | 4.55 | 3500 | 0.1840 | 0.1437 |
89
+ | 0.211 | 4.67 | 3600 | 0.1843 | 0.1403 |
90
+ | 0.2072 | 4.8 | 3700 | 0.1836 | 0.1428 |
91
+ | 0.2224 | 4.93 | 3800 | 0.1747 | 0.1412 |
92
+ | 0.1974 | 5.06 | 3900 | 0.1813 | 0.1416 |
93
+ | 0.1895 | 5.19 | 4000 | 0.1869 | 0.1406 |
94
+ | 0.1763 | 5.32 | 4100 | 0.1830 | 0.1394 |
95
+ | 0.2001 | 5.45 | 4200 | 0.1775 | 0.1394 |
96
+ | 0.1909 | 5.58 | 4300 | 0.1806 | 0.1373 |
97
+ | 0.1812 | 5.71 | 4400 | 0.1784 | 0.1359 |
98
+ | 0.1737 | 5.84 | 4500 | 0.1778 | 0.1353 |
99
+ | 0.1915 | 5.97 | 4600 | 0.1777 | 0.1349 |
100
+ | 0.1921 | 6.1 | 4700 | 0.1784 | 0.1359 |
101
+ | 0.1805 | 6.23 | 4800 | 0.1757 | 0.1348 |
102
+ | 0.1742 | 6.36 | 4900 | 0.1771 | 0.1341 |
103
+ | 0.1709 | 6.49 | 5000 | 0.1777 | 0.1339 |
104
+
105
+
106
+ ### Framework versions
107
+
108
+ - Transformers 4.16.1
109
+ - Pytorch 1.10.0+cu111
110
+ - Datasets 1.18.2
111
+ - Tokenizers 0.11.0