Hugging Face's logo

aapot
/
wav2vec2-xlsr-300m-finnish

Automatic Speech Recognition
Transformers
PyTorch
Finnish
wav2vec2
finnish
Generated from Trainer
hf-asr-leaderboard
robust-speech-event
Eval Results
Inference Endpoints
Model card Files Community
aapot commited on
Commit
9fb4b95
·
1 Parent(s): 6e7f226

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +117 -0
README.md ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: wav2vec2-xlsr-300m-finnish
7
+ results: []
8
+ ---
9
+
10
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
11
+ should probably proofread and complete it, then remove this comment. -->
12
+
13
+ # wav2vec2-xlsr-300m-finnish
14
+
15
+ This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the None dataset.
16
+ It achieves the following results on the evaluation set:
17
+ - Loss: 0.1484
18
+ - Wer: 0.1800
19
+
20
+ ## Model description
21
+
22
+ More information needed
23
+
24
+ ## Intended uses & limitations
25
+
26
+ More information needed
27
+
28
+ ## Training and evaluation data
29
+
30
+ More information needed
31
+
32
+ ## Training procedure
33
+
34
+ ### Training hyperparameters
35
+
36
+ The following hyperparameters were used during training:
37
+ - learning_rate: 0.0005
38
+ - train_batch_size: 32
39
+ - eval_batch_size: 32
40
+ - seed: 42
41
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
42
+ - lr_scheduler_type: linear
43
+ - lr_scheduler_warmup_steps: 500
44
+ - num_epochs: 10
45
+ - mixed_precision_training: Native AMP
46
+
47
+ ### Training results
48
+
49
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
50
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
51
+ | 0.973 | 0.17 | 500 | 0.5750 | 0.6844 |
52
+ | 0.713 | 0.34 | 1000 | 0.3356 | 0.4518 |
53
+ | 0.6563 | 0.5 | 1500 | 0.3007 | 0.4039 |
54
+ | 0.642 | 0.67 | 2000 | 0.2619 | 0.3674 |
55
+ | 0.6203 | 0.84 | 2500 | 0.2488 | 0.3558 |
56
+ | 0.6016 | 1.01 | 3000 | 0.2795 | 0.3835 |
57
+ | 0.5423 | 1.17 | 3500 | 0.2652 | 0.3310 |
58
+ | 0.5639 | 1.34 | 4000 | 0.2479 | 0.3462 |
59
+ | 0.586 | 1.51 | 4500 | 0.2409 | 0.3295 |
60
+ | 0.5169 | 1.68 | 5000 | 0.2728 | 0.3352 |
61
+ | 0.5176 | 1.84 | 5500 | 0.2254 | 0.3149 |
62
+ | 0.4983 | 2.01 | 6000 | 0.2169 | 0.3009 |
63
+ | 0.4982 | 2.18 | 6500 | 0.2215 | 0.3079 |
64
+ | 0.4898 | 2.35 | 7000 | 0.2174 | 0.3023 |
65
+ | 0.4922 | 2.51 | 7500 | 0.2217 | 0.3081 |
66
+ | 0.5025 | 2.68 | 8000 | 0.2002 | 0.2710 |
67
+ | 0.4745 | 2.85 | 8500 | 0.1935 | 0.2783 |
68
+ | 0.4377 | 3.02 | 9000 | 0.1859 | 0.2742 |
69
+ | 0.4511 | 3.18 | 9500 | 0.2038 | 0.2786 |
70
+ | 0.4411 | 3.35 | 10000 | 0.1863 | 0.2651 |
71
+ | 0.4501 | 3.52 | 10500 | 0.1948 | 0.2605 |
72
+ | 0.4557 | 3.69 | 11000 | 0.1872 | 0.2695 |
73
+ | 0.4493 | 3.85 | 11500 | 0.1888 | 0.2632 |
74
+ | 0.4047 | 4.02 | 12000 | 0.1818 | 0.2559 |
75
+ | 0.4319 | 4.19 | 12500 | 0.1896 | 0.2648 |
76
+ | 0.4162 | 4.36 | 13000 | 0.1953 | 0.2595 |
77
+ | 0.4046 | 4.52 | 13500 | 0.1864 | 0.2606 |
78
+ | 0.4195 | 4.69 | 14000 | 0.1843 | 0.2467 |
79
+ | 0.4146 | 4.86 | 14500 | 0.1686 | 0.2450 |
80
+ | 0.378 | 5.03 | 15000 | 0.1731 | 0.2401 |
81
+ | 0.3792 | 5.19 | 15500 | 0.1676 | 0.2325 |
82
+ | 0.3855 | 5.36 | 16000 | 0.1740 | 0.2326 |
83
+ | 0.4029 | 5.53 | 16500 | 0.1674 | 0.2345 |
84
+ | 0.386 | 5.7 | 17000 | 0.1735 | 0.2280 |
85
+ | 0.3811 | 5.86 | 17500 | 0.1692 | 0.2258 |
86
+ | 0.3607 | 6.03 | 18000 | 0.1797 | 0.2279 |
87
+ | 0.3604 | 6.2 | 18500 | 0.1651 | 0.2206 |
88
+ | 0.3362 | 6.37 | 19000 | 0.1627 | 0.2199 |
89
+ | 0.3611 | 6.53 | 19500 | 0.1652 | 0.2172 |
90
+ | 0.3671 | 6.7 | 20000 | 0.1564 | 0.2140 |
91
+ | 0.3769 | 6.87 | 20500 | 0.1525 | 0.2101 |
92
+ | 0.3539 | 7.04 | 21000 | 0.1639 | 0.2096 |
93
+ | 0.3225 | 7.21 | 21500 | 0.1611 | 0.2087 |
94
+ | 0.3323 | 7.37 | 22000 | 0.1633 | 0.2008 |
95
+ | 0.3327 | 7.54 | 22500 | 0.1692 | 0.1975 |
96
+ | 0.3456 | 7.71 | 23000 | 0.1555 | 0.1991 |
97
+ | 0.3058 | 7.88 | 23500 | 0.1590 | 0.1959 |
98
+ | 0.3034 | 8.04 | 24000 | 0.1531 | 0.1973 |
99
+ | 0.2925 | 8.21 | 24500 | 0.1583 | 0.1978 |
100
+ | 0.2967 | 8.38 | 25000 | 0.1546 | 0.1906 |
101
+ | 0.2974 | 8.55 | 25500 | 0.1540 | 0.1869 |
102
+ | 0.3131 | 8.71 | 26000 | 0.1534 | 0.1850 |
103
+ | 0.3306 | 8.88 | 26500 | 0.1482 | 0.1844 |
104
+ | 0.2842 | 9.05 | 27000 | 0.1490 | 0.1854 |
105
+ | 0.2879 | 9.22 | 27500 | 0.1463 | 0.1799 |
106
+ | 0.27 | 9.38 | 28000 | 0.1454 | 0.1798 |
107
+ | 0.2874 | 9.55 | 28500 | 0.1504 | 0.1787 |
108
+ | 0.2757 | 9.72 | 29000 | 0.1512 | 0.1784 |
109
+ | 0.3017 | 9.89 | 29500 | 0.1484 | 0.1800 |
110
+
111
+
112
+ ### Framework versions
113
+
114
+ - Transformers 4.17.0.dev0
115
+ - Pytorch 1.10.2+cu102
116
+ - Datasets 1.18.3
117
+ - Tokenizers 0.11.0