Hugging Face's logo

lucas-meyer
/
xls-r-asr_af-run5

Automatic Speech Recognition
Transformers
PyTorch
TensorBoard
wav2vec2
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
1
lucas-meyer commited on
Commit
e661171
•
1 Parent(s): ef88929

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +95 -0
README.md ADDED
@@ -0,0 +1,95 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - wer
7
+ model-index:
8
+ - name: wav2vec2-xls-r-300m-asr_af-run2
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-xls-r-300m-asr_af-run2
16
+
17
+ This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.5239
20
+ - Wer: 0.3726
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: 0.0003
40
+ - train_batch_size: 16
41
+ - eval_batch_size: 8
42
+ - seed: 42
43
+ - gradient_accumulation_steps: 2
44
+ - total_train_batch_size: 32
45
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
46
+ - lr_scheduler_type: linear
47
+ - lr_scheduler_warmup_steps: 500
48
+ - num_epochs: 30
49
+ - mixed_precision_training: Native AMP
50
+
51
+ ### Training results
52
+
53
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
54
+ |:-------------:|:-----:|:----:|:---------------:|:------:|
55
+ | 11.7451 | 0.58 | 50 | 7.2710 | 1.0 |
56
+ | 4.7931 | 1.17 | 100 | 4.0381 | 1.0 |
57
+ | 3.4944 | 1.75 | 150 | 3.1782 | 1.0 |
58
+ | 3.06 | 2.34 | 200 | 2.9951 | 1.0 |
59
+ | 3.0031 | 2.92 | 250 | 2.9964 | 1.0 |
60
+ | 2.9814 | 3.51 | 300 | 2.9652 | 1.0 |
61
+ | 2.9524 | 4.09 | 350 | 2.9419 | 0.9998 |
62
+ | 2.9014 | 4.68 | 400 | 2.8213 | 1.0 |
63
+ | 2.2569 | 5.26 | 450 | 1.6105 | 0.9433 |
64
+ | 1.3008 | 5.85 | 500 | 1.0090 | 0.8021 |
65
+ | 0.8965 | 6.43 | 550 | 0.7727 | 0.6551 |
66
+ | 0.7134 | 7.02 | 600 | 0.6579 | 0.6102 |
67
+ | 0.5275 | 7.6 | 650 | 0.5956 | 0.6005 |
68
+ | 0.4413 | 8.19 | 700 | 0.5558 | 0.5079 |
69
+ | 0.3582 | 8.77 | 750 | 0.5719 | 0.5459 |
70
+ | 0.296 | 9.36 | 800 | 0.5389 | 0.4822 |
71
+ | 0.2557 | 9.94 | 850 | 0.4608 | 0.4541 |
72
+ | 0.2051 | 10.53 | 900 | 0.4822 | 0.4290 |
73
+ | 0.1911 | 11.11 | 950 | 0.5035 | 0.4209 |
74
+ | 0.1635 | 11.7 | 1000 | 0.5319 | 0.4263 |
75
+ | 0.1582 | 12.28 | 1050 | 0.5075 | 0.4124 |
76
+ | 0.1387 | 12.87 | 1100 | 0.4759 | 0.4055 |
77
+ | 0.1251 | 13.45 | 1150 | 0.4925 | 0.3970 |
78
+ | 0.1164 | 14.04 | 1200 | 0.4933 | 0.3998 |
79
+ | 0.1052 | 14.62 | 1250 | 0.4587 | 0.3995 |
80
+ | 0.1023 | 15.2 | 1300 | 0.4863 | 0.3950 |
81
+ | 0.0918 | 15.79 | 1350 | 0.5114 | 0.3858 |
82
+ | 0.09 | 16.37 | 1400 | 0.5444 | 0.3940 |
83
+ | 0.086 | 16.96 | 1450 | 0.5071 | 0.3806 |
84
+ | 0.0798 | 17.54 | 1500 | 0.4914 | 0.3809 |
85
+ | 0.0728 | 18.13 | 1550 | 0.5425 | 0.3807 |
86
+ | 0.0678 | 18.71 | 1600 | 0.5221 | 0.3731 |
87
+ | 0.0667 | 19.3 | 1650 | 0.5239 | 0.3726 |
88
+
89
+
90
+ ### Framework versions
91
+
92
+ - Transformers 4.28.0
93
+ - Pytorch 2.0.1+cu118
94
+ - Datasets 2.14.4
95
+ - Tokenizers 0.13.3