Hugging Face's logo

saattrupdan
/
wav2vec2-xls-r-300m-ftspeech

Automatic Speech Recognition
Transformers
PyTorch
Safetensors
Danish
wav2vec2
Eval Results
Inference Endpoints
Model card Files Community
2
saattrupdan commited on
Commit
75f9b2e
•
1 Parent(s): 69d0c52

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +116 -0
README.md ADDED
@@ -0,0 +1,116 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ license: apache-2.0
3
+ tags:
4
+ - generated_from_trainer
5
+ model-index:
6
+ - name: wav2vec2-xls-r-300m-ftspeech
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-xls-r-300m-ftspeech
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: 17.8348
18
+ - Wer: 0.1186
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: 0.0001
38
+ - train_batch_size: 1
39
+ - eval_batch_size: 1
40
+ - seed: 4242
41
+ - gradient_accumulation_steps: 32
42
+ - total_train_batch_size: 32
43
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
+ - lr_scheduler_type: linear
45
+ - lr_scheduler_warmup_steps: 2000
46
+ - num_epochs: 3
47
+ - mixed_precision_training: Native AMP
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Wer |
52
+ |:-------------:|:-----:|:-----:|:---------------:|:------:|
53
+ | 311.274 | 0.02 | 500 | 329.9529 | 1.0 |
54
+ | 296.713 | 0.03 | 1000 | 305.5616 | 1.0000 |
55
+ | 69.6128 | 0.05 | 1500 | 77.2267 | 0.6466 |
56
+ | 47.4542 | 0.06 | 2000 | 56.0227 | 0.5101 |
57
+ | 39.2415 | 0.08 | 2500 | 40.1751 | 0.3483 |
58
+ | 35.9888 | 0.1 | 3000 | 33.1659 | 0.2619 |
59
+ | 34.1621 | 0.11 | 3500 | 30.4220 | 0.2296 |
60
+ | 32.3383 | 0.13 | 4000 | 28.3836 | 0.2214 |
61
+ | 31.1862 | 0.14 | 4500 | 28.7228 | 0.2220 |
62
+ | 29.818 | 0.16 | 5000 | 28.3220 | 0.2259 |
63
+ | 29.4729 | 0.18 | 5500 | 26.5646 | 0.2024 |
64
+ | 27.6171 | 0.19 | 6000 | 26.3382 | 0.1995 |
65
+ | 27.4549 | 0.21 | 6500 | 24.1257 | 0.1697 |
66
+ | 27.9176 | 0.22 | 7000 | 24.8758 | 0.1945 |
67
+ | 27.4036 | 0.24 | 7500 | 24.1006 | 0.1746 |
68
+ | 26.5633 | 0.26 | 8000 | 23.0034 | 0.1582 |
69
+ | 26.3558 | 0.27 | 8500 | 24.7499 | 0.1913 |
70
+ | 25.9604 | 0.29 | 9000 | 22.5813 | 0.1674 |
71
+ | 25.6154 | 0.31 | 9500 | 22.4642 | 0.1499 |
72
+ | 25.6231 | 0.32 | 10000 | 21.8089 | 0.1534 |
73
+ | 26.7554 | 0.34 | 10500 | 21.9619 | 0.1543 |
74
+ | 25.2901 | 0.35 | 11000 | 22.0643 | 0.1593 |
75
+ | 24.8642 | 0.37 | 11500 | 21.1113 | 0.1480 |
76
+ | 25.4664 | 0.39 | 12000 | 21.2492 | 0.1458 |
77
+ | 24.6433 | 0.4 | 12500 | 20.7650 | 0.1419 |
78
+ | 24.8455 | 0.42 | 13000 | 21.8535 | 0.1490 |
79
+ | 25.1176 | 0.43 | 13500 | 20.7491 | 0.1429 |
80
+ | 24.4585 | 0.45 | 14000 | 20.7948 | 0.1423 |
81
+ | 24.1613 | 0.47 | 14500 | 20.5817 | 0.1431 |
82
+ | 23.7281 | 0.48 | 15000 | 20.1209 | 0.1333 |
83
+ | 23.0396 | 0.5 | 15500 | 20.2883 | 0.1383 |
84
+ | 24.7056 | 0.51 | 16000 | 19.6813 | 0.1330 |
85
+ | 23.608 | 0.53 | 16500 | 20.0252 | 0.1394 |
86
+ | 23.9536 | 0.55 | 17000 | 19.9039 | 0.1341 |
87
+ | 23.1848 | 0.56 | 17500 | 19.9114 | 0.1308 |
88
+ | 23.1835 | 0.58 | 18000 | 19.7044 | 0.1345 |
89
+ | 23.9372 | 0.59 | 18500 | 19.2201 | 0.1296 |
90
+ | 23.2182 | 0.61 | 19000 | 19.3723 | 0.1350 |
91
+ | 22.3118 | 0.63 | 19500 | 19.2624 | 0.1344 |
92
+ | 22.9372 | 0.64 | 20000 | 19.5823 | 0.1387 |
93
+ | 23.1536 | 0.66 | 20500 | 18.9077 | 0.1289 |
94
+ | 22.3477 | 0.67 | 21000 | 18.7098 | 0.1257 |
95
+ | 22.3701 | 0.69 | 21500 | 19.0815 | 0.1300 |
96
+ | 22.6709 | 0.71 | 22000 | 18.4433 | 0.1242 |
97
+ | 22.2519 | 0.72 | 22500 | 18.7482 | 0.1275 |
98
+ | 21.8536 | 0.74 | 23000 | 18.6565 | 0.1236 |
99
+ | 22.4479 | 0.76 | 23500 | 18.6478 | 0.1264 |
100
+ | 21.6824 | 0.77 | 24000 | 18.4383 | 0.1257 |
101
+ | 22.1622 | 0.79 | 24500 | 18.4086 | 0.1212 |
102
+ | 22.2626 | 0.8 | 25000 | 18.4613 | 0.1230 |
103
+ | 21.0009 | 0.82 | 25500 | 18.1851 | 0.1165 |
104
+ | 20.554 | 0.84 | 26000 | 17.7352 | 0.1165 |
105
+ | 21.5141 | 0.85 | 26500 | 18.3084 | 0.1207 |
106
+ | 20.5925 | 0.87 | 27000 | 17.9997 | 0.1207 |
107
+ | 21.0997 | 0.88 | 27500 | 17.7534 | 0.1193 |
108
+ | 21.7098 | 0.9 | 28000 | 17.8348 | 0.1186 |
109
+
110
+
111
+ ### Framework versions
112
+
113
+ - Transformers 4.16.2
114
+ - Pytorch 1.10.2+cu102
115
+ - Datasets 1.18.3
116
+ - Tokenizers 0.11.0