update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,111 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
metrics:
|
6 |
+
- wer
|
7 |
+
model-index:
|
8 |
+
- name: wav2vec2-base-960h-paper
|
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-base-960h-paper
|
16 |
+
|
17 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base-960h](https://huggingface.co/facebook/wav2vec2-base-960h) on the None dataset.
|
18 |
+
It achieves the following results on the evaluation set:
|
19 |
+
- Loss: 0.8465
|
20 |
+
- Wer: 0.9629
|
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: 5e-05
|
40 |
+
- train_batch_size: 8
|
41 |
+
- eval_batch_size: 16
|
42 |
+
- seed: 42
|
43 |
+
- gradient_accumulation_steps: 4
|
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: 420
|
48 |
+
- num_epochs: 50.0
|
49 |
+
|
50 |
+
### Training results
|
51 |
+
|
52 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
53 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
54 |
+
| No log | 0.99 | 104 | 5.6840 | 1.0 |
|
55 |
+
| No log | 2.0 | 209 | 3.9772 | 1.0 |
|
56 |
+
| No log | 3.0 | 314 | 3.4204 | 1.0 |
|
57 |
+
| No log | 4.0 | 419 | 3.3692 | 1.0 |
|
58 |
+
| 5.5612 | 4.99 | 523 | 3.3945 | 1.0 |
|
59 |
+
| 5.5612 | 6.0 | 628 | 3.3426 | 1.0 |
|
60 |
+
| 5.5612 | 7.0 | 733 | 3.3333 | 1.0 |
|
61 |
+
| 5.5612 | 8.0 | 838 | 3.3296 | 1.0001 |
|
62 |
+
| 5.5612 | 8.99 | 942 | 3.1853 | 0.9999 |
|
63 |
+
| 3.2743 | 10.0 | 1047 | 2.1381 | 1.0245 |
|
64 |
+
| 3.2743 | 11.0 | 1152 | 1.6965 | 1.0142 |
|
65 |
+
| 3.2743 | 12.0 | 1257 | 1.4230 | 1.0011 |
|
66 |
+
| 3.2743 | 12.99 | 1361 | 1.2679 | 0.9873 |
|
67 |
+
| 3.2743 | 14.0 | 1466 | 1.1570 | 0.9836 |
|
68 |
+
| 1.5432 | 15.0 | 1571 | 1.0858 | 0.9784 |
|
69 |
+
| 1.5432 | 16.0 | 1676 | 1.0303 | 0.9769 |
|
70 |
+
| 1.5432 | 16.99 | 1780 | 0.9855 | 0.9746 |
|
71 |
+
| 1.5432 | 18.0 | 1885 | 0.9559 | 0.9709 |
|
72 |
+
| 1.5432 | 19.0 | 1990 | 0.9328 | 0.9728 |
|
73 |
+
| 0.902 | 20.0 | 2095 | 0.9166 | 0.9738 |
|
74 |
+
| 0.902 | 20.99 | 2199 | 0.8991 | 0.9698 |
|
75 |
+
| 0.902 | 22.0 | 2304 | 0.8717 | 0.9681 |
|
76 |
+
| 0.902 | 23.0 | 2409 | 0.8665 | 0.9669 |
|
77 |
+
| 0.7003 | 24.0 | 2514 | 0.8589 | 0.9670 |
|
78 |
+
| 0.7003 | 24.99 | 2618 | 0.8420 | 0.9659 |
|
79 |
+
| 0.7003 | 26.0 | 2723 | 0.8473 | 0.9661 |
|
80 |
+
| 0.7003 | 27.0 | 2828 | 0.8543 | 0.9666 |
|
81 |
+
| 0.7003 | 28.0 | 2933 | 0.8315 | 0.9623 |
|
82 |
+
| 0.5914 | 28.99 | 3037 | 0.8281 | 0.9626 |
|
83 |
+
| 0.5914 | 30.0 | 3142 | 0.8315 | 0.9625 |
|
84 |
+
| 0.5914 | 31.0 | 3247 | 0.8261 | 0.9620 |
|
85 |
+
| 0.5914 | 32.0 | 3352 | 0.8214 | 0.9640 |
|
86 |
+
| 0.5914 | 32.99 | 3456 | 0.8310 | 0.9634 |
|
87 |
+
| 0.5157 | 34.0 | 3561 | 0.8252 | 0.9635 |
|
88 |
+
| 0.5157 | 35.0 | 3666 | 0.8373 | 0.9638 |
|
89 |
+
| 0.5157 | 36.0 | 3771 | 0.8422 | 0.9629 |
|
90 |
+
| 0.5157 | 36.99 | 3875 | 0.8294 | 0.9632 |
|
91 |
+
| 0.5157 | 38.0 | 3980 | 0.8332 | 0.9576 |
|
92 |
+
| 0.4655 | 39.0 | 4085 | 0.8330 | 0.9595 |
|
93 |
+
| 0.4655 | 40.0 | 4190 | 0.8297 | 0.9625 |
|
94 |
+
| 0.4655 | 40.99 | 4294 | 0.8365 | 0.9621 |
|
95 |
+
| 0.4655 | 42.0 | 4399 | 0.8361 | 0.9621 |
|
96 |
+
| 0.4266 | 43.0 | 4504 | 0.8416 | 0.9625 |
|
97 |
+
| 0.4266 | 44.0 | 4609 | 0.8381 | 0.9634 |
|
98 |
+
| 0.4266 | 44.99 | 4713 | 0.8448 | 0.9645 |
|
99 |
+
| 0.4266 | 46.0 | 4818 | 0.8447 | 0.9625 |
|
100 |
+
| 0.4266 | 47.0 | 4923 | 0.8464 | 0.9641 |
|
101 |
+
| 0.4019 | 48.0 | 5028 | 0.8449 | 0.9628 |
|
102 |
+
| 0.4019 | 48.99 | 5132 | 0.8487 | 0.9626 |
|
103 |
+
| 0.4019 | 49.64 | 5200 | 0.8465 | 0.9629 |
|
104 |
+
|
105 |
+
|
106 |
+
### Framework versions
|
107 |
+
|
108 |
+
- Transformers 4.31.0.dev0
|
109 |
+
- Pytorch 2.0.1+cu118
|
110 |
+
- Datasets 2.7.0
|
111 |
+
- Tokenizers 0.13.3
|