update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,93 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
model-index:
|
6 |
+
- name: wav2vec2-demo-F01-2
|
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-demo-F01-2
|
14 |
+
|
15 |
+
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the None dataset.
|
16 |
+
It achieves the following results on the evaluation set:
|
17 |
+
- Loss: 1.6361
|
18 |
+
- Wer: 0.9025
|
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: 8
|
39 |
+
- eval_batch_size: 8
|
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: 1000
|
44 |
+
- num_epochs: 30
|
45 |
+
|
46 |
+
### Training results
|
47 |
+
|
48 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
49 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
50 |
+
| 23.7408 | 0.81 | 500 | 3.3782 | 1.0 |
|
51 |
+
| 3.348 | 1.62 | 1000 | 2.9501 | 1.0 |
|
52 |
+
| 2.8539 | 2.44 | 1500 | 2.6975 | 1.0 |
|
53 |
+
| 2.311 | 3.25 | 2000 | 1.7770 | 1.3175 |
|
54 |
+
| 1.5102 | 4.06 | 2500 | 1.3481 | 1.3515 |
|
55 |
+
| 1.0616 | 4.87 | 3000 | 1.4306 | 1.2313 |
|
56 |
+
| 0.8493 | 5.68 | 3500 | 1.2261 | 1.1701 |
|
57 |
+
| 0.7058 | 6.49 | 4000 | 1.2132 | 1.1111 |
|
58 |
+
| 0.6129 | 7.31 | 4500 | 1.4230 | 1.1429 |
|
59 |
+
| 0.5513 | 8.12 | 5000 | 1.2003 | 1.0499 |
|
60 |
+
| 0.4957 | 8.93 | 5500 | 1.5534 | 1.1043 |
|
61 |
+
| 0.4456 | 9.74 | 6000 | 1.2315 | 1.0658 |
|
62 |
+
| 0.4101 | 10.55 | 6500 | 1.1621 | 1.0680 |
|
63 |
+
| 0.3776 | 11.36 | 7000 | 1.4302 | 1.0385 |
|
64 |
+
| 0.3318 | 12.18 | 7500 | 1.3488 | 0.9977 |
|
65 |
+
| 0.3189 | 12.99 | 8000 | 1.4050 | 1.0295 |
|
66 |
+
| 0.3103 | 13.8 | 8500 | 1.4535 | 1.0385 |
|
67 |
+
| 0.2791 | 14.61 | 9000 | 1.3318 | 1.0181 |
|
68 |
+
| 0.2681 | 15.42 | 9500 | 1.5199 | 0.9909 |
|
69 |
+
| 0.2352 | 16.23 | 10000 | 1.5019 | 1.0023 |
|
70 |
+
| 0.235 | 17.05 | 10500 | 1.7984 | 0.9955 |
|
71 |
+
| 0.2319 | 17.86 | 11000 | 1.3399 | 0.9705 |
|
72 |
+
| 0.221 | 18.67 | 11500 | 1.8316 | 0.9342 |
|
73 |
+
| 0.2154 | 19.48 | 12000 | 1.6837 | 0.9637 |
|
74 |
+
| 0.1911 | 20.29 | 12500 | 1.6999 | 0.9388 |
|
75 |
+
| 0.1754 | 21.1 | 13000 | 1.4801 | 0.9274 |
|
76 |
+
| 0.1776 | 21.92 | 13500 | 1.7954 | 0.9206 |
|
77 |
+
| 0.1616 | 22.73 | 14000 | 1.7891 | 0.9320 |
|
78 |
+
| 0.1579 | 23.54 | 14500 | 1.5692 | 0.9116 |
|
79 |
+
| 0.173 | 24.35 | 15000 | 1.4928 | 0.9048 |
|
80 |
+
| 0.1561 | 25.16 | 15500 | 1.6492 | 0.9116 |
|
81 |
+
| 0.1542 | 25.97 | 16000 | 1.7356 | 0.9048 |
|
82 |
+
| 0.131 | 26.79 | 16500 | 1.7785 | 0.9048 |
|
83 |
+
| 0.1295 | 27.6 | 17000 | 1.6532 | 0.9116 |
|
84 |
+
| 0.1374 | 28.41 | 17500 | 1.6760 | 0.9093 |
|
85 |
+
| 0.1186 | 29.22 | 18000 | 1.6361 | 0.9025 |
|
86 |
+
|
87 |
+
|
88 |
+
### Framework versions
|
89 |
+
|
90 |
+
- Transformers 4.23.1
|
91 |
+
- Pytorch 1.12.1+cu113
|
92 |
+
- Datasets 1.18.3
|
93 |
+
- Tokenizers 0.13.2
|