Commit
•
1b18626
1
Parent(s):
5f3f2a5
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,87 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
model-index:
|
6 |
+
- name: wav2vec2-base-timit-demo-google-colab
|
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-base-timit-demo-google-colab
|
14 |
+
|
15 |
+
This model is a fine-tuned version of [facebook/wav2vec2-base](https://huggingface.co/facebook/wav2vec2-base) on the None dataset.
|
16 |
+
It achieves the following results on the evaluation set:
|
17 |
+
- Loss: 0.5436
|
18 |
+
- Wer: 0.3401
|
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 |
+
- mixed_precision_training: Native AMP
|
46 |
+
|
47 |
+
### Training results
|
48 |
+
|
49 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
50 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|
|
51 |
+
| 3.5276 | 1.0 | 500 | 1.9983 | 1.0066 |
|
52 |
+
| 0.8606 | 2.01 | 1000 | 0.5323 | 0.5220 |
|
53 |
+
| 0.4339 | 3.01 | 1500 | 0.4697 | 0.4512 |
|
54 |
+
| 0.3026 | 4.02 | 2000 | 0.4342 | 0.4266 |
|
55 |
+
| 0.2297 | 5.02 | 2500 | 0.5001 | 0.4135 |
|
56 |
+
| 0.1939 | 6.02 | 3000 | 0.4350 | 0.3897 |
|
57 |
+
| 0.1613 | 7.03 | 3500 | 0.4740 | 0.3883 |
|
58 |
+
| 0.1452 | 8.03 | 4000 | 0.4289 | 0.3825 |
|
59 |
+
| 0.1362 | 9.04 | 4500 | 0.4721 | 0.3927 |
|
60 |
+
| 0.1146 | 10.04 | 5000 | 0.4707 | 0.3730 |
|
61 |
+
| 0.1061 | 11.04 | 5500 | 0.4470 | 0.3701 |
|
62 |
+
| 0.0947 | 12.05 | 6000 | 0.4694 | 0.3722 |
|
63 |
+
| 0.0852 | 13.05 | 6500 | 0.5222 | 0.3733 |
|
64 |
+
| 0.0741 | 14.06 | 7000 | 0.4881 | 0.3657 |
|
65 |
+
| 0.069 | 15.06 | 7500 | 0.4957 | 0.3677 |
|
66 |
+
| 0.0679 | 16.06 | 8000 | 0.5241 | 0.3634 |
|
67 |
+
| 0.0618 | 17.07 | 8500 | 0.5091 | 0.3564 |
|
68 |
+
| 0.0576 | 18.07 | 9000 | 0.5055 | 0.3557 |
|
69 |
+
| 0.0493 | 19.08 | 9500 | 0.5013 | 0.3515 |
|
70 |
+
| 0.0469 | 20.08 | 10000 | 0.5506 | 0.3530 |
|
71 |
+
| 0.044 | 21.08 | 10500 | 0.5564 | 0.3528 |
|
72 |
+
| 0.0368 | 22.09 | 11000 | 0.5213 | 0.3509 |
|
73 |
+
| 0.0355 | 23.09 | 11500 | 0.5707 | 0.3495 |
|
74 |
+
| 0.0357 | 24.1 | 12000 | 0.5558 | 0.3483 |
|
75 |
+
| 0.0285 | 25.1 | 12500 | 0.5613 | 0.3455 |
|
76 |
+
| 0.0285 | 26.1 | 13000 | 0.5533 | 0.3480 |
|
77 |
+
| 0.0266 | 27.11 | 13500 | 0.5526 | 0.3462 |
|
78 |
+
| 0.0249 | 28.11 | 14000 | 0.5488 | 0.3429 |
|
79 |
+
| 0.0237 | 29.12 | 14500 | 0.5436 | 0.3401 |
|
80 |
+
|
81 |
+
|
82 |
+
### Framework versions
|
83 |
+
|
84 |
+
- Transformers 4.17.0
|
85 |
+
- Pytorch 1.11.0+cu115
|
86 |
+
- Datasets 1.18.3
|
87 |
+
- Tokenizers 0.12.1
|