Commit
•
24e4a5c
1
Parent(s):
b2ba397
update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,102 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- common_voice_11_0
|
7 |
+
metrics:
|
8 |
+
- wer
|
9 |
+
model-index:
|
10 |
+
- name: firstcolab3
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Automatic Speech Recognition
|
14 |
+
type: automatic-speech-recognition
|
15 |
+
dataset:
|
16 |
+
name: common_voice_11_0
|
17 |
+
type: common_voice_11_0
|
18 |
+
config: th
|
19 |
+
split: train+validation
|
20 |
+
args: th
|
21 |
+
metrics:
|
22 |
+
- name: Wer
|
23 |
+
type: wer
|
24 |
+
value: 0.6226224783861671
|
25 |
+
---
|
26 |
+
|
27 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
28 |
+
should probably proofread and complete it, then remove this comment. -->
|
29 |
+
|
30 |
+
# firstcolab3
|
31 |
+
|
32 |
+
This model is a fine-tuned version of [facebook/wav2vec2-xls-r-300m](https://huggingface.co/facebook/wav2vec2-xls-r-300m) on the common_voice_11_0 dataset.
|
33 |
+
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 0.2756
|
35 |
+
- Wer: 0.6226
|
36 |
+
- Cer: 0.0535
|
37 |
+
|
38 |
+
## Model description
|
39 |
+
|
40 |
+
More information needed
|
41 |
+
|
42 |
+
## Intended uses & limitations
|
43 |
+
|
44 |
+
More information needed
|
45 |
+
|
46 |
+
## Training and evaluation data
|
47 |
+
|
48 |
+
More information needed
|
49 |
+
|
50 |
+
## Training procedure
|
51 |
+
|
52 |
+
### Training hyperparameters
|
53 |
+
|
54 |
+
The following hyperparameters were used during training:
|
55 |
+
- learning_rate: 0.0001
|
56 |
+
- train_batch_size: 32
|
57 |
+
- eval_batch_size: 16
|
58 |
+
- seed: 42
|
59 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
60 |
+
- lr_scheduler_type: linear
|
61 |
+
- lr_scheduler_warmup_steps: 500
|
62 |
+
- num_epochs: 20
|
63 |
+
- mixed_precision_training: Native AMP
|
64 |
+
|
65 |
+
### Training results
|
66 |
+
|
67 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer | Cer |
|
68 |
+
|:-------------:|:-----:|:-----:|:---------------:|:------:|:------:|
|
69 |
+
| 7.187 | 0.75 | 1000 | 3.7705 | 1.0 | 1.0 |
|
70 |
+
| 2.0277 | 1.5 | 2000 | 0.6139 | 0.9202 | 0.1545 |
|
71 |
+
| 0.8368 | 2.24 | 3000 | 0.4351 | 0.8589 | 0.1147 |
|
72 |
+
| 0.6772 | 2.99 | 4000 | 0.3762 | 0.8200 | 0.0990 |
|
73 |
+
| 0.5702 | 3.74 | 5000 | 0.3434 | 0.7889 | 0.0891 |
|
74 |
+
| 0.5205 | 4.49 | 6000 | 0.3427 | 0.7726 | 0.0855 |
|
75 |
+
| 0.4773 | 5.24 | 7000 | 0.3073 | 0.7408 | 0.0767 |
|
76 |
+
| 0.4389 | 5.98 | 8000 | 0.2969 | 0.7421 | 0.0759 |
|
77 |
+
| 0.4069 | 6.73 | 9000 | 0.2884 | 0.7134 | 0.0711 |
|
78 |
+
| 0.3858 | 7.48 | 10000 | 0.2952 | 0.7066 | 0.0699 |
|
79 |
+
| 0.36 | 8.23 | 11000 | 0.2846 | 0.6902 | 0.0662 |
|
80 |
+
| 0.3517 | 8.98 | 12000 | 0.2729 | 0.6756 | 0.0638 |
|
81 |
+
| 0.3265 | 9.72 | 13000 | 0.2844 | 0.6756 | 0.0645 |
|
82 |
+
| 0.3127 | 10.47 | 14000 | 0.2769 | 0.6803 | 0.0640 |
|
83 |
+
| 0.3016 | 11.22 | 15000 | 0.2772 | 0.6566 | 0.0618 |
|
84 |
+
| 0.2855 | 11.97 | 16000 | 0.2791 | 0.6540 | 0.0598 |
|
85 |
+
| 0.2699 | 12.72 | 17000 | 0.2714 | 0.6455 | 0.0589 |
|
86 |
+
| 0.264 | 13.46 | 18000 | 0.2782 | 0.6472 | 0.0588 |
|
87 |
+
| 0.2518 | 14.21 | 19000 | 0.2693 | 0.6398 | 0.0578 |
|
88 |
+
| 0.2498 | 14.96 | 20000 | 0.2761 | 0.6300 | 0.0561 |
|
89 |
+
| 0.2426 | 15.71 | 21000 | 0.2796 | 0.6366 | 0.0561 |
|
90 |
+
| 0.2271 | 16.45 | 22000 | 0.2804 | 0.6336 | 0.0554 |
|
91 |
+
| 0.2271 | 17.2 | 23000 | 0.2758 | 0.6347 | 0.0552 |
|
92 |
+
| 0.22 | 17.95 | 24000 | 0.2785 | 0.6279 | 0.0544 |
|
93 |
+
| 0.2143 | 18.7 | 25000 | 0.2783 | 0.6246 | 0.0538 |
|
94 |
+
| 0.2134 | 19.45 | 26000 | 0.2756 | 0.6226 | 0.0535 |
|
95 |
+
|
96 |
+
|
97 |
+
### Framework versions
|
98 |
+
|
99 |
+
- Transformers 4.24.0
|
100 |
+
- Pytorch 1.12.1+cu113
|
101 |
+
- Datasets 2.7.1
|
102 |
+
- Tokenizers 0.13.2
|