update model card README.md
Browse files
README.md
ADDED
@@ -0,0 +1,97 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: apache-2.0
|
3 |
+
tags:
|
4 |
+
- generated_from_trainer
|
5 |
+
datasets:
|
6 |
+
- fleurs
|
7 |
+
metrics:
|
8 |
+
- wer
|
9 |
+
model-index:
|
10 |
+
- name: wav2vec2-large-xls-r-300m-kr-jw4169
|
11 |
+
results:
|
12 |
+
- task:
|
13 |
+
name: Automatic Speech Recognition
|
14 |
+
type: automatic-speech-recognition
|
15 |
+
dataset:
|
16 |
+
name: fleurs
|
17 |
+
type: fleurs
|
18 |
+
config: ko_kr
|
19 |
+
split: train
|
20 |
+
args: ko_kr
|
21 |
+
metrics:
|
22 |
+
- name: Wer
|
23 |
+
type: wer
|
24 |
+
value: 0.519593179778642
|
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 |
+
# wav2vec2-large-xls-r-300m-kr-jw4169
|
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 fleurs dataset.
|
33 |
+
It achieves the following results on the evaluation set:
|
34 |
+
- Loss: 0.9752
|
35 |
+
- Wer: 0.5196
|
36 |
+
|
37 |
+
## Model description
|
38 |
+
|
39 |
+
More information needed
|
40 |
+
|
41 |
+
## Intended uses & limitations
|
42 |
+
|
43 |
+
More information needed
|
44 |
+
|
45 |
+
## Training and evaluation data
|
46 |
+
|
47 |
+
More information needed
|
48 |
+
|
49 |
+
## Training procedure
|
50 |
+
|
51 |
+
### Training hyperparameters
|
52 |
+
|
53 |
+
The following hyperparameters were used during training:
|
54 |
+
- learning_rate: 0.0003
|
55 |
+
- train_batch_size: 4
|
56 |
+
- eval_batch_size: 8
|
57 |
+
- seed: 42
|
58 |
+
- gradient_accumulation_steps: 4
|
59 |
+
- total_train_batch_size: 16
|
60 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
61 |
+
- lr_scheduler_type: linear
|
62 |
+
- lr_scheduler_warmup_steps: 500
|
63 |
+
- num_epochs: 30
|
64 |
+
|
65 |
+
### Training results
|
66 |
+
|
67 |
+
| Training Loss | Epoch | Step | Validation Loss | Wer |
|
68 |
+
|:-------------:|:-----:|:----:|:---------------:|:------:|
|
69 |
+
| 35.084 | 1.39 | 200 | 6.8536 | 1.0 |
|
70 |
+
| 4.853 | 2.78 | 400 | 4.6246 | 1.0 |
|
71 |
+
| 4.5491 | 4.17 | 600 | 4.3815 | 1.0 |
|
72 |
+
| 2.799 | 5.55 | 800 | 1.7402 | 0.8642 |
|
73 |
+
| 1.3872 | 6.94 | 1000 | 1.2019 | 0.7448 |
|
74 |
+
| 0.9599 | 8.33 | 1200 | 1.0594 | 0.7134 |
|
75 |
+
| 0.675 | 9.72 | 1400 | 0.9321 | 0.6404 |
|
76 |
+
| 0.4775 | 11.11 | 1600 | 0.9088 | 0.5911 |
|
77 |
+
| 0.3479 | 12.5 | 1800 | 0.9430 | 0.6010 |
|
78 |
+
| 0.2712 | 13.89 | 2000 | 0.8948 | 0.5854 |
|
79 |
+
| 0.2283 | 15.28 | 2200 | 0.9009 | 0.5495 |
|
80 |
+
| 0.1825 | 16.67 | 2400 | 0.9079 | 0.5501 |
|
81 |
+
| 0.161 | 18.06 | 2600 | 0.9518 | 0.5390 |
|
82 |
+
| 0.1394 | 19.44 | 2800 | 0.9529 | 0.5399 |
|
83 |
+
| 0.1266 | 20.83 | 3000 | 0.9505 | 0.5283 |
|
84 |
+
| 0.1102 | 22.22 | 3200 | 0.9748 | 0.5328 |
|
85 |
+
| 0.101 | 23.61 | 3400 | 0.9593 | 0.5316 |
|
86 |
+
| 0.0907 | 25.0 | 3600 | 0.9832 | 0.5292 |
|
87 |
+
| 0.0833 | 26.39 | 3800 | 0.9773 | 0.5181 |
|
88 |
+
| 0.0781 | 27.78 | 4000 | 0.9736 | 0.5163 |
|
89 |
+
| 0.0744 | 29.17 | 4200 | 0.9752 | 0.5196 |
|
90 |
+
|
91 |
+
|
92 |
+
### Framework versions
|
93 |
+
|
94 |
+
- Transformers 4.25.0.dev0
|
95 |
+
- Pytorch 1.10.0+cu102
|
96 |
+
- Datasets 2.6.1
|
97 |
+
- Tokenizers 0.13.1
|