Commit
•
96c1c53
1
Parent(s):
18c7dbf
Update Readme
Browse files
README.md
CHANGED
@@ -1,38 +1,54 @@
|
|
1 |
---
|
2 |
-
language:
|
3 |
-
- ja
|
4 |
license: apache-2.0
|
|
|
|
|
5 |
tags:
|
6 |
- automatic-speech-recognition
|
7 |
-
-
|
8 |
-
-
|
|
|
9 |
model-index:
|
10 |
- name: wav2vec2-large-xlsr-53-ja
|
11 |
-
results:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
12 |
---
|
13 |
|
14 |
-
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
15 |
-
should probably proofread and complete it, then remove this comment. -->
|
16 |
-
|
17 |
# wav2vec2-large-xlsr-53-ja
|
18 |
|
19 |
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - JA dataset.
|
20 |
-
It achieves the following results on the evaluation set:
|
21 |
-
- Loss: 0.4214
|
22 |
-
- Wer: 0.3375
|
23 |
-
- Cer: 0.1587
|
24 |
-
|
25 |
-
## Model description
|
26 |
|
27 |
-
|
|
|
|
|
|
|
|
|
28 |
|
29 |
-
|
|
|
|
|
|
|
|
|
30 |
|
31 |
-
|
32 |
|
33 |
-
|
34 |
|
35 |
-
|
|
|
|
|
36 |
|
37 |
## Training procedure
|
38 |
|
|
|
1 |
---
|
|
|
|
|
2 |
license: apache-2.0
|
3 |
+
language:
|
4 |
+
- vi
|
5 |
tags:
|
6 |
- automatic-speech-recognition
|
7 |
+
- robust-speech-event
|
8 |
+
- common-voice
|
9 |
+
- vi
|
10 |
model-index:
|
11 |
- name: wav2vec2-large-xlsr-53-ja
|
12 |
+
results:
|
13 |
+
- task:
|
14 |
+
name: Speech Recognition
|
15 |
+
type: automatic-speech-recognition
|
16 |
+
dataset:
|
17 |
+
name: Common Voice 8.0
|
18 |
+
type: mozilla-foundation/common_voice_8_0
|
19 |
+
args: ja
|
20 |
+
metrics:
|
21 |
+
- name: Test WER (with LM)
|
22 |
+
type: wer
|
23 |
+
value: 16.08
|
24 |
+
- name: Test CER (with LM)
|
25 |
+
type: cer
|
26 |
+
value: 7.15
|
27 |
---
|
28 |
|
|
|
|
|
|
|
29 |
# wav2vec2-large-xlsr-53-ja
|
30 |
|
31 |
This model is a fine-tuned version of [facebook/wav2vec2-large-xlsr-53](https://huggingface.co/facebook/wav2vec2-large-xlsr-53) on the MOZILLA-FOUNDATION/COMMON_VOICE_8_0 - JA dataset.
|
|
|
|
|
|
|
|
|
|
|
|
|
32 |
|
33 |
+
### Benchmark WER result:
|
34 |
+
| | [COMMON VOICE 7.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0) | [COMMON VOICE 8.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0)
|
35 |
+
|---|---|---|---|
|
36 |
+
|without LM| 15.74 | 25.10 |
|
37 |
+
|with 4-grams LM| 15.37 | 16.09 |
|
38 |
|
39 |
+
### Benchmark CER result:
|
40 |
+
| | [COMMON VOICE 7.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_7_0) | [COMMON VOICE 8.0](https://huggingface.co/datasets/mozilla-foundation/common_voice_8_0)
|
41 |
+
|---|---|---|---|
|
42 |
+
|without LM| 9.51 | 9.95 |
|
43 |
+
|with 4-grams LM| 6.91 | 7.15 |
|
44 |
|
45 |
+
## Evaluation
|
46 |
|
47 |
+
Please use the eval.py file to run the evaluation:
|
48 |
|
49 |
+
```python
|
50 |
+
python eval.py --model_id vutankiet2901/wav2vec2-large-xlsr-53-ja --dataset mozilla-foundation/common_voice_7_0 --config ja --split test --log_outputs
|
51 |
+
```
|
52 |
|
53 |
## Training procedure
|
54 |
|