manifoldix
commited on
Commit
•
c6c8ad5
1
Parent(s):
89e6ca8
readme
Browse files
README.md
CHANGED
@@ -1,8 +1,32 @@
|
|
1 |
-
Trained on common voice FA (Persian).
|
2 |
-
On the test set, the WER is 26% without LM and 23% with LM.
|
3 |
-
|
4 |
-
|
5 |
---
|
6 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7 |
---
|
8 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language: fa
|
3 |
+
datasets:
|
4 |
+
- common_voice
|
5 |
+
tags:
|
6 |
+
- robust-speech-event
|
7 |
+
widget:
|
8 |
+
- example_title: Common Voice sample 2978
|
9 |
+
src: https://huggingface.co/manifoldix/xlsr-fa-lm/resolve/main/sample2978.flac
|
10 |
+
- example_title: Common Voice sample 5168
|
11 |
+
src: https://huggingface.co/manifoldix/xlsr-fa-lm/resolve/main/sample5168.flac
|
12 |
+
model-index:
|
13 |
+
- name: XLS-R-300m Wav2Vec2 Persian
|
14 |
+
results:
|
15 |
+
- task:
|
16 |
+
name: Speech Recognition
|
17 |
+
type: automatic-speech-recognition
|
18 |
+
dataset:
|
19 |
+
name: Common Voice fa
|
20 |
+
type: common_voice
|
21 |
+
args: fa
|
22 |
+
metrics:
|
23 |
+
- name: Test WER without LM
|
24 |
+
type: wer
|
25 |
+
value: 26%
|
26 |
+
- name: Test WER with LM
|
27 |
+
type: wer
|
28 |
+
value: 23%
|
29 |
---
|
30 |
|
31 |
+
## XLSR-300m Persian
|
32 |
+
Fine-tuned on commom voice FA
|