Update README.md
Browse files
README.md
CHANGED
@@ -1,3 +1,39 @@
|
|
1 |
---
|
2 |
license: mit
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
license: mit
|
3 |
---
|
4 |
+
|
5 |
+
---
|
6 |
+
language: fr
|
7 |
+
datasets:
|
8 |
+
- common_voice
|
9 |
+
metrics:
|
10 |
+
- per
|
11 |
+
tags:
|
12 |
+
- audio
|
13 |
+
- automatic-speech-recognition
|
14 |
+
- speech
|
15 |
+
model-index:
|
16 |
+
- name: Wav2Vec2-base French finetuned for phonemes by LMSSC
|
17 |
+
results:
|
18 |
+
- task:
|
19 |
+
name: Speech Recognition
|
20 |
+
type: automatic-speech-recognition
|
21 |
+
dataset:
|
22 |
+
name: Common Voice fr
|
23 |
+
type: common_voice
|
24 |
+
args: fr
|
25 |
+
metrics:
|
26 |
+
- name: Test PER
|
27 |
+
type: per
|
28 |
+
value: 5.52
|
29 |
+
- name: Val PER
|
30 |
+
type: per
|
31 |
+
value: 4.31
|
32 |
+
|
33 |
+
---
|
34 |
+
|
35 |
+
# Fine-tuned French Voxpopuli v2 wav2vec2-base model for speech-to-phoneme task in French
|
36 |
+
|
37 |
+
Fine-tuned [facebook/wav2vec2-base-fr-voxpopuli-v2](https://huggingface.co/facebook/wav2vec2-base-fr-voxpopuli-v2) on French using the train and validation splits of [Common Voice v13](https://huggingface.co/datasets/mozilla-foundation/common_voice_13_0).
|
38 |
+
When using this model, make sure that your speech input is sampled at 16kHz.
|
39 |
+
|