Update README.md
Browse files
README.md
CHANGED
@@ -17,20 +17,20 @@ model-index:
|
|
17 |
metrics:
|
18 |
- name: Test WER
|
19 |
type: wer
|
20 |
-
value: 0.
|
21 |
---
|
22 |
# Wav2Vec2-Large-XLSR-53-Moroccan-Darija
|
23 |
|
24 |
**wav2vec2-large-xlsr-53 new model**
|
25 |
|
26 |
-
- Fine-tuned on
|
27 |
- Fine-tuning is ongoing 24/7 to enhance accuracy.
|
28 |
- We are consistently adding data to the model every day (We prefer not to add all MDVC Corpus at once as we are trying to standardize more and more the way we write this language).
|
29 |
|
30 |
<table><thead><tr><th><strong>Training Loss</strong></th> <th><strong>Validation</strong></th> <th><strong>Loss Wer</strong></th></tr></thead> <tbody><tr>
|
31 |
-
<td>0.
|
32 |
-
<td>0.
|
33 |
-
<td>0.
|
34 |
</tr> </tbody></table>
|
35 |
|
36 |
## Usage
|
|
|
17 |
metrics:
|
18 |
- name: Test WER
|
19 |
type: wer
|
20 |
+
value: 0.146602
|
21 |
---
|
22 |
# Wav2Vec2-Large-XLSR-53-Moroccan-Darija
|
23 |
|
24 |
**wav2vec2-large-xlsr-53 new model**
|
25 |
|
26 |
+
- Fine-tuned on 39 hours of labeled Darija Audios extracted from MDVC corpus which contains more than 1000 hours of Moroccan Darija "ary".
|
27 |
- Fine-tuning is ongoing 24/7 to enhance accuracy.
|
28 |
- We are consistently adding data to the model every day (We prefer not to add all MDVC Corpus at once as we are trying to standardize more and more the way we write this language).
|
29 |
|
30 |
<table><thead><tr><th><strong>Training Loss</strong></th> <th><strong>Validation</strong></th> <th><strong>Loss Wer</strong></th></tr></thead> <tbody><tr>
|
31 |
+
<td>0.029600</td>
|
32 |
+
<td>0.213569</td>
|
33 |
+
<td>0.146602</td>
|
34 |
</tr> </tbody></table>
|
35 |
|
36 |
## Usage
|