Commit
•
d5838ae
1
Parent(s):
feda12e
Update README.md
Browse files
README.md
CHANGED
|
@@ -56,19 +56,14 @@ It achieves the following results on the evaluation set:
|
|
| 56 |
- Loss: 1.4638
|
| 57 |
- Wer: 0.5602
|
| 58 |
|
| 59 |
-
|
| 60 |
|
| 61 |
-
|
| 62 |
|
| 63 |
-
|
| 64 |
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
## Training and evaluation data
|
| 68 |
-
|
| 69 |
-
More information needed
|
| 70 |
-
|
| 71 |
-
## Training procedure
|
| 72 |
|
| 73 |
### Training hyperparameters
|
| 74 |
|
|
|
|
| 56 |
- Loss: 1.4638
|
| 57 |
- Wer: 0.5602
|
| 58 |
|
| 59 |
+
### Evaluation Commands
|
| 60 |
|
| 61 |
+
1. To evaluate on mozilla-foundation/common_voice_8_0 with test split
|
| 62 |
|
| 63 |
+
python eval.py --model_id DrishtiSharma/wav2vec2-large-xls-r-300m-or-dx12 --dataset mozilla-foundation/common_voice_8_0 --config or --split test --log_outputs
|
| 64 |
|
| 65 |
+
2. To evaluate on speech-recognition-community-v2/dev_data
|
| 66 |
+
Oriya language isn't available in speech-recognition-community-v2/dev_data
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 67 |
|
| 68 |
### Training hyperparameters
|
| 69 |
|