Instructions to use Ansu/mHubert-basque-ASR with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Ansu/mHubert-basque-ASR with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("automatic-speech-recognition", model="Ansu/mHubert-basque-ASR")# Load model directly from transformers import AutoProcessor, AutoModelForCTC processor = AutoProcessor.from_pretrained("Ansu/mHubert-basque-ASR") model = AutoModelForCTC.from_pretrained("Ansu/mHubert-basque-ASR") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -47,14 +47,6 @@ The following hyperparameters were used during training:
|
|
| 47 |
- Datasets 3.3.2
|
| 48 |
- Tokenizers 0.21.0
|
| 49 |
|
| 50 |
-
## Test results
|
| 51 |
-
|
| 52 |
-
Map: 100%|██████████| 16359/16359 [09:32<00:00, 28.58 examples/s]
|
| 53 |
-
|
| 54 |
-
Test WER: 0.137
|
| 55 |
-
|
| 56 |
-
Test CER: 0.024
|
| 57 |
-
|
| 58 |
### Sample predictions:
|
| 59 |
|
| 60 |
Test CV WER: 0.074
|
|
|
|
| 47 |
- Datasets 3.3.2
|
| 48 |
- Tokenizers 0.21.0
|
| 49 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 50 |
### Sample predictions:
|
| 51 |
|
| 52 |
Test CV WER: 0.074
|