Update README.md
Browse files
README.md
CHANGED
@@ -71,3 +71,14 @@ manually evaluate WER on test set - vietnamese part:
|
|
71 |
| this model | 26.6% | 37.1% | 18.7% |
|
72 |
|
73 |
all training + evaluation scripts are on my repo: https://github.com/phineas-pta/fine-tune-whisper-vi
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
71 |
| this model | 26.6% | 37.1% | 18.7% |
|
72 |
|
73 |
all training + evaluation scripts are on my repo: https://github.com/phineas-pta/fine-tune-whisper-vi
|
74 |
+
|
75 |
+
usage example:
|
76 |
+
```python
|
77 |
+
import torch
|
78 |
+
from transformers import pipeline
|
79 |
+
|
80 |
+
PIPE = pipeline(task="automatic-speech-recognition", model="doof-ferb/whisper-tiny-vi", device="cuda:0", torch_dtype=torch.float16)
|
81 |
+
PIPE_KWARGS = {"language": "vi", "task": "transcribe"}
|
82 |
+
|
83 |
+
PIPE("audio.mp3", generate_kwargs=PIPE_KWARGS)["text"]
|
84 |
+
```
|