jonatasgrosman
commited on
Commit
•
f333d38
1
Parent(s):
8539b2e
Update README.md
Browse files
README.md
CHANGED
@@ -73,15 +73,15 @@ The script used for training can be found here: https://github.com/jonatasgrosma
|
|
73 |
|
74 |
The model can be used directly (without a language model) as follows...
|
75 |
|
76 |
-
Using the [
|
77 |
|
78 |
```python
|
79 |
-
from
|
80 |
-
|
81 |
-
asr = ASREngine("fr", model_path="jonatasgrosman/wav2vec2-large-xlsr-53-french")
|
82 |
|
|
|
83 |
audio_paths = ["/path/to/file.mp3", "/path/to/another_file.wav"]
|
84 |
-
|
|
|
85 |
```
|
86 |
|
87 |
Writing your own inference script:
|
|
|
73 |
|
74 |
The model can be used directly (without a language model) as follows...
|
75 |
|
76 |
+
Using the [HuggingSound](https://github.com/jonatasgrosman/huggingsound) library:
|
77 |
|
78 |
```python
|
79 |
+
from huggingsound import SpeechRecognitionModel
|
|
|
|
|
80 |
|
81 |
+
model = SpeechRecognitionModel("jonatasgrosman/wav2vec2-large-xlsr-53-french")
|
82 |
audio_paths = ["/path/to/file.mp3", "/path/to/another_file.wav"]
|
83 |
+
|
84 |
+
transcriptions = model.transcribe(audio_paths)
|
85 |
```
|
86 |
|
87 |
Writing your own inference script:
|