Update README.md
Browse files
README.md
CHANGED
|
@@ -50,7 +50,7 @@ from transformers import pipeline
|
|
| 50 |
|
| 51 |
pipe = pipeline(
|
| 52 |
task="automatic-speech-recognition",
|
| 53 |
-
model="seanghay/whisper-small-khmer",
|
| 54 |
)
|
| 55 |
|
| 56 |
result = pipe("audio.wav",
|
|
@@ -61,14 +61,4 @@ result = pipe("audio.wav",
|
|
| 61 |
)
|
| 62 |
|
| 63 |
print(result["text"])
|
| 64 |
-
```
|
| 65 |
-
|
| 66 |
-
|
| 67 |
-
## whisper.cpp
|
| 68 |
-
|
| 69 |
-
|
| 70 |
-
### 1. Transcode the input audio to 16kHz PCM
|
| 71 |
-
|
| 72 |
-
```shell
|
| 73 |
-
ffmpeg -i audio.ogg -ar 16000 -ac 1 -c:a pcm_s16le output.wav
|
| 74 |
```
|
|
|
|
| 50 |
|
| 51 |
pipe = pipeline(
|
| 52 |
task="automatic-speech-recognition",
|
| 53 |
+
model="seanghay/whisper-small-khmer-v2",
|
| 54 |
)
|
| 55 |
|
| 56 |
result = pipe("audio.wav",
|
|
|
|
| 61 |
)
|
| 62 |
|
| 63 |
print(result["text"])
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 64 |
```
|