speech-to-speech-translation

Sleeping

hewliyang commited on Jan 30

Commit

21f0bae

•

1 Parent(s): 0323180

switch back to whisper-medium

Files changed (3) hide show

app.py CHANGED Viewed

@@ -16,7 +16,7 @@ print(f"Using {device} with fp {torch_dtype}")
 # load speech translation checkpoint
 asr_pipe = pipeline(  # noqa: F821
     "automatic-speech-recognition",
-    model="openai/whisper-large-v3",
     device=device,
     torch_dtype=torch_dtype,
 )

 # load speech translation checkpoint
 asr_pipe = pipeline(  # noqa: F821
     "automatic-speech-recognition",
+    model="openai/whisper-medium",
     device=device,
     torch_dtype=torch_dtype,
 )

dev.ipynb DELETED Viewed

The diff for this file is too large to render. See raw diff

example.wav DELETED Viewed

Binary file (263 kB)