Large audio file (more then 2 hours)

#59
by jonfv - opened

My code:

pipe = pipeline(
    "automatic-speech-recognition",
    model="openai/whisper-large-v2",
    generate_kwargs={"language": "br", "task": "transcribe"},
    device="cpu",
    use_fast=True
)

res = pipe(YT_AUDIO_FILE, batch_size=10, return_timestamps=True, chunk_length_s=30, stride_length_s=(4, 2))

Why the pipe finish after end of audio? The audio have more then 2 hours and less then minutes is generated.

Thx!!!

Hey @jonfv - your code looks good. Could you share the audio file so I can reproduce locally on my end?

Sign up or log in to comment