Whisper mels error.

#6
by ahamedddd - opened

Got the self.processor.tokenizer error and
getting ValueError: Whisper expects the mel input features to be of length 3000 or less, but found 6390. Make sure to pad the input mel features to 3000.
error

I am acutally trying to use this for a larger audio. Also can i use this for audio transcripts ?

Sign up or log in to comment