Error Message: RuntimeError: Argument #4: Padding size should be less than the corresponding input dimension, but got: padding (1024, 1024) at dimension 2 of input [1, 65621, 2] about auxiliaryasr HOT 3 OPEN

Which line did you add? Does this error only happen when you have your extra line?

from auxiliaryasr.

@yl4579
I just convert a wav file from this youtube video down to a 1 second 24000khz wav file.
https://www.youtube.com/watch?v=GQt4SY_6-w4.
Here is the converted wav file
https://files.catbox.moe/2jfn9u.wav
I also get the same error message if I add any more lines and wav files from this page. Note I also convert them to 24000 khz after downloading these wav files.
https://huggingface.co/Bluebomber182/Judy-Hopps/tree/main/audio

from auxiliaryasr.

As I've found out... All files need to be mono! You can do this by adding:

# Convert to mono if needed
if np.ndim(wave) > 1:
    wave = np.mean(wave, axis=1)

Under sf.read() in _load_tensor()

from auxiliaryasr.