incorrect audio shape

[sergedahdouh]

data_svc/waves-16k/
data_svc/whisper

speaker0<<<<<<<<<<
(639, 1024)
speaker1<<<<<<<<<<
Traceback (most recent call last):
File "/content/lora-svc/prepare/preprocess_ppg.py", line 54, in
pred_ppg(whisper, f"{wavPath}/{spks}/{file}.wav", f"{ppgPath}/{spks}/{file}.ppg")
File "/content/lora-svc/prepare/preprocess_ppg.py", line 26, in pred_ppg
ppg = whisper.encoder(mel.unsqueeze(0)).squeeze().data.cpu().float().numpy()
File "/usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py", line 1501, in _call_impl
return forward_call(*args, **kwargs)
File "/content/lora-svc/whisper/model.py", line 154, in forward
assert len_x <= len_e, "incorrect audio shape"
AssertionError: incorrect audio shape

any idea what is the issue speaker0 is my record voice around 11 sec and speaker1 is song which is around 57 sec

[MaxMax2016]

4 cut audio, less than 30 seconds for whisper