Problema con audios de 30 segundos

[hugozeballos]

Al intentar inferir con el moddelo para audios con una duracion cercana a 30 segundos el modelo se cae, especificamente generando este error:

Traceback (most recent call last):
File "/home/hugoz/storage/seed_vc/seed-vc-main/inference.py", line 278, in
main(args)
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/utils/_contextlib.py", line 116, in decorate_context
return func(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/seed_vc/seed-vc-main/inference.py", line 157, in main
(quantized, codes) = codec_encoder.quantizer(z, waves_input)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/mnt/ialabnas/homes/hugoz/seed_vc/seed-vc-main/modules/quantize.py", line 197, in forward
prosody_feature = self.preprocess(wave_segments)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/mnt/ialabnas/homes/hugoz/seed_vc/seed-vc-main/modules/quantize.py", line 191, in preprocess
mel_tensor = self.to_mel(wave_tensor.squeeze(1))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torchaudio/transforms/_transforms.py", line 619, in forward
specgram = self.spectrogram(waveform)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1736, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/nn/modules/module.py", line 1747, in _call_impl
return forward_call(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torchaudio/transforms/_transforms.py", line 110, in forward
return F.spectrogram(
^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torchaudio/functional/functional.py", line 126, in spectrogram
spec_f = torch.stft(
^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/functional.py", line 702, in stft
input = F.pad(input.view(extended_shape), [pad, pad], pad_mode)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/hugoz/storage/miniconda3/envs/seed-vc-env/lib/python3.11/site-packages/torch/nn/functional.py", line 5096, in pad
return torch._C._nn.pad(input, pad, mode, value)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
RuntimeError: Argument #4: Padding size should be less than the corresponding input dimension, but got: padding (1024, 1024) at dimension 2 of input [1, 1, 504]

[Plachtaa]

Please describe your issue in English as I don't understand your language.
Also attach input audio samples that caused your problem pls