torchaudio compilation error for fairseq2 0,4.0

[sleepingcat4]

I was trying to install Fairseq2 for SONAR model on Google colab and I received this error: pytorch/audio#62

Note: it goes away for Pytorch version 2.5.1 while installed with torchaudio.

pytorch/audio#62

SONAR repo: https://github.com/facebookresearch/SONAR

[sleepingcat4]

OSError                                   Traceback (most recent call last)
[<ipython-input-2-95c54e860898>](https://localhost:8080/#) in <cell line: 0>()
      1 import torch
      2 
----> 3 from sonar.inference_pipelines.text import (
      4     TextToEmbeddingModelPipeline,
      5     TextToTextModelPipeline,

18 frames
[/usr/lib/python3.11/ctypes/__init__.py](https://localhost:8080/#) in __init__(self, name, mode, handle, use_errno, use_last_error, winmode)
    374 
    375         if handle is None:
--> 376             self._handle = _dlopen(self._name, mode)
    377         else:
    378             self._handle = handle

OSError: /usr/local/lib/python3.11/dist-packages/torchaudio/lib/libtorchaudio.so: undefined symbol: _ZN2at4_ops9fft_irfft4callERKNS_6TensorESt8optionalIN3c106SymIntEElS5_INS6_17basic_string_viewIcEEE