https://github.com/KoeAI/LLVC 써봄

CUDA 11.8 버전 쓰고 있어서

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118 

설치했으나 C컴파일을 요구해서 Vistual Studio Community 설치.

아래보고 참조함

<https://wikidocs.net/168848 참조>

FAIRSEQ 관련 오류는

Using cached ruamel.yaml.clib-0.2.8-cp311-cp311-win_amd64.whl (118 kB)

Building wheels for collected packages: fairseq

  Building wheel for fairseq (pyproject.toml) ... error

  error: subprocess-exited-with-error



  × Building wheel for fairseq (pyproject.toml) did not run successfully.

  │ exit code: 1

  ╰─> [1829 lines of output]

      No CUDA runtime is found, using CUDA_HOME='C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.8'

.............................. 

          match = self._regex.search(version)

                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^

      TypeError: expected string or bytes-like object, got 'NoneType'

      [end of output]



  note: This error originates from a subprocess, and is likely not a problem with pip.

  ERROR: Failed building wheel for fairseq

Failed to build fairseq

ERROR: Could not build wheels for fairseq, which is required to install pyproject.toml-based projects

이런 것이었는데 버전이 git+https 로 되어있던 requirements.txt 중 fairseq @ git+https://github.com/One-sixth/fairseq.git 이 부분을 그냥 fairseq 로 변경

일단 넘어는 갔다.

`python infer.py

로 추론을 해보니

converted_out\174-50561-0000.wav
converted_out\652-129742-0000.wav
converted_out\777-126732-0000.wav
converted_out\1919-142785-0000.wav
converted_out\2086-149214-0000.wav
converted_out\2412-153947-0000.wav
converted_out\2902-9006-0000.wav
converted_out\5895-34615-0000.wav
converted_out\7850-73752-0000.wav
converted_out\8842-302196-0000.wav

파일이 잘 생성된다.

흠.. 실시간 생성은 어떻게 하는거지? -s 옵션 준다고 되진 않네.