RVC-Project / Retrieval-based-Voice-Conversion-WebUI

Easily train a good VC model with voice data <= 10 mins!
MIT License
24.52k stars 3.61k forks source link

Getting failed to load audio error #1329

Open DhananjayPurohit opened 1 year ago

DhananjayPurohit commented 1 year ago

I am storing the voice like this:-

Download Results:
gid   |stat|avg speed  |path/URI
======+====+===========+=======================================================
353dd9|OK  |   472KiB/s|/content/Retrieval-based-Voice-Conversion-WebUI/dataset/alia.wav

On running the cell having commands:-

%cd /content/Retrieval-based-Voice-Conversion-WebUI
sudo python3 infer-web.py --colab --pycmd python3

the UI is starting up:- Screenshot (318)

I think it's taking the location fine as it can read up the wav file inside the dataset folder but still getting the error:-

start preprocess
['trainset_preprocess_pipeline_print.py', 'dataset', '40000', '2', '/content/Retrieval-based-Voice-Conversion-WebUI/logs/mi-test', 'False']
dataset/alia.wav->Traceback (most recent call last):
  File "/content/Retrieval-based-Voice-Conversion-WebUI/my_utils.py", line 14, in load_audio
    ffmpeg.input(file, threads=0)
  File "/usr/local/lib/python3.8/dist-packages/ffmpeg/_run.py", line 325, in run
    raise Error('ffmpeg', out, err)
ffmpeg._run.Error: ffmpeg error (see stderr output for detail)

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "trainset_preprocess_pipeline_print.py", line 75, in pipeline
    audio = load_audio(path, self.sr)
  File "/content/Retrieval-based-Voice-Conversion-WebUI/my_utils.py", line 19, in load_audio
    raise RuntimeError(f"Failed to load audio: {e}")
RuntimeError: Failed to load audio: ffmpeg error (see stderr output for detail)

end preprocess

I have tried all the ways quotes and without quotes, using double slashes and single slash, don't have spaces in file or folder name.

Any suggestions or help will be much appreciated.

Thanks in advance.

ZZZ-zero commented 1 year ago

ffmpeg可能是因为它不需要将wav转为wav文件而报错,也可能是由于ffmpeg版本报错,请先尝试重装ffmpeg。

DhananjayPurohit commented 1 year ago

I have tried downgrading the ffmpeg version. Can you specify the version I should try with?

RVC-Boss commented 1 year ago

try copying the absolute path of dataset?

DhananjayPurohit commented 1 year ago

I have tried absolute path as well but getting the same error. I tried drilling down the issue and found that it's saving the feature extraction file in /logs/mute instead of /logs/mi-test. Is there anything wrong with my audio file? Any ways to fix this? FileNotFoundError: [Errno 2] No such file or directory: '/content/Retrieval-based-Voice-Conversion-WebUI/logs/mi-test/3_feature256'

My folder structure:-

.
   |-TEMP
   |-_pycache_
   |-configs
   |-dataset
   |-docs
   |-i18n
   |-infer
   |-infer_pack
   |---_pycache_
   |-logs
   |---mi-test
   |-----0_gt_wavs
   |-----1_16k_wavs
   |-----2a_f0
   |-----2b-f0nsf
   |---mute
   |-----0_gt_wavs
   |-----1_16k_wavs
   |-----2a_f0
   |-----2b-f0nsf
   |-----3_feature256
   |-----3_feature768
   |---na
   |-----0_gt_wavs
   |-----1_16k_wavs
   |-----2a_f0
   |-----2b-f0nsf
   |---nigga
   |-----0_gt_wavs
   |-----1_16k_wavs
   |-----2a_f0
   |-----2b-f0nsf
   |-pretrained
   |-pretrained_v2
   |-train
   |---_pycache_
   |-uvr5_pack
   |---_pycache_
   |---lib_v5
   |-----_pycache_
   |-----modelparams
   |-uvr5_weights
   |-weights
DhananjayPurohit commented 11 months ago

@ZZZ-zero @RVC-Boss

PasiKoodaa commented 6 months ago

I had the same error in WLS2 Ubuntu, the fix was

pip install ffmpeg-python instead of pip install ffmpeg