search

bytedance / music_source_separation

Other
1.26k stars 194 forks source link

my mp3 doen't work #18

Open mxzhao opened 2 years ago

mxzhao commented 2 years ago

I tried a mp3, but it failed: Using cuda for separating .. /home/zeppsh/anaconda3/envs/music/lib/python3.7/site-packages/librosa/core/audio.py:165: UserWarning: PySoundFile failed. Trying audioread instead. warnings.warn("PySoundFile failed. Trying audioread instead.") Traceback (most recent call last): File "/home/zeppsh/anaconda3/envs/music/lib/python3.7/site-packages/librosa/core/audio.py", line 149, in load with sf.SoundFile(path) as sf_desc: File "/home/zeppsh/anaconda3/envs/music/lib/python3.7/site-packages/soundfile.py", line 629, in init self._file = self._open(file, mode_int, closefd) File "/home/zeppsh/anaconda3/envs/music/lib/python3.7/site-packages/soundfile.py", line 1184, in _open "Error opening {0!r}: ".format(self.name)) File "/home/zeppsh/anaconda3/envs/music/lib/python3.7/site-packages/soundfile.py", line 1357, in _error_check raise RuntimeError(prefix + _ffi.string(err_str).decode('utf-8', 'replace')) RuntimeError: Error opening '2.mp3': File contains data in an unknown format.

Is there any limitation about the mp3 format? would you please share your test mp3: ./resources/vocals_accompaniment_10s.mp3

qiuqiangkong commented 2 years ago

Mp3 is OK. The mp3 file is uploaded. The installation is simplified. Try pip install bytesep and follow the updated README?