A Python script to convert the WSJ0 speech corpus to more friendly file formats.
sph2pipe
in PATH. Get it from here.ffmpeg
in PATH (not required if using the--no-flac
option)
Simply run:
python wsj0_convert.py <path-to-WSJ0> <output-dir>
This will create an audio
directory inside <output-dir>
containing all the audio files in .flac
format,
or in .wav
format if using the --no-flac
option.
The audio files are organized in sub-directories by speakers.
- Only
.wv1
files are converted..wv2
files are skipped. - The original folder structure is not respected. Output files are organized in sub-directories by speaker.
- Total output size if 3.9 GB when using
.flac
format. - For extra speaker information (e.g. gender), see here.