Hi, and thanks for the wonderful dataset. We've seen its popularity and huge contribution to the speech process area.
However, during my experience using it, there was a time that I needed to migrate the whole dataset to another server. But my jobs falled when trying to use LibriMix then. It took me half a day to find out that all the mixture_path in mixture_***_***_***.csv were in the form of AP (absolute path) so my programs couldn't find the mixture files at their new home.
Later I'll create a pull request to contribute a transcript to update those paths for convenience of others who need to migrate the whole dataset like me, instead of regenerating it.
Hi, and thanks for the wonderful dataset. We've seen its popularity and huge contribution to the speech process area.
However, during my experience using it, there was a time that I needed to migrate the whole dataset to another server. But my jobs falled when trying to use LibriMix then. It took me half a day to find out that all the
mixture_pathinmixture_***_***_***.csvwere in the form of AP (absolute path) so my programs couldn't find the mixture files at their new home.Later I'll create a pull request to contribute a transcript to update those paths for convenience of others who need to migrate the whole dataset like me, instead of regenerating it.