Speechbrain/sepformer-whamr with large files

Is it possible to separate large files with speechbrain/sepformer-whamr?
I’ve used VAD to split up larger files which can then be processed by speechbrain/sepformer-whamr but I’m concerned that there could be inconsistency with speaker separation between each file (split using VAD).

Which VAD did you use?
Anyway, I think this is the best way to work with large audio file.
Maybe you should fix the min audio duration for 4-6 seconds