Closed aedocw closed 9 months ago
Models like XTTS are the only kind that benefit from having a transcript compare the output to the original text. When using VITS, this is just a waste of time/CPU, so should be skipped.
Models like XTTS are the only kind that benefit from having a transcript compare the output to the original text. When using VITS, this is just a waste of time/CPU, so should be skipped.