Deep voices #286
Labels
No Label
bug
duplicate
enhancement
help wanted
insufficient info
invalid
news
not a bug
question
wontfix
No Milestone
No project
No Assignees
3 Participants
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: mrq/ai-voice-cloning#286
Loading…
Reference in New Issue
Block a user
No description provided.
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Has anyone had success with deep voices? I've found every model trained with a deep speaker so far to come out much higher than their normal voice.
Both low and high-pitched voices come out closer to the median. Might improve with more training cycles but I usually just pitch-shift it with ffmpeg.
I have gotten pretty accurate (meaning low pitched) results on George Takei and James Earl Jones using ~250 epochs however the results have static-y artifacts. I'm trying to finetune BigVGAN to see if that helps