This is just an idea. Some folks might have privacy concerns about sharing clear voice recordings, as they can be used to synthetize and create TTS models of other people without authorization. Maybe offering some kind of optional distortion would be a good feature to add.
creates video voice memos from audio clips; with bluesky integration.
trill.ptr.pet
offer some kind of distortion to avoid exact replication #1
open
opened by
good idea! it sounds like it wouldn't be that hard to implement, it should just be implementing some existing distortion method i guess?