Karadenizli on Nostr: By the way, is there any way to do deepfakes and that sort of video editing? Or audio ...

By the way, is there any way to do deepfakes and that sort of video editing? Or audio editing? I was trying to deepfake a person in a video with a second provided picture and couldn't get it to work at all. Even the models that have both IMG to vid and vid to vid can't seem to do this. Especially since this is the biggest normie use case for ai and all of the solutions that do it are shitty single purpose "sign up for free trial and forget" apps. And it seems OAI and Gemini refuse any deepfakes of public figures.

Also, if you could figure out a way to provide the audio and video models as tools available to a text model with visual input, it would massively help some use cases.