Not so sure how to feel about this, idea popped in my head last night at 2 AM, did a rough first draft and this is the final result, I'm not so happy with the final section, but at least I got the idea out of my head
tts skits might get a hole lot better, or worse, depending on your point of view
this would 100% pass the casual listening test, coffee shops and the like, as always, the raw, unmixed file is available upon request, convertion done with audimee.com
i have to add the fact this is not 1 response but a comp of 5 or so responses in the same conversation, i just stripped the preamble/me talking to it
what chat gpt advanced voice mode has to say about the state of things, ran through eleven labs voice voice isolation to upsample the audio, original available on request
the second singer is AI, with my own vocals as the input, seams to be some clipping through out the song, since i forgot to use a limiter on the master track, but it's not too bad
an ai cover of bad dreams using an rvc model trained on my voice, not the best ever but passable for backing vocals, and yes, the worning is there for idiots who can't read properly and screem "zis guy no good, he using AI he stupid he think people do not know?I've seen it before so decided to cover my bases just incase