Open gingerling opened 6 years ago
I agree! It would technically be possible to do this, but would require a fair amount of manual effort. The "singing" is not generated dynamically, so this will need someone to produce the viseme sequence manually and align the timing.
If anyone wants to undertake this, I'd be happy to provide advice on how to do it.
Do you think this would be possible for an intelligent technically minded NON-programmer to have a go at? It strikes me that potentially it's a bit like stop-motion animation, which I have done before, I'm wondering how repetitive the programming would be - if it's a case of you write the beginning and end code and I fill in something which says "what milliseconds, which mouth shape" for the entire song that is possibly doable...
On Thu, 27 Sep 2018, 07:43 Steve Penrod, notifications@github.com wrote:
I agree! It would technically be possible to do this, but would require a fair amount of manual effort. The "singing" is not generated dynamically, so this will need someone to produce the viseme sequence manually and align the timing.
If anyone wants to undertake this, I'd be happy to provide advice on how to do it.
— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/MycroftAI/skill-singing/issues/9#issuecomment-424976669, or mute the thread https://github.com/notifications/unsubscribe-auth/ACuIsH4HlIJfEXKoZFtApxLCY01UBTdHks5ufHN6gaJpZM4W1Z2d .
Hey would be great if mycroft's mouth would move while he sings, but he goes back to sleep