I thought of this- What if a scratcher creates a song?(Or uses an ai generator like I do?) and they want to share it on scratch. However they can't sing it for safety concerns and they don't want to only post the lyrics. If the text to speech could have singing voices, it would fix this problem.

Support, this would be a good idea if you don't want to record your own voice.

use fakeyou (am I allowed to mention that website there’s nothing inherently bad on it)

Gamer_Logan819 wrote:

use fakeyou (am I allowed to mention that website there’s nothing inherently bad on it)

I looked at it. I meant like singing.

Horsekatto wrote:

Gamer_Logan819 wrote:

use fakeyou (am I allowed to mention that website there’s nothing inherently bad on it)

I looked at it. I meant like singing.

It has voice to voice. Sing it and put it through that filter.

Gamer_Logan819 wrote:

Horsekatto wrote:

Gamer_Logan819 wrote:

use fakeyou (am I allowed to mention that website there’s nothing inherently bad on it)

I looked at it. I meant like singing.

It has voice to voice. Sing it and put it through that filter.

I'm not gonna put my voice in a random website I know basically nothing about.

Support. This would bring a bunch of new opportunities for people and it would be interesting mixed with other things (like animation, a game etc.)

Do you guys have any idea how hard it is to make an ai voice talking, much less an ai voice singing? No support this is a waste of resources that could be used for example fixing the forum servers.

5_g wrote:

snip

But singing voices would create more possibilities for tts voices. Plus, what if someone doesn't want to put their voice in the project, but they want singing, but they don't want AI voice websites either because they think they might be hiding malware?

Horsekatto wrote:

I thought of this- What if a scratcher creates a song?(Or uses an ai generator like I do?) and they want to share it on scratch. However they can't sing it for safety concerns and they don't want to only post the lyrics. If the text to speech could have singing voices, it would fix this problem.

that's a great idea, on paper. all of scratch's text to speech voices are from google, and if google doesn't have a tts voice for singing, we can't really do that.
also, how would the tts know the melody / how to sing it?

How would it know what notes/rhythms/articulations to sing?

k7e wrote:

How would it know what notes/rhythms/articulations to sing?

exactly, this would be impossible, scratch does not have the resources to get close to this

Horsekatto wrote:

I thought of this- What if a scratcher creates a song?(Or uses an ai generator like I do?) and they want to share it on scratch. However they can't sing it for safety concerns and they don't want to only post the lyrics. If the text to speech could have singing voices, it would fix this problem.

I don't see how this would work - what would define the melody that the voice would sing?