Originally Posted by jpettit
Yes because this tool works off of phonics You can transcend languages.
Unlike Vocaloid, which tried to be language neutral by recording international phonemes, SynthV records voice providers in their native language, with their regional dialect.

More simply said, SynthV voices are not language neutral.

Dreamtonics uses AI to convert the voices from one language to another via a process called "cross lingual synthesis", but accents still can come through. I find that Japanese often retains a bit of an accent when converted into English, especially for phonemes that require tongue thrusts. YMMV.

Quote
The other thing that you need to realize is that choirs are not going to have voices that stand out. They went to a lot of work to make the voices blend.
Yes, but how choral vs. soloistic depends on who they recorded and how they sang. My impression is that the English voices seem to be a bit more less homogeneous than the others, but that may just be that I don't pick up on Chinese or Japanese that well. That is, I think the non-English voices sound a bit less choral than the others. But I'll have a chance to test that out next week.

Quote
It's more of a question of does it sound like a choir or not.
Remember you can test them all and I think you have till the end of February.
100% this. If it sounds good to you, it is good. smile

-- David Cuny

My virtual singer development blog
Vocal control, you say. Never heard of it. Is that some kind of ProTools thing?

BiaB 2025 | Windows 11 | Reaper | Way too many VSTis.