Search results

omg they killed miss synthv herself?? i knew it was inevitable for her but they could've waited until after i bought her or something

I'm trying to work on getting good at making her sing English, but I'm never sure if I'm just biased when it comes to intelligibility because I actually know the lyrics to the songs. If someone could listen to this short Defoko English test and tell me if they can understand the words, I'd...

ever since eleanor forte plus came out, i've been saying twindrill needs to make an official ai voicebank trained off teto's utau voicebanks. it'd probably end up being synthv, but it'd be amazing if it wasn't (it'd be amazing if it happened at all, really)

dear beloved by iroharingo, flight code by aisole, evangelic girl is a gun by yeule, yubisaki halation by kisara + every shiohiirie song is rlly good, especially kyodai na r and sariyuku anata e. i needddd him to come back to music so bad it was all peak... obligatory pinocchiop mention...

i use firefox almost purely because UBlock still works on it (originally used chrome), but i've heard enough about chromium based browsers to not feel very inclined to ever switch back

i can't open it either. if it's the same as this, then it's just unlisted

mimi o sumashite! from what i can tell, it's only called "To You" on amazon music and apple music. the original utaite version by rumdarjun and the vocaloid self-cover are both album exclusives, so there's no music videos therefore very few people know it </3

4769

did you make a whole voicebank that way?? if so, that's still impressive, even if it sounds weird. i've only gotten as far as vowels as of now, but the "a" sound is the only one that doesn't just sound like vaguely speech sounding humming 1748343462 YESSS!! this was my favorite song a few years...

heresy=rotten heresy and chocolate=pinocchiop=peak 1748300789 i started trying to make an utau with drawn waveforms, but i think i need to reference higher quality audio than utaus have :ehhh: making a surgext utau's been going pretty well though

i need them to add it back before iroha synthv gets released cause i'd love to get her as my first ever paid synthv voicebank, but i simply would not be able to deal with having to tune around the ai autopitch

Do you have a general idea for what you want the instrumental to be like? Also, just a heads-up in case you don't already know, but ProSEKA NEXT doesn't accept songs made with vocalists other than the Cryptonloids (Miku, Luka, Len, Rin, KAITO, MEIKO) so you'd need to choose a different vocalist...

I'm not sure what the issue is, but I think you might've written the names of those files wrong. Are the files with dakuten named the kana themselves (が, ゔ, etc.) or the base kana with an added dakuten as a separate character (かﾞ, うﾞ, etc.)? If it's the latter, it might be causing problems with...

I've seen some fairly good Diffsingers with around 10 minutes of data, and I know there's some way to do XLS so you wouldn't even need to record in Japanese. If you really want Japanese data, and a lot of it, you could record for a voice conversion AI such as RVC or Diff-SVC first and use that...

I'm sorry, I wasn't familiar with this voicebank when I said that. Unless this is a different voicebank than the one I found, I think setting the phonemizer to [JA VCV] and only having "a" as the lyrics should work.

It seems like you don't have the right phonemizer selected. On the part where you select the singer, there's another setting under the name that says "DEFAULT". If you click that, you can change the phonemizer to an English one.