Mikoto Studio is a singing synthesizer with UTAU support.
Just recently they announced their first 2 vocals for the engine.
Just recently they announced their first 2 vocals for the engine.
Other Mikoto Studio Discussion Thread
- Thread starter sapplerx
- Start date
Gotta admit my alarm bells went off at first (I've been in the fandom too long fsdjlk) but looking into it more I'm willing to give it the benefit of the doubt for now. They have some more info in the Discord but apparently you're not allowed to share info from the Discord outside of it, if I understood the rules correctly?? But anyone can join it if they join the waitlist, so IDK what the point of that is honestly unless it's mostly meant for the people who are beta testing
I literally just wanted to make a thread about it sgfagagagagagagsgsysyshs
anyways, there seems to be some sort of waitlist for it
EDIT: I’ve just realised that this thing exists since February…
Last edited:
It'll be exciting to see how this turns out! In the past, I would have been much more sceptical of new engines as well, but AI, for better or worse, is a far more mainstream trend than concatenative synthesis ever was. This means resources for developers are very accessible, with plenty of code to freely study. It's only a matter of time before someone duck-tapes OpenUTAU and Diff-SVC together, and gives AI development a simpler workflow.
My one concern is that they gave their program the same name as an already existing vocal synth (Vocaloid Mikoto); this could cause confusion and potentially some trademark issues. The difference between an engine and a voicebank is obviously clear if you follow vocal synth development closely, but it could be an issue for casual fans.
As an aside, this makes me reflect again on Maghni AI, which has now been in development for so long that a whole other engine (beginning with M) has been announced before they've got a GUI screenshot. I wonder if Maghni's running out of runway to establish themselves before a competitor ticks all the same boxes. But it's too early in either engine's development to say for sure: they could be completely different in their own cool ways.
My one concern is that they gave their program the same name as an already existing vocal synth (Vocaloid Mikoto); this could cause confusion and potentially some trademark issues. The difference between an engine and a voicebank is obviously clear if you follow vocal synth development closely, but it could be an issue for casual fans.
As an aside, this makes me reflect again on Maghni AI, which has now been in development for so long that a whole other engine (beginning with M) has been announced before they've got a GUI screenshot. I wonder if Maghni's running out of runway to establish themselves before a competitor ticks all the same boxes. But it's too early in either engine's development to say for sure: they could be completely different in their own cool ways.
This is interesting. Two things I noticed:
1) Mikoto is a very weird name for a voice synthesizer.
2) The UTAU support— I wonder if this means they’re building off of the openutau codebase?
OpenUtau is licensed under MIT, which means commercial use is allowed.
1) Mikoto is a very weird name for a voice synthesizer.
2) The UTAU support— I wonder if this means they’re building off of the openutau codebase?
OpenUtau is licensed under MIT, which means commercial use is allowed.
Any info from the discord server is prohibited from being shown to those who aren’t in the server so no. But if you want to get into the discord server you just need to register on the waitlist
Hi, developer of Mikoto Studio here! I'd like to clear up a few things.
If you want to get more info, I'd recommend you to sign up for the waitlist and join the Discord server - it's a rather active community and we share plenty of insights and information there! Very happy to see all this interest in my little graduation project, and I'm excited to see what the future holds.
Mikoto Studio is developed completely from scratch, using C++ and the JUCE framework. This includes the UTAU support as well. I've obviously looked at the OpenUTAU code for inspiration in desperate moments, but there is no code that was directly copied from there.
I realized this after I'd already been plastering the name all over my codebase for a few months, so it's rather unlikely it'll change now. If anyone is wondering, the synth is not named after the voicebank, but rather inspired by a certain anime character.
If you want to get more info, I'd recommend you to sign up for the waitlist and join the Discord server - it's a rather active community and we share plenty of insights and information there! Very happy to see all this interest in my little graduation project, and I'm excited to see what the future holds.
A First Look at Mikoto Studio - ExpressiveLabs blog
To everyone who put "a new commercial singing synthesizer" on their 2024 vocal synth bingo cards: this one is for you! Today, after almost two years of extensive research and development, we can finally share the first snippets of Mikoto Studio.
blog.expressivelabs.net
a new blog post and a sneak peak at the UI
Great update, it looks like it's coming along well! Looking forward to hearing the first sample.
Mikoto has posted the first sample of their engine, I assume this is their concatenative UTAU compatability on display:
There's one more demo releasing very shortly:
Also moved this thread to 'News' to be consistent with Maghni AI.
There's one more demo releasing very shortly:
Also moved this thread to 'News' to be consistent with Maghni AI.