Me and a friend have been working on a synth for 2 years now called QVoice

Mayim

Momo's Minion
It's going to be an AI concatenative hybrid and you can learn more about it here > https://bit.ly/3mKsbVT (Some information has become dated)

This is going to be our senior project for high school so we're mainly building off of already existing things like Kontakt and NNSVS. We're planning on running an Indiegogo in November through December and we're hopping to have samples ready by the end of October.

There will be 6 vocalists. A baritenor suited for rock & metal, an androgynous alto, a well rounded bass, a brisk mezzo-soprano, an opera based contralto and a Miku-esk soprano. We haven't hired voice providers yet but we do have ideas. Most of which are in the community. Our Indiegogo will mainly cover paying them.

Earlier today we released our third input survey which asks stuff along the lines of rewards, phonetics and more. > https://forms.gle/c7nBSv2BagLNf2ks6

If anyone has any questions, feel free to ask!

Website | Twitter
 

Kiyoteru

UtaForum power user
Supporter
Defender of Defoko
Software development is a rather substantial ordeal that I don't think one can easily put aside in favor of character design and marketing. This is primarily going to be a completely new piece of software, right? I would be more interested in seeing plans regarding that aspect than in UTAU-like documents about reclists.

If you haven't started yet, these are some template documents you can use. I would recommend starting with the software requirements specification, then moving onto a more detailed design document.

 
  • Like
Reactions: Kazumimi

Mayim

Momo's Minion
Thread starter
Software development is a rather substantial ordeal that I don't think one can easily put aside in favor of character design and marketing. This is primarily going to be a completely new piece of software, right? I would be more interested in seeing plans regarding that aspect than in UTAU-like documents about reclists.

If you haven't started yet, these are some template documents you can use. I would recommend starting with the software requirements specification, then moving onto a more detailed design document.

QVoice will run off of Kontakt but if we can't do what we want in it, we'll temporarily switch to using OpenUTAU as a GUI and NNSVS as the full synthesis until we can make a new synthesis engine. Here's our phonetic script. All of our SRD documents are already done.
 

Kiyoteru

UtaForum power user
Supporter
Defender of Defoko
I understand if you don't feel the need to publish documents regarding the software development side, but it doesn't inspire a lot of confidence to have absolutely zero indication of any concrete planning other than "we'll use Kontakt and NNSVS". This is especially due to the fact that it's a high school project and that it's planned to be crowdfunded, and it's not very clear how this will actually be built. If "OpenUTAU + NNSVS" is the simpler method until a new GUI is developed, maybe we can see a prototype of this in action.

It seems like attention is divided between two completely separate things.
1. Developing the software itself (creating the Kontakt instrument, integrating NNSVS, working on the GUI and synthesis engine)
2. Designing and marketing characters (writing reclists, creating merchandise, planning voicebanks)

And most of the attention is on the latter.
If this imbalance of focus is purposeful, I think you should reconsider the software plans entirely and focus on using existing engines to develop high quality voicebanks, characters, merch, etc. There are a few groups who are taking this professional approach to UTAU such as Tasteloid and VOISTAR Project. I've heard that Empathy Studios LLC is working on a new UTAU project along these lines. There are also startup companies like Eclipsed Sounds that are working with Dreamtonics to produce a SynthV voicebank, or how VoxWave previously collaborated with Plogue on Alter/Ego.
If the imbalance is accidental, then I would suggest reevaluating your priorities and making sure you can deliver a functional singing synthesis plugin first and foremost before thinking about characters and voicebanks.
 
  • Like
Reactions: Nohkara

Mayim

Momo's Minion
Thread starter
I understand if you don't feel the need to publish documents regarding the software development side, but it doesn't inspire a lot of confidence to have absolutely zero indication of any concrete planning other than "we'll use Kontakt and NNSVS". This is especially due to the fact that it's a high school project and that it's planned to be crowdfunded, and it's not very clear how this will actually be built. If "OpenUTAU + NNSVS" is the simpler method until a new GUI is developed, maybe we can see a prototype of this in action.

It seems like attention is divided between two completely separate things.
1. Developing the software itself (creating the Kontakt instrument, integrating NNSVS, working on the GUI and synthesis engine)
2. Designing and marketing characters (writing reclists, creating merchandise, planning voicebanks)

And most of the attention is on the latter.
If this imbalance of focus is purposeful, I think you should reconsider the software plans entirely and focus on using existing engines to develop high quality voicebanks, characters, merch, etc. There are a few groups who are taking this professional approach to UTAU such as Tasteloid and VOISTAR Project. I've heard that Empathy Studios LLC is working on a new UTAU project along these lines. There are also startup companies like Eclipsed Sounds that are working with Dreamtonics to produce a SynthV voicebank, or how VoxWave previously collaborated with Plogue on Alter/Ego.
If the imbalance is accidental, then I would suggest reevaluating your priorities and making sure you can deliver a functional singing synthesis plugin first and foremost before thinking about characters and voicebanks.
We've been working on publishing a version of our SRD with things such as potential voice providers and other information we don't want to be public taken out. We have other plans such as main features that aren't talked about on that page in the drop down menu labeled QVoice on our website. There will be some prototype demos available before the Indiegogo goes live.

Currently, we haven't been able to actually develop the Kontakt library due to my friend not having access to the full version of it (I have a physical version of Komplete). Development will start on September 7th. In the meantime we've been doing things like prepping the UTAUs we have permission to use for demos, getting art done, finishing the diphonetic and triphonetic reclists, a better GUI mockup and much more that will be talked about in the August progress update. All of the character info was created in my free time since I've created a work schedule for myself. I recommend looking at the previous progress updates if you think we're focusing on the characters.
 

Kiyoteru

UtaForum power user
Supporter
Defender of Defoko
Best of luck. I hope what you manage to accomplish can at least let you graduate high school on time.
 

feelingiconiic

Ruko's Ruffians
Defender of Defoko
I'm sorry Kiyoteru, but this is so rude and unnecessary, and you do this on multiple posts belittling the work of others in the vocal synth community. This is supposed to be a forum for vocal synthesis, not to discourage others. Mayim, I've recently been researching AI singing and speech synthesis and this seem's like a interesting project that I'd love to follow the development of! Is there any where updates are being posted?
 

Kiyoteru

UtaForum power user
Supporter
Defender of Defoko
I'm sorry Kiyoteru, but this is so rude and unnecessary, and you do this on multiple posts belittling the work of others in the vocal synth community. This is supposed to be a forum for vocal synthesis, not to discourage others. Mayim, I've recently been researching AI singing and speech synthesis and this seem's like a interesting project that I'd love to follow the development of! Is there any where updates are being posted?
If you've been following them on Twitter and on their website, you will have noticed that their most recent document is using the template I provided them, and that have taken my advice to focus more on software development and the necessary tools to do so.
 

cloudyyskie

Ritsu's Renegades
Defender of Defoko
If anything, you're the one being hostile here. I'm saying they have a twitter and a website.
I don't believe they were being hostile, simply calling you out on your condescending and rude sounding replies. When confronted about it, your reply was essentially "Well, they took my advice, so it's OK."
...No. You still sounded rude, and after OP pointed out that you were wrong about their priorities, all you said was "Best of luck. I hope what you can accomplish will let you graduate on time.", which.. is extremely rude and belittling?? I'm not saying they may not have made some mistakes, but I am saying maybe you should do a better job of pointing it out and not sounding like a huge prick.
 
  • Like
Reactions: arekkusu_tobayashi

buffin_muffin

Momo's Minion
so you saying "but it doesn't inspire a lot of confidence to have absolutely zero indication of any concrete planning other than "we'll use Kontakt and NNSVS"." doesnt come off as the littlest bit of rude to you? not to say you cant critique this person but its the matter of phrasing things thats extremely important.
If anything, you're the one being hostile here. I'm saying they have a twitter and a website.
 
  • Like
Reactions: arekkusu_tobayashi

arekkusu_tobayashi

Momo's Minion
If anything, you're the one being hostile here. I'm saying they have a twitter and a website.
I don't know why I am replying to this too but if they're trying their best (think, two years) and if they are happy with what they do, just don't say stuff that makes them overthink. It is common sense that if you see them being happy and announcing a bunch of stuff online you should encourage them and, of course, not say stuff. Also, they're 17-18, and you're 22 or 23, and also beginners. Please reanalyse your sentences before transmitting them.