I am working at a big international IT company. We are currently working on advanced text to speech software for English, German and Japanese.
In it's most basic form it enables you to create any voice you can imagine.
At first you create a basic voice, either with an extensive set of parameters or with a software that can "digitize" a voice if you feed it with specific samples, this process needs manual tuning though.
Then you can let it speak written text and modify the virtual voices feelings/speed/etc. For example there are depression and aggression sliders.
It doesn't work like current t2v software, it has only one huge basic voice (female, male, children...) and the modifier files that are responsible for the voice you want to hear are surprisingly small.
We are doing negotiations with several big players that are interested in our product. (Hollywood & translator services *wink wink (google (you guys are the only smart people we interact with, everyone else looks mentally disabled compared to you))*)
Additionally we already have an anime studio on board.
In the end of 2016 an original anime with only digital "voices actors" will begin airing, as a cross collaboration and advertisement for involved companies.
It is not a question whether our voices sound convincing enough, they do, it's rather a question of when real voices are not used anymore in animated media.
It is revolutionary and Vocaloid will look like a bad joke after we launch our product. It is indistinguishable from real voice and allows you to recreate real voices in a way that will fully eradicate any reliability of voice recordings.
Voctro, Yamaha and Serra can go die, they had their chances.
I am breaking several NDAs and my career would be over if they ever found out who I am, but at this point there are enough people involved for me to keep my anonymity and guess what, I am behind 9 proxies and a VPN.
See you in < three years.
Too bad anime will be dead by then.
>It is indistinguishable from real voice and allows you to recreate real voices in a way that will fully eradicate any reliability of voice recordings.
I doubt you can recreate all the slight and subtle variations and aspects of a real voice filled with the personality and emotions of a real person.
But it still sounds interesting.
Cool lie bro. In any case, it won't be used in anime because seiyuu already work for peanuts and they also serve as an attraction to many anime fans.
Doubt, ichi yoku. This is complete nonsense. You cannot emulate a person's feelings with a computer, no matter how extensive the parameter set may be. You can never replace popular seiyuu with a program. People may accept teh Rie, but no way in hell they're going to accept an artificial Rie.
Sugoi monogatari, Onii-chan.
It might revolutionize indie productions in that it could save the producers money (provided the price of your product won't be too jewed up), but VA's are also idols of a kind, with their own fans and such. I don't think you'll be able to get that kind of success with a program, if it won't be successfully marketed in the same way Hatsune Miku was.
This sounds fucking terrible.
I'd rather have an actual VA doing voices.
Say the company's name already, faggot.
You can't even tell a lie, OP.
The farther we get from 3DPD, the better.
>It is revolutionary and Vocaloid will look like a bad joke after we launch our product.
You don't even understand the appeal of vocaloid, shill
>English, Japanese, German
Oh great, just the three most complicated Text-to-Speech languages ever spoken! I'm sure this'll be PERFECT!
Verpiss dich aus meiner Sprache, Hurensohn.
So you think this will be used on sexrobots to sound just like real humans?
> sex robots
First 3D Glasses and fleshlights. Sexbots won't be affordable till after 4chan's death.
We were born to early.
Videogames with fully-voiced everything.