This is important, because it means that the Crypton characters will not be moving to VOCALOID5.
As of now, Crypton has confirmed that they will continue to work with YAMAHA, as this is not to be a divorce from them altogether. Rather, they will be looking to create their own software instead.
Unfortunately, it is unclear what plans Crypton has for their characters. However, they have stated they are looking to Piapro Studio for their potential future synth.
Currently, Piapro Studio is a VST that allows users to edit their VOCALOIDs while composing music.
Piapro Studio is currently limited to Hatsune Miku, Kagamine Rin and Len, Megurine Luka, MEIKO and KAITO. This would make it seem more reasonable that they would want to focus on expanding Piapro. It would maintain focus on their vocals while relying on in-house technology.
While there are still several details yet to be revealed about this exciting turn of events, there are some facts we have already!
We know for sure that the software is being developed for a 2020 release and it is currently focused on vocals that Crypton has already produced.
The software will be separate from VOCALOID, but it is possible YAMAHA is working with Crypton on this release, as well.
New Technology and Demo
Shortly after the initial announcement, a blog post was posted by T.Ryo with details of the new audio technology which Crypton is developing.
According to the post, Crypton is using audio signal processing and deep learning to develop this new technology, based on research with various speech-related societies.
Two VST vocal effectors which have been developed so far are Vocal Drive and Cherry Pie.
The Vocal Drive effector adds a distortion effect to the voice.
Depending on the settings used, you can create various vocal effects such as soft breathy vocals, pop growl, grunts and even a guttural growl for harsher vocals.
The Cherry Pie effector uses real-time, high-quality speech analysis and synthesis technology.
This effector has various settings to change the voice, such as flexible pitch operation, spectral controls and non-periodic index deformation.
The voice quality conversion function reads various network files, acquired by deep learning, and converts a voice into the voices of other people.
For example, you could convert a mature vocal to a youthful one or even convert a male vocal to female vocal.
For those of you who are wondering how all this works, check out the demo video below to see these new vocal effectors in action.
This is a developing story, so be on the look out for more details coming in the next few months!
We’re all excited and curious as to what Crypton could be planning for the future of our number one diva.