Mikoto Studio is a new application for Singing Voice Synthesis, combining existing community-made sound libraries (voicebanks) with the newest Deep Learning technologies.

Visual for the project, by Lovesick Hospital

"The Snow White Princess Is...", original song by Noboru, VSQx by UltimaV123, vocals using Macne Nana 2S.

A screenshot of Mikoto Studio, featuring first-party virtual singer WouterCV.

Artist statement

My passion for Singing Voice Synthesis is deeply rooted in my connection with the SVS community. After many years of struggling with the limitations and boundaries of existing software, and with no viable alternative in sight, I decided to take up the challenge and develop something myself.
In a broader sense: I took the lessons I learned as a user and applied those to my own design process. In collaboration with a diverse group of test users, I was then able to create a prototype that exceeds all of our expectations.

Ambitions

Helped by my unbiased and fresh approach, I've been able to achieve much more than I expected to be able to within a year. Extending this process into the future, and with the help of the Mikoto community, I believe that this is just the beginning of a longer journey. In five years, I hope to still be working on this project, albeit a much more mature version of the same idea.

Learned during the studies

During the past 4 years I have developed a broad knowledge on the subject of digital audio processing and music software development. I've been exposed to many different views and design approaches, and met a lot of people with very different wishes and requirements for the tools that are a part of their workflow.

List of publications / exhibitions / prices / concerts / shows etc.

- Night of the Nerds 2023: https://www.nightofthenerds.com/portfolio/ontdek-en-speel-met-een-ai/