The world's first
voice synthesis solution
for singing and acting.

We create a hyper-realistic and
expressive voice that cannot be distinguishable from real humans.

APPLICATIONS

Commercial content production
Meet your favorite legends again.
Artist and Repertoire
Spread your wings of imagination. Find a voice that fits your work.
Virtual Celeb
Get a whole new voice without royalty issues.

DEMO

PREV
NEXT

PROCESS

1. Voice sample collection

We acquire voice samples for the raw material for training AI models. The valuable voice data you provided is never disclosed to the public and is only used for model training purposes. For security, watermarking and voice modulation technologies are applied and strictly managed with the minimal access permission.

2. Preprocessing training data

If noise or music is mixed with voice, it is first clearly enhanced by Supertone's high-quality source separation technology. Then we make music or text annotations with our experts in the field of music/audio content production.

3. Model training

Using the elaborately processed training data, training a neural network is done with our high-performance computing server. Developed through long-term original research, Supertone's voice synthesis and cloning neural network boasts the world's best hyper-realistic quality.

4. Model deployment for creators

Limited rights to use the trained model are granted only to creators approved by the right party of the voice. For music creation, we provide an API based on the MIDI protocol to creators so that they can make music easily in the same working environment.

5. Post-correction (optional)

When content is needed to be at a commercial level, we proceed with a detailed direction just like the conventional recording and filmmaking process.

PORTFOLIO

"AI Music Project: Once Again"

AI music project "Once Again,"
which presents a new stage by recreating the voices of missing artists such as Kim Hyun-Sik and Turtleman who have passed away a long time ago. It is also a warm drama in which artists sing a new song that has been never performed in their lifetime but is familiar to the public, presenting them to bereaved families, acquaintances, and fans to commemorate the deceased together.

"Battle of the Century: AI vs. Human"

Planned as a special program for the 30th anniversary of SBS,
which is one of the major broadcasting channel in Korea, the show introduces AI from various fields to compete with human experts. It is a cultural entertainment program designed to suggest a direction in which artificial intelligence and humans can coexist.

ETHICAL AI

We understand the resistance people can feel when AI first crosses the "uncanny valley".

  • We have also been deeply concerned about the problems
  • that can arise when this technology is used for wrong purposes.
Supertone

does not do:

We never monetize any synthetic voice without the permission of the right holder.
Non-commercial research is limited to those who have publicity or have passed away.
Supertone takes care of

the data you provide:

We manage to minimize access to training data and synthesized voice.
We have our original audio watermarking technology that can identify whether it is synthetic or not.
Supertone respects

the creative culture:

We can easily create all kinds of content, from simple text reading to artwork.
A completely recreated voice will relieve you of worrying about privacy issues.