Microsoft studio adopts Edinburgh firm’s software

The Coalition, a Microsoft games studio, has been awarded a license to use software created by a university spin-out which generates facial animation from audio.

Speech Graphics has built a reputation for advances in the field, providing facial animation for the video games industry and working with companies like Warner Brothers and artists such as Kanye West.

Its SGX software, the result of five years of research and development,  allows video game studios to create animation from thousands of lines of dialogue; for Microsoft’s Gears of War 4 it animated more than 35,000.

“Speech Graphics provided us with a robust system for automatically creating quality facial animation for the many thousands of lines of dialogue,” said David Coleman, the game’s animation director.

“SGX goes beyond good lip sync. Speech contains energy and emotion, and that too can be decoded from the voice and synchronized in the face. Using all available acoustic information, algorithms drive not just the mouth but the entire face from audio input, from syllables to scowls.”

The eigtht-strong Speech Graphics team brings together decades of research into machine learning, speech recognition, phonetics and computer graphics.

Michael Berger, co-founder and chief technology officer, said: “Automatic, accurate lip sync is one of the holy grails of computer facial animation. Our task is to create the impression that the animated face you see is the source of the sound you hear.

“This illusion is notoriously difficult to achieve; the movements of speech are fast, complex and subtle and the viewer is highly sensitive to any mismatch between face and the voice.”

The company, which is planning to hire three more people in the next year, aims to be the main provider of voice animation, a market forecast to be worth £375m.


Stay up to speed on Scottish tech news