Sophia the Robot, built by Hong Kong’s Hanson Robotics, has performed a duet with Jimmy Fallon on the Tonight Show using speech technology developed by Edinburgh-based CereProc.

The text-to-speech (TTS) specialist is the only firm of its kind in the UK and has collaborated with Hanson Robotics to help Sophia showcase its TTS synthesiser.

The Scottish company, which created Sophia’s “characterful” voice, as well as its new singing voice, said it had “pushed its technology to the next level” and that its voice cloning technology has the “potential to change the music industry”.

CereProc trained a Deep Neural Network system and built a database designed for singing synthesis in order to recreate the timbre, expression and other human characteristics associated with singing.

Last week, Jimmy Fallon and Sophia sang Say Something by A Great Big World featuring Christina Aguilera – thought to be the first robot/human duet ever performed live on television.

Following Sophia’s performance, Chris Buchanan CereProc audio development engineer, tweeted: “The world’s first robot-human duet, broadcast live on #FallonTonight. Proud to have been part of its development at @cereproc and @hansonrobotics. Nice improvisation at the end too @RealSophiaRobot 😉 (PS Jimmy genuinely did choose that song).”

CereProc plans to release a beta version of its technology next year and is seeking commercial opportunities.

Matthew Aylett, chief scientific officer at CereProc, said: “At the moment the market is not clear, but there is a lot of potential as we’ve seen with synthetic instruments over the years. What we would need to commercialise this technology is to find a good partner.”

The technology, he believes, could be utilised in toys and gaming, or could be used to create backing vocals, for example.

He said: “This will probably get a mixed response, just as people had very mixed responses to pitch-changing technology, which was controversial in various ways.

“We’re not trying to replace human singing. The synthesiser doesn’t know what it’s singing and one of the most fundamental elements of a human singer is that they understand the sentiment, the reality of what they’re singing. But at the same time, as computers enter the social domain, being able to do these things becomes more and more important.”