The spinout, named after the Zulu for "sounds/noise", offers sound design services based on their innovative procedural audio technology. The new platform provides a browser-based service that can be used to create almost any sound effect from scratch, and in real-time, based on intuitive controls guided by the user.
Nemisindo's transformative procedural audio technology aims to completely remove the reliance on sample libraries for sound effects by offering intuitive tools to generate any sound. Unlike other technologies that simply use pre-recorded sounds, Nemisindo's platform generates sounds that have never been recorded. These sound effects can also be shaped and crafted at the point of creation by the user, breaking through limitations of sampled sounds.
"It's digital Foley," said Nemisindo co-founder and Queen Mary Professor Josh Reiss. "The idea is that no one should need to own massive sound effect libraries to have great sound in their content. If someone wants a 'whoosh' sound for their game, or footsteps or gunshots, they just tell the system what they're looking for and adjust the sound while it's being created. Then it's ready for use."
"We think our service could become a valuable tool for anyone needing support on sound design for games, film, TV, VR as well as podcasters or bloggers. And our intuitive system means that high-quality sounds can be created by professionals and amateurs alike."
Nemisindo co-founder, Professor Josh Reiss, previously co-founded the intelligent music production service LandR, and is an academic co-founder of another start-up, Tonz, offering neural net audio processing plug-ins. He is also President-Elect of the Audio Engineering Society.
Unique approach
The technology is based on years of research and development by Queen Mary's Audio Engineering research team and builds on a large body of work by the research community into sound synthesis. By integrating it into a common framework, and making it lightweight and real-time, the Nemisindo team have made this technology applicable to a wide variety of contexts. Machine learning approaches have also been applied to improve the sound quality of the models.
Existing efforts that offer procedural audio usually focus on a few specific sounds, or still involve manipulating sound samples. Nemisindo's unique approach instead aims to generate any sound effect, without storing any sound samples.
The innovative technology offered by Nemisindo has already caught the attention of leading interactive entertainment company Epic Games, with the spinout receiving an Epic MegaGrant to develop procedural audio for the Unreal game engine.
