AI approach to acoustic response extrapolation for Bonza Music immersive rendering
Project Profile
Project Lead
Dr Michael McLoughlin (Academic Lead)
Fiona Ryder (Industry Lead)
Associated People
University of York
Dr Karolina Prawda
Professor Gavin Kearney
Professor Helena Daffern
Research Theme
AI and Creative Industries
Project Funder
Innovate UK
Project Partners
Bonza Music Ltd
Project Description
Immersion in virtual and augmented reality solutions is reliant on plausible spatial audio. However, performing acoustic measurements often requires many individual measurements of source-receiver pairs with specialist spatial microphones, making the procedure time-consuming and expensive. In this project, we explored a method utilising machine learning and signal processing techniques to gather a set of spatially extrapolated ambisonic measurements from a single mono Room Impulse Response (RIR). We evaluated the effectiveness of our method in providing perceptually plausible extrapolated RIRs using 3-Alternative Forced Choice (3AFC) listening test.