Contribution

Binaural room impulse responses upsampling using physics-informed neural network for personal sound zones

Authors

* Presenting author
Day / Time: 20.03.2025, 11:00-11:40
Room: Hall E, Poster Island H
Typ: Poster
Session: Electroacoustics and Audio Signal Processing (Poster Island H)
Information: The posters will be exhibited in Hall E north from Tuesday to Thursday, sorted by thematic context in the poster island indicated in the session title. The poster session at the specified time offers the opportunity to enter into discussion with the authors.
Abstract: Binaural room impulse responses (BRIRs) are a combination of room impulse responses (RIRs) and head-related impulse responses (HRIRs) that contain both room information and spatial perception information, where in-situ measurements are difficult. BRIRs interpolation is solved using traditional DSP methods. We propose to use deep learning to learn the implicit representation of BRIRs from a sparse measurements and add physics constraints to improve the performance. Our results show that the proposed method can overcome the problems in traditional methods.