Towards HRTF Personalization using Denoising Diffusion Models

2025/01/06 by Luca Comanducci, Sánchez, Juan Camilo Albarracín, Mirco Pezzoli +4 · 4 citations
Business, Management and Accounting · #AI and HR Technologies

paper · pdf · doi:10.48550/arxiv.2501.02871

Abstract

Head-Related Transfer Functions (HRTFs) have fundamental applications for realistic rendering in immersive audio scenarios. However, they are strongly subject-dependent as they vary considerably depending on the shape of the ears, head and torso. Thus, personalization procedures are required for accurate binaural rendering. Recently, Denoising Diffusion Probabilistic Models (DDPMs), a class of generative learning techniques, have been applied to solve a variety of signal processing-related problems. In this paper, we propose a first approach for using DDPM conditioned on anthropometric measurements to generate personalized Head-Related Impulse Response (HRIR), the time-domain representation of HRTF. The results show the feasibility of DDPMs for HRTF personalization obtaining performance in line with state-of-the-art models.

Cited by

• Head-Related Transfer Function Individualization Using Anthropometric Features and Spatially Independent Latent Representation
• Deep Learning for Personalized Binaural Audio Reproduction
• VR-PTOLEMAIC: A Virtual Environment for the Perceptual Testing of Spatial Audio Algorithms
• Low-Rank Adaptation of Deep Prior Neural Networks For Room Impulse Response Reconstruction

Related

• A Machine Learning Approach for Denoising and Upsampling HRTFs
• A Cross-Evaluated Database of Measured and Simulated HRTFs Including 3D Head Meshes, Anthropometric Features, and Headphone Impulse Responses
• Predicting Global HRTFs From Scanned Head Geometry Using Deep Learning and Compact Representations
• Head-Related Transfer Function Individualization Using Anthropometric Features and Spatially Independent Latent Representation
• Global HRTF Interpolation via Learned Affine Transformation of Hyper-conditioned Features