Sona: Real-Time Multi-Target Sound Attenuation for Noise Sensitivity

2026-03-31

AI Summary

This work proposes a mobile real-time soundscape modulation system based on a target-conditioned neural network to address the challenge faced by noise-sensitive individuals, who often lose auditory awareness due to excessive suppression of ambient sounds by conventional noise reduction techniques. The system simultaneously attenuates multiple overlapping interfering sound sources while preserving sounds of user interest, overcoming the limitation of traditional approaches that support only single-target suppression. It enables real-time multi-target processing and allows users to dynamically expand the set of recognizable sound categories through in-situ audio examples without requiring model retraining. Experimental results demonstrate that the system effectively reduces interfering sounds under low-latency constraints, significantly enhancing user experience while maintaining situational auditory awareness.

Abstract

For people with noise sensitivity, everyday soundscapes can be overwhelming. Existing tools such as active noise cancellation reduce discomfort by suppressing the entire acoustic environment, often at the cost of awareness of surrounding people and events. We present Sona, an interactive mobile system for real-time soundscape mediation that selectively attenuates bothersome sounds while preserving desired audio. Sona is built on a target-conditioned neural pipeline that supports simultaneous attenuation of multiple overlapping sound sources, overcoming the single-target limitation of prior systems. It runs in real time on-device and supports user-extensible sound classes through in-situ audio examples, without retraining. Sona is informed by a formative study with 68 noise-sensitive individuals. Through technical benchmarking and an in-situ study with 10 participants, we show that Sona achieves low-latency, multi-target attenuation suitable for live listening, and enables meaningful reductions in bothersome sounds while maintaining awareness of surroundings. These results point toward a new class of personal AI systems that support comfort and social participation by mediating real-world acoustic environments.

Problem

• noise sensitivity
• real-time sound attenuation
• multi-target sound processing
• selective audio filtering
• acoustic environment mediation

Innovation

• multi-target sound attenuation
• target-conditioned neural pipeline
• real-time on-device processing
• user-extensible sound classes
• soundscape mediation