Project outline
This project aims to create a new generation of personalised headphone audio technology that adapts intelligently to an individual’s unique hearing profile and listener preferences. All users, whether they have hearing loss or not, want an effortless, high-fidelity listening experience.
Project Partners
You will be working closely with Sonos and using the latest technology, and acoustics facilities at the University of Salford, to develop a set of psychoacoustic measures for understanding users better and improving audio processing to enhance the listening experience.
Further project information
Moving beyond the static presets seen often in consumer products, this project will develop a sophisticated framework for modelling hearing and listener preference to deliver personalised headphone audio. The goal is to deliver a perfectly balanced sound output, optimised for preference such as speech intelligibility, music enjoyment, or spatial immersion. The research will specifically address complex auditory characteristics and diverse listening needs such as hearing loss, loudness recruitment and deficits in temporal processing.
The core objective is to create predictive, individual models of the users’ hearing and listener preferences. This will be achieved by developing a suite of efficient psychoacoustic measures that go beyond simple auditory thresholds. These tests will probe spatial audio perception, loudness, temporal acuity, and speech-in-noise abilities. We will employ a “user-in-the-loop” optimisation strategy, where interactive tests gather rich data while simultaneously engaging the user in their own auditory discovery process.
The data will be used to train a personalised computational model that can predict the preferred audio processing strategies for that specific listener, given the listening context. The resulting low-latency model will be able to do real-time adaptive processing on the device to enhance the listening experience. The results will also be useful to inform the user about potential concerns about their hearing health.
Subject Areas
- Acoustics
- Acoustical Engineering
- Electronic Engineering
- Mathematics
- Machine Learning
- Artificial Intelligence
- Physics
Required qualifications/skills
Essential
- You must have or are expected to achieve a first-class or 2:1 Honours degree or Master’s degree in Acoustics, Engineering, Physics, or a related field
- Experience with acoustics and psychoacoustics, audio digital signal processing, mathematics, and some understanding of human hearing. Strong analytical and quantitative research skills.
- Excellent written and verbal communication skills, with the ability to present complex information clearly and concisely.
Desirable
- Experience in Machine Learning and optimisation algorithms, user research and software engineering.
Student
Commencing in October 2026