Listening in Difficult Conditions
Modelling Human Behaviour, Attention and Listening Effort in Conversation
- Project Supervisor: Prof Jon Barker
- Partner: Mitsubishi Electric Research Laboratories (MERL)
- Host: University of Sheffield
- Project Reference: Sh.27.1
- Mode of Attendance: Full-time only
Funded EPSRC CDT studentship for 2027 entry.
Application Deadline: 11 January 2027, 10:00 GMT
Project Outline
Difficult listening conditions can change how we take part in conversations: where we look, how we orient ourselves, when we speak and whether we ask someone to repeat themselves. This project asks whether such behaviour can reveal when listeners are following a conversation successfully and when they are struggling.
Further project information
The research will combine analysis of multi-person conversational recordings, statistical and machine-learning models, and controlled experiments. It will begin with the ECHI dataset, using its audio, head-tracking and speaker-activity data to investigate relationships between listening conditions and behaviour. Experiments will test promising findings and relate behaviour to comprehension and perceived effort.
In collaboration with MERL, the project aims to inform future audio systems that respond to listeners' needs. See the full Sound Futures project advert for further information.
Required qualifications/skills
A relevant undergraduate degree at 2:1 or above, or a master's degree, in a quantitative discipline. Useful experience includes programming, statistics or machine learning, audio processing, or behavioural data analysis; applicants are not expected to have all these skills.
How to Apply
Application Deadline: 11 January 2027, 10:00 GMT
This four-year studentship is funded through the EPSRC Centre for Doctoral Training in Sustainable Sound Futures, with tuition fees, a stipend and research training support. For current funding details and application instructions, visit the Sound Futures CDT application page. Applications are handled by the CDT through the University of Salford; the project is based at the University of Sheffield.