Person-specific Automatic Speaker Recognition: understanding the behaviour of individual speakers for applications of ASR

Posted on 21 March 2022

A new linguistic project on automatic speaker recognition has been awarded a major ESRC grant

The Department of Language and Linguistic Science is delighted to announce that Dr Vincent Hughes (PI), Prof. Paul Foulkes (Co-I) and Phil Harrison (Co-I) have been awarded an Economic and Social Research Council grant for the project “Person-specific Automatic Speaker Recognition: understanding the behaviour of individual speakers for applications of ASR” (ES/W001241/1; £1,012,570).The project is due to start in summer 2022 and will run for 3 years. The project is made possible by collaborations between the department and Oxford Wave Research and two project partners including the Netherlands Forensic Institute.

Automatic speaker recognition (ASR) software processes and analyses speech to make decisions about whether two voices belong to the same or different individuals. Such technology is becoming an increasingly important part of our lives; used as a security measure when gaining access to personal accounts (e.g. banks), or as a means of tailoring content to a specific person on smart devices. Around the world, ASR systems are commonly used for investigative and forensic purposes, to analyse recordings of criminal voices where identity is unknown.

The overarching aim of this project is to systematically analyse the factors that make individuals easy or difficult to recognise within automatic speaker recognition (ASR) systems. By understanding these factors, we can better predict which speakers are likely to be problematic, tailor systems to those individuals, and ultimately improve overall accuracy and performance. We will use innovative methods and large-scale data, uniting expertise from linguistics, speech technology, and forensic speech analysis, from the academic, professional, and commercial sectors.

“We are delighted to have been awarded this grant and to be able to collaborate with our partner organisations. It gives us the opportunity to do large-scale systematic testing with automatic systems which will have considerable benefit for forensics and other applied fields.” - Dr Vincent Hughes, Department of Language and Linguistic Science

More information can be found on the project website: Person-specific automatic speaker recognition