Article in JASA, Acoustical Society of America

Monday 22 Mar 21


Axel Ahrens
Academic employee
DTU Health Tech
This month, the paper "Speech intelligibility with various head-related transfer functions: A computational modelling approach" is featured in JASA, the Journal of the Acoustical Society of America. By Axel Ahrens, Maria Cuevas-Rodriguez and W. Owen Brimijoin.

The intelligibility of speech is commonly improved when the target voice is spatially separated from interfering noises. The reason is a spatial filter due to the anatomy of our head and ears. This spatial filter is also referred to as the head-related transfer function (HRTF). We hypothesized that the advantage in speech intelligibility due to spatial separation of the sources is dependent on the individual HRTF and consequently on the anatomy of each person.
In this project in collaboration with partners from the University of Malaga (Spain) and Facebook Reality Labs (USA), we employed a computational speech intelligibility model that has previously been shown to correctly predict speech intelligibility in certain conditions. Using the model, we showed that the speech intelligibility cues are largely different for various HRTFs. Thus, some listeners might have speech intelligibility advantages over others. However, it remains unclear if these differences can also be found in real-life scenarios or if other factors contribute to a larger extend.
Following, one might choose the HRTF that leads to the largest speech intelligibility advantage when virtually separating sources. For example, spatially separating the voices in video chat applications such as Skype or Zoom might improve the communication experience.
Link to Open Access article

Published Online: 02 March 2021, Volume 1, issue 3

Illustration: Colourbox