Using virtual sound reproduction for studying L1/L2 speech perception in varying acoustic environments

Abstract

Studying the effect of varying acoustic environments on speech perception is often challenging due to the logistical difficulties in moving participants undertaking listening tests between environments as well as lack of controllability and reproducibility in terms of the acoustics of the test environments. In addition, when the study requires to compare first and second language listeners, it often becomes infeasible due to the availability of participants with required language background in a single geographical location where the study is conducted. Virtual sound reproduction could address the challenge thanks to its ability to reproduce the acoustics of arbitrary spaces at multiple geographic locations in a controllable manner. However, it has not been studied well if the results collected using virtual sound reproduction would be a valid alternative to the results collected in real spaces. This talk will introduce a recent study that investigated the difference of speech perception in varying acoustic environments between first and second language New Zealand English listeners using virtual sound reproduction technology. The talk will particularly focus on how the results collected under virtual acoustic environments assimilates to that collected in the original real acoustic environments between first and second language listeners.