How do voices become familiar? Speech intelligibility and voice recognition are differentially sensitive to voice training
Emma E. Holmes, Grace To, Ingrid S. Johnsrude · 2020
When people listen to speech in noisy places, they can understand more words spoken by someone familiar, such as a friend or partner, than someone unfamiliar. Yet, we know little about how voice familiarity develops. We exposed participants to three voices for different lengths of time (speaking 88, 166, or 478 sentences). Trained voices were recognizable and were more intelligible when presented with a competing talker than novel voices. However, recognition and intelligibility improved at different rates with longer exposures, demonstrating that recognition and intelligibility of familiar voices are partially dissociable. Whereas recognition was similar for all trained voices, intelligibility was best for the most extensively trained voice. The speech intelligibility benefit for the most extensively trained voice (10–15%) is as large as that reported for naturally highly familiar voices (friends and spouses)—demonstrating that intelligibility of a voice can be improved substantially after only an hour of training.