Abstract
People often mishear spoken utterances. Semantic context, background noise, and pronunciation clarity (how clearly words are pronounced) are potential causes of such misperceptions. The present study investigated the influence of these three factors on spoken word recognition. We took recordings of a female speaker and manipulated pronunciation clarity using Praat (Boersma & Weenink, 2015) by gradually changing the target word from 3 moras to 2 moras in 5 steps. Additionally, we made three semantic contexts by adding phrases that were meaningful for the 2 moras word, 3 moras word, or both. Finally, we presented the stimuli with cafeteria noise ratio( at three signal-to-noise ratios: ±0dB, -5dB, none). The results indicated that the effect of semantic context differs in degree depending on the speech clarity and signal-to-noise ratio.
