Intelligibility of voiced and whispered speech in noise in listeners with and without musical training

Musicians have been shown to exhibit more robust neural coding of periodicity than non-musicians; they have also been reported to exhibit an advantage in understanding speech in noise. This study tested the hypothesis that the musicians' speech intelligibility advantage arises from more efficient coding of voiced (periodic) speech. This was tested by comparing intelligibility of normal speech in noise with that of whispered (unvoiced) speech in musicians and non-musicians. Listeners with less than 2 years of formal musical training were categorized as non-musicians; listeners who began musical training before age 10 and who currently play more than 10 h/wk were included as musicians. Listeners heard grammatically correct nonsense sentences that were either 1. Voiced, 2. Whispered, or 3. Whispered with subband amplitude distributions matched to the voiced speech. Masking noise was either continuous or gated with a 16 Hz, 50% duty cycle. In contrast to the earlier study, preliminary data suggest no advantage for musicians over non-musicians in understanding voiced or whispered speech in either continuous- or gated-noise conditions. The results suggest that more investigation is needed to fully understand the nature of auditory and speech processing advantages imparted by musical training.