Intelligibility of dysphonic speech in primary schools

Abstract

School children need clear auditory signals and low background noise to learn. When classroom acoustics are poor, teachers often compensate by raising their voices, usually with limited effect against background noise, and, long-term, this makes vocal overuse the primary cause (60%) of the high prevalence of voice problems in teachers. Speech intelligibility tests were performed in primary schools with normal hearing students using words produced by an actor with normal voice quality and simulating a dysphonic voice. The speech was played by a Head and Torso Simulator. Artificial classroom noise and classrooms with different reverberation times were used to obtain a range of Speech Transmission Index from 0.2 to 0.7 (from bad to good). Results showed a statistically significant decrease in intelligibility when the speaker was dysphonic with a maximum of 15% intelligibility loss. This study extends an important pairing of problems related to student learning: classroom acoustics and teachers with voice disorders. It provides important insights into the enormous variability in speech intelligibility in classrooms by characterizing students’ intelligibility when students receive degraded auditory input. The degraded auditory input results from the intersection of classroom acoustics and poor teacher voice quality