Effects of speech and sung speech training on speech prosody production by trilingual children with autism spectrum disorder

Speech prosody can be used to distinguish old (topic) versus new (focus) information and rejecting incorrect alternative statement. It has been reported that children with autism spectrum disorder (ASD) may show abnormal prosody of either monotonic or exaggerated intonation and some of them may fail to mark focus. Though speech and musical training have shown to improve speech production by ASD children, no specific training methods have been proposed to improve the use of speech prosody to mark focus and few studies investigated tonal language speakers. We aim to test (1) whether Cantonese-speaking ASD children fail to mark focus in their native tonal language (2) whether trainings may improve the speech prosody processing (3) whether sung speech training is more effective than speech training? We recruited two training groups of Cantonese-speaking ASD children, a control group of ASD children and TD children. In the training tasks, we focused on improving the mapping between the acoustic cues and information structure. Our pilot results showed that speech and musical training improved the use of prosodic cues such as intensity and f0 in marking focus across various positions. However, ASD children may have difficulties in integrating all the prosodic cues across conditions.