This paper investigates the use of multi-distribution deep neural networks (MD-DNNs) for automatic lexical stress detection and pitch accent detection, which are useful for suprasegmental mispronunciation detection and diagnosis in second-language (L2) English speech. The features used in this paper cover syllable-based prosodic features (including maximum syllable loudness, syllable nucleus du...

This paper describes an eye tracking experiment to study the processing of diphone synthesis, unit selection synthesis, and human speech taking segmental and suprasegmental speech quality into account. The results showed that both factors influenced the processing of human and synthetic speech, and confirmed that eye tracking is a promising albeit time consuming research method to evaluate synt...

This study addresses turn taking in everyday Mandarin conversation from a phonetic perspective, in particular, suprasegmental analysis. The acoustic data are based on 103 exchanges from two conversations – one structured and one free. The results show that turn final cues are mostly signaled via various falling intonation patterns in Mandarin conversation. Some anomalies exist and these are ana...

To guarantee a good quality of synthesized speech in a rule-based synthesis, detailed and accurate segmental and suprasegmental rules for a given text are necessary. Especially, much work on the prosodic structure is needed. The goal of this study is to examine roles of syntactic information, intonation pattern, and phrase final lengthening in formation of prosodic units or perception of a phra...

During acoustic communication in humans, information about a speaker's emotional state is predominantly conveyed by modulation of the tone of voice (emotional or affective prosody). Based on lesion data, a right hemisphere superiority for cerebral processing of emotional prosody has been assumed. However, the available clinical studies do not yet provide a coherent picture with respect to inter...

In light of the fact that L2 pronunciation errors are often caused by the transfer of well-established L1 sound systems, this paper examines some of the characteristic phonological differences between Japanese and English. Comparing segmental and suprasegmental aspects of both languages, this study also discusses several problematic areas of pronunciation for Japanese learners of English. Based...