Abstract-In this paper, we investigated the speech perception under effects of coarticulation within syllables, including estimating the nuclei and coarticulated transition intervals of phonemes, and investigating the perception of nuclei and coarticulated transition intervals of phonemes within syllables. To complete these objectives, we firstly proposed the folded spectral transition measure to estimate the boundary points between the nuclei interval and coarticulated transition intervals. Then, we conducted experiments to investigate the syllable identification and syllable quality evaluation under effects of coarticulation. The experimental results show that human can identify the syllables with the retaining intervals equal or wider the coarticulated transition intervals. Besides, the experimental results show that the nuclei intervals are still important for speech quality perception; the quality of truncated syllables retaining the two phoneme nucleus and coarticulated transition interval between the two phonemes within syllable is equivalent with the quality of original two-phoneme syllables.

No abstract

Coarticulation is a phonological phenomenon, always occurring in all sequences of sounds not separated by pauses. Analyses in coarticulation of speech reveal that articulation targets are incomplete in neutralized sound, and it is difficult to estimate incomplete articulatory targets of phonemes due to their sensitivity. In this paper, we firstly proposed a acoustical model of coarticulation of phonemes within syllables. After that, we investigated the stability of spectral targets affected by the vowel neutralization phenomenon. The experimental results show that the proposed coarticulation model decomposed speech into context-insensitive event targets, which are close with articulatory targets, and the context-sensitive event functions, which closely represent the movements between the adjacent targets. In addition, the PLP-LSF was shown as a stable spectral target against vowel neutralization phenomenon in our proposed coarticulation model.