Deriving disyllabic word variants from a Chinese conversational speech corpus.

Motivated by the quasi-categorical reduced forms of disyllabic words produced in Chinese conversational speech, a frequency-based selection procedure of typical pronunciation by disyllabic word type and reduction degree is proposed in this paper. This variant-selection algorithm utilizes techniques of free phone recognition and phonetic similarity score calculation to account for Chinese syllable structure. Four reduction types are suggested by considering the presence of a within-word syllable boundary: Citation form-like reduction, marginal segment deletion, nuclei merger, and syllable merger. The results show that the most frequent reduction types for disyllabic words in Chinese conversation are citation form-like reduction and syllable merger. In particular, high-frequency disyllabic words preferentially take the extreme syllable-merger form. As shown in the analysis, segmental reduction in Chinese disyllabic words is morphology-dependent. It is also related to the prosodic position at which a disyllabic word is produced as well as the temporal quality of the word. Finally, in the automatic speech recognition experiments, the performance was improved by adding a small number of variants selected by the algorithm to the pronunciation dictionary of the system.

Acoustic characteristics of vowels and plosives/affricates of Mandarin-speaking hearing-impaired children.

This article presents the results of an acoustic analysis of vowels and plosives/affricates produced by 45 Mandarin-speaking children with hearing impairment. Vowel production is represented and categorized into three groups by vowel space size calculated with normalized F1 and F2 values of corner vowels. The correlation between speech intelligibility and language abilities assessed by the level of word comprehension and the complexity of sentence structure is statistically significant. Vowel space grouping is correlated with speech intelligibility and spike percentage of plosives/affricates production. The generalized linear model analysis also shows that the level of word comprehension and the degree of hearing loss are the two most significant factors in predicting speech intelligibility. The statistical results suggest that the interplay of acoustic characteristic and speech ability is complex.