In this paper, a data pruning approach is presented for building acoustic unit inventory for syllable-based concatenative embeddable Chinese TTS system. A 3-portion segmentation of a syllable is proposed based on the nature of voiced/unvoiced structure of Chinese syllable. Individual factorial acoustic measurement of syllable is used to calculate the penalty of perceptual unsatisfactory for con...

This paper reports the results of syllable recovery from speech using an articulatory model of the syllable. The contribution of syllable recovery to the overall process of speech recognition is discussed and speech recognition results are presented.

27 28 Behavioral variability is important for motor skill learning but continues to be 29 present and actively regulated even in well-learned adult behaviors. In adult songbirds, 30 two types of song variability can persist and are modulated by social context: variability 31 in syllable structure and variability in syllable sequencing. The degree to which the 32 control of both types of adult v...

Fast speech contains pronunciation reductions to different extents. This paper examines perceptually identified syllable contractions with a corpus of spontaneous Mandarin data. By means of quantitative analyses and acoustical measurements, we find that syllable contraction is a continuous process with a predictable target syllable. Our statistic analyses show that contraction forms of a given ...

The whole, that which is compounded out of something, is one, not like a heap, but like a syllable. Now the syllable is not its elements; ba is not the same as b and a, nor is flesh the same as fire and earth (for when these are separated the wholes, i. e. the flesh and the syllable, no longer exist, but the elements of the syllable exist, and so do fire and earth); the syllable, then, is somet...

An approach to automatic detection of syllable structure is presented. We demonstrate a novel application of EM-based clustering to multivariate data, exempliied by the induction of 3-and 5-dimensional probabilistic syllable classes. The 3-dimensional models were subjected to a pseudo-disambiguation task, the result of which shows that the onset is the most variable, or least predictable, part ...

This paper introduces a Syllable HMM based Mandarin TTS system. 10-state left-to-right HMMs are used to model each syllable. We leverage the corpus and the front end of a concatenative TTS system to build the Syllable HMM based TTS system. Furthermore, we utilize the unique consonant/vowel structure of Mandarin syllable to improve the voiced/unvoiced decision of HMM states. Evaluation results s...