Human listeners can distinguish between languages of different rhythmic classes (e.g. stressand syllable-timed languages). The present study investigated the role of speech rate in this process. Acoustic data suggests (experiment I) that speech rate can distinguishes as reliable between stressand syllable-timed languages as previously proposed correlates of speech rhythm (%V, VarcoC and nPVI). ...

A real-world speech recognition system encounters several speaking styles and speaking rates and its accuracy depends highly on the speaking rate, i.e., degrades sharply with very fast or very slow speech (including hyperarticulated speech) In this paper, we propose a generic modeling scheme to capture a range of speaking rates from very slow to very fast with the use of decision trees. This ap...

Objectives: One of the most effective methods to describe speech disorders is the measurement of speech intelligibility. The speech intelligibility indicates the extent of acoustic signals that correctly speaker produces and hearer receives. The purpose of this study was to investigate the speech intelligibility in the Persian children with Down syndrome, age range was 3 to 5 years, who had spo...

This paper details two perception experiments based on synthetic British English obtained with CART models predicting phone durations in slow speech from normal speed speech. Speech rate and naturalness were assessed by 6 English natives. Synthetic slow speech was rated as both slower and less natural than natural slow speech; however, the insertion of the pauses produced in natural slow speech...

A variety of methods are used for speaker adaptation in speech recognition. In some techniques, such as MAP estimation, only the models with available training data are updated. Hence, large amounts of training data are required in order to have significant recognition improvements. In some others, such as MLLR, where several general transformations are applied to model clusters, the results ar...