Segmentation of Malay Syllables in Connected Digit Speech Using Statistical Approach

摘要

This study present segmentation of syllables in Malay connected digit speech.Segmentation was done in time domain signal using statistical approachesnamely the Brandt?￠a??a?￠s Generalized Likelihood Ratio (GLR) algorithm andDivergence algorithm. These approaches basically detect abrupt changes ofenergy signal in order to determine the segmentation points. Patterns used in thisexperiment are connected digits of 11 speakers spoken in read mode in labenvironment and spontaneous mode in classroom environment. The aim of thisexperiment is to get close match between reference points and automaticsegmentation points. Experiments were conducted to see the effect of number ofthe auto regressive model order p and sliding window length L in Brandt?￠a??a?￠salgorithm and Divergence algorithm in giving better match of the segmentationpoints. This paper reports the finding of segmentation experiment using fourcriterions ie. the insertion, omissions, accuracy and segmentation match betweenthe algorithms. The result shows that divergence algorithm performed onlyslightly better and has opposite effect of the testing parameter p and L comparedto Brandt?￠a??a?￠s GLR. Read mode in comparison to spontaneous mode has bettermatch and less omission but less accuracy and more insertion.