Segmentation of Speech Sounds
D. R. Reddy · The Journal of the Acoustical Society of America · 1966
A program for segmentation of an acoustical continuum of speech sounds into discrete parts suitable for further analyses is described. The speech wave is read into the computer using an A-D converter. Analysis is performed directly on the acoustic waveform, using an IBM-7090. The pattern-recognition techniques used to segment the acoustic waveform into sustained and transitional parts are discussed. Some results obtained by the computer program are given.