STUDIES IN THE PHONOLOGY OF ASIAN LANGUAGES. V. ACOUSTIC FEATURES IN THE MANNER-DIFFERENTIATION OF KOREAN STOP CONSONANTS.

Abstract

In Korean nine stop consonants--the aspirated bilabial, dental, and velar stops; the weak bilabial, dental, and velar stops; and the strong bilabial, dental, and velar stops--contrast with each other. In order to determine those acoustic features involved in the manner differentiation of these stops, a fairly large amount of data was collected, and a number of speech synthesis experiments were carried out using the tape cutting and splicing method. These studies revealed that aspirated stops are distinguished from weak and strong stops primarily by the timing of the voice onset. Aspirated stops were found to be 2.4 to 5.3 times longer than weak stops and even longer than this compared to strong stops. The cues for the distinction between weak and strong stops seem to be (1) the intensity build-up in the first few centiseconds of voicing following stop release, which is generally slower with weak stops than with strong stops and (2) the peak amplitude of the first period of voicing. These findings indicate that the difference between these stops is a function of the slope of the leading edge of the intensity contour during the first few centiseconds of voicing following the stop release. Relative to a given speaker, if the slope rises abruptly, the stop will be heard as strong, and if it rises gradually, the stop will be heard as weak. (Author)

Document Details

Document Type
Technical Report
Publication Date
Jul 15, 1967
Accession Number
AD0657384

Entities

People

  • Mieko S. Han
  • Raymond S. Weitzman

Organizations

  • University of Southern California

Tags

DTIC Thesaurus Topics

  • Amplitude
  • Consonants
  • Contrast
  • Intensity
  • Language
  • Leading Edges
  • Linguistics
  • Phonemes
  • Phonology
  • Social Sciences

Readers

  • Plasma Physics / Magnetohydrodynamics
  • Speech Processing/Speech Recognition.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems
  • AI & ML - Machine Learning Algorithms