A comparison of several speech-spectra classification methods

An important consideration in speech processing involves classification of speech spectra. Several methods for performing this classification are discussed. A number of these were selected for comparative evaluation. Two measures of performance-accuracy and stability-were derived through the use of...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on acoustics, speech, and signal processing Vol. 24; no. 4; pp. 289 - 295
Main Authors Silverman, H., Dixon, N.
Format Journal Article
LanguageEnglish
Published IEEE 01.08.1976
Subjects
Acoustic signal processing
Automatic speech recognition
Classification algorithms
Frequency
Performance evaluation
Signal processing algorithms
Speech processing
Stability
Steady-state
Vectors
Online AccessGet full text

Cover

More Information
Summary:An important consideration in speech processing involves classification of speech spectra. Several methods for performing this classification are discussed. A number of these were selected for comparative evaluation. Two measures of performance-accuracy and stability-were derived through the use of an automatic performance evaluation system. Over 3000 hand-labeled spectra were used. Of those evaluated, a linearly mean-corrected minimum distance measure, on a 40-point spectral representation with a square (or cube) norm was consistently superior to the other methods.
ISSN:0096-3518
DOI:10.1109/TASSP.1976.1162814