Logo image
Feature selection via dynamic programming for text-independent speaker identification
Journal article

Feature selection via dynamic programming for text-independent speaker identification

R Cheung and B Eisenstein
IEEE transactions on acoustics, speech, and signal processing, v 26(5), pp 397-403
Oct 1978

Abstract

Acoustic testing Cepstral analysis Costs Dynamic programming Humans Libraries Loudspeakers Speaker recognition Speech
Dynamic programming is applied to the selection of feature subsets in text-independent speaker identification. Each feature is long-term averaged in order to reduce its variability to text information. The resulting subset of features shows a lower average identification error in comparison to that of the "knock-out" strategy, the cepstral coefficients, and the PARCOR coefficients.

Metrics

10 Record Views
33 citations in Scopus

Details

Logo image