PATTERNS OF CONFUSION MADE BY A MODEL OF DOUBLE-VOWEL IDENTIFICATION: A COMPARISON WITH HUMAN DATA
SJ MAKIN, GJ BROWN · 2024
BackgroundIn a real acoustic environment, signals reaching the ears of a listener are usually mixtures of several sound sources.In order to make sense of such complex auditory scenes .the perceptual system must analyse the mixture to recover the constituent sources.This problem has come to be known as Auditory Scene Analysis" (ASA).[4].In ASA.acoustic mixtures are thought to be decomposed into elements and groups of these elements recombined to form auditory streams .on the basis of the likelihood of them having arisen from the same source.There are numerous cues which have been shown to play a part in this process.such as onset/offset synchrony.common AM. common spatial origin.etc.A particularly important factor is fundamental frequency (f0).Harmonically related components tend to perceptually fuse .whereas differences in ll) promote segregation.There have been numerous studies showing that listeners are better