The role of timing cues in speech perception

Lacocque, Jeremy André

Permalink

https://hdl.handle.net/2142/49322 Copy

Description

Title

The role of timing cues in speech perception

Author(s)

Lacocque, Jeremy André

Issue Date

2011

Director of Research (if dissertation) or Advisor (if thesis)

Wickesberg, Robert E.

Department of Study

Psychology

Discipline

Psychology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

B.S.

Degree Level

Thesis

Keyword(s)

Speech perception

Language

en

Abstract

Theories of speech perception posit that recognizing key frequencies in speech sounds is the primary way we recognize speech. These theories, however, do not explain how we understand speech with shifted, key-frequency information such as whispered speech. In an effort to determine characteristics of speech vital to speech recognition beyond just frequency, timing patterns were examined in auditory nerve fibers’ action potentials throughout the presentation of speech sounds. Since chinchillas have similar peripheral auditory systems as do humans, they were used in the study. Eighteen chinchillas were anesthetized and their auditory nerves exposed to record their activity. Alveolar stop consonants speech sounds /pap/, /paep/, /bab/, and /baeb/ were presented to chinchillas’ peripheral auditory systems by a female speaker in two ways: normally spoken, and whispered, each at three different intensities. Despite the fact that whispered speech contains different frequency information than does normally spoken speech, the experiment is expected to reveal that each consonant evokes a similar timing pattern in the auditory nerve fibers across method of speech production, but distinct between consonants, regardless of whether the speech sound was whispered or normally spoken. Timing patterns in the auditory nerve fibers’ responses were gathered and analyzed by pooling the responses from each auditory nerve fiber from each chinchilla to create an ensembleaverage, time histogram. For each speech sound, a specific number and pattern of peaks in action potentials at certain points in time appeared in each histogram, sometimes independent of method of speech production, whispered or normally spoken. Certain histograms revealed similar timing patterns encoded in the chinchillas peripheral auditory system for each pair of speech sounds, showing that timing information is consistent between different forms of speech production and therefore might be an important cue in how we process speech.

Type of Resource

text

Permalink

http://hdl.handle.net/2142/49322

Copyright and License Information

Copyright 2011 Jeremy A. Lacocque

Owning Collections