Dissertations and Theses @ UNI

Availability

Open Access Thesis

Keywords

Esophageal speech;

Abstract

This study was designed to investigate the perceivability of emotion in esophageal speech. Specifically, the questions under study were: (1) Are naive listeners able to accurately perceive emotional intention from auditory samples of esophageal speech? (2) Are the listeners' judgments of emotion from esophageal speech made with as much confidence as similar judgments made on normal laryngeal speech? A perceptual paradigm was utilized in which listeners made emotion judgments from speech samples of one laryngeal and one esophageal speaker. Two master audiotape recordings were constructed. Each contained a listening corpus of 75 randomly distributed grammatically complete, content-constant sentences. Each audiotape represented a speaker's portrayal of five emotions (angry, bored, happy, sad, and surprised) imposed onto five distinct sentences, with three replications of each token (5 x 5 x 3= 75). Thirty college students served as listening subjects. Each listener was required to: (1) determine the emotion being represented by each sentence token, making judgments on the basis of the five preselected emotion labels; (2) indicate the confidence with which the judgments of emotion were made. Evaluation of the results pertaining to Question 1 was based on a procedure which allowed a comparison of the observed accuracy attained by listeners with the accuracy levels expected by random guessing. The observed mean proportion of 0.38 (± 0.02) was clearly significant at the 99% confidence limits. The number of correct judgments of esophageal speech (865/2250 judgments) was approximately 22 standard deviations above the expected mean value (0.2 ± 0.01) as determined by a z-statistic of 21.87. Listeners were obviously able to accurately determine the emotion intended by the speaker at a significantly higher rate than is expected by random guessing. Question 2 involved the use of a five point confidence rating scale. Listeners were asked to rate their confidence in making emotion judgments according to the following scale: very confident, more than confident, confident, less than confident, and not confident. Scores were assigned to each of these confidence categories, with five assigned to ''very confident" and one assigned to each "not confident" rating. The group mean confidence rating scores were compared using a t-test. Results of the t-tes.t showed that there is a significant difference between confidence rating scores in normal and esophageal speech. Judgments of esophageal speech were made with significantly less confidence than were similar judgments of laryngeal speech. The study's conclusions are: (1) listeners are apparently able to identify, recognize, and analyze acoustic information related to emotion from audiotape recordings of esophageal speech; (2) listeners are more accurate and more confident in their judgments of emotion from normal speech.

Year of Submission

1978

Degree Name

Master of Arts

Department

Department of Speech Pathology and Audiology

First Advisor

Ralph Schwartz

Second Advisor

Roy E. Eblen

Third Advisor

Jack B. Yates

Comments

If you are the rightful copyright holder of this thesis and wish to have it removed from the Open Access Collection, please submit a request to scholarworks@uni.edu and include clear identification of the work, preferably with URL.

Date Original

1978

Object Description

1 PDF file (84 leaves)

Language

en

File Format

application/pdf

Download

Comments?

Share

COinS