Date of Award

5-1-1970

Document Type

Thesis

Degree Name

Master of Science (M.S.) in Speech

Department

Speech and Hearing Sciences

Physical Description

1 online resource (4, vii, 104 leaves)

Subjects

Compressed speech

DOI

10.15760/etd.682

Abstract

Speech compression is a method of electromyographically reducing the speech signal, resulting in a decrease in the time required for presentation of a message by increasing the number of words per minute. Research in the area of speech compression has to this time been focused on determining the listener response by means of comprehension testing. While comprehension testing uncovers the individual's ability to resolve, organize, and recall a message at accelerated rates, it gives no information about the listener's psychophysical response to compression. Does the individual find listening to compression to be tension-provoking or stressful? Or can a listener accommodate accelerated speech without stress? The purpose of this study was to determine the individual's acceptance of compressed speech. "Acceptability" was operationally defined as finding no significant differences between tension levels produced while listening to normal speech rate as compared to tensions produced while listening to compressed speech rates. Tension levels were determined by an electromyograph which measures the amount of tension within a muscle. The specific hypotheses researched were: Hypothesis 1. Sampled muscular tension levels will differ significantly with Order of Presentation. Hypothesis 2. Sampled muscular tension levels will differ significantly with Rate of Presentation. Hypothesis 3. Sampled muscular tension levels will differ significantly between Males and Females. Hypothesis 4. Sampled muscular tension levels will show significant interaction between Order of Presentation and Rate of Presentation. Hypothesis 5. Sampled muscular tension levels will show significant interaction between Order of Presentation and Sex of the Listener. Hypothesis 6. Sampled muscular tension levels will show significant interaction between Rate of Presentation and Sex of the Listener. Hypothesis 7. Sampled muscular tension levels will show significant interaction among Order of Presentation, Rate of Presentation, and Sex of the Listener. A 6 x 3 x 2 factorial de sign was constructed to examine muscular tension levels. The controlled factors were Order of Presentation, Rate of Presentation, and Sex of Listener. Rate of Presentation included three speech sample s: normal (189 words per minute), speech compressed 25% (252 words per minute), speech compressed 50% (378 words per minute). These rates were arranged into six presentational patterns, composed of the three rates varied by its position within the sequence. Each Order of Presentation was presented to a male and a female listener, resulting in twelve subjects. The dependent variable was muscular tension within the trapezius muscle produced by the listener as he heard the experimental presentation. The sample was selected from speech students enrolled in Fundamentals of Speech at Portland State University. Winter Term, 1970. Apparatus used to collect tension level readings were: bipolar surface electrodes, differential amplifier, oscilloscope, stereo tape recorder, and multi-functioned voltmeter. The muscle tension levels were analyzed statistically by the three -factor analysis of variance test for significance. The finding of no significant differences at the. 05 level of confidence for any of the experimental conditions warranted rejection of the research hypotheses. The conclusion drawn from this finding was that compression within the limits used in this study is an acceptable mode of presentation based on the criterion that tension levels produced by compression differ only by chance from tension produced while listening to normal speech. It seems feasible, therefore, to substitute compression for normal speech in conditions where usage would be desirable.

Description

Portland State University. Dept. of Speech and Hearing Sciences

Persistent Identifier

http://archives.pdx.edu/ds/psu/9388

Share

COinS