Recognition of spoken vowel sounds using a hybrid computer simulation, 1971

Item — Multiple Containers

Identifier: b2088576

Staff Only

Scope and Contents

From the Collection:

The collection consists of theses written by students enrolled in the Monmouth College and Monmouth University graduate Electronic Engineering programs. The holdings are bound print documents that were submitted in partial fulfillment of requirements for the Master of Science degree.

Dates

Creation: 1971

Creator

Schweitzer, Philip F. (Author, Person)
Drucker, Harris (1943-2024) (Thesis advisor, Person)

Conditions Governing Access

All analog collection holdings are limited to library use only.

Researchers seeking to photocopy collection materials must complete an Application to Photocopy Form.

In some cases, photocopying of collection materials may be performed by the Monmouth University Library staff.

The Monmouth University Library reserves the right to limit or refuse duplication requests subject to the condition of collection materials and/or restrictions imposed by the collection creators or by the United States Copyright Act.

Permission to examine, or copy, collection materials does not imply permission to publish or quote. It is the responsibility of the researcher to obtain such permissions from both the copyright holder and Monmouth University.

Extent

2 Items (print book) : 45 pages ; 8.5 x 11.0 inches (28 cm).

Language of Materials

English

Additional Description

Abstract

This paper will be an investigation into a hybrid computer simulation to recognize four (A, E, I, O) spoken vowels. The hybrid simulation makes available two modes of operation to achieve the recognition process. A learning mode will provide the algorithms required to categorize each vowel into parameters that can later be accessed by the second or recognition mode. The recognition mode takes the input vowel sound and correlates the recognition mode parameters against the learning mode parameters. When a best fit of parameters is completed, the recognition process is terminated.

The hybrid computer selected was the Electronic Associates 690 computing system. The analog computer subsystem provides the necessary parallel processing of analog signals, and the digital computer subsystem provides the necessary digital processing of the recognition parameters. Upon completion of the hybrid simulation, it is expected that an actual recognition system could be built from the simulated design.

The premise of this study was to propose a voice recognition process that is limited in numbers of input sounds, requires a small digital core size, is high in recognition probability and still be low in cost.

At the present time, the proposed system has a recognition probability of 91 percent on the four vowels. Problems still exist in areas as computer environment background noise levels and loudness of speech. Corrections to the problem areas should increase the probability of recognition beyond that of the present 91 percent recognition level.