Vocabulary Specification for Automatic Speech Recognition in Aircraft Cockpits.

Abstract

The general focus of this research was to design a communication media (a vocabulary) that is advantageous to both machine recognition and human production of speech events. The problem was analyzed from a human factors perspective that centered upon the man-computer dialogue (interaction) required for cockpit application of ASR. The results indicated that phrase familiarity and stimulus familiarity had major impact on the learning and utilization of the phrases in the paired-associate task. Phrase length and meaningfulness did not appear to differentially affect either the learning or utilization of the paired associate. In addition, pretraining of stimulus familiarity did not seem to result in improved performance. Acoustic lexical confusability also was discussed in general methodological terms. The results of the study were interpreted in terms of a contextualist viewpoint with the necessity of a broader contextual manipulation being pointed out as a requirement for further research.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Aug 31, 1979
Accession Number
ADA073703

Entities

People

  • Catherine Meyn
  • Elaine Regelson
  • John J. Petersen
  • Nancey Lee
  • William Satzer

Tags

Communities of Interest

  • Air Platforms
  • Weapons Technologies

DTIC Thesaurus Topics

  • Accuracy
  • Aircrafts
  • Algorithms
  • Analysis Of Variance
  • Artificial Intelligence
  • Automated Speech Recognition
  • Automatic Direction Finders
  • Cognition
  • Cognitive Science
  • Control Systems
  • Information Exchange
  • Information Processing
  • Language
  • Natural Language Processing
  • Psychology
  • Recognition
  • Tacan

Fields of Study

  • Psychology

Readers

  • Brain and Cognitive Science; Experimental Psychology; Cognitive Neuroscience
  • Speech Processing/Speech Recognition.
  • Systems Analysis and Design

Technology Areas

  • AI & ML
  • AI & ML - Machine Translation