The Effects on Computer Recognition of Speech When Speaking through Protective Masks

Abstract

The purpose of this experiment was to investigate the effects on computer recognition of speech when speaking through aviator protective masks as compared to the standard boom-mounted microphone. Twelve Army aviators were tested with a 50-word vocabulary under three conditions: no mask (boom-mounted microphone), M24 aviator protective mask which has an internally mounted microphone, and the XM33 developmental aviator protective mask which has an externally mounted microphone. The conclusions were (1) of the two masks designs, the M24 mask with the internally mounted microphone provided better results with the speech recognition system and (2) although there was no significant difference in performance between the boom-mounted microphone and the protective masks, there was an indication of inconsistent performance obtained with the boom-mounted microphone which was not clearly understood and warrants further investigation.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
May 01, 1983
Accession Number
ADA129371

Entities

People

  • Frank J. Malkin

Organizations

  • United States Army Materiel Command

Tags

Communities of Interest

  • Air Platforms
  • Biomedical

DTIC Thesaurus Topics

  • Analysis Of Variance
  • Army Aircraft
  • Artificial Intelligence
  • Automated Speech Recognition
  • Computer Programming
  • Computer Programs
  • Computers
  • Engineering
  • Helicopters
  • Human Factors Engineering
  • Maryland
  • Microphones
  • Military Aircraft
  • Protective Masks
  • Recognition
  • Standards
  • Vocabulary

Readers

  • Aerospace Test and Evaluation
  • Materials Science
  • Speech Processing/Speech Recognition.

Technology Areas

  • AI & ML
  • AI & ML - Autonomous Systems