Physiological Sensors for Speech Recognition

Abstract

Systems designers are expressing greater interest in speech-based user interfaces for a variety of civilian and military applications. Such interfaces provide hands free operation and a more natural way for humans to interact with systems. One difficulty with speech-based user interfaces is poor operation in noisy environments such as military operations. The Physiological Sensor, developed at ARL, is an example of an alternative sensor for automatic speech recognition This sensor detects speech by measuring acoustic signals through the speaker's skin. While the signal produced is not typical of that from an airborne acoustic microphone, the possibility exists for using this sensor as a microphone. We investigate several possible methods for using the Physiological Sensor as a microphone for automatic speech recognition.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 12, 2002
Accession Number
ADP014015

Entities

People

  • Francis Fisher
  • Mike Scanlon
  • Steve Chen

Organizations

  • United States Army Research Laboratory

Tags

Communities of Interest

  • Sensors

DTIC Thesaurus Topics

  • Acoustic Detectors
  • Acoustic Signals
  • Ambient Noise
  • Automated Speech Recognition
  • Computer Science
  • Frequency
  • Hidden Markov Models
  • Human-Computer Interfaces
  • Language
  • Markov Models
  • Microphones
  • Military Operations
  • Models
  • Physiology
  • Recognition
  • Training
  • User Interface

Readers

  • Sensor Fusion and Tracking Systems.
  • Speech Processing/Speech Recognition.
  • Systems Analysis and Design

Technology Areas

  • AI & ML