Speech Controlled Radio Channel Selector.

Abstract

The report describes the design and development of a breadboard model of a radio channel selector activated by spoken call signs. The breadboard model was designed to be capable of operating with speech input in a typical aircraft environment. It is capable of recognizing with high accuracy a set of six words as spoken by six talkers. The speech used to test the performance of the equipment was recorded in a simulated aircraft environment with a noise level of 107 dB to approximate the noise conditions in the cockpit of an F4B aircraft in a holding configuration. The equipment is capable of direct interfacing with an aircraft radio type AN/ARC-144. The frequency of the radio, as well as the transmit/receive function, is controlled by voice command. With the aid of an airborne version of such a speech controller, it would be possible for a pilot to automatically tune the radio to a desired frequency by speaking the appropriate call sign into his microphone. The speech controlled radio channel selector responded with an accuracy of 91 percent to three repetitions of each of six call signs as spoken by five talkers in the noise background. Recognition circuitry was designed to recognize six selected call signs with the capability of expansion to twelve call signs designed into the equipment. (Author)

Document Details

Document Type
Technical Report
Publication Date
Oct 01, 1971
Accession Number
AD0888850

Entities

People

  • Jerry R. Richards
  • Phillips B. Scott

Tags

Communities of Interest

  • Air Platforms

DTIC Thesaurus Topics

  • Accuracy
  • Airborne
  • Aircrafts
  • Breadboard Models
  • Channel Selectors
  • Communication Equipment
  • Electrical Equipment
  • Environment
  • Frequency
  • Mechanical Equipment
  • Microphones
  • Models
  • Recognition
  • Vehicle Equipment
  • Vehicles

Readers

  • Computer Science/Computer Engineering/Data Science/Digital Signal Processing.
  • Radio communications and signal processing.
  • Speech Processing/Speech Recognition.