AN EXPERIMENT IN THE AUTOMATIC RECOGNITION OF A SPEAKER,
Abstract
The author describes a method for selecting key words which best display the individual characteristics of a speaker and presents general requirements for such words. Final selection is made on the basis of actual recording and analysis. The speech signal at the output of the analyzer is determined according to frequency, time, and amplitude. For the computer input, the signals are quantized in time and amplitude. Further processing is based essentially on these two quantities. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Jun 20, 1968
- Accession Number
- AD0682818
Entities
People
- G. S. Ramishvili
Organizations
- National Air and Space Intelligence Center