Accession Number : AD0635488

Title :   MODEL 245C SIMORAMIC DOUBLE-SPECTRUM ('CEPSTRUM') ANALYZER FOR EXTRACTION OF PITCH DATA.

Descriptive Note : Final rept. Sep 64-Sep 65.

Corporate Author : FEDERAL SCIENTIFIC CORP NEW YORK

Personal Author(s) : Weiss, Mark R.

Report Date : 15 FEB 1966

Pagination or Media Count : 134

Abstract : A new technique of pitch extraction has been implemented which makes use of the double spectrum analysis of speech signals. In contrast to previous applications of this method of extracting pitch, the system described here operates in real-time and does not require the use of a computer. In this implementation, the waveforms of speech spectra, obtained by use of a Model 254 SIMORAMIC Spectrum Analyzer, are scanned by an analyzer of the stepped-filter type to obtain the desired pitch values. Tests of this instrument show that it provides reliable pitch data even when the signal-to-noise ratio is very poor and the fundamental component of voiced speech is missing. Spurious signals such as hum and narrow band noise, even when much greater than the speech signal, do not affect the accuracy of pitch extraction. The analyzer potentially is capable of detecting and extracting pitch data for multiple talker inputs. (Author)

Descriptors :   (*PITCH DISCRIMINATION, *SPECTRUM ANALYZERS), (*SPEECH, SPECTRUM ANALYZERS), (*AUDITORY SIGNALS, SPECTRUM ANALYZERS), INFORMATION THEORY, SPEECH RECOGNITION, SOUND PITCH, WAVE ANALYZERS, WIRING DIAGRAMS

Subject Categories : Psychology
      Cybernetics
      Acoustics
      Voice Communications

Distribution Statement : APPROVED FOR PUBLIC RELEASE