SIGNAL PROCESSING BY INFINITE CLIPPING AND RELATED TECHNIQUES.
Abstract
Infinite peak clipping was studied as a technique for processing speech and other signals. Current and past utilization of this technique was surveyed briefly, and limited experimental and theoretical studies were performed. A new explanation is advanced for the high intelligibility of infinitely clipped speech, and a technique is suggested for achieving an acceptable compromise between the improved resistance to noise or jamming (or, alternatively, reduction in transmitted power) provided by infinitely clipped speech with amplitude modulation, and the concomitant reduction in quality and voice recognition. A possible system for speech bandwidth compression using a signal obtained by infinite clipping is described, and results of simple experiments are given. Also described is a type of visual display which produces patterns on speech signals that tend to show high correlation with the spoken word and its enunciation, and low correlation with the particular voice speaking. Fields of application for the techniques discussed include: radio communication systems, antijamming measures, acoustic systems, speech bandwidth compression, speech therapy and training, automatic recognition of speech, doppler radar signal analysis, and analysis of phonocardiographic and other physiological signals for medical diagnosis. (Author)
Document Details
- Document Type
- Technical Report
- Publication Date
- Apr 01, 1964
- Accession Number
- AD0600863
Entities
People
- Berry O. Pyron
- Frank R. Williamson Jr.
Organizations
- Georgia Tech