Separation of speech from interfering speech by means of harmonic selection

Abstract

A common type of interference in speech transmission is that caused by the speech of a competing talker. Although the brain is adept at clarifying such speech, it relies heavily on binaural data. When voices interfere over a single channel, separation is much more difficult and intelligibility suffers. Clarifying such speech is a complex and varied problem whose nature changes with the moment-to-moment variation in the types of sound which interfere. The principal subproblem, the separation of vocalic speech, was attacked. Separation was done by selecting the harmonics of the desired voice in the Fourier transform of the input. In implementing this process, techniques were developed for resolving overlapping spectrum components, for determining pitches of both talkers and for assuring consistent separation. These techniques were described, their performance on test utterances was summarized and the possibility of using this process as a basis for the solution of the general 2-talker problem was briefly considered.