Event

Decoding Speech Using Neural Rhythmicity and Synchrony

Oded Ghitza

Boston University

Tuesday, May 05, 2009
12:30

Speech is an inherently rhythmic phenomenon. Phonetic segments are articulated in syllabic "packages," which are spoken in cadence and reflect energy modulations between 3 and 20 Hz. The intonation contour is also rhythmic. This rhythmic aspect of speech is important for intelligibility and naturalness. The main thesis of our ongoing project is that many temporal properties of spoken language reflect not merely articulatory constraints but also higher-order principles of cortical function. In particular, we believe that neural rhythmicity may play an important role in decoding spoken language. The range over which these rhythms operate (0.5–80 Hz) could serve as the basis of a hierarchical synchronization mechanism by which the central nervous system processes and integrates linguistic information. For example, beta (12−30 Ηz) and gamma (30−80 Ηz) rhythms may reflect neural processing of phonetic segments and features, theta (3−10 Ηz) rhythms the processing of syllables and words, and delta (0.5−3 Ηz) rhythms the processing of sequences of syllables and words embedded within the prosodic phrase. In this talk I'll describe our efforts in quantifying the possible role of internal brain rhythms in perceiving speech by generating speech sentences with disrupted syllabic rhythm and measuring intelligibility as a function of the degree of disruption. This data is intended to be used to determine parameter values for a computational model of speech processing based upon neural rhythmicity ("TEMPO").

Bio:
Oded Ghitza received the B.Sc., M.Sc. and Ph.D. degrees in electrical Engineering from Tel-Aviv University, Israel, in 1975, 1977 and 1983, respectively. From 1968 to 1984 he was with the Signal Corps Research Laboratory of the Israeli Defense Forces. During 1984-1985 he was a Bantrell post-doctoral fellow at MIT, Cambridge, Massachusetts, and a consultant with the Speech Systems Technology Group at Lincoln Laboratory, Lexington, Massachusetts. From 1985 to early 2003 he was with the Acoustics and Speech Research Department, Bell Laboratories, Murray Hill, New Jersey, where his research was aimed at developing models of hearing and at creating perception based signal analysis methods for speech recognition, coding and evaluation. Since early 2003 he is with Sensimetrics Corp., Malden, Massachusetts, where he continues to acquire and model basic knowledge of auditory physiology and of perception for the purpose of advancing speech, audio and hearing-aid technology. From 2005 to 2008 he was with the Sensory Communication Group at MIT. Since mid 2006 he is with the Center for Biodynamics and with the Hearing Research Center at Boston University, where he studies the role of brain rhythms in speech perception.