Principal Investigator Stefanie Hufnagel
Co-investigator Joseph S Perkell
Project Website http://www.rle.mit.edu/speech/
Speech is highly variable, especially in spontaneous conversation (words and sounds are acoustically very different in different contexts). The variation is systematic and predictable, and easily handled by human listeners, but presents a challenge for automatic speech recognition and for current models of human speech perception and production. The objective is to integrate knowledge-based processing with statistical processing for speech analysis.
The Speech Communication Group (SPE), part of the Research Laboratory of Electronics at MIT, is a multidisciplinary laboratory engaged in teaching and research on the production and perception of speech by humans and machines. The approach is to combine the principles and tools of electrical engineering and computer science with the insights of linguistic theory and cognitive science to construct integrated models of the physics, physiology and cognition of speech processing.
Current Research Projects:(*) A model for determining consonantal features (*) Constraints and strategies in speech production(*) Effects of hearing status on adult speech production(*) Phonetic modification of function words(*) Physiological and acoustic studies of speech(*) Segmental and prosodic aspects of speech planning