Voicing detection by Kalman filtering
A. Arcese · The Journal of the Acoustical Society of America · 1975
In this talk, we present an algorithm for the automatic detection of voiced and unvoiced speech segments. The algorithm makes use of the Kalman filter where the state transition matrix is taken as the identity matrix. The attributes used for discrimination are the partial correlation coefficients PARCOR). Initially, the algorithm requires a teacher; thereafter, it becomes self-learning. Audio tapes demonstrate speech processed with this algorithm on an LPC-10 system.