A mixed-source model for speech compression and synthesis.

John Makhoul, R. Viswanathan, Richard M. Schwartz, A. W. F. Huggins

Browse the full ICASSP paper archive.