This project concerns the study of both forward and backward adaptive speech coding structures that operate at 6.4 to 16 kilobits per second. Techniques involved are integer and fractional rate trees, unweighted and weighted squared error distortion measures, the (M,L) tree search algorithm, and incremental path map symbol release. Speech compression to these rates has a number of important applications, including speech storage, voice mail, military communications, commercial telephony, and land mobile radio.
