Research in Speech Understanding

Abstract : This annual report describes the work performed during the first year of a two-year effort to develop improved techniques for acoustic-phonetic recognition, with eventual application to automatic speech understanding. The work during this year was devoted to three major activities: speech database generation, computer tool building, and development of acoustic-phonetic rules. A large database of speech was designed and then recorded by four male speakers. Many utility programs that had been developed for the PDP-10 TOPS-20 system were modified to run on the VAX 11/780 under VMS. We initiated the development of a set of context-dependent acoustic-phonetic features to distinguish among phonemes with greater accuracy.