Speech Compression and Synthesis
Quarterly progress rept. no. 7, 8 Dec-7 Mar 1980
BOLT BERANEK AND NEWMAN INC CAMBRIDGE MA
Pagination or Media Count:
This document reports on work toward a very low rate phonetic vocoder and a multirate speech compression system. The phonetic vocoder consists of a phonetic recognizer based on a trained diphone network, and a natural phonetic synthesizer which also uses diphone templates as a model for speech. This quarter several new diphones were added to the data base. There are currently 2845 diphones. We tested the phonetic recognition program, and made improvements to improve its speed and performance.
- Non-Radio Communications