Combining Multiple Knowledge Sources for Speech Recognition
Abstract
This annual report summarizes the work on using multiple knowledge sources in a speech recognition system. Included is research in search strategies, statistical language modeling and phonological rules. We also describe the testing of the system using the standard DARPA Resource Management Database. Additional topics of demonstrations, database documentation and porting of the system from the Symbolics Lisp machine to the SUN4 in C are also discussed.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 15, 1988
- Accession Number
- ADA198928
Entities
People
- John Makhoul
- Richard Schwartz
Organizations
- BBN Technologies