Abstract
Since the earliest days of computing automatic speech recognition technology has ridden the technology wave that has come to be known as Moore’s Law. This is vividly illustrated by the market introduction of several general-purpose continuous speech recognition products. Besides Moore’s law the two things that have made this possible have been advances in acoustic modeling, especially adaptation technologies and advances in decoding techniques that permit real-time performance on today’s PC’s.I will discuss a broad range of these decoding techniques including, beam search, A-star and variants, multi-pass search, incremental application of knowledge, tree-organization, heuristic pruning and look-ahead techniques. Additionally I will discuss the relative merits of these search techniques with respect to memory size and performance.
Biography
Fil Alleva received the BS degree in mathematics from Carnegie Mellon University in 1980. As an undergraduate he worked on the Harpy speech recognition system. Later as a Project Scientist at CMU he contributed to the Agora, Sphinx I and Sphinx II system, for which he was awarded the 1992 Allen Newell Research Excellence Medal. He joined Microsoft Research in 1993 and is currently managing CSR system development. Mr. Alleva is a member of the IEEE and has published numerous papers on spoken language technology. His current professional interests are in all areas of spoken language technology, particularly heuristic search and language modeling.