|nikkostrom | NICO | Quite BASIC|
N. Ström, L. Hetherington, T.J. Hazen, E. Sandness, & J. Glass (1997): "Acoustic Modeling Improvements in a Segment-Based Speech Recognizer," Proc. of the 1999 IEEE Workshop on Automatic Speech Recognition and Understanding, Keystone, Colorado.
Abstract - In this paper we report on some recent improvements
on the acoustic modeling in a segment-based speech recognition system.
Context-dependent segment models and improved pro-nunciation modeling are
shown to reduce word error rates in a phone-based, conversational system
by over 18%, while the technique of Gaussian selection reduces overall
computation by more than a factor of two.