In addressing the problem of achieving high-accuracy real-time speech recognition systems, we focus on recognizing speech from ARPA's20,000-word Wall Street Journal (WSJ) tas...
Hy Murveit, Peter Monaco, Vassilios Digalakis, Joh...
An ongoing debate in text understanding efforts centers on the use of pattern-matching techniques, which some have characterized as "designed to ignore as much text as possib...
The BYBLOS continuous speech recognition system is applied to on-line cursive handwriting recognition. By exploiting similarities between on-line cursive handwriting and continuou...
John Makhoul, Thad Starner, Richard M. Schwartz, G...
The key problem to be faced when building a HMM-based continuous speech recogniser is maintaining the balance between model complexity and available training data. For large vocab...
In this paper, we describe a reversibleletter-to-sound/soundto-letter generation system based on an approach which combines a rule-based formalism with data-driven techniques. We ...