Fuzzy reasoning in confidence evaluation of speech recognition
G. Hernandez-Abrego, José Bernardo Mariño Acebal · RECERCAT (Consorci de Serveis Universitaris de Catalunya) · 1999
Confidence measures represent a systematic way to express reliability of speech recognition results. A common approach to confidence measuring is to take profit of the information that several recognition-related features offer and to combine them, through a given compilation mechanism, into a more effective way to distinguish between correct and incorrect recognition results. We propose to use a fuzzy reasoning scheme to perform the information compilation step. Our approach opposes the previously proposed ones because ours treats the uncertainty of recognition hypotheses in terms of "possibility" contrasted to the "probability" notion of similar works. Experimental results over isolated words, continuous speech and keyword spotting recognition tasks show higher performance of our system compared against standard compilation methods. Here we demonstrate that, due to their approach to uncertainty; to their capabilities to handle expert knowledge and to their versatility, Fuzzy Inferenc...