BTH: an efficient parsing algorithm for word-spotting

This paper presents a new parsing algorithm, BTH, which is capable of efficiently parsing a keyword lattice that contains a large number of false alarms. The BTH parser runs without unfolding the given keyword lattice, and thus it can efficiently obtain a set of word sequences acceptable to the given grammar as the parser result. The algorithm has been implemented on Windows-based PCs and is tested by applying it to a car navigation task that has a scale of practical applications. The result indicates promise in implementing the function of a sentence utterance-based spontaneous speech understanding in next -generation car navigation systems.