Time Map Phonology Finite State Models And Event Logics In Speech Recognition

Time Map Phonology addresses key areas of sound structure at which the two technologies of natural language processing and speech technology are beginning to converge. Solutions are presented to the problems of how to process words which have not been heard before and how to develop fine-grained knowledge representation and processing techniques for linguistic units smaller than the word. The solutions are based on a careful comparison of linguistic theories and on the investigation of computational techniques for the next generation of flexible spoken language input and output devices. The approach has been fully implemented for the vocabulary of German and subjected to quantitative evaluation.