Advanced search


Joan-Andreu Sánchez, Francisco Casacuberta, Pablo Aibar, David Llorens, María-José Castro. Fast Phoneme Look-Ahead in the ATROS System. Proceedings of the VIII Simposium Nacional de Reconocimiento de Formas y Análisis de Imágenes, 1999. pp. 77-84. AERFAI.

Current speech recognition systems require a lot of computational resources to decode an input utterance. Many efforts have been done in order to reduce these requirements. One of the techniques that is being explored is the fast phoneme look­ahead. The idea is to compute quickly approximate scores in order to prune little promising hypothesis. These scores are computed by using simple phone­like units and analysing an acoustic segment look­ahead. ATROS is an automatic speech recognition system whose acoustic, lexical, and syntactical models can be learnt automatically from training data. An improved version of ATROS which can deal with large vocabularies is presented. This version supports several kinds of acoustic models. It also incorporates improved search algorithms in the decoding process in order to reduce the computational requirements. In this paper exhaustive experiments are reported to evaluate the application of the fast phoneme look­ahead technique in the ATROS system. The system has been tested on a Spanish task of queries to a geographical database (with a vocabulary of 1,208 words).