Phoneme Lattice Based A* Search Algorithm for Speech Recognition

Abstract : This paper presents the Speeral continuous speech recognition system developed in the LIA. Speeral uses a modified A* algorithm to find in the search graph the best path taking into account acoustic and linguistic constraints. Rather than words by words, the A* used in Speeral is based on a phoneme lattice previously generated. To avoid the backtraking problems, the system keeps for each frame the deepest nodes of the partially explored lexical tree starting at this frame. If a new hypothesis to explore is ended by a word and the lexicon starting where this word finishes has already been developed, then the next hypothesis will " jump " directly to the deepest nodes. Decoding performances of Speeral are evaluated on the test set of the ARC B1 campaign of AUPELF '97. The experiments on this French database show the efficiency of the search strategy described in this paper.
Document type :
Conference papers
Complete list of metadatas

https://hal.archives-ouvertes.fr/hal-01319837
Contributor : Bibliothèque Universitaire Déposants Hal-Avignon <>
Submitted on : Monday, May 23, 2016 - 10:06:15 AM
Last modification on : Saturday, March 23, 2019 - 1:22:39 AM

Links full text

Identifiers

Collections

Citation

Pascal Nocera, Georges Linares, Dominique Massonié, Loïc Lefort. Phoneme Lattice Based A* Search Algorithm for Speech Recognition. TSD 5th International Conference, Sep 2002, Brno, Czech Republic. ⟨10.1007/3-540-46154-X_41⟩. ⟨hal-01319837⟩

Share

Metrics

Record views

64