Towards the optimal minimization of a pronunciation dictionary model

Authors:
Simon Dobriýek;Janez Žibert;France Mihelič
Affiliations:
University of Ljubljana, Faculty of Electrical Engineering, Ljubljana, Slovenia;University of Primorska, Primorska Institute of Natural Sciences and Technology, Koper, Slovenia;University of Ljubljana, Faculty of Electrical Engineering, Ljubljana, Slovenia
Venue:
TSD'10 Proceedings of the 13th international conference on Text, speech and dialogue
Year:
2010

Citing 4
Cited 0

Compilers: principles, techniques, and tools

Compilers: principles, techniques, and tools
Minimisation of acyclic deterministic automata in linear time

Theoretical Computer Science - Selected papers of the Combinatorial Pattern Matching School
Statistical methods for speech recognition

Statistical methods for speech recognition
OpenFst: a general and efficient weighted finite-state transducer library

CIAA'07 Proceedings of the 12th international conference on Implementation and application of automata

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents the results of our efforts to obtain the minimum possible finite-state representation of a pronunciation dictionary. Finite-state transducers are widely used to encode word pronunciations and our experiments revealed that the conventional redundancy-reduction algorithms developed within this framework yield suboptimal solutions. We found that the incremental construction and redundancy reduction of acyclic finite-state transducers creates considerably smaller models (up to 60%) than the conventional, nonincremental (batch) algorithms implemented in the OpenFST toolkit.