FST-Based Pronunciation Lexicon Compression for Speech Engines
Žiga Golob, Jerneja Žganec Gros, Mario Žganec, Boštjan Vesnicer, Simon Dobrišek · International Journal of Advanced Robotic Systems · 2012
Finite-state transducers are frequently used for pronunciation lexicon representations in speech engines, in which memory and processing resources are scarce. This paper proposes two possibilities for further reducing the memory footprint of finite-state transducers representing pronunciation lexicons. First, different alignments of grapheme and allophone transcriptions are studied and a reduction in the number of states of up to 30% is reported. Second, a combination of grapheme-to-allophone rules with a finite-state transducer is proposed, which yields a 65% smaller finite-state transducer than conventional approaches.