Title :
Speech recognition with syllables and concepts
Author :
De Palma, P. ; Wooters, C.
Author_Institution :
Dept. of Comput. Sci., Gonzaga Univ., Spokane, WA, USA
Abstract :
The transformation of speech to words is not necessary to emulate human linguistic performance in some contexts. A large vocabulary continuous speech recognition (LVCSR) system can perform more accurately with a syllable-based, as opposed to a word-based, language model. This accuracy can be further enhanced by the addition of a concept model, where a concept is an equivalence class of words and phrases.
Keywords :
computational linguistics; equivalence classes; speech recognition; vocabulary; LVCSR system; equivalence class; human linguistic performance; large vocabulary continuous speech recognition system; speech transformation; syllable-based language model; syllables; word-based language model; concept; speech recognition; syllable;
Conference_Titel :
Soft Computing and Intelligent Systems (SCIS) and 13th International Symposium on Advanced Intelligent Systems (ISIS), 2012 Joint 6th International Conference on
Conference_Location :
Kobe
Print_ISBN :
978-1-4673-2742-8
DOI :
10.1109/SCIS-ISIS.2012.6505412
