Abstract : In this paper we present a speech recognition system based on articulatory dynamics. We do not extend the acoustic feature with any explicit articulatory measurements but instead the ar- ticulatory dynamics of speech are structurally embodied within episodic memories. The proposed recognizer is made of differ- ent memories each specialized for a particular articulator. As all the articulators do not contribute equally to the realization of a particular phoneme, the specialized memories do not per- form equally regarding each phoneme. We show, through phone string recognition experiments that combining the recognition hypotheses resulting from the different articulatory specialized memories leads to significant recognition improvements.
https://hal.inria.fr/inria-00602414
Contributeur : Slim Ouni
<>
Soumis le : mardi 6 octobre 2015 - 15:39:07
Dernière modification le : jeudi 11 janvier 2018 - 06:19:57
Document(s) archivé(s) le : jeudi 7 janvier 2016 - 10:41:36
Sébastien Demange, Slim Ouni. Continuous episodic memory based speech recognition using articulatory dynamics. ISCA. 12th Annual Conference of the International Speech Communication Association - Interspeech 2011, Aug 2011, Florence, Italy. 2011. 〈inria-00602414〉
