José David Romeral

Learn More
+P VJKU RCRGT YG FGUETKDG VJG UVGRU ECTTKGF QWV HQT FGXGNQRKPI C 4CKNYC[ +PHQTOCVKQP 5GTXKEG HQT 5RCPKUJ 6JKU YQTM KPVTQFWEGU C OGVJQFQNQI[ HQT FGUKIPKPI FKCNQIWG OCPCIGTU KP URQMGP FKCNQIWG U[UVGOU HQT TGUVTKEVGF FQOCKPU +P VJKU OGVJQFQNQI[ UGXGTCN UQWTEGU QH KPHQTOCVKQP CTG EQODKPGF KPVWKVKQP QDUGTXCVKQP CPF UKOWNCVKQP HQT FGHKPKPI UGXGTCN FKCNQIWG(More)
In the context of large vocabulary speech recognition systems, it is crucial to accurately model the allophonic variations that will be found in a real world task. In this paper we describe a study on the use of data driven pronunciation variations, considering the generation and reduction strategies, as well as their impact in the system performance. The(More)
  • 1