On-the-fly term spotting by phonetic filtering and request-driven decoding - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2008

On-the-fly term spotting by phonetic filtering and request-driven decoding

Résumé

This paper addresses the problem of on-the-fly term spotting in continuous speech streams. We propose a 2-level architecture in which recall and accuracy are sequentially optimized. The first level uses a cascade of phonetic filters to select the speech segments which probably contain the targeted terms. The second level performs a request-driven decoding of the selected speech segments. The results show good performance of the proposed system on broadcast news data : the best configuration reaches a F-Measure of about 94% while respecting the on-the-fly processing constraint.
Fichier principal
Vignette du fichier
Onthefly.pdf (82.8 Ko) Télécharger le fichier
Loading...

Dates et versions

hal-01320210 , version 1 (14-11-2017)

Identifiants

Citer

Mickael Rouvier, Georges Linares, Benjamin Lecouteux. On-the-fly term spotting by phonetic filtering and request-driven decoding. IEEE Spoken Language Technology Workshop (SLT), Dec 2008, Goa, India. ⟨10.1109/SLT.2008.4777901⟩. ⟨hal-01320210⟩

Collections

UNIV-AVIGNON LIA
38 Consultations
102 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More