Short Utterance-based Video Aided Speaker Recognition - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

Short Utterance-based Video Aided Speaker Recognition

Résumé

Embedded speaker recognition in mobile devices could involve several ergonomic constraints and a limited amount of computing resources. Even if they have proved their efficiency in more classical contexts, GMM/UBM based systems show their limits in such situations, with good accuracy demanding a relatively large quantity of speech data, but with negligible harnessing of linguistic content. The proposed approach addresses these limitations and takes advantage of the linguistic nature of the speech material into the GMM/UBM framework by using client-customised utterances. Furthermore, the acoustic structure is then reinforced with video information. Experiments on the MyIdea database are performed when impostors know the client utterance and also when they do not, highlighting the potential of this new approach. A relative gain up to 47% in terms of EER is achieved when impostors do not know the client utterance and performance is equivalent to the GMM/UBM baseline system in other configurations.
Fichier principal
Vignette du fichier
Short Utterance-based Video Aided Speaker Recognition.pdf (253.01 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01927792 , version 1 (20-11-2018)

Identifiants

  • HAL Id : hal-01927792 , version 1

Citer

Anthony Larcher, Jean-François Bonastre, John Mason. Short Utterance-based Video Aided Speaker Recognition. IEEE International Workshop on Multimedia Signal Processing (MMSP), Oct 2010, Cairns, France. ⟨hal-01927792⟩

Collections

UNIV-AVIGNON LIA
26 Consultations
85 Téléchargements

Partager

Gmail Facebook X LinkedIn More