Combining Acoustic Name Spotting and Continuous Context Models to improve Spoken Person Name Recognition in Speech - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2013

Combining Acoustic Name Spotting and Continuous Context Models to improve Spoken Person Name Recognition in Speech

Benjamin Bigot
  • Fonction : Auteur
  • PersonId : 942554
Gregory Senay
  • Fonction : Auteur
  • PersonId : 942555
Georges Linarès
Corinne Fredouille
Richard Dufour

Résumé

Retrieving pronounced person names in spoken documents is a critical problematic in the context of audiovisual content indexing. In this paper, we present a cascading strategy for two methods dedicated to spoken name recognition in speech. The first method is an acoustic name spotting in phoneme confusion networks. It is based on a phonetic edition distance criterion based on phoneme probabilities held in confusion networks. The second method is a continuous context modelling approach applied on the 1-best transcription output. It relies on a probabilistic modelling of name-to-context dependencies. We assume that the combination of these methods, based on different types of information, may improve spoken name recognition performance. This assumption is studied through experiments done on a set of audiovisual documents from the development set of the REPERE challenge. Results report that combining acoustic and linguistic methods produces an absolute gain of 3% in terms of F-measure compared to the best system taken alone.
Fichier principal
Vignette du fichier
i13_2539.pdf (241.18 Ko) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...

Dates et versions

hal-02102829 , version 1 (26-04-2019)

Identifiants

  • HAL Id : hal-02102829 , version 1

Citer

Benjamin Bigot, Gregory Senay, Georges Linarès, Corinne Fredouille, Richard Dufour. Combining Acoustic Name Spotting and Continuous Context Models to improve Spoken Person Name Recognition in Speech. Interspeech 2013, Aug 2013, Lyon, France. ⟨hal-02102829⟩

Collections

UNIV-AVIGNON LIA
71 Consultations
46 Téléchargements

Partager

Gmail Facebook X LinkedIn More