Person Instance Graphs for Named Speaker Identification in TV Broadcast - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Person Instance Graphs for Named Speaker Identification in TV Broadcast

Résumé

We address the problem of named speaker identification in TV broadcast which consists in answering the question " who speaks when? " with the real identity of speakers, using person names automatically obtained from speech transcripts. While existing approaches rely on a first speaker diarization step followed by a local name propagation step to speaker clusters, we propose a unified framework called person instance graph where both steps are jointly modeled as a global optimization problem, then solved using integer linear programming. Moreover, when available, acoustic speaker models can be added seamlessly to the graph structure for joint named and acoustic speaker identification – leading to a 10% error decrease (from 45% down to 35%) over a state-of-the-art i-vector speaker identification system on the REPERE TV broadcast corpus.
Fichier principal
Vignette du fichier
27.pdf (1.48 Mo) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...

Dates et versions

hal-01690272 , version 1 (23-01-2018)

Identifiants

  • HAL Id : hal-01690272 , version 1

Citer

Hervé Bredin, Antoine Laurent, Achintya Sarkar, Viet-Bac Le, Sophie Rosset, et al.. Person Instance Graphs for Named Speaker Identification in TV Broadcast. Odyssey 2014, Jun 2014, Joensuu, Finland. ⟨hal-01690272⟩
56 Consultations
90 Téléchargements

Partager

Gmail Facebook X LinkedIn More