
Audio-Visual Speaker Conversion using Prosody Features
Author(s) -
Adela Barbulescu,
Thomas Hueber,
Gérard Bailly,
Rémi Ronfard
Publication year - 2013
Publication title -
hal (le centre pour la communication scientifique directe)
Language(s) - Uncategorized
Resource type - Conference proceedings
Subject(s) - computer science , speech recognition , prosody , mixture model , speaker recognition , joint (building) , identity (music) , speaker diarisation , artificial intelligence , acoustics , engineering , architectural engineering , physics