Active speaker detection with audio-visual co-training.

Punarjay Chakravarty Jeroen Zegers Tinne Tuytelaars Hugo Van hamme

Published in: ICMI (2016)

Keyphrases

audio visual
co training
multi modal
visual information
semi supervised
semi supervised learning
multi view
single view
unlabeled data
visual data
multimedia
speaker verification
training examples
text classification
supervised learning
multi stream
labeled data
machine learning
low level
feature extraction