Robust audio speaker segmentation using one class SVMS.
Hachem KadriManuel DavyAsma RabaouiZied LachiriNoureddine EllouzePublished in: EUSIPCO (2008)
Keyphrases
- audio visual
- robust segmentation
- level set
- image segmentation
- multimedia
- region growing
- speaker identification
- signal processing
- medical images
- energy function
- multiscale
- fully unsupervised
- multibody
- noisy environments
- multiple cues
- segmentation accuracy
- object segmentation
- visual data
- segmentation method
- segmentation algorithm
- motion segmentation
- shape prior
- computationally efficient
- broadcast news
- active contours
- speaker recognition
- audio signals
- markov random field
- image sequences
- audio stream
- automatic transcription
- moving object extraction