Joint speaker segmentation, localization and identification for streaming audio.
Joerg SchmalenstroeerReinhold Haeb-UmbachPublished in: INTERSPEECH (2007)
Keyphrases
- audio visual
- optic disc
- segmentation algorithm
- accurate localization
- multimedia
- speaker identification
- data streams
- segmentation method
- audio stream
- fully automatic
- image segmentation
- level set
- prosodic features
- joint estimation
- image analysis
- medical images
- multiscale
- speech recognition
- media streams
- object segmentation
- region growing
- multi modal
- background subtraction
- energy function
- reliable detection
- graph cuts
- feature extraction
- shape prior
- visual data
- object localization
- audio signals
- visual information
- speaker diarization
- real time