Multi-speaker voice activity detection using a camera-assisted microphone array.
Trond F. BerghInes HafizovicSverre HolmPublished in: IWSSIP (2016)
Keyphrases
- voice activity detection
- speech recognition
- field of view
- hand held
- video camera
- single camera
- focal length
- surveillance system
- camera calibration
- vision system
- structure from motion
- real time
- stereo camera
- noisy environments
- autocalibration
- multi camera
- ccd camera
- speaker recognition
- speaker verification
- multiple cameras
- camera motion
- hidden markov models