On using Gaussian mixture model for double-talk detection in acoustic echo suppression.
Ji-Hyun SongKyu-Ho LeeYun-Sik ParkSang-Ick KangJoon-Hyuk ChangPublished in: INTERSPEECH (2010)
Keyphrases
- gaussian mixture model
- mixture model
- speaker recognition
- em algorithm
- maximum likelihood
- mouth region
- feature vectors
- expectation maximization
- detection algorithm
- gaussian distribution
- gaussian mixture modeling
- feature space
- object detection
- background subtraction
- gaussian mixture
- speaker identification
- detection method
- mixture of gaussians
- generative model
- image representation
- data mining
- active learning
- pattern recognition
- multiscale
- bayesian networks
- training data
- computer vision