A recursive expectation-maximization algorithm for speaker tracking and separation.
Ofer SchwartzSharon GannotPublished in: EURASIP J. Audio Speech Music. Process. (2021)
Keyphrases
- expectation maximization
- em algorithm
- computational complexity
- np hard
- real time
- kalman filter
- dynamic programming
- optimization algorithm
- probabilistic model
- worst case
- recursive algorithm
- maximum likelihood estimation
- parameter estimation
- generative model
- maximum likelihood
- neural network
- cost function
- k means
- optimal solution
- detection algorithm
- kalman filtering
- target tracking
- moving target
- learning algorithm
- gaussian mixture
- bayesian framework
- similarity measure
- object tracking
- computational cost
- objective function
- mixture model
- significant improvement
- graph cuts
- particle swarm optimization