EM-Based TDOA Estimation of a Speech Source via Gaussian Mixture Models in Noisy and Anechoic Environments.
Zhihua LuJoão P. J. da CostaTai FeiPublished in: IEEE Access (2021)
Keyphrases
- gaussian mixture model
- speaker recognition
- speaker identification
- expectation maximization
- mixture model
- em algorithm
- probability density
- density estimation
- log likelihood function
- feature vectors
- maximum likelihood estimation
- gaussian mixture
- parameter estimation
- speech music discrimination
- gaussian distribution
- maximum likelihood
- probability density function
- noisy environments
- generative model
- feature space
- mel frequency cepstral coefficients
- speech recognition
- finite mixtures
- variational bayes
- covariance matrices
- speech signal
- audio visual
- image compression
- language model
- model selection
- unsupervised learning
- bayesian networks