Filling the gaps: Gaussian mixture models from noisy, truncated or incomplete samples.
Peter MelchiorAndy D. GouldingPublished in: Astron. Comput. (2018)
Keyphrases
- gaussian mixture model
- incomplete data
- mixture model
- em algorithm
- noisy data
- speaker recognition
- feature vectors
- data sets
- expectation maximization
- maximum likelihood
- missing data
- probability density
- gaussian mixture
- density estimation
- noisy environments
- training set
- probability density function
- missing values
- speaker identification
- finite mixtures
- pairwise