Comparison of Gaussian process regression and Gaussian mixture models in spectral tilt modelling for intelligibility enhancement of telephone speech.
Emma JokinenUlpu RemesPaavo AlkuPublished in: INTERSPEECH (2015)
Keyphrases
- gaussian process regression
- speaker recognition
- gaussian mixture model
- speaker identification
- gaussian processes
- speech recognition
- gaussian process
- mixture model
- feature vectors
- em algorithm
- spectral features
- speech signal
- expectation maximization
- speaker verification
- image processing
- mel frequency cepstral coefficients
- facial expression recognition
- bayesian framework
- covariance matrices
- maximum likelihood