Supervector Dimension Reduction for Efficient Speaker Age Estimation Based on the Acoustic Speech Signal.
Gil DobryRon M. HechtMireille AvigalYaniv ZigelPublished in: IEEE Trans. Speech Audio Process. (2011)
Keyphrases
- dimension reduction
- speech signal
- acoustic features
- speech recognition
- speaker recognition
- automatic speech recognition
- speaker identification
- gaussian mixture model
- age estimation
- principal component analysis
- mel frequency cepstral coefficients
- low dimensional
- feature extraction
- vocal tract
- high dimensional data
- high dimensional
- speaker verification
- manifold learning
- singular value decomposition
- unsupervised learning
- linear discriminant analysis
- feature selection
- feature space
- partial least squares
- noisy environments
- data sets
- high dimensionality
- face images
- non stationary
- cluster analysis
- hidden markov models
- sound source
- least squares
- pattern recognition
- information retrieval
- neural network