Nonlinear prediction of speech
- 1 January 1991
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 15206149,p. 425-428 vol.1
- https://doi.org/10.1109/icassp.1991.150367
Abstract
Measurements were made of the correlation dimension of normally spoken speech from a single speaker, and the results reveal that most of the points in the state space of the signal lie very close to a manifold of a dimensionality of less than three. This result indicates that one should be able to construct a nonlinear predictor for speech that significantly outperforms linear predictors. To validate this conclusion, a nonparametric predictor was constructed which was able to produce a prediction gain approximately 3 dB better than an equivalent linear predictor. Similar improvements in signal-to-noise ratio were also observed when the nonlinear predictor was added to a simple speech coder.Keywords
This publication has 7 references indexed in Scilit:
- A dynamical systems approach to speech processingPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Nonlinear prediction of chaotic time seriesPhysica D: Nonlinear Phenomena, 1989
- Locally Weighted Regression: An Approach to Regression Analysis by Local FittingJournal of the American Statistical Association, 1988
- Extracting qualitative dynamics from experimental dataPhysica D: Nonlinear Phenomena, 1986
- The dimension of chaotic attractorsPhysica D: Nonlinear Phenomena, 1983
- Characterization of Strange AttractorsPhysical Review Letters, 1983
- Differentiable ManifoldsAnnals of Mathematics, 1936