Subspace techniques in speech enhancement

Author

Smith, Gavin ; Niranjan, Mahesan ; Robinson, Tony

Author_Institution

Dept. of Eng., Cambridge Univ., UK

fYear

1999

fDate

36373

Firstpage

449

Lastpage

458

Abstract

This paper concerns speech enhancement. Speech is first modelled and then model parameters used to design a Kalman smoother to reduce the background noise. The 4SID (subspace state space system identification), DOA (direction-of-arrival) and polynomial techniques are described within a common subspace state space framework, and assumptions for each technique stated. It is shown that 4SID methods make least assumptions about the data, provide detailed spectra for the vocal tract, and can be used effectively for speech enhancement when combined with a Kalman smoother. An iterative technique is also possible

Keywords

Kalman filters; direction-of-arrival estimation; iterative methods; noise; polynomials; smoothing methods; spectral analysis; speech enhancement; state-space methods; 4SID methods; DOA; Kalman smoother; background noise removal; direction-of-arrival; iterative technique; model parameters; polynomial techniques; speech enhancement; speech modelling; subspace state space framework; subspace state space system identification; subspace techniques; vocal tract spectra; Background noise; Computer science; Equations; Gaussian noise; Iterative methods; Kalman filters; Parameter estimation; Speech enhancement; State-space methods; System identification;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing IX, 1999. Proceedings of the 1999 IEEE Signal Processing Society Workshop.

Conference_Location

Madison, WI

Print_ISBN

0-7803-5673-X

Type

conf

DOI

10.1109/NNSP.1999.788164

Filename

788164