An adaptive KLT approach for speech enhancement.

An adaptive Karhunen-Loeve transform (KLT) tracking-based algorithm is proposed for enhancement of speech degraded by colored additive interference. This algorithm decomposes noisy speech into its components along the axes of a KLT-based vector space of clean speech. It is observed that the noise en...

ver descrição completa

Detalhes bibliográficos
Publicado no:	IEEE Transactions on speech and audio processing 9, 2 (2001).
Autor principal:	Rezayee, A.
Formato:	Artigo
Idioma:	English
Assuntos:	KLT tracking-based algorithm. KLT-based vector space. Adaptive KLT. Adaptive Karhunen-Loeve transform. Clean speech. Colored additive interference. Computer generated white Gaussian noise. Distortion minimisation. Efficient voice activity detector. Eigenvalue. Eigenvector. Inverse KLT. Listening tests. Noise energy. Noise samples. Noisy speech decomposition. Noisy speech processing. Principle component eigenfilter. Residual noise power. Silence intervals. Speech enhancement. Speech samples.