2005 | OriginalPaper | Buchkapitel
Nonlinear Speech Processing: Overview and Possibilities in Speech Coding
verfasst von : Marcos Faundez-Zanuy
Erschienen in: Nonlinear Speech Modeling and Applications
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
In this paper we give a brief overview of nonlinear predictive models, with special emphasis on neural nets. Several well known strategies are discussed, such as multi-start random weights initialization, regularization, early stop with validation, committee of neural nets, different architectures, etc. Although the paper is devoted to ADPCM speech coding (scalar and vectorial schemes), this study offers a good chance to deal with nonlinear predictors, as a first step towards a more sophisticated applications. Thus, our main purpose is to state new possibilities for speech coding.