Speech Separation Based on Higher Order Statistics Using Recurrent Neural Networks
Independent Component Analysis (ICA) (Comon, 1994; Lee, 1998; Karhunen et al,1997; Haykin, 1998) is an unsupervised technique, which tries to represent the data in terms of statistically independent variables. ICA and the related blind source separation (BSS) and application topics both in unsupervised neural learning and statistical signal processing.
KeywordsIndependent Component Analysis Recurrent Neural Network Independent Component Analysis Blind Source Separation Processing Node
Unable to display preview. Download preview PDF.
- Cardoso J.F. (1989) Source Separation Using Higher Order Moments, IEEE Transactions on Signal Processing, Vol. No. pp. 2109–2112Google Scholar
- Cardoso JF and Comon P (1996). Independent component analysis, a survey of some algebraic methods. In Proc. ISCAS’96, vol. 2, pp. 93–96.Google Scholar
- Chan, D. C. B. (1997). Blind Signal Processing, PhD thesis, Signal Processing and Communications Laboratory Department of Engineering, University of Cambridge, UK.Google Scholar
- Haykin S. (1998), Neural networks- A comprehensive Foundation, 2nd ed. Englewood Cliffs, NJ: Prentice-Hall.Google Scholar
- Lee T. W. (1998), Independent component analysis-theory and application, Dordrecht, The Netherlands: Kluwer.Google Scholar
- Li Y., Powers D. and Wen P. (2001), Separation and Deconvolution of Speech Using Recurrent Neural Networks, pp. 1303–1309, Vol. III, Proceedings of the International Conference on Artificial Intelligence (IC-AI’01), June 25–28, 2001, Las Vegas, Nevada, USA.Google Scholar
- Mansour A. and Ohnishi N. (1999), Multichannel Blind Separation of Sources Algorithm Based on Cross- Cumulant and the Levenberg-Marquardt Method, Vol. 47, No. 11, November, pp. 3172–3179.Google Scholar
- Schobben, D. (1999), Efficient adaptive multi-channel concepts in acoustics: Blind signal separation and echo cancellation, PhD thesis.Google Scholar
- Westner, A. G. (1999). Object based audio capture: Separating acoustically mixed sounds. Master’s thesis, MIT.Google Scholar