Loading...

Speech modeling and voiced/unvoiced/mixed/silence speech segmentation with fractionally gaussian noise based models

Oveisgharan, Sh ; Sharif University of Technology | 2004

167 Viewed
  1. Type of Document: Article
  2. Publisher: 2004
  3. Abstract:
  4. The ARMA filtered fractionally differenced Gaussian Noise (FdGn) model and a new AR Filtered FdGn Added up model are applied to speech signal and performance of their parameters on speech Unvoiced/Voiced/Mixed/Silence classification is evaluated against Zero Crossing Rate (ZCR) feature. For parameter estimation of AR filtered FdGn model two methods were applied: iterative Maximum Likelihood (ML) method of Tewfik and a new computationally efficient Linear Minimum Square Error (LMSE) algorithm Also for parameters estimation of new Added up model two approaches were implemented: an Expectation-Maximization (EM) based approach and an iterative MSE approach. The described models and methods were applied to speech signal and also its real Cepstrum. The performance of described models on V/U/M/S speech classification was obtained based on J1 parameter in this order: Added up model on real Cepstrum of speech, Filtered FdGn model on real Cepstrum of speech (LMSE method), Filtered FdGn model on speech (LMSE method), ZCR, and Filtered FdGn model on speech (Tewfik method)
  5. Keywords:
  6. Speech enhancement ; Gaussian noise ; Iterative methods ; White noise ; Speech analysis ; Parameter estimation ; Maximum likelihood estimation ; Cepstrum ; Filtering ; Nonlinear filters
  7. Source: Proceedings - IEEE International Conference on Acoustics, Speech, and Signal Processing, Montreal, Que, 17 May 2004 through 21 May 2004 ; Volume 1 , 2004 , Pages I613-I616 ; 15206149 (ISSN)
  8. URL: https://ieeexplore.ieee.org/document/1326060