Skip to main content
Proceedings - International Conference on Pattern RecognitionVolume 3, 1994, Pages 211-21312th IAPR International Conference on Pattern Recognition - Conference C: Signal Processing - Conference D: Parallel Computing, ICPR 1994; Jerusalem; Israel; 9 October 1994 through 13 October 1994; Category number94CH3440-5; Code 171434

Time-varying AR speech analysis using robust RLS algorithm with variable forgetting factor(Conference Paper)

  • Kovačević, B.D.,
  • Milosavljević, M.M.,
  • Veinović, M.Dj.
  Save all to author list
  • aFaculty of Electrical Engineering, Bulevar Revolucije 73, Belgrade, Serbia
  • bInstitute of Applied Math, and Electronics, Kneza Miloša 37, Belgrade, Serbia

Abstract

In this paper a new robust recursive method of estimating the linear prediction (LP) parameters of an auto-regressive (AR) speech signal model using weighted least squares (WLS) with variable forgetting factors (VFF's) is described. The proposed robust recursive least squares (RRLS) differs from the conventional recursive least squares (RLS) by the insertion of a suitable chosen nonlinear transformation of the prediction residuals. The RRLS algorithm takes into account the contaminated Gaussian nature of the excitation for voiced speech. In addition, VFF is adapted to a nonstationary speech signal by a generalized likelihood ratio (MGLR) algorithm, which accounts for the nonstationarity of a speech signal. The proposed method has a good adaptability to the nonstationary parts of a speech signal, and gives low bias and low variance at the stationary signal segments. © 1994 IEEE.

Indexed keywords

Engineering controlled terms:Audio signal processingMathematical transformationsSpeechSpeech analysisSpeech communication
Engineering uncontrolled termsAuto-regressiveLinear predictionNonstationaryPrediction parametersRecursive Least Square algorithmRecursive least squaresRecursive methodsSpeech signalsTime varyingVariable forgetting factors
Engineering main heading:Least squares approximations
  • ISSN: 10514651
  • ISBN: 0818662751
  • CODEN: PICRE
  • Source Type: Conference Proceeding
  • Original language: English
  • DOI: 10.1109/ICPR.1994.577162
  • Document Type: Conference Paper
  • Sponsors: The International Association for Pattern Recognition (IAPR)
  • Publisher: Institute of Electrical and Electronics Engineers Inc.


© Copyright 2021 Elsevier B.V., All rights reserved.

Cited by 4 documents

Pradhan, S. , Patel, V. , Patel, K.
Acoustic feedback cancellation in digital hearing aids: A sparse adaptive filtering approach
(2017) Applied Acoustics
Kovačević, B. , Milosavljević, M.M. , Veinović, M.
Robust digital processing of speech signals
(2017) Robust Digital Processing of Speech Signals
Kovačević, B. , Banjac, Z. , Kovačević, I.K.
Robust adaptive filtering using recursive weighted least squares with combined scale and variable forgetting factors
(2016) Eurasip Journal on Advances in Signal Processing
View details of all 4 citations
Inform me when this document is cited in Scopus:
Set citation alert Set citation feed

SciVal Topic Prominence

Topic:
Prominence percentile: