Loading...

Cepstral-domain HMM-based speech enhancement using vector Taylor series and parallel model combination

Veisi, H ; Sharif University of Technology | 2012

618 Viewed
  1. Type of Document: Article
  2. DOI: 10.1109/ISSPA.2012.6310563
  3. Publisher: 2012
  4. Abstract:
  5. Speech enhancement problem using hidden Markov model (HMM) and minimum mean square error (MMSE) in cepstral domain is studied. This noise reduction approach can be considered as weighted-sum filtering of the noisy speech signal in which the filters weights are estimated using the HMM of noisy speech. To have an accurate estimation of the noisy speech HMM, vector Taylor series (VTS) is proposed and compared with the parallel model combination (PMC) technique. Furthermore, proposed cepstral-domain HMM-based speech enhancement systems are compared with the renowned autoregressive HMM (AR-HMM) approach. The evaluation results confirm the superiority of the cepstral domain approach in comparison with AR-HMM and indicate that VTS-based method provides higher noise reduction rate than the PMC-based method
  6. Keywords:
  7. Accurate estimation ; Auto-regressive ; Cepstral domain ; Evaluation results ; Minimum mean square errors (MMSE) ; Noisy speech ; Noisy speech signals ; Parallel model combinations ; Speech enhancement system ; Vector taylor series ; Weighted-sum ; Hidden Markov models ; Information science ; Noise abatement ; Signal processing ; Speech enhancement ; Speech recognition
  8. Source: 2012 11th International Conference on Information Science, Signal Processing and their Applications, ISSPA 2012, 2 July 2012 through 5 July 2012 ; July , 2012 , Pages 298-303 ; 9781467303828 (ISBN)
  9. URL: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6310563