Reconstruction of missing features for robust speech recognition


Speech recognition systems perform poorly in the presence of corrupting noise. Missing feature methods attempt to compensate for the noise by removing noise corrupted components of spectrographic representations of noisy speech and performing recognition with the remaining reliable components. Conventional classifier-compensation methods modify the recognition system to work with the incomplete representations so obtained. This constrains them to perform recognition using spectrographic features which are known to be less optimal than cepstra. In this paper we present two missing-feature algorithms that reconstruct complete spectrograms from incomplete noisy ones. Cepstral vectors can now be derived from the reconstructed spectrograms for recognition. The first algorithm uses MAP procedures to estimate corrupt components from their correlations with reliable components. The second algorithm clusters spectral vectors of clean speech. Corrupt components of noisy speech are estimated from the distribution of the cluster that the analysis frame is identified with. Experiments show that, although conventional classifier-compensation methods are superior when recognition is performed with spectrographic features, cepstra derived from the reconstructed spectrograms result in better recognition performance overall. The proposed methods are also less expensive comput-ationally and do not require modification of the recognizer.

DOI: 10.1016/j.specom.2004.03.007

Extracted Key Phrases

13 Figures and Tables

Showing 1-10 of 32 references

Speech Communication

  • B Raj
  • 2004

Speech in noisy conditions using missing feature approach

  • P Renevey
  • 2001
2 Excerpts
Showing 1-10 of 141 extracted citations


Citations per Year

254 Citations

Semantic Scholar estimates that this publication has received between 189 and 342 citations based on the available data.

See our FAQ for additional information.