Missing-Feature Reconstruction by Leveraging Temporal Spectral Correlation for Robust Speech Recognition in Background Noise Conditions

• Published in: IEEE transactions on audio, speech, and language processing speech, and language processing, 2010-11, Vol.18 (8), p.2111-2120
• Main Authors: Wooil Kim, Hansen, J
• Format: Article
• Language: English
• Subjects: Applied sciences; Background noise; Correlation; Error analysis; Exact sciences and technology; Frequency estimation; Information, signal and communications theory; Mathematical models; missing-feature; Noise; Noise robustness; Performance analysis; Reconstruction; Reconstruction algorithms; robust speech recognition; Signal and communications theory; Signal processing; Signal representation. Spectral analysis; Signal, noise; Spectra; Spectral analysis; Speech; Speech analysis; Speech enhancement; Speech processing; Speech recognition; Studies; Telecommunications and information theory; temporal correlation; Temporal logic; temporal spectral feature
• ISSN: 1558-7916; 2329-9290; 1558-7924; 2329-9304
• DOI: 10.1109/TASL.2010.2041698

This paper proposes a novel missing-feature reconstruction method to improve speech recognition in background noise environments. The existing missing-feature reconstruction method utilizes log-spectral correlation across frequency bands. In this paper, we propose to employ a temporal spectral feature analysis to improve the missing-feature reconstruction performance by leveraging temporal correlation across neighboring frames. In a similar manner with the conventional method, a Gaussian mixture model is obtained by training over the obtained temporal spectral feature set. The final estimates for missing-feature reconstruction are obtained by a selective combination of the original frequency correlation based method and the proposed temporal correlation-based method. Performance of the proposed method is evaluated on the TIMIT speech corpus using various types of background noise conditions and the CU-Move in-vehicle speech corpus. Experimental results demonstrate that the proposed method is more effective at increasing speech recognition performance in adverse conditions. By employing the proposed temporal-frequency based reconstruction method, a +17.71% average relative improvement in word error rate (WER) is obtained for white, car, speech babble, and background music conditions over 5-, 10-, and 15-dB SNR, compared to the original frequency correlation-based method. We also obtain a +16.72% relative improvement in real-life in-vehicle conditions using data from the CU-Move corpus.