New Features Using Robust MVDR Spectrum of Filtered Autocorrelation Sequence for Robust Speech Recognition

Figure 5

Comparison of the average estimated subband SNRs obtained over the created compact database (containing both clean and noisy data at SNRs from 20 dB to −5 dB) for two cases of using the RAS filter and without it. The SNR in the figure refers to estimated subband SNRs at the output of Mel filter banks followed by equal loudness curve and power law of hearing, as shown in Figure 3(b).