Skip to main content
Fig. 2 | EURASIP Journal on Audio, Speech, and Music Processing

Fig. 2

From: Quadratic approach for single-channel noise reduction

Fig. 2

PESQ and STOI scores of TIMIT speech signals as a function of the temporal smoothing preprocessing parameter τy for N=3 in the presence of white Gaussian noise: a PESQ scores with iSNR=− 5 dB, b PESQ scores with iSNR=15 dB, c STOI scores with iSNR=− 5 dB, and d STOI scores with iSNR=15 dB. The PESQ scores of the input noisy observation signal are 1.47 and 2.78 with iSNR=− 5 dB and iSNR=15 dB, respectively; their corresponding STOI scores are 0.61 and 0.97

Back to article page