Skip to main content
. Author manuscript; available in PMC: 2012 Mar 1.
Published in final edited form as: Speech Commun. 2011 Mar 1;53(3):340–354. doi: 10.1016/j.specom.2010.10.005

Figure 3.

Figure 3

Top two panels show spectrograms of a sentence in quiet and corrupt sentence processed by the RDC spectral-subtractive algorithm [22] respectively. The input sentence was corrupted by babble at 0 dB SNR. Bottom panel shows the frame SNRLOSS values. The average SNRLOSS was 0.86 (SNRLIM was set to 5 dB).