Skip to main content

Table 2 Results of speech quality, intelligibility and dereverberation measures for Scenario #2. The input signal refers to one of the microphones. NLMS refers to applying NLMS alone with any dereverberation algorithm. The values of LSD, SRMR and SER are in dB and the values of the STOI measure are the percentage of correct words

From: An online algorithm for echo cancellation, dereverberation and noise reduction based on a Kalman-EM Method

  LSD SRMR STOI ΔSER
Microphone signal 5.52 3.35 82.15 0
Proposed 3.54 7.40 95.57 23.16
NLMS 4.89 3.38 83.64 1.84
NLMS-MCSE 3.74 6.32 90.07 6.03
NLMS-RKEMD-1 3.89 5.28 71.81 11.04
NLMS-RKEMD-2 3.79 5.44 74.49 10.96