Skip to main content

Table 2 Recognition rate (%) obtained by proposed and standard methods with suburban train noise

From: A bio-inspired feature extraction for robust speech recognition

  Recognition rate with HMM-4-GM
  SNR level PLPaGc PLP LPCC LPC MFCC
  0 dB 38.55 27.77 21.79 11.86 26.95
  5 dB 65.59 50.16 40.48 13.62 49.42
Suburban train noise 10 dB 84.71 72.74 60.96 18.47 71.66
  15 dB 92.74 85.82 77.90 28.96 86.30
  20 dB 95.77 91.72 87.06 41.96 92.60
  Average 75.47 65.64 57.64 22.97 65.39