Skip to main content

Table 2 Classification results of different algorithms with the context voting ensemble learning (-v) at the second-stage TsEL (unit: %)

From: Robust sound event classification with bilinear multi-column ELM-AE and two-stage ensemble learning

  Clean 20 dB 10 dB 0 dB Mean
DNN 98.87 95.33 92.40 78.87 91.37
ELM 93.76 ± 0.61 91.35 ± 0.62 86.99 ± 1.06 72.81 ± 1.29 86.23 ± 0.90
ELM-AE 95.53 ± 2.35 92.28 ± 1.67 89.40 ± 1.70 81.93 ± 1.86 89.78 ± 1.90
B-ELM-AE 98.23 ± 0.57 94.38 ± 0.67 92.50 ± 1.25 84.43 ± 1.76 92.39 ± 1.06
MC-ELM-AE-C 96.93 ± 0.65 93.70 ± 0.97 91.33 ± 0.83 83.25 ± 1.16 91.30 ± 0.90
MC-ELM-AE-TsEL-V 97.23 ± 0.70 93.90 ± 0.67 91.50 ± 1.22 84.20 ± 1.29 91.71 ± 0.97
MC-ELM-AE-TsEL-W 97.95 ± 0.62 95.35 ± 0.76 92.38 ± 1.49 84.78 ± 2.07 92.62 ± 1.24
B-MC-ELM-AE-C 98.13 ± 0.36 94.33 ± 0.73 92.88 ± 0.90 84.45 ± 1.36 92.45 ± 0.84
B-MC-ELM-AE-TsEL-V 98.33 ± 0.57 94.73 ± 0.58 92.98 ± 0.99 86.23 ± 1.54 93.06 ± 0.92
B-MC-ELM-AE-TsEL-W 98.95 ± 0.52 96.45 ± 0.55 94.15 ± 1.27 86.93 ± 1.43 94.12 ± 0.94