Skip to main content

Table 1 Classification results of different algorithms with the mean probability value ensemble learning (-b) 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 96.73 94.60 90.27 76.47 89.52
ELM 97.08 ± 0.45 90.31 ± 0.92 85.75 ± 1.16 72.39 ± 1.18 86.38 ± 0.93
ELM-AE 97.68 ± 1.31 94.25 ± 1.53 92.28 ± 2.04 86.63 ± 1.43 92.71 ± 1.58
B-ELM-AE 99.08 ± 0.34 95.88 ± 0.49 94.93 ± 0.93 89.85 ± 0.85 94.93 ± 0.65
MC-ELM-AE-C 98.63 ± 0.46 95.88 ± 0.67 94.35 ± 0.86 87.90 ± 1.27 94.19 ± 0.82
MC-ELM-AE-TsEL-V 98.53 ± 0.52 95.63 ± 0.55 93.73 ± 0.91 88.33 ± 1.48 94.05 ± 0.86
MC-ELM-AE-TsEL-W 99.03 ± 0.34 96.55 ± 0.52 95.53 ± 0.92 89.08 ± 1.61 95.05 ± 0.85
B-MC-ELM-AE-C 99.08 ± 0.35 96.08 ± 0.69 94.65 ± 1.07 89.93 ± 1.19 94.94 ± 0.83
B-MC-ELM-AE-TsEL-V 99.15 ± 0.31 96.20 ± 0.42 95.10 ± 0.88 90.45 ± 0.93 95.23 ± 0.63
B-MC-ELM-AE-TsEL-W 99.45 ± 0.30 97.00 ± 0.46 96.15 ± 1.26 91.63 ± 1.29 96.06 ± 0.83