Skip to main content

Table 4 The performance of the model with additional features

From: Automatic segmentation of infant cry signals using hidden Markov models

Features Acccuracy F score (%)
  (%) Ins Exp
MFCCs 87.5 ±1.1 41.5 ±2.8 80.6 ±1.4
MFCCs + Δ and ΔΔ 88.6 ±0.9 45.9 ±2.5 82.2 ±1.1
MFCCs + F0 88.5 ±0.9 47.8 ±2.9 81.6 ±1.3
MFCCs + F0 + 88.8 ±0.9 47.1 ±2.5 82.3 ±1.1
Δ and ΔΔ    
MFCCs + ap0 89.2 ±0.9 48.9 ±2.8 83.3 ±1.3
MFCCs + ap0+ 89.0 ± 0.8 47.4 ±2.5 82.8 ±1.2
Δ and ΔΔ    
  1. The expiratory and inspiratory phase classes are denoted by Exp and Ins, respectively