On the use of speech parameter contours for emotion recognition

EURASIP Journal on Audio, Speech, and Music Processing

Table 11 Top 5 UARs obtained on the Aibo corpus and the corresponding feature set utilised

Feature set $ν^{(i)}$	UAR (%)
$ν^{(i)} = [s_{F_{0}}^{(i)}, b_{F_{0}}^{(i)}, m_{F_{0}}^{(i)}, m_{F_{g}}^{(i)}, m_{A_{g}}^{(i)}, s_{F_{1}}^{(i)}, b_{F_{1}}^{(i)}, s_{A_{1}}^{(i)}, b_{A_{1}}^{(i)}]$	40.3
$ν^{(i)} = [b_{F_{0}}^{(i)}, m_{F_{0}}^{(i)}, m_{A_{g}}^{(i)}, s_{F_{1}}^{(i)}, b_{F_{1}}^{(i)}, s_{A_{1}}^{(i)}, b_{A_{1}}^{(i)}, τ^{(i)}]$	39.9
$ν^{(i)} = [s_{F_{0}}^{(i)}, b_{F_{0}}^{(i)}, m_{F_{g}}^{(i)}, m_{A_{g}}^{(i)}, m_{F_{1}}^{(i)}, m_{A_{1}}^{(i)}, τ^{(i)}]$	39.9
$ν^{(i)} = [s_{F_{0}}^{(i)}, m_{F_{0}}^{(i)}, m_{A_{g}}^{(i)}, m_{F_{1}}^{(i)}, m_{A_{1}}^{(i)}, τ^{(i)}]$	39.8
$ν^{(i)} = [s_{F_{0}}^{(i)}, m_{F_{g}}^{(i)}, m_{A_{g}}^{(i)}, m_{F_{1}}^{(i)}, m_{A_{1}}^{(i)}, τ^{(i)}]$	39.8
$ν^{(i)} = [s_{F_{0}}^{(i)}, b_{F_{0}}^{(i)}, m_{F_{g}}^{(i)}, m_{A_{g}}^{(i)}, m_{F_{1}}^{(i)}, m_{F_{2}}^{(i)}, m_{F_{3}}^{(i)}, m_{A_{1}}^{(i)}, m_{A_{2}}^{(i)}, m_{A_{3}}^{(i)}, τ^{(i)}]$	37.5