Skip to main content


Figure 5 | EURASIP Journal on Audio, Speech, and Music Processing

Figure 5

From: Towards expressive musical robots: a cross-modal framework for emotional gesture, voice and music

Figure 5

Experiment 1: Visualization of confusion matrices for gesture and voice. Intended emotion is shown in the titles, and the average percentage of raters that selected each emotion are given along the dimensional axes. Pointed triangles indicate that the one emotion was greatly perceived on average. Similar shapes for a given number indicate similar perceived emotion for both input gesture and output voice.

Back to article page