Fig. 4From: Transforming an embodied conversational agent into an efficient talking head: from keyframe-based animation to multimodal concatenation synthesisVariation of the ten articulatory parameter values for the sentence: “Look at the windows and see if it’s raining”. The variations are nonlinear and cannot be replicated with linear interpolation animation. The database is segmented into phonemes. A multimodal diphone dictionary is created with this segmented databaseBack to article page