Animated Lombard speech: Motion capture, facial animation and visual intelligibility of speech produced in adverse conditions

► We have created the first (to our knowledge) 3D face animation of Lombard speech. ► An audio/video/motion capture corpus of 180 sentences recorded in quiet, noise and whispering drive the animation. ► A perceptual study shows audio-visual intelligibility of the animation to be on par with the recorded video. ► Lombard animation combined with neutral acoustic speech yields better intelligibility than neutral animation. ► A subjective test shows high acceptance of the face animations, also for AV incongruent stimuli.