Development of a lip-sync algorithm based on an audio-visual corpus.