Using Vision, Acoustics, and Natural Language for Disambiguation

doi 10.1145/1228716.1228727