2017 INTERSPEECH INTERSPEECH 2017

Motion Analysis in Vocalized Surprise Expressions

Abstract

The background of our research is the generation of natural human-like motions during speech in android robots that have a highly human-like appearance. Mismatches in speech and motion are sources of unnaturalness, especially when emotion expressions are involved. Surprise expressions often occur in dialogue interactions, and they are often accompanied by verbal interjectional utterances. In this study, we analyze facial, head and body motions during several types of vocalized surprise expressions appearing in human-human dialogue interactions. The analysis results indicate an inter-dependence between motion types and different types of surprise expression (such as emotional, social or quoted) as well as different degrees of surprise expression. The synchronization between motion and surprise utterances is also analyzed.

πŸŒ‰ Interdisciplinary Bridge β€” Artificial Intelligence and Computer Vision
🧭 Keyword Pioneer β€” body motion
🐣 Hot Topic Early Bird β€” emotion recognition
🐝 Cross-Pollinator β€” Artificial Intelligence, Computer Science, Computer Vision, Data Science & Analytics, Deep Learning, Healthcare & Medicine, Interdisciplinary, Knowledge & Reasoning, Machine Learning, Mathematics & Optimization, Natural Language Processing, Reinforcement Learning, Robotics, Speech & Audio