The GENEA Challenge 2022: A large evaluation of data-driven co-speech gesture generation

TL;DR

The GENEA Challenge 2022 evaluated data-driven co-speech gesture generation systems, revealing some synthetic gestures can surpass human motion in human-likeness but are less appropriate for speech, enabling direct comparison of methods.

Contribution

This study provides a large-scale, standardized benchmark for co-speech gesture generation, decoupling human-likeness from appropriateness and enabling direct system comparisons.

Findings

Synthetic gestures can be rated more human-like than real motion capture.

All synthetic gestures are less appropriate for speech than real motion.

The evaluation framework allows direct comparison of different gesture-generation methods.

Abstract

This paper reports on the second GENEA Challenge to benchmark data-driven automatic co-speech gesture generation. Participating teams used the same speech and motion dataset to build gesture-generation systems. Motion generated by all these systems was rendered to video using a standardised visualisation pipeline and evaluated in several large, crowdsourced user studies. Unlike when comparing different research papers, differences in results are here only due to differences between methods, enabling direct comparison between systems. This year's dataset was based on 18 hours of full-body motion capture, including fingers, of different persons engaging in dyadic conversation. Ten teams participated in the challenge across two tiers: full-body and upper-body gesticulation. For each tier we evaluated both the human-likeness of the gesture motion and its appropriateness for the specific speech signal. Our evaluations decouple human-likeness from gesture appropriateness, which previously was a major challenge in the field. The evaluation results are a revolution, and a revelation. Some synthetic conditions are rated as significantly more human-like than human motion capture. To the best of our knowledge, this has never been shown before on a high-fidelity avatar. On the other hand, all synthetic motion is found to be vastly less appropriate for the speech than the original motion-capture recordings. Additional material is available via the project website at https://youngwoo-yoon.github.io/GENEAchallenge2022/