GENEA Challenge 2022: Co-Speech Gesture Generation

When a game character gestures as it talks, does it feel lifelike? And if you had to benchmark just how realistic the gestures were, how would you do that?

Check out the paper here: https://www.ea.com/seed/news/genea-ch...

Benchmarking the realism of co-speech gestures is the purpose of the GENEA Challenge. Co-organized by SEED’s own Taras, the challenge featured teams from all over the world who competed to make the most realistic automatically-generated human gestures based on recorded speech.

The evaluation results were 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.

----------

SEED is a pioneering group within Electronic Arts, combining creativity with applied research. We explore, build, and help determine the future of interactive entertainment.

Learn more about SEED at https://seed.ea.com

Find us on:

Twitter:   / seed  
LinkedIn:   / seed-ea