----- If you use any data contained in this subdirectory, please cite the following: -----
Ferstl, Ylva, Michael Neff, and Rachel McDonnell. "Understanding the predictability of gesture parameters from speech and their perceptual importance." Proceedings of the 20th ACM International Conference on Intelligent Virtual Agents. 2020.

BibTeX:
@inproceedings{ferstl2020understanding,
  title={Understanding the predictability of gesture parameters from speech and their perceptual importance},
  author={Ferstl, Ylva and Neff, Michael and McDonnell, Rachel},
  booktitle={Proceedings of the 20th ACM International Conference on Intelligent Virtual Agents},
  pages={1--8},
  year={2020}
}

--- Instructions for use:
The gesture_parameter_values folder contains the ground truth calculated parameter values for all gestures, for both datasets.
The automatic_gesture_parameter_estimation folder contains the python script to estimate gesture parameters from corresponding speech audio segments. This uses the trained models contained in the trained_models folder. (Model predicitions represent normalized data; the script accesses the de-normalization statistics from the prediction_denormalization_stats folder.)