GaTech Center for Music Technology director Gil Weinberg and PhD candidate Richard Savery pose with a Shimi robot. (source: Georgia Tech)

Georgia Tech’s robotic music lab awarded new National Science Foundation grant

EarRelevant Staff | 02 SEP 2019

ATLANTA, GA—  The Georgia Tech School of Music announced on Friday that the National Science Foundation (NSF) has awarded Georgia Tech Center for Music Technology director Gil Weinberg and PhD candidate Richard Savery a new research grant totaling $669,912 for the purpose of researching trust and emotions between robots and humans by means of non-verbal communication.

The premise is that as “co-robots” become prevalent at home, work, and public spaces, they need to become trust-worthy and socially believable agents in order to become accepted and integrated into human society.

Prosodic features such as pitch, loudness, tempo, timbre, and rhythm bear strong resemblance to musical features, which can inform a novel approach for generative emotion-driven robotic prosody. This project will perform fundamental research contributing to the establishment of trust between humans and robots through the development of novel, non-linguistic emotional communication channels and implement them in a group of newly developed personal robots.

  • AD Spivey Hall Children's Choir 23-12-08
  • JSCO - Earrelevant Ad (1)

The first phase of this project will focus on developing machine learning techniques to derive features from a newly created emotionally-labeled musical dataset. It will use these features to drive a non-linguistic robotic voice synthesizer that conveys emotional content and builds trust. The results of this study will be integrated with previous work on conveying robotic emotions through physical gestures.

The second phase of the project will focus on experiments that study users’ preference to a variety of robotic emotional responses when interacting with a single robot. It will use the learned features to design a larger scale robotic emotional contagion engine in an effort to improve and enrich emotion-driven human interaction with large groups of robots.

AD ECMSA 2023-24 season

Results are expected to lead to novel approaches for creating open, meaningful interactions between groups of humans and robots in large scale interactive scenarios such as personal robots in private and public spaces, work place training, education and combat. The project will take an interdisciplinary approach to these goals, addressing fields such as cognitive science, communication and music while fostering progress in science and engineering.

Awarded under the NSF’s National Robotics Initiative, the nearly $670 thousand grant, spans a two year research period beginning this November 1 and ending October 31, 2022.

The absract,“Creating Trust Between Groups of Humans and Robots Using a Novel Music Driven Robotic Emotion Generator,” can be accessed on the NSF website. ■