Short Description

Backchannels such as “mhm”, “yeah”, or “aha” play an important role in making conversation feel natural. The social robot Furhat, developed by the Stockholm-based startup Furhat Robotics, offers a unique opportunity to investigate these aspects in human–robot interaction.

In this project, students will design and run an experiment where Furhat presents pre-selected speech stimuli (taken from a spontaneous speech corpus) to participants. The participants will then be instructed to produce backchannels at moments they find appropriate. The aim of the study is to better understand human backchanneling behaviour in human–robot interaction.

This work can build upon a previous study where only audio stimuli were used, but now extends the experimental setting to include the embodied interaction with Furhat.

1–2 students are welcome!

Your Tasks (depending on specific project):

• Implementing the speech stimuli into Furhat
• Conducting the experiment with participants
• Analyzing the data and reporting your results (thesis writing)

Your Profile

• Basic knowledge of sound engineering and/or speech communication
• Good programming skills (Python, advantage with Kotlin know-how)

Contact:

Michael Paierl (paierl@turgaz.at)
Barbara Schuppler (b.schuppler@tugraz.at)