Improving Non-Native Speakers’ Participation with an Automatic Agent

Context

  • At NTT Technology Labs in Japan, I had the opportunity to create a conversational robot to enhance multilingual communication.

  • Previous solutions failed because they did not fully account for the complexities of group dynamics.

  • After conducting thorough research, I proposed a novel approach that diverged significantly from prior work to design the robot.

  • I led a controlled experiment with 48 participants to test the robot, which showed it increased participation by 40%.

What Did I do?

  • Collaborated with a cross-functional team to design a conversational robot, increasing participation of Non-Native speakers by 40%. 

  • Designed quantitative experiments, created Qualtrics surveys and proved the effect of the conversational robot. 

  • Conducted in-depth user interviews and generated actionable design recommendations to inform future iterations. 

  • Published and presented ACM research article, which led to interdisciplinary discussion and news coverage

More details about this project.

Impact

  • Increased participation of Non-Native speakers by 40% .

  • Research insights directly employed to develop a new generation of conversational agent.

  • Reached wider audience through presentations at a large ACM conference, news coverage, and podcast.

Goals

  • Use technology to facilitate more balanced communication in multilingual groups.

  • Prove the effectiveness of the technology with quantitative experiments.

  • Gather user feedback with in-depth, qualitative interviews to inform the design of future platforms.