3M-TRANSFORMER: A Multi-Stage Multi-Stream Multimodal Transformer for Embodied Turn-Taking Prediction
This work addresses turn-taking prediction for human-computer/robot interaction, representing an incremental improvement over prior multimodal transformer approaches.
The paper tackles turn-taking prediction in multiparty conversations using synchronized multi-perspective egocentric data, achieving a performance improvement of up to 14.01% on average over existing baselines on the EgoCom dataset.
Predicting turn-taking in multiparty conversations has many practical applications in human-computer/robot interaction. However, the complexity of human communication makes it a challenging task. Recent advances have shown that synchronous multi-perspective egocentric data can significantly improve turn-taking prediction compared to asynchronous, single-perspective transcriptions. Building on this research, we propose a new multimodal transformer-based architecture for predicting turn-taking in embodied, synchronized multi-perspective data. Our experimental results on the recently introduced EgoCom dataset show a substantial performance improvement of up to 14.01% on average compared to existing baselines and alternative transformer-based approaches. The source code, and the pre-trained models of our 3M-Transformer will be available upon acceptance.