The Impact of Background Speech on Interruption Detection in Collaborative Groups

2025-07-09

AI Summary

This study addresses the challenge of overlapping speech interference in collaborative learning classrooms, where multiple concurrent group dialogues impede accurate detection of conversational interruptions. Departing from conventional interruption detection paradigms—typically designed for single-turn, clean-audio settings—we propose a robust multimodal interruption detection framework tailored to authentic pedagogical contexts. Our approach integrates acoustic signal processing with linguistically and prosodically informed features. Crucially, it constitutes the first systematic investigation into the acoustic and discourse-level characteristics of interruptions under overlapping speech conditions. Evaluated on real-world classroom recordings featuring multi-group collaboration, the model achieves significant improvements in interruption detection accuracy and demonstrates strong generalizability across speakers and robustness to speech overlap. These findings provide both deployable technical solutions and foundational theoretical insights for group dialogue analysis, intelligent instructional feedback systems, and educational process modeling.

Technology Category

• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Interruption plays a crucial role in collaborative learning, shaping group interactions and influencing knowledge construction. AI-driven support can assist teachers in monitoring these interactions. However, most previous work on interruption detection and interpretation has been conducted in single-conversation environments with relatively clean audio. AI agents deployed in classrooms for collaborative learning within small groups will need to contend with multiple concurrent conversations -- in this context, overlapping speech will be ubiquitous, and interruptions will need to be identified in other ways. In this work, we analyze interruption detection in single-conversation and multi-group dialogue settings. We then create a state-of-the-art method for interruption identification that is robust to overlapping speech, and thus could be deployed in classrooms. Further, our work highlights meaningful linguistic and prosodic information about how interruptions manifest in collaborative group interactions. Our investigation also paves the way for future works to account for the influence of overlapping speech from multiple groups when tracking group dialog.

Problem

• Detecting interruptions in multi-group collaborative settings with overlapping speech
• Developing robust AI methods for interruption identification in noisy classrooms
• Analyzing linguistic and prosodic cues of interruptions in group interactions

Innovation

• Robust interruption detection in overlapping speech
• AI-driven method for multi-group dialogue settings
• Linguistic and prosodic analysis of interruptions