Speech Interruption

Speech interruption, the act of one speaker beginning before another finishes, is a multifaceted research area focusing on its automatic detection and impact across various contexts. Current research employs machine learning models, including convolutional neural networks, recurrent neural networks (like LSTMs), and transformer-based architectures (like RuBERT and WavLM), to classify interruption types (cooperative vs. competitive) and filter out unwanted speech in real-time applications such as human-robot interaction and remote meetings. Understanding the costs and benefits of interruptions, particularly in human-robot teaming and multi-robot supervision, is crucial for optimizing collaborative systems and improving user experience. This work has implications for improving the design of conversational AI, human-computer interfaces, and collaborative robotics.