EgoSpeak: Learning When to Speak for Egocentric Conversational Agents in the Wild

TL;DR

EgoSpeak is a real-time framework that predicts when egocentric conversational agents should speak, using first-person video and multimodal data to improve natural interactions in complex environments.

Contribution

The paper introduces EgoSpeak, a novel approach for speech initiation prediction in egocentric videos, and provides a new dataset YT-Conversation for large-scale pretraining.

Findings

EgoSpeak outperforms baselines in real-time speech initiation tasks.

Multimodal input improves prediction accuracy.

Longer context enhances decision quality.

Abstract

Predicting when to initiate speech in real-world environments remains a fundamental challenge for conversational agents. We introduce EgoSpeak, a novel framework for real-time speech initiation prediction in egocentric streaming video. By modeling the conversation from the speaker's first-person viewpoint, EgoSpeak is tailored for human-like interactions in which a conversational agent must continuously observe its environment and dynamically decide when to talk. Our approach bridges the gap between simplified experimental setups and complex natural conversations by integrating four key capabilities: (1) first-person perspective, (2) RGB processing, (3) online processing, and (4) untrimmed video processing. We also present YT-Conversation, a diverse collection of in-the-wild conversational videos from YouTube, as a resource for large-scale pretraining. Experiments on EasyCom and Ego4D demonstrate that EgoSpeak outperforms random and silence-based baselines in real time. Our results also highlight the importance of multimodal input and context length in effectively deciding when to speak.