AURA: Always-On Understanding and Real-Time Assistance via Video Streams

TL;DR

AURA is an end-to-end streaming visual interaction framework that enables continuous video understanding and real-time assistance, supporting open-ended questions and proactive responses with state-of-the-art performance.

Contribution

It introduces a unified VideoLLM system for live video streams, integrating context management, data construction, and deployment optimization for stable long-horizon interaction.

Findings

Achieves state-of-the-art performance on streaming benchmarks.

Supports real-time question answering and proactive responses.

Runs at 2 FPS on high-end accelerators with integrated ASR and TTS.

Abstract

Video Large Language Models (VideoLLMs) have achieved strong performance on many video understanding tasks, but most existing systems remain offline and are not well-suited for live video streams that require continuous observation and timely response. Recent streaming VideoLLMs have made progress, yet current approaches often rely on decoupled trigger-response pipelines or are limited to captioning-style narration, reducing their effectiveness for open-ended question answering and long-horizon interaction. We propose AURA (Always-On Understanding and Real-Time Assistance), an end-to-end streaming visual interaction framework that enables a unified VideoLLM to continuously process video streams and support both real-time question answering and proactive responses. AURA integrates context management, data construction, training objectives, and deployment optimization for stable long-horizon streaming interaction. It achieves state-of-the-art performance on streaming benchmarks and supports a real-time demo system with ASR and TTS running at 2 FPS on two 80G accelerators. We release the AURA model together with a real-time inference framework to facilitate future research.