T2AV-Compass: Towards Unified Evaluation for Text-to-Audio-Video Generation

TL;DR

This paper introduces T2AV-Compass, a comprehensive benchmark and evaluation framework for text-to-audio-video generation, addressing fragmented assessment methods and revealing significant gaps in current model performance.

Contribution

It presents a unified benchmark with diverse prompts and a dual-level evaluation framework combining objective metrics and subjective judgment for T2AV systems.

Findings

Current models underperform in realism and cross-modal alignment

Persistent issues in audio realism and synchronization

Benchmark reveals significant room for improvement

Abstract

Text-to-Audio-Video (T2AV) generation aims to synthesize temporally coherent video and semantically synchronized audio from natural language, yet its evaluation remains fragmented, often relying on unimodal metrics or narrowly scoped benchmarks that fail to capture cross-modal alignment, instruction following, and perceptual realism under complex prompts. To address this limitation, we present T2AV-Compass, a unified benchmark for comprehensive evaluation of T2AV systems, consisting of 500 diverse and complex prompts constructed via a taxonomy-driven pipeline to ensure semantic richness and physical plausibility. Besides, T2AV-Compass introduces a dual-level evaluation framework that integrates objective signal-level metrics for video quality, audio quality, and cross-modal alignment with a subjective MLLM-as-a-Judge protocol for instruction following and realism assessment. Extensive evaluation of 11 representative T2AVsystems reveals that even the strongest models fall substantially short of human-level realism and cross-modal consistency, with persistent failures in audio realism, fine-grained synchronization, instruction following, etc. These results indicate significant improvement room for future models and highlight the value of T2AV-Compass as a challenging and diagnostic testbed for advancing text-to-audio-video generation.