Abstract: Many video-to-audio (VTA) methods have been proposed for dubbing silent AI-generated videos. An efficient quality assessment method for AI-generated audio-visual content (AGAV) is crucial for ensuring audio-visual quality. Existing audio-visual quality assessment methods struggle with unique distortions in AGAVs, such as unrealistic and inconsistent elements. To address this, we introduce **AGAVQA-3k**, the first large-scale AGAV quality assessment dataset, comprising $3,382$ AGAVs from $16$ VTA methods. AGAVQA-3k includes two subsets: AGAVQA-MOS, which provides multi-dimensional scores for audio quality, content consistency, and overall quality, and AGAVQA-Pair, designed for optimal AGAV pair selection. We further propose **AGAV-Rater**, a LMM-based model that can score AGAVs, as well as audio and music generated from text, across multiple dimensions, and selects the best AGAV generated by VTA methods to present to the user. AGAV-Rater achieves state-of-the-art performance on AGAVQA-3k, Text-to-Audio, and Text-to-Music datasets. Subjective tests also confirm that AGAV-Rater enhances VTA performance and user experience. The dataset and code is available at https://github.com/charlotte9524/AGAV-Rater.

Lay Summary:

Can LMMs be utilized to evaluate the quality of audio-visual content (AGAV) generated by video-to-audio (VTA) methods? Our goal is to adapt LMMs to score AGAVs like humans.We introduce AGAVQA-3k, the first large-scale AGAV quality assessment dataset, comprising 3,382 AGAVs from 16 VTA methods. We further propose AGAV-Rater, an LMM-based model that can score AGAVs, as well as audio and music generated from text, across multiple dimensions. Remarkably, AGAV-Rater achieves state-of-the-art performance and can help VTA methods select the highest-quality AGAVs to present to users.Our research contributes to the study of AGAVs' perceptual quality and demonstrates its potential for supervising and controlling the quality of AGAVs.