FoleyBench: A Benchmark For Video-to-Audio Models

TL;DR

FoleyBench is a new large-scale benchmark dataset designed specifically for evaluating video-to-audio models in Foley sound generation, addressing the mismatch in existing datasets and enabling more accurate assessment of audio-visual alignment and sound quality.

Contribution

This paper introduces FoleyBench, the first dedicated benchmark for Foley-style video-to-audio generation, with a large, diverse dataset and comprehensive evaluation metrics.

Findings

Existing datasets have poor audio-visual correspondence for Foley tasks.

FoleyBench covers a wider range of Foley sound categories.

State-of-the-art models show room for improvement in audio-visual alignment.

Abstract

Video-to-audio generation (V2A) is of increasing importance in domains such as film post-production, AR/VR, and sound design, particularly for the creation of Foley sound effects synchronized with on-screen actions. Foley requires generating audio that is both semantically aligned with visible events and temporally aligned with their timing. Yet, there is a mismatch between evaluation and downstream applications due to the absence of a benchmark tailored to Foley-style scenarios. We find that 74% of videos from past evaluation datasets have poor audio-visual correspondence. Moreover, they are dominated by speech and music, domains that lie outside the use case for Foley. To address this gap, we introduce FoleyBench, the first large-scale benchmark explicitly designed for Foley-style V2A evaluation. FoleyBench contains 5,000 (video, ground-truth audio, text caption) triplets, each featuring visible sound sources with audio causally tied to on-screen events. The dataset is built using an automated, scalable pipeline applied to in-the-wild internet videos from YouTube-based and Vimeo-based sources. Compared to past datasets, we show that videos from FoleyBench have stronger coverage of sound categories from a taxonomy specifically designed for Foley sound. Each clip is further labeled with metadata capturing source complexity, UCS/AudioSet category, and video length, enabling fine-grained analysis of model performance and failure modes. We benchmark several state-of-the-art V2A models, evaluating them on audio quality, audio-video alignment, temporal synchronization, and audio-text consistency. Samples are available at: https://gclef-cmu.org/foleybench