Model-Guided Dual-Role Alignment for High-Fidelity Open-Domain Video-to-Audio Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Kang, Pham, Trung X., Lee, Suyeon, Niu, Axi, Senocak, Arda, Chung, Joon Son
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917047848402944
author Zhang, Kang
Pham, Trung X.
Lee, Suyeon
Niu, Axi
Senocak, Arda
Chung, Joon Son
author_facet Zhang, Kang
Pham, Trung X.
Lee, Suyeon
Niu, Axi
Senocak, Arda
Chung, Joon Son
contents We present MGAudio, a novel flow-based framework for open-domain video-to-audio generation, which introduces model-guided dual-role alignment as a central design principle. Unlike prior approaches that rely on classifier-based or classifier-free guidance, MGAudio enables the generative model to guide itself through a dedicated training objective designed for video-conditioned audio generation. The framework integrates three main components: (1) a scalable flow-based Transformer model, (2) a dual-role alignment mechanism where the audio-visual encoder serves both as a conditioning module and as a feature aligner to improve generation quality, and (3) a model-guided objective that enhances cross-modal coherence and audio realism. MGAudio achieves state-of-the-art performance on VGGSound, reducing FAD to 0.40, substantially surpassing the best classifier-free guidance baselines, and consistently outperforms existing methods across FD, IS, and alignment metrics. It also generalizes well to the challenging UnAV-100 benchmark. These results highlight model-guided dual-role alignment as a powerful and scalable paradigm for conditional video-to-audio generation. Code is available at: https://github.com/pantheon5100/mgaudio
format Preprint
id arxiv_https___arxiv_org_abs_2510_24103
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Model-Guided Dual-Role Alignment for High-Fidelity Open-Domain Video-to-Audio Generation
Zhang, Kang
Pham, Trung X.
Lee, Suyeon
Niu, Axi
Senocak, Arda
Chung, Joon Son
Sound
Artificial Intelligence
Multimedia
Audio and Speech Processing
We present MGAudio, a novel flow-based framework for open-domain video-to-audio generation, which introduces model-guided dual-role alignment as a central design principle. Unlike prior approaches that rely on classifier-based or classifier-free guidance, MGAudio enables the generative model to guide itself through a dedicated training objective designed for video-conditioned audio generation. The framework integrates three main components: (1) a scalable flow-based Transformer model, (2) a dual-role alignment mechanism where the audio-visual encoder serves both as a conditioning module and as a feature aligner to improve generation quality, and (3) a model-guided objective that enhances cross-modal coherence and audio realism. MGAudio achieves state-of-the-art performance on VGGSound, reducing FAD to 0.40, substantially surpassing the best classifier-free guidance baselines, and consistently outperforms existing methods across FD, IS, and alignment metrics. It also generalizes well to the challenging UnAV-100 benchmark. These results highlight model-guided dual-role alignment as a powerful and scalable paradigm for conditional video-to-audio generation. Code is available at: https://github.com/pantheon5100/mgaudio
title Model-Guided Dual-Role Alignment for High-Fidelity Open-Domain Video-to-Audio Generation
topic Sound
Artificial Intelligence
Multimedia
Audio and Speech Processing
url https://arxiv.org/abs/2510.24103