A Simple but Strong Baseline for Sounding Video Generation: Effective Adaptation of Audio and Video Diffusion Models for Joint Generation

TL;DR

This paper proposes a simple yet effective baseline for sounding video generation by adapting audio and video diffusion models with novel mechanisms to improve audio-video alignment, outperforming existing methods.

Contribution

It introduces two novel mechanisms, timestep adjustment and CMC-PE, to enhance audio-video synchronization in joint diffusion-based generation models.

Findings

The proposed methods improve audio-video alignment.

The model outperforms existing sounding video generation techniques.

The mechanisms enhance temporal consistency in generated videos.

Abstract

In this work, we build a simple but strong baseline for sounding video generation. Given base diffusion models for audio and video, we integrate them with additional modules into a single model and train it to make the model jointly generate audio and video. To enhance alignment between audio-video pairs, we introduce two novel mechanisms in our model. The first one is timestep adjustment, which provides different timestep information to each base model. It is designed to align how samples are generated along with timesteps across modalities. The second one is a new design of the additional modules, termed Cross-Modal Conditioning as Positional Encoding (CMC-PE). In CMC-PE, cross-modal information is embedded as if it represents temporal position information, and the embeddings are fed into the model like positional encoding. Compared with the popular cross-attention mechanism, CMC-PE provides a better inductive bias for temporal alignment in the generated data. Experimental results validate the effectiveness of the two newly introduced mechanisms and also demonstrate that our method outperforms existing methods.