AnyTalker: Scaling Multi-Person Talking Video Generation with Interactivity Refinement
This work addresses the challenge of generating realistic multi-person talking videos for applications in video synthesis and human-computer interaction, representing an incremental improvement over existing methods.
The paper tackles the problem of generating multi-person talking videos from audio by addressing high data collection costs and achieving coherent interactivity, resulting in a framework that achieves remarkable lip synchronization, visual quality, and natural interactivity with favorable data cost and scalability.
Recently, multi-person video generation has started to gain prominence. While a few preliminary works have explored audio-driven multi-person talking video generation, they often face challenges due to the high costs of diverse multi-person data collection and the difficulty of driving multiple identities with coherent interactivity. To address these challenges, we propose AnyTalker, a multi-person generation framework that features an extensible multi-stream processing architecture. Specifically, we extend Diffusion Transformer's attention block with a novel identity-aware attention mechanism that iteratively processes identity-audio pairs, allowing arbitrary scaling of drivable identities. Besides, training multi-person generative models demands massive multi-person data. Our proposed training pipeline depends solely on single-person videos to learn multi-person speaking patterns and refines interactivity with only a few real multi-person clips. Furthermore, we contribute a targeted metric and dataset designed to evaluate the naturalness and interactivity of the generated multi-person videos. Extensive experiments demonstrate that AnyTalker achieves remarkable lip synchronization, visual quality, and natural interactivity, striking a favorable balance between data costs and identity scalability.