Towards Streaming Synchronized Spatial Audio Generation via Autoregressive Diffusion Transformer
This work is significant for users seeking immersive experiences in virtual reality and augmented reality by enabling real-time, high-fidelity spatial audio generation, representing an incremental improvement over existing methods.
This paper addresses the challenge of real-time, high-quality spatial audio generation from panoramic videos and text prompts, which is often hindered by latency and difficulty in capturing precise spatial information. The proposed SwanSphere framework achieves superior performance in both video-to-spatial and text-to-spatial audio generation tasks.
Real-time and accurate spatial audio generation is pivotal for delivering an immersive experience. However, existing spatial audio synthesis technologies are often encumbered by a tradeoff between generation quality and high inference latency, as well as difficulty in capturing precise spatial information from multimodal inputs. To address these challenges, we propose SwanSphere, a unified streaming framework for high-fidelity spatial audio generation from panoramic videos and text prompts. SwanSphere mainly makes the following contributions: 1) We introduce a causal autoregressive diffusion transformer architecture that enables streaming high-quality spatial audio generation. 2) We design a Spatial Video-Audio Contrastive (SVAC) learning strategy to align the video encoder with the acoustic domain, and further employ a multi-objective online direct preference optimization (ODPO) scheme, resulting in strong spatial perception and robust multimodal spatial audio synthesis. 3) To alleviate the current scarcity of spatial audio datasets, we also develop an automated annotation pipeline for generating detailed spatial captions. Experimental results demonstrate that SwanSphere achieves superior performance in both video-to-spatial and text-to-spatial audio generation tasks. Demos can be found at: https://swanaigc.github.io.