Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2512.11423 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917370939834368 |
|---|---|
| author | Li, Chaochao Wang, Ruikui Zhou, Liangbo Feng, Jinheng Luo, Huaishao Zhang, Huan Wu, Youzheng He, Xiaodong |
| author_facet | Li, Chaochao Wang, Ruikui Zhou, Liangbo Feng, Jinheng Luo, Huaishao Zhang, Huan Wu, Youzheng He, Xiaodong |
| contents | Existing DiT-based audio-driven avatar generation methods have achieved considerable progress, yet their broader application is constrained by limitations such as high computational overhead and the inability to synthesize long-duration videos. Autoregressive methods address this problem by applying block-wise autoregressive diffusion methods. However, these methods suffer from the problem of error accumulation and quality degradation. To address this, we propose JoyStreamer-Flash, an audio-driven autoregressive model capable of real-time inference and infinite-length video generation with the following contributions: (1) Progressive Step Bootstrapping (PSB), which allocates more denoising steps to initial frames to stabilize generation and reduce error accumulation; (2) Motion Condition Injection (MCI), enhancing temporal coherence by injecting noise-corrupted previous frames as motion condition; and (3) Unbounded RoPE via Cache-Resetting (URCR), enabling infinite-length generation through dynamic positional encoding. Our 1.3B-parameter causal model achieves 16 FPS on a single GPU and achieves competitive results in visual quality, temporal consistency, and lip synchronization. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_11423 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | JoyStreamer-Flash: Real-time and Infinite Audio-Driven Avatar Generation with Autoregressive Diffusion Li, Chaochao Wang, Ruikui Zhou, Liangbo Feng, Jinheng Luo, Huaishao Zhang, Huan Wu, Youzheng He, Xiaodong Computer Vision and Pattern Recognition Existing DiT-based audio-driven avatar generation methods have achieved considerable progress, yet their broader application is constrained by limitations such as high computational overhead and the inability to synthesize long-duration videos. Autoregressive methods address this problem by applying block-wise autoregressive diffusion methods. However, these methods suffer from the problem of error accumulation and quality degradation. To address this, we propose JoyStreamer-Flash, an audio-driven autoregressive model capable of real-time inference and infinite-length video generation with the following contributions: (1) Progressive Step Bootstrapping (PSB), which allocates more denoising steps to initial frames to stabilize generation and reduce error accumulation; (2) Motion Condition Injection (MCI), enhancing temporal coherence by injecting noise-corrupted previous frames as motion condition; and (3) Unbounded RoPE via Cache-Resetting (URCR), enabling infinite-length generation through dynamic positional encoding. Our 1.3B-parameter causal model achieves 16 FPS on a single GPU and achieves competitive results in visual quality, temporal consistency, and lip synchronization. |
| title | JoyStreamer-Flash: Real-time and Infinite Audio-Driven Avatar Generation with Autoregressive Diffusion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2512.11423 |