Real-time long-form avatar audio--video generation requires causal, continuous synthesis while maintaining audiovisual synchronization and visual consistency. Adapting a pretrained bidirectional model to this setting presents two key dilemmas.
TEMPO is a unified model that adds temporally‑grounded capabilities to large audio‑language models, enabling timestamping of events, speakers, and sounds in audio, speech, and music. It introduces a supervised fine‑tuning stage featuring atomic timestamp tokens, a time‑aware projector with sinusoidal encodings, and a distance‑aware Gaussian loss, trained via a synthetic‑to‑real curriculum. Additionally, TEMPO employs reinforcement learning (GRPO) as a refinement step, and achieves state‑of‑the‑art performance on a benchmark of 10K samples across five timestamping tasks, surpassing Audio Flamingo Next and Qwen3‑Omni.
By Apoorva Kulkarni, Kaousheik Jayakumar, Sreyan Ghosh, Utathya Aich, Ramani Duraiswami, Dinesh Manocha
Recent advances in generative video modeling have enabled diverse generation, reference-based synthesis, extension, and editing, but existing approaches often rely on fragmented task-specific models. A general model must distinguish heterogeneous target, source, and reference signals to determine what to generate, preserve, or use as guidance, while reducing interference among tasks.
arXiv:2606. 01031v1 Announce Type: cross Abstract: Audio-driven talking-head generation has advanced rapidly, yet existing evaluation protocols mainly rely on frame-wise metrics that assume strict temporal correspondence between generated and reference videos.
By Zhicheng Zhang, Lei Wang, Yu Zhang, Yongsheng Gao
arXiv:2608. 04902v1 Announce Type: cross Abstract: Video-to-audio (V2A) generation extends image-to-audio generation (I2A) by introducing consecutive frames that provide essential temporal cues for audio synthesis.
By Zehua Chen, Junyou Wang, Yuxuan Jiang, Zhenying Fang, Yusheng Dai, Jianfei Chen, Ziwei Liu, Jun Zhu
arXiv:2602. 10230v2 Announce Type: replace Abstract: Audio language models process input audio into rich frame-level representations, but the standard approach to temporal localization generates timestamps as sequences of text tokens, which discards the frame-level representations in favor of autoregressive decoding.
By Joseph An, Phillip Keung, Jiaqi Wang, Orevaoghene Ahia, Noah A. Smith