--- license: mit task_categories: - video-text-to-text - question-answering language: - en tags: - realtime - multimodal - audio-video - streaming pretty_name: Aero Realtime Data --- # Aero Realtime Data
Training data for [Aero Realtime](https://huggingface.co/kcz358/aero-realtime-4B), a native proactive audio-video language model whose input and output share one aligned autoregressive stream. ## ⚠️ This is only part of the data This repository hosts **only the self-packaged assets** that are not available on an official Hugging Face release: - `qaego4d` videos and TTS audio - all TTS audio for `EgoIT-99K` - the `lmms_engine_parquet` / `sft_parquet` manifests used by the training configs The full dataset additionally pulls from these **official public releases**, which are downloaded separately to keep this repo lightweight and avoid re-hosting: - [`lmms-lab/LLaVA-Video-178K`](https://huggingface.co/datasets/lmms-lab/LLaVA-Video-178K) - [`lmms-lab/EgoIT-99K`](https://huggingface.co/datasets/lmms-lab/EgoIT-99K) - [`chenjoya/Live-WhisperX-526K`](https://huggingface.co/datasets/chenjoya/Live-WhisperX-526K) ## Preparing the full dataset Use the one-shot preparation script in the code repository to assemble the complete blob layout the stage1/stage2 configs expect. It downloads the sources above, merges in the assets from this repo, and extracts every archive in place. ```bash git clone https://github.com/kcz358/aero-realtime.git cd aero-realtime bash scripts/data/prepare_data.sh --dest /scratch/aero-ws/data ``` Re-running is cheap: finished datasets are skipped via marker files. Select a subset with `--only llava,egoit,livecc,assets` or force a refresh with `--force`. See `scripts/data/prepare_data.sh --help` and the header of that script for the full environment overrides and the resulting on-disk layout. ## Citation If you find this work useful, please cite: ```bibtex @misc{zhang2026aerorealtimefullyaligned, title={Aero Realtime: Fully Aligned Input-Output Streams for Low-Latency Streaming Multimodal Generation}, author={Kaichen Zhang and Wei Huang and Keming Wu and Bo Li and Xiaojuan Qi}, year={2026}, eprint={2608.08469}, archivePrefix={arXiv}, primaryClass={cs.AI}, url={https://arxiv.org/abs/2608.08469} } ```