AI RESEARCH
Speech-Synchronized Whiteboard Generation via VLM-Driven Structured Drawing Representations
arXiv CS.LG
•
ArXi:2603.25870v1 Announce Type: cross Creating whiteboard-style educational videos demands precise coordination between freehand illustrations and spoken narration, yet no existing method addresses this multimodal synchronization problem with structured, reproducible drawing representations. We present the first dataset of 24 paired Excalidraw nstrations with narrated audio, where every drawing element carries millisecond-precision creation spanning 8 STEM domains.