01Orality: A Semantic Canvas for Externalizing and Clarifying Thoughts with SpeechSpeech becomes editable spatial material, not a transcript to clean up later.Presented by Feng Liang, June 8thVenueCHI 2026ACM Digital Library LinkAuthorsCity University of Hong KongWengxi Li, Jingze Tian, Can Liu (劉燦).Why careLLM voice needs structureVoice is important in meetings.Voice is becoming important in vibe coding.Current voice/chat artifacts stay too linear for sensemaking.Novelty/ContributionSpeech becomes a semantic canvasThe system turns spoken thought into editable spatial objects with embedded AI scaffolds.Figure 1: semantic canvas, speech input, AI stimulation, history, and export.
Why careLLM voice needs structureVoice is important in meetings.Voice is becoming important in vibe coding.Current voice/chat artifacts stay too linear for sensemaking.
Novelty/ContributionSpeech becomes a semantic canvasThe system turns spoken thought into editable spatial objects with embedded AI scaffolds.
