Peer-reviewed research
Just Speak It: Minimize Cognitive Load for Eyes-Free Text Editing with a Smart Voice Assistant
Jiayue Fan et al., Proceedings of UIST 2021, ACM Symposium on User Interface Software and Technology, 2021 · DOI 10.1145/3472749.3474795
Read the originalRead on October 16, 2026
- What it is
- Researchers at Tsinghua University compared ways of editing dictated text by voice without looking at a screen. Revising by speaking the target words again had a lower failure rate than describing the edit in a command or re-speaking the whole phrase.
- How much weight it carries
- Peer-reviewed UIST paper. This entry reads the abstract from the Semantic Scholar record for the paper's DOI, because the full PDF at the Tsinghua address could not be read by the quote checker on 2026-10-16. The full PDF (fan-2021-just-speak-it-smartedit) carries the same sentence. The study is eyes-free and uses earphone buttons, so it is not a desktop dictation test.
Facts taken from this source
- In the 2021 SmartEdit study, the authors' editing techniques lowered the average failure rate by 54.1 percent compared with descriptive commands or re-speaking.
“decrease the average failure rate by 54.1% compared to descriptive command or re-speaking” (Abstract)