Fact · Accessibility and health
In a 2026 qualitative study of four students with dyslexia using speech to text, the most frequent technology-based errors were misrecognized single words and omitted words or parts of sentences.
Checked against its source on October 16, 2026
In the source's own words
“The most frequent were misrecognition of single words and omission of words or parts of sentences.”
Technology-Based Writing: Qualitative Study of the Writing Process When Students With Dyslexia Use Speech-to-Text, Helle Bundgaard Svendsen et al., Dyslexia, 2026, Abstract.
Open the original (DOI 10.1002/dys.70024)- What it measures
- Students aged 10 to 12 wrote in a seven-week speech-to-text program, and their screencasts were transcribed and coded for errors.
- What it does not show
- Four students. The study found no conventional spelling errors, so the errors came from recognition, not from the students' spelling.