Fact · Dictation software
0.50percent mean word error rate
When linguistic errors were excluded, the Whisper-based model had a mean word error rate of 0.50 percent.
Checked against its source on October 16, 2026
In the source's own words
“When excluding linguistic errors, Whisper outperformed Dragon with a mean WER of 0.50% versus 1.34%”
The use of generative artificial intelligence-based dictation in a neurosurgical practice: a pilot study, Benjamin S. Hopkins et al., Neurosurgical Focus, 2025, Results.
Open the original (DOI 10.3171/2025.4.FOCUS24834)- What it measures
- A secondary result reported in the study's results section, alongside the all-error figures.
- What it does not show
- This is a secondary result from a small pilot. The paper's own definition of linguistic errors matters for the comparison.