In a 2025 9to5Mac test of one 7-minute-31-second episode, Apple's speech API had a word error rate of 10.3 percent, against 1.5 percent for Whisper Large V3 Turbo.

Checked against its source on October 16, 2026

In the source's own words

“Whisper Large V3 Turbo 40 seconds 0.2% 1.5% Apple 9 seconds 1.9% 10.3%”

What it measures: The author ran the same episode through each engine on an M2 Pro MacBook Pro with 16GB of RAM and scored the transcripts with the same word error rate method. The author reached Apple's engine through Yap, a third-party app, not through the Dictation feature itself.
What it does not show: One recording, one author, one error-rate method, from 2025. Apple has since changed its dictation models, so the figure describes that test, not the Dictation you use today.