AI scribe accuracy benchmark — Australia
The methodology behind these results, including how each number was measured and its limitations, is in the methodology document.
Against six other AI scribes on the same recordings, Hanah's notes captured 11% more of the required clinical details than the average (93.7% against 84.2%), with no major hallucination flags against an average of 0.96 per note, and Hanah's transcripts had 66% fewer clinically significant errors. See the comparison.
Headline metrics
Coverage breakdown
Jury composition
Each detail is graded by every juror configured in the panel. Verdicts disagree on hard cases — that's the signal. Per-juror reasoning is shown verbatim below.
Sessions
Click any recording to drill in: source audio, ground-truth transcript, what the STT pipeline heard, every generated document, and every per-LLM verdict with its reasoning.
Methodology & provenance
Word Error Rate, panel-of-LLM jury, prompt-scoped expectations, calibration sanity-check, and corpus graduation roadmap (synthetic → role-play → real-world).
Read the methodology document →The full machine-readable artefact this page renders from. Every number on this page is computable directly from this file.
Open raw data →Corpus, juror prompts, and judge model versions are all version-controlled. Anyone can re-run the harness against the same commit and reproduce these numbers.
Trust Center home →