Notes on transcript formatting, certification, and the court record.
· Court Transcript Platform
Our benchmark corpus tripled: six U.S. Supreme Court arguments, 8.4 hours of audio, 89,328 certified words. Every error graded for whether it changes the record, and every one counted as caught or missed by our review flags. The numbers, including the ones that don't flatter us.
· Court Transcript Platform
A per-account encryption tier for transcripts nobody here should be able to read. How the sealing works, what is ciphertext at rest, and the four limits we will not round up.
· Court Transcript Platform
A court hearing, a deposition, and a 911 call are all transcripts, and almost nothing about their pages is the same. What the apparatus actually differs in, and what supporting all three cost us in the schema.
· Court Transcript Platform
On August 10, 2026 the California Supreme Court ordered superior courts to provide an official verbatim record to litigants who cannot afford a private court reporter, by electronic recording when no reporter is available. What the ruling says, who it covers, and what it leaves unresolved.
· Court Transcript Platform
A trusted transcript, a courtroom, and a paragraph attributed to the wrong speaker. The embarrassment that started Court Transcripts App, and why we publish our accuracy numbers.
· Court Transcript Platform
We benchmarked our pipeline against Rev.com on two U.S. Supreme Court arguments, scored against the official certified transcripts. Fewer word errors before review, less than half after it. Fewer name errors. Full measurement reports available on request.
· Court Transcript Platform
Word accuracy, speaker attribution, and what human review buys — measured on certified court transcripts of public hearings. No adjectives, just the figures.
· Court Transcript Platform
Everyone says 99%. Here is what the number actually describes at each service, and why 'claimed' and 'measured' are different things — including how to apply both tests to us.
· Court Transcript Platform
Seven ways to transcribe the same two Supreme Court arguments — our three modes, extreme mode with its flags resolved by a human, each engine by itself, and Rev.com — ranked on one table by errors against the certified record.
· Court Transcript Platform
If a human resolved every word our system flags — and nothing else — how accurate is the transcript? We computed it against certified references: 98.6% combined, reviewing one word in twenty-five. Here is that number, its cost, and what the residual actually is.
· Court Transcript Platform
We put Rev.com's transcripts of the same two Supreme Court arguments through our own benchmark, added the attribution metrics legal work actually turns on, and measured whether feeding the engines the case caption helps. Two of three results flatter us. The third is the roadmap.
· Court Transcript Platform
We ran our pipeline against the Supreme Court's own certified transcripts — and found our own benchmark had been scoring the better tier as the worse one. Here are the corrected numbers, now including the three-engine tier.
· Court Transcript Platform
Line numbers, the box, running headers, Q&A indentation: why a transcript page looks the way it does, and what courts actually check.
· Court Transcript Platform
Where we write about transcript formatting, certification workflows, and what it takes to make a record filing-ready.