Advanced processing

Standard transcription produces an accurate transcript without telling you who said what. Advanced processing re-transcribes the meeting with speaker detection, attributing each line to a speaker and improving coverage on difficult audio.

It runs per meeting rather than as a global setting, so you can spend the extra processing time only on the calls where attribution actually matters — a design debate with five people, rather than a solo voice memo.

Speaker labels flow through to transcript exports and share pages. Advanced processing is available on Team plans and above, from the meeting page.

Something unclear or missing? Tell us and we'll fix the page.