Speech-to-text API, coming soon

Who said what, and when.

Send a whole meeting, interview, or episode. Get back every line with its speaker and timestamp, as JSON, SRT, or VTT.

episode-41.vtt

WEBVTT

SPEAKER_00

SPEAKER_01

SPEAKER_02

Send the whole file

Hours of audio in one request, not cut into short clips first.

Each segment carries a speaker label, so you can follow a conversation without listening back.

Structured segments for your code, or caption files ready for a video player.