Skip to content

Speech to transcript

Recorded speech in.A transcript out.

Speakers separated, and named where the recording says who they are. Readable on a phone, copyable six ways, fetchable by any program.

or drop a file · or record on your phone

Three ways in

  • A YouTube link

    Public or unlisted. We fetch the audio and nothing else.

  • An audio or video file

    Large files transfer resumably. The upload bar is real: it knows the size.

  • A conversation, live

    Record on your phone and put it face-down. If the connection drops, the take stays on the phone.

Six formats

  • Prose
  • Speakers
  • Timestamped
  • SRT
  • VTT
  • JSON

What is promised

  • Source audio is deleted within 24 hours.The transcript stays until you delete it.
  • It doesn’t summarize, translate or edit.You get the words that were said, by who said them.
  • It hands off, then gets out of the way.Copy six formats, download, fetch by API, or send to bullshit.doctor.
Every promise, in the privacy policy

For programs

Another program can submit a run with a key and either wait for a signed webhook or poll. The documentation is enough to integrate against without asking anyone a question.

Read the quickstart