Sign in Start free

Live Transcription

Last updated 15 Sep 2026

Overview

Live Transcription converts speech into text in real time. Open the Transcribe page, start a session, and your words appear on screen as live text while you speak.

How to use it

  1. Open Transcribe from the menu.
  2. Choose a transcription model — options from OpenAI and ElevenLabs, each with its per-minute rate — and the spoken language, or leave the language on auto-detect.
  3. Press Start and speak into your microphone — text streams in as you go, with a running timer next to the status badge.
  4. Press Stop when finished; the transcript stays on screen so you can select and copy it.

Cost and metering

Transcription is billed per minute of connected session time at the rate shown next to the model — the estimate is displayed before you start — and every session writes billing records with a full pricing snapshot. Bring-your-own-key (BYOK) routing covers live transcription too: usage runs through your own provider key and ToRun charges only the small per-minute platform fee.

For spoken-word output (text-to-speech) and file-based speech-to-text, see the Text-to-Speech and Speech-to-Text doc. For two-way spoken translation, see Live Translation.

Tips

  • A good microphone and a quiet room dramatically improve accuracy.
  • Set the spoken language explicitly when you know it — auto-detect adds a little latency.
  • You are billed for connected minutes — stop the session as soon as you are done.
Using Live Translation Deep Research (Cited, Multi-Step)