Speak to Ask Adapt: dictate and Talk

Use the microphone to dictate into the Ask box, or the blue Talk button for a spoken conversation.

How-toChecked 2026-10-10

There are two ways to use your voice. Both need your browser's microphone permission. Listening and speaking use ElevenLabs through Adapt's own server. Recordings are thrown away: only the words are kept.

Dictate

  1. Press the microphone in any Ask box.
  2. Allow the microphone if the browser asks. If it is blocked you see "The microphone is blocked. Allow it in the address bar, then try again."
  3. Talk. A moving line follows your voice, with a timer.
  4. Press Done (or Enter). It says Writing it down, then the words land in the box. Press Cancel (or Escape) to throw it away.

Nothing is sent until you send it. Recording stops by itself after 3 minutes. If you had text in the box, the new words are added to it.

Talk

Talk is on the Ask Adapt page only. It is the blue button with bars beside the microphone.

  1. On the Ask Adapt page press Talk. The microphone opens.
  2. Speak, then pause. After about a second and a half of quiet your turn ends. If your sentence does not end with a full stop, question mark or exclamation mark, it waits for the rest.
  3. Adapt asks Ask Adapt your question, then reads the answer aloud and listens again.
  4. Press Stop to end. It also stops if you leave the page.

The orb shows the stage: Listening, Thinking, Speaking or Writing it down. If it hears nothing twice it stops: "I could not hear anything, so I stopped. Press Talk to carry on."

Limits

  • 200 listens and 200 spoken replies per account per day. Beyond that you see "That is plenty for today. Type your answer instead, or try again tomorrow."
  • A spoken reply reads the first part of the answer only (up to about 380 characters). The full answer is on screen.
  • Your browser must support recording. Otherwise: "This browser cannot record. Type instead."
  • If voice is switched off on the server: "Voice is not switched on yet."

Privacy

The recording is sent to ElevenLabs by our server to be turned into words, and the audio is not kept. A question you ask by voice is handled like a typed one. See Are my chats with Ask Adapt saved?.