Skip to Content
ChatVoice input

Voice input

Speak a prompt and Anuma transcribes it before sending. Nothing you say is uploaded as audio.

Available on: web, iOS, Android.

Turn it on

Voice input is opt-in.

  • Web: Settings → Accessibility → Enable voice input. The first use downloads a speech model (about 250 MB) that then runs in your browser. On iOS Safari, Anuma uses the device’s own speech recognition instead. Once on, a Transcription language option lets you pin a language if short clips get mistranscribed.
  • App: the app uses your phone’s built-in speech recognition.

The Accessibility section of Settings on the web with the Enable voice input switch, Language, Theme, Reduce effects and Keyboard shortcuts

Use it

  1. Tap the microphone in the composer.
  2. Speak. Words appear as you talk.
  3. Tap stop, fix anything, and send.

Good to know

  • Available on every plan.
  • Spoken answers (text-to-speech) are not part of Anuma today.

Related articles

Last updated on
Personality customizationLive thinking