Skip to content

Text-to-Speech

Text-to-speech (TTS) lets Atlas speak its responses out loud. Instead of reading each reply, you can listen — useful when you’re away from the screen, doing something else with your hands, or just prefer an audio experience.

Go to Settings > Voice and toggle Text-to-Speech on. Once enabled, Atlas will automatically read each response aloud after it arrives.

You can turn TTS off at any time from the same settings panel, or quickly mute it from the chat panel without going into settings.

You can control how fast Atlas speaks using the Speed slider in Settings > Voice:

SpeedEffect
0.5xSlow — good for careful listening or working in a non-native language
1.0xDefault — natural conversational pace
1.5xFaster — comfortable for most people after a short adjustment period
2.0xFast — great for power users who want information quickly

Speed changes apply to the next response. You can adjust it at any time and it takes effect immediately.

TTS uses streaming playback — meaning the audio starts playing before the full response has been generated. You’ll hear the beginning of Atlas’s reply within a second or two, and the rest follows as it’s synthesized. This makes the experience feel much more natural and responsive compared to waiting for the entire response before playback begins.

If you want to stop Atlas mid-sentence, click the stop button in the chat panel or press the push-to-talk button. Atlas will stop speaking and you can respond immediately.

TTS works well alongside push-to-talk and wake word for a fully hands-free experience:

  1. Say “Hey Jarvis” (or click the mic button)
  2. Speak your message
  3. Atlas responds in text and speaks the reply aloud
  4. Use the wake word again for your next message

This creates a back-and-forth voice conversation without touching the keyboard at all.