Voice and read aloud

Talk instead of type, and listen instead of read. Voice chat is a natural spoken conversation, Dictate turns speech into text you can edit, and Read Out Loud reads any answer or note in one of 13 voices.

Read Out Loud is free. Voice chat and dictation are part of Pro and included in the 3-day trial.

Voice chat

Choose Voice chat Pro in the composer and start talking. The assistant listens, answers out loud, and can still use its tools while you talk.

• On desktop, voice chat opens as a dock above the conversation, so the chat stays visible and every spoken turn appears as a message.
• On mobile, it opens full screen: "Just start talking".

The status always says what is happening: Listening, Hearing you…, Thinking, Speaking, or Using a tool. Speak at any time to interrupt, or use Interrupt, Done speaking and End voice chat. Mobile adds Pause voice chat and Resume voice chat.

For the most accurate recognition, set the language you speak in Settings → General → Voice chat language. Naming it beats Auto detect on short replies.

Dictation

Dictate Pro works anywhere you type a prompt: the chat composer and the image playground. Speak, then choose Stop & Transcribe. Your words appear as text in the composer, ready to edit before you send.

Read Out Loud

Every reply has Read Out Loud, and so does every note (the whole note, or a selection). Open its options to choose:

• Voice: Alloy, Ash, Coral, Echo, Fable, Nova, Onyx, Sage or Shimmer, plus Ballad, Verse, Marin and Cedar with Mini TTS.

• Quality: Standard (fast), HD, or Mini TTS, the most natural and expressive.

• Style Instructions (Mini TTS): presets such as Calm, Energetic, Storytelling, Formal and Friendly, an Intensity, or your own direction.

• Speed: from 0.5x to 2x.

Download audio on desktop (and Share Audio on mobile) saves the reading as a file, ready for a commute or a podcast app.

Free accounts read aloud in Standard quality. Pro unlocks HD and Mini TTS and makes Mini TTS the default. With your own OpenAI key, every quality is available on any plan.

Providers and cost

Settings → Default models sets the Audio provider (read aloud and voice) and the Transcription provider (dictation):

• 123sudo uses your credits and needs no setup.
• OpenAI uses your own API key and is billed by OpenAI.

Audio is processed online by the provider you choose, even when your chat model is local. See Privacy and data flow.

If the microphone does not work

Allow 9xChat to use the microphone in your system settings. On macOS that is System Settings → Privacy & Security → Microphone; on a phone, the app's permissions in system settings.