Dictation turns your speech into text in the message box, so you can talk instead of type. These answers cover how it works, where your audio goes and what stays on your device. They describe the web and desktop apps.
How do I dictate a message?
- Select the microphone button in the composer, labeled "Start voice dictation". Allow microphone access if your browser or computer asks.
- Speak. ZeroTwo transcribes your words in real time.
- When you are done, choose one of the buttons that replace the microphone: Stop dictation inserts the text into the message box, Send dictated message inserts it and sends it, and Cancel dictation discards it.
Dictated text is inserted like typed text, so you can edit it before you send. If your browser cannot record audio, the button is disabled and its tooltip reads "Voice dictation not supported".
Where does my audio go?
When you start dictating, the app asks ZeroTwo's server for a short-lived, single-use token. It then streams your microphone audio to ElevenLabs' real-time Scribe speech-to-text service, which returns the transcript. ZeroTwo's own ElevenLabs key stays on ZeroTwo's server; your device receives only the single-use token.
The text you dictate becomes part of your message. ZeroTwo's Privacy Policy lists audio among the content you create or upload; read it for how ZeroTwo handles that data.
Does ZeroTwo keep my recordings?
When you finish a dictation with text, a copy of the recording is saved on the device you used. Settings > Voice shows them under Recent recordings, with the note "Your last 20 recordings are saved on this device." Each row shows the transcript and the time, with buttons to play or delete that recording. ZeroTwo keeps your latest 20, and a dictation you cancel is not saved.
If a recording cannot be saved, ZeroTwo keeps your text and tells you: "Your text is kept, but this recording could not be saved."
Can I help dictation recognize names and terms?
Yes. In Settings > Voice, find the Dictation section and choose + Add entry under "Dictation dictionary" ("Words or phrases dictation should recognize"). You can add up to 50 entries of up to 20 characters each. ZeroTwo passes the entries to the transcription service as key terms.
Which microphone does dictation use?
The one you pick in Settings > Voice > Microphone, which is used for voice chat and dictation. It starts on "System default". If the microphone you chose is disconnected mid-dictation, dictation stops with the message "Microphone disconnected." and the next dictation uses your system default.
Can I dictate into other apps?
In the desktop app, the Dictation section of Settings > Voice has two hotkeys:
- Hold-to-dictate hotkey: hold it anywhere on your desktop to dictate where your cursor is.
- Toggle dictation hotkey: press it once to start dictating anywhere, then press again to stop.
Both read "Off" until you set them. Select a hotkey field, then press the combination you want; it must include Command, Control or Option.
What if dictation stops or does not start?
- "Allow microphone access to dictate messages." means microphone permission was declined. Allow it in your browser or system settings and try again.
- "Voice dictation is temporarily unavailable", "could not connect" and "could not start" messages mean dictation could not start or lost its connection, for example because the transcription service could not be reached. Try again in a moment.
- If dictation stops after you have already spoken, ZeroTwo keeps the text it captured in the error message with a Copy text button, so you can paste it into the message box.
How is dictation different from voice chat?
Dictation turns your speech into text for the message box. Voice chat is a spoken conversation: when the message box is empty, the send button becomes "Start voice conversation". Pick the voice for voice chats under Settings > Voice > Voice chat.