Voice
You can talk to your agents in the dashboard — speak instead of type, and hear replies read aloud.
Two ways to use your voice
When voice is available, a microphone appears next to the chat input, and a headphones toggle slides out beside it once you start speaking:
- Dictation — tap the mic and speak. Your words transcribe live into the input as you go, building on anything you've already typed. Tap again to stop, then send when ready.
- Hands-free voice mode — toggle the headphones and just talk. When you pause, your message sends automatically; the agent's reply plays aloud as it streams back, sentence by sentence. When the reply finishes, it listens again for your next message — or tap the microphone to stop playback and talk right away. The loop continues until you switch it off.
Setup
Voice is configured by an admin on the Setup → Audio page (admin only). An agent can hear and speak in two ways:
- The device's built-in voice — the browser or phone's own speech engine. Nothing to connect; it works out of the box.
- A provider — a pluggable speech-to-text provider (to hear you) and text-to-speech provider (to speak), each with an account you bring. OtoDock currently supports Deepgram for speech-to-text and Cartesia for text-to-speech — create an account with each and paste its API key into the provider card. Admins can pick a voice per language, so each person hears a natural voice in their own language. Supported languages include English, Greek, German, Spanish, French, and Italian.
Admins choose a policy for which is used:
- Native only — use only the device's built-in voice. Works even with no provider connected.
- Native preferred — use the device voice when available, otherwise the configured provider.
- User choice — each person picks device vs. provider in their audio preferences, and the choice follows them across devices.
The voice controls appear on their own wherever a usable voice engine exists under the chosen policy — there's nothing to switch on per user. An admin can also turn chat audio off entirely from the same page.
note
Voice usage (transcription time and characters spoken) is tracked alongside your other usage — see Usage & Limits.
Next steps
- Chat → — the conversation voice plugs into.
- Usage & Limits → — where voice usage is tracked.