> ## Documentation Index
> Fetch the complete documentation index at: https://docs.agenium.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Talk to an agent by voice

> Dictate into a session's composer with Deepgram, or hold a voice call with a session through Gemini Live or GPT Live.

> [!NOTE]
>
> Draft: these steps await verification against the current release of Agenium.

**Outcome:** you speak to an agent instead of typing: either your words appear in the composer as text (dictation), or you hold a call where a voice model talks with you and hands your words to the session (voice mode).

| | Dictation | Voice mode |
| - | - | - |
| What it does | Turns speech into text in the composer. | A spoken conversation; the voice tells you the agent's replies. |
| Provider | Deepgram | Gemini Live or GPT Live |
| Where to set it up | Settings > Transcription | Settings > Voice |
| Works in | Agenium's composers | Sessions Agenium runs |

```mermaid theme={null}
flowchart LR
  you(("You")) -- "speak" --> voice["Voice model<br/>(Gemini Live / GPT Live)"]
  voice -- "your words, as your message" --> session["Session"]
  session -- "agent's reply" --> voice
  voice -- "speaks it in its own words" --> you
```

## What you need

* A session Agenium runs, such as one started with **New** on an agent's page.
* For dictation: a Deepgram API key.
* For voice mode: a Gemini Live or GPT Live key.
* Microphone permission for Agenium; macOS asks on first use.

## Steps in the app

> [tab] Dictation
>
> **1. Add your Deepgram key**
>
> Open **Settings > Transcription** and paste your Deepgram API key. Agenium validates it. Optionally choose the model, the language (**Automatic** by default), the dictation key (right Option by default), key terms and sounds.
>
> **2. Start listening**
>
> In a session's composer, press the microphone beside Send, or tap the dictation key once. The start sound plays once the microphone is listening, and the button shows the level of your voice.
>
> **3. Speak**
>
> Text appears at the cursor as you speak, dimmed while it is still being recognised.
>
> **4. Stop or send**
>
> Press the microphone or the key again to stop, Escape to stop and keep the text, or Send to wait for your last words and send. Listening stops by itself after 30 seconds of silence.

> [tab] Voice mode
>
> **1. Turn on voice mode**
>
> Open **Settings > Voice**, turn voice mode on, choose **Gemini Live** or **GPT Live**, and enter its key. Each engine has a voice prompt you can edit.
>
> **2. Start a call**
>
> Open a session Agenium runs. A microphone stands left of the composer; press it to start the call.
>
> **3. Talk**
>
> The voice starts from its prompt, the agent's name and description, and the session's last few turns. When you ask for something, it hands what you said to the session as your message, headed *Voice mode messages*, and tells you the agent's reply in its own words. You can keep typing throughout.
>
> **4. Mute or end**
>
> While live, the button widens with **Mute** and **End call**.

## Check it worked

1. **Dictation**: your words appear in the composer and settle as ordinary text.
2. **Voice mode**: the button runs a light round its outline and a glow follows the voice; after a handover, the session's feed shows your message headed *Voice mode messages* with a line per utterance, and the agent answers it.

## Common failures

| What you see | Why, and what to do |
| - | - |
| The microphone opens Settings > Transcription | No Deepgram key is set. Add it. |
| Dictation stops with a short note | The network dropped or Deepgram rejected the key. The text already dictated stays; check the key. |
| Nothing is heard | macOS has not allowed Agenium the microphone. Allow it in System Settings > Privacy & Security > Microphone. |
| No voice-mode button beside the composer | Voice mode is off in Settings > Voice, or the session is not one Agenium runs. |
| The call shows live but there is no sound | Update to the latest release, which restarts the sound when macOS reconfigures the audio device during a call. |
| The dictation key does nothing | It works only while Agenium is in front and a composer is on screen. |

## Related

* [Sessions: dictation and voice mode](https://docs.agenium.io/app/sessions.md#dictation)
* [Settings > Transcription](https://docs.agenium.io/app/settings.md#transcription) and [Settings > Voice](https://docs.agenium.io/app/settings.md#voice)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.