# ElevenLabs Configuration

## Overview

ElevenLabs is a cloud-based text-to-speech service that produces significantly more natural-sounding speech than Windows built-in voices. iReporter integrates with ElevenLabs as an optional voice engine — when configured, all button voice output uses ElevenLabs instead of Windows TTS.

ElevenLabs requires an internet connection during races. A free-tier account provides enough monthly character credits for most iReporter use cases.

## Getting an ElevenLabs Account and API Key

1. Go to [https://elevenlabs.io](https://elevenlabs.io) and click **Sign Up**.
2. Create a free account using your email address or a Google/Apple login.
3. Once logged in, click your profile avatar (top-right) and select **Profile + API Key**.
4. Under the **API Key** section, click **Copy** to copy your key.

Keep this key private — it controls access to your account credits.

## Configuring iReporter

1. In the iReporter settings panel, tick **Enable Voice** in the top bar.
2. In the **Voice Options** panel, set the **Voice Engine** dropdown to **Eleven Labs**.
3. Paste your API key into the **API Key** field.
4. Click **Load Voices** to fetch the list of voices available on your account.
5. Select a voice from the **Voice** dropdown.
6. Click **Test Voice** to hear a sample phrase spoken with the selected voice.

## Voice Output Device

By default, voice output goes to your system's default audio device. You can route it to a specific device using the **Voice Output Device** dropdown:

- **Default (system audio)** — plays through your normal speakers or headset
- **VB-Audio CABLE Input** — routes audio into a virtual microphone, which can then be used as your iRacing in-car microphone so race control hears the spoken message

VB-Audio Virtual Cable is free software available at [https://vb-audio.com/Cable/](https://vb-audio.com/Cable/).

## Choosing a Voice

ElevenLabs offers dozens of pre-made voices as well as the ability to clone custom voices. For race reporting, clear and authoritative voices work best. After clicking **Load Voices**, try a few with **Test Voice** until you find one that reads race messages clearly at speed.

You can browse voices on the [ElevenLabs Voice Library](https://elevenlabs.io/voice-library) to preview them before loading in iReporter.

## Troubleshooting

<table id="bkmrk-issuefixload-voices-"><thead><tr><th>Issue</th><th>Fix</th></tr></thead><tbody><tr><td>Load Voices returns empty list</td><td>Check your API key is correct and your account is active. Free accounts may have usage limits.</td></tr><tr><td>Test Voice fails with error</td><td>Ensure your API key is saved (click outside the field first). Check your internet connection.</td></tr><tr><td>Voice plays through wrong device</td><td>Use the Voice Output Device dropdown to select the correct output.</td></tr><tr><td>No voice heard during race</td><td>Confirm Enable Voice is ticked globally. Check the per-button Speak Text is not empty and Disable Voice for this button is not ticked.</td></tr></tbody></table>

## Switching Back to Microsoft Voice

To revert to Windows TTS, use the **Voice Engine** dropdown and select **Microsoft Voice**, or click the **Switch to Microsoft Voice** button inside the ElevenLabs panel. This clears the stored API key and voice selection.