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

# Audio & Microphone

> Select and test your microphone.

Scout records audio from your microphone and streams it to the speech engine. You can choose which microphone to use in **Settings → Audio**.

## Microphone selection

By default, Scout uses your system's default input device. You can override this by selecting a specific microphone in Settings.

If you've explicitly selected a microphone, Scout uses that device as long as it's available. If the selected device is disconnected, Scout falls back to the system default.

## Bluetooth auto-switch

If your system's default input device is a Bluetooth microphone (like AirPods) and you haven't explicitly selected a microphone in Scout's settings, Scout tries to switch to your built-in microphone instead.

**Why?** Bluetooth microphones typically use a low-quality audio profile for input, which reduces transcription accuracy. Your built-in microphone usually produces better results.

<Note>
  This only applies when the OS default is Bluetooth, you haven't manually selected a device, **and** Scout can find a built-in microphone. If no built-in mic is available, Scout uses the Bluetooth device. If you explicitly choose a Bluetooth mic in Settings, Scout respects that choice.
</Note>

## Audio cues

Scout plays brief audio cues when dictation starts and stops, so you know when it's listening.

## Mic check

Use the mic check feature in Settings to verify your microphone is working correctly. It shows a real-time audio level indicator so you can confirm Scout is receiving audio from the right device.

## Tips for best audio

* **USB or 3.5mm microphones** give the best transcription accuracy.
* **Built-in laptop microphones** work well in quiet environments.
* **Bluetooth microphones** can work but may reduce accuracy. Consider using your built-in mic instead.
* **Reduce background noise** when possible — the speech engine handles some noise, but a quieter environment gives better results.