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

# Voice-to-Text

> Powerful on-device speech recognition

## Local Speech Recognition

Pipit uses advanced on-device speech recognition to transcribe your voice. Your audio is processed entirely on your Mac — nothing is sent to the cloud.

## How It Works

<Steps>
  <Step title="Activate">
    Press and hold your hotkey (default: Right Option) to start recording.
  </Step>

  <Step title="Speak">
    A subtle overlay appears showing the recording status. Speak naturally — Pipit handles punctuation and formatting.
  </Step>

  <Step title="Release">
    Let go of the hotkey to stop recording. Pipit processes your audio instantly.
  </Step>

  <Step title="Insert">
    The transcribed text is automatically inserted at your cursor position in any application.
  </Step>
</Steps>

## Recording Modes

<CardGroup cols={2}>
  <Card title="Push-to-Talk">
    **Default mode.** Hold the hotkey while speaking, release when done. Best for quick dictations.
  </Card>

  <Card title="Toggle Mode">
    Press once to start, press again to stop. Better for longer dictations where holding a key is tiring.
  </Card>
</CardGroup>

To change recording mode, go to **Settings** → **Recording** → **Recording Mode**.

## Tips for Better Accuracy

* **Use a quality microphone** — While built-in mics work, a USB microphone or headset improves results significantly.
* **Minimize background noise** — Find a quiet space when possible for the best transcription quality.
* **Speak clearly and consistently** — Natural speech works best, but enunciate at a steady volume.

## Supported Languages

Pipit supports multiple European languages with automatic detection. Enable multilingual mode in **Settings** → **Languages** to have Pipit automatically detect and transcribe the language you're speaking. Switch to English-only mode for optimized English transcription.