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

# Sotto overview

> Fully local voice dictation for Linux. Press a shortcut, speak, and the formatted text lands in whatever app has focus.

Sotto is voice dictation that runs entirely on your own machine. Press a global shortcut, speak,
and the text appears in whatever application has focus.

**Every sample is processed locally.** No account, no server, no telemetry. The only network
request Sotto ever makes is downloading a speech model, once, when you ask for it.

## How it works

<Steps>
  <Step title="Press the shortcut">
    **Meta+Alt+D** by default. A small dark pill appears at the bottom of your **active** monitor.
  </Step>

  <Step title="Speak">
    The pill shows a live level meter and a live transcript, so you can see it is hearing you.
    Text appears as you talk, not only at the end.
  </Step>

  <Step title="Press it again">
    Sotto finishes the last utterance, formats everything, and puts the result into the focused
    application.
  </Step>
</Steps>

## What makes it different

* **Local.** [whisper.cpp](https://github.com/ggml-org/whisper.cpp) runs on your GPU or CPU. Audio
  never leaves the machine, and there is nothing to sign in to.
* **Proper writing.** Punctuation and capitalisation come from the model; Sotto adds paragraph
  breaks when you pause, understands spoken "new line" and "new paragraph", and strips the noise
  markers a transcriber emits.
* **It goes where you are typing.** Text is pasted into the focused app, typed key by key, or left
  on the clipboard — your choice.
* **It never steals focus.** The dictation pill is a layer-shell surface: it cannot be focused, and
  tiling compositors will not tile it.

## Requirements

|                      |                                                                                 |
| -------------------- | ------------------------------------------------------------------------------- |
| **Operating system** | Linux                                                                           |
| **Display server**   | Wayland (X11 works with a degraded overlay)                                     |
| **Desktop**          | KDE Plasma / KWin is the primary target                                         |
| **GPU**              | Optional but strongly recommended — AMD via ROCm, or Vulkan, or NVIDIA via CUDA |
| **Disk**             | 78 MB to 3.1 GB for a speech model                                              |

Sotto is early v1, developed and tested on Arch Linux with Plasma 6 on Wayland and an AMD GPU with
ROCm. Expect rough edges elsewhere.

<Note>
  There is no Windows or macOS build. The audio capture, the global shortcut, the text injection and
  the overlay are all Linux-specific, so a port is real work rather than a rebuild.
</Note>

## Where to go next

<CardGroup cols={2}>
  <Card title="Installing" icon="download" href="/sotto/installing">
    AppImage, or build with the GPU backend that suits your hardware.
  </Card>

  <Card title="Dictating" icon="microphone" href="/sotto/dictation">
    The shortcut, the overlay, and getting text into other applications.
  </Card>

  <Card title="Speech models" icon="brain" href="/sotto/models">
    Which model to download, and what it costs you in speed and disk.
  </Card>

  <Card title="Formatting" icon="paragraph" href="/sotto/formatting">
    Paragraph breaks, voice commands, and what gets cleaned up.
  </Card>

  <Card title="Notepad and CLI" icon="terminal" href="/sotto/notepad-and-cli">
    Dictate without inserting, and script Sotto from anywhere.
  </Card>

  <Card title="Privacy and licence" icon="lock" href="/sotto/privacy">
    What the local-only guarantee covers, and the licence.
  </Card>
</CardGroup>

## Compared to cloud dictation

|                          | Sotto        | Cloud dictation services           |
| ------------------------ | ------------ | ---------------------------------- |
| Where audio is processed | Your machine | Someone else's                     |
| Works offline            | ✅            | ❌                                  |
| Account required         | ❌            | Usually                            |
| Speed depends on         | Your GPU     | Your connection                    |
| Cost                     | Free         | Usually per-minute or subscription |
| Platforms                | Linux        | Usually everything                 |