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

# Introduction

> Open source AI-powered offline voice-to-text dictation for macOS and Windows

<img className="block dark:hidden" src="https://mintlify.s3.us-west-1.amazonaws.com/moinulmoin-voicetypr-56/images/hero-light.svg" alt="VoiceTypr Hero Light" />

<img className="hidden dark:block" src="https://mintlify.s3.us-west-1.amazonaws.com/moinulmoin-voicetypr-56/images/hero-dark.svg" alt="VoiceTypr Hero Dark" />

## What is VoiceTypr?

VoiceTypr is an **open source AI-powered voice-to-text dictation tool** that runs completely offline on your device. It's a privacy-first alternative to Wispr Flow and SuperWhisper, designed for developers, writers, and anyone who wants to type faster without sacrificing privacy.

Press a hotkey anywhere on your system, speak your thoughts, and watch them appear instantly at your cursor—no cloud, no tracking, no subscriptions.

<CardGroup cols={2}>
  <Card title="Quick Installation" icon="download" href="/installation">
    Get VoiceTypr running on macOS or Windows in minutes
  </Card>

  <Card title="Quickstart Guide" icon="rocket" href="/quickstart">
    Complete your first transcription in under 5 minutes
  </Card>

  <Card title="Voice Transcription" icon="microphone" href="/features/voice-transcription">
    Learn how system-wide voice dictation works
  </Card>

  <Card title="AI Models" icon="brain" href="/features/ai-models">
    Understand model sizes and choose the right one
  </Card>
</CardGroup>

## Key Features

### Instant Voice-to-Text

Transcribe your voice in real-time with a simple system-wide hotkey. Text appears automatically at your cursor position—works in Cursor, VS Code, Slack, ChatGPT, and every other application.

<CodeGroup>
  ```bash macOS Default Hotkey theme={null}
  Cmd + Shift + Space (hold to record)
  ```

  ```bash Windows Default Hotkey theme={null}
  Ctrl + Shift + Space (hold to record)
  ```
</CodeGroup>

<Info>
  **Pro tip**: Release the hotkey to stop recording and start transcription. Double-tap ESC while recording to cancel.
</Info>

### 100% Offline & Private

All transcription happens locally on your device using OpenAI's Whisper AI models. Your voice never leaves your computer—no cloud processing, no data collection, no tracking (except for license verification).

* **Zero cloud dependency** for transcription
* **Complete data ownership** of your recordings
* **Open source transparency** with AGPL-3.0 license
* **Local model storage** in your home directory

### Powered by Local AI

VoiceTypr uses state-of-the-art Whisper AI models that run entirely on your device:

* **Multiple model sizes**: Tiny (39MB) to Large (1.5GB)
* **99+ languages supported** out of the box
* **Automatic language detection**
* **Hardware acceleration** (Metal on macOS, GPU on Windows)
* **Speed vs accuracy tradeoffs** based on your needs

<Note>
  The **base** model offers the best balance of speed and accuracy for most users. Start there and adjust based on your needs.
</Note>

### AI Enhancement (Optional)

Transform raw transcriptions with AI-powered post-processing:

* **Smart presets**: Convert voice to prompts, emails, commit messages, or formatted notes
* **Cloud AI integration**: Groq (fast & free) or Gemini
* **Secure API storage**: Keys stored in system keychain
* **Internet required**: Enhancement only (transcription stays offline)

<Warning>
  AI Enhancement requires an internet connection and API key (Groq or Gemini). This is optional—transcription works perfectly without it.
</Warning>

### Native Performance

Built with Rust (Tauri v2) backend and React 19 frontend for maximum efficiency:

* **Blazing-fast transcription** with hardware acceleration
* **Minimal resource usage** during idle and recording
* **Small app footprint** (\~50MB installed, excluding models)
* **Platform-native** integration and feel

## Supported Platforms

<Tabs>
  <Tab title="macOS">
    **Requirements**

    * macOS 13.0 (Ventura) or later
    * 3-4 GB free disk space (for AI models)
    * Intel or Apple Silicon processor
    * Microphone access permission
    * Accessibility access permission (for auto text insertion)

    **Hardware Acceleration**

    * Automatic Metal GPU acceleration
    * Apple Neural Engine support (with Parakeet models)
    * Optimized for M1/M2/M3 chips

    **App Security**

    * Fully signed and notarized by Apple
    * No security warnings on installation
    * Distributed as .dmg for easy installation
  </Tab>

  <Tab title="Windows">
    **Requirements**

    * Windows 10/11 (64-bit)
    * 3-4 GB free disk space (for AI models)
    * Modern x64 processor
    * Microphone access permission

    **GPU Acceleration**

    * **5-10x faster** with compatible GPUs
    * NVIDIA, AMD, and Intel GPU support
    * Automatic fallback to CPU if unavailable
    * Update graphics drivers for best performance:
      * [NVIDIA Drivers](https://www.nvidia.com/drivers)
      * [AMD Drivers](https://www.amd.com/support)
      * [Intel Drivers](https://www.intel.com/content/www/us/en/support/products/80939/graphics.html)

    **Installation**

    * NSIS installer for smooth setup
    * Auto-installs WebView2 if needed
  </Tab>
</Tabs>

## Use Cases

VoiceTypr excels in scenarios where typing is slow, inefficient, or impossible:

<CardGroup cols={2}>
  <Card title="Coding" icon="code">
    Dictate comments, documentation, commit messages, and code explanations faster than typing
  </Card>

  <Card title="Writing" icon="pen-fancy">
    Draft articles, blog posts, and documentation by speaking your thoughts naturally
  </Card>

  <Card title="Communication" icon="message">
    Compose Slack messages, emails, and chat responses at the speed of thought
  </Card>

  <Card title="Note-taking" icon="note-sticky">
    Capture meeting notes, ideas, and thoughts without breaking your flow
  </Card>
</CardGroup>

## Why VoiceTypr?

### Privacy First

Unlike cloud-based alternatives, VoiceTypr keeps your voice data on your device. Perfect for sensitive work environments, confidential communications, or anyone who values privacy.

### Open Source

Fully transparent [AGPL-3.0 licensed](https://github.com/moinulmoin/voicetypr) codebase. Audit the code, contribute features, or fork it for your needs.

### Pay Once, Use Forever

No subscriptions, no monthly fees. One-time purchase with lifetime updates. Trial version available to test before buying.

### Offline Operation

Works perfectly without internet (except for optional AI Enhancement). Ideal for flights, remote locations, or offline-first workflows.

## Getting Started

Ready to experience voice dictation? Follow our quickstart guide:

<Steps>
  <Step title="Download VoiceTypr">
    Get the latest release for [macOS or Windows](https://github.com/moinulmoin/voicetypr/releases/latest)
  </Step>

  <Step title="Install and Launch">
    Run the installer and grant required permissions
  </Step>

  <Step title="Download AI Model">
    Choose and download your preferred Whisper model (recommended: base)
  </Step>

  <Step title="Start Transcribing">
    Press your hotkey, speak, and watch the magic happen
  </Step>
</Steps>

<Card title="Continue to Installation" icon="arrow-right" href="/installation" horizontal>
  Follow the complete installation guide for your platform
</Card>

## Community and Support

<CardGroup cols={2}>
  <Card title="GitHub Repository" icon="github" href="https://github.com/moinulmoin/voicetypr">
    View source code, report issues, and contribute
  </Card>

  <Card title="Troubleshooting" icon="life-ring" href="/troubleshooting">
    Common issues and solutions
  </Card>

  <Card title="FAQ" icon="question" href="/faq">
    Frequently asked questions
  </Card>

  <Card title="Development Docs" icon="code" href="/development/setup">
    Build VoiceTypr from source
  </Card>
</CardGroup>