๐ค AI-Powered Voice Dictation for Your Desktop
Turn your voice into text anywhere on your computer โ powered by cloud AI transcription
- Global Hotkey - Press a single key (default:
F1) to start/stop dictation from anywhere - Floating Widget - Minimal, always-on-top microphone button for quick access
- Two Activation Modes:
- Tap to Talk - Press to start, press again to stop
- Push to Talk - Hold to record, release to process
- Cloud Speech-to-Text - Supports multiple providers:
- OpenAI Whisper API
- Groq API
- Z.ai (ๆบ่ฐฑ AI)
- Any OpenAI-compatible custom endpoint
- Multi-language Support - Transcribe in any language supported by the provider
- Intelligent Cleanup - Automatically removes filler words, fixes grammar, and formats text
- Agent Mode - Address your AI assistant by name (e.g., "Hey Jarvis, make this more formal")
- Cloud & Local AI - Choose between cloud models (GPT, etc.) or local models via Ollama
- Prompt Studio - Customize how AI processes your speech
- Transcription History - View and manage all past transcriptions
- One-Click Copy - Easily copy or re-use previous transcriptions
- SQLite Database - Fast, local storage of your history
- Bilingual Interface - English and Simplified Chinese (ไธญๆ็ฎไฝ)
- Control Panel - Settings, history, and quick-start guide in one place
- Auto-Paste - Transcribed text automatically pastes at your cursor
- Auto-Start - Launch at system startup (optional)
- Auto-Updates - Built-in update checker and installer
- Debug Logging - Enable detailed logs for troubleshooting
- Custom API Endpoints - Connect to self-hosted models
- Node.js 18 or higher
- Rust toolchain (stable)
- Tauri Prerequisites - See Tauri Setup Guide
# Clone the repository
git clone https://github.com/Charlo-O/typefree1.git
cd typefree1
# Install dependencies
npm install
# Run in development mode
npm run tauri:devnpm run tauri:buildOutput files:
- Windows:
src-tauri/target/release/bundle/nsis/Typefree_x.x.x_x64-setup.exe - MSI:
src-tauri/target/release/bundle/msi/Typefree_x.x.x_x64_en-US.msi
- Click in any text field where you want to type
- Press
F1(or your configured hotkey) to start recording - Speak your text clearly
- Press
F1again to stop and process - Your transcribed text will automatically paste at your cursor!
When AI Text Enhancement is enabled, you can give your agent specific instructions:
"Hey Jarvis, write a formal email about the meeting tomorrow"
"Hey Assistant, convert this to bullet points"
"Hey Jarvis, make this more professional"
Regular dictation (without mentioning the agent name) will simply clean up your speech:
"um so like I think we should uh schedule a meeting for next week"
โ "I think we should schedule a meeting for next week."
- OpenAI - Get API key from platform.openai.com
- Groq - Get API key from console.groq.com
- Z.ai (ๆบ่ฐฑ) - Get API key from open.bigmodel.cn
- Custom - Any OpenAI-compatible
/v1/audio/transcriptionsendpoint
- Cloud AI: OpenAI, Groq, Z.ai, or custom OpenAI-compatible chat endpoints
- Local AI: Ollama or any local model serving OpenAI-compatible API
- Check for conflicts with other applications
- Change the hotkey in Settings โ General โ Dictation Hotkey
- Ensure the app has permission to register global shortcuts
- Check that the correct microphone is selected in Settings
- Grant microphone permission when prompted
- Try enabling "Prefer Built-in Microphone" if using Bluetooth headphones
- Ensure Accessibility permissions are granted (macOS)
- Try manually copying from the history if auto-paste fails
For detailed troubleshooting, enable Debug Logging in Settings โ Developer to capture detailed logs.
- Frontend: React 19 + Vite + Tailwind CSS
- Desktop: Tauri v2
- Database: SQLite (via rusqlite)
- UI Components: Radix UI + shadcn/ui
MIT License. See LICENSE for details.
- Tauri - Desktop application framework
- OpenAI Whisper - Speech-to-text model
- Radix UI - UI primitives