Skip to content

ashish8485/talkink

Repository files navigation

🎙️ talkink - Private desktop speech to text

talkink provides private, secure dictation directly on your computer. It converts your speech to text without sending audio data to the cloud. The application runs locally using your computer hardware to ensure your conversations remain private.

🛠️ System Requirements

To run talkink, verify these items:

  • A modern Apple Mac computer using Apple Silicon (M1, M2, M3, or newer).
  • macOS 14.0 or later.
  • At least 8GB of memory.
  • A stable internet connection for the first installation.

📥 How to Install

Follow these steps to set up talkink on your machine:

  1. Visit the official releases page to download the software.
  2. Look for the file ending in .dmg.
  3. Click the download link to save the file to your computer.
  4. Once the file finishes downloading, open it by double-clicking the icon.
  5. Drag the talkink icon into your Applications folder.
  6. Open your Applications folder and double-click the talkink icon to start the program.

Note: Since this is a new application, your Mac might ask for permission to open it. If this happens, right-click the icon, select Open, and confirm the action in the box that appears.

⚙️ Choosing a Model

talkink uses advanced speech recognition models to transcribe your voice. You can choose different models based on your needs:

  • Qwen3-ASR: Best for daily dictation tasks.
  • NVIDIA Nemotron: Provides high accuracy for complex technical terms.
  • Voxtral: Optimized for speed and low memory usage.

To change your model, click the settings icon in the top right corner of the application window. Select the Model menu and choose the version that fits your current task. The application will download the necessary files the first time you select a new model.

🎙️ How to Use

The interface uses a simple design. Follow these steps to start dictating:

  1. Click the talkink icon in the menu bar at the top of your screen.
  2. Select your microphone from the list if the app does not detect your default headset.
  3. Press and hold your selected hotkey to start recording. By default, this is the Command key.
  4. Speak clearly into your microphone. As you talk, the text will appear in the active window.
  5. Release the hotkey to stop recording.

🔒 Privacy

talkink operates 100% on your device. No audio or text data leaves your computer. The software uses Apple MLX to process speech locally. This design prevents third-party companies from accessing

About

Transcribe your voice directly to text using local MLX models on macOS. Press a key to dictate and paste text instantly while keeping your data private.

Topics

Resources

License

Contributing

Security policy

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors