Skip to content

Latest commit

 

History

History
85 lines (58 loc) · 2.6 KB

README.md

File metadata and controls

85 lines (58 loc) · 2.6 KB

Whisper Voice Transcription App

A real-time voice transcription desktop application powered by faster-whisper (large-v3 model). Record audio with a hotkey and have it automatically transcribed and pasted at your cursor position.

Features

  • Real-time voice recording — toggle recording with a keyboard shortcut
  • Automatic transcription — uses the Whisper large-v3 model via faster-whisper
  • Auto-paste — transcribed text is automatically inserted at your cursor position
  • GPU acceleration — automatically detects and uses CUDA when available
  • Visual indicators — on-screen overlays show recording and transcription status
  • Accent support — full support for accented characters (French, etc.)
  • Re-inject — re-paste the last transcription with a shortcut

Requirements

  • Python 3.9+
  • Windows (uses Windows-specific GUI overlays and hotkeys)
  • A microphone
  • (Optional) NVIDIA GPU with CUDA for faster transcription

Installation

  1. Clone the repository:

    git clone https://github.com/BerthalonLucas/whisper_app.git
cd whisper_app

  2. Create and activate a virtual environment:

    python -m venv venv
venv\Scripts\activate

  3. Install dependencies:

    pip install -r requirements.txt

  4. (Optional) Install CUDA support for GPU acceleration — install PyTorch with CUDA or ensure ctranslate2 can detect your GPU.

Usage

Run the application

python app.py

Or use the provided batch file on Windows:

launch_whisper.bat

Keyboard shortcuts

Shortcut Action
Ctrl + F9 Start / Stop recording
Ctrl + Shift + F9 Re-paste the last transcription
Ctrl + F10 Quit the application

Transcribe an audio file

You can also transcribe an existing audio file from the command line:

python transcribe.py path/to/audio.wav

How it works

  1. Press Ctrl + F9 to start recording — a red ● REC indicator appears on screen.
  2. Press Ctrl + F9 again to stop — the audio is sent to the Whisper model for transcription.
  3. A blue progress overlay shows transcription status.
  4. Once complete, the transcribed text is automatically pasted at your current cursor position.

License

This project is licensed under the MIT License. See the LICENSE file for details.