WhisperWrap is a powerful macOS application that wraps the faster-whisper library to provide local, privacy-focused speech-to-text capabilities. It offers system-wide dictation, file transcription, and text-to-speech features.
- System-Wide Dictation: Press a global hotkey (default:
Option+Space) to start dictating anywhere. The transcribed text is automatically copied to your clipboard and can be auto-pasted. - File Transcription: Drag and drop audio or video files to transcribe them using the Whisper model.
- Text-to-Speech (TTS): Convert text to speech using system voices.
- Local Processing: All speech processing is done locally on your device using
faster-whisper, ensuring data privacy. - Private Python Environment: Automatically manages a dedicated Python environment for dependencies, keeping your system clean.
- Menu Bar Access: Quick access to dictation and settings from the menu bar.
- Download the latest release from the Releases Page.
- Unzip
WhisperWrap.zipif needed. - Move
WhisperWrap.appto your Applications folder. - Open the app. On first launch, click "Set Up Speech Engine" to install the necessary local AI models and dependencies.
- Ensure the app is running.
- Place your cursor where you want to type.
- Press
Option + Spaceto start recording. - Speak your text.
- Press
Option + Spaceagain to stop. - The text will be transcribed and pasted automatically (if "Auto Paste" is enabled in settings).
- Model Selection: Choose between different Whisper model sizes (Tiny, Base, Small, Medium, Large) to balance speed vs. accuracy.
- Auto Copy/Paste: Configure clipboard behavior.
- Hotkeys: Customize the global dictation hotkey.
Requirements:
- macOS 13.0+
- Swift 5.9+
- Python 3.10+ (for runtime environment creation)
# Clone the repository
git clone https://github.com/akeslo/WhisperWrap.git
cd WhisperWrap
# Build the app bundle
./generate_app.shThe compiled WhisperWrap.app will be in the project root.