Whisper Transcribe
Welcome to the Whisper Transcribe repository. This is a simple Python
command line application that allows you to record audio and transcribe it using one of the
Whisper ASR models by OpenAI. It uses a terminal user interface, making it easy to use.
Prerequisites
- Python 3.10 or higher (probably works with lower versions, but not tested)
- ffmpeg
Installation from PyPI
pip install whisper-clipboard
And then run transcribe
and start transcribing.
Please make sure you have ffmpeg
installed properly on your system (this may vary between different operating systems).
Installation using the repo
For NixOS Users:
git clone https://github.com/data-stepper/whisper-clipboard
cd whisper-clipboard
nix-shell
python transcribe.py
For Non-NixOS Users:
git clone https://github.com/data-stepper/whisper-clipboard
cd whisper-clipboard
python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
python transcribe.py
Usage
Start the program using
python transcribe.py
After starting the program, you will see a prompt ('>').
- Press 'Space' to start recording. While recording, the terminal will display "Recording..."
- Press 'Space' again to stop recording. The audio will be saved, and the transcription will be displayed in the terminal and copied to the clipboard.
- Press 'Q' to quit the application.
Troubleshooting
Ensure that you have the required version of Python and FFMPEG installed. If you encounter any issues, feel free to create an issue on this repository.
Happy transcribing!