Command-line Knowledge Base
OpenAI Whisper
Linux Command Line
Install dependencies
sudo apt install python3 python3-pip python3-venv ffmpeg -y
Create and activate a Python virtual environment (recommended to avoid system conflicts)
python3 -m venv whisper_env source whisper_env/bin/activate
Install the openai-whisper package using pip
pip install -U openai-whisper
Specify a model
Whisper offers different model sizes (tiny, base, small, medium, large). Larger models provide better accuracy but are slower.
whisper audio.mp3 --model medium
Specify language and output format
whisper audio.mp3 --language English --output_format txt
Deactivate the environment when finished
deactivate
GUI Applications
Buzz
- Provides an easy-to-use desktop interface for Linux, Windows, and macOS that runs on your local machine using Whisper.
- It supports importing audio and video files and exporting transcripts in TXT, SRT, or VTT formats.
- You can install it via Flathub.
Speech Note
- Another local, private option available on Flathub, Speech Note lets you transcribe audio and supports multiple languages.
- You will need to download the specific language models within the app after installation.