A open source python bot that transcribe voice notes to text with OpenAI Whisper.
Before using the bot, make sure ffmpeg is installed in PATH
You can install most required modules/library by doing pip install -r requirements.txt
You might need other requirements for OpenAI Whisper if you want to run it with GPU.
You also need to create a file .env under the same directory with your token in it. (token=INSERT_YOUR_TOKEN, check .env.example)
This bot has a queue system where it will transcribe the voice notes one by one.
It also cache transcribed notes for 7 days so it doesn't have to transcribe the same voice note again.
There is no slash commands in this bot.
The only interactions is a message command where you right click on a message and click "Transcribe Voice Note" in the apps menu.
This project is under MIT License. You can check the details in LICENSE.