Although I always keep your and my own privacy in mind while developing software, I offer this experimental service for free to anybody who would like to use it, but I do not guarantee anything, including your privacy. Use this bot at your own risk!
This bot uses OpenAI's whisper and Python Telegram Bot to transcribe your voice messages, video messages and even raw audio and video files from Telegram and other messengers like WhatsApp or Signal. You can just send the message to your Telegram-Bot and it responds with the containing text.
I recommend using Docker on a server to build and run this repository. To do this, run these commands as root:
- Clone the repository & cd into it
git clone https://github.com/silasmartin/telegram-transcription-bot.git && cd telegram-transcription-bot
- Edit the .env file and insert your token. You can get it via BotFather.
nano .env
- Build the image
docker build . --tag ttbot
- Run the image
docker run -d --restart unless-stopped ttbot
It takes a few moments to download the whisper model, depending on your internet speed. You could always check the status with docker logs.
You can run this on a local machine as well. Just clone the repository, edit .env, install requirements.txt with pip and run main.py - that's it!
I'm happy to take a look. Reach out!