This project is a Twitch chatbot that uses OpenAI's GPT-3.5 model to generate responses to user queries. The bot can respond to various types of input, from simple questions to more complex requests.
To run this bot, you will need to install the following libraries:
pathlibdotenvtwitchioplaysounddatetimeopenai
You will also need to create a .env file with the following content:
TMI_TOKEN=<your Twitch bot's token>
CLIENT_ID=<your Twitch bot's client ID>
log_dir=<folder path for log files>Additionally, you will need to provide your OpenAI API key and the name of the GPT-3.5 model you want to use.
To start the bot, run the twitch_chatbot.py file. The bot will then be ready to respond to user messages in the Twitch chat.
The bot will respond to messages that begin with "@wuyodo". If a user sends a message without this prefix, the bot will not respond.
The bot will greet users once per day. If a user sends a message multiple times in a day, the bot will not greet them again.
The bot can also respond to various types of input, from simple questions to more complex requests. To make a request to the bot, type "@wuyodo" followed by your query.
The bot can also respond with a variety of emotes. Here are some examples of the emotes the bot can use:
- Hand gestures: ✋, ✌️, 👐, 👋, 🤚, 🤙
- Nose: 👃, 🐽, 👃🏻, 👃🏿, 👃🏽, 👃🏼, 👃🏾
- Tongue: 😛, 😜, 😝, 😋
- Shy: 🤗, 👐, 🤭, 😄, 🥰, 😼, 😙, 😍, 😻, 😅
- Laugh: 🍑, 🤣, 😂, 😹, 😆, 🙈
- Kiss: 👄, 💋, 😘, 😚, 😙, 😽
If you would like to contribute to this project, please open a pull request.