The purpose of this repository is to create demo use cases for AI tools that can act as conversation partners. More importantly, these agents open the door for the automation and feedback of common scenarios wherein the dependency of a knowledgeable individual is too taxing or inaccessible for the user. This includes:
- Learning languages
- Learning how to perform conversational tasks (patient histories, interviews, etc.)
- Any problem-solving tasks where question and answer is necessary
You'll need the following to use this repository:
- An OpenAI API key with credits
- Docker
Clone the repository:
git clone https://github.com/Birmingham-AI/conversation-partner.gitAdd a new .env in the core/src/core directory for your API key. You can copy and paste this command with your key's
value to instantly create the file:
echo 'OPENAI_API_KEY=<YOUR_KEY>' > core/.envStart the services by running the following from the root of the project:
docker compose up -dOnce all services are up, you can check the health of your server by running:
curl localhost:3000/healthcheckYou should receive a response like this:
{ "message": "I'm alive you fools!" }You can also run docker ps to see container with the name conversation-bot-server running and its logs visible
using:
docker logs -f conversation-bot-serverYou can use this Postman collection to explore the API; select the Language Bot collection after opening: