Ever thought of how a Heavy Metal song about the last tweet of Elon Musk would sound? Well, this is your repo then ...

This repository contains a crewAI application for automatically generating songs given a topic and a musc genre. The song generation is done using the "unofficial" Suno AI API.

The crew consists of three agents:

1️⃣ Web Researcher Agent

The Web Researcher Agent searches for relevant information in the web about a provided topic.

2️⃣ Lyrics Creator Agent

The Lyrics Creator Agent takes the previous research and generates high quality lyrics from it. In addition, it adapts the lyrics to the provided music genre.

3️⃣ Song Generator Agent

The Son Generator Agent takes the generated lyrics and uses the Suno custom tool to interact with the Suno API and download two candidate songs.

Thr first thing you'll need to do is to create a .env and provide values for the following variables:

SUNO_COOKIE: You'll need this to use the Suno API, please check the instructions in the unofficial Suno API repo. GROQ_API_KEY: Your Groq API Key. Why Groq? Because it's free! SERPER_API_KEY: Your Serper API KEY.

Check the .env.example if you have doubts about the .env structure

Now that you have all the env variables configured, it is time to let the Docker magic begin! This project is configured using docker-compose, where you'll be creating two containers; one container for the Suno API logic and the other for the streamlit + crewAI logic. You can check the docker-compose specification here.

To start the application, simply run:

docker compose build & docker compose up

A Streamlit application will be launched on your localhost, on port 8501.

If you click the URL, you should see something like this: