🚀 FastAPI Project

This is a FastAPI project setup using uv, a blazing fast Python package manager.

Follow the steps below to get started.

---

📦 Requirements

• Python 3.12 (uv handles virtual environments automatically)
• uv installed

curl -LsSf https://astral.sh/uv/install.sh | sh

• (Optional) Docker & Docker Compose

---

⚙️ Environment Variables

Create a .env file in the project root with the following values:

# Spotify API Keys
SPOTIFY_CLIENT_ID=
SPOTIFY_CLIENT_SECRET=

# Google YT API key
YOUTUBE_API_KEY=

# Supabase API keys
SUPABASE_API_KEY=
SUPABASE_URL=
BUCKET_NAME=

# Atlas MONGODB Connection String
MONGODB_URL=

# Replicate API Key
REPLICATE_API_TOKEN=

# API Key
GROQ_API_KEY=

# Logging
LOG_LEVEL=INFO   # Options: NOTSET, DEBUG, INFO, WARNING, ERROR, CRITICAL

⚠️ Make sure not to commit .env to version control.

---

How to get your MongoDB connection string

1. Create an Account on MongoDB Atlas

• Visit MongoDB Atlas Login.

2. Create a Cluster

• Once logged in, you'll be redirected to the MongoDB Atlas Dashboard.
• Click on "Create Cluster".
• Select the Free Plan (M0) as your cluster type.
• Follow the steps for creating a cluster (choose cloud provider and region).
• Click on "Create Cluster".

3. Create Database Users

• After the cluster is created, you will be taken to a "Connect to _cluster_name" popup.

• Under "Create Database User", you need to create a user with a username and password:

  • Choose a Username (e.g., myUser).
  • Choose a Password (e.g., myPassword123).
  • Click on the "Create Database User" button.

4. Connect to Your Cluster

• After creating your database user, go to the next step in the popup.

• Choose "Drivers" as the connection method.

• Select Python as your driver.

• In the next step, select the version of Python you’re using.

• MongoDB will now show you the connection string. It will look something like this:

  mongodb+srv://<username>:<password>@cluster0.mongodb.net/test?retryWrites=true&w=majority
  

---

🎵 Audio Processing Setup

Saves processed stems file to Supabase Storage.

Before running:

1. Go to your Supabase project dashboard

2. Navigate to Storage → Buckets

3. Create a new bucket with the name:

   Media-Files
   

4. Enable ✅ “Public bucket”

5. Inside this bucket, create a directory:

   audio_stems
   

Now, Splitter AI will save all audio files under:

Media-Files/audio_stems/

---

🏷️ API Versioning & Pricing

The API supports two versions for stem extraction, each with different processing methods and costs:

API Version	Processing Method	Cost	Notes
v1	Local Processing	Free	Uses the local logic with HTDemucs model. Requires sufficient local resources.
v2	Replicate	~$0.039 / run	Uses cloud-based processing via Replicate. Ideal for scalable deployments without local GPU.

💡 Tip: Use v2 if you want to offload processing power, or stick to v1 for free local testing.

---

⚙️ Local Setup Instructions

1. Clone the repository

   git clone https://github.com/NafeesMadni/Stratum-Backend.git
   cd Stratum-Backend

2. Install dependencies

   uv sync

3. Activate the virtual environment (optional if you run via uv run)

   source .venv/bin/activate   # Linux / MacOS
   .venv\Scripts\activate      # Windows

---

▶️ Running the App (Local)

uv run main.py

• App: http://127.0.0.1:8000
• Docs: http://127.0.0.1:8000/docs
• ReDoc: http://127.0.0.1:8000/redoc

---

⚡ Running Celery Worker (Local)

celery -A app.core.celery_app worker --loglevel=info

---

🧪 Running Tests

Run all tests:

pytest

Run Audio Processing API tests:

pytest tests/test_audio_api_workflow.py

---

🐳 Running with Docker & Docker Compose

1. Build & Start Services

docker-compose up --build

Services started:

• Redis (broker & backend for Celery)
• Web (FastAPI app) → http://localhost:8000
• Celery Worker (background task processor)

2. Stop Services

docker-compose down

Remove volumes (e.g., Redis data):

docker-compose down -v

3. Logs

docker-compose logs -f        # all services
docker-compose logs -f web    # only FastAPI
docker-compose logs -f celery-worker