zbhavyai/melody-engine repository overview

⁠Melody Engine

Melody Engine is a thin but practical wrapper on MagentaRT⁠, providing:

1. REST API for programmatic music generation
2. Serialized job queue to avoid GPU thrashing
3. Containerized runtime for reproducible deployment
4. Web interface that hides MagentaRT’s operational complexity

⁠🧰 Requirements

To run Melody Engine, you would need

• a machine with a powerful Nvidia GPU, and
• a linux host with podman or docker installed.

MagentaRT's official repository suggests a GPU with at least 40GB VRAM, but as per my limited experiments, NVIDIA RTX A5000 with 24GB VRAM works just fine.

⁠🚀 How to use this image

1. Pull the latest image.

   podman image pull docker.io/zbhavyai/melody-engine:latest
   

   Copy

2. Create output and cache directories on the host

   mkdir -p "$(pwd)/outputs" && mkdir -p "$HOME/.cache/melody-engine"
   

   Copy

3. Run the container. This will load MagentaRT on your GPU and start a uvicorn server on port 8080.

   podman container run \
      --detach \
      --name melody-engine \
      --restart unless-stopped \
      --publish 8080:8080 \
      --security-opt=label=disable \
      --env TF_GPU_ALLOCATOR=cuda_malloc_async \
      --device=nvidia.com/gpu=all \
      --volume "$(pwd)/outputs:/opt/app/outputs:rw,Z" \
      --volume "$HOME/.cache/melody-engine:/magenta-realtime/cache:rw,Z" \
      docker.io/zbhavyai/melody-engine:latest
   

   Copy

4. Access the web interface at localhost:8080⁠ or explore the API documentation at localhost:8080/docs⁠.

5. Generated audio files are saved locally in the outputs⁠ directory, or can be downloaded from the web interface or REST API.

⁠👨‍💻 Source code

The source code is here: zbhavyai/melody-engine⁠. Please star the repository if you come across it ⭐

You may also choose to sponsor it here⁠ if you would like to show your support 🤜🤛