heartmula

Set up HeartMuLa and generate songs locally from lyrics and tags.

Updated Apr 30, 2026
One-click install
npx skills add https://github.com/photonics-dhl/Hermes --skill heartmula-photonics-dhl
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: heartmula
Source: https://github.com/photonics-dhl/Hermes/tree/main/hermes-home/skills/media/heartmula
Command: npx skills add https://github.com/photonics-dhl/Hermes --skill heartmula-photonics-dhl

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

HeartMuLa provides an open-source, local music generation solution that enables offline creation of songs from lyrics and tags, avoiding reliance on proprietary cloud services.

Core Features & Use Cases

  • HeartMuLaGen model family for generation from lyrics and tags
  • HeartCodec for high-fidelity audio reconstruction
  • HeartTranscriptor and HeartCLAP for transcription and alignment Use cases include offline music generation, local experimentation, and open-source research.

Quick Start

Install dependencies and run the included run script to generate a song from lyrics and tags.

Frequently Asked Questions about heartmula

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate music locally from lyrics without using cloud services?

To generate music locally from lyrics, you can use an open-source pipeline like HeartMuLa that runs offline, allowing you to create songs from text inputs without relying on proprietary cloud APIs.

What do I need to set up an offline AI music generation environment?

Setting up an offline AI music generation environment requires CUDA-enabled hardware, reproducible setup steps, and patching dependencies to run the model locally on your machine.

How does multimodal AI music generation work from text inputs?

Multimodal AI music generation works by using models like HeartMuLaGen to process lyrics and tags, while HeartCodec handles high-fidelity audio reconstruction to create the final song.

Can I transcribe and align audio files for local music research?

Yes, you can transcribe and align audio files locally using tools like HeartTranscriptor and HeartCLAP, which support open-source research and offline experimentation workflows.

Why patch dependencies when running an open-source music generation model locally?

Patching dependencies is necessary to ensure reproducible setup and compatibility when running an open-source music generation model locally on CUDA-enabled hardware.

Is offline music generation suitable for hobbyists and developers?

Offline music generation is suitable for researchers, developers, and hobbyists seeking an open-source pipeline for local experimentation and creating songs without cloud dependencies.