heartmula

Generate MP3 music from lyrics and tags using HeartMuLa models.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/matlee0409/cronus --skill heartmula-matlee0409
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: heartmula
Source: https://github.com/matlee0409/cronus/tree/main/skills/media/heartmula
Command: npx skills add https://github.com/matlee0409/cronus --skill heartmula-matlee0409

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

HeartMuLa enables high-quality music creation directly from lyrics and user-provided tags using open-source models, offering offline capability and multilingual support.

Core Features & Use Cases

  • Open-source HeartMuLa models generate music conditioned on lyrics and tags.
  • HeartCodec provides high-fidelity audio reconstruction; HeartTranscriptor and HeartCLAP handle lyrics transcription and audio-text alignment.
  • Use cases include offline song generation on Linux, macOS, and Windows with multilingual lyrics support and local deployment.

Quick Start

Run the music generation script with your lyrics and tags to produce an MP3.

Frequently Asked Questions about heartmula

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

FAQPage Schema
How do I generate music from lyrics and tags offline?▼

To generate music offline, run the music generation script with your lyrics and tags to produce an MP3 using open-source models on your local machine.

Does open-source music generation support multilingual lyrics?▼

Yes, multilingual lyrics support is included for offline song creation, allowing you to condition the music generation using text in various languages.

What do I need for local song creation on Linux, macOS, and Windows?▼

Local song creation requires HeartMuLa checkpoints and HeartCodec for audio reconstruction, with optional HeartTranscriptor and HeartCLAP for transcription and alignment.

How does audio reconstruction work in offline music generation?▼

Audio reconstruction uses HeartCodec to build high-fidelity audio outputs from the generated music, while HeartTranscriptor and HeartCLAP handle lyrics transcription and audio-text alignment.

Can I use my own text tags to condition AI music generation?▼

Yes, you can condition open-source AI music generation directly by providing your own custom text tags alongside your lyrics to guide the output style.

What are the limitations of open-source offline music generation?▼

Limitations include the mandatory setup of specific checkpoints and audio codec dependencies for high-fidelity output, restricting immediate deployment without these model components.