heartmula

Generate original music from lyrics using HeartMuLa models.

13|3|Updated May 12, 2026
One-click install
npx skills add https://github.com/kevinnft/ai-agent-skills --skill heartmula-kevinnft
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: heartmula
Source: https://github.com/kevinnft/ai-agent-skills/tree/main/skills/media/heartmula
Command: npx skills add https://github.com/kevinnft/ai-agent-skills --skill heartmula-kevinnft

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

HeartMuLa addresses the challenge of creating music from text by turning lyrics and descriptive tags into full songs using open-source models.

Core Features & Use Cases

  • Open, offline music generation from lyrics and tags for creative workflows.
  • Multilingual lyric-conditioned synthesis with configurable model sizes and hardware.
  • Use Case: artists can generate draft songs from a simple text description and refine them locally.

Quick Start

Run a basic music generation by providing your lyrics and tags files.

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 locally?

To generate music from lyrics locally, you provide your text descriptions and lyrics files to the HeartMuLa models. The Skill synthesizes original songs offline using configurable model sizes and hardware setups.

Can I use multilingual lyrics for AI audio generation?

Yes, you can use multilingual lyrics for AI audio generation. The Skill supports multilingual lyric-conditioned synthesis, allowing you to create songs from text descriptions across different languages.

Do I need CUDA and PyTorch to run offline music generation?

Yes, you need a PyTorch environment with CUDA support for offline music generation. This setup ensures efficient, high-fidelity audio synthesis and allows the process to scale from single-GPU to multi-GPU configurations.

What components are required for text-to-music synthesis?

Text-to-music synthesis requires HeartMuLa model checkpoints and HeartCodec integration. These open-source components process your lyrics and tags to produce high-fidelity original audio tracks.

Does open-source music generation work without an internet connection?

Yes, open-source music generation works without an internet connection. The Skill applies offline local music creation, allowing artists to generate and refine draft songs entirely from their local machine.

How do I scale audio synthesis across multiple GPUs?

You scale audio synthesis across multiple GPUs by configuring the hardware setup in your environment. The Skill supports scalable generation from single-GPU to multi-GPU setups for efficient high-fidelity output.