heartmula

Generate full music tracks from lyrics and comma-separated tags.

Updated May 16, 2026

Search Tags:#music generation #audio generation #multilingual #tags #lyrics #local ai #gpu inference

Ready to run

npx skills add https://github.com/jaredcroxton/hermes-mirror --skill heartmula-jaredcroxton

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: heartmula
Source: https://github.com/jaredcroxton/hermes-mirror/tree/main/agents/skills/media/heartmula
Command: npx skills add https://github.com/jaredcroxton/hermes-mirror --skill heartmula-jaredcroxton

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

HeartMuLa turns lyrics plus tag-style prompts into complete, music-like audio tracks without requiring cloud production services.

Core Features & Use Cases

Song generation from lyrics + tags: Builds full songs using bracketed section-structured lyrics and comma-separated tags.
Local/offline music workflow: Runs on your machine with configurable GPU splitting between the music model and codec.
Transcription and alignment models included: Supports related components (transcription and audio-text alignment) to expand creative workflows around the generated music.

Quick Start

Use heartmula to generate a song by providing your lyrics file and a tags file to the HeartMuLa music generation example script.

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 locally?▼

To generate music locally, you provide a lyrics file and a comma-separated tags file to the music generation script. The system runs entirely on your machine using configurable GPU or CPU device parameters.

Can I generate songs with multilingual lyrics on-device?▼

Yes, you can generate songs with multilingual lyrics on-device. The system processes bracketed section-structured lyrics like Intro, Verse, and Chorus to build full audio tracks without cloud services.

Do I need a GPU to run local audio generation workflows?▼

A GPU is not strictly required for local audio generation workflows, but it is recommended. You can configure the script to run on either a GPU or CPU, with options for splitting GPU memory between the music model and codec.

What is the best way to structure lyrics for AI music generation?▼

The best way to structure lyrics for AI music generation is using bracketed section markers like Intro, Verse, and Chorus. This structured format helps the model build complete, Suno-style song drafts accurately.

How do I install HeartMuLa and its checkpoints for offline song generation?▼

To install HeartMuLa for offline song generation, you must install the heartlib package and download both the HeartMuLa and HeartCodec checkpoints before running the provided music generation entry point script.

Are there limitations to generating Suno-style music tracks offline?▼

Generating Suno-style music tracks offline requires downloading large model checkpoints and demands significant hardware resources. You must manually configure device parameters to split GPU inference between the music model and the audio codec.