754,000+ skills indexed

LamseyahEliasCOMMUNITY

Explore all skills by this author →

Verified:GitHub

heartmula

Generate complete songs locally from lyrics and tags using HeartMuLa.

Updated Jun 11, 2026

Search Tags:#audio #open-source #codec #lyrics #music-generation #ai-music #heartmula

One-click install

npx skills add https://github.com/LamseyahElias/jarvis-cloud-v2 --skill heartmula-lamseyahelias

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: heartmula
Source: https://github.com/LamseyahElias/jarvis-cloud-v2/tree/main/hermes-agent/skills/media/heartmula
Command: npx skills add https://github.com/LamseyahElias/jarvis-cloud-v2 --skill heartmula-lamseyahelias

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

HeartMuLa provides local, open-source music generation by turning lyrics and descriptive tags into full songs, enabling offline creation without relying on cloud services.

Core Features & Use Cases

Open-model music generation: HeartMuLa Music Language Model (3B/7B) produces songs conditioned on lyrics and tags.
High-fidelity audio: HeartCodec reconstructs audio with a 12.5Hz codec for quality.
Offline & multilingual: Works locally with multilingual lyrics support and CPU/GPU options; includes optional GPU acceleration.

Quick Start

Run the example script to generate a 4-minute track from your lyrics and tags using the HeartMuLa pipeline.

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 without using cloud services?▼

Open-source AI music generation locally uses lyrics and descriptive tags to produce complete songs offline. By running a local music language model pipeline, it generates full tracks without relying on cloud APIs, enabling private creation.

Can I generate multilingual lyrics with offline AI music generation?▼

Yes, offline AI music generation supports multilingual lyrics input. It processes text in multiple languages alongside descriptive tags to condition the local model, producing high-fidelity audio tracks for diverse linguistic requirements.

Does local AI music generation require a GPU to produce high-fidelity audio?▼

Local AI music generation does not strictly require a GPU, as it supports both CPU and GPU options. However, optional GPU acceleration significantly speeds up the local model loading and audio generation pipeline for high-fidelity output.

What is the best way to create a 4-minute track from text tags and lyrics?▼

The best way to create a 4-minute track from text tags and lyrics is using a local music generation pipeline. Running the example script conditions a music language model on your inputs to output a complete MP3 stereo track at 48kHz.

What audio format and sample rate does offline AI music generation output?▼

Offline AI music generation outputs high-fidelity MP3 audio at 48kHz stereo. It reconstructs audio using a 12.5Hz codec, ensuring quality while maintaining an accessible file format for the generated complete song.

Related Skills

Explore More Design & Creative Skills→

bitwize-music-studio

clipboard

Copy lyrics, style prompts, or streaming lyrics from music track files to the system clipboard.

heartmula

Generates full songs from user lyrics and style tags as MP3 files.

heartmula

Generate full MP3 songs from lyrics and tags using HeartMuLa models.

heartmula

Generate music and songs from lyrics and descriptive tags using HeartMuLa.