heartmula

Generate songs from lyrics and tags using PyTorch, torchtune, and HeartMuLa library.

Updated Apr 29, 2026

Search Tags:#AI #audio #music generation #tags #lyrics

Ready to run

npx skills add https://github.com/fikriaf/agentos --skill heartmula-fikriaf

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: heartmula
Source: https://github.com/fikriaf/agentos/tree/main/src/agentos/skills/media/heartmula
Command: npx skills add https://github.com/fikriaf/agentos --skill heartmula-fikriaf

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pytorch, torchtune, datasets, transformers, and includes scripts (resource) and assets (resource) components.

What problem does it solve?

This Skill enables the generation of music from provided lyrics and tags, addressing the need for users to create custom music without traditional music creation skills.

Core Features & Use Cases

Music Generation: Create full songs based on given lyrics and tags.
Multilingual Support: Generate music in multiple languages.
Use Case: For users seeking to produce music inspired by a specific set of lyrics or a theme without the need for in-depth musical knowledge or instruments.

Quick Start

Generate a song using the HeartMuLa skill by providing lyrics and tags, e.g., 'play song --lyrics "I want to go to the mountains" --tags "adventure, nature, acoustic".'

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?▼

You can generate multilingual music by inputting lyrics and tags into the Skill. It supports multiple languages for song creation, allowing you to produce audio from text descriptions without needing traditional musical instruments or knowledge.

Do I need PyTorch to generate audio from text descriptions?▼

Yes, you need PyTorch, torchtune, and the datasets and transformers libraries to generate audio from text. These dependencies are required for the model inference and audio generation processes used to create full songs.

Can I create a full song based on specific lyrics without musical instruments?▼

Yes, you can create a full song based on specific lyrics without musical instruments. The Skill is designed for users who want to produce music inspired by lyrics or themes without needing in-depth musical knowledge or physical instruments.

What are the limitations of generating music with PyTorch and torchtune?▼

Generating music with PyTorch and torchtune requires setting up a Python environment with specific dependencies like datasets and transformers. The Skill is limited to processing provided lyrics and tags, relying entirely on text inputs for audio generation.

How does multilingual music generation work?▼

Multilingual music generation works by processing text descriptions and lyrics in multiple languages. The Skill uses model inference to translate your provided tags and lyrics into personalized audio, creating songs across different languages.