754,000+ skills indexed

ossoolliCOMMUNITY

Explore all skills by this author →

Verified:GitHub

heartmula

Generates full songs from lyrics and style tags using open-source HeartMuLa music models.

Updated May 18, 2026

Search Tags:#open source #music generation #AI audio #heartmula #lyrics to song #text to music #gpu inference

One-click install

npx skills add https://github.com/ossoolli/Nexum-Core --skill heartmula-ossoolli

Or copy as Structured Prompt for Agent▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Creating original music typically requires composition skills, instruments, or expensive commercial services like Suno. This Skill lets you generate complete songs locally from plain-text lyrics and style tags using the open-source HeartMuLa model family, with no cloud subscription required. ## Core Features & Use Cases - Lyrics-to-Song Generation: Produces full MP3 tracks (48kHz stereo) conditioned on bracketed lyrics and comma-separated style tags, with multilingual support. - Local GPU Inference: Runs the 3B/7B HeartMuLa language model plus HeartCodec audio decoder on your own hardware, with lazy loading to fit in 8GB VRAM. - Use Case: A content creator needs a custom background song for a video. They write lyrics with [Verse]/[Chorus] structure, specify tags like "piano,happy,wedding", and generate a 4-minute track in roughly 4 minutes on a single GPU. ## Quick Start Generate a song from my lyrics file and the tags "rock,energetic,guitar" using the HeartMuLa 3B model and save it as 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 a song from lyrics with AI locally?▼

Install the heartlib package, download the HeartMuLa and HeartCodec checkpoints from Hugging Face, then run run_music_generation.py with a lyrics file using [Verse]/[Chorus] tags and a comma-separated tags file. Output is a 48kHz stereo MP3.

What open-source alternative to Suno exists for music generation?▼

HeartMuLa is an Apache-2.0 open-source music model family comparable to Suno, generating full songs from lyrics and tags. It includes a 3B/7B music language model, the HeartCodec audio codec, and lyrics transcription tools.

How much VRAM does HeartMuLa music generation need?▼

HeartMuLa needs a minimum of 8GB VRAM with lazy_load enabled, which peaks around 6.2GB for the 3B model. 16GB or more is recommended, and multi-GPU setups can split the language model and codec across devices.

Can HeartMuLa run on CPU or macOS?▼

CPU mode works via --mula_device cpu but is extremely slow, taking 30-60+ minutes per song and requiring 12GB+ RAM. Triton GPU acceleration is Linux/CUDA only, so macOS users should use a cloud GPU or the online demo.

Why are my style tags ignored during song generation?▼

Tag ignoring is a known upstream issue (#90) where lyrics tend to dominate the conditioning signal. Experimenting with tag ordering and phrasing can help the tags influence the generated style.

Related Skills

Explore More Design & Creative Skills→

music-lyrics

Extract timestamped lyrics from audio using faster-whisper.

KaiserWhoLearns

FFmpeg Audio Processing

Extract, normalize, and mix audio from video files using FFmpeg commands.

yue2-music

Generate, cover, transcribe, and edit songs with YuE2 and SheetSage2 via ABC scores.

audiocraft-audio-generation

Generate music and sound effects from text prompts with AudioCraft models.