754,000+ skills indexed

brittb-devCOMMUNITY

Explore all skills by this author →

Verified:GitHub

heartmula

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

1|Updated Feb 17, 2026

Search Tags:#music generation #audio synthesis #heartmula #gpu inference #lyrics to music #open-source suno #ai songs

One-click install

npx skills add https://github.com/brittb-dev/zerogravityclaw --skill heartmula-brittb-dev

Or copy as Structured Prompt for Agent▼

Please help me install this Agent Skill.
Skill: heartmula
Source: https://github.com/brittb-dev/zerogravityclaw/tree/main/src/hermes-core/skills/media/heartmula
Command: npx skills add https://github.com/brittb-dev/zerogravityclaw --skill heartmula-brittb-dev

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Generating complete songs with vocals from text requires either paid services like Suno or complex local model setups. This Skill guides you through installing and running HeartMuLa, an open-source music foundation model family, to produce full songs locally from lyrics and style tags. ## Core Features & Use Cases - Song Generation from Lyrics + Tags: Produce MP3 songs (48kHz stereo) conditioned on bracketed lyrics structure and comma-separated style tags using 3B/7B models. - VRAM-Constrained Operation: Run on GPUs with as little as 8GB VRAM using lazy loading, or split models across multiple GPUs. - Dependency & Patch Guidance: Includes required fixes for transformers 5.x compatibility, RoPE cache initialization, and HeartCodec checkpoint loading. - Use Case: A musician wants an open-source Suno alternative to prototype song ideas locally. Provide lyrics with [Verse]/[Chorus] tags and style tags like "piano,happy,wedding" to generate a 4-minute song in roughly 4 minutes on GPU. ## Quick Start Ask the AI to install HeartMuLa and generate a song from your lyrics file and style tags using the 3B model with lazy loading enabled.

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 with an open-source model?▼

Install the heartlib package, download the HeartMuLa-oss-3B 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 is a good open-source Suno alternative for local song generation?▼

HeartMuLa is an Apache-2.0 licensed family of music foundation models comparable to Suno. It generates full songs conditioned on lyrics and style tags, with multilingual support, running entirely on your own hardware.

How much VRAM does HeartMuLa need to generate a song?▼

The 3B model runs in about 6.2GB VRAM with lazy_load enabled, making 8GB the minimum. 16GB or more is recommended, and multi-GPU setups can split the language model and codec across devices.

Can HeartMuLa run on CPU or without an NVIDIA GPU?▼

Yes, by setting --mula_device cpu and --codec_device cpu, but generation is extremely slow, potentially 30-60+ minutes per song with 12GB+ RAM required. A cloud GPU or the online demo at heartmula.github.io is recommended instead.

Why does HeartMuLa fail after upgrading transformers?▼

Newer transformers versions break RoPE cache initialization and HeartCodec checkpoint loading. Apply the documented patches: reinitialize Llama3ScaledRoPE caches in modeling_heartmula.py and add ignore_mismatched_sizes=True to HeartCodec.from_pretrained calls.

What are the limitations of HeartMuLa music generation?▼

Style tags may be ignored since lyrics dominate conditioning, Triton GPU acceleration is unavailable on macOS, RTX 5080 incompatibility has been reported, and HeartCodec must use float32 rather than bfloat16 to preserve audio quality.

Related Skills

Explore More Design & Creative Skills→

heartmula

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

video-audio-extractor

Extract audio streams from video files as WAV or MP3.

music-lyrics

Extract timestamped lyrics from audio using faster-whisper.

music-generation

Generate instrumental or vocal music tracks from natural-language prompts.