754,000+ skills indexed

moistalgiaCOMMUNITY

Explore all skills by this author →

Verified:GitHub

heartmula

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

Updated Jul 27, 2026

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

One-click install

npx skills add https://github.com/moistalgia/hermes-tools --skill heartmula-moistalgia

Or copy as Structured Prompt for Agent▼

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Generating original music locally without relying on paid cloud services like Suno is difficult, and setting up open-source music generation models involves dependency conflicts, source patches, and multi-gigabyte model downloads. This Skill walks through the complete installation, patching, and generation workflow for HeartMuLa, an Apache-2.0 music foundation model family. ## Core Features & Use Cases - Lyrics-to-Song Generation: Produces full MP3 songs (48kHz stereo) from bracketed lyrics and comma-separated style tags using the 3B or 7B HeartMuLa models. - Guided Installation & Patching: Documents the required dependency upgrades (datasets, transformers) and two source-code patches needed to run HeartMuLa with transformers 5.x. - VRAM-Constrained Operation: Supports lazy loading to run on GPUs with as little as 8GB VRAM, or multi-GPU splits across devices. - Use Case: A user wants a royalty-free birthday song with custom lyrics. Provide lyrics with [Verse]/[Chorus] structure and tags like "piano,happy,celebration", then run the generation script to get a finished MP3 in about four minutes on a GPU. ## Quick Start Ask the assistant to install HeartMuLa and generate a song from your lyrics file and style tags, saving the result 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 music from lyrics with HeartMuLa?▼

Clone the heartlib repository, install it in a Python 3.10 virtual environment, download the model checkpoints from Hugging Face, then run examples/run_music_generation.py with paths to your lyrics file, tags file, and output MP3. Lyrics use bracketed structural tags like [Verse] and [Chorus].

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

HeartMuLa is an Apache-2.0 family of music foundation models that generates full songs from lyrics and style tags, comparable to Suno. It includes the HeartMuLa language model, HeartCodec for audio reconstruction, and HeartTranscriptor for lyrics transcription.

How much VRAM does HeartMuLa need to run?▼

HeartMuLa requires a minimum of 8GB VRAM using the --lazy_load true flag, which loads and unloads models sequentially, peaking around 6.2GB for the 3B model. 16GB or more is recommended, and multi-GPU setups can split models across devices.

Can HeartMuLa run on CPU without a GPU?▼

Yes, by passing --mula_device cpu --codec_device cpu, but generation is extremely slow, potentially 30-60+ minutes per song versus about 4 minutes on GPU, and requires 12GB+ of free RAM. A cloud GPU or the online demo is recommended instead.

Why does HeartMuLa fail after installing with transformers 5.x?▼

The pinned dependencies conflict with newer packages, requiring upgrades to datasets and transformers plus two source patches: reinitializing RoPE caches in modeling_heartmula.py and adding ignore_mismatched_sizes=True to HeartCodec loading calls.

Why are my style tags ignored during HeartMuLa generation?▼

Tag ignoring is a known upstream issue where lyrics tend to dominate the generation. Experimenting with tag ordering can help, but there is no guaranteed fix in the current release.

Related Skills

Explore More Design & Creative Skills→

minimax-music-gen

Generate vocal songs, instrumentals, and covers via the MiniMax Music API using the mmx CLI.

video-audio-extractor

Extract audio streams from video files as WAV or MP3.

music-lyrics

Extract timestamped lyrics from audio using faster-whisper.

heartmula

Generate complete AI music tracks from lyrics and style descriptions.