ai-music-generation

Generate music and songs via the inference.sh CLI using ElevenLabs, Diffrythm, and Tencent models.

Updated Jul 28, 2026
One-click install
npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill ai-music-generation-aadi-110i
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ai-music-generation
Source: https://github.com/Aadi-110i/PEP-PROJECT/tree/main/skills/ai-music-generation
Command: npx skills add https://github.com/Aadi-110i/PEP-PROJECT --skill ai-music-generation-aadi-110i

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Creating original music for videos, podcasts, games, or social media normally requires composition skills, licensing negotiations, or expensive stock music subscriptions. This Skill lets you generate royalty-free tracks, full songs with vocals, and jingles directly from text prompts through the inference.sh CLI. ## Core Features & Use Cases - Multiple Music Models: Choose between ElevenLabs Music (up to 10 minutes with commercial license), Diffrythm (fast song generation), and Tencent Song Generation (full songs with vocals and custom lyrics). - Text-to-Music Generation: Describe genre, mood, instruments, and structure in a prompt to produce instrumental tracks, soundtracks, or jingles. - Use Case: A content creator needs background music for a YouTube video. They run a single CLI command with a prompt like "calm lo-fi hip hop beat, study music" and receive a ready-to-use audio track. ## Quick Start Ask the AI to generate an upbeat electronic dance track using the ai-music-generation skill via the belt CLI.

Frequently Asked Questions about ai-music-generation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate AI music from a text prompt?

Run belt app run infsh/diffrythm with a JSON input containing your prompt, such as an upbeat electronic dance track description. The inference.sh CLI returns a generated audio file based on your genre, mood, and instrument keywords.

Which AI music model should I use for songs with vocals?

Use Tencent Song Generation (infsh/tencent-song-generation) for full songs with vocals, since it accepts both a prompt and custom lyrics. Diffrythm is better for fast instrumental generation, while ElevenLabs Music supports tracks up to 10 minutes.

Can I use AI-generated music commercially?

ElevenLabs Music includes a commercial license and supports tracks up to 10 minutes long, making it suitable for commercial projects. Check the licensing terms of each model on inference.sh before publishing generated audio.

How do I add custom lyrics to an AI-generated song?

Run belt app sample infsh/tencent-song-generation --save input.json to create a template, then edit the lyrics field with your text. Pass the edited file with belt app run to generate a song using your lyrics.

What are the limitations of AI music generation?

Output length and quality depend on the model chosen; Diffrythm targets fast generation while ElevenLabs supports up to 10 minutes. Results vary with prompt specificity, so include genre, mood, instrument, and structure keywords for better control.