From lyria-audio
Generates AI music using Gemini Lyria (realtime streaming). Handles duration, BPM, brightness, density, scale, loop detection, and MP3/WAV output. Use when asked to "generate music", "create BGM", "make a soundtrack", "generate audio", or any music generation task.
How this skill is triggered — by the user, by Claude, or both
Slash command
/lyria-audio:lyria-audioThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
AI music generation CLI. Powered by Gemini Lyria (realtime streaming).
AI music generation CLI. Powered by Gemini Lyria (realtime streaming).
When the user says "init", "setup lyria", or "install lyria-audio", run these commands. No sudo required.
Prerequisites: Bun must be installed. If not: curl -fsSL https://bun.sh/install | bash
# 1. Copy tool from plugin to ~/tools/lyria
cp -r "${CLAUDE_PLUGIN_ROOT}/tools" ~/tools/lyria
# 2. Install dependencies
cd ~/tools/lyria && bun install
# 3. Link globally (creates `lyria` command via Bun - no sudo)
cd ~/tools/lyria && bun link
# 4. Set up API key
mkdir -p ~/.lyria
echo "GEMINI_API_KEY=<ask user for their key>" > ~/.lyria/.env
After init, the user can type lyria "prompt" from anywhere.
If bun link fails or the command is not found after linking, fall back to:
mkdir -p ~/.local/bin
ln -sf ~/tools/lyria/src/cli.ts ~/.local/bin/lyria
# Then ensure ~/.local/bin is on PATH:
# macOS/Linux (zsh):
echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.zshrc && source ~/.zshrc
# Windows (Git Bash):
echo 'export PATH="$HOME/.local/bin:$PATH"' >> ~/.bashrc && source ~/.bashrc
Windows note: On Windows with Git Bash, use the full Bun path instead of bun:
# Find your Bun path:
where bun # (cmd) or: ls /c/Users/*/AppData/Local/Microsoft/WinGet/Packages/Oven-sh.Bun*/bun-windows-x64/bun.exe
# Then run directly:
/path/to/bun.exe run ~/tools/lyria/src/cli.ts "your prompt"
Get a Gemini API key at: https://aistudio.google.com/apikey
lyria "prompt" [options]| Option | Default | Description |
|---|---|---|
-o, --output | lyria-{timestamp} | Output filename (no extension) |
-d, --dir | current directory | Output directory |
--duration | 30 | Duration in seconds |
--bpm | auto | Beats per minute (60-200) |
--brightness | auto | Tonal brightness: 0.0 (dark) to 1.0 (bright) |
--density | auto | Arrangement density: 0.0 (sparse) to 1.0 (full) |
--guidance | 4.0 | Prompt adherence: 0 (free) to 6 (strict) |
--temperature | 1.1 | Generation randomness |
--scale | auto | Musical key: C_MAJOR, A_MINOR, etc. |
--mode | QUALITY | Generation mode: QUALITY, DIVERSITY, VOCALIZATION |
--format | mp3 | Output format: mp3 or wav |
--seed | random | Seed for reproducible output |
--no-loop | - | Disable loop-point detection |
--api-key | - | Gemini API key (overrides env/file) |
--costs | - | Show cost summary |
| Mode | Description |
|---|---|
QUALITY | Default. Best audio fidelity and coherence |
DIVERSITY | More variation and exploration in the output |
VOCALIZATION | Generates vocal-like sounds and textures |
0.0 - Dark, somber, melancholic tones0.5 - Neutral, balanced mood1.0 - Bright, uplifting, cheerful tones0.0 - Sparse, ambient, minimal instrumentation0.5 - Moderate arrangement1.0 - Full, complex, layered instrumentation60-80 - Slow, ambient, contemplative90-120 - Moderate, walking pace, relaxed120-150 - Energetic, driving150-200 - Fast, intense, action-packed0 - Free generation, loosely follows prompt3-4 - Balanced (default)6 - Strict prompt adherence, less creative freedomlyria "epic orchestral battle theme, fantasy RPG" --duration 60 --bpm 140 --density 0.8 -o battle-bgm
lyria "mysterious dark ambient, ancient ruins" --duration 90 --brightness 0.2 --density 0.3 -o ambient
lyria "calm hopeful piano melody, warm" --duration 45 --brightness 0.7 --bpm 90 -o menu
lyria "intense cinematic orchestral, rising tension" --duration 30 --density 0.9 --guidance 5.0 -o cinematic
lyria "ethereal choir, floating harmonies" --mode VOCALIZATION --duration 30 -o choir
lyria "retro synthwave beat" --seed 42 --bpm 120 -o synthwave
Every generation is logged to ~/.lyria/costs.json. View summary:
lyria --costs
Lyria is free during the experimental preview period.
The CLI resolves the Gemini API key in this order:
--api-key flagGEMINI_API_KEY environment variable.env file in current directory.env file next to the CLI script~/.lyria/.envGet a key at: https://aistudio.google.com/apikey
npx claudepluginhub whiskychoy/whisky-claude-plugins --plugin lyria-audioGenerates AI music, songs with vocals (Suno/Udio), and instrumentals (Google Lyria). Guides API selection, prompt crafting, and workflow for genres, moods, BPM.
Generates vocal/instrumental songs and audio covers using MiniMax Music API via mmx CLI. Supports prompts, lyrics optimization, moods, genres, and reference audio.
Generates AI music tracks via ElevenLabs API from text prompts or JSON-defined multi-section compositions. Ideal for soundtracks, jingles, background music, and instrumentals up to 5 minutes.