music-generation

Generate ambient background music using Lyria via the Interactions API.

npx skills add https://github.com/google-gemini/gemini-managed-agents-templates --skill music-generation

Music Generation

Generate background music for the radio show using the Lyria model via the Interactions API.

Embedded Script

python3 skills/music-generation/scripts/generate_music.py --workspace ./workspace --mood tech

Arguments

Argument	Default	Description
`--workspace`	`workspace`	Root workspace directory
`--mood`	`default`	Music mood (see moods below)

Moods

Mood	Style	Pair with `--style`
`tech` (default)	Clean synths, electronic pulse, Silicon Valley startup vibe	`explainer`, `debate`
`chill`	Soft pads, gentle piano, lo-fi warmth	`roundtable`, `explainer`
`debate`	Building tension, brass-like synths, panel discussion opener	`debate`, `interview`

What it does

Selects the music prompt based on --mood.
Calls the Interactions API with the lyria-3-clip-preview model.
Generates a ~30-second music clip.
Saves as MP3 to {workspace}/audio/music/background.mp3.
Gracefully handles failures — the pipeline continues without music.

Dependencies

google-genai (>= 2.0.0)

Output

File: {workspace}/audio/music/background.mp3
Format: MP3
Duration: ~30 seconds

Fallback

Lyria is experimental. If generation fails with a policy error or returns no music, the script will attempt to retry once with a simpler fallback prompt: "Create a 30-second simple ambient background track. Instrumental only, calm and neutral."

If the fallback attempt also fails, the pipeline proceeds without background music — the audio-mixing step handles this gracefully.