Generate Music via Google's Gemini API — Full Songs or 30-Second Clips
Describe the music you want and Lyria writes it. Lyria 3.5 composes a full song with verses, chorus, and vocals; Lyria 3 Clip returns a 30-second loop. An optional image can set the mood. Returns a 44.1 kHz stereo MP3 and, for songs with vocals, the lyrics. Runs on Google's side against your configured Google/Gemini API key.
🔑 Requires a Google / Gemini API key
What it does
Lyria turns a written description — genre, instruments, tempo, mood, and the words to sing if you have them — into a 44.1 kHz stereo MP3. Two variants are available: Lyria 3.5 writes a full-length song (about a couple of minutes) whose length you steer in the prompt, and Lyria 3 Clip always returns a 30-second clip at half the price.
An image on the optional port is read as a mood board, so the music can follow the look of a frame or product shot from earlier in the flow. Songs with vocals also come back with their lyrics on a separate text port. It is an external node: nothing runs locally, and Google bills your provider account.
Problem it solves
- Background tracks — A track for a short-form video, matched to its mood
- Full songs with vocals — A complete song from a theme and a few lines of lyrics
- Loops and drafts — 30-second cues for previews and drafts on Lyria 3 Clip
- Image-driven mood — Music that follows the look of an image wired in from earlier
- Lyrics for captions — The returned lyrics can feed subtitle or caption steps downstream
Input/Output
Input
- Text (required) — description of the music: genre, instruments, mood, tempo, and lyrics to sing
- Image (optional) — an image whose mood the music should follow
Parameters
Option | Values · what it does |
Model | Lyria 3.5 (default) — full-length song, about a couple of minutes, length steerable in the prompt. Lyria 3 Clip — always a 30-second clip; half the price, for loops and previews |
Style / extra instruction | Optional instruction prepended to the text input — e.g., "instrumental only, 90 BPM lo-fi." Leave blank to use the text input alone. Handy for keeping one style fixed while the text input changes |
Output
- Sound — generated music (MP3, 44.1 kHz stereo)
- Lyrics — lyrics of the generated song (empty for instrumentals)
How to Use
- Configure your Google/Gemini API key under your avatar → Workspace Settings → Integrations.
- Write a music description in Text Input and connect it to Lyria. Optionally connect an image to set the mood.
- Pick the variant: Lyria 3 Clip for quick 30-second drafts, Lyria 3.5 for the final full-length track.
- Connect Sound to Sound Output (and Lyrics to Text Viewer if the song has vocals), then run.
- Mix the approved music under narration with Audio Replace / Mix, using voice ducking where appropriate.
Example Workflow
Prompt: "Warm instrumental background music for a calm product demonstration: soft electric piano, light percussion, steady moderate tempo, about one minute, no vocals."
Flow: Text Input → Lyria → Sound Output.
Image-to-music: Image Loader → Lyria image port, with a short text prompt such as "music that matches this scene."
Use Case Tips
- Say what to avoid in the prompt — There is no negative-prompt field; write "no vocals" or "no drums" in the description.
- Steer length in words — There is no exact duration field on Lyria 3.5; write "about one minute" in the prompt.
- Draft on Clip, finish on 3.5 — Iterate on the cheaper 30-second variant, then switch once the style is right.
- Vocals — Put the lyrics in the text input; the generated lyrics come back on the Lyrics port.
Limitations
- No seed, negative prompt, or exact duration field — say what to avoid and how long it should run in the prompt itself.
- Lyria 3 Clip always returns 30 seconds regardless of what the prompt asks for.
- Runs on your own Google AI API key. Without one configured in the workspace, the node cannot run.
How it compares
What you need | Use |
Full songs with vocals, or image-driven mood, via your Google key | Lyria |
CNAPS-hosted music with lyrics, key, tempo, and time-signature control (48 kHz stereo, 5–600 s) | Text to Music (ACE-Step 1.5) |
Sound effects and ambience | Text to Sound Effect (MOSS-SoundEffect v2) |
Spoken narration | Text to Speech (Qwen3 / Qwen3 Voice Design / Gemini) |
Model Source
- CNAPS model page: Lyria
- Lyria — official music-generation guide: Google AI for Developers
- Lyria — official prompt guide: Lyria prompt guide
Compliance & Provenance
Item | Detail |
Provider | Google · GPAI |
License / service terms | Google Cloud Terms, and the terms applicable to your provider account |
EU AI Act risk class | Limited Risk |
Art. 50 transparency | Required — AI-generated outputs are marked |
For platform classification, acceptable use, and output marking, see the CNAPS AI Policy.
Run it now
Open CNAPS Studio, find Lyria in the Component panel, and start with the example workflow above.