Is the difference actually audible?

In a bed under a voiceover played from a phone, probably not. In a track that carries the film, on headphones, yes — mostly in the top end and in the sense of space.

Then why keep the standard version?

For pace. When you are trying three or four moods, what matters is how quickly you get back to comparing them, not how the piano decay sounds.

Can vocals reliably be turned off?

Yes, but through the brief rather than a switch. The app appends an instrumental-only request to what you wrote — probed on the same sentence with and without it: with the suffix it came back as guitar and cello, without it a vocal came in.

Music model · Google

Lyria 3 Pro when the track carries the whole spot

A bed running under a voiceover can be ordinary. A track the film stands on — no words, just picture and music — cannot. This version audibly holds the top end and the sense of space better, and it costs exactly what the standard one does, so the only reason not to pick it is the wait.

Claim 40 credits See pricing

Rendered by this model

Turn the sound on and listen

–:–– / –:––

A calm bed for a home fragrance advert: felt piano, sustained strings, one low swell. No vocals.This was generated by the model this page is about, exactly the way the app runs it. It is not a photo shoot and not client work.

Specification

Built by: Google
Starts from: a written brief · your own photo
Track length: the engine decides
Your own lyrics: does not take your own words
Vocals: steered by the brief, with no switch
Render time: about 45s
Cost: 1 credit per track

What it is good for

Detail you hear on good speakers: Piano action, bow noise, the decay of the room. These are the things that smear together on the standard version and make the difference on a large screen with real sound.
The same one credit: The provider bills us twice what the standard engine costs, and both still land inside one credit — so the better version costs you nothing extra.
Takes a picture too: Attach a finished graphic and the engine scores that shot. It is the shortest route there is from a finished ad to music that suits it.

When to pick something else

Three times slower than the standard version: about forty-five seconds against fifteen. While you are auditioning four moods in a row that is two minutes of waiting for fidelity nobody is judging yet — use Lyria 3 for that. The rest of the limits are the same: it will not sing your words (that is MiniMax Music 2.6) and it takes no length (that is ElevenLabs Music). Its content checker refused one of our briefs here in full without naming the word; if that happens, rewrite the brief with plain instrument names.

See every model

How to brief this model

The things this engine handles differently from the rest. The habits that hold for every model are collected separately, in the help centre.

Use plain instrument names. The content checker refused one of our briefs outright without naming the word; reworded to "piano, strings, gentle percussion" it went through.
Take it for the final version, not for finding the mood. Forty-five seconds a track is three times longer than Lyria 3, and at the audition stage nobody is judging that difference.
Listen to the result on decent equipment or headphones. The advantage of this version is in detail a phone speaker will not reproduce.
Write in that there should be no vocal if a voiceover goes over the track. There is no switch here, same as in the basic version.
It does take a photograph as a hint about mood, so a bed for a specific creative is worth ordering with that graphic attached.

How to write a brief that works

Questions about this model

Other models

Try this model on your own product

Sign up without a card and get 40 credits. Switching model is one click, so you can send the same brief through several of them and keep the best frame.

Claim 40 credits Work out your saving