Gemini · Status at source date: Product release
Google brings Lyria 3.5 music generation to Gemini
Google has released Lyria 3.5 in the Gemini app and the Gemini API, adding more flexible options for vocal or instrumental tracks and for shorter or longer output.
What changed
Google’s 4 September announcement introduces Lyria 3.5 in Gemini and the Gemini API. It describes more expressive vocals and richer arrangements, with choices for vocal or instrumental styles and short or longer tracks. It also names Google AI Studio, Flow Music and Vids as access routes. These are the company’s release statements, not an assessment of a particular generated track.
For this historical update, the relevant change is a wider set of ways to shape an audio concept. The announcement does not by itself establish the commercial permissions, restrictions or suitability of every output for an advertising campaign. Those checks depend on the product, applicable terms and the actual material the team wants to use.
Why it matters for advertisers
PMC analysis: audio generation can be useful early in a creative brief, when a team needs to compare mood, pacing and the relationship between music and a spoken message. Start with the communication problem: an explainer may need unobtrusive backing, whereas a short brand treatment may depend on a memorable musical phrase. Compare a few intentional directions instead of generating many tracks without a way to choose between them.
Evaluate the selected track in the edit where it will be heard. Music that sounds strong on its own can obscure a voice-over or change the tone of a product claim. Listen on the devices and at the levels relevant to the placement, and have someone review any generated vocal content. Retain the prompt and output version so revisions can be discussed precisely. Faster concept development is a plausible benefit; improved campaign performance must still be established through the creative test.
What to check next
- Create a short audio brief describing the intended mood, duration and role of music, then compare vocal and instrumental options where useful.
- Review the actual output in the proposed video or audio placement for speech clarity, pacing and unintended lyrics or associations.
- Check the applicable product terms and the selected asset’s suitability before campaign use, retaining the output version and creative-review decision.
Sources & contributor credit
- Newsletter coverage · Paid Media Collective newsletter
Original newsletter text, contributor labels and media for this update.
- Source referenced in newsletterGoogle Blog
Linked from the original newsletter. The source publication date has not been independently confirmed.
Source attribution checked
Google is the official announcement publisher. The inspected article text supplied no individual author or audio-asset creator.
Attribution evidence and limitations
Read full cached Google announcement, including release date, stated quality improvements, style/duration options and named product routes. No generated audio auditioned; no commercial-rights or independent quality claim.
Preserved original newsletter and announcement URLs. Inspected links to Lyria model, Gemini API and named products; usage terms were not separately reviewed and remain a pre-use check.
No attached newsletter or source visual/video in the record. No audio was generated or listened to, and no audio creator is inferred.
Attribution checked . This is a review date, not the original publication date.
- Published on this site
- Article updated
This update reflects the dated source reporting. Availability may have changed. Further coverage of this same development will be added to this page.
Original newsletter text and archive evidence
Google brings Lyria 3.5 music generation to Gemini.
Google has released Lyria 3.5 in the Gemini app and the Gemini API, adding more flexible options for vocal or instrumental tracks and for shorter or longer output. The company says the model improves vocals and musical arrangements and is also available through tools such as Google AI Studio and Vids. For you, this means another option for developing audio concepts, with creative quality and the applicable usage terms reviewed before campaign use.
Thank you to every practitioner who documented a change, added context, or helped verify the original source. Your work is what makes this a collective rather than another news roundup. 🤝
Source captured . No explicit first-contributor label was provided for this update.
