Google has added its Lyria 3.5 music model to the Gemini app, allowing users to choose a genre and style and generate either vocal or instrumental tracks. Templates for background music and personalized birthday songs provide starting points for people who do not want to build a detailed prompt.
The release extends beyond the consumer chatbot. Lyria 3.5 is available through Google AI Studio for developers, in Google Vids and through Flow Music with additional controls. That distribution lets the same model support casual creation, video soundtracks and applications built through an API.
Google claims the update produces more expressive vocals and richer arrangements than its predecessor. Those quality claims come from the company and will need practical comparison across genres, languages and longer compositions. The source does not provide pricing, regional limits or detailed generation lengths, so users should check the live product terms before planning production work.
The company says Lyria was trained only on licensed content, a material distinction in a market facing copyright lawsuits. It has not disclosed which catalogs were licensed or the terms, leaving creators unable to independently assess coverage. Generated music also remains subject to each product’s usage rules and any applicable rights in prompts, voices or requested styles.