Google launches Lyria 3: Artificial intelligence that turns your photos and memories into songs
The new Gemini model allows you to generate music with vocals and lyrics from a simple description or an image from your gallery.
Generative artificial intelligence is no longer just a tool for writing emails or creating static images. Google has taken a bold step into the world of digital entertainment with the launch of Lyria 3, its most sophisticated music generation model to date. This update, which is currently in beta, aims to democratize artistic creation by allowing anyone, regardless of their technical knowledge, to compose a complete music track in a matter of seconds.
What really sets Lyria 3 apart from its competitors is not only its ability to generate beats, but also its artistic coherence. The system is now capable of interpreting the context of a prompt to write rhyming verses that fit the requested genre. Rather than being a tool for professional producers, Google presents it as a “pocket studio” designed for personal expression and everyday fun.
How can you create your own music with this AI?
The process has been simplified as much as possible to ensure a smooth and enjoyable experience. Currently, the platform offers two main ways to bring a composition to life:
- From text to melody: You can describe a feeling or an anecdote. For example, if you request an Afrobeat song about nostalgia for home cooking, the AI will automatically generate the instrumentation, lyrics, and vocals.
- From image to sound: This is perhaps the most innovative feature. When you upload a photo or video, Gemini “analyzes” the visual content to compose a matching soundtrack. If you upload a photo of your pet in a park, the system will detect the environment and create a track that captures that specific atmosphere.
Each creation is 30 seconds long and includes custom cover art. Users have the flexibility to download the result as a video with the cover image or simply the audio track in 192 Kbps stereo MP3 format, ideal for sharing on social media.
Tests we performed
We already had access to the beta version and decided to test it with an old Meristation article titled: “What is a squeaker? 10 keys to identifying them.” We asked it to create a song about the topic in a heavy metal style, and it gave us a very interesting result in thrash metal that you can find on our social media.
What about copyright and security?
One of the big debates in the industry is the use of famous artists’ voices. Google has been categorical: Lyria 3 is designed for original expression. If a user tries to force the style of a well-known singer, the system will only take the reference as general creative inspiration to avoid intellectual property infringements.
In addition, to ensure transparency in the age of deepfakes, all tracks include SynthID. This is a watermark imperceptible to the human ear that allows the audio to be identified as having been created with AI. The Gemini app will even allow external audio files to be uploaded to verify whether they contain this seal of authenticity.
When can you start using Lyria 3?
The rollout of this technology began today on the desktop version of Gemini and will be integrated into the mobile app in the coming days. The feature is available to users over the age of 18 and supports several languages.
Related stories
Although the tool is free, Google has set usage levels. Subscribers to Google AI Plus, Pro, and Ultra plans will have higher generation limits, allowing them to create more soundtracks per day. After our initial tests, such as creating a reggaeton song based on cultural articles, the result is surprisingly fun and opens a new door to digital creativity.
Follow MeriStation USA on X (formerly known as Twitter). Your video game and entertainment website for all the news, updates, and breaking news from the world of video games, movies, series, manga, and anime. Previews, reviews, interviews, trailers, gameplay, podcasts and more! Follow us now!
Complete your personal details to comment