Photo Credit: Google Flow Music
Google’s Flow Music Spaces gets a big update, bringing song editing, stem splitting, video creation, and more into an AI “vibe coding” workspace.
Google has announced a major update for its Flow Music Spaces, unveiling a suite of new AI-powered creative tools for collaborative workspaces. The update means users can now generate and edit songs, split stems, create images, write lyrics, and generate video without swapping between different apps.
“Spaces, our vibe coding tool for building any custom instrument or music app, just got a massive upgrade,” writes Google. “You can now do it all right inside Spaces: generate and edit songs (cover, replace, extend), split stems, generate images, generate lyrics, (and) generate videos.”
Not having to jump between multiple AI tools or sessions is a huge upgrade for Google’s creative AI suite and sets it apart from its contemporaries. Using “vibe coding” tools, a musician can now generate a song, replace a verse, separate vocals from instrumentals, extend the track, create album artwork, write lyrics, and even produce a video all from the same workspace.
What began as a space for collaboration has rapidly evolved into a central hub for AI-assisted music production. It certainly won’t replace your professional digital audio workspace, but the update makes Flow Music significantly more useful for actual musicians, producers, and (of course) hobbyists.
You might remember Riffusion back in 2022, an open-source project that went viral before officially launching as ProducerAI last year. Google scooped it up in February and quickly rebranded it to align with its Flow suite of AI video and image tools.
Since then, the company has been adding new features over the past several months, expanding its AI-powered creative capabilities for creators. The new update makes the workspace far more capable and all-inclusive, meaning people are more likely to actually use it beyond just playing around with AI tools.
Flow Music utilizes Lyria 3 Pro, Google DeepMind’s music-generation model. It also draws on Google’s Gemini chat model, as well as Veo and Nano Banana for video and images. Outputs are embedded with SynthID, Google’s watermark for easily identifying AI-generated content.
The latest updates bring Flow Music more on par with AI-music giants Suno and Udio, both of which also have their own editing and production suites. Suno Studio launched in September, while Udio released its Sessions workspace in June last year.
Like Suno and Udio, Google’s Lyria 3 Pro has been the subject of controversy and litigation. A group of independent musicians sued the company in March, claiming it had trained Lyria 3 on copyrighted recordings on YouTube without permission or payment. Google asserts that it has the right to use anything on YouTube to train its models per its terms of service and partner agreements.
Leave a Reply