Google has added song creation and editing to Spaces, a music tool housed within its Flow Music platform.
Users of Spaces can now generate and edit songs, split recordings into stems, and generate images, lyrics and videos, all inside the tool.
The song features within Flow Music include the ability to create covers, replace sections and extend tracks.
The company announced the update via its Flow Music account on X on Friday (July 24).
“Spaces, our vibe coding tool for building any custom instrument or music app, just got a massive upgrade,” Google wrote in the post.
“You can now do it all right inside Spaces,” the company added, before listing song generation and editing, stem splitting, and image, lyric and video creation among the new capabilities.
When Google brought the platform – then called ProducerAI – into Google Labs in February, it pitched Spaces as a way for artists to build their own instruments and effects.
“As we continue to build ProducerAI at Google, we will be laser-focused on creative control for artists, including through features like Spaces, which allows artists to use natural language to create completely new instruments, effects and more,” said Elias Roman, Senior Director of Product Management at Google Labs, at the time.
“These mini-apps are shareable and remixable across users,” Roman added.
Flow Music runs on Lyria 3 Pro, a music-generation model built by Google DeepMind, and also draws on Google’s Gemini, Veo and Nano Banana models for chat, video and images. All output is embedded with SynthID, Google’s watermark for identifying AI-generated content.
The platform began as Riffusion, an open-source project that went viral in 2022, and launched as ProducerAI in 2025.
Google acquired the platform in February and rebranded it Flow Music in April, aligning it with the company’s Flow suite of AI image and video tools.
In May, Believe partnered with Google to offer Flow Music to artists across the distributor and its self-release arm, TuneCore.
Google and Believe described the tool as a “creative collaborator,” with a group of artists set to meet weekly with Google’s product team to shape its development.
For Believe, the deal formed one side of a two-pronged approach to AI: the company has been blocking the distribution of tracks made on unlicensed “pirate studios” while investing in what it calls “value-creative AI.”
“The adoption of Gen-AI is going to enhance human creativity,” Believe founder and CEO Denis Ladegaillerie told MBW at the time.
“I genuinely believe that Gen-AI will democratize creativity. And out of that democratization will come new forms of very powerful music,” Ladegaillerie said.
The additions move Flow Music closer to rivals Suno and Udio, two AI music platforms that have built their own editing and production tools.
Suno launched its own digital audio workstation, Suno Studio, in September 2025, combining AI music generation with multi-track editing and stem separation.
Udio released a visual editing workstation called Sessions in June 2025, which lets users move, extend and replace sections within a track.
“We know artists and producers use Udio alongside other tools, often jumping between platforms to shape their sound,” Andrew Sanchez, Udio’s co-founder and CEO, said at the launch.
“We’ve built Sessions to seamlessly integrate into those workflows, making it easier to visualize, edit, and experiment with tracks in one unified place.”
Google has said Lyria 3 Pro was built using materials that it and YouTube have “a right to use” under their terms of service, partner agreements and applicable law.
In March, a group of independent musicians sued Google, claiming it had trained Lyria 3 on copyrighted recordings taken from YouTube without permission or payment.Music Business Worldwide
