Google is consolidating its creative AI suite under a unified banner. The company’s experimental music generation tool, formerly known as ProducerAI, has been rebranded as Google Flow Music. This shift aligns the service with Google’s "Flow" video editor within the Google Labs ecosystem, signaling a move toward a more integrated, multimodal creative environment for digital creators.
The platform is built on a foundation of the Lyria 3 audio model and the Gemini large language model, enabling the generation of full tracks that include vocals and complex arrangements. Recent updates to the underlying architecture have extended the maximum duration of these synthetic compositions to three minutes. Unlike earlier iterations of text-to-audio tools that functioned as simple prompt boxes, Flow Music centers its interface on a "Producer Agent"—a conversational system designed to mimic the collaborative dynamics of a professional recording studio.
This update also introduces more granular editing capabilities, most notably a "Remix" feature that allows users to replace or extend specific segments of an existing track through natural language commands. Currently available as a web-based application, Flow Music operates on a daily credit system without requiring a subscription, offering a low-friction entry point for users to experiment with high-fidelity generative audio.
With reporting from [Canaltech].
Source · Canaltech
