Google has released an upgraded version of its AI music generator, Lyria 3 Pro. The new model significantly expands the length of songs it can create, now producing full three-minute tracks compared to the 30-second limit of the initial version launched just last month. This update introduces much deeper customization for users. People can now prompt the AI to craft specific sections of a song, such as an intro, verse, chorus, or bridge. Google states that Lyria 3 Pro has a better understanding of musical composition than its predecessor, making it suitable for experimenting with various styles and creating songs with complex structural transitions. The tool is already accessible to paying Gemini subscribers and enterprise clients using Vertex AI. Developers can also integrate it via the Gemini API and Google AI Studio. Furthermore, Google is incorporating Lyria 3 Pro into its AI-powered video creation platform, Google Vids. Google emphasizes that responsibility was a core principle in the model’s design and training. The company asserts it only used materials for which it holds proper rights. All audio output from Lyria 3 Pro is also embedded with SynthID, a digital watermark designed to identify content as AI-generated. This launch arrives amidst a massive surge of AI-created music flooding online platforms. Current estimates indicate approximately 50,000 AI-generated tracks are uploaded to Spotify every single day. The streaming service reported deleting a staggering 75 million such tracks last year as part of its efforts to manage spam and low-quality content, highlighting the scale of AI music production. The introduction of Lyria 3 Pro, a more powerful tool from a major tech player, raises questions about the future volume and quality of AI music in an already saturated digital landscape.
