On Wednesday (local time in the US), Google introduced its latest music generation model, Lyria 3 Pro. As an enhanced iteration of Lyria 3, this model empowers users to create music tracks extending up to 3 minutes in length. It also offers refined creative control and customization features, allowing users to define the structure of their compositions with greater precision. Initially, Lyria 3 Pro will be accessible on the Gemini platform, exclusively for paid subscribers. There are also plans to incorporate it into Google Vids and ProducerAI in the future. For enterprise-level deployments, Google will facilitate music generation through Vertex AI, the Gemini API, and AI Studio. Regarding training data and adherence to copyright regulations, Google leverages partner and 'licensable data' to prevent 'mimicking' the styles of particular artists. Moreover, all generated tracks will incorporate SynthID markers to trace their origin. Furthermore, streaming services such as Spotify and Deezer are also expediting the deployment of tools to tackle AI-generated music.
