Google DeepMind has released Lyria 3.5, the latest version of its artificial intelligence model designed to generate music. The update arrives as the tech industry continues to push generative AI into creative fields, from text and images to audio.
What Lyria 3.5 does
Lyria 3.5 is an AI model that creates original music based on user prompts. It can produce instrumental tracks, melodies, and full compositions. The model builds on earlier versions of Lyria, which Google DeepMind first introduced as a research project in AI-driven music creation.
The company has not disclosed specific technical details about the new version, such as the size of the training dataset or the number of parameters. It also hasn't announced a public release date or pricing for Lyria 3.5. The model appears to be in an early or research stage, with no immediate consumer-facing product confirmed.
Lyria 3.5 enters a crowded field. Other companies, including OpenAI with Jukebox and startups like Soundraw and AIVA, have developed AI music generators. Google itself has experimented with music AI through its Magenta project and earlier tools like MusicLM. Lyria 3.5 is distinct because it comes from DeepMind, the AI research lab known for breakthroughs in reinforcement learning and protein folding.
The launch signals that Google sees music generation as a key area for AI development. But the company faces questions about copyright, artist compensation, and the potential for AI to displace human musicians. DeepMind has not addressed those issues in connection with Lyria 3.5.
Google DeepMind has not set a timeline for a wider rollout of Lyria 3.5. The company may release more details at an upcoming conference or in a research paper. For now, the model remains a demonstration of what AI can do in music — without a clear path to commercial use.
