Google has unveiled Lyria 3 Pro, an upgraded AI music generation model that extends track lengths to three minutes-ten times longer than its predecessor, Lyria 3, which capped compositions at just 30 seconds. The new model allows users to create longer, more structured music tracks with better artistic control.

Beyond longer tracks, Lyria 3 Pro offers enhanced control over music structure. Users can specify individual sections like intros, verses, choruses, and bridges in their prompts, as the model better understands the architecture of songs.

Integration of Lyria 3 Pro in Google’s apps and services

The original Lyria 3 was integrated into Google’s Gemini app when it launched last month, allowing users to generate music snippets. The Pro version will also be available in Gemini, but only to paid subscribers. It is furthermore set to appear in Google’s new video editor, Vids, and in ProducerAI-a generative AI music production tool Google acquired recently.

Google is embedding Lyria 3 Pro into its enterprise AI services as well, including Vertex AI (now in public preview), Gemini API, and AI Studio. These tools will harness the new model to deliver advanced music generation capabilities to developers and businesses.

Training data and transparency features of Lyria 3 Pro

To train Lyria 3 Pro, Google used datasets sourced from partners and licensed content from YouTube and its own platforms. The company emphasizes that the AI doesn’t replicate exact performances or artists but can reflect a general style when a user references a particular musician in their prompt.

All AI-generated tracks made with Lyria 3 and Lyria 3 Pro include SynthID technology tags, which mark the content as produced by artificial intelligence to promote transparency.

How Lyria 3 Pro compares to AI music tools from Apple and Microsoft

Unlike Apple’s more experimental AI-assisted music features in Logic Pro, or Microsoft’s integration of AI audio tools into gaming via Xbox, Google is targeting both creators and businesses with Lyria 3 Pro’s scalable, structured music generation. This approach could shift how music content is created for personal projects and professional media production alike.

Impact of Google’s advanced AI music model on creative workflows

Keep an eye on how Google’s expanded AI music models influence creative workflows and content authenticity standards, especially as AI-generated music gains traction beyond short loops to full-length compositions. The balance between user control, AI creativity, and licensing transparency will be key to widespread adoption.

Source: Techcrunch

Leave a comment

Your email address will not be published. Required fields are marked *