Alphabet has significantly upgraded its AI music generation capabilities with the launch of the Lyria 3 Pro model, which can create musical pieces lasting up to three minutes. On Wednesday, March 25, the company announced the new model, arriving just one month after the previous Lyria 3 version.

The new model dramatically increases the maximum single-generation track length from 30 seconds to 3 minutes. It also supports detailed customization of musical structural elements. Lyria 3 Pro will be simultaneously deployed across the Gemini application (for paying users only), the video creation tool Google Vids, the AI music production platform ProducerAI, and enterprise-focused services like Vertex AI, the Gemini API, and AI Studio. This rollout covers multiple core scenarios from consumer to enterprise levels.

This series of deployments indicates that Alphabet is accelerating the integration of AI music capabilities into its product ecosystem and seeking to open new revenue streams in the enterprise market.

**Core Upgrade: From Fragments to Complete Compositions** The most significant changes in Lyria 3 Pro compared to Lyria 3 are the generation duration and structural awareness. The new model supports the creation of tracks up to 3 minutes long and can recognize and generate specific musical structure elements, including intros, verses, choruses, and transition passages. This allows users to specify a complete song architecture directly in their prompts, rather than just describing a style or mood.

Alphabet stated that Lyria 3 Pro is particularly suited for creative scenarios requiring complex passage transitions or experimentation with multiple styles. It offers a more complete solution for content creation such as podcast scoring, Vlog background music, and tutorial videos.

The company is attempting to enhance the credibility of its Lyria series within the creative community through collaborations with professional musicians. Grammy Award-winning producer Yung Spielburg has already used Lyria for scoring and in the production process of a Google DeepMind short film. DJ and producer François K has used Lyria iteratively to create a soon-to-be-released new song. Alphabet's Music AI Sandbox project also continues to provide experimental tools for musicians, producers, and songwriters, with related feedback incorporated into the development of Lyria 3.

**Multi-Platform Deployment for Developers, Enterprises, and Creators** Alphabet is extending the reach of Lyria 3 Pro to several different audience groups. For enterprise users, Lyria 3 Pro is now in public preview on Vertex AI, supporting on-demand, large-scale generation of high-fidelity audio for applications like rapid customization of game soundtracks and integration into creative tools.

For developers, Lyria 3 Pro is available in AI Studio, offered alongside Lyria RealTime and accessible via the Gemini API for building next-generation creative tools. For general users, the AI video creation app Google Vids will begin offering Lyria 3 and Lyria 3 Pro features this week to Google Workspace customers and Google AI Pro and Ultra subscribers, allowing them to add custom soundtracks to creative projects or marketing videos. The long-track generation feature is also being made available to paying users within the Gemini app.

Regarding ProducerAI, Alphabet completed its acquisition of the AI music production tool last month. Following the integration of Lyria 3 Pro, ProducerAI will provide an agent-like creative experience for artists, producers, and songwriters and is now available to free and paid users globally.

**Copyright and Provenance: SynthID Marking and Training Data** Amid ongoing industry focus on copyright issues surrounding AI-generated content, Alphabet has disclosed the model's data sources and content identification mechanisms. The company stated that Lyria 3 Pro's training data comes from partner-licensed data, as well as available data from YouTube and its other platforms.

The company emphasized that the model will not directly imitate a specific artist's style. However, if a user specifies an artist in the prompt, the model will draw "broad inspiration" from that artist's style to generate a track. All audio generated through Lyria 3 and Lyria 3 Pro will be embedded with a SynthID watermark to identify its AI-generated origin.

This move aligns with broader industry trends. It was reported that Spotify introduced a new tool this week allowing artists to verify songs released under their name to prevent misattribution of AI-generated content. Streaming platform Deezer has also launched a tool for identifying AI music that can be used by any streaming service.