Google has expanded its Lyria music generation platform with a Pro tier capable of producing tracks up to 3 minutes long.
The tech giant is positioning the technology as a potential alternative to royalty-free production music services like Epidemic Sound, with the company suggesting creators could use it to generate “personalized tracks for vlogs, podcasts or tutorial videos”.
Lyria 3 Pro, an upgrade to the Lyria 3 model introduced last month, adds what the company calls “structural awareness,” which means the model now better understands musical compositions, following prompts for specific elements like intros, verses, choruses and bridges.
The previous version only allowed users to describe a genre, mood, or scenario and receive an AI-generated track complete with vocals and lyrics. Lyria 3 also limited users to 30-second audio clips.
Lyria 3 Pro now lets users create tracks of up to 3 minutes long. Google said: “Lyria 3 Pro’s enhanced customization offers more space to experiment and play with longer tracks. So now, you can add more details to bring your full vision to life.”
“Lyria 3 Pro’s enhanced customization offers more space to experiment and play with longer tracks. So now, you can add more details to bring your full vision to life, or create personalized tracks for vlogs, podcasts or tutorial videos.”
In addition to the Gemini app, the Pro tier is also embedded in Google Vids, the company’s AI-powered video creation tool, letting users add custom music to match their video content.
The Google Vids integration is rolling out to Google Workspace and Google AI Pro & Ultra starting this week.
Google is also opening Lyria 3 Pro to enterprise customers through Vertex AI, now in public preview, and to developers through Google AI Studio and the Gemini API.
The company is touting the Vertex AI integration toward gaming studios and video platforms that need to generate custom soundtracks at scale.
Google has not specified how Lyria 3 was trained and is carefully choosing the language around the training that went into it. The platform reiterated that the design and training of Lyria 3 used “materials that YouTube and Google has a right to use under our terms of service, partner agreements, and applicable law.”
Google added in its blog post: “To protect original expression, Lyria 3 and Gemini do not mimic artists; if a prompt names a creator, the model takes that as broad inspiration. Additionally, we employ filters to check outputs against existing content and users must adhere to the Terms of Service and Gen AI prohibited use policies, which prohibit violating others’ intellectual property and privacy rights.”
“To protect original expression, Lyria 3 and Gemini do not mimic artists; if a prompt names a creator, the model takes that as broad inspiration.”
Billboard reported in January 2024 that Google had previously trained its AI music models on copyrighted recordings before approaching rights holders for licensing deals.
Universal Music Group and Alphabet unit YouTube struck a licensing deal last year that included what UMG CEO Sir Lucian Grainge called “really important guardrails and protection for our artists and writers around gen AI content.”
Lyria 3 follows Lyria 2, which MBW reported on in September, and which powers YouTube’s “Speech to Song” tool for converting dialogue into musical soundtracks.
Google says in its latest blog post that all Lyria 3 outputs are embedded with SynthID, Google’s watermark for AI-generated content. The company also says outputs are filtered against existing content before delivery.
In developing Lyria, Google said it has brought in working musicians like Grammy-winning producer Yung Spielburg, who used Lyria in composing the score for a Google DeepMind short film called Dear Upstairs Neighbors. DJ and producer François K has been collaborating with the team on a song he described as part of an iterative process.
“Having explored the vast landscape of music tools, the progress on Lyria 3 is incredible, especially the fidelity and musicality.”
François K, DJ and producer
François K said: “Having explored the vast landscape of music tools, the progress on Lyria 3 is incredible, especially the fidelity and musicality. The fashion in which I use generative AI tools never boils down to ‘one-button-click’ prompting. Instead it’s becoming a versatile part of my arsenal, allowing me to refine ideas with realism and precision.”
Google also runs Music AI Sandbox, a program giving musicians, producers and songwriters access to experimental tools in exchange for feedback to help develop Lyria 3.
Music Business Worldwide
