Google adds music-generation capabilities to the Gemini app

Google has added music-generation capabilities to its Gemini app using the Lyria 3 model, allowing users to create original tracks and lyrics based on descriptions. The feature is in beta and supports multimedia inputs. Lyria 3 will also be available globally on YouTube's Dream Track, fostering original expression while ensuring AI content is identifiable. The rollout reflects broader trends in AI music amidst copyright concerns in the industry.

Key Points

• Google announced a music-generation feature in the Gemini app utilizing DeepMind's Lyria 3 model.
• Users can describe desired songs, and the app generates both music tracks and lyrics.
• The app can create a song in response to multimedia inputs, ensuring the output matches the media's mood.
• Lyria 3 enhances realism and complexity compared to previous models, allowing user control over various musical elements.
• The Dream Track feature broadens access for YouTube creators globally, not just in the U.S.
• Tracks created with Lyria 3 will carry a SynthID watermark to identify them as AI-generated.
• Google emphasizes the generation of original content rather than mimicking specific artists, with filters to prevent copyright issues.
• AI-generated music faces a mixed reception, as companies adopt it while artists express concerns over copyright violations.

Relevance

• The rise of AI in creative fields, particularly music, is a significant trend in 2025, with both opportunities and challenges.
• Issues surrounding copyright and originality in AI-generated art echo ongoing legal battles within the music industry.
• Major platforms like YouTube and Spotify are evolving their strategies to incorporate AI-generated content, reflecting industry shifts.

The introduction of music-generation features in the Gemini app signals a pivotal moment in AI's role in creativity, balancing innovation with the importance of copyright protection and original artistry.