Gemini can now generate a 30-second approximation of what real music sounds like
Summary
Google's new Lyria 3 model enhances its Gemini platform, allowing users to create 30-second music tracks or remix existing ones. This feature, also integrated into YouTube's Dream Track, offers improved control and realism in music generation.
Key Insights
Does Lyria 3 actually mimic existing artists, or does it create original music?
Lyria 3 is designed for original expression, not for mimicking existing artists. If you name a specific artist in your prompt, Gemini treats it as broad creative inspiration and creates a track that shares a similar style or mood rather than copying the artist's work. Google has implemented filters to check outputs against existing content to prevent unauthorized reproduction of copyrighted material.
How does Google identify AI-generated music, and can I verify if a song was created by Lyria 3?
All songs created with Lyria 3 are embedded with a SynthID watermark that identifies them as AI-generated content. Additionally, you can upload any audio file to Gemini and ask whether it was generated using Google AI. Gemini will check for the SynthID watermark and use its own reasoning to determine if the track is AI-generated, similar to how it verifies AI-generated images and videos.