Google adds music generation capabilities to the Gemini app

In a groundbreaking announcement on Wednesday, Google revealed that its Gemini app will soon include a music generation capability powered by DeepMind's Lyria 3 model. Currently in beta, this feature allows users to craft unique songs by simply describing their desired creation. For example, if a user requests a "comical R&B slow jam about a sock finding its match," Gemini will generate a 30-second track along with custom cover art designed by Nano Banana. What makes this feature even more intriguing is its ability to adapt to user-generated media. Participants can upload photos or videos, and the AI tool will create a song that complements the mood of the content. Google boasts that Lyria 3 significantly enhances the quality of music production, offering more realistic and intricate soundscapes compared to its predecessors. Users will also have the flexibility to modify various aspects of the music, including style, vocals, and tempo. In addition to integrating Lyria 3 into Gemini, Google is expanding its reach by making it available to YouTube creators through the newly enhanced Dream Track feature, which facilitates the creation of AI-generated tracks. Previously limited to U.S. users, Dream Track is now accessible globally. While the tool allows users to draw inspiration from specific artists, Google emphasizes that it is meant for original expression rather than imitation. Should users reference an artist, Gemini will interpret this as a creative prompt and generate music that embodies a similar vibe without directly replicating the artist's style. The company has implemented safeguards to ensure that outputs do not infringe on existing content rights. To further ensure transparency, all tracks produced with the Lyria 3 model will feature a SynthID watermark, signifying their AI-generated nature. Additionally, Gemini will provide users with the ability to verify whether a submitted track is AI-generated. The rollout of this music generation feature is targeted at all users aged 18 and over, supporting multiple languages including English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese. As AI-generated music gains traction, it has sparked a mixed response from both artists and audiences. While platforms like YouTube and Spotify are embracing AI and forming partnerships with music labels to monetize this new genre, the music industry has raised concerns, leading to legal actions against AI model companies regarding copyright issues. In response, platforms such as Deezer are developing tools to clearly label AI-generated music, aiming to mitigate fraudulent streaming practices.