When users provide a mood, genre, memory, or a personal joke as input, Gemini will produce a complete track with lyrics, instrumental arrangements, and cover design.
In the system where lyrics are automatically generated by artificial intelligence, the Nano Banana model will be used for visual designs.
This feature, accessible to users aged 18 and over, will be offered with higher usage limits for Google AI Plus, Pro, and Ultra subscribers. Designed as a fun tool for musical experiments and sharing with friends rather than creating a commercial hit, the system will deliver results in seconds.
Each generated track will include a "SynthID" digital watermark, imperceptible to the human ear, which identifies the content as AI-generated.
Gemini is designed to produce entirely original material rather than copying existing works. When real artists are referenced in commands, the AI will only draw general stylistic inspiration without copying copyrighted music.
Google will implement content filters to ensure that outputs do not infringe upon existing works. Users will be able to report content they believe infringes copyright.
The feature will support multiple languages, including English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese.
