Gemini can now create AI music

Google is launching music generation in its Gemini app via the new Lyria 3 model developed by Google Deepmind. The feature is in beta and allows you to create 30-second pieces of music based on text prompts and/or uploaded images and videos, says the announcement post.

For example, you can describe a genre, mood, or even a memory, and Gemini will generate a song and lyrics according to your description. It’s also possible to control the style, tempo, and type of vocals used in the generation, giving you more creative control.

All songs are marked with SynthID, Google’s invisible watermark for AI-generated content. Gemini can also check whether audio files have been created with Google’s AI. Google emphasizes that the model is intended for original creation and not for imitating specific artists.

Lyria 3 is available globally in English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese. Users must be over 18 years of age. Paying Gemini subscribers get higher usage limits.