Google Releases Music Generation AI Lyria 3.5, Supporting Tracks Up to 3 Minutes

The new model supports music generation of up to 3 minutes. It can create consistent audio tailored to the required length of a project, ranging from short 60-second clips to full-sized tracks.

Additionally, it features a capability to convert an input image into a custom, high-resolution track based on the visual content.

Detailed control via prompts is possible. By specifying vocal styles, acoustic preferences, BPM, and other parameters, users can generate music tailored to specific genres or moods.

Lyria was developed incorporating feedback from producers and musicians and is designed to understand musicality, from rhythm to arrangement.