Google Launches AI Music Generator Lyria 3.5 in Gemini App

Google has officially launched Lyria 3.5, its next-generation AI model for music and singing generation, now integrated into the Gemini app for mobile and web users worldwide. Developed by DeepMind, this advanced model aims to democratize music creation, allowing individuals without prior musical knowledge to produce complete songs.

Lyria 3.5 boasts significant improvements in vocal quality and sound processing, generating more dynamic singing voices and multi-channel audio with natural flow, free from metallic artifacts. The model offers enhanced understanding of complex creative prompts, precise control over tempo (beats per minute), customizable song structures, and the ability to export entire musical pieces. Within the Gemini app, the user experience is designed for simplicity, enabling users to select genres, choose between instrumental or vocal tracks, and utilize pre-made templates.

The system facilitates the easy creation of original songs for birthdays, background music for videos, business jingles, or custom ringtones, with options for both short clips and longer, complex compositions. Google's Gemini app also features templates for specific uses like birthday songs and jingles, with controls for tempo and style.

Beyond the Gemini app, Lyria 3.5 powers the Flow Music platform for creators and artists, enabling real-time music creation in collaborative workspaces. It is also available to developers via Google AI Studio and is integrated into the Google Vids app, allowing content creators to add original soundtracks to videos and presentations effortlessly. The Lyria 3.5 model is available globally and in Israel, free of charge with premium options, marking a new phase in Google's competition with independent AI music generators.