Last Updated:
Gemini users get the new Lyria 3 AI model that lets you generate music out of prompts, images and even videos.
Lyria 3 is the improved version with more power to create.
Gemini users are getting new abilities and with the help of Lyria 3 AI generation model you can create music with prompts or by uploading a photo in seconds. This new model is part of the Gemini app, as announced by Google this week and it is currently in beta for users. We’ve got Google making AI models that can generate videos with prompts, edit slides and create presentations with simple details.
Lyria is further moving into the multimedia arena for AI models and the new version looks to give better results.
Lyria 3 AI Model For Music: What’s New
The Lyria 3 AI model improves on its earlier versions in these ways:
- Get auto-generated lyrics based on text prompts
- Better control over genre, tempo and even vocal style for the tracks
- Music that sounds closer to reality and not AI generated
Lyria 3 supports AI music generation in multiple ways:
- Text to track: Feed basic prompts with your preference and see the AI model generate the music in seconds.
- Ask Gemini: You can seek help from Gemini and creatively get the music generated
AI content generates obvious concerns of originality and not violating policies of artists. Google is using the SynthID watermarking to identify the content generated using the Lyria 3 AI model. In fact, the Gemini app can now also help you verify if any video or audio has been generated using AI tools.
Lyria 3 AI is coming to YouTube via Dream Track and is available in the US for now, with creators in other regions getting the AI model soon. The beta version in Gemini app will work for users above 18 years and gets support in German, Spanish, Hindi, French and Japanese among others.
The desktop version will be released soon and people with Google AI Plus, Pro or Ultra subscription will be offered higher limits to generate music.
California, USA
February 20, 2026, 09:12 IST
Read More
