Google Introduces Lyria 3 Pro For Three Minute AI Music Tracks

Google just upgraded its music generation tools with the release of Lyria 3 Pro. This new artificial intelligence model lets users create full songs that last up to three minutes, which is a massive jump from the thirty-second limit on older versions.

It is now available inside the Gemini app for paid subscribers, giving creators a simple way to produce custom audio tracks for videos, podcasts, or personal projects.

What makes the new model different

Lyria 3 Pro does much more than just make tracks longer. The system actually understands how real music is put together. You can ask it to build a song with a specific intro, distinct verses, a catchy chorus, and a bridge. It also handles different vocal styles and instruments based on simple text descriptions or even uploaded photos.

The company worked closely with music industry professionals, including Grammy-winning producers, to ensure the output sounds natural rather than robotic.

If you want to try it out, you will need a Google AI Plus, Pro, or Ultra subscription. Once you open the Gemini app, you just type in what you want to hear. You can specify the genre, tempo, and the type of instruments you prefer. The tool also lets you write your own lyrics or rely on the system to generate them for you.

Aside from Gemini, the company is adding this technology to other platforms like Google Vids and AI Studio to help developers and video creators access the tool directly.

The new version keeps the music safe and original

With any artificial intelligence tool, copyright is a big concern. Google built filters into Lyria 3 Pro to stop it from directly copying real artists. If you ask for a song that sounds exactly like a famous singer, the model will just take general inspiration for the mood instead of mimicking the voice.

Every track generated also includes a hidden watermark called SynthID. This helps people identify when a piece of audio was created by a computer.

Via TechCrunch