Google's Lyria 3.5 Brings Smarter Lyrics, Improved Vocals And More Creative Controls

It looks like Google will move further with its artificial intelligence in music development through the new Lyria 3.5. The software is intended to help create tunes that look closer to what humans have developed rather than being developed by artificial intelligence.

All the aspects of creating music have been improved through the update – including melodies, vocals, and lyric writing. Besides, users will now have more options when it comes to editing their tunes without having to start the process again.

ALSO READ: Samsung Open To More Affordable Foldables As Category Matures, Says Samsung Southwest Asia CEO & President JB Park

Songs Should Sound More Natural

The most significant improvement in Lyria 3.5 comes in terms of the actual quality of the generated music. According to Google, with the updated Lyria 3.5, users will be able to generate even more intricate melodies, and the lyrics of songs will become more prompt-sensitive, and they will retain consistency throughout the whole composition.

The vocals also underwent some improvements, as the company reports that the generated music will feature much more expressive singing and clear pronunciation.

Finally, with Lyria 3.5, users will get the opportunity to modify the tempo and length of the songs right in the Flow Music interface.

Flow Music Is Becoming a Complete Music Studio

Google isn't treating Flow Music as just another text-to-song generator anymore. Over the past few months, the platform has steadily picked up editing tools that let users tweak individual parts of a song instead of generating an entirely new one every time.

Users can already rewrite or translate lyrics, swap instruments and make changes to specific sections of a track. At Google I/O 2026, the company went a step further by introducing conversational editing for music videos, along with more precise controls for editing individual elements within a composition.

Latest Upgrade to Google's AI Music Platform

Lyria was first introduced by Google DeepMind in 2023 as the company's AI music generation model. Earlier this year, Google expanded its capabilities with Lyria 3, allowing users to create songs from text prompts, images and even videos. The model could also generate vocals, lyrics and cover art, while offering controls for genre, mood and tempo.

Google later followed that up with Lyria 3 Pro, which extended the maximum song length to three minutes and introduced additional creative tools for paid Gemini subscribers. The technology has since found its way into YouTube Shorts through Dream Track and Google Vids, where it's used to generate AI-powered background music.

ALSO READ: Samsung Galaxy Z Fold8 Series Crosses 2.71 Lakh Pre-Orders In 72 Hours, Tier 2 Cities Drive Demand

For now, Google hasn't said whether Lyria 3.5 will eventually replace Lyria 3 Pro across Gemini or its developer platforms. The company has, however, confirmed that access to Flow Music will continue to depend on a user's Google AI Plus, AI Pro or AI Ultra subscription tier.