AI Models

Google Releases Lyria 3.5 Music Generation Model with New Editing Capabilities

Google released Lyria 3.5, its latest AI music generation model, on July 29, 2026, featuring more natural melodies, improved lyrics, and realistic vocals. The update introduces Selective Section Painting, allowing users to edit specific parts of a track without regenerating the entire song. Users can control tempo and track length from 30 seconds to 3 minutes. The model is available through Google Flow Music, but training data details remain undisclosed.

Neura News

Neura News

Neura Market Editorial

July 29, 20262 min read
Google Releases Lyria 3.5 Music Generation Model with New Editing Capabilities

Google released Lyria 3.5, the latest version of its AI music generation model, on July 29, 2026. The update brings more natural-sounding melodies, improved lyrics, and more realistic vocals with clearer pronunciation, according to the company.

Improved Sound Quality and Control

Lyria 3.5 produces melodies that sound more natural than its predecessor, Lyria 3. Google said the new model also delivers better lyrics and more realistic vocals, with clearer pronunciation. Users can now control tempo and track length more precisely than before. Generated tracks can run from 30 seconds to 3 minutes, giving creators more flexibility.

The model is available through Google Flow Music, the company's platform for AI-powered music creation.

Selective Section Painting

A new feature called "Selective Section Painting" lets users edit specific parts of a track without starting over. For example, a short melody can be turned into a full song by painting in new sections. Users can fine-tune the tempo and duration of individual elements such as vocals, drums, bass, and other instruments.

This editing capability marks a shift from earlier versions, which required regenerating entire tracks to make changes.

The #1 Newsletter in AI

Stay ahead of the AI curve

The most important updates, news, and content — delivered weekly.

No spam. Unsubscribe anytime.

Training Data Questions Remain

Google has not disclosed details about the training data used for Lyria 3.5. When asked, the company did not immediately respond to a question about the data. For Lyria 3, Google said it was trained on materials that YouTube and Google had rights to use under partner agreements and applicable law. The company did not provide further specifics about Lyria 3's training data.

The lack of transparency around training data is a recurring issue for AI music models, which have faced legal and ethical scrutiny over copyright concerns.

Background on Lyria Series

Google has been developing AI music models under the Lyria brand for some time. Lyria 3 was the previous version, though Google did not specify its release date. The new version builds on that work with improved sound quality and editing tools.

Google Flow Music serves as the distribution channel for Lyria 3.5, integrating the model into a platform for music creation.

Related on Neura Market

More from Neura News

AI Models

42 Mathematicians Urge Royal Society to Warn Government and Media About AI Existential Risk

Forty-two mathematical fellows, including Fields Medal winners Martin Hairer, Peter Scholze, and Wendelin Werner, have signed an open letter urging the Royal Society to warn the UK government and media about existential risks from advanced AI. The letter follows recent breakthroughs in which leading models solved open research problems, including a Millennium Problem. None of the signatories are affiliated with AI companies. The group warns that AI labs' estimates of existential risk above ten percent must not be dismissed as hype, and that by the time the situation becomes obvious to the public, it may be too late to act.

Sep 18·2 min read
Developer

Steve Yegge Shuts Down Gas Town After Failing to Build Anything Else With It

Steve Yegge shut down Gas Town, his ultra-vibed coding agent orchestrator, after admitting he never built anything else with it despite heavy subscription spend. Databricks reported a 60% coding spend increase after rolling out GPT-6 Astra to 3,500 engineers, OpenAI published a misalignment disclosure framework with six case reports, and Xiaomi ran MiMo-V2.6 RL training in public with live telemetry.

Sep 18·21 min read