Google expands Lyria 3 Pro: AI composer now creates tracks up to three minutes

Google updated its music neural network and presented Lyria 3 Pro — a significantly more powerful version that generates tracks up to three minutes long. Previously, Lyria was strictly limited to 30-second clips: this was enough for demonstrating capabilities, but clearly insufficient for real-world use in content production. The difference is fundamental: 30 seconds is a snippet, three minutes is a full composition.

A sixfold increase in length moves the tool from the demo category to the working tool category for content creators. The main change is not just length, but structural control. Lyria 3 Pro allows users to explicitly specify what elements should make up the track: intro, verse, chorus, bridge.

This fundamentally changes the logic of work. Instead of just describing the mood and waiting for the result, you can build a complete musical structure. For those who know anything about music production, this is a significant improvement compared to competitors, who usually give control over arrangement to the model itself without the ability to explicitly influence it.

Otherwise, the working principle is standard for the market: you describe the genre, mood, instrumentation — the neural network creates the track. The same logic applies to Suno and Udio, which were the first to widely popularize generative music. Lyria 3 Pro can also write song lyrics based on a prompt.

But there is a non-standard feature — creating music based on a reference image. You upload a photo or cover, and the model generates an appropriate soundtrack. This is an atypical approach, but potentially useful for videographers, marketers, and designers who need music for a specific visual idea.

A key difference between Lyria and competitors is built-in integration into Google's ecosystem. Lyria 3 Pro will appear in several Google products, although the specific list is not yet disclosed. Independent services like Suno or Udio have to fight for distribution on their own, invest resources in marketing, and convince users to visit a new platform.

Google's position is different: it can simply add a "create music" button to YouTube, Google Slides, or any other service with an audience of hundreds of millions. This is the main advantage — not just model quality, but the scale of distribution. Lyria 3 Pro is part of Google's consistent strategy to cover all generative media formats.

The company already has Veo for video and Imagen for images. Musical AI was perhaps the only major media direction where Google was significantly lagging: while Suno and Udio attracted tens of millions of users, Google remained on the sidelines. The launch of Lyria 3 Pro with expanded capabilities looks like a serious entry into this segment, not just a technical update.

Three minutes is already a real working format. Background music for a YouTube video, an advertising jingle, a podcast intro, a soundtrack for a presentation or corporate video — all of this fits within the three-minute limit. For content creators who currently either pay for licensed music or use limited free libraries, a built-in generator right in a familiar service could be a real solution.

Especially if Google integrates Lyria into YouTube Studio or similar tools for creators. The generative music market is still small but growing rapidly — and Google is entering it with an obvious structural advantage. Suno and Udio showed that there is significant demand for such tools.

Now the world's largest media platform with built-in distribution is joining them. If Lyria 3 Pro turns out to be of sufficient quality, it risks becoming the default tool for millions of users — not because it's better than competitors, but simply because it will be there where people already work every day.