Google DeepMind has released Lyria 3.5, its most expressive AI music generation model yet, featuring meaningfully better vocals, sharper lyric coherence, and direct tempo control. The update arrived Wednesday in Google Flow Music at no additional cost for all users, marking the third significant Lyria milestone in five months. Meanwhile, its two main rivals, Suno and Udio, remain mired in copyright litigation that Google has deliberately structured itself to avoid through licensed training data.

What Makes Lyria 3.5 Different From Earlier AI Music Models?

The update addresses four distinct technical areas that have historically limited AI music generation. On musicality, the model now generates richer and more complex melodic structures that sound more natural across the length of a track, directly tackling a persistent criticism of earlier releases: melodic loops that felt mechanical or repetitive after the first 30 seconds.

On lyrics, Lyria 3.5 produces higher-quality text with improved prompt adherence and structural awareness. This means the model is more likely to write lyrics that actually fit the mood, narrative, and genre you described, rather than producing grammatically correct but thematically hollow text. The vocal improvements bring more realistic and emotionally nuanced performances, along with improved pronunciation, which matters particularly for creators working with non-English lyrics or genre-specific vocal styles.

Independent comparisons placed Lyria 3 Pro "comparable to Suno v4.5 in quality" on overall audio fidelity, not yet at Suno v5's level, but closing the gap. The Lyria 3.5 release specifically targets the vocal territory where Udio has historically been strongest.

How Does Google's Licensing Strategy Give It a Legal Advantage?

The clearest differentiator Google emphasizes is its training data provenance. Lyria 3 Pro and Lyria 3.5 are trained on licensed data using materials that YouTube and Google have the right to use under their terms of service, partner agreements, and applicable law. Google confirmed the model does not replicate any specific artist's voice or catalog.

This distinction matters enormously in 2026. The Recording Industry Association of America filed landmark copyright lawsuits against both Suno and Udio in June 2024 on behalf of Sony Music, Universal Music Group, and Warner Music Group. Universal Music Group settled with Udio in October 2025, and Warner Music Group settled with Suno in November 2025. However, Sony Music Entertainment has settled with neither company, and its cases against both remain fully active.

By contrast, Google's licensed-data approach means Lyria operates within a clear legal framework that protects both the company and its users from the copyright exposure that has defined the competitive landscape for its rivals.

How to Understand the Technical Improvements in Lyria 3.5

• Audio Quality: The model outputs 48 kilohertz, 16-bit PCM stereo audio, which is professional production quality. The underlying stack combines Lyria for audio generation, Gemini for natural language understanding that interprets creative prompts, and Veo for visual components in the broader Flow ecosystem.
• Structural Coherence: The improvements that debuted in Lyria 3 Pro come from training the model to understand songs as structured compositions with intros, verses, choruses, and bridges, rather than as continuous audio generated token by token without knowledge of where it stands in a song's arc.
• Creative Control: Users can now set tempo and duration with more precision, a practical upgrade for producers who have found earlier Lyria outputs useful as sketch material but too unpredictable for structured compositions.

Google acquired ProducerAI in February 2026, a startup built on a different architecture using the original Riffusion model, which fine-tuned Stable Diffusion on spectrograms by treating music as an image generation problem. Google replaced that architecture entirely with Lyria 3 at acquisition.

Why Does Google's Watermarking Strategy Matter for the Industry?

Every track generated by Lyria carries a SynthID watermark, an imperceptible cryptographic signal embedded directly into the audio waveform at generation time. This watermark exploits frequency ranges below human hearing sensitivity and the masking effect during louder segments, creating a digital signature that is part of the sound itself, not a removable metadata tag. It survives MP3 compression, including aggressive 128 kilobits per second encoding, speed and pitch shifts, and even re-recording through a speaker.

In May 2026, OpenAI adopted SynthID watermarking for images generated by ChatGPT, and NVIDIA integrated it into its Cosmos world models. SynthID is becoming the default AI content provenance standard across the industry, which means Lyria-generated music sits inside the only verifiable chain of custody that streaming platforms, rights management organizations, and regulators can use for automated detection and policy enforcement. Deezer's 99.8% AI-detection system, Spotify's AI Credits disclosure framework, and TIDAL's July 15 AI-labeling policy all depend on being able to identify AI-generated audio.

What Is Google's Timeline for Lyria Releases?

The pace of Google's updates signals that the company is treating AI music not as a research experiment but as a priority product line. Lyria 3 launched February 18, 2026, generating 30-second high-fidelity tracks from text or image prompts through the Gemini app for users 18 and older at 48 kilohertz with lyrics, vocals, and custom cover art. Lyria 3 Pro followed March 25, 2026, extending the maximum track length to three minutes and introducing the structural song awareness that distinguishes proper full-length compositions from extended loops.

Google rolled Lyria 3 Pro out across six platforms simultaneously: the Gemini app, Google Vids, ProducerAI (now Flow Music), Vertex AI, the Gemini API, and AI Studio. Lyria 3.5 now builds on the 3 Pro foundation, focusing on expressive quality rather than duration or distribution breadth.

Google Flow Music, the platform where Lyria 3.5 lives, is the renamed and rebuilt version of ProducerAI, itself formerly known as Riffusion. The platform lets users generate songs from text, image, audio, and voice prompts; remix existing tracks; edit compositions at the section level; rewrite lyrics in place; publish creations; and collaborate through shareable workspaces. In May 2026, Google partnered with Believe, the global artist development company that distributes for TuneCore, to offer Flow Music directly to Believe's artists and songwriters, marking a deliberate shift toward positioning Flow Music as infrastructure for professional and semi-professional music creation.

Flow Music also integrates with Google Vids, Google AI Studio, Vertex AI, and the Gemini app, a distribution footprint neither Suno nor Udio can match. As the copyright landscape continues to shift, Google's combination of licensed training data, industry-standard watermarking, and broad platform integration positions it as the only major AI music player operating without active litigation from major record labels.