AiPhreaks ← Back to News Feed

We’re launching Lyria 3.5 in Google Flow Music, with advances across musicality, lyrics, vocals, and creative control

By Jakub Antkiewicz

2026-07-30T10:24:03Z

Google announced today the release of Lyria 3.5, its newest music generation model, which is being integrated directly into the Google Flow Music application starting July 29, 2026. The update targets core aspects of AI music creation, aiming to provide users with more refined and controllable tools. This move signals a strategic focus on improving the practical usability and output quality of generative audio tools, moving beyond novelty to address specific creator needs.

Core Updates in Lyria 3.5

The Lyria 3.5 model introduces several specific enhancements designed to elevate the quality of generated tracks. According to the company, these improvements are centered on making the output sound more natural and providing creators with more direct influence over the final song. The key advancements include:

  • Improved Musicality: The model can now generate more complex and natural-sounding melodic structures.
  • Enhanced Lyrics: Generation capabilities have been updated for better prompt adherence and structural coherence in lyrics.
  • Improved Vocals: Vocal tracks are designed to be more emotionally nuanced and expressive, with better pronunciation.
  • Creative Control: Users now have more direct control over the tempo and total duration of the musical output.

By embedding these upgrades within Google Flow Music, Google is aiming to make its platform more competitive in a growing market of AI-powered creative tools. The emphasis on specific controls like tempo and duration suggests the industry is evolving to serve more sophisticated users who require tools for co-creation rather than simple generation. This focus on polish and control may set a new baseline for what prosumers and hobbyists expect from music generation platforms.

Google's focus on granular controls like tempo and duration with Lyria 3.5 indicates the market for generative music AI is maturing from a novelty into a practical toolset, where usability and refinement are becoming key differentiators.
End of Transmission
Scan All Nodes Access Archive