Google drops Lyria 3.5 music model inside Flow Music
Google just released Lyria 3.5, its upgraded music generation model rolling out inside Flow Music. The update packs richer melodies, more expressive vocals with cleaner pronunciation, and sharper lyric formatting that actually follows prompts. It also adds direct control over track tempo and song length, giving creators significantly more say over the final track.
Anyone who has tried generating AI music knows the drill: either the vocals sound like a flat synthesizer, or the track ends abruptly without matching the rhythm you actually needed.
Google has now launched Lyria 3.5, rolling it directly into Google Flow Music with a set of practical upgrades aimed squarely at creators.
Expressive vocals and tighter structure
Lyria 3.5 focuses on four specific areas. The biggest shift is in the vocals, which Google says now feature more realistic emotion, nuanced delivery, and cleaner pronunciation.
Lyrics and melodies also got an overhaul. The model follows prompts more closely and shows better awareness of song structure, resulting in richer and more complex melodic compositions.
Crucially for editing workflows, Google added better creative control. Users can now more easily tweak the tempo and set the duration of generated tracks right from the prompt.
Why it matters
Most generative audio tools have suffered from the slot-machine problem: you pull the lever and hope the output fits your project. By letting users dictate tempo and length, Google is pushing generative music from a party trick into a usable production tool. We think this is a smart pivot that prioritizes creator utility over raw novelty, making it instantly more appealing for video editors and podcasters who just need a track that fits a specific timeline.
Sources (1)
- blog.google(opens in a new tab)Originalblog.google