Back in February, Google acquired a music generator startup called Producer. In April, Google rebranded Producer, turning it into Google Flow Music. That product has been available as part of the Google Labs suite. But, as is usually the case with such experiments, Google Flow Music hasn’t really set the world on fire with its popularity—even at a time when music generators like Suno have, for better or worse, made headlines. But Google’s new music model, Lyria 3.5, has just been made available through Google Flow Music. In a blog post on Wednesday, Google wrote that the Lyria 3.5 boasts, “richer, more complex melodic structures that sound more natural,” as well as lyrics that are supposed to adhere better to user prompts, and have better attentiveness to song structure. The vocals are supposed to be more expressive and “emotionally nuanced.” Users also get more control over the absolute bedrock basics of song creation: tempo and song length. As if to prove its point, the blog post comes with an embed of a YouTube video featuring an unnamed song in the genre I call Taco Bell pop. Y’know, the kind of not-quite-good-enough-to-be-on-the-radio pop you hear in Taco Bell, and see affiliated with Taco Bell, and which makes you feel like you’ve just consumed Taco Bell for your ears? Everyone calls it that, right? The song is pleasant and the lyrics are about spinning around:
Help! I Generated a Song With Google's New Music Model and I Kind of Like It
It reminds me of what AI products were like in 2022.
Google deployed Lyria 3.5 with improved melodic complexity and vocal control via Flow Music. Music generators transitioned from experimental tools to commodity content at scale—governance, copyright enforcement, and platform risk now critical for tech stack decisions.










