As GenAI tools begin to rework the music industry in incredible — and in some cases ethically problematic — ways, Google is ramping up its investments in AI tech to create latest songs and lyrics.
The search giant today unveiled MusicFX, an upgrade to MusicLM, the music-generating tool Google released last 12 months. MusicFX can create ditties as much as 70 seconds in length and music loops, delivering what Google claims is “higher-quality” and “faster” music generation.
MusicFX is on the market in Google’s AI Test Kitchen, an app that lets users check out experimental AI-powered systems from the corporate’s labs. Technically, MusicFX launched for select users in December — but now it’s generally available.
Image Credits: Google
And it’s not terrible, I need to say.
Like its predecessor, MusicFX lets users enter a text prompt (“two nylon string guitars playing in flamenco style”) to explain the song they need to create. The tool generates two 30-second versions by default, with options to elongate the tracks (to 50 or 70 seconds) or mechanically stitch the start and end to loop them.
A brand new addition is suggestions for alternative descriptor words in prompts. For instance, for those who type “country style,” you may see a drop-down with genres like “rockabilly style” and “bluegrass style.” For the word “catchy,” the drop-down might contain “chill” and “melodic.”
Image Credits: Google
Below the sector for the prompt, MusicFX provides a word cloud of additional recommendations for relevant descriptions, instruments and tempos to append (e.g. “avant garde,” “fast,” “exciting,” “808 drums”).
So how’s it sound? Well, in my transient testing, MusicFX’s samples were… tremendous? Truth be told, music generation tools are attending to the purpose where it’s tough for this author to differentiate between the outputs. The present state-of-the-art produces impressively clean, crisp-sounding tracks — but tracks tending toward the boring, uninspired and melodically unfocused.
Perhaps it’s the SAD attending to me, but certainly one of the prompts I went with was “a house music song with funky beats that’s danceable and uplifting, with summer rooftop vibes.” MusicFX delivered, and the tracks weren’t bad — but I can’t say that they arrive near any of the higher DJ sets I’ve heard recently.
Listen for yourself:
Anything with stringed instruments sounds worse, like an inexpensive MIDI sample — which is probably a mirrored image of MusicFX’s limited training set. Listed below are two tracks generated with the prompt “a soulful melody played on string instruments, orchestral, with a powerful melodic core”:
And for a change of pace, here’s MusicFX’s interpretation of “a weepy song on guitar, melancholic, slow tempo, on a moonlight [sic] night.” (Forgive the spelling mistake.)
There are specific things MusicFX won’t generate — and that may’t be faraway from generated tracks. To avoid running afoul of copyrights, Google filters prompts that mention specific artists or include vocals. And it’s using SynthID, an inaudible watermarking technology its DeepMind division developed, to make it clear which tracks got here from MusicFX.
I’m undecided what kind of master list Google’s using to filter out artists and song names, but I didn’t find all of it that arduous to defeat. While MusicFX declined to generate songs within the kind of SZA and The Beatles, it happily took a prompt referencing Lake Street Dive — although the tracks weren’t writing home about, I’ll say.
Lyric generation
Google released a brand new lyrics generation tool, TextFX, in AI Test Kitchen that’s intended as a kind of companion to MusicFX. Like MusicFX, TextFX has been available to a small user cohort for a while — but it surely’s now more widely available, and upgraded when it comes to “user experience and navigation,” Google says.
As Google explains within the AI Test Kitchen app, TextFX was created in collaboration with Lupe Fiasco, the rap artist and record producer. It’s powered by PaLM 2, certainly one of Googles’ text-generating AI models, and “[draws] inspiration from the lyrical and linguistic techniques [Fiasco] has developed throughout his profession.”
Image Credits: Google
This reporter expected TextFX to be a kind of automated lyrics generator. Nevertheless it’s actually not that. As an alternative, TextFX is a set of modules designed to assist within the lyrics-writing process, including a module that finds words in a category starting with a selected letter and a module that finds similarities between two unrelated things.
Image Credits: Google
TextFX takes some time to get the hang of. But I can see it becoming a useful resource for lyricists — and writers usually, frankly.
You’ll wish to closely review its outputs, though. Google warns that TextFX “may display inaccurate info, including about people,” and I indeed managed to prompt it to suggest that climate change “is a hoax perpetrated by the Chinese government to harm American businesses.” Yikes.
Image Credits: Google
Questions remain
With MusicFX and TextFX, Google’s signaling that it’s heavily invested in GenAI music tech. But I wonder if its preoccupation with maintaining with the Joneses reasonably than addressing the tough questions surrounding GenAI music will serve it well in the long run.
Increasingly, homemade tracks that use GenAI to conjure familiar sounds and vocals that could be passed off as authentic, or at the very least close enough, have been going viral. Music labels have been quick to flag AI-generated tracks to streaming partners like Spotify and SoundCloud, citing mental property concerns. They’ve generally been victorious. But there’s still an absence of clarity on whether “deepfake” music violates the copyright of artists, labels and other rights holders.
A federal judge ruled in August that AI-generated art can’t be copyrighted. Nevertheless, the U.S. Copyright Office hasn’t taken a stance yet, only recently starting to hunt public input on copyright issues as they relate to AI. Also unclear is whether or not users could find themselves on the hook for violating copyright law if they struggle to commercialize music generated within the kind of one other artist.
Google’s attempting to forge a careful path toward deploying GenAI music tools on the YouTube side of its business, which is testing AI models created by DeepMind in partnership with artists like Alec Benjamin, Charlie Puth, Charli XCX, Demi Lovato, John Legend, Sia and T-Pain. That’s greater than could be said of a number of the tech giant’s GenAI competitors, like Stability AI, which takes the position that “fair use” justifies training on content without the creator’s permission.
But with labels suing GenAI vendors over copyrighted lyrics in training data and artists registering their discontent, Google has its work cut out for it — and it’s not letting that inconvenient fact slow it down.
