Calvin Wankhede / Android Authority
Have you ever wished you could create music without any knowledge of musical theory or instruments whatsoever? That’s now possible, thanks to a new Google AI experiment. Dubbed MusicLM, it’s a new kind of generative AI that can create brand-new music. It’s a lot like AI image generators where you simply type in a few lines of text and get beautiful art in return. So how well does Google’s new MusicLM work and how can you try it for yourself? Here’s everything you need to know.
What is MusicLM?
MusicLM is a language model that’s capable of generating music based on a text-based description. For example, the prompt “Calming, soft music I can study to” will generate a lo-fi track. Likewise, a prompt along the lines of “Epic orchestral track that builds tension in an action-adventure movie scene” yields music with a distinct sense of urgency that would feel right at home in Terminator or Mission Impossible.
While its current release is limited, MusicLM can do much more than just turn text into music. In an academic paper, Google’s researchers demonstrated that it can generate audio tailored to images and paintings. And in the future, you might even be able to convert a recorded clip of your humming into a full-blown music track.
MusicLM can now transform text descriptions into audio. Google is also working on adding support for image and melody inputs.
Google trained MusicLM on over 200,000 hours of music spanning over 5,500 clips. These were manually categorized by human experts, helping the machine learning model distinguish jazz from techno and even specific eras (90s pop) and styles (Afro-Cuban dance music). The company has released the labeled audio dataset with captions on Kaggle, allowing other AI developers to develop their own AI tools.
No, Google has released MusicLM for free but you’ll need to join a waitlist to gain access as outlined in a later section of this article.
How well does MusicLM work?
Calvin Wankhede / Android Authority
I’ll walk you through how to sign up for MusicLM in the next section, but let me first explain what using it is like. In MusicLM’s current state, you can only type in a description and hit the generate button. Think of it like Midjourney, but instead of four AI-generated images, you get two 20-second music tracks instead.
MusicLM surfaces three suggestions below the input box, including some bizarre ones in my case like “Bubbly, optimistic, cyber pizza party music at the underwater arcade”. I hit generate on that one out of curiosity and the result was an upbeat electronic mix. Not quite what I’d expect from a “cyber pizza party”, but it was certainly bubbly and optimistic.
I found MusicLM a bit hit-or-miss in practice, but it’s still a promising first glance.
Moving on, I entered the prompt “Spooky, slow music that you’d hear in a haunted forest” and the resulting track fit the description quite well. However, it didn’t take long to see why Google considers MusicLM an experimental project.
Simply entering “rap music” produced a track with shrill, piercing sounds and a synthetic-sounding background vocal track. The latter seemed like MusicLM was reproducing something straight from its training data.
So while MusicLM doesn’t require any musical skill, it’s also not quite a musician. The results may or may not impress you, similar to chatbots before ChatGPT disrupted that landscape. Still, MusicLM represents the world’s first generative AI for music and that’s impressive in its own right. Here are some more facts about it:
- You cannot ask for music in the style of a specific artist. MusicLM also cannot generate vocals reliably at this time.
- In many instances, MusicLM generated music with unintelligible vocals. Even when I specifically asked for instrumental tracks, the AI ignored my request.
- The current iteration of MusicLM only generates 20-second tracks, even though the underlying model is capable of much longer generations.
- You get two audio clips per generation, which I found were always mostly similar to each other. Still, Google lets you vote for your favorite via a trophy button.
- According to Google, MusicLM can reproduce electronic and classical instruments better than other kinds of music. However, it did decently well when I tried to generate Indian folk music and 8-bit soundtracks in the style of old video games.
How to sign up for MusicLM
Calvin Wankhede / Android Authority
Google hasn’t released MusicLM publicly yet, opting instead to limit access via a waitlist. We saw Microsoft and Google adopt a similar strategy when they first released their chatbots in early 2023. But perhaps because MusicLM isn’t as popular, I only had to wait a few days to receive an invite. However, that can change at any time so don’t expect immediate access.
Here’s a quick guide on how to sign up for the MusicLM waitlist:
- Navigate to Google’s AI Test Kitchen homepage and click on the Get started button.
- Click on Register your interest.
- Fill out the form. It essentially asks for your country of residence, profession, the reason for using AI Test Kitchen, and whether you’ll participate in voluntary surveys.
- In the next step, sign in using your Google account.
- Once logged in, you should see a success message confirming that you’ve been added to the waitlist.
Google’s AI Test Kitchen isn’t available worldwide at the moment. So if you don’t receive an invitation email within a few weeks, that might be the reason.
As for a public launch, it may happen later in 2023 or early 2024 if Google commits to improving MusicLM. One of the biggest hurdles remains copyright infringement. The company’s researchers found that the AI would replicate existing work in around one percent of cases, which could invite lawsuits from music labels and artists.
Moreover, with virtually no major competition in the generative music space, it remains to be seen whether it remains a priority for the search giant. It’s entirely possible that Google prioritizes all of its AI resources on its Bard chatbot and the upcoming Gemini language model instead.
FAQs
It’s unclear if you can use audio clips from MusicLM commercially at this point. So in the absence of a license, it’s best to assume that you do not have rights to Google’s AI-generated music.