Google has rolled out examples of its latest AI, which can take a text prompt and turn it into a musical composition – or at least something closely resembling one.
MusicLM can generate high-fidelity music from text descriptions, such as “a calming violin melody backed by a distorted guitar riff”. Google claims it “outperforms previous systems both in audio quality and adherence to the text description.”
In addition, it can “transform whistled and hummed melodies according to the style described in a text caption,” meaning you might even get a co-writing credit.
Google isn’t letting you fiddle around with the AI just yet, but that have provided a series of 30 second audio snippets, and text used to generate them. For example”
“A fusion of reggaeton and electronic dance music, with a spacey, otherworldly sound. Induces the experience of being lost in space, and the music would be designed to evoke a sense of wonder and awe, while being danceable.”
The sample (which you can listen to here, along with others) sounds loosely like what is requested.
While this is impressive, AI-generated music has existed since the 1990s.
In 1997, an album named Classical Music Composed By Computer was released by a machine named Experiments in Musical Intelligence, or EMI.
That same year, EMI composed a full-scale Mozart symphony and piano concerto, which was performed at the Santa Cruz Baroque Festival.
Linda Burman-Hall, who played the piano solo, told the New York Times: ”It felt a little different than playing a normal Mozart work. But it was very much like a work of the same period. It was certainly in the ball park.”
So Google isn’t exactly swimming in new waters here, but it is likely to provide the internet with another fun AI-based distraction for a week or so.
