This week, Google researchers published a paper detailing the results of AI (AI) tool designed to create music.
The tool, called MusicLM, isn’t the first AI music tool to launch. But the examples Google provides demonstrate musical creativity based on a limited set of descriptive words.
AI shows how complex computer systems have been trained to behave in human-like ways.
Tools like ChatGPT can quickly produce or generate written documents that compare well with human work. ChatGPT and similar systems require powerful computers to handle complex machine learning models. San Francisco-based company OpenAI launched ChatGPT late last year.
Developers train such systems on massive amounts of data to learn methods for recreating different forms of content. For example, computer-generated content may include written materials, design elements, art, or music.
ChatGPT has recently received a lot of attention for its ability to generate complex texts and other content from just a simple natural language description.
Google’s MusicLM
Google engineers explain the MusicLM system as follows:
First, the user comes up with a word or words that describe the kind of music they want the instrument to create.
For example, a user might enter this short phrase into the system: “a long soothing violin backed by a soft guitar sound.” The description entered may include different music stylesinstruments or other existing sounds.
Several different musical examples produced by MusicLM have been posted online. Some of the music generated comes from just one or two word descriptions, such as “jazz”, “rock” or “techno”. The system produced other examples of more detailed descriptions containing full sentences.
In one example, Google researchers included these instructions to MusicLM: “The essential soundtrack of an arcade game. He is hectic and optimistic, with a catchy electric guitar riff. It’s the music repetitive and easy to remember but with unexpected sounds…”
In the resulting recording, the music appears to be very close to the description. The team said the more detailed the description, the better the system can try to create it.
The MusicLM model works similarly to the machine learning systems used by ChatGPT. Such tools can produce human-like results because they are trained on massive amounts of data. Many different materials are fed into the systems to allow them to learn complex skills to create realistic works.
In addition to generating new music from written descriptions, the team said the system can also create examples based on a person’s own singing. hummingwhistling or playing an instrument.
The researchers said the instrument “produces high-quality music … for several minutes while it is faithful to the text conditioning signal.’
As of now, the Google team has not released the MusicLM models for public use. This differs from ChatGPT, which was available online for users to experiment with in November.
However, Google has announced that it is releasing a “high-quality dataset” of more than 5,500 music writing pairs prepared by professional musicians called MusicCaps. The researchers took this step to help develop other AI music generators.
MusicLM researchers said they believe they have designed a new tool to help anyone quickly and easily create high-quality music selections. However, the team said it also recognizes some risks associated with the machine learning process.
One of the biggest problems identified by the researchers is “bias present in the training data.” Addiction can involve too much of one side and not enough of the other. The researchers said this raises the question “of relevance to generate music for cultures underrepresented in the training data.”
The team said it plans to continue studying any systemic results that might be taken into account cultural appropriation. The goal would be to limit bias through more development and testing.
In addition, the researchers said they plan to continue improving the system to include text generation, text conditioning, and better voice and music quality.
I’m Brian Lynn.
Brian Lin wrote this story for VOA Learning English based on reports from Google.
_____________________________________________________________
Words in this story
AI – n. the development of computer systems that have the ability to perform work that would normally require human intelligence
style -n. a particular shape or design, usually used when comparing forms of art or handicraft
instruction -n. a description of how to do something
arcade – n. area containing many electronic and other coin games
optimistic – adj. full of hope and happiness
repetitive – adj. saying or doing something over and over again
humming – v. to make a musical sound without opening your mouth
whistle – v. to produce a high-pitched sound by forcing air through a small hole in the mouth
faithful – adj. standing firm on an idea or belief
appropriate – adj. the degree to which something is appropriate for a given situation
cultural appropriation – n. when members of one culture in a society, often the mainstream culture, use a practice of another, often a minority culture, without fully understanding the meaning or importance of the practice.
_________________________________________________________________
What do you think of this story? We want to hear from you. We have a new comment system. Here’s how it works:
- Write your comment in the field.
- Below the box you can see four images for social media accounts. They are for Disqus, Facebook, Twitter and Google.
- Click on an image and a box appears. Enter the login details for your social media account. Or you can create one in the Disqus system. It’s the blue circle with a “D” on it. It’s free.
Every time you come back to comment on the English learning site, you can use your account and see your comments and replies to them. Our comment policy is here.