This week, Google researchers published a paper describing results from an AI tool built to create music. The tool, called MusicLM, is not the first AI music tool to be launched. But the examples Google provided demonstrate musical creative ability based on a limited set of descriptive words.
Several different music examples produced by MusicLM were published online. Some of the generated music came from just one- or two-word descriptions, such as "jazz", "rock" or "techno（高科技舞曲）". The tool also created other examples from more detailed descriptions containing whole sentences.
In one example, Google researchers included these instructions to MusicLM, "It is a soundtrack of a game. It is fast-paced and upbeat, with a catchy electric guitar riff（反复乐节）. The music is repetitive and easy to remember, but with unexpected sounds ..." In the resulting recording, the music seemed to keep very close to the description. The team said the more detailed the description was, the better the system could attempt to produce it.
The MusicLM model operates similarly to the machine-learning systems used by ChatGPT. ChatGPT and MusicLM both require powerful computers to operate complex machine-learning models. The San Francisco-based company OpenAI launched ChatGPT late last year. ChatGPT has recently received a lot of attention because of its ability to generate complex writings and other content from just a simple description in natural languages. Such tools can produce human-like results because they are trained on huge amounts of data. Many different materials are fed into the systems to permit them to learn complex skills to create realistic works.
However, the MusicLM researchers said there were some risks linked to the new tool. One of the biggest issues the researchers identified was biases（偏好）present in the training data. A bias might be including too much of one side and not enough of the other. The researchers said this raised a question about appropriateness for music generation for cultures underrepresented in the training data.
