The rapid rise of OpenAI’s artificial intelligence chatbot ChatGPT has left many wondering what else generative AI tools will replace. If a Google research paper released this week is anything to go by, songwriters will be—and possibly the music industry.
The paper describes a tool called MusicLM that “can convert whistled and hummed melodies according to the style described in the text caption.” It can also generate “high-fidelity music from text descriptions such as ‘quiet violin melodies backed by distorted guitar riffs.'”
On the paper’s website, examples show the results generated by the tool. In one example, someone sings “Bella Ciao,” an Italian folk song from the late 19th century. Then, based on the tool, the tool produces music with a variety of instruments and styles, including guitar solos, string quartets, and jazz with saxophone.
“Wow, this is bigger than ChatGPT for me. Google almost completed the music generation, I would say,” tweeted Keunwoo Choi, AI scientist at Gaudio Lab, an AI audio technology company.
“Think of MusicLM as ChatGPT for music,” tweeted entrepreneur Martin Uetz, added, “I can’t wait for this to become mainstream.”
Generative AI vs artists
Less enthusiastic may be musicians who have spent decades mastering their instruments, just as illustrators and graphic artists are outraged by AI tools that create beautiful images from text messages.
Among the AI art tools are Midjourney, Stable Diffusion, and DALL-E 2. One person recently used Midjourney to illustrate a children’s book. Impressed with the tool, he shared his experience on social media—and was surprised by the reaction from illustrators. And last year, a picture made with Midjourney won a prize at an art festival, which also angered the artists.
The problem artists have with these tools is that they train themselves in large collections of digital artwork without permission. A lawsuit recently filed in San Francisco by working artists describes Stable Diffusion and Midjourney as “collage tools that infringe upon the rights of millions of artists.”
Indeed, copyright issues prevent Google AI from releasing MusicLM to the public. But startups may prefer to release the technology into the wild.
No, Big Tech isn’t funneling resources into generative AI either.
DALL-E is offered by ChatGPT maker OpenAI. Microsoft is investing billions into OpenAI and will use the technology in various products, including the Bing search engine. That in turn has ignited fire at Google parent Alphabet, which is working on similar tools to answer the challenge.
As a tool, MusicLM is far from perfect, but it hints at its purpose. The same can be said of ChatGPT itself. As billionaire Mark Cuban recently said about AI chatbots, “imagine what GPT 10 would be like.”
Learn how to navigate and strengthen trust in your business with The Trust Factor, a weekly newsletter that examines what leaders need to succeed. Log in here.