The advent of AI technology is already disrupting the business world. Especially, the rise of AI-based applications that can auto generate images, text, and whatnot. Amid the rising popularity of AI and controversies surrounding the increasing use cases, Google has announced the launch of MusicLM, a bot that can generate music from text prompts.
According to a research paper published by Google, “MusicLM is a model that generates high-fidelity music using text prompts such as a melodious violin tune along with distorted guitar riff.”
The Google bot is capable of generating music based on text prompts such as “enchanting jazz song with a memorable saxophone solo and a solo singer” or “Berlin 90s techno with a low bass and strong kick.” Google has even shared audio examples of the corresponding results on its Github page.
Also Read: A Surge In Tech Layoffs 2022 As Industry Struggles To Recover
However, the public shouldn’t get their hopes up for a chance to use MusicLM just yet. Google has stated that it has no plans to release the bot to the public due to various risks including programming biases, technological glitches and the potential for copyright infringement. The study found that identifiable existing songs were present in 1% of examples.
The launch of MusicLM comes amid the rising popularity of OpenAI’s chatbot ChatGPT and Google’s response to the competition, with the company ramping up the release of 20 new products and a version of Google Search with AI chatbot features.
Also Read: Samsung Galaxy S23: Leaked Pricing Confirms Cost Ahead of Unveiling
Despite its limitations, such as the use of negations and temporal ordering in text prompts and vocal quality, researchers plan to work towards “modeling of high-level song structure like introduction, verse, and chorus.” The study highlights the need for further work to address the risks associated with music generation.
Now it will be interesting to see when Google rolls out this amazing bot and how effectively it works.
