YouTube, one of the leading platforms for video content and owned by Google, has announced significant advances in the integration of generative artificial intelligence to revolutionize music creation. These are the tool, Dream Track based on DeepMind’s Lyria, and Music AI Tools.

In addition to these two innovations, YouTube is also committed to conducting additional testing with songwriters, producers and artists before the official launch. This focus on tiered testing highlights the importance of responsible development of AI in music.

Dream Track is presented as the gateway to the creation of unique songs, imitating the style of famous artists. YouTube has taken proactive steps to address concerns about the ethical use of AI in music, collaborating directly with artists such as T-Pain, Demi Lovato and Sia. This tool uses Google DeepMind’s Lyria model to generate tracks of up to 30 seconds. This collaborative approach between creators and technology seeks to offer users an authentic and unique musical experience.

Thanks to this new tool, a user can generate 30-second audio clips in the style of 9 singers: Alec Benjamin, Charlie Puth, Charli XCX, John Legend, Papoose, T-Pain, Troye Sivan, and the aforementioned Lovato and Sia. To do this, it is enough to give a brief indication, a prompt, to Dream Track, explaining the genre of music, the theme and the style, as explained by the same YouTube platform through an introductory video.

In collaboration with Google DeepMind, YouTube has introduced another new tool called Music AI Tools, as part of its initiative to explore the possibilities of AI in music creation. This revolutionary suite of utilities allows users to generate instrumental tracks and vocal accompaniments from scratch, or even transform existing music from one style to another.

Its versatility stands out, since not only can you describe the musical idea you have in mind, but it also offers the option of humming a melody, allowing the AI ??to incorporate it into the composition. This approach opens new doors for musical creativity, offering artists and creators a unique tool to experiment and customize their compositions in ways never seen before. Importantly, although Music AI Tools is currently in an experimental phase, it promises to unleash a wave of creative opportunities for music lovers and content creators on the YouTube platform.

Given the controversy generated by the use of generative AI to “clone” style or draw strong inspiration from artists’ creations, Google has announced the implementation of SynthID. This is a special watermark that adds a layer of transparency and authenticity to the content generated by these new technologies.

As DeepMind explains on its blog, SynthID “converts the audio waveform into a two-dimensional visualization that shows how the frequency spectrum of a sound evolves over time.” And it’s something that “is unlike anything that exists today, especially in the context of audio.”

This “watermark is designed to maintain detectability even when the audio content undergoes many common modifications,” they add, such as “additions of noise, MP3 compression, or acceleration and deceleration of the track.” In addition, this new tool in full development also has the ability to detect “the presence of a watermark in a track to help determine if Lyria generated parts of a song.”

Lyria, developed by Google DeepMind, calls itself “the world’s most advanced generative AI model.” It addresses the complexity of replicating instruments and vocals in longer sequences, allowing users to create songs from scratch from a single phrase or instruction. Furthermore, its ability to combine instruments and voices in longer sequences stands out, marking a milestone in the evolution of artificial intelligence applied to music.

With these tools, YouTube places itself at the forefront of musical creativity, harnessing the potential of AI to empower creators and artists. This breakthrough not only redefines music creation but also highlights the need to ethically and responsibly address the impact of artificial intelligence on the world of entertainment.