Are they real or AI deepfakes?

When Snowd4y, a parody rapper from Toronto, released the track “Wah Gwan Dillaah” featuring Drake via Soundcloud on Monday (June 3), it instantly went viral.

“This should be AI,” one commenter wrote of the song. It was a sentiment shared by many others, especially given the track's ridiculous lyrics and the off-kilter audio quality of Drake's vocals.

To date, the two rappers have not confirmed or denied the AI ​​rumor. Although Drake posted the track on his Instagram story, that's hardly confirmation that the sounds in question are AI-free. (As we learned during Drake's recent beef with Kendrick Lamar, the rapper isn't afraid of deep fake vocals).

To try to get to the bottom of the “Wah Guan Delilah” mystery, Billboard To review the track, we contacted two companies that specialize in AI audio detection. The answer, unfortunately, was not very satisfactory.

“Our first analysis shows something. Signs of [generative] AI, but it seems to involve a lot of mix,” wrote Romain Semiandchief product officer of Ircam Amplify, a French company that makes audio tools for rights holders, in an email response.

Larry Mills, senior VP of sales at Pex, which specializes in tracking and monetizing music consumption on the Web, also had mixed results. He told Billboard The Pex research and development team played the song [their] Voice ID Matcher” and that Drake’s voice on the “Wah Gwan Delilah” verse doesn’t quite match Drake’s…[as his voice on] Official release [does], but it's close enough to confirm that it could be Drake's own voice or a good AI copy. Notably, Pex's VoiceID tool alone isn't good enough to distinguish between real and AI voices, but the singer/rapper's voice on “Wah Gwan Delilah” and his other, officially released songs Its detection of differences between can indicate some level of AI manipulation.

A representative for Drake did not immediately respond. of the billboard Request for comment.

How to Screen AI in Song

There are a variety of tools currently used to distinguish between AI-generated music and human-produced music, but these nascent products are still in development and not definitive. Like Pax Jacob Galka In a recent company blog post about the topic, the company wrote, “AI-generated music recognition [is] A particularly difficult task.”

Some detectors, such as Ircam's, identify AI music using “artifact detection”, meaning they detect parts of a work that are far from reality. A clear example of this is seen with AI-generated images. Early AI images often contain hands with extra or misshapen fingers, and few tools exist to detect these errors.

Other detectors rely on reading watermarks embedded in AI-generated music. Although these watermarks are not perceptible to the human ear, they can be detected with certain tools. Galka writes that “since the purpose of watermarking can be discovered by watermark detection algorithms, such algorithms can also be used to show how watermarks embedded in audio be removed or modified so that it is no longer discoverable”—something he sees as a major flaw. This detection system.

Pex's approach to using Voice ID, which can determine if a singer matches between multiple recordings, could also be useful in AI detection, though it's not a clear answer. . This technology is especially helpful when users take to the internet and release random tracks with Drake vocals, whether they're leaked songs or AI deepfakes. With VoiceID, Pex can tell a rights holder that their voice was featured on another track that may not have been officially released by them.

When Voice ID is combined with the company's other product, Automatic Content Recognition (ACR), it can sometimes determine whether a song uses an AI voice, but the company states that there is not enough information on “Wah Guan Delah” to complete a full ACR. check

The role of parody in AI music

Although it cannot be determined without a doubt whether “Wah Guan Delilah” contains AI voices, parody songs in general have played an important role in popularizing and normalizing AI music. This is especially evident on TikTok, which is full of so-called “AI Covers,” pairing famous singers with unexpected songs. Popular examples of this trend include Natasha Bedingfield's song “Pocketful of Sunshine” by Kanye West, Juice Wrld's song “Viva La Vida” by Coldplay, George Michael's song “Careless Whisper” by Michael Jackson, and more. are

Most recently, AI comedy music took center stage with Metro Boomin's SoundCloud-released track “BBL Drizzy” — which sampled the AI-produced song of the same name. The track poked fun at Drake and his perceived “Brazilian butt lift” during the rapper's beef with Lamar, and in the process, became the first major use of an AI-generated sample. Later, Drake and Sexy Redd sampled the original AI-generated “BBL Drizzy” on their own song, “You My Everything,” taking “BBL Drizzy” to new heights.

WhatsApp Group Join Now
Telegram Group Join Now
Instagram Group Join Now

Leave a Comment