If you’ve skimmed our news coverage over the past few months, you’ve probably noticed a growing trend: fans using AI to turn their own voices or the voices of existing artists into a voiceover. Popular names like drake, weekend, And Liam Gallagher of Oasis.
This is a fascinating development, and has seismic implications for the music industry. Although only a few notable examples have caught the public’s attention so far, thousands of fans have already used AI to remake popular songs with new voices.
Have to ask sometime AI Paul McCartney sings Billy Joel’s Piano Man? how AI Kanye West and AI Rihanna take on Eminem and Dido’s Stan? Fancy listening AI covers Michael Jackson Rick AstleyOr AI does Kurt Cobain blur? Now you can.
How does it work? Most of the people making these cards use an open source software called So-VITS-SVC or RVC to process the voices. These are AI-powered deep learning models that can convert voice recordings into the singing voice they’ve trained using audio files of any voice. By ripping vocal chords from an artist’s recordings and training the samples, users can teach the software how to sound like the artist in question.
There are several ways to access these models. The software can be downloaded and run locally, but the process is a bit more complicated than installing a regular application. If you want to run RVC on your computer and learn how to train models and process voice, YouTuber p3tro Great training for that
Alternatively, there are many paid services that offer a more streamlined way to achieve the same results. Let’s take a look at six great websites you can use to create AI voices.
Before you start processing a vocal, you want to make sure you have a clean a cappella vocal recording. This can be achieved by using many paid tools that extract vocals from the entire mix (a process called stemming) but there are free options as well. VocalRemover And Audio Studio. Check out our guide to the best cord separation tools to learn more.
Kids AI works with a 'partner artist community’, so if you use one of these AI artist voices, you can use any results for personal use, but if you want to publish it commercially you must go past the original artist. Then you work with the artist on the royalty split. Seems reasonable. Only a dozen artists have signed up so far, but there are plenty of royalty free models to use and even AI tools for modeling.
Another expensive site, but to clone your own voice, you can use parody models – we have Donald Trump singing – or choose from artist models (not as popular as some sites). Musicfy acknowledges that its service raises ethical concerns when using famous voices, but says: „You can use AI voice cloning freely, but it’s important to exercise responsibility”.
One of the most fully featured, yet most expensive, AI voice platforms. Voicify offers voice training, text-to-speech, samples of famous names, famous singers and innovative, funny characters and a good sense of community. Overcharged compared to some others, so choose it only if you need a lot of voices and have particularly deep pockets…
Singify seems like a new platform, but is free to use (so far) and has some famous singer AIs, including the very young Taylor Swift and Ariana Grande. We tried the Swift model and got good downloadable results, so if you want to experiment with voice AI, this is a good site.
Revocalize is a bit different from other sites here, in that it’s less about modeling famous voices (there are only four models to choose from), and more about creating new models. You can download a free VST plugin to make the changes for you. Some of the features on the website aren’t working and the prices haven’t been ironed out yet, but the recent addition of AI mastering is great. A must see.
If you like popular voices, this is a great choice, although the app doesn’t mention any credits. It’s all about these changes and you have to be careful to get the right source material, models are great.
Before concluding, it is our duty to remind you that the legal status of AI cards is very murky. While the legal limitations of this type of AI-powered voice use have not yet been removed, we recommend erring on the side of caution here.
While we maintain it firmly No While it’s a good idea to try to monetize AI-generated music that imitates an existing artist’s voice — for example, by uploading it to Spotify or YouTube — there’s no harm in playing with this technology at home for your own entertainment.