Here, I'll do you a solid. There's an open source separation library known as Deezer Spleeter - it can separate the voice from the instrumental (and even more). The following site, Melody ML, is a site that's hosted by some generous soul that uses Deezer Spleeter. You can use this tool for free, you just need to provide an e-mail address (yes, you can use a bunk email address and it should still work). I've used it before when I want to get fancy with house mixes.
https://melody.ml/
Get yourself an MP3 version of the track you're talking about and give it a try, you'll just need to upload the track on the site and choose how you want it to be separated. When it's done, you can download the zip file from the site containing the isolated tracks - just extract the zip and there you go. It aint perfect, but it's pretty damn good. It runs on Python and Tensorflow (a machine learning library) if you're interested in the technicalities.
https://research.deezer.com/projects/spleeter.html