How to remove vocals from a song at this site?

To do so, you need to sign up first. It is super easy to make an account in our site. To separate vocals from a song you need a credit. After you obtained a credit, you will be able to upload your song. Once you upload your song, an AI-based algorithm processes it and separates the vocal from instruments. It creates two mp3 files (vocal and music) which you can download them. The music file only contains the music where the vocal is removed and the vocal file only contains the vocal sound where the music is removed.
Click here to upload your song

Your file should be in mp3 format and it only take about a minute to process it. Please read our FAQ at the bottom of the page if you have any further question.

"Sometime music is the only medicine the heart and soul needs. We are very pleased that we could provide such a service for people to be entertained. Please share our site with your friends if you liked it."

What is Easy-Karaoke?

Easy-karaoke is a platform for separating vocals from music tracks. It uses one of the world’s best AI-powered technology to extract vocal and instrumental tracks from any audio. This online service aims to make music creation and mixing easier. As a next-generation music separation service, it allows anyone to quickly and effortlessly remove vocal and instrumental tracks without quality loss.

What technology is used?

Finding an instrumental or karaoke version of a song can be pretty easy unless you're dealing with a song that isn't popular. That leaves you with just the full version of the song, complete with vocals. So how do you convert it into an instrumental or karaoke track?
Accompaniment tracks are made in the studio and are often referred to as "minus-one" tracks because they are missing one track. In our case, that would be the vocal track. The only way you would be able to remove the vocal track perfectly from the music entirely is if you have a multitrack version of the song, which is unlikely. However, there are ways to remove vocals or significantly lower them on any stereo song file to create a near-perfect instrumental or karaoke version.
One of the methods that allow us to achieve this is training a machine learning model to recognize the vocal content of a song via a set of isolated vocal and music tracks. A well-trained model makes the best results possible. A new open-source AI tool has such property and allows to isolate vocals from music tracks in a fast and easy way.
The software is called Spleeter and was developed by music streaming service Deezer for research purposes. In November 2019 the company released it as an open-source package, putting the code up on Github for anyone to download and use. However, you’ll need some tech expertise to use it. Unless you’re regularly playing with software like Python or Google’s AI toolkit TensorFlow (which was used to train Spleeter) you’ll have to download a few programs to get everything up and running. And you’ll have to be comfortable using a command-line input (albeit a very simple one) instead of a more accessible visual interface.
To make everything simple, we have provided this platform where you can simply upload your song and all the process will be done on the server automatically without the need for installing any programs like Python or TensorFlow. You can listen to some examples of the vocal separation that has been done via Spleeter below: