How do I separate vocals from a song on Windows PC?

Copper Contributor

I've recently been working on a project where I need to separate vocal from a song. I'm using a Windows PC, and I've tried a few programs but the results are not very good, either the operation is too complicated, or the quality of the separated sound is not good enough. I wonder if there is any methods that can do this simply and effectively? If anyone has a good recommendation or experience to share, it's really so needed! Thanks everyone!

15 Replies


Isolating vocals in a song can be a bit of a challenge, especially if you want to maintain the sound quality. I'd like to recommend you to try AudioKies, It’s simple to use and has a full range of features, including an “Vocals Extractor” function.


I followed this tutorial: (100% works)


Look, this is a screenshot I took on my computer, which can perfectly separate the vocals and background music!



I hope this helps! What other software have you tried?

You can use FFmpeg to attempt to separate vocals from music files on a Windows PC. While FFmpeg is a powerful command-line tool for audio and video processing, its ability to separate vocals from a song is limited compared to specialized software. However, you can use a technique known as "center channel extraction," which works best on stereo recordings where the vocals are mixed equally in both channels.

FFMPEG is not recommended if you never use the command prompt. The text command could be complex for separating vocals from a song. In addition, it is also a time consuming process to install FFMPEG tool on a Windows 11 or Windows 10 PC. It is much better to use some dedicated software to Isolate vocals from a song on Windows PC.

Spleeter is an open-source tool developed by Deezer that uses machine learning to separate vocals from music. It requires Python to be installed on your computer.


1. Install Spleeter using the following commands in the Command Prompt:


pip install spleeter


2. Download and Prepare Your Audio File. Use the following command in the Command Prompt to separate the vocals:

spleeter separate -i input.mp3 -p spleeter:2stems -o output


Replace input.mp3 with the path to your audio file. The separated files will be saved in the output folder. You can use the above trick to isolate vocals in a song on Windows 11 or Windows 10 PC.

@LottieLopezess  LALAL.AI is an online AI-powered service for vocal and instrument separation. It offers a free tier that you can use to separate vocals from a song.


Visit LALAL.AI: Go to the LALAL.AI website.


Upload Your Audio File: Click on the Select File button and upload the song you want to process.
Process the File: Choose the type of separation you want (e.g., Vocal and Instrumental).


Download the Separated Tracks: Once processing is complete, download the separated vocal and instrumental tracks.


These method provides reliable and free option for separating vocals from music tracks on a Windows PC and it is suitable for various levels of technical expertise.


Do you mind if I share my experience? I've recently uploaded a couple songs on Smule. Personally, I would spend about a dollar to buy karaoke tracks from the internet and then upload them to Smule.


As someone who loves to sing, I really hate opening a song on Smule only to find out that it's still mixed with the original vocals or the sound quality is poor, that's a really bummer feeling. I've tried using vocal remover software to separate vocals from a song , but it doesn't work well. Although you can remove some of the voice, but you can't remove it completely, and the sound quality will still be bad.


So, if you also want to upload songs, I suggest you can just search for karaoke tracks online. I myself have found a lot of good resources on a site called Karaoke Version.

Before separating vocal from a song, try adjusting the EQ settings in your chosen software to focus on the vocal range (around 200-400 Hz).
If the separation isn't perfect, try using additional software like iZotope RX or Waves C4 for further processing.

LALAL.AI's algorithm is based on machine learning, which means it may not always produce the desired output quality. You may need to compromise on the balance between vocal clarity and instrument separation.
Spleeter uses AI-powered algorithms to separate vocals from instrumental tracks. While it can produce good results, it may not always be perfect. You can refine your results by adjusting settings or using other tools for further editing.

If you have a digital audio workstation (DAW) like FL Studio or Ableton Live, you can use plug-ins specifically designed for vocal separation from a song. For example, iZotope RX 8 or Melodyne Studio are popular plug-ins that can help separate vocals from an instrumental track.

Honestly, Spleeter is not the ideal tool to separate vocals from music. Although Spleeter is relatively fast, processing high-quality, long audio files can still take a significant amount of time and computational resources. In songs where vocals overlap significantly with other instruments (e.g., guitar, synths), Spleeter might struggle to separate them cleanly, leading to bleeding of instrumental sounds into the vocal track and vice versa.

@LottieLopezess  I am using PhonicMind and it works good for normal usage. PhonicMind is a robust online tool specifically designed for separating vocals from music. It uses advanced AI algorithms to provide high-quality vocal and instrumental separation.


Go to the official website. Click on the "Upload" button to upload the audio file you want to separate. PhonicMind supports various formats, including MP3, WAV, FLAC, etc. Once the file is uploaded, PhonicMind will process the track to separate the vocals from the music. This usually takes a few minutes depending on the length and complexity of the track.


After the processing is complete, you can preview the separated tracks and download them. PhonicMind provides both vocal and instrumental versions of the track. This is the best tool I have used for separating vocals from music on my Windows 10 PC.


I usually use Izotope's RX7 software for various music edits, and I'm told that it can separate vocals from music tracks, but I haven't tried that feature yet. The RX7 is no longer available, and the latest RX9 software is quite expensive. 


If you're just working with individual tracks, you should be able to get what you need by the end of the trial period. Good luck with that!


BTW, I don't think you have to use RX9 if you just want to remove vocals from songs. you can try some free open source software like Audacity, which sometimes works pretty well. Of course, if you are looking for more professional processing, I still recommend you to try the trial version of RX9 and experience its powerful functions. All in all, it depends on your needs and budget to choose the tool.


Actually, no audio editing software (DAW) can perfectly separate vocals from a 2-channel stereo MP3 or WAV mix. If you're working on a music project, find someone else to share the individual tracks ("backbone") of their project, or ask them to make a mix that mutes certain tracks, for example, by turning off the guitar part, so that you can use that track as a backdrop to practice different guitar parts.


What exactly are you trying to accomplish? Do you want to remove vocals form music? or do you want to get separate instrument parts?  If you just want to practice a particular instrumental part, why not just ask another member or partner in the band for their track, it's much better than software processing.


Alternatively, if you do need software help, try some specialized tools like the free trial version of AudioKies, or some open source audio processing software like Audacity ( ), which won't remove the vocals perfectly, but may be able to achieve what you need.


I actually wasn't originally in music production and don't know anyone who owns those songs. However, I used to use Virtual DJ to separate vocals from a music for fun. Recently I switched to a Manjaro system and had to look for some new alternatives, and Ardour looked like it could take the place of Mixpad, which is a bit more complicated, but quite versatile. I'm using Mixxx instead of Virtual DJ, but it doesn't seem to be able to separate vocals directly.


I'm thinking of using Audacity to try separating voices, what do you think? I wonder if there are any special plugins or toolkits that can enhance Audacity for sound separation? Or, is there any other audio editing software for Linux that makes this separation easy?