Vocal reducer
Reduce components shared by the left and right channels of stereo audio. Compare the original with the processed audio, then save the processed result.
What this tool does
Reduce sound shared by both channels
Take the difference between the two channels of stereo audio to reduce center components. There is no strength control or AI-based separation of vocals alone.
Compare the original and processed audio
After creating the processed audio, switch between Original and Processed previews. Centered instruments and reverberation may also change, so check the accompaniment as well as the vocals.
Save two-channel audio
Choose MP3, WAV, M4A, FLAC, OGG, MP2 or M4R. Output is fixed at two channels; AMR and mono output, which combine the channels, are unavailable.
How to use
Choose stereo audio
Use Choose an audio file to select one audio or video file, or drop it into the tool. Files fetched through public URL input go through the same reading checks. After reading it, the tool checks that the first audio track has exactly two channels. Mono audio and audio with three or more channels cannot be processed.
Choose output settings
Choose the output format and sample rate. Bitrate is also available for formats other than WAV and FLAC. The sample rate defaults to Match the original. M4R requires audio no longer than 40 seconds.
Reduce center vocals and check in stereo
Select Reduce center vocals. When the result is ready, compare the original and processed previews. Use stereo playback that keeps the channels separate to check changes to the vocals and accompaniment.
Save the processed audio
Download with Save the processed audio. The processed result is saved even while you preview the original. Changing output settings invalidates the previous result; create it again before saving.
When to use it
Reduce vocals for singing practice
Reduce centered vocals and compare with the original to choose audio for practice. Vocal placement and reverberation may leave vocals audible and alter the accompaniment. The tool does not guarantee a completely vocal-free karaoke track.
Hear the effect of centered sound
Compare how centered vocals or instruments affect your own stereo material. The processed channels have opposite phase, so listen in stereo after saving as well.
Use processed audio from a video
Read the first audio track from a video and save an audio file with reduced center components. Video is not saved, and the input audio must have exactly two channels.
Stereo audio and center-component reduction
How center components are reduced
For left and right channels L and R, the tool calculates side=(L−R)/2, outputs side on the left and −side on the right. Components that are identical in both channels cancel in the difference, reducing sounds such as centered vocals.
The tool does not identify audio content and select vocals alone. Centered instruments may also be reduced, while vocals spread across the channels and reverberation may remain. There is no reduction-strength setting or separation-model selector.
Why stereo playback must be preserved
The same difference signal is placed in the two output channels with opposite phase. Combining them into mono can cancel that signal too, making the audio quieter or silent.
Output is therefore fixed at two channels. AMR and mono output are unavailable. Check whether your playback device uses a mono setting or combines the channels in another way.
Preview audio and the saved file
The original/processed switch is for comparing the effect of reduction. The save button always targets the processed file; selecting Original does not change what is saved.
For MP2, a WAV version is used for preview while the selected MP2 file is downloaded. Recreate the result after changing the output format or quality.
Choosing format and quality
The default is MP3 at 192 kbps. WAV and FLAC do not use the bitrate setting. Increasing the sample rate cannot restore sound that is absent from the original audio.
M4R is limited to audio no longer than 40 seconds. This tool does not automatically shorten longer audio. Use Audio trimmer first if you need a particular section.
Frequently asked questions
Can this completely remove only the vocals?
It cannot fully isolate vocals. Centered instruments may change, while off-center vocals and reverberation may remain. Compare the original and processed audio to decide whether the result suits your use.
Can I convert mono audio to stereo to use this tool?
Duplicating the same audio into both channels causes the left-right difference to cancel. This tool accepts audio that already has two channels when read; it does not duplicate mono audio into stereo.
Why can the processed audio disappear in mono playback?
The left-right difference is output with opposite phase, so combining the channels may cancel it. Play it as two-channel stereo and check your device for a mono setting.
What is saved while I preview the original?
The processed audio is saved. The preview switch is for comparison and does not change the saved file to the original.
File handling
- Input
- Add a file by local selection, drag and drop, or public URL. The first audio track of one audio or video file is used. Exactly two channels are required. The original file is not overwritten.
- Processing
- Audio reading, center-component reduction and re-encoding run in your browser.
- Output
- Processed audio is saved with two channels. The original filename extension is replaced by the selected output format. M4R is limited to 40 seconds.