Record voice, music or video
Sing over a YouTube video, keep your voice separate for editing, and download the vocal track, an MP3 mix or an MP4 video.
Online karaoke recording and pitch correction
Paste a YouTube video link and sing along. Your voice stays separate from the music, ready for volume, pitch and timing adjustments before you download an MP3 or MP4.
You sing along to music that reaches you slightly late, so your voice lands slightly late too. This puts it back in place. The default usually works — and you can fix it after recording, by ear.
Plays three clicks and measures how long the sound takes to go out and come back through the mic. Hold your headphones against the mic and stay quiet.
For the first two, Chrome asks what to share. Pick this tab and tick “Also share tab audio” — without it the music is not recorded.
If your voice sounds early or late against the music, drag the orange track until the shapes line up. Drag the background to move through time, and use the mouse wheel to zoom in.
Free voice studio
YouSing turns your browser into a karaoke studio: record your voice with a video, correct pitch, adjust timing and export the result. There is nothing to install and no account to create.
Sing over a YouTube video, keep your voice separate for editing, and download the vocal track, an MP3 mix or an MP4 video.
Analyze the music to find its likely key, then control the scale, correction strength and response speed for anything from a natural result to a pronounced effect.
Your voice and mix are processed on your device. YouSing does not upload your recording to a server.
Yes. You can record, edit and download your karaoke without paying or creating an account.
No. Recorded audio and voice processing stay in your browser; the file only leaves your device when you choose to upload or share it.
Voice-only recording works in modern browsers. To capture the video audio too, use Chrome or Edge on a computer because mobile browsers do not allow audio capture from another tab.
YouSing detects the notes you sing and moves them toward the selected scale. It can also analyze the recorded music to suggest a key, while you control the correction strength and naturalness.
The device that picks up your voice. If you use a headset, choose it here — names only appear once you allow access.
How strongly your voice comes in. Hit test, sing at the volume you will actually sing, and raise it until the bar nearly fills.
If the bar turns red, lower it: sound that arrives clipped cannot be repaired later.
A few thousandths of a second pass between the music leaving your headphones and your voice returning through the mic. Uncorrected, your voice sits slightly behind the music.
Wireless headphones lag far more than wired ones. When in doubt, use the measure button.
Ready-made combinations of EQ, compression and room. It is the same treatment you hear here and the one that goes into the file — what sounds good is what you get.
The sense of space, like singing in a large hall instead of a small bedroom.
A little gives the voice body; too much pushes it away from the music.
Lifts the high end. To the right the voice is more present and clear; to the left, softer and darker. Useful when the voice sounds dull or too sibilant.
How hard the note is pulled toward correct pitch. At 100% the correction is strongest; lower settings fix only part of the drift and preserve more of your phrasing.
How long the correction takes to act. At 0 ms the voice jumps from note to note for a robotic pop effect. Between 30 and 60 ms usually sounds more natural.
Which key the song is in. With the right key, correction only uses notes that fit — in the wrong key it makes things worse.
The “find the key” button analyses the recorded music, estimates the most likely key and suggests it here.