Skip to content

Features

A free online multitrack audio editor that opens instantly

MixSongs is a browser-based audio editor built for one job: turning several songs into one great-sounding mix. No installer, no plugins, no account. Here is everything inside.

Cut and move in seconds

Cut Move Export 0:000:040:080:120:160:20 Sample 1M S Sample 2M S Sample 2 Sample 1 Sample 1 Sample 1
Animation: the scissors cut a song into three pieces; the middle piece is dragged onto another track and the last one slides back to close the gap.

The timeline

A real multitrack arrangement view, the same idea as a desktop DAW, trimmed to what mixing songs actually needs.

  • Unlimited tracks and clips

    Add as many songs as your computer’s memory allows. Move clips freely, including between tracks, and each clip keeps its own gain and fades.

  • Three ways to place a song

    At the start, at the playhead, or in sequence after the previous song, with an automatic overlap and crossfade built from volume checkpoints.

  • Cut, trim, move, fade

    Cut at the playhead with S or the scissors, drag clip edges to trim, drag any piece to another track or moment, and drag the corner handles for fade-in and fade-out.

  • Snap and grid

    Snap to the grid, clip edges or the cursor. Show the grid in time or in bars and beats at your BPM; hold Alt to move freely.

  • Loop and zoom

    Drag on the ruler to set a loop region. Zoom with Ctrl + wheel or +/−, and fit the whole project with F.

  • Undo everything

    Every edit can be undone and redone (Ctrl+Z / Ctrl+Shift+Z), with up to 200 steps of history.

See checkpoints in action

0:06.500 Export 0:000:040:080:120:160:20 Sample 1M S −15.0 dB
Animation of volume checkpoints being added and dragged on a track in MixSongs, with the waveform following the curve.

Volume checkpoints: crossfades you can draw

Most online tools offer a crossfade as a single slider: one length, one shape. In MixSongs every track has a volume line. Click it to add a checkpoint, drag a checkpoint to move it in time and level, drag a segment to move two checkpoints together, and double-click to delete. Between checkpoints the level changes linearly and sample-accurately.

Because you draw the curve, you control the transition. Hold the outgoing song at full level until the new chorus lands, duck it under a vocal intro, or make a slow two-bar blend. The waveform updates to show exactly what you will hear, and you can edit while the song is playing without clicks or jumps.

  • Shift while dragging: vertical only (change the level, keep the time).
  • Alt: no snapping, for fine adjustments.
  • Ctrl + click anywhere on a track: add a checkpoint exactly there.
  • Separated stems get their own volume line too, so you can fade the vocals out while the beat keeps going.

Speed and pitch, drawn like volume

0:00.000 Export 0:000:040:080:120:160:20 Speed 1.00×1.05×1.10×1.15×1.20×1.25×120126132138144150 BPM3:303:203:113:032:552:48 −0:42 Same key Pitch 0 st+1 st+2 st+3 st Same length
Animation: a song’s speed goes from 1.00× to 1.25× and the clip gets shorter in the same key; then its pitch goes up 3 semitones and the length stays the same.

Every track also has a speed line and a pitch line. Speed goes from 0.25× to 4× and keeps the key; pitch moves up to 24 semitones either way and keeps the length. Set one value for the whole song, or add checkpoints for a song that speeds up into the drop, or a key change for the last chorus. It all runs in your browser, for free.

Natural voice: a new key, the same singer

Vocals 0 st+1 st+2 st+3 st+4 st+5 st Ordinary pitch shift Sounds like a chipmunk MixSongs Natural voice Still the singer’s own voice NotesVoice characterOriginal
Animation: an ordinary pitch shift moves the whole voice up like a chipmunk, while MixSongs Natural voice moves the notes and keeps the singer’s voice character.

Raise a song by a few semitones with an ordinary pitch shifter and the singer starts to sound like a chipmunk; lower it and they sound like a robot. That happens because the shifter moves the voice’s formants, the resonances that make a voice recognisable, along with the notes.

With Natural voice on, MixSongs shifts the separated vocals on our servers with a studio-grade, formant-preserving engine, and the instruments on their own. The notes move to the new key, and the voice keeps its own colour. In our listening tests it beat every alternative we tried, including an AI voice model. It works on any song you have separated into stems.

Sound quality first

  • 32-bit floating-point engine

    All mixing happens in 32-bit float, so there is no internal clipping. The project runs at the highest sample rate among your files, and lower-rate files are resampled with a high-quality resampler.

  • Honest file information

    When you add a song you see its real codec, bitrate (CBR or VBR), sample rate, bit depth and channels, plus a quality badge, so you know whether your source is a 128 kbps MP3 or a 24-bit master.

  • Master limiter and clip protection

    A transparent master limiter keeps peaks under −1 dBFS during playback and export. With the limiter off, export can still scale the whole mix down so nothing clips (−0.3 dBFS for lossless formats, −1 dBFS for MP3). No compressor colours your sound.

  • Proper export

    MP3 at 128 or 192 kbps for free. Pro export, unlocked with any pack, adds WAV 16/24-bit or 32-bit float, FLAC 16/24-bit and MP3 320. Triangular dither is applied when reducing bit depth, normalizing is optional, and the export dialog recommends a format based on your sources.

Private by design

Your songs are decoded, played, edited and exported inside your browser. They are not uploaded to mix or export anything, so mixing with MixSongs costs us nothing, and that is why it is free. Projects autosave to your browser’s local storage (including the audio), so a refresh or a closed tab does not lose your work. Autosave is best-effort: browsers can clear storage, so export anything you care about.

The only feature that sends audio over the internet is optional AI stem separation, which needs heavy processing. Those files are processed and then deleted, as described in our Privacy Policy.

AI stems, built into the mix

0:00.000 Export 0:000:040:080:120:160:20 Sample 2M S ✦ AI Separating… Vocals Drums Bass Guitar Piano Music
Animation: AI splits a song in MixSongs into six lanes: vocals, drums, bass, guitar, piano and music.

Split any song into six stems with studio-grade AI, and keep working in the same project.

  • Vocals

    Clean lead and backing vocals for acapellas and karaoke.

  • Drums

    Kick, snare and cymbals as their own lane.

  • Bass

    The low end, isolated, for remixes and practice.

  • Guitar

    Electric and acoustic guitars on their own lane, to learn a part or play along.

  • Piano

    Piano and keys, isolated, with falling notes, hands and fingers to learn the part.

  • Music

    Everything else: strings, synths, pads and brass.

  • Per-stem control

    Each stem has its own fader, mute, solo and volume checkpoints.

  • Reverb control for vocals

    Take the reverb and room echo out of the vocals, then choose how much of it stays, live. Remove reverb from vocals.

  • Adds back to the original

    Together, the six stems reconstruct the original song, so nothing goes missing in your mix.

Why our stems sound cleaner

Typical stem splitter ×2 2 passes ×2 MixSongs ×4 4 passes ×4
Animation: a typical stem splitter analyses every moment of a song twice; MixSongs analyses it four times, with overlapping windows.

Two dedicated models do the work. One, trained only on vocals, lifts the voice out first; a second splits what is left into drums, bass, guitar, piano and music. Whatever traces of voice the second model still finds are kept out of the instrument stems.

AI separation listens to a song through short, overlapping windows. MixSongs overlaps them so that every moment is separated 4 times and the four results are blended, where many online splitters settle for 2 to save processing. More passes smooth out the seams and artefacts between windows, and you hear it as cleaner, steadier stems.

Supported formats

Import Export
Lossless WAV, FLAC, AIFFWAV (16/24-bit, 32-bit float), FLAC (16/24-bit), with Pro export
Lossy MP3, M4A/AAC, OGG Vorbis, Opus, WebM audioMP3: 128–192 kbps free, 256–320 kbps with Pro
Sample rates Any; the project uses the highest44.1, 48, 88.2 or 96 kHz (MP3 up to 48 kHz)

Made for everyone

  • 31 languages

    The interface is available in 31 languages, including right-to-left Hebrew, Arabic, Persian and Urdu, and follows your browser’s language automatically.

  • Keyboard first

    Space to play, S to split, Ctrl+D to duplicate, arrows to move the playhead, ? for the full list of shortcuts.

Learn more

Mixing and editing are free

Open MixSongs and drop in your first two songs. Lossless export comes with any pack.

Start mixing