On-device AI — professional-grade, fully private.

Free Online Audio Editor

Mix, trim and fade audio on a multitrack timeline, then export your mix.

Free, with no watermarks.

  • Multitrack mixing
  • Trim, split & fade clips
  • Export to MP3, M4A or OGG

What do you want to do?

Import audio, arrange and fade clips on a multitrack timeline, then export your mix.

Edit audio with a real multitrack timeline — import clips, arrange and trim them, split at the playhead, add fades, balance the mix, and export a mix. It all runs on your device, not our servers.

A real multitrack mixer

Stack clips on separate tracks and mix them exactly how you want. Drag clips to reposition them, grab an edge to trim, split at the playhead, and shape each clip with fade in, fade out and per-clip gain.

  • Unlimited audio tracks
  • Drag to move, handles to trim
  • Per-clip fades and gain, per-track volume, mute & solo

Private and offline

Every file is decoded and mixed at full resolution, so the export matches what you hear on the timeline. The final mix is rendered as MP3, M4A or OGG for free (WAV and FLAC are Pro), and the tool keeps working offline once the page has loaded.

Multitrack Audio Editing: Tracks, Clips and Fade Curves

The editor works like a multitrack studio in your browser. Import MP3, WAV, M4A, FLAC or OGG files into the media pool, then double-click a file to place it as a clip on the timeline. Keep one sound per track — a voice note on Track 1 and a music bed on Track 2, for example — so each part of the mix is handled on its own. Every track has Mute, Solo, a volume level and a pan control; in the tutorial the music track is turned down to 40% so it sits under the voice.

Every clip edge can fade, and each fade has a shape. There are six curves, including Equal power, Exponential and Smooth. Equal power keeps the loudness steady through a fade, which suits a 2-second music fade-in; Exponential holds and then falls away, which suits a 3-second fade-out. Play it back and the music rises under the voice, then drifts out.

  • Import MP3, WAV, M4A, FLAC or OGG into the media pool
  • Double-click a file to place it on a track
  • Mute, Solo, volume and pan on every track
  • Six fade shapes, including Equal power and Exponential

BPM Detection and Snap-to-Beat Editing for Music

Cutting music by ear usually lands near a beat, not on it. Find the tempo listens to the mix, measures its tempo and draws the beat grid on the timeline. On the tutorial loop, which was made at 90 BPM, it measured 90.07 BPM and drew 31 beat lines. The grid division can be switched from beats to eighths (61 lines on the same loop) or to sixteenths for finer edits.

Once the grid is on, editing follows it. Put the playhead on a beat and press S to split the clip there. Drag a clip and a drop near a beat lands right on it; trim a clip by one beat and its edge stops on the line. Find the tempo first, and every cut, move and trim you make stays in time with the music.

  • Find the tempo: 90.07 BPM measured on a 90 BPM loop
  • Grid division: beats, eighths or sixteenths
  • S splits the clip at the playhead
  • Drags and trims snap to the nearest beat line

Markers, Ripple Delete and Exporting a Time Range

Long recordings are easier to manage with a few landmarks. Press M to drop a marker at the playhead, click a marker to jump back to it and double-click it to remove it. Ripple delete removes a piece and closes the gap behind it, so no silence is left: in the tutorial a 30.7-second recording became 20.7 seconds in one step. Ctrl + C and Ctrl + V copy a clip and paste it at the playhead.

The session sets the sample rate of the whole mix, from 44.1 up to 192 kHz; the tutorial session runs at 48 kHz, while Free exports stay at 44.1 kHz. When you only need part of the project, Export a time range writes just that section instead of the whole mix — choosing 2 to 7 seconds produced a five-second file.

  • M adds a marker; click to jump, double-click to remove
  • Ripple delete: 30.7 s → 20.7 s with no gap
  • Ctrl + C / Ctrl + V paste at the playhead
  • Session sample rate 44.1–192 kHz (Free exports at 44.1 kHz)
  • Export a time range: 2–7 s gives a 5 s file

Clean Up a Voice Recording: Normalize, EQ and Remove Silence

A voice note that is quiet, muddy and full of pauses is fixed in the order that matters: level first, then tone, then gaps. Normalize brings the loudest peak to −1 dB, so the whole recording sits at a usable level. The track EQ then fixes the tone — in the tutorial the Low band is cut by 8 dB to clear the mud and the High band is lifted by 6 dB — and the compressor is switched on to even out loud and quiet words.

Remove silence then takes the gaps out. On the tutorial recording it removed almost 11 seconds of pauses, turning 30.7 seconds into 20.0. For recordings with background noise or an echoey room, Pro adds Denoise and Remove room. Everything runs on your device, and the recording is never uploaded.

  • Normalize: the loudest peak to −1 dB
  • Track EQ: Low −8 dB, High +6 dB
  • Compressor evens out loud and quiet words
  • Remove silence: 30.7 s → 20.0 s
  • Pro: Denoise and Remove room

Speed Up Audio Without Chipmunk Voice, Change Pitch and Crossfade

Speed and pitch are separate controls, so tightening a recording does not raise the voice and retuning it does not change its timing. Under Time, Speed 1.25× took a 30.7-second voice note down to 24.6 seconds while keeping its pitch — a 440 Hz test tone stayed at 440 Hz. Pitch +2 moves the voice up two semitones and leaves the length exactly as it was.

Crossfade joins two clips without a click: it overlaps them, by 0.25 seconds in the tutorial, with a fade on each side of the join, so it belongs at every point where two takes meet. With Pro, the Transcript lets you edit speech by its words — delete a word in the text and the matching audio is cut.

  • Speed 1.25×: 30.7 s → 24.6 s, same pitch
  • Pitch +2 semitones, same length
  • Crossfade: a 0.25 s overlap with a fade on each side
  • Pro: Transcript — delete a word to cut the audio

Voice Changer Presets and MP3 Title, Artist and Album Tags

Change voice offers six presets: Higher, Lower, Brighter, Deeper, Anonymous and Cartoon. Each one moves pitch and vocal tract separately, and both values are shown, so you can see exactly what a preset does to the voice. Change voice is a Pro feature.

An exported mix is a new file with no tags, so a music library files it as Unknown Artist. The export dialog fixes that: fill in Title, Artist, Album, Year and Comment, and they are written straight into the file with no re-encode. A field left blank writes no tag at all. In the tutorial the exported MP3 was opened in the Audio Tag Editor, which reported 4 tags found, so the file arrives in any music library with its name.

  • Six Change voice presets, including Anonymous and Cartoon (Pro)
  • Pitch and vocal tract shown separately
  • Export fields: Title, Artist, Album, Year, Comment
  • A blank field writes no tag
  • Check the result in the Audio Tag Editor

Split a Song into Vocals and Instrumental Stems

Split into stems takes the vocals out of a song, or the music out from under them, on your own device with nothing uploaded. Quick preview splits the song in under half a second (0.4 s in the tutorial) and puts Vocals and Instrumental on two new tracks, lined up with the song, while the original is muted so you can compare. Solo a stem to hear it alone.

Quick preview tells you whether separation helps; Full quality is for the stem you keep. It uses a 30 MB separation model that runs on your device — a 25-second song took about 27 seconds. With Pro, a song can be split into four stems: vocals, drums, bass and everything else.

  • Quick preview: Vocals + Instrumental in 0.4 s
  • Stems land on new tracks, lined up with the song
  • Solo a stem to hear it alone
  • Full quality: a 30 MB model, about 27 s for a 25 s song
  • Pro: four stems — vocals, drums, bass, other

Pan, Stereo to Mono, Swap Left/Right and Mid/Side

Pan places a whole track in the stereo field; in the tutorial a track is set 40% to the left (L 40). The Channels tools go further and rewrite what the two channels of a clip contain. Mono folds left and right into one, so L equals R — the fix for a lopsided recording. Swap L/R exchanges the channels, for microphones plugged in the wrong way round.

Side keeps only what differs between the channels, so anything mixed dead centre cancels: on the tutorial song it came out 10 dB down, a quick way to hear what was mixed wide. Mid is the same as Mono. Each tool writes a new clip and leaves the original in the media pool, and undo restores it.

  • Pan: a track set 40% left
  • Mono: left and right folded into one (L = R)
  • Swap L/R for channels recorded the wrong way round
  • Side: the centre cancels, −10 dB
  • Every tool makes a new clip; undo restores the original

Auto-Duck Music Under a Voice with Volume Automation

Ducking, turning the music down under every word, is the mixing job anyone with music under a voice has to do. Open the music track's volume automation and choose Auto-duck under Track 1. The editor writes 24 points: the music drops to 20% (−14 dB) under the speech and comes back up in every pause.

What auto-duck writes is an ordinary envelope you can still edit. Drag any point by hand to shape one moment, click the line to add a point and double-click a point to remove it. Each track also has its own effects: Reverb, set to 24% in the tutorial, adds a touch of space that glues a voice into the mix, and Echo sits beside it.

  • Track volume automation: a line you shape
  • Auto-duck under Track 1: 24 points, music at 20% (−14 dB)
  • Drag, add or remove points by hand
  • Track effects: Reverb and Echo

LUFS Loudness Meter, True Peak and Mastering for YouTube

Streaming platforms turn every upload to the same loudness, so a mix that is too hot simply gets turned down with its clipping intact. The loudness meter measures the mix the way they do: integrated loudness, true peak, loudness range and short-term max. A voice and music mix in the tutorial read −15.0 LUFS with a true peak of +2.3 dBTP, and the clip report found 179 samples at full scale.

Fix the clipping at the faders first: pulling the music fader to 40% brought the mix to −16.6 LUFS and −1.7 dBTP, with no clipping. Then pick a target — Spotify and YouTube −14, Apple Music and Podcast −16, EBU R128 −23 — and the meter states what reaching it would take before anything changes. For YouTube it said turn up 2.6 dB and add a limiter. With Pro, the export masters to the target.

  • Integrated loudness, true peak, loudness range, short-term max
  • Clip report: samples at full scale vs true-peak overshoot
  • Music fader at 40%: −15.0 → −16.6 LUFS, +2.3 → −1.7 dBTP
  • Targets: Spotify/YouTube −14, Apple Music/Podcast −16, EBU R128 −23
  • Pro: master to the target at export

Record Your Voice Online and Export MP3, M4A or OGG

Record mic records your voice straight into the editor. When you press Stop, the take lands in the media pool just like an imported file; double-click it to put it on a track, then edit it like any other clip. The microphone stream never leaves your device, and nothing is uploaded.

Export renders the mix on your device and encodes it in the format you choose. MP3, M4A and OGG are free; WAV and FLAC are Pro. Before you save, a summary line states exactly what the file will be: format, sample rate, channels and size. In the tutorial a six-second mono voice take, exported as MP3 at 44.1 kHz and 192 kbps, came out at 145 KB.

  • Record mic: the take lands in the media pool
  • Double-click the take onto a track
  • Export: MP3, M4A, OGG free; WAV and FLAC Pro
  • Summary line: format, sample rate, channels, size
  • 6 s at 192 kbps → 145 KB

How to edit audio online

  1. 1
    Import your audio

    Click Import audio and choose your files — they decode straight into the media pool.

  2. 2
    Build & mix your timeline

    Drop clips onto the multitrack timeline, drag to arrange them, trim the edges, split at the playhead, and set fades and per-track volume.

  3. 3
    Preview & export

    Press play to preview the mix, then export your mix. The full mixdown is rendered at full resolution.

Frequently asked questions

Is the audio editor really free?

Yes. Editing and exporting MP3, M4A or OGG is free, with no watermark; WAV, FLAC and higher quality come with Pro.

Is my audio uploaded to a server?

No. Every file is decoded and mixed at full resolution, so the export matches the timeline.

What formats can I import?

Any audio your device can read — commonly MP3, WAV, OGG, M4A and FLAC.

What does it export?

It renders your full mix offline and downloads an MP3, M4A or OGG file for free, at up to 44.1 kHz and 16-bit. WAV and FLAC, higher sample rates and 24/32-bit depth are Pro.

Can I mix multiple tracks?

Yes. Add as many tracks as you need, layer clips on each, and control volume, mute and solo independently before exporting the mix.

Related audio tools

Editing, mixing and export all happen on one timeline, with no round-trips between steps.

4.8out of 5

from 184 ratings

Rate this tool

Tap a star — it takes a second