Skip to main content

Online audio mixer

Multi-track mixer with per-track volume, pan, mute/solo, waveforms, and offset timing.

Free & unlimited
Master volume
100%
SilentLoud
Tracks
0/2 loaded
No file
No file
All processing happens in your browser. No data is sent to any server.

Layer up to eight audio files into one mix. Each track gets its own volume fader, a stereo pan control, mute and solo buttons, and a start offset so it can begin later than the others — which is how you put a voiceover on top of a music bed, or line up a sound effect with a moment in a recording. The signal path is exactly what the interface implies: every track runs through its own GainNode, then a StereoPannerNode, then a shared master gain, and the whole graph is rendered offline in a single pass. Vertical meters show each track's level against the master. Four mix presets set all faders at once for equal, voiceover, ducking and stacked layouts.

Key facts about Online audio mixer

Key facts about Online audio mixer
Track limit8 tracks maximum
Signal pathPer track: GainNode for volume, StereoPannerNode for position, then a shared master gain, rendered in one OfflineAudioContext pass
Per-track controlsVolume, pan, mute, solo, start offset in seconds, and a label
Mix presetsEqual (all at full), voiceover (first track to 30 percent), ducking (first track to 15 percent), fade stack (each track 25 percent quieter than the one before)
Mix lengthThe longest track's duration plus its own start offset
Output channelsAlways at least stereo, so panning is preserved even when every source is mono
Sample rateTaken from the first loaded track; other tracks are resampled into it by the audio graph
MeteringDual-channel vertical segment meters per track, driven by the measured peak times the effective volume
Output16-bit PCM WAV named mixed-audio.wav
Accepted inputMP3, WAV, OGG, FLAC, AAC and M4A, up to 500 MB per track

What happens to your file

Every track stays in your browser. Each file is read into an ArrayBuffer in this tab, decoded by the browser's own decodeAudioData, and mixed by an OfflineAudioContext running on your CPU; the result is written to a 16-bit WAV by JavaScript on this page. No file is uploaded, no project is stored, and no processing engine is downloaded — after the page loads, mixing produces no network request at all. The finished mix is a blob: URL in your tab's memory and disappears when you close it.

About this tool

  1. 1

    Add your first track

    Drop an audio file onto a track slot. It is decoded immediately and its waveform and peak level appear on the channel strip.

  2. 2

    Add up to seven more

    Use the add-track button. Each new track gets its own strip with volume, pan, mute, solo, offset and a label you can rename.

  3. 3

    Set the balance

    Pull background music down and leave voice at full, or tap a preset: voiceover puts the first track at 30 percent, ducking at 15 percent.

  4. 4

    Position tracks in time

    Use each track's offset to delay its start. This is how you place a sting at the 12-second mark or start music under the second half of a narration.

  5. 5

    Pan and check the meters

    Spread tracks across the stereo field with the pan control. The vertical meters show each channel's level so you can spot a track that will dominate.

  6. 6

    Mix and download

    The whole graph is rendered offline in one pass and comes back as a WAV called mixed-audio.wav.

Specs & compatibility
Input formatsMP3, WAV, OGG, FLAC, AAC, M4A
Maximum tracks8
Maximum file size500 MB per track
Output format16-bit PCM WAV, stereo or wider
Pan rangeFull left to full right via StereoPannerNode's equal-power law
OffsetPer-track start delay in seconds; the mix grows to fit the latest-finishing track
BrowsersChrome, Edge, Firefox and Safari 14.1+ — Web Audio only, no WebAssembly
MemoryEvery track is held decoded as 32-bit floats simultaneously, so eight long tracks is the real constraint, not the per-file cap
  • Set the loudest element first and build everything else around it. Starting with all faders at full and pulling down is harder than starting from the lead track.
  • A music bed under speech usually wants to sit somewhere between 15 and 30 percent — the ducking and voiceover presets are exactly those two points.
  • Use solo to audit one track in isolation, then unsolo before you render. Solo affects the mix as rendered, not just the preview.
  • Offsets are absolute seconds from the start of the mix, not relative to another track, so sketch your timeline before entering numbers.
  • Panning a mono source hard left or right halves its perceived contribution to the centre. For narration, leave pan at centre.
  • Normalize tracks before mixing if they came from different sources; matched loudness makes the faders behave predictably.
  • There is no automation in the mixer — each track's volume is one constant value. For a fade within a track, run it through the fade tool first.
  • Up to eight simultaneous tracks
  • Per-track volume, pan, mute and solo
  • Per-track start offset for timeline placement
  • Four mix presets including voiceover and ducking
  • Dual-channel vertical meters per track
  • Master volume control
  • Waveform preview per track
  • Single-pass offline render to WAV
  • Putting a music bed under a recorded voiceover at a sensible level
  • Combining separately recorded podcast microphone tracks into one file
  • Adding a sound effect at a specific moment in a narration
  • Layering ambience under a field recording
  • Building a simple audio montage from several clips with staggered starts
  • Merging a backing track and a vocal take into one deliverable
Eight. That limit is a deliberate interface choice rather than a technical wall — beyond eight channel strips the controls stop being usable on a normal screen, and the mixer has no automation, grouping or bussing to manage a larger session. In practice the binding constraint arrives earlier anyway: every track is held decoded as 32-bit float audio simultaneously, so eight long files can occupy several gigabytes of memory while you work. For anything larger, a desktop DAW is the right tool.
Yes, that is what the per-track offset control does. Set it in seconds and the track begins that far into the mix, with silence before it. The mix's total length grows to accommodate the track that finishes last, counting its own offset. This is how you place a sound effect at a specific moment or bring music in halfway through a narration. Offsets are measured from the start of the mix, not relative to another track, so work out the absolute times first.
Almost certainly because the summed tracks exceed full scale. Mixing is addition: two tracks each peaking near 0 dBFS sum to roughly twice that, which clips. The mixer applies your faders and master gain faithfully and does not insert a limiter, so it is your job to leave headroom. Pull the master down, or bring individual faders back until the meters stop hitting the red segments. Alternatively mix at a conservative level and raise the result afterwards with the audio normalizer.
Not within the mixer. Each track's volume is a single constant value applied across its whole length — there is no envelope or automation. To fade a track, run it through the audio fade tool first and bring the faded version into the mixer. For a crossfade between two pieces of music, the fade tool's crossfade mode handles that directly and produces one file, which you can then bring in here as a single track.
The sample rate of the first track you loaded. Everything else is resampled into it by the audio graph as it renders, which is transparent for normal material. If you are mixing sources at different rates and it matters — say a 48 kHz video-derived track with 44.1 kHz music — load the one whose rate you want the output to be first. The output WAV carries that rate, and the channel count is at least stereo so panning survives even with mono sources.
No. Every file is decoded and mixed entirely inside your browser tab using the Web Audio API — per-track gain and pan nodes feeding a master gain, all rendered by an OfflineAudioContext on your own CPU. The finished mix is written to WAV by JavaScript on this page. This tool does not even download a processing engine, so after the page has loaded there is no network traffic whatsoever. The mix exists only as a blob in your tab until you download it.

Related tools

View all

Part of Audio tools that run in your browser

Updated

We use anonymous analytics to improve ToolChamp. No personal data is stored or sold. Privacy Policy