Adding audio to a video without losing quality means handing the editor one uncompressed PCM WAV that is already mixed, ordered, and at the correct bit depth, so the timeline never has to re-encode the audio layer. Audio Joiner is built for exactly that handoff: it decodes each selected audio file in your browser, concatenates the decoded samples per channel in the order you set, and writes one locally downloadable 16-bit PCM WAV whose header reports the decoded sample rate, channel count, and exact data byte length. Because the samples are concatenated directly, with no inserted silence, no crossfade, no normalization, and no overlap, the joined file preserves the splice points you previewed. Nothing is uploaded to a server during this workflow: file reading, Web Audio decoding, channel concatenation, WAV encoding, preview, and download all happen in the current tab, which means the source files stay on your machine and the output you import into Premiere, DaVinci Resolve, CapCut, iMovie, or any other NLE is a faithful 16-bit PCM rendering of the decoded mix the joiner built. The combined WAV is uncompressed PCM, so it can be much larger than the MP3, AAC, Ogg, or FLAC files you selected, but every audio sample in the output came from a real decoded source sample rather than a recompression pass.
Why Adding Audio to Video Often Loses Quality
Most non-linear editors handle a new audio track the same way: import the file, drop it on a timeline track, mix it with the existing clip audio, and export the timeline. Behind that simple workflow, several quality-reducing steps can fire at once. If your source music is an MP3 or AAC file at 192 kbps, the decoder pulls it back to a PCM representation that is already missing the data the codec threw away during its own encode. When the timeline then mixes that PCM with the video's original audio, it may resample both sources to a single project rate—often 48 kHz for video work—which forces another interpolation step on every sample. On export, the timeline encodes the mixed audio again, this time into the codec and bitrate chosen in the export preset. Three encode or decode passes in one project is normal, and each one can shave transients, soften sibilants, or shift phase on bass-heavy material.
Compression is not the only culprit. Editors routinely apply automatic loudness normalization, dialogue enhancement, or a default ducking curve when a new audio track lands on the timeline. Those features are useful, but they change the samples in ways you did not ask for. Even a track labeled "no effects" can be pulled through a project-level gain stage or sample-rate converter before it reaches the export encoder. The cleanest way to avoid these invisible transformations is to deliver your video editor a single, already-finalized WAV that contains exactly the audio mix you want, at the bit depth and channel count your project expects. That is the job Audio Joiner is built to do: build the WAV outside the editor, then import one finished asset that the timeline treats as audio it should not modify.
How the PCM16 WAV Join Preserves Audio Data
Audio Joiner uses the browser's Web Audio API to decode every selected file through one shared AudioContext. Each decoded track arrives as an AudioBuffer of Float32 samples organized by channel. The joiner then copies each channel's samples into one growing Float32Array at a known cumulative frame offset, walks every channel independently, and feeds the combined buffer to a local encoder. The encoder clips float values to the range from -1 through 1, maps negative full scale to -32768 and positive full scale to 32767, treats non-finite values as silence, interleaves channels in frame order, and writes a 44-byte little-endian RIFF/WAVE header with the PCM format tag, channel count, decoded sample rate, byte rate, block alignment, 16-bit depth, and the exact data byte length. The header layout follows the standard documented in the Microsoft RIFF/WAVE format reference and the McGill WAVE PCM sound file format reference, so the file opens correctly in any editor or player that reads PCM WAV.
Two properties of this pipeline matter for video work. First, decoded samples are concatenated directly: the last sample frame of one track is followed immediately by the first sample frame of the next, with no inserted silence, no overlap, no crossfade, no normalization, no transition, and no automatic trimming. That gap-free behavior means a music intro that ends on a snare hit flows straight into a voiceover that begins on a vowel—exactly the splice point you hear in the previews. Second, because every file in one selection passes through the same AudioContext, the working sample rate can resample sources to a single rate. The sample rate shown in the preview and in the finished WAV header is that working rate, not necessarily the rate stored in the original files, and the output does not preserve independent source sample rates. Reordering changes the decoded-buffer array and visible list together, and clears any older WAV so a stale result cannot claim a previous order.
Combine Audio Tracks for Your Video in Three Steps
- Open Audio Joiner and select between 2 and 10 files from the same project folder. The picker recognizes common MP3, WAV, M4A, AAC, Ogg, WebM, and FLAC names and MIME types, but a recognized extension does not guarantee that every codec variant can be decoded by Safari, Chrome, Firefox, or another browser. Damaged, encrypted, incomplete, mislabeled, or unsupported files produce a clear error rather than a guess.
- Wait for the browser to decode every file, then preview each track in the order shown. Use the Move up and Move down controls to walk tracks into the exact sequence your video timeline expects—for example, ambience first, dialogue next, music last. Reordering updates both the decoded-buffer order and the visible list together, and clears any older WAV so a stale preview cannot mislead you.
- Select Join in this order, preview the PCM16 WAV the tool produces, and download it. Verify the duration, the join points, and the channel playback in a media player before importing the file into your video editor. Keep the source files until you have confirmed that the downloaded WAV sounds the way you expect on the timeline.
Input Limits and Channel Rules That Affect Quality
Audio Joiner enforces explicit budgets so a join cannot silently truncate, downmix, or downsample your material. Every limit below is checked against the actual decoded buffers, not just the file sizes on disk, which is why a short-but-high-sample-rate selection can still hit the sample budget before the duration cap.
| Constraint | Value or rule |
|---|---|
| Files per join | 2 to 10 |
| Maximum file size per track | 25 MiB (26,214,400 bytes) |
| Maximum total selection size | 100 MiB (104,857,600 bytes) |
| Maximum total decoded duration | 30 minutes |
| Maximum sample rate | 192 kHz |
| Maximum channel count | 8 channels |
| Maximum total channel-samples | 30,000,000 across all channels |
| Channel-count rule | All tracks must decode to the same channel count; mono/stereo mixes are rejected |
| Inserted silence, crossfade, or normalization | None—samples concatenate directly in list order |
| Output container | RIFF/WAVE, PCM format tag, 16-bit signed little-endian, interleaved |
If a selection fails a check, the joiner rejects it with a specific message and produces no output. Nothing is shortened, downmixed, downsampled, or omitted to fit a budget. This is the key contrast with editors that quietly remix a 5.1 source down to stereo on import: Audio Joiner refuses the mismatch instead of remapping it, which protects the channel layout you planned for in your final video mix.
When to Prepare the Audio Track Outside the Video Editor
Building one lossless WAV before you open the NLE is the right move whenever you know the final mix in advance. That covers most narration-over-music beds, podcast-style audio for talking-head footage, soundscape layers for short films, and ad spots where the voiceover, music sting, and room tone are produced separately and simply need to land in a fixed order on the timeline. For these jobs, Audio Joiner gives you a single import that the editor treats as one finished asset—no automatic gain, no ducking, no surprise resample, no metadata rewrite.
There are jobs the joiner is not built for. If you need fades, loudness matching across very different sources, channel conversion from mono to stereo or vice versa, a compressed deliverable such as MP3 or AAC, or any preservation of tags, album art, chapter markers, or cue points from the source files, reach for a full audio editor or a video editor's audio mixer instead. The joined WAV is a clean PCM representation of the browser-decoded samples; it does not preserve the original codec, bitrate, encoder settings, or container metadata. Use Audio Cutter when you need a precise time range from a single track rather than a concatenation of full tracks, and consult a dedicated editor when you need sample-level repair. For a walkthrough of the same joining workflow tuned specifically for video projects, see Add Audio for Video Free: Merge Clips Locally.