Avoid Mistakes When You Join Audio Locally

A mistake-free audio join requires every source to decode to the same channel count, a list order that matches your intended sequence, and files that fall inside the browser-enforced 25 MiB per-file and 100 MiB total budgets. The Audio Joiner decodes each file in your current browser tab through one shared AudioContext, so any track that the browser cannot decode is rejected with a clear error instead of producing a broken result. Reordering happens through Move up and Move down controls, and selecting a new order also clears any older previewed WAV so a stale result can never claim a previous sequence. The tool is intentionally strict about limits: between two and ten files, no inserted silence, no crossfade, and no automatic remixing. Once you know those rules, most "mistakes" are really mismatches between what you expected the joiner to do automatically and what it actually does by design. This guide walks through the checkpoints that prevent each common mistake before, during, and after the click.

What Goes Wrong When You Join Audio

Joining audio is a small task on paper — combine two or more tracks into one — but the small mistakes are the ones that ruin the result. The most common problems are not visible in the file list; they only show up when you play the downloaded WAV back.

A wrong order is the easiest mistake to make. Browsers let you select files in any sequence, and many tools just concatenate them in that selection order. If you select your outro before your intro, that is the order you get. A codec that the browser cannot decode is the next silent failure: a file with a recognized extension like .m4a may still fail if the codec variant is not registered in your current browser. A mono and stereo mix is another trap — some tools silently duplicate or drop channels, while the Audio Joiner rejects the selection outright. The output also tends to surprise first-time users: a 3 MB MP3 can produce a 40 MB WAV because PCM16 is uncompressed.

The mistake behind all of these is assuming the joiner will quietly fix things. A well-behaved local joiner does not fade, does not normalize, does not remix channels, and does not preserve tags. Once you accept that, the join becomes predictable, and predictable joins are joins without mistakes.

Get the Files Ready Before You Open the Tool

A short pre-flight check removes most failure modes before you ever click Join. The Audio Joiner does its work in one shared AudioContext, which means every file has to satisfy the same conditions or none of them will join.

• Confirm the channel count matches. All files must decode to the same number of channels — all mono or all stereo. Mixing a mono voice memo with a stereo music bed is the most common reason a join is rejected. For background on mono versus stereo behavior, see what happens when you try to mix mono and stereo files.
• Check the file size. Each file can be up to 25 MiB (26,214,400 bytes), and the full selection can be up to 100 MiB (104,857,600 bytes). Files above those limits are rejected before decoding starts.
• Use a recognized format. The picker recognizes common MP3, WAV, M4A, AAC, Ogg, WebM, and FLAC names and MIME types. A recognized extension does not guarantee that every variant decodes in your current browser, so be ready for a clear error if the codec is unsupported.
• Keep the originals. Nothing on the page is uploaded, but the joiner writes a new RIFF/WAVE file with interleaved, little-endian, signed 16-bit PCM samples — the original codec, bitrate, tags, and metadata are not preserved. Hold on to your source files until you have confirmed the downloaded WAV.

This checklist is not optional decoration. Each item maps to a hard limit enforced by the tool, and a failure at any of them produces a rejection with a specific message rather than a silent workaround.

Join Audio in Three Steps

Once the files pass the pre-flight check, the actual join is short. The Audio Joiner reads the files through your browser, decodes them, and concatenates them in the order shown in the list — nothing more, nothing less.

1. Choose 2 to 10 audio files. Use the file picker to select all of the tracks you want to combine. The picker accepts recognized MP3, WAV, M4A, AAC, Ogg, WebM, and FLAC files. Files outside the 25 MiB per-file or 100 MiB total limits, or selections below two or above ten files, are rejected with a clear message.
2. Preview the decoded tracks and reorder the list. Wait for the browser to decode each file. Once decoded, use the Move up and Move down controls to put the list in the order you want. The decoded order and the visible list are kept in sync, and reordering clears any older previewed WAV so a previous sequence cannot leak into the result.
3. Join in this order, preview the WAV, then download. Click Join to concatenate the channel samples directly, without gaps, overlaps, crossfades, or normalization. Preview the resulting PCM16 WAV, then download it. The file is encoded locally as a RIFF/WAVE file with interleaved, little-endian, signed 16-bit PCM samples at the shared working sample rate.

For a deeper walkthrough of the join plan and what each preview step actually shows, see how to join audio files in your browser step by step.

Limits the Browser Enforces Before You Click Join

The limits below are checked against the selection before decoding begins. They are explicit and enforced as rejections, not silent trims. If any value is exceeded, the selection is rejected with a specific message and nothing is shortened, downmixed, downsampled, or omitted to fit.

Limit	Maximum value	What it covers
Files per selection	2 to 10	Outside this range the selection is rejected before decoding.
Per-file size	25 MiB (26,214,400 bytes)	Each encoded input file measured before decoding.
Total selection size	100 MiB (104,857,600 bytes)	Sum of all selected files before decoding.
Total decoded duration	30 minutes	Measured across all tracks after Web Audio decoding.
Channel count	8 channels	All tracks must decode to the same channel count; mono and stereo mixtures are rejected.
Sample rate	192 kHz	Rates above this are rejected.
Total channel-samples	30,000,000	Counted across all channels; a high-rate multichannel selection can hit this before the duration limit.

Decoding uses the Web Audio API, and the relevant specifications are documented at MDN's BaseAudioContext decodeAudioData and AudioBuffer references. Because each file is decoded through the same AudioContext, the browser may resample sources to that context's working sample rate — so the rate you see after decoding can differ from the rate stored in the original files.

What the Output Keeps and What It Drops

The output is a brand-new RIFF/WAVE file. Knowing exactly what is preserved and what is rewritten avoids the next common mistake: assuming the downloaded file is the same as the inputs in everything but length.

What stays the same:

• Track order, set by the Move up and Move down controls in the visible list.
• Channel identity for each track — left and right channels are concatenated independently before WAV interleaving, so a stereo file's left and right channels survive the join in their original positions.
• The decoded sample sequence for every track.

What changes:

• The codec. PCM16 WAV is uncompressed and is not a re-wrap of the original MP3, AAC, Opus, Vorbis, or FLAC.
• The sample rate. Web Audio may resample sources to the AudioContext's working rate, and the output uses that single shared rate.
• The bitrate. PCM16 has a fixed depth, while compressed formats had a variable bitrate that is discarded.
• Container metadata. Tags, album art, chapters, cue points, loop points, loudness fields, and timestamps from the original files are not carried into the new WAV.
• File size. A 5 MB compressed file can produce a 50 MB WAV after decoding and re-encoding — this is expected, not an error.

Float samples are clipped to the signed PCM16 range: negative full scale maps to -32768, positive full scale to 32767, and any non-finite values are written as silence. The header is a standard 44-byte little-endian RIFF/WAVE block — RIFF size, WAVE and fmt identifiers, PCM format tag, channel count, sample rate, byte rate, block alignment, 16-bit depth, data identifier, and the exact data byte length. Every download is freshly encoded with those fields, so the result is a clean PCM file rather than a re-wrap of the original compressed container.

The practical consequence is that you should keep the source files and confirm the downloaded duration, order, joins, channel playback, and file size before deleting any original.

When the Join Is Rejected Instead of Silently Fixed

A "mistake" in this workflow usually surfaces as a rejection rather than a wrong-sounding WAV. Knowing the failure messages in advance lets you fix the selection without guessing.

• "Damaged, unsupported, or mislabeled file." The browser's Web Audio decoder cannot read the file, even though the extension looked right. Damaged, encrypted, incomplete, or unsupported files fall into this bucket. There is no server-side codec fallback, so trying a different browser or re-exporting the file from its source application is the fix.
• "All files must decode to the same channel count." You tried to mix mono and stereo. The joiner does not silently duplicate, drop, average, or remap channels — the selection is rejected so you can convert the file yourself first.
• "Selection exceeds the per-file, total, duration, channel, rate, or sample budget." One of the limits in the table above was tripped. The specific message names which limit, so you can remove a track, split a long file, or convert to a supported rate before retrying.
• "Selection must contain between 2 and 10 files." Either too few or too many tracks were chosen. For a single-file time-range edit, the Audio Cutter is the right tool instead.

Decode work carries a job identifier, and component cleanup invalidates older jobs so a slow read cannot overwrite a newer selection. If you re-select the same files, the file input is reset after every choice, so the picker behaves the same way each time.

When to Use a Different Tool

The Audio Joiner is built around one job — concatenating full tracks in a chosen order, gap-free, locally. A few common tasks are not what it is for, and trying to force them through the joiner is its own category of mistake.

• For a time range from one file, use the Audio Cutter instead. It trims a range and downloads a local PCM16 WAV without joining anything.
• For fades, loudness matching, channel conversion, compressed output, metadata preservation, or sample-level repair, a dedicated editor is the right tool. The Audio Joiner does none of those by design, and it is not a quiet substitute for them.
• For reversible experiments, keep the source files until you have verified the downloaded WAV. The new file is independent of the inputs and is not a re-encoding of them.

Treat the tool as a precise concat: it does one thing, it does it locally, and it tells you up front when the inputs do not fit. That predictability is what makes the join mistake-free in the first place.