ContextCleaner: see every entry in the Windows 11 right-click menu, and which app added it — on the Microsoft Store.

Last updated:

Guided workflow variation

A folder of WAVs in,every file at the target LUFS.

Normalizing one file in an editor takes a minute; normalizing forty and proving it takes the evening. WavSpec sets a whole folder to any LUFS or RMS target, limits overs, converts sample rate and bit depth, and writes a report of every file.

Pick a built-in spec or type your own target, add the folder, and run once: every WAV is normalized, peak-limited where needed, re-measured, and reported.

Pay once. Own it forever.

One-time purchase on the Microsoft Store · No subscription · Instant install

Get it from Microsoft

Buy through the Microsoft Store—instant install, automatic updates, and a Microsoft receipt for easy expensing. No account needed on our site, and no subscription on Store purchases.

Full WavSpec overview
One-time purchase No telemetry Local-first

windows · re-measured from disk · originals never overwritten · nothing uploads

If this sounds like your week

Peak normalizing is not loudness normalizing.

Normalizing to a peak level makes the loudest sample hit a number, but two files with the same peak can sound several LU apart. Clients and platforms ask for an integrated LUFS figure instead, and most audio editors set it one file at a time, through a dialog, with the result checked on a meter afterward.

Where tooling usually breaks

The batch is where small errors hide.

Forty stems normalized by hand means forty chances to skip one, pick the wrong preset, or export at the wrong bit depth. Raising quiet files pushes their transients past the ceiling. Some bulk tools rewrite the WAV and drop the bext and iXML chunks the client’s system reads, and nobody keeps a list of what each file measured.

Where WavSpec lands

Set the target once and run the folder.

WavSpec aims each file at the exact target, not just inside the window, then applies look-ahead true-peak limiting only where gain would cause an over. It converts rates with a windowed-sinc resampler and bit depth with TPDF dither, carries bext, iXML, LIST and cue chunks across, and re-reads every written file to confirm the numbers.

Pay once. Own it forever.

One-time purchase on the Microsoft Store · No subscription · Instant install

Get it from Microsoft

Buy through the Microsoft Store—instant install, automatic updates, and a Microsoft receipt for easy expensing. No account needed on our site, and no subscription on Store purchases.

Learn more about WavSpec

windows · re-measured from disk · originals never overwritten · nothing uploads

What you get

Feature highlights

Named client specs

ten built-in presets for broadcast, podcast, ad and audiobook delivery, each cited to its source document, plus your own specs, exported and imported as files to share.

Measured per ITU-R BS.1770-5

integrated, short-term and momentary loudness, loudness range, and 4× oversampled true peak, checked against the EBU Tech 3341 and 3342 test signals, plus RMS, noise floor and room tone.

Corrected in one run

gain to the exact target, look-ahead true-peak limiting only where needed, windowed-sinc resampling, bit depth with TPDF dither, and PCM WAV or constant-bit-rate MP3 delivery.

Re-measured from disk

every delivered file is read back and measured again, with up to three passes to land on target, and files that already meet the spec pass unchanged.

A report to send

PDF, HTML, CSV and JSON with before and after values, a loudness timeline per file, client and project details, and a SHA-256 digest for every delivered file.

Audiobook heads and tails

room tone padded with the file’s own room tone or trimmed, digital silence replaced, plain file names, and 192 kbps CBR MP3 for ACX, with originals never overwritten.

Especially when

  • Check a whole batch against the client’s loudness, true-peak and format spec instead of metering and fixing files one at a time
  • Correct loudness, true peaks, sample rate and bit depth in one run, without the gain change pushing peaks over the ceiling
  • Get audiobook chapters within the ACX rules for RMS, peaks, room tone and MP3 format before the upload
  • Send a report with before and after values and a SHA-256 digest for every file, with the audio kept on the PC and the originals untouched

Who benefits

Real-world scenarios

Stems for a game audio team

A composer delivers a library of music stems that a game audio team wants at -18 LUFS integrated, -1 dBTP, 48 kHz, 24-bit WAV, with a list of what each stem measured.

Choose New in Client specifications, set the target to -18, the tolerance to 1 and the peak ceiling to -1, enter 48000 and tick 24-bit PCM, and save. Every stem in the folder is normalized and delivered, and the CSV report goes to the team with the files.

A mixed batch of formats

The client sent source audio as WAV, AIFF from a Mac session, FLAC from an archive, and a couple of MP3 references, and wants all of it back as 48 kHz WAV at one loudness.

WavSpec reads all of them, decodes the compressed files, resamples where the rate differs, normalizes to the target, and writes PCM WAV at the spec’s bit depth. The Format column shows each source, and the report shows each result.

Normalizing without losing timecode

Production sound arrives as Broadcast Wave files whose bext and iXML chunks carry the time reference and the scene and take data the editor’s system relies on.

WavSpec copies bext, iXML, LIST and cue chunks into each delivered file in their original order and copies unknown chunks as they are, so time references and scene data survive the gain change. The verified loudness goes into bext when the spec asks for it.

Deep dive

What a batch loudness normalizer should do, and how WavSpec does it

Measure loudness, not peaks. Integrated loudness per ITU-R BS.1770 is the figure clients and platforms quote. WavSpec measures it natively with BS.1770-5 K-weighting and gating, along with short-term and momentary loudness and loudness range per EBU Tech 3341 and 3342, and its tests check the meter against the EBU test signals.

Aim at the target. A tolerance window is for checking, not aiming. WavSpec computes the gain that puts each file on the target itself, writes the file, measures it again from disk, and adjusts in up to three passes. Files that already meet the spec pass unchanged, and Copy already-compliant files to the delivery folder puts them beside the corrected ones so the delivery is complete.

Limit only what needs it. When gain would push a true peak over the ceiling, a look-ahead limiter with a 2 ms window and 4× oversampled detection holds it, with a 0.1 dB margin for WAV and 0.5 dB for MP3. Files that never come near the ceiling get gain only, so their dynamics stay as mixed.

Handle the format in the same pass. Allowed sample rates, bit depths and channel layouts are part of the spec. WavSpec resamples with a windowed-sinc converter, converts float and other bit depths with TPDF dither, and can deliver constant-bit-rate MP3 through the Windows encoder when a spec asks for it. It reads WAV, BWF, RF64, AIFF, FLAC, MP3, AAC, M4A and WMA in mono, stereo, 5.1 and 7.1.

Keep the metadata. Broadcast Wave bext, iXML, LIST and cue chunks stay in their original order, unknown chunks are copied verbatim, and verified loudness values can be written into bext as EBU Tech 3285 version 2 fields.

Write somewhere else. Delivered files go to the delivery folder, written to a temporary file and moved into place. Originals are opened read-only.

Keep a record. Every run saves a report as PDF, HTML, CSV and JSON. The CSV has one row per file with integrated loudness and true peak before and after, sample peak, RMS, noise floor, maximum short-term loudness, loudness range, the correction, the output file, and the SHA-256 of the output and the source.

Automate when it repeats. A watch folder checks and corrects new files as they finish writing, and the command line runs a named spec on files or folders with exit codes for scripts and DAW post-export hooks.

How it is built

Built to change only what the spec needs, and show the numbers.

Verified from disk

Every delivered file is read back and measured again, with up to three passes to land on target. The report’s After columns and SHA-256 digests describe the exact bytes delivered.

Originals never overwritten

Sources are opened read-only. Delivered files go to a separate delivery folder as a temporary file moved into place, and Broadcast Wave bext, iXML and LIST chunks are carried across.

Nothing leaves your PC

No account, no telemetry, no update check and no cloud processing. Measurement and correction run on your PC. One-time purchase in the Microsoft Store.

Head-to-head

How WavSpec compares

Feature WavSpec Audio editor, file by file Online loudness service
Whole folder in one run Yes, files and folders, plus a watch folder for new files One file per dialog Upload, process, download
True-peak limiting after gain Yes, look-ahead, with 4× oversampled detection A separate limiter pass Varies by service
Sample rate and bit depth set to the spec Yes, windowed-sinc resampling and TPDF dither in the same run Chosen in the export dialog each time Varies by service
BWF bext and iXML kept Yes, in original order, with verified loudness written into bext Depends on the editor Varies by service
Report for every file PDF, HTML, CSV and JSON with before and after values and SHA-256 Notes you keep yourself Varies by service
Where the audio goes Stays on this PC. Nothing uploads. Stays on this PC Uploaded to the service

FAQ

Straight answers—no glossary dump

What LUFS target should I normalize WAV files to?

The one your client or platform names. European broadcast uses -23 LUFS under EBU R 128, US television uses -24 LKFS under ATSC A/85, and podcast platforms commonly use around -16 LUFS. WavSpec ships presets for each of these and lets you type any target, tolerance and true-peak ceiling into a spec of your own.

How do I batch normalize a folder of WAV files?

Choose Add folder (Ctrl+Shift+O) or drag the folder into WavSpec, pick a Client specification, set the Delivery folder, and press Check and correct batch (F5). Every file is measured, normalized, peak-limited where needed, converted to the spec’s format, re-measured from disk, and written to the delivery folder with the report beside it.

Can I normalize to an RMS level instead of LUFS?

Yes. In the spec editor, tick Judge level by whole-file RMS in dBFS instead of integrated loudness in LUFS, then set the target and tolerance in dBFS. The ACX and Author’s Republic audiobook presets work this way. You can also apply the ceiling to sample peak in dBFS instead of true peak in dBTP when a spec states it that way.

What happens to my original audio files?

They stay exactly as they were. WavSpec opens originals read-only and writes every delivered file to the delivery folder you choose, first as a temporary file and then moved into place. If the delivery folder is the source folder, the delivered file gets a new name, so the original is never overwritten. Copies renamed for a platform’s file-name rule are new files too, and the source keeps its name.

Does my audio leave the PC, and how do I get WavSpec?

Your audio stays on this PC. WavSpec opens no network connection: there is no account, no telemetry, no update check and no cloud processing, and settings, specs and the log stay in your local app data folder. It runs on Windows 11 version 22H2 or later on an x64 PC. WavSpec is a one-time purchase in the Microsoft Store, and the Store runs a 15-day free trial of the full product.

Related pages

-23 LUFS and -1 dBTP, across the whole delivery. -16 LUFS for every episode, from the pilot to the back catalog. Every chapter checked against ACX, and the fixable ones fixed. -24 LKFS for every spot, checked and corrected in a batch.

A batch of audio in.
A verified delivery and its report out.

WavSpec checks every file against the client’s spec, corrects loudness, true peak, sample rate and bit depth in one run, re-measures each delivered file from disk, and writes a PDF, HTML, CSV and JSON report. Originals are never overwritten, nothing uploads, and the Microsoft Store runs a 15-day free trial.

Pay once. Own it forever.

One-time purchase on the Microsoft Store · No subscription · Instant install

Get it from Microsoft

Buy through the Microsoft Store—instant install, automatic updates, and a Microsoft receipt for easy expensing. No account needed on our site, and no subscription on Store purchases.

Open full product page

ACX and Audible are trademarks of Audible, Inc. or its affiliates. Apple Podcasts is a trademark of Apple Inc. Spotify is a trademark of Spotify AB. EBU R 128, ATSC A/85 and the other named specifications belong to their publishers. WavSpec is not affiliated with or endorsed by any of them.

Technical specifications

Technical specifications for procurement

Spec Implementation
Data Sovereignty Reads only the files and folders you add, or the watch folder you choose, and opens originals read-only; writes delivered files and reports only to the delivery folder you pick, as a temporary file moved into place; settings, specs and the log stay in your local app data folder on this PC
Telemetry Status None; no account, no telemetry, no update check, and the app opens no network connection
Core Runtime C# on a self-contained .NET 10 runtime with WPF, packaged as MSIX for the Microsoft Store; loudness metering, true-peak detection, resampling and limiting are implemented natively; AIFF is read with NAudio, and MP3, AAC, FLAC and WMA use the Windows Media Foundation codecs
Network Requirements Fully functional offline; WavSpec makes no network connections, and the Microsoft Store handles the trial and purchase
Deployment Compatibility Windows 11 version 22H2 or later, x64, with 4 GB of RAM and about 150 MB of disk; one-time purchase on the Microsoft Store with a 15-day free trial
Inputs WAV, BWF, RF64, AIFF, FLAC, MP3, AAC/M4A and WMA; mono, stereo, 5.1 and 7.1; files, whole folders, or a watch folder
Measurement ITU-R BS.1770-5 integrated loudness, EBU Tech 3341 short-term and momentary loudness, Tech 3342 loudness range, 4× oversampled true peak, sample peak, whole-file RMS, noise floor, and head and tail room tone
Correction Gain to the exact LUFS or RMS target, look-ahead true-peak limiting, windowed-sinc resampling, bit depth and float conversion with TPDF dither, CBR MP3 through the Windows encoder, room-tone padding and trimming, and plain file names on delivered copies
Built-in specs EBU R 128 v5.0, EBU R 128 s1, UK DPP AS-11, ARD/ZDF/ORF TV (2025), ATSC A/85 (2026) full-mix / short-form, Apple Podcasts, Spotify audio ads, ACX audiobook, Author’s Republic audiobook, and generic 48 kHz / 24-bit; user specs export and import as files
Verification Every delivered file is read back from disk and measured again, with up to three passes to land on target
Outputs Corrected PCM WAV or CBR MP3 in a separate delivery folder, with bext, iXML, LIST and cue chunks kept and verified loudness written to bext (EBU Tech 3285 v2)
Report PDF, HTML, CSV and JSON with before and after values per criterion, loudness timelines, a batch consistency line, client and project details, and SHA-256 digests of delivered and source files
Automation Watch folder in the app (Ctrl+W); command line with --spec, --out, --check-only and exit codes for scripts