Voxengo SPAN: How to Set It Up, What Every Reading Means, and a Settings Chart by Job

The short answer: Voxengo SPAN is a free spectrum analyzer plugin (VST, VST3, AU and AAX, for Windows and macOS) that also carries a stereo correlation meter, true peak and clipping statistics, and LUFS and K-system level metering. Put it last on the master, or on any track you want to inspect, and set it to a slow average so the curve stops jumping. Then read it against a reference in your genre at matched loudness, not against a flat line. SPAN tilts its display 4.5 dB per octave by default, because the spectrum of contemporary music closely follows that slope, which is why its top end sits higher than on most other analyzers. Hover over a peak and the readout shows the frequency and the nearest note: the fastest way to name a boomy note or find an 808's key. An analyzer confirms what your ears suspect. It can't tell you what sounds good.

I master my own releases (Sony, Universal, Ultra, Dim Mak, 50M+ streams) and other people's songs as a service, and SPAN is the free analyzer I recommend in the best metering plugins. I also sell mastering chains, disclosed at the end.

What SPAN shows (it's more than a spectrum)

Download it from voxengo.com and nowhere else. SPAN is free; SPAN Plus is the paid edition, covered further down.

Max Crest Factor is the gap between the average level and the loudest short burst (Voxengo uses a 50 ms window for the peak RMS). The TP switch flips the peak readout between true peak and sample peak, and Reset starts the statistics over.

Two features most people miss. Hold Ctrl (Command on macOS) and drag across the spectrum to solo a narrow band, so you hear exactly what you're looking at; SPAN doesn't touch the sound once you let go. And right-click the spectrum to copy the frequency under the cursor, ready to paste into an EQ. The Hold switch freezes the display for a moment when you want to study it.

The mode settings that matter

Click the Mode selector to pick and edit a display mode. Voxengo calls the editing window the Spectrum Mode Editor, and these are the settings in it worth knowing.

About the slope. Real music has less energy in the highs than in the lows, so on an untilted display every song runs downhill. SPAN's 4.5 dB per octave default compensates, and that's why a mix looks brighter on SPAN than on a stock analyzer with no slope. Some engineers prefer 3 dB per octave, the slope at which pink noise reads as a flat line. Either works. What matters is using the same slope for your mix and your reference, and never comparing a SPAN screenshot with another analyzer's.

Labels and options can change between versions, so treat the plugin window and Voxengo's user guide as the authority.

SPAN settings chart, by job

These are starting points, not Voxengo's defaults. Voxengo recommends a higher Block Size together with a higher Overlap for accurate bass readings, and notes that very large blocks can need a bigger audio buffer in your DAW to avoid clicks. Large blocks also make the display lag behind the music; that's expected, not a fault.

How to set it up

1. Install it from voxengo.com in the format your DAW uses.
2. Put it last on the master, after the limiter, with the master fader at 0 dB, so it reads what you'll export. To inspect one sound, put it on that track instead.
3. Pick a mode and edit it to the job in the chart. Save your setup with the Presets button so every session opens the same way.
4. Set up a reference the right way. Put a song in your genre on its own track, routed around your master bus processing, and match its loudness to your mix; how to use reference tracks covers both steps, and Youlean Loudness Meter gives you the numbers.
5. Load a second SPAN on the reference with the same preset, and open the two windows side by side. Free SPAN draws curves from one instance only; Voxengo points to SPAN Plus for comparing spectrums across tracks.
6. Compare the same section, one region at a time. Drop against drop: the sub, then the low mids, then the presence range, then the air. Write down the gap, fix it in the mix, compare again.
7. Read the correlation meter on the loudest section. Voxengo calls 0 to +1 the acceptable range for stereo. Below 0, the channels are working against each other and the mix won't hold up in mono. What stereo imaging is explains the numbers.
8. Check the peaks. Switch TP on, reset the statistics, and play the song through. Then close SPAN and listen.

Reading the curve

The low end. Compare the sub (below about 60 Hz) and the low mids (roughly 200 to 400 Hz) with your reference. More low-mid energy than the reference usually means mud; a sub that towers over the reference usually means the mix will feel weak at loudness, because the low end is eating the headroom.

An 808 or bass note. Solo it, hover over the lowest strong peak (the fundamental, not the louder harmonics above it) and read the note and cents. If the 808 glides, read the sustained part, not the attack. In FL Studio, the FL Studio 808 chain uses this reading to set the sample's root note.

Harshness. Set the 2nd Type to RT Max and watch 2 to 5 kHz for narrow peaks that jump above the average on the loud notes. Ctrl-drag across them to hear which part it is before you cut anything; a dynamic EQ like TDR Nova is often the right tool, because it only cuts when the peak appears.

Width. SPAN can also analyze mid and side separately through its routing settings. A side channel with lots of energy below about 100 Hz is a wide low end, and it will lose power in mono. The best stereo imaging plugins covers the fix.

Readings that mislead

• Louder looks better. Level shifts the whole curve up. Match loudness first, or set Offset to Center so the overall level drops out of the comparison.
• Fast averaging. RT Avg with a short Avg Time jumps with every hit. Mixing to it is chasing noise.
• Heavy smoothing. It hides the narrow resonance you're looking for. Smooth less when you're hunting one frequency.
• The flat-line myth. A finished master isn't supposed to look flat on any analyzer. It's supposed to look close to records in its genre that already work.
• SPAN's integrated loudness on long tracks. Voxengo notes that SPAN's integrated reading only considers the most recent five minutes and updates twice a second. For a whole-song integrated number, plus LRA and PLR, use a dedicated loudness meter like Youlean.

SPAN Plus, and SPAN vs the rest

SPAN Plus is the paid edition. Voxengo lists static spectrum display, real-time spectrum import and export for multi-track comparisons, PNG export, and a wide display mode; check Voxengo's site for the current price. If you compare against references every day, the overlay is what the upgrade buys you.

For everything SPAN doesn't do: Voxengo's free Correlometer splits correlation into frequency bands, Youlean covers loudness, and iZotope Insight 2 and ADPTR Metric AB are the paid all-in-one and referencing picks. They're compared in the best metering plugins. When SPAN shows a problem, the fix is usually an EQ move, and the best EQ plugins has the options by job.

Where I land (and the disclosure)

SPAN shows you where a master differs from a reference: too much low mid, a sub that's too wide, a dull top. The fixes happen in EQ, low-end and width stages, at small amounts, before the limiter. The Ultimate Master Chain for Ozone has those stages sequenced and gain-staged: two EQ stages, Low End Focus, Exciter and Imager, then the Maximizer. It's built in Ozone, not SPAN, so keep SPAN last on the master to check the result. I'm biased; SPAN is free and worth installing whatever you master with.

Quick answers

Is Voxengo SPAN free? Yes. SPAN is free from voxengo.com, in VST, VST3, AU and AAX for Windows and macOS. SPAN Plus is the paid edition, adding static spectrums, spectrum import and export for multi-track comparison, PNG export and a wide view.

What are the best Voxengo SPAN settings? For whole mixes and masters: Type RT Avg, Block Size 8192, Avg Time around 5,000 to 6,000 ms, smoothing around 1/6 octave, and the slope left at 4.5 dB per octave. For bass and 808 notes, raise the Block Size and the Overlap.

Why does Voxengo SPAN use a 4.5 dB slope? Voxengo sets it because the power spectrum of contemporary music closely follows that slope, so tilting the display makes the spectrum easier to read. It's why the highs look higher than on most other analyzers. Use the same slope for your mix and your reference.

How do I read the correlation meter in SPAN? +1 means left and right are identical, which is mono. Around 0 means wide, uncorrelated stereo. Below 0 means the channels are working against each other, and parts will cancel in mono. Most mixes should sit between 0 and +1.