01
Start listening
Allow the microphone and pick the input: the phone’s own mic, a measurement mic or a USB interface fed from the desk. Type the label of the mix — Wedge 3, IEM 2.
Free ring-out assistant · A Squaresmile app
Push it till it rings. SpectraSense writes it down.
Ring out a wedge with the phone in your hand. SpectraSense names the frequency that rings, the note, the GEQ fader — and the neighbour when it sits between two — and suggests a narrow cut to start with. Every ring goes into a Found list per wedge or mix, ready to copy as text. Free, in the browser, nothing to install. Next to it, for measuring: an SPL meter with a level protocol and a transfer function.
Free — 30-second signup, no card
Nothing recorded or uploaded
Phones, tablets and laptops
How it works
The ring-out you know — with the numbers kept for you.
02
Hold it at the singer’s mic
Where the singer’s mouth will be, not against the wedge. Music off, unused mics muted — only the mic and the mix you are ringing out stay open.
03
Push the send until it rings
As it starts to ring, SpectraSense flags it: frequency, note, GEQ band and a cut to start with. Pull the fader back first — the card holds the last ring for you to read.
04
Cut, repeat, copy
Make the cut on that mix’s EQ and go up again for the next ring. When you are done, copy the Found list as text into the advance sheet or a message.
Features
Which frequency, which fader, how much.
Feedback, not notes
A ring is flagged only when three tests agree for 0.3 seconds: a narrow peak, a lone tone without overtones, and a level that is steady or growing. A sung note with its overtones stays unflagged.
The fader to pull
The nearest band of a 31-band graphic EQ — and the neighbour when the ring sits between two, so you pull both a little rather than one a lot. The note comes with cents, e.g. D7 −7¢.
A cut to start with
−3 dB and narrow, about 1/10 octave (Q ≈ 14). If the same ring comes back, a further −3 dB, to −9 dB at most — then the mic or the wedge is simply too loud for the room.
The Found list
Every ring with its time, note, band, cut and the wedge or mix it belongs to. It stays through Stop and a reload; Copy as text sends it on, one ring per line.
Made for the phone in your hand
On a phone the answer card sits on top with the frequency in large type, readable at arm’s length with the other hand on a fader. The screen stays awake while it listens, and the app speaks 11 languages.
Nothing recorded
The sound is analysed in the browser and never leaves the device. The Found list stays on this device; only your display and detection settings follow your account.
In the tool
The ring, the card, the list.
Measurement suite
After the ring-out: measure.
Two more tabs and a Setup dialog sit next to Ring-out — same free account, same browser, nothing to install. Ring-out stays the first tab.
Setup: routing, calibration, test tone
Pick the input channel your measurement mic is on, with a live meter per channel. Load the mic’s calibration file — miniDSP UMIK, REW or plain frequency / dB text. Calibrate SPL with a 94 or 114 dB calibrator or against a reference meter. Pink or white noise, a sine or a log sweep, starting at −40 dB FS and never by itself.
Level: SPL meter and protocol
LAF, LAeq over 10 s, 1 and 15 minutes, LCeq, LCpeak and LAFmax — checked against DIN 15905-5, the WHO standard and other audience limits. Log a show one level per second and print the protocol on A4 with a SHA-256.
Transfer: magnitude, phase, coherence
Measure a system against a reference — a second input from the desk, or the test tone as sent. A delay finder compensates the arrival time; smoothing from 1/3 to 1/24 octave; store up to eight traces and compare them.
- Indicative with a phone mic. A phone or laptop microphone is no measurement mic — calibrated or not, its levels are a guide. SpectraSense is not a sound level meter for a record a permit or the law asks for: use one, and import its CSV log into the protocol.
- Transfer needs an interface and a wider screen. It runs on a tablet or computer from 768 px wide; on a phone the tab explains why. One interface for mic and reference is best — Bluetooth is unusable for timing.
- The browser decides the channel count. Chrome and Edge pass every input of an interface; Safari and Firefox often only the first one or two. Setup shows what really arrives.
- Calibration stays on this device. The mic file, the SPL calibration, level logs, protocols and traces stay in this browser and are never synced to your account — export them as a file to take them along.
Honest limits
What SpectraSense is — and what it is not.
- A ring-out tool, not a show-time suppressor. It never touches your desk; you make every cut. Under music, a held pure tone — a flute, a whistle, a synth pad — looks like feedback and gets flagged.
- It flags a ring as it starts to ring. A ring has to stand out from the spectrum around it (14 dB by default, the Sensitivity setting) before it is flagged — push the send slowly and keep a hand on the fader.
- Low hoots read smaller on a phone mic. Phone microphones roll off the low end, so a hoot around 100–300 Hz looks weaker than it sounds. Trust your ears there, or use a measurement mic or a USB interface.
- Ring-out hears one channel. Your device’s microphone or the first input of an audio interface — or the channel you pick in Setup. Its display is dBFS; SPL is read on the Level tab once the input is calibrated.
FAQ
Quick answers.
Does SpectraSense remove feedback by itself?
No. It tells you where to cut — the frequency, the note, the nearest GEQ band and a suggested cut — and you make the cut on your desk. SpectraSense never touches the console.
Can I use it during the show?
It is made for the ring-out before the show, not for running under music. A held pure tone — a test oscillator, a whistle, a flute or a synth pad with few overtones — looks exactly like steady feedback and will be flagged.
Does it work with my phone’s microphone?
Yes. Hold the phone where the singer’s mouth will be — at the mic, not against the wedge: a phone microphone overloads long before a wedge does. Low hoots around 100–300 Hz read smaller on a phone than they sound in the room, because phone microphones roll off the low end; a measurement mic or a USB interface reads them better.
How does it tell feedback from a sung note?
It flags a frequency only when three tests agree for 0.3 seconds: a narrow peak standing well above the spectrum around it, a lone tone without the overtones a musical note has, and a level that is steady or growing. A note with overtones stays unflagged; you can switch the note check off.
What does the suggested cut mean?
A starting point the way most engineers ring out: −3 dB and narrow, about 1/10 octave (Q ≈ 14). If the same ring comes back, a further −3 dB, to −9 dB at most. On a graphic EQ it names the nearest band — and the neighbouring one when the ring sits between two. It is a suggestion, not a setting.
What is the Found list?
Every ring SpectraSense confirms, with the time, frequency, note, GEQ band, suggested cut and the label of the wedge or mix you were ringing out. It stays through Stop and a reload until you clear it, and Copy as text puts it on the clipboard, one ring per line. It is kept on this device only — not synced, never uploaded.
Which input does it listen to?
Your device’s own microphone, a measurement mic or a USB interface fed from the desk. Ring-out hears the first input of an interface until you pick another channel in Setup. How many channels arrive depends on the browser: Chrome and Edge pass them all, Safari and Firefox often only the first one or two. The Ring-out display is in dBFS; for SPL, calibrate in Setup and read the Level tab.
Is anything recorded or uploaded?
No. The sound is analysed in your browser; no audio is recorded or sent anywhere. Your display and detection settings follow your account; the Found list, the chosen input and channel, the calibration, level logs, protocols and stored traces stay on the device.
Can SpectraSense measure SPL?
Yes, on the Level tab — once the input is calibrated with a calibrator or against a reference meter; until then it reads dB FS and says so. With a phone or laptop microphone the levels are indicative, and SpectraSense does not replace a sound level meter when a permit or the law asks for a record: use one, and import its CSV log into the protocol.
What do I need for the transfer function?
A tablet or computer screen (768 px and wider) and, best, an audio interface with two inputs: the measurement mic and a feed from the desk as the reference. With one input, the test tone as sent can be the reference. Bluetooth outputs are unusable for timing, and two devices with separate clocks drift apart.
Is SpectraSense free?
Yes. Free with a free Squaresmile account, which also unlocks every other free Squaresmile browser tool. It runs on phones, tablets and laptops, with nothing to install, and speaks 11 languages.
Ready when you are
Ring out the next wedge with the phone in your hand.
Create a free Squaresmile account and you’re in — one login unlocks SpectraSense and every other free Squaresmile app.
