Sound Measurement Glossary: Acoustics and Hearing Terms Explained
This glossary collects the most common terms in sound measurement, hearing health, and occupational noise — defined in plain language, with cross‑links to deeper coverage on the rest of the site. Use it as a quick reference when reading the science page, the workplace standards page, or any technical document that assumes familiarity with the vocabulary.
A
A‑weighting (dBA)
A frequency weighting filter applied to sound pressure level measurements that approximates the response of human hearing at moderate listening levels. A‑weighting heavily attenuates low frequencies (about −30 dB at 50 Hz) and slightly attenuates above 6 kHz, leaving the 1 – 5 kHz range nearly flat where the ear is most sensitive. Almost every modern noise standard — NIOSH REL, OSHA PEL, WHO community guidelines, ISO 1996, EU 2003/10 — specifies A‑weighting. See frequency weightings.
Acoustic trauma
A sudden, severe hearing loss caused by a single high‑intensity exposure — typically an explosion, a gunshot at close range, or an industrial accident. Distinct from gradual noise‑induced hearing loss because the mechanism is mechanical disruption rather than metabolic exhaustion of the cochlear hair cells. Damage may include ruptured eardrum and disrupted ossicular chain in addition to cochlear injury.
Action level
The exposure level at which a hearing‑conservation program becomes mandatory under occupational regulations. In OSHA 29 CFR 1910.95 the action level is 85 dBA / 8 h TWA (the PEL is 90 dBA, but mandatory audiometric testing and worker training begin at 85). EU Directive 2003/10/EC has two action values: lower at 80 dBA, upper at 85 dBA. See workplace standards.
AGC (Automatic Gain Control)
A signal‑processing feature found in most consumer microphones and
recording chains that automatically adjusts input gain to keep the
recorded level near a target. AGC is helpful for voice calls and
casual recording but actively harmful for sound‑level measurement
because it makes the relationship between the actual SPL and the
digital sample value time‑varying and non‑linear. Our browser meter
requests autoGainControl: false from the OS; consumer devices vary
in how strictly they honor that flag, which is one reason
calibration matters per device.
ANSI S1.4
The American National Standard specifying the performance and calibration requirements for sound‑level meters. The international counterpart is IEC 61672‑1. Both define Class 1 (precision, ±1 dB tolerance) and Class 2 (general purpose, ±2 dB tolerance) performance grades. Most occupational compliance instruments are Class 2.
Audiogram
A graph of hearing thresholds across frequencies, used to characterize a person's hearing. Produced by an audiometer in a sound‑treated booth. A baseline audiogram before occupational exposure and an annual follow‑up are required by OSHA 29 CFR 1910.95 for workers above the 85 dBA action level. The classic NIHL signature is a "notch" — a sharper drop at 4 kHz than at adjacent frequencies — visible even when the rest of the audiogram is normal.
B
Background noise
The ambient sound level present at a location with the source of interest absent. For a measurement to be valid, the background should be at least 10 dB below the target level (the source is at least 10 dB above the background). Background between 6 and 10 dB below requires a correction; background within 6 dB of the target makes the measurement unreliable.
Basilar membrane
The membrane inside the cochlea on which the sensory hair cells sit. Its mechanical tuning maps frequency to position: high frequencies vibrate the base near the oval window, low frequencies vibrate the apex. Damage to the base is the early signature of NIHL because the base sees the highest acoustic stress.
C
C‑weighting (dBC)
A flat‑ish frequency weighting used for peak measurements and for sources dominated by low‑frequency content (concerts, subwoofers, thunder, fireworks). C‑weighting is nearly flat from 31.5 Hz to 8 kHz, with only a few dB attenuation at the extremes. Always specified alongside the level: "100 dBC" not just "100 dB."
Cochlea
The fluid‑filled spiral organ in the inner ear that transduces sound pressure into neural signals. Contains roughly 15,000 sensory hair cells arranged on the basilar membrane. Mammalian cochlear hair cells do not regenerate, which is why noise damage is permanent. See hearing health.
D
Decibel (dB)
A logarithmic unit expressing the ratio between two values, used in
acoustics for sound pressure level (SPL), sound intensity, and sound
power. The decibel is not a unit in the SI sense; it requires a
reference value to be meaningful. For sound pressure, the reference
is 20 µPa (the threshold of hearing at 1 kHz), and the formula is
L = 20 × log10(p / 20 µPa). See the decibel scale.
dBA, dBC, dBZ
Decibel sound pressure level with A, C, or Z (zero, flat) weighting applied. Each weighting yields a different number for the same physical sound. Always specify which weighting was used when reporting a measurement.
dB FS (full scale)
A digital‑audio scale in which 0 dB FS is the maximum representable sample value. Used in DAWs and audio production. Cannot be directly compared to dB SPL without a reference calibration that maps a known acoustic pressure to a known sample value.
dB SPL (sound pressure level)
A decibel value with reference to 20 micropascals (20 µPa) of air pressure variation. The default scale for environmental and occupational acoustics. The number you see on a sound‑level meter is dB SPL unless explicitly noted otherwise.
Dynamic range
The ratio between the loudest and quietest signals a system can handle. The healthy human ear has a dynamic range of about 120 dB (from threshold of hearing at 0 dB SPL to threshold of pain near 120 dB SPL). Sound‑level meters typically cover 25 – 130 dB SPL.
E
Equal‑loudness contour
A curve on a frequency‑vs‑SPL plot showing combinations of frequency and SPL that are perceived as equally loud. Standardized in ISO 226:2003 and originally measured by Fletcher and Munson in 1933. The curves explain why A‑weighting is shaped the way it is and why it works best at moderate listening levels.
Exchange rate
The number of decibels by which the safe exposure time halves. NIOSH and most international standards use 3 dB (energy‑equivalent: every 3 dB doubles the acoustic energy). OSHA uses 5 dB for historical reasons. The exchange rate is one of the largest practical differences between the OSHA PEL and the NIOSH REL.
F
FFT (Fast Fourier Transform)
An efficient algorithm for transforming a time‑domain signal into a frequency‑domain representation. The visualizer in our meter performs an FFT on each window of microphone samples and displays the magnitude of each frequency bin. FFT bin resolution equals sample_rate / FFT_size — for a 48 kHz signal with 2048 samples per window, each bin spans 23.4 Hz.
Free field
An idealized acoustic environment with no reflective surfaces — sound propagates outward from the source unimpeded. Approximated by anechoic chambers and (less perfectly) by open outdoor spaces. In free field, the inverse‑square law applies: doubling the distance from a point source reduces the SPL by 6 dB.
Frequency
The number of complete oscillations per second in a periodic signal, measured in hertz (Hz). The audible range for healthy young humans is 20 Hz to 20 kHz. Frequencies below 20 Hz are infrasound; above 20 kHz, ultrasound.
H
Hair cell
A sensory cell in the cochlea that converts mechanical motion of the basilar membrane into neural signals. Mammals have ~15,000 in each ear: 3,500 inner hair cells (the actual transducers) and 12,000 outer hair cells (mechanical amplifiers). Hair cells damaged by noise do not regenerate.
Hearing Conservation Program (HCP)
The set of activities required by occupational noise standards once exposure exceeds the action level. Includes monitoring, audiometric testing, hearing protectors, training, and recordkeeping. Full breakdown on the workplace standards page.
Hertz (Hz)
The SI unit of frequency: cycles per second. 1 Hz = one cycle per second; 1 kHz = 1,000; 20 kHz = 20,000.
Hyperacusis
Reduced tolerance to ordinary sound levels — sounds that other people find comfortable feel uncomfortably loud. Often associated with tinnitus and noise‑induced damage; sometimes a symptom of underlying neurological conditions.
I
IEC 61672‑1
The international standard specifying the performance and calibration of sound‑level meters. Defines Class 1 (precision, ±1 dB) and Class 2 (general purpose, ±2 dB) performance categories. Compliance instruments for occupational and environmental noise must be IEC 61672‑1 certified.
Impulse weighting (I)
A time weighting with a fast attack (35 ms) and slow decay (1.5 s), designed to capture brief loud events. Used for measurements of impulsive noise sources — gunshots, hammer strikes, balloon pops — and required by some standards for occupational measurements with prominent impulses.
ISO 1996
The international standard for environmental noise measurement. Two parts: 1996‑1 covers terminology and quantities; 1996‑2 covers field measurement procedures. The reference for community noise mapping, traffic noise studies, and most environmental impact assessments.
L
LAeq
The A‑weighted equivalent continuous sound pressure level — the
steady‑state SPL that would deliver the same total acoustic energy as
the actual variable signal over a measurement period. Formally:
LAeq,T = 10 × log10((1/T) × integral(10^(LA(t)/10) dt)). The
energy‑equivalent quantity used in nearly every modern occupational and
community noise standard.
Lden
A 24‑hour weighted average level used in EU community noise mapping. Penalizes evening noise (+5 dB) and night noise (+10 dB). The WHO 2018 guideline value for road traffic is 53 dB Lden outdoors.
Leq
The equivalent continuous sound pressure level — same definition as LAeq but without specifying a frequency weighting. Most commonly seen as LAeq (A‑weighted) or LCeq (C‑weighted).
L10, L50, L90
Statistical descriptors of variable noise. L10 is the level exceeded 10 % of the measurement period (a "peak typical" level), L50 is the median, L90 is the level exceeded 90 % of the time (a "background" level). Useful for community noise where Leq alone hides the distribution of events.
Lmax, Lpeak
Single‑event maxima. Lmax is the highest time‑weighted level during the measurement (using Fast, Slow, or Impulse weighting). Lpeak is the highest instantaneous unweighted sample value, which catches very brief peaks that time weighting would average out.
Logarithmic scale
A scale on which equal increments correspond to equal ratios rather than equal differences. The decibel scale is logarithmic; +10 dB is a 10× ratio in intensity at every level on the scale. Logarithmic scales compress wide ranges into manageable numbers and approximately match how humans perceive many sensory quantities (loudness, brightness, pitch).
M
Microphone (omnidirectional vs directional)
A transducer that converts air pressure into an electrical signal. An omnidirectional mic has equal sensitivity in all directions; a directional mic (cardioid, hypercardioid, shotgun) has reduced sensitivity off‑axis. Sound‑level meters typically use omnidirectional mics so that the reading depends on the actual SPL at the mic position, not on which way the mic is pointing. Phone microphones are often directional or have software beam‑forming, which is one reason calibration is per‑device.
N
NIHL (Noise‑Induced Hearing Loss)
Permanent sensorineural hearing loss caused by exposure to loud sound, either gradually over years (chronic NIHL) or from a single high‑level event (acoustic trauma). The second most common form of sensorineural hearing loss after age‑related decline. Almost entirely preventable; not currently reversible. Background on the hearing health page.
NIOSH REL (Recommended Exposure Limit)
NIOSH's recommended occupational noise exposure limit: 85 dBA over an 8‑hour TWA, with a 3 dB exchange rate. More protective than the OSHA PEL. The de facto international standard for non‑US employers and for US employers practicing best‑in‑class hearing conservation.
NRR (Noise Reduction Rating)
The labeled attenuation of a hearing protector, in decibels, derived from laboratory measurement under ANSI S3.19. Real‑world attenuation is typically lower than the labeled NRR; OSHA recommends derating foam plug NRR by 50 % and muff NRR by 25 % for practical exposure calculations. Modern fit‑testing systems measure attenuation in the user's actual ear.
O
OSHA PEL (Permissible Exposure Limit)
OSHA's regulated occupational noise exposure limit: 90 dBA over an 8‑hour TWA, with a 5 dB exchange rate. Codified at 29 CFR 1910.95; unchanged since 1983. Hearing‑conservation programs are mandatory at the 85 dBA action level. See workplace standards.
P
Pascal (Pa)
The SI unit of pressure: one newton per square meter. Atmospheric pressure is about 101,325 Pa. The threshold of hearing is about 20 µPa (0.00002 Pa). The threshold of pain is about 20 Pa.
Phon
A unit of loudness level defined as the SPL of a 1 kHz tone perceived as equally loud as the test tone. A 60 phon tone at any frequency sounds as loud as a 60 dB SPL tone at 1 kHz. Distinct from the sone, which measures perceived loudness on a linear scale.
R
Reverberation time (RT60)
The time it takes for sound in a room to decay by 60 dB after the source stops. Standardized as RT60. Lower reverberation time improves speech intelligibility; ANSI S12.60 recommends 0.6 s RT60 for classrooms. Living rooms typically have 0.3 – 0.6 s; cathedrals can exceed 5 s.
S
Sone
A linear unit of perceived loudness. 1 sone = 40 phon (a 40 dB SPL tone at 1 kHz); 2 sones = 50 phon (twice as loud); 4 sones = 60 phon. Used in equipment loudness specifications (HVAC, kitchen appliances) where a perceptually linear measure is more useful than the logarithmic dB.
Sound Pressure Level (SPL)
The decibel value of a sound pressure measurement, referenced to 20 µPa. Always reported in dB; usually further qualified by weighting (dBA / dBC / dBZ) and time weighting (Fast / Slow / Impulse).
STS (Standard Threshold Shift)
A change in audiometric thresholds of an average of 10 dB or more at 2, 3, and 4 kHz, measured against a baseline audiogram. STS is a trigger for required follow‑up under OSHA's hearing‑conservation program: refit hearing protection, retest, consider removing the worker from the high‑noise area.
T
Threshold of hearing
The lowest sound pressure level a healthy young human ear can detect at a given frequency. Defined by convention as 0 dB SPL at 1 kHz (corresponding to 20 µPa pressure). Higher (less sensitive) at very low and very high frequencies — see equal‑loudness contour.
Time weighting (Fast / Slow / Impulse)
Exponential averaging applied to the squared pressure signal before computing the SPL display. Fast uses a 125 ms time constant — the default for most measurements. Slow uses 1 s — for steady ambient. Impulse uses 35 ms attack and 1500 ms decay — for transient events.
Tinnitus
The perception of sound (ringing, buzzing, whining) in the absence of an external source. Often follows acoustic exposure; chronic tinnitus is associated with measurable hearing loss in most cases. The CDC estimates 50 million Americans live with tinnitus.
TWA (Time‑Weighted Average)
The 8‑hour energy‑equivalent of a worker's variable noise exposure
during a shift. Calculated from dose: TWA = 16.61 × log10(D/100) + 90
under OSHA's 5 dB exchange rate, or TWA = 10 × log10(D/100) + 85
under NIOSH's 3 dB rate. The compliance‑relevant quantity in every
occupational noise standard.
Z
Z‑weighting (zero / flat)
A flat frequency response across 10 Hz to 20 kHz. Replaces the older "linear" or "unweighted" terminology, which was inconsistent across manufacturers. Used for research, instrument verification, and any measurement where weighting would add unwanted distortion. Defined in IEC 61672‑1.
If a term you're looking for isn't here, it is likely either covered on the science page (technical concepts) or the hearing health page (medical and physiological concepts). The comparison chart ties many of these terms back to real‑world sound levels.