Reference

The audiophile glossary.

296 plain-English definitions for the terms that show up across every review on the site - planar magnetic, R-2R, PRaT, soundstage, comb filtering, MQA, and the rest. Tooltips on review pages pull from the same data, so the definition under "stealth magnets" on an Arya review is the same definition you\'ll find here, linkable and bookmarkable.

#

2-Pin Connector: A 0.78mm two-pin IEM cable connector that's mechanically stiffer than MMCX and offers better long-term reliability, though without the rotational freedom; standard on most mid-range and high-end IEMs.
4.4mm Pentaconn: A 5-conductor balanced headphone connector designed for portable audio, now standard on many high-end DAPs and portable amplifiers.
4N Copper: Copper of 99.99% purity ("four nines"); standard for premium audio cables. Higher purity grades (5N, 6N, 7N) carry diminishing audible returns and increasing marketing premium.

A

AAC: Advanced Audio Coding - the lossy codec used by Apple Music, YouTube, and most iOS Bluetooth devices. At 256 kbps, AAC is subjectively transparent for most listeners on most recordings; iOS encodes AAC over Bluetooth at lower rates.
Absorber: An acoustic treatment material (foam, fiberglass, mineral wool) that converts sound energy to heat, reducing reflections and decay time.
Absorption Coefficient: A material's efficiency at absorbing sound at a given frequency, from 0 (perfect reflection) to 1.0 (perfect absorption); varies strongly with frequency.
AC Line Filter: A passive LC network in the mains path of an audio component that attenuates high-frequency noise above ~10kHz before it reaches the transformer; standard in pro audio gear, increasingly common in consumer equipment.
Active Crossover: A crossover filter operating at line level before the power amplifiers, enabling bi- or tri-amping and providing more precise filter implementation.
AES/EBU: A professional digital audio standard electrically identical to SPDIF but using balanced XLR connectors with higher voltage for longer cable runs.
Air: The sense of spaciousness and extension above 10kHz; "airy" recordings reveal the acoustic space of the venue, and "airy" headphones resolve that space accurately.
AMT: Air Motion Transformer - a driver where a folded membrane squeezes air laterally, achieving very fast transient response and low distortion.
Analytical: A presentation that prioritizes detail retrieval and accuracy over harmonic richness; analytical systems reveal flaws clearly but may lack emotional engagement.
Anechoic Chamber: A room treated to absorb essentially all sound reflections, used for measuring speaker behaviour in isolation from room effects. Practical measurement increasingly uses quasi-anechoic windowed measurements in normal rooms.
Anti-Skate: A counterforce applied to the tonearm to cancel the bias force pulling the arm toward the center groove, keeping the stylus centered in the groove.
Antinode: A point in a standing wave where sound pressure is at maximum; a subwoofer or speaker placed at an antinode excites that room mode most strongly.
Apodizing Filter: A digital filter designed to remove pre-ringing artifacts introduced by recording-chain anti-aliasing filters, at the cost of some high-frequency rolloff.
aptX Adaptive: A variable-bitrate Bluetooth codec from Qualcomm that adjusts between 280-420kbps based on RF conditions; balances aptX HD's quality against the reliability of standard aptX.
aptX HD: Qualcomm's 576 kbps Bluetooth codec for 24-bit/48kHz audio. Widely supported on Android phones and audiophile receivers; lower-latency and more battery-efficient than LDAC at the cost of bit depth.
aptX Lossless: Qualcomm's claimed-lossless 1Mbps Bluetooth codec, available only on devices with Snapdragon Sound certification (chipset + firmware support on both ends). The first Bluetooth codec to claim bit-perfect CD-quality transmission.
ASIO: Audio Stream Input/Output - a low-latency audio driver standard for Windows that bypasses the Windows audio mixer for bit-perfect, sample-accurate playback.
Asynchronous USB: A USB audio mode where the DAC controls the data clock (not the computer), dramatically reducing jitter compared to adaptive or synchronous USB.
Authority: A system's ability to maintain control of the bass under demanding passages without compression, congestion, or loss of articulation. High-authority bass keeps the kick drum distinct from the bass guitar even at concert levels.
Auto-Bias: A tube amplifier circuit that automatically sets the operating bias point for each output tube as it ages, eliminating the manual bias adjustment that older tube amps required every few months.
Axial Mode: A room resonance involving only one dimension (length, width, or height); the strongest type, calculated as f = n x 343 / (2 x dimension_in_meters).

B

Baffle Step: A 6dB low-frequency boost that occurs as a driver transitions from radiating into half-space (near the baffle) to full-space, typically between 100-500Hz.
Balanced: A signal transmission method using two opposite-polarity signal lines plus a ground; noise induced on both lines is cancelled at the differential input.
Balanced Armature: A miniature driver used in IEMs where a metal armature pivots between magnets, allowing multiple drivers to be stacked in a very small shell.
Bass Extension: How low in frequency a system accurately reproduces sound; good bass extension means 20Hz output, not just 60Hz.
Bass Management: The crossover and level-routing logic that decides which frequencies go to which speakers in a multi-speaker system; standard in home-theatre receivers, increasingly common in stereo preamps with subwoofer outputs.
Bass Reflex: A speaker enclosure with a tuned port or slot that allows the rear wave to reinforce bass output, extending low-frequency response.
Bass Trap: An acoustic absorber designed to reduce the Q of low-frequency room modes; most effective in corners where axial, tangential, and oblique modes all converge.
Beaming: Progressive narrowing of a driver's radiation pattern as frequency rises, occurring when the wavelength becomes comparable to the driver diameter. Causes the high-frequency response to dim sharply off-axis.
Beryllium: An extremely lightweight and rigid metal used in high-end tweeter and headphone diaphragms for its exceptional stiffness-to-mass ratio.
Bi-Amping: Driving the low-frequency and high-frequency sections of a speaker with separate amplifier channels, requiring either passive or active crossover filtering.
Bi-Amplification: See Bi-Amping - separate amplifier channels for low and high frequency drivers of a loudspeaker.
Bias Current: The constant quiescent current flowing through an output stage that determines the operating class and sets the thermal operating point.
Binaural: Two-channel audio processed with HRTF to simulate 3D sound over headphones, placing instruments outside the head rather than between the ears.
Bit Depth: The number of bits per sample - determines dynamic range (approximately 6dB per bit, so 16-bit gives ~96dB, 24-bit gives ~144dB).
Bit-Perfect: Playback where the audio data arrives at the DAC unchanged from the file - no sample rate conversion, no bit depth change, no digital volume attenuation.
Bite: An incisive character in the upper midrange or lower treble that adds energy and presence to instruments like brass, electric guitar, and snare drum. Too much bite is fatiguing; just enough is exciting.
Bloom: A pleasing expansion of a sustained note, particularly in the lower midrange and upper bass, that gives instruments natural weight and decay. Excessive bloom becomes "thickness"; insufficient bloom sounds "lean."
Boogie Factor: The subjective quality of how effectively a system communicates rhythmic energy and groove; closely related to PRaT but more focused on overall musical engagement than pure timing accuracy.
Break-in: Synonym for burn-in; the supposed period during which new gear "improves" with playback time. Most documented changes happen in the first 24 hours; claims of "200-hour break-in" are typically the listener adjusting to the sound.
Brick Wall Filter: A very steep anti-aliasing filter that completely eliminates all content above the Nyquist frequency but introduces pre-ringing in the time domain.
Brittle: A thin, fragile-sounding treble with audible grain, often paired with insufficient body in the upper midrange. A common failure mode of inexpensive digital sources and bright headphones.
Burn-in: The hypothesised mechanical or electrical settling-in period for new audio components. Real for moving-cone speakers and woven-cone drivers (first 5-50 hours); largely auditory acclimation rather than physical change for solid-state amplifiers and DACs.
Butterworth: A crossover (or filter) topology providing the flattest possible passband response (-3dB at cutoff), used for its gentle rolloff characteristic.

C

Cans: Audiophile slang for headphones, particularly over-ear models. Originates from the metal "cans" that held early aviation and broadcast headphones.
Cantilever: The tiny arm that holds the stylus tip and transmits groove vibration to the cartridge generator; its material and geometry affect compliance and trackability.
Cascode: A two-transistor amplifier topology stacking a common-emitter (or source) stage under a common-base (or gate) stage. Improves bandwidth and reduces Miller capacitance, common in high-performance preamp input stages.
Cathode Follower: A tube circuit where the output is taken from the cathode rather than the plate; provides voltage gain less than 1 but very low output impedance, often used as a buffer at the output of a tube preamp.
Channel Separation: The isolation between left and right channels, measured in dB; insufficient separation causes stereo crosstalk that narrows the soundstage.
Class A: An amplifier topology where the output transistors or tubes conduct current at all times, eliminating crossover distortion at the expense of significant heat and inefficiency.
Class AB: The most common amplifier class, biasing the output stage into Class A for small signals and transitioning toward Class B at higher levels for better efficiency.
Class D: A switching amplifier using pulse-width modulation to achieve 90%+ efficiency; modern Class D designs rival linear amplifiers in audio performance.
Closed-Back: Headphones with sealed ear cups providing isolation from ambient noise; the trapped air behind the driver affects bass tuning and often produces a more intimate sound.
Coaxial Driver: A speaker design mounting a tweeter in the center of a woofer, creating a coincident point source that improves off-axis coherence.
Coloration: Any consistent deviation from accurate reproduction that imposes the system's own character on recordings; can be pleasant (euphonic) or fatiguing.
Comb Filtering: The frequency-response pattern produced when a direct signal mixes with a delayed reflection - alternating peaks and notches every few hundred Hz, audible as hollowness and coloration in the affected band.
Common Mode Rejection: The ability of a balanced circuit to cancel noise that appears equally (in common) on both signal lines; measured in dB, higher is better.
Comply Tips: Memory-foam IEM tips that expand to fill the ear canal, producing a deeper seal and stronger bass response than silicone tips - at the cost of some treble extension and a shorter usable life (1-3 months per pair).
Confirmation Bias: The cognitive tendency to weight evidence supporting an existing belief and discount evidence against it. In audio reviewing, leads to "the new DAC sounds better" once the listener has paid for it, regardless of measurable change.
Constant Directivity: A speaker design goal where the radiation pattern stays the same width across the audio band; produces consistent off-axis tonality and is the principle behind waveguide-loaded tweeters and horn-loaded systems.
Convolution: A mathematical operation used to apply complex filter shapes (HRTFs, room corrections, speaker impulse responses) to audio in the digital domain.
Critically Damped: A system tuned so that after excitation it returns to rest in the shortest time without oscillation - the ideal for accurate transient reproduction.
Crossfeed: A DSP process that mixes a small, slightly delayed and filtered portion of each channel into the other, simulating the natural crosstalk of loudspeaker listening.
Crossover: A network of filters that divides the audio signal into frequency bands before each reaches its appropriate driver - passive (in the speaker) or active (before the amp).
Crossover Distortion: Distortion in Class B/AB amplifiers at the zero-crossing point where the two output devices transition, producing a characteristic "zipper" artifact.
Crossover Slope: The rate of attenuation beyond the crossover frequency, measured in dB/octave; common slopes are 6dB (1st order), 12dB (2nd order), 18dB (3rd order).
Crosstalk: Leakage of signal from one channel to the other; in headphones, some deliberate crossfeed is beneficial, but unintentional crosstalk reduces stereo separation.
Current Drive: Driving a speaker or headphone with a high-impedance current source rather than a low-impedance voltage source, which can change the effective frequency response.

D

DAC: Digital-to-Analog Converter - a device that translates binary audio data into an analog electrical signal that can be amplified and heard.
Damping Factor: The ratio of load impedance to amplifier output impedance; a higher damping factor gives the amplifier tighter control over bass transients.
DC Blocker: A device placed in series with the AC mains that removes any DC offset on the power line, preventing toroidal transformers in audio equipment from saturating and humming. Cheap fix for "buzzing transformer" symptoms.
DC Offset: An unwanted DC voltage at an amplifier's output; can cause speaker damage, bias the woofer cone off-center, and create turn-on/off clicks.
DDC: Digital-to-Digital Converter - a device that converts one digital audio interface to another (USB to I²S, USB to S/PDIF) while reclocking the data with a low-jitter local oscillator, often as a "bridge" between a computer and a digital input on a DAC.
Decibel: A logarithmic ratio unit; 3dB represents a doubling of power, 6dB a doubling of voltage or pressure, and 10dB a perceived doubling of loudness.
Delta-Sigma: The dominant DAC architecture today, using high oversampling and noise shaping to push quantization noise above the audible range.
Diaphragm: The vibrating membrane in a transducer that converts between electrical energy and acoustic waves; its mass, stiffness, and damping determine driver character.
Diffraction: Sound bending around objects or edges, causing interference patterns; speaker cabinet edges cause treble-band diffraction that affects off-axis response.
Diffusion: An acoustic treatment that scatters reflections in multiple directions, reducing the audibility of early reflections without removing acoustic energy.
Digital Filter: An algorithm applied before DAC conversion that manages the transition from audio band to alias zone; choices include brick-wall, minimum-phase, and apodizing designs.
Dipole: A loudspeaker radiating equally from front and back, producing a figure-8 polar pattern; open-back headphones are acoustic dipoles.
Dipole Cancellation: The acoustic short-circuit that occurs at low frequencies in open-baffle speakers where front and rear waves cancel; requires large baffles or equalization to extend bass.
Directivity Index: The decibel ratio between a speaker's on-axis output and its total radiated sound power. Higher DI means more directional radiation; consistent DI across frequency is the marker of a well-designed speaker.
Dithering: Adding controlled low-level noise to audio data when reducing bit depth, which randomizes quantization error and prevents the audible distortion of truncation.
Double-Blind Test: A listening test where neither the listener nor the test administrator knows which option is playing - the gold standard for separating real audible differences from cognitive bias.
Driver Flex: A faint crackling or popping sound when inserting closed-back IEMs, caused by the seal trapping air pressure against the dynamic driver. Mostly harmless; relieve by inserting slowly or with the IEM partially seated.
DSD: Direct Stream Digital - a 1-bit, high-sample-rate audio format used on SACD, encoding audio through rapid single-bit switching rather than multi-bit PCM words.
DSP: Digital Signal Processing - manipulation of audio in the digital domain for room correction, crossover implementation, EQ, and delay alignment.
Dynamic Driver: The most common transducer type, using a voice coil in a magnetic gap to push a cone or dome diaphragm - the same principle as a traditional loudspeaker.
Dynamic Range: The decibel span between a system's maximum undistorted output and its noise floor; 16-bit audio has ~96dB, 24-bit has ~144dB of theoretical range.

E

Ear Gain: The combined acoustic gain of the outer ear and ear canal (typically 8-15dB centred near 3kHz). Headphone manufacturers tune frequency response to compensate for the fact that the headphone bypasses some natural ear gain.
Edge: A slightly forward, lean character in the treble that can read as either "detailed" (positive) or "etched/harsh" (negative) depending on the listener and recording. Distinct from sibilance, which is band-specific.
Efficiency: A speaker's acoustic output power relative to its electrical input power; a typical home speaker converts only 0.5-2% of electrical power to acoustic power.
Electrostatic: A driver using an ultra-thin membrane suspended between two charged stator plates; requires a high-voltage energizer and produces vanishingly low distortion.
Endgame: Audiophile slang for a piece of gear so completely satisfying that the listener stops shopping for upgrades - usually a flagship headphone, DAC, or amplifier. Often used aspirationally rather than literally.
Equalization: The boosting or cutting of specific frequency ranges to correct tonal imbalances, compensate for room acoustics, or achieve a target response.
Etched: An overly defined, sharp-edged character in the treble that exaggerates micro-detail at the expense of natural timbre. Common in systems that maximize resolution measurements without ear-tuning the implementation.
Euphonic: Pleasing to the ear through harmonically benign distortion - typically even-order harmonics that enhance perceived richness and warmth.
Expectation Bias: A perceptual distortion where what the listener anticipates hearing influences what they actually perceive; closely related to placebo and confirmation bias but specifically about prior expectation.

F

Ferrite Bead: A small ring of ferrite material clamped around a cable that suppresses high-frequency RF noise on the cable without affecting the audio signal. Effective on USB and digital interconnects; often built into the connector molding.
Ferrite Choke: A larger ferrite assembly attached to a power cable to filter common-mode noise; the cylindrical lumps on the power cords of laptops and printers are ferrite chokes preventing the cable from radiating switching noise.
First Reflection: The earliest room reflection to reach the listening position after the direct sound from the speaker; typically arrives 1-10ms later and can smear imaging.
FLAC: Free Lossless Audio Codec - an open-source lossless compression format that reduces file size by 40-60% with zero degradation to the audio data.
Fletcher-Munson: Equal-loudness contours (ISO 226) showing that human hearing is most sensitive around 3-4kHz and requires much more bass and treble at low volumes.
Forgiving: A tonal character that softens harsh or compressed recordings, making poor source material more listenable at the cost of some accuracy on well-recorded material. The opposite of "revealing" or "ruthless."
Forward: A tonal character with elevated upper midrange or lower treble that pushes vocalists and lead instruments ahead of the mix; can sound exciting or fatiguing.
FPGA: Field-Programmable Gate Array - a reconfigurable chip used in high-end DACs to implement custom digital filter algorithms outside the standard silicon.
Free Air Resonance: The natural resonant frequency of a driver's suspension system when operated in free space; denoted Fs, a fundamental Thiele-Small parameter.
Frequency Response: A graph showing output amplitude vs. frequency - the most fundamental measurement of any audio component's tonal character.

G

Gain: The multiplication factor applied to a signal by an amplifier, expressed in dB; proper gain staging is critical for minimizing noise.
Galvanic Isolation: Electrical isolation between circuit sections via transformer, optocoupler, or capacitor coupling - allows signal transfer while blocking DC and breaking ground loops.
Glare: An unpleasant emphasis in the upper midrange (~2-4kHz) that makes voices and instruments sound strident or pushed forward. Often a sign of crossover error or an unhappy driver-amplifier interaction.
Ground Loop: An unwanted current path through the ground conductor between two devices, creating a 50/60Hz hum that's audible through the audio chain. Cured by galvanic isolation, ground lifts (carefully), or routing every device through a single grounded point.
Group Delay: The variation in signal propagation delay across frequency; flat group delay means all frequencies arrive at the listener simultaneously.

H

Harman Curve: A headphone frequency response target developed by Sean Olive at Harman through large-scale listener preference research; adds bass shelf over flat response.
Harmonic Distortion: Distortion products at integer multiples of the input frequency; even-order harmonics (2nd, 4th) are generally less audible than odd-order (3rd, 5th).
Head Tracking: Real-time HRTF adjustment based on detected head orientation, maintaining stable virtual soundstage positioning as the listener moves.
Headroom: The decibel margin between the loudest expected signal and an amplifier's clipping point. 10-20dB of headroom is generally needed for unclipped reproduction of dynamic recordings at realistic listening levels.
Hi-Res Audio: A marketing certification (JAS / Hi-Res Music) for audio with sample rates ≥ 96kHz and bit depths ≥ 24-bit. The audible benefit over 16-bit/44.1kHz is debated; the legitimate benefit is reduced quantization error in the mixing chain.
High-Gain Mode: An amplifier setting that increases voltage gain, intended for low-sensitivity headphones (e.g., HIFIMAN HE-6, Susvara) or low-output sources. Using high gain with sensitive IEMs raises the noise floor unnecessarily.
Holographic: A three-dimensional imaging quality where instruments occupy distinct positions in width, depth, and height, with believable acoustic space around them. The headphone audiophile's holy grail.
Horn Loading: Coupling a driver to a flared horn to dramatically increase sensitivity and control directivity - common in professional and high-efficiency home speakers.
HRTF: Head-Related Transfer Function - the acoustic filter created by your head, pinnae, and torso that enables 3D localization; unique to each individual.
Hum: Audible low-frequency interference (usually 50Hz, 60Hz, or harmonics) caused by ground loops, induced AC fields, or transformer leakage. The most common form of "noise" in audio chains and often the easiest to track down.

I

I²S: Inter-IC Sound - a digital audio interface standard using separate lines for data, bit clock, and word clock. Used inside DACs natively; some high-end DACs expose I²S via HDMI for connection to dedicated transports.
Imaging: The ability to place individual instruments in precise, stable positions within the soundstage - good imaging means you can "point" to a violin in the mix.
IMD: Intermodulation Distortion - new frequencies created when two or more tones interact non-linearly; IMD products are harmonically unrelated to the original tones and sound harsh.
Impedance: The total opposition (resistance + reactance) a speaker or headphone presents to the driving current, measured in ohms and varying with frequency.
Impedance Hump: A peak in headphone impedance at the driver's mechanical resonance frequency; with a high-output-impedance amplifier, this produces a bass emphasis.
Impedance Matching: Choosing amplifier and load impedances to optimize power transfer, frequency response, and stability; less critical for modern low-impedance voltage-source amps.
Impulse Response: How a system responds to a single sample of audio; reveals ringing, pre-ringing, and time-domain behavior not visible in a steady-state frequency response.
Inner Detail: The lowest-level sonic information in a recording - fingers on strings, breath between phrases, the sound of the room - that reveals what a system can truly resolve at low volumes.
Input Impedance: The impedance seen looking into an amplifier's input; must be much higher than the source output impedance to avoid signal loss and frequency response changes.
Interaural Level Difference: The difference in level between your two ears for a sound source at a given angle; the brain uses ILD along with ITD to localize sounds.
Interaural Time Difference: The difference in arrival time between your two ears for a sound source to one side; the brain uses ITD for low-frequency localization.
Inverse Square Law: The principle that SPL drops by 6dB every time the listening distance doubles in a free field; a speaker 4m away is 6dB quieter than at 2m.

J

Jitter: Timing irregularities in the digital audio clock that cause correlated noise sidebands, audible as a subtle smearing or loss of focus.

L

Laid-Back: A tonal character with a slight upper-midrange recession that pushes the presentation back, creating a relaxed, spacious quality but potentially lacking immediacy.
Layering: The system's ability to render multiple instruments at different perceived depths in the soundstage, rather than collapsing them onto a single plane. Strong layering reveals the spatial structure of a recording.
LC3: Low Complexity Communications Codec - the mandatory codec for Bluetooth LE Audio. More efficient than SBC at low bitrates (≥160 kbps gives near-AAC quality) and supports multi-stream and broadcast applications.
LDAC: Sony's proprietary Bluetooth audio codec supporting up to 990 kbps - the highest available over Bluetooth. Quality depends heavily on signal strength; in clean RF environments LDAC at 990kbps is subjectively transparent.
LE Audio: The Bluetooth standard introduced in 2020 that adds the LC3 codec, multi-stream support, Auracast broadcast, and Hearing Aid Profile - the biggest functional upgrade to Bluetooth audio since the original spec.
Level Matching: Setting two pieces of gear to identical SPL output before comparing them, typically with a measurement microphone and 0.1dB precision. Without level matching, the louder option will reliably be preferred regardless of any actual quality difference.
Linear Phase: A filter applying equal delay to all frequencies, preserving phase relationships but causing symmetrical pre- and post-ringing around transients.
Linkwitz-Riley: A crossover alignment where the outputs sum to a flat amplitude response with correct phase relationship, requiring both drivers to be at -6dB at the crossover point.
Liquid: A smooth, fluid, fatigue-free presentation in the midrange and treble, free from grain and edge artifacts. Often described as "analog" sounding even on digital sources.
Listening Triangle: The geometric arrangement where the two speakers and the listening position form an equilateral triangle (or close to it); the standard starting point for two-channel speaker placement.
Listening Window: The averaged frequency response across a ±30° horizontal and ±10° vertical arc in front of a speaker - one of the key curves in a Spinorama, representing what listeners actually hear in a normal room.
Low-Gain Mode: The default amplifier setting on amps with switchable gain; provides cleaner output and lower noise floor for sensitive headphones and IEMs. Always start in low-gain mode unless you can't reach acceptable listening levels.

M

Macro-Dynamics: The large-scale dynamic contrasts in music - from soft passages to loud climaxes; limited by amplifier clipping and driver excursion limits.
Magnetic Saturation: The condition where an iron core reaches maximum magnetization and cannot store additional flux; saturation in speaker motors causes distortion.
Micro-Detail: The very lowest-level sonic information in a recording - the subtle sounds like breath, bow rosin, and room ambience that reveal what a system can truly resolve.
Micro-Dynamics: Small-scale dynamic gradations within individual notes and phrases; limited by noise floor and the ability to resolve low-level detail.
Midrange: The frequency range from approximately 250Hz to 5kHz where most musical information, vocals, and instrument fundamentals reside.
Midrange Driver: A driver optimized for the critical 500Hz to 5kHz band where most musical information and vocal intelligibility resides.
Minimum Impedance: The lowest impedance a speaker reaches at any frequency - this represents the worst-case current demand on the amplifier.
Minimum Phase: A filter (or system) with the minimum possible phase shift for its amplitude response, concentrating any time smearing after the transient rather than before.
MMCX: A miniature coaxial connector used to attach removable cables to IEMs; rotates 360°, which is both a feature (cable freedom) and a wear concern (intermittent contact over time).
MOSFET: Metal-Oxide-Semiconductor Field-Effect Transistor - a transistor type used in audio output stages, valued for smooth overload behavior and high current capacity.
Moving Coil: A phono cartridge type with a very tiny coil attached to the cantilever, producing extremely low output (0.2-1mV) but with very low moving mass.
Moving Magnet: The most common cartridge type with a magnet on the cantilever and fixed coils; produces higher output (3-5mV) and uses standard phono stage gain.
MQA: Master Quality Authenticated - a proprietary lossy compression format claiming to preserve master-quality information in a smaller file. Critics argue it adds distortion and serves rights-holder DRM more than fidelity; Tidal removed MQA from its catalogue in 2024.
Mu: The voltage amplification factor (μ) of a vacuum tube - the maximum theoretical voltage gain the tube can produce in a triode-connected configuration. Higher mu (e.g., 12AX7's μ=100) means more gain but typically higher noise.
Musical: A subjective quality where a system seems to convey the emotional content of music effectively, often (though not always) involving some euphonic coloration.

N

Nearfield: Listening at very close range (typically 0.5-1.5m) where the direct sound from the speakers overwhelms room reflections; standard for studio monitoring.
Negative Feedback: Feeding a portion of the output signal back to the input inverted, reducing distortion, output impedance, and frequency response errors.
Neutrality: A frequency response and tonal character that imposes no consistent bias toward warmth, brightness, or any particular frequency range.
Node: A point in a standing wave where sound pressure is at minimum; sitting at a node at a room mode frequency means you hear almost no bass at that frequency.
Noise Shaping: A DSP technique that moves quantization noise from the audible band into higher, less perceptible frequencies.
Nominal Impedance: The rated impedance value for a loudspeaker - typically its average across the frequency range, often lower than the stated value at resonance.
Nominal Impedance (Speaker): The manufacturer's rated load impedance, typically the average across the audio band; actual impedance varies significantly with frequency.
NOS DAC: Non-Oversampling DAC - processes audio at the original sample rate with no digital filtering, producing a different frequency/time-domain characteristic than conventional DACs.
Notch Filter: A very narrow-band filter that deeply attenuates a specific frequency; commonly used to suppress driver resonances or diffraction peaks.

O

Oblique Mode: A room resonance involving all three dimensions; 6dB weaker than axial modes; higher-frequency oblique modes become very dense and diffuse.
OCC: Ohno Continuous Casting - a copper or silver purification process that produces very long single-crystal grains, potentially reducing grain boundary resistance.
OCXO: Oven-Controlled Crystal Oscillator - a precision clock kept at a constant temperature in a heated enclosure, achieving sub-100ppb frequency stability for low-jitter digital audio applications.
OFC: Oxygen-Free Copper - copper refined to remove dissolved oxygen, claimed to improve conductivity and reduce long-term oxidation. Real but small effect on bulk conductivity; mostly a marketing distinction in audio cables.
Open-Back: Headphones with perforated or meshed ear cups allowing free air exchange; produces a more natural, spacious presentation with no isolation from ambient sound.
Orthodynamic: An alternative term for planar magnetic headphones, emphasizing that the magnetic field acts perpendicular to the diaphragm across its full surface.
OTL: Output Transformerless - a tube amplifier design that connects tubes directly to the load, avoiding transformer colorations but requiring many tubes in parallel.
Out-of-Head: A headphone listening experience where the stereo image escapes the area between the ears and extends into the surrounding space, approaching the externalisation of loudspeaker listening. Rare without DSP head tracking.
Output Impedance: The impedance looking back into an amplifier's output terminals - a high output impedance interacts with headphone impedance curves to alter frequency response.
Output Stage: The final amplification block of an amp that directly drives the load (speakers or headphones), supplying the current and voltage the load demands. Unrelated to "stage" / "soundstage" - that's a listening term about perceived spatial width and depth; this is a circuit block.
Overdamped: Excessive damping causing bass rolloff to begin at a higher frequency than optimal; sounds "dry" or "thin" in the bass.
Oversampling: Processing audio at a multiple of the original sample rate before conversion, improving noise performance and relaxing the requirements on the analog output filter.

P

PCM: Pulse-Code Modulation - the standard digital audio format, encoding amplitude as binary integers at fixed time intervals (e.g., 16-bit/44.1kHz for CD).
Pentode: A five-element vacuum tube producing higher gain than a triode but with more complex distortion products; often ultralinear-connected in audio amplifiers.
Phase Coherence: The alignment of phase relationships between drivers, channels, or frequency bands; phase coherence determines depth and focus in the soundstage.
Phase Response: How a system shifts the relative timing of different frequencies; minimum-phase systems have the least phase shift for a given frequency response.
Phono Stage: A preamplifier for turntable cartridges that applies RIAA equalization and provides 40-60dB of gain to bring cartridge output to line level.
Pinna Gain: The acoustic amplification provided by the outer ear (pinna) in the 2-7kHz region, caused by the ear's shape acting as a resonant horn. Headphone tuning must account for the pinna gain that a speaker would otherwise produce.
Pinpoint Imaging: Precise, focused, stable localisation of individual instruments and voices within the soundstage; the opposite of "diffuse" or "smeared" imaging where the source location wanders or blurs.
Placebo Effect: A perceived improvement in sound caused by expectation alone, with no actual change in the audio signal. Real, measurable, and the reason serious audio comparisons require blind testing to isolate from genuine audible differences.
Planar Magnetic: A driver using a thin membrane with embedded conductors suspended between magnets, producing sound from the entire surface for very low distortion.
Post-Ringing: Oscillation following a transient in digital filter implementations; present in minimum-phase filters and always in speakers and room acoustics.
Power Bandwidth: The frequency range over which an amplifier can deliver its rated output power; limited at low frequencies by output coupling and at high frequencies by slew rate.
Power Conditioning: Any treatment of the AC mains supply before it reaches audio equipment - filtering noise, blocking DC, surge protection, or balancing voltage. Benefits range from "obvious" to "imaginary" depending on the listener and the implementation.
Power Regenerator: A class of power conditioner that completely rebuilds the AC waveform from DC, producing a clean sine wave regardless of the input quality. Effective but power-limited (typically 1000-1500W), expensive, and only relevant where mains are demonstrably bad.
PRaT: Pace, Rhythm, and Timing - the subjective sense of how confidently a system communicates the timing and groove of music; high-PRaT components make you tap your foot, low-PRaT components flatten rhythmic energy.
Pre-Out: A line-level output on an integrated amplifier or receiver that taps the signal after the preamplifier stage but before the power amplifier, allowing a separate power amp to be added without bypassing the integrated's volume control.
Pre-Ringing: A temporal artifact of linear-phase digital filters where a low-level oscillation precedes a transient - some listeners find this more disturbing than post-ringing.
Punch: Bass impact in the 60-150Hz region - the chest-thump of a kick drum or the snap of a slap-bass note. Distinct from slam, which extends lower; punch is about the leading edge of bass transients, not the depth.
Push-Pull: An amplifier topology using two matched output devices - one handling positive, one negative half-cycles - which cancels even-order distortion.

Q

Quantization Error: The rounding error introduced by representing a continuous analog signal as discrete digital levels; audible as distortion when the error is correlated with the signal.

R

R-2R: A DAC architecture using a resistor ladder network with only two resistance values, often associated with a more "natural" tonal character than delta-sigma designs.
Recessed: A perceived dip in a frequency region (commonly the upper midrange or lower treble) that pulls instruments backward in the soundstage and softens overall presence.
Reclocking: Regenerating a digital audio signal's timing reference using a low-jitter local clock, effectively stripping the original clock's jitter from the data.
Refraction: The bending of sound waves as they travel through media of differing density; temperature gradients in a listening room cause subtle sound path changes.
Resolution: A system's ability to retrieve and reproduce fine detail in the recording; high resolution reveals micro-dynamics, spatial cues, and timbral nuance.
Resonant Frequency: The frequency at which a mechanical or electrical system oscillates with minimum energy input; for a speaker driver, this is where cone excursion peaks.
RIAA: Recording Industry Association of America - a standardized equalization curve applied during vinyl cutting and reversed on playback to restore flat frequency response.
Ribbon Driver: A transducer where a thin corrugated metal ribbon suspended in a magnetic field acts as both conductor and diaphragm, known for extreme high-frequency extension.
Room Correction: DSP-based EQ that flattens a speaker's in-room frequency response, primarily addressing the bass region below the Schroeder frequency where room modes dominate. Examples: Dirac Live, Audyssey, Trinnov, Anthem ARC.
Room Gain: The acoustic bass reinforcement caused by room boundary reflections; sealed rooms can provide up to 6dB of gain below 100Hz.
Room Mode: A standing wave resonance at a frequency determined by room dimensions; axial modes (one dimension) are the strongest and most audible.
Roon: A music management and playback platform with advanced DSP, network streaming to Roon Ready endpoints, and integration with streaming services like Qobuz and Tidal.
Roon Ready: A Roon Labs certification for streaming endpoints that natively support Roon's RAAT protocol for bit-perfect transport with metadata. Roon Ready devices appear in Roon's zone list automatically.
Roon Tested: A lower-tier Roon certification confirming that a device works with Roon via standard protocols (AirPlay, Chromecast, Squeezebox) without native RAAT support.
RT60: Reverberation Time 60 - the time in seconds for a room's reverberant energy to decay by 60dB after the source stops; target is 0.3-0.5s for critical listening.

S

Sample Rate: How many audio samples are captured per second; 44.1kHz captures frequencies up to 22.05kHz (Nyquist limit), well above the 20kHz hearing limit.
SBC: Sub-Band Coding - the mandatory baseline Bluetooth audio codec, supported by every Bluetooth device. Quality is acceptable at 328 kbps but most implementations default to 200-256 kbps which is audibly compressed.
SBIR: Speaker Boundary Interference Response - bass cancellation caused by destructive interference between direct sound and floor, wall, or ceiling reflections near the speaker.
Schroeder Frequency: The transition frequency above which room modes become so dense they overlap, creating smooth but reverberant sound; approximately 2000 x sqrt(RT60/Volume).
Sealed Enclosure: A completely sealed speaker cabinet; produces a gentle 12dB/octave rolloff below resonance and tighter, more controlled bass than ported designs.
Sensitivity: The output sound pressure level for a standardized input - typically dBSPL at 1W/1m for speakers, or dBSPL at 1mW or 1V for headphones.
Sensitivity (Headphone): The output sound pressure level for a standardized input power (1mW) or voltage (1V); high-sensitivity headphones (>100dBSPL/mW) can be driven by portable devices.
Sibilance: Excessive "s," "sh," and "ts" consonants caused by a peak in the 6-10kHz region; can be a characteristic of the headphone, the recording, or a bright source.
Sighted Listening: Comparison done with full knowledge of which gear is playing; the default mode of audio reviewing. Useful for forming opinions but cannot, by itself, prove audible differences exist - that's what blind testing is for.
SINAD: Signal-to-Noise And Distortion - a single-number audio quality metric combining SNR and THD+N; DAC rankings by SINAD are published at AudioScienceReview.
Single-Ended: An amplifier configuration using one output device for the complete audio waveform; produces even-order harmonic distortion considered "euphonic" by many.
Slam: The visceral impact of low-frequency transients - kick drums, bass drops, tympani - felt as much as heard. High-slam systems combine deep extension with fast transient attack and high SPL capability without compression.
Slew Rate: The maximum rate of output voltage change (V/microsecond); inadequate slew rate causes high-frequency distortion when the amplifier cannot keep up with the signal.
Snap: The clarity and speed of a transient attack, particularly on percussion - a snare hit, hand clap, or rim shot. Snap requires both fast driver response and the absence of ringing in the recovery.
SNR: Signal-to-Noise Ratio - the decibel gap between the maximum signal level and the noise floor; higher is better.
Soft Start: A circuit that ramps mains current up gradually when an amplifier powers on, preventing the inrush current that would otherwise blow fuses or stress capacitors. Almost universal on amplifiers above 100W per channel.
Soundstage: The perceived three-dimensional acoustic space in a stereo recording - width beyond the speakers, depth front-to-back, and sometimes height information.
SPDIF: Sony/Philips Digital Interface - a coaxial (RCA) or optical (TOSLINK) serial digital audio interface supporting up to 24-bit/192kHz stereo PCM or Dolby/DTS.
Spider: The flexible, corrugated surround that centers and restores the voice coil assembly; its stiffness is the primary source of a driver's mechanical compliance.
SpinFit: A third-party silicone IEM tip brand with a rotating-nozzle design that follows the ear canal's natural curve, often improving seal and comfort over stock straight tips.
Spinorama: A standardized speaker measurement protocol (CTA-2034) that captures on-axis response, listening window, early reflections, and total power across a 360° sphere; the foundation of Sean Olive's preference research.
SPL: Sound Pressure Level - measured in dBSPL relative to 20 micropascals (the threshold of hearing); conversation is ~60dB, live rock is ~110dB, pain threshold is ~130dB.
Spotify Connect: Spotify's native streaming protocol that pushes audio directly from the Spotify cloud to compatible network speakers and DACs, bypassing the controlling phone's audio output.
Stage: Short for soundstage; the perceived three-dimensional acoustic space of a stereo recording. Often used to describe headphone presentation specifically ("the Arya has a deep stage").
Standing Wave: A stationary interference pattern formed when a sound wave and its reflection interact; produces fixed pressure peaks (antinodes) and nulls (nodes).
Stealth Magnets: HIFIMAN's asymmetric magnet geometry on their planar magnetic drivers, designed to present a more acoustically transparent surface to the diaphragm and reduce wave reflections that would otherwise distort the response.
Stereo Imaging: The ability of a stereo system to create the illusion of a soundstage with specific, stable positions for each instrument and voice.
Stylus: The diamond-tipped needle assembly that contacts the record groove and converts mechanical vibration into electrical signal via the cartridge.
Sub-Bass: Frequencies below approximately 60Hz; felt as much as heard, sub-bass conveys pipe organ fundamentals, kick drum body, and concert hall size.
Subwoofer Integration: The process of seamlessly blending a subwoofer's output with the main speakers in level, phase, and crossover frequency. Done well, the sub is inaudible as a separate source; done poorly, it's a humming box in the corner.
Susvara: HiFiMAN's flagship planar magnetic headphone, famous for requiring enormous amplifier power (at least 1-2W) due to its extremely low sensitivity of ~60dBSPL/mW.
Sweet Spot: The narrow listening position where stereo imaging is most accurate - typically the apex of an equilateral triangle with the two speakers. Outside the sweet spot, the image collapses toward whichever speaker is closer.

T

Tangential Mode: A room resonance involving two dimensions simultaneously; carries 3dB less energy than axial modes but still contributes audible coloration.
TCXO: Temperature-Compensated Crystal Oscillator - a clock with on-chip temperature compensation, less stable than an OCXO but cheaper and faster to lock; used in mid-range DACs and most consumer DAC stages.
Tesla Driver: Beyerdynamic's high-flux-density (over 1 Tesla) dynamic driver platform used in the T1, T5, and DT 1990 Pro lines; the strong magnetic field allows lower voice-coil mass and faster transient response.
THD: Total Harmonic Distortion - the ratio of harmonic overtones to the fundamental frequency, expressed as a percentage; lower is generally better.
THD+N: Total Harmonic Distortion plus Noise - a more complete distortion measurement that includes both harmonic products and broadband noise.
Thiele-Small Parameters: A set of electromechanical specifications (Fs, Vas, Qts, Xmax, etc.) that mathematically describe a woofer's behavior for enclosure design.
Tidal Connect: A native Tidal streaming protocol that bypasses Bluetooth or AirPlay to deliver bit-perfect audio directly from the Tidal cloud to compatible network endpoints.
Tight Bass: Low frequencies with fast attack, well-controlled decay, and clear pitch definition - the opposite of "loose" or "bloated" bass that booms and overhangs into subsequent notes.
Timbre: The tonal quality of a sound - what makes a violin sound like a violin vs. a trumpet at the same pitch and volume; determined by harmonic content and envelope.
Time Alignment: Adjusting the delay of individual drivers so their acoustic output arrives at the listening position simultaneously, maintaining phase coherence through the crossover region.
Toe-In: The inward angle of loudspeakers relative to the listening position. Greater toe-in narrows the soundstage but improves on-axis tonality; less toe-in widens the stage but exposes off-axis response variations.
Tonal Balance: The overall perceived distribution of energy across bass, midrange, and treble; correct tonal balance is the foundation of accurate reproduction.
Topology Diaphragm: HIFIMAN's patented diaphragm coating featuring a precisely-engineered surface pattern that controls breakup modes and damps unwanted resonances across the driver's operating range.
TOSLINK: An optical S/PDIF digital audio interface using a plastic fibre and a red LED, common on TVs, soundbars, and consumer DACs. Limited to 24-bit/96kHz stereo PCM in practice; longer runs need glass fibre for stability.
Tracking Force: The downward force (in grams) applied by the stylus to the groove; too light causes skipping and distortion, too heavy accelerates groove and stylus wear.
Transformer-Coupled: An amplifier output stage using a transformer to match high tube plate impedance to low speaker impedance; the transformer's quality is critical.
Transient Response: How accurately and quickly a system reproduces the onset and decay of sounds; slow transient response produces a "veiled" or "smeared" character.
Transmission Line: A speaker enclosure using a long, tapered, internally damped tube to absorb the rear wave and extend low-frequency response without a resonant port.
Transparency: The quality of a system that conveys the recording with minimal added coloration or character of its own; transparent components are "invisible" in the chain.
Transparency Ratio: The ratio of direct sound to total sound (direct + reflections) at the listening position; higher transparency ratio means clearer, more precise imaging.
Treble Extension: How high in frequency a system accurately reproduces; good treble extension captures the fundamental and harmonics of cymbals and strings.
Tribrid Driver: A multi-driver IEM combining three driver types - typically dynamic (DD) for bass, balanced armature (BA) for mids, and electrostatic (EST) for treble - in a single shell with a passive crossover network.
Triode: A vacuum tube with three elements (cathode, grid, and plate) that produces predominantly second-harmonic distortion, which many listeners find musically pleasing.
Tube Rolling: Replacing vacuum tubes in audio equipment with different types, brands, or production vintages to alter tonal character - a practice unique to tube audio.
Tweeter: A small, lightweight driver designed for frequencies above approximately 2kHz, using a dome or ribbon diaphragm for low mass and fast response.

U

UAC2: USB Audio Class 2 - a USB device protocol for high-bandwidth digital audio supporting up to 32-bit/384kHz PCM and DSD256 over standard USB; the dominant standard since 2010 and supported natively by every major OS.
Unbalanced: Single-ended signal transmission using one signal conductor and a shared ground, more susceptible to hum and interference over long cable runs.
Underdamped: Insufficient damping allowing the driver or port to resonate; creates a "one-note" or "boomy" bass characteristic.
Upsampling: Converting audio from its native sample rate to a higher rate, either in the DAC chip or in software, using a mathematical interpolation algorithm.
USB Isolator: A device that uses optocouplers or capacitors to electrically isolate the USB data path, preventing computer ground noise and USB switching noise from reaching the DAC.

V

Veil: A subjective loss of resolution and immediacy that makes recordings sound as if a thin curtain hangs between the listener and the performers. Can result from high distortion, limited bandwidth, or excessive damping.
Voice Coil: The coil of wire wound on a cylindrical former that sits in the magnetic gap of a speaker motor; carrying audio current creates the force that moves the cone.
Voltage Gain: The ratio of output voltage to input voltage in an amplifier, expressed in dB; as opposed to power gain, relevant when the amplifier has low output impedance.
Volume Matching: A less rigorous form of level matching done by ear or with line-level voltage measurement; sufficient for casual comparison but introduces error that can mask small differences in a blind A/B test.
VTA: Vertical Tracking Angle - the angle of the stylus cantilever relative to the record surface; affects tonal balance and channel separation if not set correctly.

W

Warmth: A subjective description of elevated bass and lower-midrange energy giving a sense of fullness; can be a tonally accurate or an artificial coloration.
WASAPI: Windows Audio Session API - the modern Windows audio subsystem; WASAPI Exclusive Mode bypasses the mixer for bit-perfect output similar to ASIO.
Waterfall Plot: A 3D decay measurement (frequency vs. time vs. level) showing how long resonances ring after a signal stops - critical for room and speaker analysis.
Waveguide: A flared horn or lens around a tweeter that controls directivity and can improve crossover behavior by matching the tweeter's radiation to the woofer's pattern.
Woofer: A loudspeaker driver optimized for bass frequencies, typically from 20Hz to 500Hz, with a large cone and long voice coil travel.
Word Clock: A precision timing reference (typically 44.1kHz, 48kHz, or 10MHz) distributed across multiple digital audio devices to synchronize their internal clocks for jitter-free transfer.

X

XLR: A professional locking audio connector with three pins: ground, positive signal, and negative signal; the standard for balanced studio connections.

Z

Zobel Network: A resistor-capacitor network placed across a speaker's binding posts to flatten the rising inductive impedance at high frequencies, improving crossover performance.