Glossary - Sensing Sound and Music

This page collects the technical vocabulary of the book in one place, as a lookup list for revision and exam preparation. Every term is explained in context in the chapters, so treat these short definitions as reminders rather than as first introductions.

Absolute pitch: The ability to name or produce a specific pitch without hearing a reference tone first.
Acousmatic listening: Listening to sound separated from its visible source, typically over loudspeakers, which directs attention to the qualities of the sound itself.
Acoustics: The science of how sound is produced, transmitted, and received; the word also refers to the sonic qualities of a room.
Action: The subjective experience of a movement sequence: a chunk of continuous motion with a fuzzy beginning and end, often but not always goal-directed. Actions are inferred by an observer; they cannot be read directly from motion data.
Action–perception loop: The continuous cycle in which movement shapes what we perceive and perception shapes the next movement, central to embodied accounts of music.
Action–sound coupling: The lawful, mechanical link between an action and the sound it produces on an acoustic instrument, which listeners internalise through experience.
Action–sound mapping: The designed relation between action and sound in an electronic instrument, which can be arbitrary and reprogrammed, unlike a mechanical coupling.
ADSR envelope: A four-stage description of how a sound’s level unfolds through attack, decay, sustain, and release, widely used in synthesisers.
Affordance: What an object or environment invites a person to do, applied to sound when a heard event signals possible actions.
Aliasing: The distortion that arises when a signal contains frequencies above half the sampling rate, folding them down into false lower frequencies.
Ambisonics: A recording and reproduction technique that captures the whole sound field so it can be rotated and decoded to different loudspeaker layouts.
Amplitude: The maximum displacement of a vibrating medium from its resting position, related to the perceived loudness of a sound.
Arousal: The dimension of emotion running from calm to activated, one axis of the valence–arousal model.
Asymmetric metre: A metre whose beats are of unequal length, built by adding groups of two and three pulses, as in 7/8 as 2+2+3; called additive metre or, in Balkan and Turkish music, aksak.
Asynchrony: The difference in timing between two sounds meant to fall together, measured in milliseconds; the microrhythmic dimension of playing early, late, or on the beat relative to another part.
Attack time: The time from the physical onset of a sound to its energy peak; sounds with slow attacks are heard as beginning later than their onsets.
Auditory scene analysis: The process by which the auditory system sorts a single incoming pressure wave into separate sound sources and streams.
Autonomic nervous system: The involuntary control system of the body whose sympathetic and parasympathetic branches regulate arousal and recovery.
Beat: The basic pulse listeners tap along to, typically most comfortable at intervals of roughly half a second; also called the tactus.
Beat delay and anticipation: Playing systematically behind or ahead of the beat by tens of milliseconds; a slight delay is heard as laid-back, a slight anticipation as pushed, and too much of either as dragging or rushing.
Beat tracking: Estimating the times of the beats in a recording, usually from onsets and a tempo estimate; prone to the octave error of choosing a level twice or half as fast as a listener would tap.
Binaural audio: Headphone reproduction that delivers to each eardrum the signal it would receive in the recorded space, using dummy-head recording or HRTF filtering.
Bouba–kiki effect: The near-universal pairing of a rounded shape with the word bouba and a spiky shape with kiki, the standard demonstration that crossmodal correspondences between sound and shape are shared rather than learned.
Bit depth: The number of bits used to store each sample of a digital signal, which sets the amplitude precision and the available dynamic range.
Centroid of motion: The average position of the pixels that changed between two video frames, a rough measure of where in the image the motion happens.
Chills: Brief, intense responses to music with shivers and goosebumps, also called frisson, accompanied by measurable changes in skin conductance and heart activity.
Chromagram: A representation showing the energy in each of the twelve pitch classes over time, with octave information discarded, useful for analysing harmony.
Chronobiology: The study of biological rhythms, from hormonal cycles of an hour or two through the 24-hour circadian cycle to monthly and annual cycles; the field that first used the word entrainment.
Clave: A short repeating rhythmic pattern in Afro-Cuban music, spanning two bars, that serves as the timeline against which the other parts are heard.
Coarticulation: The blending of neighbouring actions or sounds into a larger chunk, so that each is shaped by those around it, as when a pianist’s hand prepares the next chord while playing the current one.
Cochlea: The spiral, fluid-filled organ of the inner ear where hair cells convert mechanical vibration into nerve impulses, organised by frequency.
Cocktail party effect: The ability to focus on one voice among many competing sounds, combining spatial hearing with selective attention.
Communicative gestures: Performer movements directed at other musicians or the audience, such as cueing an entrance, one category in the taxonomy of music-related body motion.
Consonance: The perceived stability of simultaneous pitches, as against the tension of dissonance; shaped both by the auditory system and by cultural exposure.
Continuity illusion: Hearing a tone as continuing through an interrupting noise burst, because the auditory system reconstructs what the noise plausibly masked.
Coupled oscillators: Two oscillating systems that influence each other, however weakly, and therefore settle into a stable phase or frequency relationship; the physical model behind entrainment, first described by Huygens for two pendulum clocks on one beam.
Critical band: A frequency region within which the ear analyses sound as a unit, so that components falling in the same band interact and mask each other strongly.
Crossmodal correspondence: A systematic association between the senses shared by most people, such as linking high pitch with brightness or small size.
Cross-rhythm: The overlap of rhythmic streams whose periodicities do not fit into each other as whole numbers, such as three against two, creating metrical ambiguity.
Crossmodal perception: The case in which one sense changes what another reports, as when seen lip movements change the heard syllable (the McGurk effect) or a seen source captures a sound’s location; integration showing itself when the senses disagree.
Decibel: A logarithmic unit for expressing the ratio between two sound levels, matching the ear’s compressive response to intensity.
Deep learning: Machine learning with artificial neural networks of many layers, which discover their own features from large amounts of data instead of relying on features chosen by hand.
Diffuseness: A measure, from an ambisonic recording, of how far the sound comes from everywhere at once rather than from one direction; low for a single talker, high for a reverberant murmur.
Doppler effect: The shift in perceived frequency when a sound source and listener move relative to each other, with pitch rising on approach and falling on retreat, exploited musically in the Leslie speaker.
Downbeat: The first beat of a bar, usually the strongest metric position.
Duty cycle: The fraction of a repeating on–off cycle during which a machine or signal is on, used in the book for the slow rhythms of appliances such as refrigerators.
Dynamic attending: The theory that metre is a behaviour of attention, with expectancy peaking periodically so that events on strong positions become more salient.
Echoic memory: The sensory memory that holds raw sound for a fraction of a second; it fuses fast events into a single tone and fixes where an attack is heard to begin.
Electrocardiography (ECG): Recording the electrical activity of the heart through electrodes on the skin, the standard method for precise beat-to-beat cardiac timing.
Electrodermal activity (EDA): Measurement of the skin’s electrical conductance, which tracks sweat-gland activity and serves as an index of sympathetic arousal.
Electroencephalography (EEG): Recording the summed electrical activity of the brain from the scalp, with excellent temporal resolution but limited spatial precision.
Electromyography (EMG): Recording the electrical activation of muscles through surface electrodes, capturing effort and tension even when little movement is visible.
Embedding: A list of numbers produced by a trained model to summarise a clip, a lyric line, or an image, so that similar items get similar numbers and can be compared or searched.
Embodied music cognition: The view that musical meaning arises through the whole body’s engagement with sound rather than through listening as a purely mental act.
Enculturation: The gradual, implicit learning of the norms of surrounding music, which makes familiar rhythms and tonal hierarchies feel natural.
Entrainment: The synchronisation of one rhythmic system with another, as when movements, heartbeats, or breathing align with a musical pulse.
Envelope: The overall shape of a sound’s amplitude over time, from its onset to its decay, an important cue for timbre.
Equal-loudness contours: Curves showing the sound pressure level needed at each frequency for tones to sound equally loud, with hearing most sensitive around 2–5 kHz.
Euclidean rhythm: A pattern made by spreading a number of onsets as evenly as possible over a number of pulses, which reproduces many traditional timelines such as the tresillo; a standard module in drum machines.
Event-related potential (ERP): An EEG pattern time-locked to a specific stimulus, used to study auditory attention, expectation, and memory.
Eye tracking: Measuring where the eyes are directed over time, used to study visual attention during score reading, performance, and multimedia.
Feature extraction: Computing compact descriptors from audio, such as spectral centroid, chroma, or onset strength, that stand in for perceptual or musical qualities in later analysis.
Fixation: A pause of the gaze on one location, typically a few hundred milliseconds, during which detailed visual processing takes place.
Formant: A resonance peak of the vocal tract that amplifies some harmonics of the voice, with the lowest two formants distinguishing the vowels.
Four levels of description: The book’s organising framework of physical, digital, perceptual, and interpretive levels, used to locate whether a claim is about the air, the file, the ear, or the listener.
Fourier transform: The mathematical operation that decomposes a complex signal into sinusoids, converting the time-domain view into a frequency-domain view.
Frame differencing: Subtracting each video frame from the one before, pixel by pixel, and keeping the size of the difference, so that only what moved remains; the basis of the motion image and the motiongram.
Frequency: The number of vibration cycles per second, measured in hertz, the physical counterpart of perceived pitch.
Functional magnetic resonance imaging (fMRI): Brain imaging that tracks activity indirectly through blood oxygenation, giving fine spatial detail but slow temporal resolution and a noisy scanner environment.
Functional near-infrared spectroscopy (fNIRS): An optical method measuring blood oxygenation in the cortex, silent and tolerant of movement, and therefore usable with infants and performing musicians.
Fundamental frequency: The lowest frequency component of a periodic tone, which normally determines the perceived pitch.
Generative music AI: Systems that produce new music, as notation or as audio, by learning the statistical patterns of large collections of existing music and continuing or combining them on request.
Gestalt principles: Perceptual grouping rules such as proximity, similarity, and continuity, by which listeners organise sounds into coherent lines and phrases.
Gesture: The meaning-bearing component of an action; “visible action as utterance.” The same gesture, such as waving goodbye, can be carried out with different physical motion while keeping its meaning.
Groove: A repeated rhythmic pattern that establishes a clear beat, and equally the pleasurable urge to move that such patterns induce.
Harmonics: Frequency components at whole-number multiples of the fundamental, whose relative strengths shape the timbre of a tone.
Head-related transfer function (HRTF): The filtering that the head, torso, and outer ear impose on incoming sound, individual to each person and essential for judging elevation and front–back position.
Hearing: The physiological detection of sound waves by the auditory system, contrasted in this book with listening as an active, interpretive process.
Heart rate variability (HRV): The variation in the intervals between successive heartbeats, with higher variability indicating relaxed, parasympathetic states.
Hi-hat: A pair of cymbals on a stand in a drum kit, played with sticks and opened and closed with a foot pedal; it often carries the subdivision in pop and dance music.
Impulsive sound: In Schaeffer’s typology, a short sound made by a single burst of energy, such as a drum hit.
Inertial measurement unit: A small sensor combining an accelerometer, a gyroscope, and often a magnetometer, found in phones, watches, and wearable research devices; it gives orientation and acceleration at the point where it is worn, but not position.
Interaural level difference: The loudness difference between the two ears caused by the shadowing of the head, a localisation cue that works best at high frequencies.
Interaural time difference: The tiny difference in arrival time of a sound at the two ears, the main cue for left–right localisation at low frequencies.
Interdisciplinarity: Genuinely integrating theories and methods from several disciplines, rather than only placing them side by side.
Inter-onset interval: The time between the starts of successive sounds, the basic measurement unit in studies of rhythm and timing.
Isochrony: Equal spacing of events in time; a metre whose beats are all the same length has an isochronous beat level, and non-isochrony is the deviation from it, whether as swing at the subdivision level or as unequal beats.
Iterative sound: In Schaeffer’s typology, a sound made by a rapid series of bursts that fuse into one object, such as a fast drum roll.
Just noticeable difference: The smallest change in a stimulus that a listener can reliably detect, such as about one decibel of level or a fraction of a percent of frequency.
Keynote sound: In soundscape studies, a background sound so fundamental to a place that it is mostly heard unconsciously, such as distant traffic.
Kinematics: The description of motion itself, covering position, velocity, and acceleration, without reference to the forces involved.
Kinetics: The study of the forces and torques that cause motion, which must be measured with force sensors or inferred, since cameras cannot see them.
Labanotation: A notation system for human movement developed by Rudolf Laban, with symbols on a vertical staff for direction, level, duration, and body part.
Latency: The delay between a sound arriving and a system’s response to it, set by the analysis window, any lookahead, and the processing time; the central constraint of real-time machine listening.
Listening: The active, attentive interpretation of sound, as opposed to the passive registration of hearing.
Long-term memory: Memory lasting minutes to years, holding the form of a piece and the metres and grooves of a style, and supplying the expectations a listener brings to music.
Loudness: The subjective impression of a sound’s strength, which depends on frequency and context and therefore does not map directly onto sound pressure level.
Machine listening: The use of signal processing and machine learning to turn sound into structured information: features, events, labels, and transcriptions; also called computer audition.
Masking: One sound making another harder or impossible to hear, whether simultaneously or just before or after it, a principle exploited by perceptual audio codecs.
McGurk effect: An illusion in which seen lip movements change which speech sound is heard, demonstrating that vision participates in speech perception.
Mel-frequency cepstral coefficients (MFCCs): A compact numerical description of the spectral envelope on a perceptually spaced frequency scale, widely used as a correlate of timbre, long a standard feature in speech and music classification.
Metre: A hierarchical framework of nested pulse levels, from fast subdivisions through the beat to bar-length groupings, against which rhythms are heard.
Metric accent: The emphasis a listener feels on certain positions in the metre, such as beat 1 in 4/4; in groove-based styles practice can shift it, as when the backbeat on 2 and 4 feels strongest.
Metronome: A device that produces clicks at a set rate; the clockwork version of Maelzel from 1815 gave rise to the metronome marking, such as a quarter note equals 120.
Micromotion: The tiny involuntary movements, at millimetre scale, that remain when a person tries to stand completely still, measurable with motion capture.
Microrhythm: Timing and shaping of events at a scale below about 100 milliseconds, around the beat and subdivision levels, including asynchrony, beat delay, and swing.
MIDI: A protocol and file format that stores performance instructions such as pitch, velocity, and duration rather than recorded sound.
Mirror neurons: Neurons reported to fire both when performing an action and when observing it, often invoked, though still debated, as a basis for understanding others’ sound-producing actions.
Missing fundamental: The phenomenon of hearing a pitch at an absent fundamental frequency inferred from the spacing of its harmonics, evidence that pitch is constructed by the auditory system.
Motion: The physical displacement of an object or body in time and space; continuous, with no natural beginning or end, and measurable objectively with motion capture systems.
Motion capture: Any systematic recording of body motion over time, from marker-based infrared camera systems to wearable inertial sensors and video-based pose estimation.
Motiongram: A visualisation that compresses the motion in a video into a single image over time, allowing movement to be read somewhat like a spectrogram.
Movement: The experienced, continuous excursions of the body in time and space; the everyday, experiential counterpart to motion, describing how moving is felt and perceived rather than measured.
Multimodal perception: The ordinary case in which several senses contribute to one percept at the same time, as when a drummer’s stroke is seen and heard as one event; each sense adds information the others lack.
Music information retrieval (MIR): The research field concerned with extracting structured information from musical audio and related data, from beat tracking to recommendation.
Musicking: Small’s term for music understood as an activity that includes performing, dancing, listening, and organising, rather than as an object.
Musique concrète: Schaeffer’s compositional practice of shaping recorded sounds directly, treating environmental and instrumental recordings as raw material.
Negative mean asynchrony: The tendency for taps in time with a metronome to fall slightly before the click, by 20 to 50 milliseconds for most people, showing that tapping is predicted rather than reacted to.
Neuroplasticity: The brain’s capacity to reorganise its connections through experience, evident in the effects of musical training.
Noise: In signal processing, the unwanted part of a sound or measurement, defined relative to the signal of interest; in musical contexts an aesthetic judgement rather than a physical category, as noise music shows.
Note value: The relative duration of a note in Western notation, from the whole note down through half, quarter, eighth, and sixteenth notes, each half the length of the one before; a rest of the same length marks silence.
Nyquist frequency: Half the sampling rate, the highest frequency a digital system can represent without aliasing.
Offbeat: A metric position between the beats, such as the “and” in “1-and-2-and”.
Onset: The beginning of a sound event; the physically measurable start can differ from the moment a listener perceives the sound as beginning.
Onset detection: Finding the moments at which new sound events begin in a recording, usually from sudden rises in energy or spectral change; the first step of beat tracking and transcription.
Onset displacement: How many milliseconds an onset falls early or late relative to a reference grid, negative for early and positive for late; the basic measure of asynchrony and swing.
Organology: The academic study and classification of musical instruments, with the Hornbostel–Sachs system as the best-known scheme.
Overtone: Any partial above the fundamental of a complex tone; the counting is shifted by one relative to harmonics, so the first overtone is the second harmonic.
Parasympathetic nervous system: The calming branch of the autonomic nervous system, dominant during rest and relaxing music, and reflected in higher heart rate variability.
Partial: Any of the sinusoidal components that make up a complex tone, whether harmonically related or not; the most general term for a tone’s frequency components.
Participatory discrepancies: Charles Keil’s term for the small timing and tuning differences between players that give music its feel.
Passage of time: The felt speed of time, judged separately from duration; a long concert can pass in a flash while a short wait feels endless.
Perceived and induced emotion: The distinction between recognising the emotion a piece of music expresses and actually feeling an emotion oneself, which any study of musical emotion must specify.
Perceiver: A person experiencing music through any of the senses; used instead of “listener” to include watching, feeling, and moving.
Perceptual centre: The moment at which a listener places a sound in time, usually after its physical onset and before its energy peak; also called the P-centre or perceived attack time. It depends on the attack, so sounds that start together may not be heard together.
Period: The time taken by one cycle of a periodic signal, the inverse of its frequency; for a rhythm, the inter-onset interval between repeating events.
Phase: The position within a cycle at a given moment; what a tapping finger aligns when it entrains to a beat.
Pitch: The perceived height of a tone, closely related to fundamental frequency but produced by the auditory system rather than read directly from the signal.
Pitch class: The set of all pitches that are whole octaves apart, such as every C on a keyboard, the basis of the chromagram.
Polyrhythm: A common name for cross-rhythm, the overlap of rhythmic streams whose periods do not fit into each other as whole numbers.
Pose: A posture with a meaning-bearing component: communication without moving. Like a gesture, but static rather than dynamic.
Pose estimation: The use of computer-vision models to find the positions of body landmarks, such as wrists and elbows, in ordinary video, the basis of markerless motion capture.
Position: The physical location of a body, body part, or object in space, typically given as coordinates and measurable with a motion capture system.
Posture: The subjective description of how a person holds their body; related to position the way action is related to motion.
Precedence effect: The dominance of the first-arriving wavefront in determining a sound’s perceived direction, which keeps room reflections from confusing localisation.
Psychoacoustics: The study of how physical sound relates to what listeners actually perceive, covering loudness, pitch, timbre, and spatial hearing.
Pulse: The regular series of moments that a listener feels the music moving in; the same as the beat, though the word is also used for any regular stream of events, such as an eighth-note pulse. A pulse can be felt even when no sound marks it.
Pupillometry: Measuring changes in pupil diameter as an index of arousal, attention, and mental effort, with careful control of lighting.
Quantity of motion: A single summary measure of how much motion occurs in a recording, often computed by comparing successive video frames.
Reduced listening: Schaeffer’s practice of attending to the qualities of a sound itself while setting aside its cause and meaning.
Resonance: The tendency of an object or air volume to vibrate strongly at particular frequencies, amplifying and colouring a sound source.
Reverberation: The persistence of sound in a room after the source stops, produced by many overlapping reflections and quantified by the T60 decay time.
Rhythm: A pattern of durations formed by the intervals between perceived sound events, which may or may not be periodic.
Room modes: The resonance frequencies of a room at which standing waves form between its surfaces, causing uneven bass response.
Rubato: The performer’s freedom to stretch and compress the beat expressively; literally stolen time.
Saccade: A rapid jump of the eyes between fixations, lasting tens of milliseconds, during which visual intake is largely suppressed.
Sampling rate: The number of amplitude measurements taken per second when digitising a signal, which must exceed twice the highest frequency to be captured.
Schizophonia: Schafer’s term for the split between a sound and its original source that recording technology makes possible.
Self-similarity matrix: A square image that compares every moment of a recording with every other moment, so that repetitions show up as stripes and changes of section as changes of texture; used for both audio and video.
Sensorimotor synchronisation: The task of moving in time with an external rhythm, typically tapping a finger to a metronome, which is the standard laboratory method for studying timing.
Shepard tone: An illusion built from octave-spaced components that seems to rise or fall endlessly in pitch.
Short-term memory: Memory lasting a few seconds and holding about seven items, the span of a phrase or a metric cycle; its active part, used for counting beats or holding one bar while playing the next, is called working memory.
Singer’s formant: A resonance peak around 2,500–3,000 Hz produced by clustering the third, fourth, and fifth formants, giving classically trained voices the ring that carries over an orchestra.
Sonification: The use of non-speech sound to represent data, the auditory counterpart of visualisation.
Sound object: Schaeffer’s term for the perceptual unit of listening, a chunk of sound of roughly half a second to five seconds heard as one event.
Sound-accompanying actions: Movements made to sound rather than for producing it, such as dancing or air-instrument playing, one category in the taxonomy of music-related body motion.
Sound-facilitating actions: Support movements that do not themselves produce sound but help performance, such as a pianist’s arm trajectories or breathing.
Soundmark: A sound with special significance for a community, the sonic counterpart of a landmark, such as particular church bells.
Sound-producing actions: Movements that generate sound directly, divided into excitation, such as striking or bowing, and modification, such as fingering.
Soundscape: The acoustic environment as perceived and experienced by people, including natural, human, and mechanical sources.
Soundwalking: The practice of walking through an environment while listening attentively, used both as an artistic practice and a research method.
Source separation: Computationally splitting a mixed recording into individual parts such as vocals, drums, and bass, the machine counterpart of auditory scene analysis.
Source–filter model: The standard account of the voice, in which the vibrating vocal folds provide a harmonically rich source and the vocal tract filters it through its resonances.
Spectral centroid: The centre of mass of a spectrum, a computational feature that correlates with the perceived brightness of a timbre.
Spectrogram: A representation of sound with time on the horizontal axis, frequency on the vertical axis, and energy shown as brightness or colour.
Spectrum: A snapshot of how the energy of a signal is distributed across frequencies, summarising the whole signal without showing change over time.
Spontaneous motor tempo: The rate at which a person taps when asked to tap at a comfortable pace with no music, around 100 to 120 beats per minute for most people, stable within a person and slowing with age.
Standing wave: A stable pattern of reinforcement and cancellation formed when sound reflects between parallel surfaces, the cause of room modes.
Standstill paradigm: The experimental approach of asking participants to stand as still as possible while music plays, so that involuntary micromotion can be measured.
Stem: One instrument or voice of a mix as a separate track, either kept from the production or estimated afterwards by source separation.
Stream segregation: The splitting of a sound sequence into separate perceptual streams, as when a fast alternation of high and low tones falls apart into two melodies.
Subdivision: The division of each beat into smaller equal units, such as eighth or sixteenth notes or triplets, heard as subdivisions when their intervals are roughly 100 to 500 milliseconds.
Sustained sound: In Schaeffer’s typology, a sound made by a continuous supply of energy, such as a bowed or blown note.
Swing: A systematic long–short timing pattern at the subdivision level, with ratios in performance ranging from nearly even to well beyond triplet spacing.
Swing ratio: The ratio between the durations of the first and second note of a pair of subdivisions; 1:1 is straight and 2:1 is the triplet feel, with performed values ranging between and beyond.
Sympathetic nervous system: The activating branch of the autonomic nervous system, raising heart rate and skin conductance during intense musical moments.
Synaesthesia: A condition in which stimulation of one sense consistently triggers experiences in another, such as seeing colours for tones, distinct from the crossmodal correspondences most people share.
Synchresis: Chion’s term for the automatic perceptual welding of a sound and a visual event that occur at the same moment, the foundation of film sound design.
Syncopation: A local contradiction of metrical expectation, such as an accent on a weak beat or a silent strong beat, which creates rhythmic tension.
Tala: The rhythmic cycle of North and South Indian classical music, which can be additive at the beat level, such as jhaptal as 2+3+2+3.
Telespringar: A Norwegian fiddle-and-dance tradition whose bars divide into stable long, medium, and short beats, a documented case of a non-isochronous beat level.
Tempo: The rate of the beat, measured in beats per minute; in a score it is given by a metronome marking or by an Italian tempo marking such as andante.
Tempo marking: An Italian word at the head of a score, such as adagio or allegro, that names a character as much as a speed; conventional ranges in beats per minute are printed on metronomes.
Tempogram: A plot of how strongly each tempo is present in a recording over time, computed from the onset strength; the machine-listening counterpart of tapping along to find the beat.
Timbre: The sound quality that distinguishes two tones of identical pitch and loudness, arising from spectral content and the temporal envelope.
Time signature: The pair of numbers at the start of a staff giving the note value that counts as one pulse and how many of them make a bar, such as 3/4 or 6/8.
Timeline: A short repeating pattern, usually on a bell or by clapping, against which everything else in much West African and Afro-diasporic music is heard.
Tonality: The organisation of pitches and chords around a central tonic, producing a learned hierarchy of stability among scale tones.
Tonotopy: The orderly arrangement of the auditory system by frequency, from the cochlea up to the auditory cortex.
Transduction: The conversion of energy from one form to another, as when a microphone turns pressure waves into voltage or a loudspeaker does the reverse.
Tresillo: The 3+3+2 grouping of eight eighth notes, the first half of the son clave and the basis of countless popular-music rhythms.
Two-level timing model: The account, from continuation tapping, that timing variability comes from two sources: a central timekeeper that decides when the next event is due, and the motor delay between that decision and the movement.
Upbeat: The beat or part of a beat that leads into a downbeat; a melody that starts there begins on an upbeat.
Valence: The dimension of emotion running from unpleasant to pleasant, one axis of the valence–arousal model.
Valence–arousal model: A two-dimensional map of emotion, also called the circumplex model, that places affective states by pleasantness and activation instead of sorting them into discrete categories.
Ventriloquism effect: The capture of a sound’s perceived location by a plausible visual source, as when film dialogue seems to come from the actors rather than the loudspeakers.
Verticality: The habit of describing pitch as “high” and “low”, a conceptual metaphor drawn from bodily experience rather than a physical property of sound.
Vibrato: A regular fluctuation of pitch, around five to seven cycles per second in singing, that adds warmth and helps a voice carry.
Videogram: An image made by collapsing each video frame into a single line of averaged pixels and placing the lines side by side over time, so that the movement in a whole video can be seen at a glance.
Window: The short segment of signal, tapered at its edges, that is analysed at each step when computing a spectrogram, with its size setting the trade-off between time and frequency resolution.