A small listening laboratory

What makes two notes sound different?

Keep the same note. Add harmonics one by one. Hear the change and watch what happens to the waveform.

45 projects on music, voices and sound

Synthesised note · 220 HzPlayback on request

Four periods of the fundamental frequency, drawn as a waveform.

Frequency and amplitude. The bars show the components used to make the sound.

Fundamental220 HzHarmonic 2440 HzHarmonic 3660 HzHarmonic 4880 HzHarmonic 51100 HzHarmonic 61320 Hz

This is a simple synthetic tone, not a recording of an instrument. Press “Play” when you want to hear it.

01 / Harmonics

A note has a fundamental frequency and may have others above it.

Each component here is a sine wave at a whole-number multiple of 220 Hz. Change their levels; the graph and sound use the same values. Real instruments have attacks, resonances and variations this simple demonstration does not reproduce.

Inside a note — An instrument for curious ears What a sound looks like The room sings back The Speed of Sound The Deep Ocean Highway

02 / Listening

The ear selects, measures and recognises.

A voice can be understood in a crowded room. A syllable changes pace from one language to another. Other projects explore hearing thresholds, direction and timbre without treating one short test as a full portrait of any listener.

What a baby hears in the womb: your mother's voice as you first heard it Would you catch a hidden name in a crowded room? Take the 1959 test How fast Romanian is spoken: the measured pace of speech, in syllables and bits per second One paragraph, 21 British voices

03 / Music

The same note belongs to different histories.

Improvisation, a rhapsody, a tuning system and a recording preserve different ways of making music. Hear what remains and see who wrote, performed or passed on each example.

The doina — The unwritten song invented on the spot Dimitrie Cantemir: the prince who wrote down the Turks' music (1673–1723) Twelve minutes that made a composer The 23-Cent Comma The Anatomy of Wooden Fragility: Why Gustavo Santaolalla’s Theme for The Last of Us Is an Acoustic Masterpiece

04 / Soundscapes

Forests, wetlands and animals have their own sounds.

Some pages preserve real recordings; others model waves or show spectrograms. Each project description says what you hear and where its material comes from.

The woodpeckers of Băneasa Forest: seven species on the edge of Bucharest, each with its own sound Listen to the wetland that grew in Ceaușescu's lake Yellowhammer song dialects: a map of Europe drawn from the birds’ own song endings Bats aren't blind, but they hunt by ear. Find the moth by its echo The 52 Hz whale, ‘the loneliest in the world’, and the blue whales singing ever deeper since 1963

Audio starts only when you press the button. Playback is kept quiet; stop it at any time.

Sound · 45 projects

The complete collection

Search titles, descriptions and topic terms, or choose a theme.

Search for a project or topic

45 projects

↑ Back to the experience

Sounds of other species
Yellowhammer song dialects: a map of Europe drawn from the birds’ own song endings
The yellowhammer ends its song differently from region to region. 1,178 public recordings, classified twice without knowing the place, on a map drawn from the birds’ own spectrograms. Tap a tile to hear the bird. In English and Romanian.
The physics of sound
Inner tempo: how fast 3,576 people tap a finger when nothing sets the pace
Asked to tap at a comfortable pace, each person settles on a pace of their own. A 2021 University of Hamburg study timed 3,576 people and published every tap: the middle person leaves 0.73 s between taps, and only 15% tap near the textbook 600 ms. A 15-second test places your pace among them, as one dot among 3,576 dots that each flash at their own pace.
Sounds of other species
The woodpeckers of Băneasa Forest: seven species on the edge of Bucharest, each with its own sound
Seven of Romania’s ten woodpecker species have been found in Băneasa Forest, on the northern edge of Bucharest. Hear how each one sounds, with drum rolls measured strike by strike, and drum a rhythm of your own. With photographs taken in the forest in 2026 and a map of what has been built around it since 1990. In Romanian and English.
Voice and language
Which language is closest to Romanian? Three measures, three answers
Romanian, the Romance language of Romania and Moldova, compared with the world's languages on words, grammar and sounds, using open data. On words the nearest is Megleno-Romanian, then Italian; on grammar Italian and Albanian score the same; on sounds the list depends on how Romanian is described. Any language can be looked up. In Romanian, English and Chinese.
The physics of sound
Bucharest noise map by address: modelled road-noise levels around every building, school and park
The City's official noise maps existed only as PDF images. Type a street and number to see the modelled level around the building, at night and as a 24-hour average, against the WHO limit of 45 dB. Nearly 9 in 10 homes exceed it at night.
Sounds of other species
Elephants may call each other by name
Wild elephants in Kenya seem to use calls meant for one listener. Hear a real rumble in a dusk scene, then see what 469 recorded calls showed, the loudspeaker test, and the 2026 reanalysis that says it is not proven.
The physics of sound
The Earth rang like a bell for weeks
After the earthquake of 26 December 2004, which with the tsunami it set off killed or left missing 227,898 people, the whole planet vibrated like a bell. Hold a button to hear a month of the real record played back 500,000 times faster while a globe takes the shape of each tone; the Earth's breath, 0.05 mm every 20½ minutes, was recorded in Romania too.
Sounds of other species
The 52 Hz whale, ‘the loneliest in the world’, and the blue whales singing ever deeper since 1963
What US Navy hydrophones heard from 1989 to 2004, and what is not known about the 52 Hz whale. From 1963 to 2019 the song of northeast Pacific blue whales dropped by a third; type your birth year and hear a NOAA recording played at that year's note and at the 2019 note. In Romanian and English.
Music and memory
London church bells: 178 towers, their practice nights, and what the bells sound like
A night map of London's bell towers: pick a night and the towers listing a practice in Dove's Guide light up, tap one to hear its bells, synthesised at the frequencies in the ringers' database, while the order of the ropes changes row by row. Four towers, including St Mary-le-Bow, also have real openly licensed recordings.
Voice and language
One sound, five accepted forms
In one example from the public Monsoon collection for Hindi (hi-IN), the same transcript has a 33.3% or 0% error rate depending on which forms are accepted as the reference.
Sounds of other species
Bats aren't blind, but they hunt by ear. Find the moth by its echo
Hear a pipistrelle's call 10 times slower, so you can hear it at all, and find a moth in the dark by its echo. Then a whole attack, slowed down, the moth that hears the bat from 30 metres, and where the 1,000-mosquitoes-an-hour myth comes from. In Romanian and English.
The physics of sound
What a sound looks like
A brass plate computed from real physics, strewn with sand. Bow it or sing to it, and the sand runs onto the lines that stand still, as it did for Ernst Chladni in 1787. Then find eight of his figures. In Romanian and English.
Music and memory
Guess what a song is for: 118 recordings from 86 societies, heard by 29,357 people
Hear melodies recorded around the world and guess whether each is for dancing, healing, love or putting a baby to sleep. Lullabies and dance songs are recognised anywhere; love songs barely above chance.
Voice and language
One paragraph, 21 British voices
21 speakers across Britain read the same paragraph: pick a town and watch six words change shape in phonetic transcription.
Music and memory
Twelve minutes that made a composer
Eight taraf tunes sewn into an orchestral score: what the twelve minutes hold.
Music and memory
Maria Tănase (1913–1963): life, songs, the ban
From the Cărăbuș to New York 1939 and the 1940 ban: Maria Tănase's life in 12 stops and 12 songs, with years as attributed by the cited discography.
Music and memory
The conductor who refused the record — Sergiu Celibidache
Sergiu Celibidache led the Berlin Philharmonic at 33, then largely refused the studio and the commercial record after a few early sessions — while accepting radio. Why, how his concerts survived, and how slow his Bruckner is in these recordings.
Music and memory
Dimitrie Cantemir: the prince who wrote down the Turks' music (1673–1723)
Hostage at Tsargrad, prince one year at Iași, beaten at Stănilești, exiled in Russia: Dimitrie Cantemir's life, with his music notation to try and a game of the 1711 decisions.
The physics of sound
Inside a note — An instrument for curious ears
Take a sound apart. Shape its harmonics, see the waveform, and listen to the same pitch change character.
Voice and language
The Same Mouth — How 32,000 Baby Name Spellings Collapsed Into 4,000 Sounds
US birth records contain 31,904 distinct baby-name spellings in 2024, but a phonetic algorithm reduces them to 3,998 sounds. The interactive charts trace how the gap has widened since 1950.
The physics of sound
The Speed of Sound
From Yakutsk to Bangkok, sound gains 6.83% in speed — nearly a full second over every 5 kilometres.
Ears and listening
Your Ear Has Its Own Amplifier
The quantitative biophysics of the active cochlear amplifier: how 12,000 outer hair cells leverage a 150 mV transmembrane gradient to oscillate nanomechanically at over 20,000 Hz, generating otoacoustic emissions and the Tartini phantom tone (2f1 - f2).
Ears and listening
Why You Don't Deafen Yourself Talking
The quantitative biophysics of the stapedius reflex: anticipatory activity observed during vocalisation and frequency-dependent filtering in the middle ear. The simulator uses 20 decibels as an explicit model example.
Ears and listening
Ten Microseconds Between the Ears: How the Brain Computes Sound Location
The human brainstem detects sound arrival disparities down to 10 microseconds (3.4 mm in air) using axonal delay lines and coincidence detector neurons in the medial superior olive.
Ears and listening
Ten Picometres in the Eardrum
If sound struck inner ear fluid directly, 99.89% of energy would bounce off like a wall. To bridge the impedance mismatch, evolution built a mechanical three-bone lever that amplifies pressure over 22-fold — enabling the eardrum to detect vibrations of just 10 picometres.
The physics of sound
The Deep Ocean Highway
At one thousand metres depth, the ocean forms a natural waveguide carrying sound across global basins. In 1991, a tone emitted near Antarctica was recorded in Bermuda, 18,200 kilometres away.
The physics of sound
The 23-Cent Comma
Why every piano lies to your ear.
The physics of sound
Two Hands, One Rhythm
At 180°, cross-finger pairs alternate; as tempo rises, the pattern can fall abruptly into 0° symmetry. A test on your own hands reveals your personal threshold.
The physics of sound
Why no piano can be perfectly in tune
Twelve pure fifths miss the C seven octaves up by 23.46 cents. Tune a fifth by ear, climb twelve like it, and hear where Pythagorean, meantone and equal tuning hid the difference.
The physics of sound
The Ninth Bell Needs Eight Days Without a Pause
Eight tower bells can ring in 40,320 different orders, with no repeats allowed. One team has ever rung the complete list without a break, in 1963, over almost eighteen hours. One more bell would take over eight straight days.
Music and memory
The Anatomy of Wooden Fragility: Why Gustavo Santaolalla’s Theme for The Last of Us Is an Acoustic Masterpiece
A musicological analysis of Gustavo Santaolalla’s score: the Andean ronroco, asymmetric 6/8 meter, negative space, and unadorned acoustic percussion.
Voice and language
Hear the Difference: A Romanian Perception Pilot — Six Contrasts
Compare six Romanian sound contrasts with Mandarin, using IPA and explanations of tongue and lip position. The guide has no audio recordings.
Music and memory
Why do the bells ring at noon? Belgrade, 1456
July 1456: Hunyadi and Capistrano hold Belgrade against Mehmed II — and the noon bells had already been ordered.
Music and memory
How did our anthem wait 142 years? 1848–1990
“Un răsunet” at Brașov in June 1848, six official anthems, two bans and 24 January 1990: the song that waited 142 years.
Ears and listening
What You Can No Longer Hear
Hearing at very high frequencies weakens first, long before speech becomes difficult to follow.
The physics of sound
The room sings back
Walls give a room notes of its own. Where you listen changes how loud each mode sounds.
Ears and listening
The faintest sound you can hear moves your eardrum less than an atom
Go down into your own body, order of magnitude after order of magnitude: from a candle seen kilometres away, through two needles on the skin, a red cell and a 10-nanometre turbine, down to an atom. At every step, a measurement from 43 other pages.
Music and memory
The doina — The unwritten song invented on the spot
The doina is a formulaic improvisation system, not a fixed melody: a skeleton of long notes each singer dresses differently. Shape three phrases and hear them change. UNESCO, 2009.
Voice and language
Italian dialects that sound like Romanian: the AIS atlas, village by village
A century ago, an atlas led by two Swiss linguists wrote down how people spoke in 407 villages of Italy and Switzerland. Pick a Romanian word and see what each village said.
Ears and listening
What a baby hears in the womb: your mother's voice as you first heard it
For months before birth, you could hear. Listen to a mother's voice the way it reaches the child inside her, rebuilt from studies that placed microphones in the womb, then record your own.
Ears and listening
Would you catch a hidden name in a crowded room? Take the 1959 test
Put on headphones. Follow one voice and ignore the other, then discover the name hidden in the voice you ignored: a 60-second shadowing test rebuilt from the original studies of attention.
Ears and listening
The voice arrives late
On the fitted curve for 30 listeners, the 50% boundaries were 131 milliseconds of audio lead and 225 milliseconds of audio lag. Two metres of air is 5.8 milliseconds.
Sounds of other species
Listen to the wetland that grew in Ceaușescu's lake
Pick a month and hear the birds recorded in Văcărești Nature Park, from 276 public checklists, over the city's hum recorded in the park. Below, the pit from space, 1984 to 2025.
The physics of sound
Rhythmic applause: why an audience claps in unison, and why the beat falls apart
A theatre audience claps in time with no conductor. Clap along with a simulated hall of 300 people and find out why the beat appears only when people clap more slowly, and why it breaks.
Voice and language
How fast Romanian is spoken: the measured pace of speech, in syllables and bits per second
Read Romanian runs at 5.6 syllables a second, measured on 265 speakers from real recordings — the same pace as Italian measured with the same ruler. Every speaker is one dot on a single line; three voices can be played, the parliamentary rung and the 2019 study's scale sit alongside without mixing, and any voice can be timed in the page at the end.

About the collection

This collection brings together relevant projects from the site’s library. Each entry has a title, a description and a link to its available edition. The complete projects carry their sources and evidence limits.

Reviewed 2026-10-15 · Collection as text

Sources

Harmonics and hearing · Institute of Sound and Vibration Research, University of Southampton
Explains fundamental frequency, harmonics and timbre; the sound on this page is synthesised in the browser.
The science behind sounds · Physics Classroom
Doina · UNESCO Intangible Cultural Heritage