Acoustic phonetics studies the physical properties of the speech signal as it propagates through air between speaker and listener. It treats speech as sound waves and measures their frequency, amplitude, and duration, providing an objective record that complements the articulatory and auditory branches of Phonetics.
The three primary acoustic dimensions of a speech sound are frequency (pitch correlate, measured in hertz), amplitude (loudness correlate, measured in decibels), and duration (length, measured in milliseconds). A waveform plots amplitude against time. A spectrogram, the central tool of the field, plots frequency on the vertical axis and time on the horizontal axis, with darkness indicating intensity at each frequency–time point.
Vowels and other resonant sounds show distinctive concentrations of acoustic energy at particular frequencies, called formants. The formants are numbered from lowest to highest: F1, F2, F3, and so on. F1 is inversely related to tongue height — a high vowel like /iː/ has a low F1, an open vowel like /ɑː/ has a high F1. F2 correlates with tongue frontness — front vowels have a high F2, back vowels a low F2. F3 is sensitive to lip rounding and rhoticity. F1 and F2 alone are usually sufficient to identify a vowel, and the F1–F2 plot reproduces the auditory shape of the vowel quadrilateral.
The standard model treats speech as the product of a source (vocal-fold vibration or turbulent airflow) filtered by the resonances of the vocal tract above the larynx. Different vocal-tract shapes produce different filter functions, which is why the same laryngeal source yields different vowels.
Acoustic analysis software such as Praat allows learners and researchers to visualise pitch contours, formant trajectories, and segment durations, supporting work on stress, intonation, and vowel-quality contrasts that are difficult to teach by ear alone.