formant

Formant refers to a concentration of acoustic energy around a particular frequency in the speech spectrum, produced by the resonant properties of the vocal tract during vowel and other voiced sounds. Formants arise from the vocal tract acting as a system of resonators; changing the shape of the tract by moving the tongue, lips, and jaw shifts the resonant frequencies, altering the spectral envelope of the sound.

The primary formants F1, F2, and F3 are the most prominent resonances in the vowel spectrum. F1 is associated with vowel height, with higher vowels exhibiting lower F1 frequencies; F2 is associated with frontness, with front vowels showing higher F2 than back vowels; F3 is related to more subtle qualities, including rounding and certain vowels.

Measurement: Formants are typically estimated from recorded vowel segments using spectral analysis methods such as linear predictive coding (LPC), cepstral analysis, or FFT-based peak picking of the sound spectrum. In practice, formant values are reported as frequencies in Hz and may include bandwidths.

In linguistics and speech technology, formants are central to vowel identification and vowel inventory descriptions, as well as to speech synthesis and voice transformation. They also vary with speaker anatomy, gender, age, and speaking style.

Limitations: Formant estimates depend on the recording, phonation, and analysis method; formants are not fixed properties of a sound but emerge from the vocal tract configuration; across contexts, higher formants are less robust.