PHONETICS AND PHONOLOGY
Segmental Aspects

The term speech intelligibility is understood as comprehensibility of speech, while the degree of comprehensibility is dependable on the acoustic conditions of the environment in which the oral manifestation of speech takes place. “Human natural speech is intelligible in the sound level range of about 80 dB, it is fully intelligible even at very low sound levels down to 40 dB” (Hudcovičová, Petrášová, Rychtáriková, 2022, p. 349).  Speech can remain intelligible even when the noise is more intense than speech, or if the speech signal comes from a different direction than the noise. It is a consequence of directional hearing (Rychtáriková, 2011). In our modern, highly developed society, there are many ways how to analyse speech and visualise its characteristic aspects.

Analysis of acoustic parametres of speech is the task for acoustic/ experimental phonetics. The basic acoustic parameters of vowel sounds are frequency, intensity, duration and quality, i.e. wave structure determined by formants in vowels.

The following information clearly explains how the terms of auditory and acoustic phonetics correspond:

FREQUENCY

• = the number of complete repetitions (cycles) of variations in air pressure during one second
• the shorter the cycles, the higher the frequency
• the unit of frequency = Hertz
• human hearing organs can catch the sounds between 20 and 20 000 Hertz (most human sounds are between 80 – 350 Hz)
• a male voice = cca 120 Hz, a female voice = 220 Hz
• frequency = the term used in acoustic phonetics corresponding with the term pitch in auditory phonetics

INTENSITY

• = the amount of energy carried by a sound wave (owing to the variation of air pressure coming from the lungs)
• the unit = decibels (dB), 100 dB = quite loud sound – subjective
• the higher the intensity, the bigger the amplitude of a wave
• intensity (the term used in acoustic phonetics) = loudness (the term used in auditory phonetics)

QUANTITY (duration)

• duration (the term used in acoustic phonetics) = length (the term used in auditory phonetics)
• two categories of length:
• absolute phonetic duration – measurable
• functional phonological length – can cause a change in the meaning of words
• in connected speech – cca 6 – 20 sounds pronounced in one second

QUALITY (wave structure)

• the quality of vowels is determined by formants
• FORMANTS – seen on spectrogram as thick black horizontal bars (characteristic for all voiced sounds)
• vowels – distinguishable by F1 and F2, the formants above F4 determine the speaker´s voice quality
• voiceless consonants don´t have formant structure

(Pavlík, 2000)

Spectrograms depict the acoustic information manifested in the speech signal and allow the visualisation of the three main properties of sounds: frequency, length, amplitude. The bars seen on the spectrogram are called formants. “The presence of formants in a spectrogram is a characteristic feature of all voiced sounds” (Pavlík, 2000, p. 56). The F1 (formant 1) and F2 (formant 2) values are directly connected to the height and location of vowels:

vowels /i, u/, which are classified as high (close), have low F1 values

vowels /a, o, æ/, which are classified as low (open), have high F1 values.

vowels /i, e/, which are classified as front, have high F2 values

vowels /u, o, a, æ/, which are classified as back, have low F2 values

(Ashby, Maidment, 2005).

The average F2 value is 1500 Hz. “When the vocal folds vibrate, they produce what are called harmonics of their fundamental frequency of vibration. Harmonics are vibrations at whole-number multiples of the fundamental frequency. Thus, when the vocal folds are vibrating at 100 Hz, they produce harmonics at 200 Hz, 300 Hz, 400 Hz, and so on” (Ladefoged, Johnson, 2011, p. 209).

“A spectrogram is a graphical analysis of a sound and offers an analysis in three dimensions. The first (horizontal) dimension represents time, the second (vertical) dimension represents frequency, and the third dimension (distinguished according to the colour intensity of the vertical lines) represents amplitude” (Pavlík, 2000, p. 55). As a result, resonant frequencies of vowel sounds will appear as darker horizontal lines at certain frequencies, and those lines will correspond with the formant values of a vowel sound. F3 parameters reveal the level of nasality in vowel pronunciation (Clark, Yallop, 1995).