Formant
Formants are resonant frequencies in the vocal tract that contribute to the distinct sounds of vowels and consonants in human speech. They are created by the shaping of the vocal tract and are crucial for distinguishing between different speech sounds. Formants play a key role in the study of phonetics and are used to analyze and understand the acoustic properties of speech sounds.
6 Key excerpts on "Formant"
eBook - ePubThe Neurolinguistics of Bilingualism
An Introduction
- Franco Fabbro(Author)
- 2013(Publication Date)
- Psychology Press(Publisher)
...A given vowel is defined by two or more Formants (Fl, F2). All speakers of a language, when producing the same vowel, filter the glottal signal in the same manner: the sound thus produced presents the (two) Formants that characterize the vowel. What changes during the production of the same vowel is the fundamental frequency, namely the voice pitch, which is higher in children and women as opposed to men. Even though men have a fundamental frequency that is lower than that of women during the production of a given speech sound, e.g. [a], the vocal tract in both men and women is shaped into the same configuration and produces two sounds with the same features, i.e. similar Formants F1, F2, and F3. What makes the difference to listeners, who will clearly distinguish the male voice from the female voice, is the typical fundamental frequency of the two speakers: about 100Hz in men and 300Hz in women. The more the vocal folds vibrate in one second, the higher the fundamental frequency and hence the voice pitch of the speaker. FIG. 2.2. Different modalities of acoustic analysis of a vowel. ACOUSTIC ANALYSIS OF SPEECH In the 1940s, American researchers developed an instrument called a spectrograph, which visually shows the Formants of speech sounds by marking them on paper. The original idea behind this instrument was to give deaf people the possibility of “seeing” words (visible speech), but it turned out that it was rather difficult for them to interpret a spectrogram. On the other hand, this device became extremely useful for the acoustic analysis of oral speech. The production of speech sounds into a microphone connected to a spectrograph permits one to visualize in real time the Formants of a sound and hence the Formants of phonemes (Liberman, 1996). The Formants are plotted on a spectrogram and appear like vertical lines of different lengths (cf. Fig. 2.3, showing the Formants of the vowels [i], [a], and [u])...
eBook - ePubExperimental Phonetics
An Introduction
- Katrina Hayward(Author)
- 2014(Publication Date)
- Routledge(Publisher)
...Chapter 6 The acoustic description of vowels 6.1 Introduction In this chapter, we shall be concerned with the acoustic description of vowels. With regard to the source-filter theory, these sounds are the most straightforward, since it is not necessary to take into account any source other than air flow through the vibrating vocal folds. Their spectra show peaks and valleys, defined by the increasing and decreasing amplitude of their harmonics, as one moves up or down the frequency scale. We have seen that particular importance attaches to the frequencies of the peaks. These correspond to the Formants, that is, the resonant frequencies of the vocal tract (Section 4.2.2.2). We have seen that, given a set of Formant frequencies for a (non-nasal) vowel, it is possible to predict the output spectrum for that vowel (provided that the spectrum of the source is known, Section 4.4). In Chapter 4, Formants were treated as part of the realm of speech production, and we concentrated on their role in generating vowel spectra. However, we may also think of Formants in purely acoustic terms, that is, as peaks in the spectrum which will be heard by the listener. On this view, the primary role of Formants is to characterise the input to speech perception rather than to determine the output of speech production. In acoustic studies of vowel quality, the focus of attention is typically on the steady state of the vowel. In natural speech, vowels are rarely pronounced in isolation. In pronouncing a CV, VC or CVC sequence, the organs of speech must move from the consonantal position to the vocalic position and/or from the vocalic position to the consonantal position. These transitional movements are reflected in so-called transitions in the vowel’s Formant pattern. A good example is the spectrogram of the word door, shown in Figure 3.6...
eBook - ePub- Mike Davenport, S.J. Hannahs(Authors)
- 2020(Publication Date)
- Routledge(Publisher)
...voiced fricatives like [v] and [z]: with [v] and [z] the fricative part of the sound is aperiodic while the voicing part is periodic. Quality also plays a role in distinguishing speech sounds. Differences in vowels have to do in large part with differences in quality: [i] and [u] differ because of differences in the shape of the oral tract. The position of the tongue changes the shape of the air in the oral cavity, thus [u] and [i] have a different quality. 5.2.1 Vowels and sonorants For voiced speech sounds we distinguish the fundamental frequency (symbol: F0), the frequency at which the vocal cords are vibrating. Given the differences in the size of the vocal apparatus, men, women and children tend to have different fundamental frequencies: roughly speaking, the human voice produces speech sounds at fundamental frequencies of about 80–200 Hz for adult males, 150–300 Hz for adult females and 200–500 Hz for children. In addition to the fundamental frequency the production of a voiced sound causes the vocal tract to resonate in specific ways depending on the shape of the tract. Thus, apart from the fundamental frequency, this resonating emphasises certain frequencies above the fundamental frequency, as with the harmonics associated with musical instruments. With a particular vowel, for example, these emphasised harmonics are multiples of the fundamental frequency and correspond to the resonances of the vocal tract shape that accompany a particular vowel. In dealing with speech, resonances that are above F0 are called Formants or Formant frequencies (Figure 5.5). To take a concrete example, consider the vowel sound in the word ‘sad’. During the production of [æ] the vocal cords may be vibrating at about 100 Hz and the first Formant (F1) is about 500 Hz. This indicates that for that vowel the fifth harmonic, i.e. five times the frequency of the fundamental frequency, is emphasised, and it therefore appears darker on a spectrogram...
eBook - ePub- Peter Ladefoged(Author)
- 2017(Publication Date)
- University of Chicago Press(Publisher)
...We are therefore using a development of the model representing the vocal tract as coupled tubes in figure 8.7, and the model considering the vocal tract as (in part) a Helmholtz resonator as in figure 8.10. Fig. 8.12. The frequencies of the first three Formants when there is a constriction in the vocal tract at different distances from the glottis. Frequencies of F1 are represented by solid points, those of the second Formant by shaded points, and those of the third Formant by open circles. The line of points above the arrow on the left of figure 8.12 represents the Formants that are produced when the series of tubes simulates a constriction at a point almost 14 cm from the glottis, which is in the alveolar region. As you can see, the first Formant is very low, the second Formant is in the neighborhood of 1800 Hz, and the third Formant is a little over 3,000 Hz. Whenever the articulators move toward a constriction of the vocal tract in the alveolar region, as they do when going from a vowel toward one of the consonants [t, d, n, 1], the Formant frequencies will move toward these values. If the constriction in the vocal tract is a little further back, in the palatal region, then the Formant frequencies will be as indicated by the second column of points. As we saw earlier, these are very like the frequencies that occur in the production of the vowel [i] as in see. A constriction still further back, in the velar region, produces a situation in which the second and third Formants are very close to each other. This is one of the marks of many velar consonants. In a word such as gag, the second and third Formants seem to start from a common origin, move apart for the [æ] vowel, and then move together again for the final consonant. If the constriction is even further back, in the uvular region, then the second and third Formants are not so close together, and there is a noticeable increase in the first Formant...
eBook - ePub- Peter Ladefoged, Sandra Ferrari Disner(Authors)
- 2012(Publication Date)
- Wiley-Blackwell(Publisher)
...5 Charting Vowels 5.1 Formants One and Two Earlier in this book, when we were considering the information conveyed by the pitch of the voice, we noted that men’s and women’s voices differed substantially in pitch, but they conveyed the same information. What mattered was the relative pitch within a sentence, whether it went up or down at the end, or which word stood out from the others because of a difference in pitch. It is the pattern of pitch changes that counts, not the exact frequencies involved. We must keep the same considerations in mind when discussing vowels. The vowels of a particular speaker can be described precisely by stating their Formant frequencies. But some speakers with big heads will have large resonating cavities, producing Formants with comparatively low frequencies; and others will have higher Formant frequencies because they have smaller vocal tracts. In order to represent the vowels of a language we need to show the relative values of the Formants. The most useful representation of the vowels of a language is a plot showing the average values of Formant one and Formant two for each vowel as spoken by a group of speakers. We can also get this plot to reflect the approximate tongue positions in vowels by arranging the scales appropriately. When you say the vowel i as in heed, you pull the tongue up so that it is high and in the front of the mouth. Breathe in while holding the position for i, and you will be able to feel the cold air rushing through the narrow gap between the tongue and the roof of the mouth. Now say the vowel u as in who’d. If you hold this position while breathing in you will find that the tongue is still high in the mouth, but you can feel the rush of cold air further back in the mouth. The vowel ɑ as in hod has a very low tongue position. Look in a mirror and you will see that the mouth is wide open and the body of the tongue is very far back in the mouth...
eBook - ePubScott-Brown's Otorhinolaryngology and Head and Neck Surgery
Volume 3: Head and Neck Surgery, Plastic Surgery
- John Watkinson, Ray Clarke, John C Watkinson, Ray W Clarke(Authors)
- 2018(Publication Date)
- CRC Press(Publisher)
...fire). Vowel articulations are made by varying the resonating shape of the oral and pharyngeal cavities. Changing the shape of the vocal tract subsequently changes its resonating behaviour; different shapes responding to different components of the harmonic structure of the glottal sound source. The resonance peaks of the vocal tract are called Formants. These Formant structures vary for each vowel and are easily identifiable on a sound spectrograph. 5 It is possible to distinguish between vowels by changing (i) the height of the tongue raising in the mouth, (ii) the part of the tongue which is raised (front, centre, back) and (iii) the position of the lips (spread or rounded). For example, the vowel / i : / as in ‘see’ is made with the front of the tongue raised and with lips spread. In contrast, the vowel / u : / as in ‘sue’ is made with back tongue raising and rounded lips. Diphthongs (i.e. ‘beer’, ‘air’) start with one oral tract articulatory shaping and glide to another. Speech consonants are defined by their much clearer articulation (and often obstruction of the airflow) within the oral tract. Distinction between consonants can be made using three main elements: the place of articulation (i.e. lips, alveolar); the manner of articulation (i.e. plosive, fricative); and the state of the larynx (voiced or voiceless). All of these elements require further explanation. The place of articulation Consonants are clearly articulated at different places within the oral tract. Bi-labial consonants refer to articulation between the upper and lower lips (p, b, m, w). Labio-dental consonants require top teeth and lower lip articulation (f, v). Dental articulation refers to tongue tip and top teeth occlusion (th). Alveolar consonants are made by the tongue tip touching the ridge behind the teeth (t, d, n, s, z, r, ch, dj). Articulation of the middle tongue with the hard palate produces ‘y’. Velar consonants require back of tongue and soft palate articulation (k, g, ng)...
