Vowels

This activity allows students to understand how the shape of the oral cavities influences the production of vowels. It establishes a link between acoustic physics and articulatory phonetics.

The human voice is a remarkably versatile instrument that can produce dozens of distinct vowel sounds using the same basic mechanism: air from the lungs vibrates the vocal cords, producing a buzzing sound rich in harmonics. This raw sound, the same for all vowels, is then shaped by the cavities of the mouth, throat, and nasal passages, which act as adjustable acoustic filters. By changing the position of the tongue, jaw, and lips, we alter the resonant properties of these cavities, selectively amplifying certain frequency bands called formants. Each vowel has a characteristic pattern of formants that our brain recognizes instantly. The vowel 'i' (as in 'see') has a widely spaced formant pattern with energy concentrated at high frequencies, giving it a bright, cutting quality. The vowel 'o' (as in 'go') has closely spaced low-frequency formants, producing a dark, round quality. Using FizziQ's spectrum analyzer, students can visualize these formant patterns and discover the acoustic fingerprint of each vowel.

Activity overview:

The student uses the FizziQ Sound Spectrum tool to compare the frequencies that make up different sung vowels. By analyzing the number and intensity of the frequency peaks for each vowel (notably 'i' and 'o'), the student discovers that certain vowels generate more harmonics and understands that the oral cavities act as acoustic filters modifying the timbre of the sound produced by the vocal cords.

Level:

Middle school

FizziQ

Author:

Duration (minutes) :

30

What students will do :

- Analyze the frequency spectra of different sung vowels using FizziQ
- Identify formant frequencies as bands of enhanced harmonic amplitude
- Compare the spectral signatures of vowels like 'i', 'e', 'a', 'o', 'u'
- Understand that the vocal cords produce the harmonics while the mouth cavities shape the formant pattern
- Connect acoustic physics with articulatory phonetics and speech science

Scientific concepts:

- Vowel formants
- Acoustic resonance
- Articulatory phonetics
- Spectral filtering
- Vocal harmonics

Sensors:

- Microphone (frequency analysis)
- FizziQ spectrum analyzer (FFT)

Material needed:

- Smartphone with the FizziQ application
- A quiet environment for voice recordings
- FizziQ experience notebook

Experimental procedure:

Open FizziQ and select the Spectrum Analyzer (FFT) tool.
In a quiet environment, sing the vowel 'a' (as in 'father') at a comfortable, steady pitch. Hold the note for at least 3 seconds.
Observe the spectrum. You should see a series of harmonic peaks with certain frequency regions showing higher amplitudes (formants).
Record the frequencies of the two or three strongest spectral regions (formant frequencies).
Now sing the vowel 'i' (as in 'see') at the same pitch. Observe how the spectral envelope changes dramatically.
Record the formant frequencies for 'i'. Note that the formants are spaced differently from 'a'.
Repeat for vowels 'e' (as in 'say'), 'o' (as in 'go'), and 'u' (as in 'you').
Create a comparison table with the first two formant frequencies (F1 and F2) for each vowel.
Note the pattern: for 'i', F2 is very high (>2000 Hz); for 'u' and 'o', F2 is low (<1000 Hz); for 'a', F1 is high (>700 Hz).
Sing the same vowel at different pitches and verify that the formant frequencies remain approximately the same (they depend on mouth shape, not vocal cord frequency).
Try whispering the vowels (no vocal cord vibration) and observe that the formant structure is still visible as broad noise bands.
Discuss: how does the shape of the mouth cavity determine the formant frequencies for each vowel?

Expected results:

Each vowel should show a distinct formant pattern. Typical formant frequencies for an adult male voice are: 'i' (F1 ≈ 270 Hz, F2 ≈ 2300 Hz), 'e' (F1 ≈ 400 Hz, F2 ≈ 2000 Hz), 'a' (F1 ≈ 730 Hz, F2 ≈ 1100 Hz), 'o' (F1 ≈ 570 Hz, F2 ≈ 840 Hz), 'u' (F1 ≈ 300 Hz, F2 ≈ 870 Hz). Female voices show higher formant frequencies (shifted up by about 15-20%). The key observation is that while the individual harmonics shift when singing at different pitches, the formant envelope (the spectral regions that are enhanced) remains approximately constant for each vowel, proving that formants depend on mouth shape rather than vocal cord frequency.

Scientific questions:

- Why do the formant frequencies remain the same when you sing the same vowel at different pitches?
- What physical property of the mouth cavity determines the first formant frequency?
- How do the vocal cords produce harmonics? Why are there many harmonics rather than a single frequency?
- Why does whispering still allow vowels to be distinguished even without vocal cord vibration?
- How does speech recognition technology use formant analysis to identify spoken vowels?
- Why do different languages have different numbers of vowel sounds?

Scientific explanations:

The production of vowels is a fascinating acoustic phenomenon which involves two complementary mechanisms: 1) A sound source: the vocal cords which, by vibrating, produce a sound rich in harmonics (the raw voice); 2) A system of filters: the cavities of the mouth, pharynx and nose which modify this sound by amplifying certain frequencies and attenuating others. These areas of amplification, called formants, are characteristic of each vowel and allow them to be distinguished aurally.

FizziQ's Sound Spectrum tool allows you to visualize this unique acoustic signature. The vowel 'i' typically has a low first formant (around 250-300 Hz) and a very high second (around 2200-2500 Hz), creating a big difference between the two.

To produce it, the tongue is brought forward and elevated, creating a small cavity in front and a large one in the back. The 'o' shows two low, close formants (around 400 Hz and 800 Hz), resulting from a retracted tongue position and a rounded labial opening.

Vowels like 'i' and 'e' generally generate more visible harmonics because their high second formant amplifies more of the high-frequency components of the speech spectrum. This spectral richness explains why these vowels “carry” better in a song.

Singers exploit this phenomenon by subtly adapting the shape of their oral cavities to optimize vocal projection while maintaining speech intelligibility. The acoustic analysis of vowels has applications in singing teaching, speech therapy, speech recognition and even historical phonetics to understand the evolution of languages.

This experience constitutes an accessible introduction to speech acoustics, a field at the intersection of physics and language sciences.

Extension activities:

Frequently asked questions:

Q: The spectrum shows so many peaks that I cannot identify the formants. How do I distinguish them?
R: Formants are broad spectral envelopes that group several harmonics. Look for regions where several consecutive harmonics are stronger than their neighbors. The formant is the center frequency of this reinforced region, not a single peak.

Q: My formant frequencies are different from the typical values listed. Is this normal?
R: Yes, formant frequencies vary significantly between individuals depending on vocal tract size. Men typically have lower formants than women, and children have the highest. The relative pattern between vowels is more important than absolute values.

Q: When I change pitch, the formants seem to change too. Why?
R: At higher pitches, the harmonics are more widely spaced, making it harder to see the formant envelope clearly. The formant frequencies themselves do not change significantly with pitch, but the discrete sampling of the formant envelope by the harmonics can shift the apparent maximum.

Q: Can I analyze consonants the same way?
R: Consonants are more complex because many are produced by turbulent noise (fricatives like 's' and 'f') or brief transients (plosives like 'p' and 't') rather than sustained vocal cord vibration. They require different analysis techniques.