Great! Let’s go through these three pages step by step in a simple and clear way.
I’ll break them
into smaller ideas so you can easily follow.
3.1 Introduction (Simplified)
1. What is Speech?
Speech is how we communicate information using sounds.
Humans developed a very efficient system for producing speech, which requires little
effort.
A speech signal is basically the air pressure variations that travel from the speaker’s
mouth to the listener’s ears.
This signal is time-varying (nonstationary), meaning it constantly changes as muscles
in the vocal tract (mouth, tongue, lips, etc.) move.
2. Sound Segments
Speech can be divided into sound segments (short pieces of sound).
Segments share acoustic and articulatory (mouth movement) features.
Sounds are made by articulators:
o Vocal folds (cords)
o Tongue, lips, teeth, velum, jaw
Sounds fall into two main categories:
a) Vowels – unrestricted airflow, louder, more intense.
b) Consonants – airflow restricted, weaker intensity.
3. How We Produce Speech
Before speaking, we inhale air into the lungs. Then, while exhaling, air flows out
through the pulmonic egressive airstream (most common type of speech airflow).
Speech production = sound source excites a vocal tract filter.
o Sound source = vibrating vocal folds.
o Vocal tract = acts like a filter, shaping the sound.
Two types of sound:
o Voiced speech – periodic (regular vibration of vocal folds, e.g., vowels).
o Unvoiced speech – noisy, irregular (e.g., “s,” “f”).
4. Voicing Mechanism
Vocal folds can open and close (abduction/adduction).
Pulses of air from folds create periodic vibrations → voiced sounds.
Voiced speech = smoother, periodic.
Unvoiced speech = noisy, irregular, created by airflow at narrow constrictions.
5. Frequency & Harmonics
Voiced sounds have a fundamental frequency (F0) = vibration rate of vocal folds.
Harmonics = multiples of F0, give richness to the sound.
Voiced sounds are quasi-periodic (almost regular, but not perfectly).
6. Time & Transitions in Speech
Each sound lasts for tens of milliseconds.
Normal speech: smooth transitions between sounds.
Even when a sound suddenly changes, small signal adjustments occur.
Articulators rarely stay fixed for more than 40 ms → speech is always changing.
7. Scope of the Chapter
Focus is only on human speech (not animals or infants).
Sections in this chapter:
o 3.2: Articulators
o 3.3: Relationship between vocal tract and sound
o 3.4: Acoustic properties of phonemes
o 3.5: Math models of speech production
o 3.6: Digital versions of models
o 3.7: Context effects in speech
o 3.8: Intonation
✅ Summary in Plain Words:
Speech is moving air shaped by our mouth and throat muscles. Vowels are produced with free
airflow, consonants by restricting it. Sounds can be voiced (vocal folds vibrate, periodic) or
unvoiced (noisy, irregular). The vocal tract acts like a filter, changing sound characteristics.
Speech is dynamic—constantly changing in milliseconds as articulators move.
Do you want me to make diagrams or flowcharts to visually explain how speech is produced
(lungs → vocal folds → vocal tract → sound)? That would make it even clearer.
3.2 Anatomy and Physiology of the Speech Organs
Ear vs. Speech Organs:
o The ear is designed only for hearing.
o But speech organs (lungs, larynx, mouth, etc.) also do other jobs (like breathing,
eating, and smelling).
o So, they are not perfectly designed for speech only. They are kind of “multi-
purpose tools.”
Human Vocal Tract:
o Because it has to do many things, it’s not the most efficient for speech compared
to how well the ear is for hearing.
o Example: breathing while talking is less efficient than breathing when silent.
Connection to Electronics:
o In electronics, a good signal uses the least energy and shortest time to send
information clearly.
o Similarly, in speech, we try to use minimum effort (“energy + time”) but still keep
words clear.
o Human speech uses a frequency range up to 7 kHz. Our voices and ears work best
in this range.
3.2.1 The Lungs and the Thorax
Speech Organs:
o Three main groups: lungs, larynx, vocal tract.
o Lungs push air → through larynx (voice box) → through vocal tract → out of the
mouth as speech.
Breathing Basics:
o Normal breathing: we inhale and exhale about 0.5 liters of air every 3–5 seconds.
o Breathing uses the diaphragm (muscle under the lungs).
o Exhaling is about 60% of the breathing cycle.
o Other muscles (ribs, abdomen) also help in breathing.
Speaking vs. Breathing:
o Almost all speech sounds (like in English) happen when we exhale (egressive
sounds).
o So, speaking is more efficient when exhaling.
o That’s why we spend longer exhaling than inhaling while talking (ratio ≈ 10:1).
Special Cases:
o Some unusual sounds (like clicks in African languages) happen when air is
sucked in (ingressive sounds).
o Ejectives = when air is trapped and then suddenly released.
Airflow and Speech Loudness:
o To talk, we push out air. More airflow = louder speech.
o Normal lung capacity: 4–5 liters, but not all air is used. About 0.5 liter used per
breath.
o Everyday speech uses about half of our lung capacity. Very loud speech can use
up to 80%.
Speed of Airflow (Volume Velocity):
o Air leaves lungs at about 0.2 liters/second during sustained speech.
o Muscles help control airflow, especially during long sentences.
Pressure in Speech:
o Speaking needs air pressure.
o Conversational speech = about 10 cm H₂O pressure (imagine blowing into a
straw underwater at 10 cm depth).
o Loud speech = up to 20 cm H₂O.
o Normal breathing = only about 1–2 cm H₂O.
o So speaking requires more effort than just breathing.
o But compared to atmospheric pressure, it’s still very small.
In short:
Lungs provide air, the diaphragm controls breathing, and air pressure + airflow make
speech possible.
Speaking uses exhaled air, and normal conversation takes a bit more effort than regular
breathing.
Our organs aren’t built only for speech, but they work well enough for communication.
Got it 👍 Let’s go step by step and make this simple. Your text is about Larynx, Vocal Folds,
and Sound Production.
1. Larynx and Vocal Folds
The larynx (voice box) sits in your throat.
It has cartilages (thyroid, cricoid, arytenoid, epiglottis) that keep its shape.
Air from your lungs passes through it.
Inside are the vocal folds (commonly called vocal cords).
These are small folds of tissue that can open and close.
👉 When you breathe normally:
The folds stay open → air passes freely, no sound.
👉 When you speak/sing:
The folds come closer and vibrate when air passes → produces sound.
The gap between them is called the glottis.
2. How Sound Is Produced
Air from the lungs pushes through the glottis.
If folds are close enough, they vibrate like guitar strings.
The vibration creates voice (phonation).
The pitch (high/low) depends on:
o Length of folds (longer = lower pitch)
o Tension of folds (tighter = higher pitch)
3. Noisy Sounds Without Vibration
If the folds don’t vibrate but the air squeezes through a narrow gap, it makes a
hissing/turbulent noise.
o Example: whispering or sounds like /h/, /f/, /s/.
These are called fricatives (like “fff,” “sss”).
4. Phonemes and Features (from the table)
Phonemes = the smallest sound units in language.
They are described by:
Manner of articulation → how the sound is made (stop, nasal, fricative, vowel, etc.)
Place of articulation → where in the mouth/throat the sound is made (lips, teeth, tongue,
palate, velum, glottis).
Voiced or not → whether vocal folds vibrate.
Example:
/p/ → stop, lips, no voice (as in pop).
/b/ → stop, lips, voice (as in bib).
/s/ → fricative, alveolar (tongue near teeth ridge), no voice (as in sass).
/z/ → fricative, alveolar, with voice (as in zoo).
5. Voiced vs. Voiceless Sounds
Voiced: vocal folds vibrate → e.g., b, d, g, z, m, n, vowels.
Voiceless: no vibration → e.g., p, t, k, f, s, h.
6. Special Terms
Fundamental frequency (F0) → the basic pitch of your voice.
Period (T0) → time between two vibrations.
Subglottal pressure → air pressure below the folds (from lungs).
Supraglottal pressure → air pressure above the folds (toward mouth).
✅ In short:
Lungs = air power
Larynx/vocal folds = sound generator
Mouth/tongue/lips = sound shapers
Would you like me to also make a simple diagram with labels so you can visualize this better?