The Physics of Sound: What Your Ears Don't Tell Your Brain

The Physics of Sound: What Your Ears Don't Tell Your Brain

When you speak into your phone and ask it to play your favorite song, something almost magical happens. Your voice—made of air molecules bumping into each other in a specific pattern—gets converted into numbers. Those numbers get processed by software that somehow understands what you said. And then music plays. This chain of events, which takes less than a second, relies on a series of physical principles and clever engineering decisions that we'll unpack throughout this book.

The core question driving this chapter is deceptively simple: How does sound, a physical phenomenon involving vibrating molecules, become data that a computer can process? To answer it, we need to understand what sound actually is at a physical level, how we measure its properties, and how we translate the continuous, analog world of real-world sounds into the discrete, digital representation that computers require. This foundation will serve as the bedrock for everything that follows—from understanding how speech recognition works to building systems that can transcribe conversations, detect emotions in voice, or even clone someone's voice from samples.

Let's start by dispelling a common misconception. When you hear a friend call your name across a playground, your ears are not actually receiving your friend's voice traveling through the air. What travels through the air is a pattern of pressure changes—a wave of compressed and rarefied air molecules. Your friend's vocal cords vibrate, pushing air molecules together in waves that propagate outward like ripples in a pond. When these pressure waves eventually reach your ear, they cause your eardrum to vibrate in exactly the same pattern. Your brain interprets these vibrations as sound, and specifically as your friend's voice saying your name.

Sound as Vibration: Understanding Pressure Waves and Their Propagation

Sound begins with vibration. Whenever an object moves back and forth rapidly, it pushes the surrounding air molecules together and pulls them apart in a repeating pattern. This creates a sound wave—a traveling disturbance of air pressure that moves through space at a characteristic speed.

In dry air at room temperature (approximately 20°C or 68°F), sound travels at about 343 meters per second (roughly 767 miles per hour). This speed is fast enough that we perceive sounds almost instantaneously over everyday distances, but it's slow enough to create noticeable delays over larger distances. If you've ever seen a thunderstorm from a distance and noticed that the flash of lightning arrives before the rumble of thunder, you've observed the finite speed of sound firsthand. Light travels approximately 880,000 times faster than sound, which is why we see the lightning flash almost instantly while the sound takes seconds to reach us depending on how far away the strike occurred.

The key insight about sound waves is that they are longitudinal waves, meaning the molecules move parallel to the direction the wave travels. Imagine a slinky stretched out on a table. If you push one end sharply toward the other end and then pull it back, you'll see a compression pulse travel down the length of the slinky—the coils bunch together and then spread apart as the pulse passes. Sound waves in air work the same way: molecules cluster together (compression) and then spread apart (rarefaction) as the wave propagates.

```mermaid graph LR A[Sound Source<br/>Vibrating Object] --> B[Air Molecules<br/>Compressed] B --> C[Pressure High] C --> D[Air Molecules<br/>Rarefied] D --> E[Pressure Low] E --> F[Air Molecules<br/>Compressed] F --> G[Pressure High] G --> A style A fill:#e1f5fe style C fill:#c8e6c9 style E fill:#ffcdd2 ```

This diagram illustrates how sound waves alternate between regions of high and low pressure. The wavelength of a sound is the physical distance between two consecutive points that are in the same phase of the vibration—two adjacent compressions, for instance. Wavelength relates to the wave's frequency (how many cycles occur per second) and the speed of sound through the fundamental relationship:

$$v = f \lambda$$

Where $v$ is the velocity of sound (approximately 343 m/s in air), $f$ is the frequency in hertz (Hz), and $\lambda$ (lambda) is the wavelength in meters. This equation tells us that higher-frequency sounds have shorter wavelengths. A 100 Hz sound has a wavelength of about 3.4 meters, while a 10,000 Hz sound (10 kHz) has a wavelength of only 3.4 centimeters.

Understanding this relationship matters for audio technology because the physical dimensions of microphones, speakers, and acoustic spaces often need to interact with specific wavelengths. The human ear, for instance, is most sensitive to frequencies between 2-5 kHz—a range that happens to correspond to the resonant properties of the ear canal itself, which is about 2.5 centimeters long.

When sound encounters obstacles, it reflects, absorbs, or diffracts depending on the relationship between the wavelength and the obstacle's size. Low-frequency sounds (long wavelengths) diffract around obstacles easily, which is why you can hear the bass from a car stereo even when it's around a corner. High-frequency sounds (short wavelengths) travel in more straight lines and are absorbed or reflected more readily by typical building materials. This is why rooms with lots of soft furnishings (curtains, carpets, upholstered furniture) tend to "muffle" high frequencies while bass sounds persist—the soft materials absorb treble more effectively than bass.

The human vocal system exploits these physics to produce recognizable speech. Your larynx (voice box) contains two folds of tissue called vocal folds (commonly called vocal cords, though they're not actually cords). When you speak, air from your lungs causes these folds to vibrate, opening and closing in a periodic pattern. This vibration creates a quasi-periodic stream of air pulses that serves as the raw "source" of voiced speech sounds.

The rate at which your vocal folds vibrate determines the fundamental frequency of your voice—roughly corresponding to the pitch you perceive. Adult males typically have fundamental frequencies between 85-180 Hz, while adult females typically range from 165-255 Hz. Children have higher fundamental frequencies, often above 250 Hz. These differences contribute to our perception of voices as "deeper" or "higher" and are part of what gives each person's voice its distinctive quality.

But the vocal fold vibration alone doesn't produce the complex sounds of human speech. Your vocal tract—the throat, mouth, and nasal cavity above the larynx—acts as a resonating chamber that filters the raw sound coming from your larynx. Different shapes of this tract amplify certain frequencies while suppressing others, creating the distinct vowels and consonants we use to encode language. This source-filter theory of speech production, developed by acoustic phonetics researchers in the mid-20th century, provides the framework for understanding how the same basic physical principles of vibration and resonance give rise to the rich variety of human speech sounds.

Frequency and Amplitude: The Dual Dimensions of Sound

Every sound you hear can be characterized along two fundamental dimensions: frequency and amplitude. Together, these properties describe almost everything we need to know about a sound wave from a physical perspective.

Frequency measures how many cycles of compression and rarefaction occur per second. The unit of frequency is the hertz (Hz), named after Heinrich Hertz, the German physicist who first demonstrated the existence of radio waves. One hertz equals one cycle per second. Human hearing typically spans from about 20 Hz to 20,000 Hz (20 kHz), though this range narrows with age and exposure to loud sounds. Most people over 30 cannot hear freque