Introduction to Sound and Hearing

Sound is vibration that can be perceived by the auditory system of humans and other animals. It consists of propagated waves of disturbance through a medium, usually air, though it can also travel through liquids and solids. When sound waves travel, they cause changes in air pressure that occur over time at regular intervals, creating a complex interplay between frequency, amplitude, and timbre.

Pure Tone

The simplest form of sound is a pure tone, represented mathematically as a sine wave or cosine wave, characterized by its single frequency and waveform shape. Pure tones are rarely found in nature but are essential in understanding more complex sounds.

Time Domain Representation

Sound can be graphed in the time domain, where:

  • x-axis: represents time, showing how sound changes over a specific duration.

  • y-axis: represents changes in air pressure, illustrating the waveform's amplitude variation.

Wave Attributes

Understanding sound involves several key attributes:

  • Period: The time taken for one complete cycle of the waveform, typically measured in milliseconds.

  • Frequency: The rate of repetition of the sound wave, quantified in Hertz (Hz), indicating how many cycles occur per second.

  • Amplitude: The maximum extent of the waveform, relating to the energy and volume of the sound.

  • Phase: Refers to the starting point in the cycle of the wave, crucial for understanding sound interference.

Relationship Between Period and Frequency

Period and frequency are inversely related:

  • As the period increases, the frequency decreases and vice versa. For instance, if the period is 10 milliseconds, the frequency will be 100 Hertz, calculated by the formula:
    ext{Frequency} = rac{1}{ ext{Period}}.

Attributes and Perception

Different attributes correlate with human perception of sound:

  • Amplitude: Relates closely to loudness—the greater the amplitude, the louder the sound perceived.

  • Frequency: Influences pitch; higher frequencies correspond to higher pitches. This can be analogized to color and wavelength—different wavelengths correspond to different colors, yet they are not the same concept.

  • Timbre: The quality or color of sound that makes it distinct, not attributable solely to loudness or pitch; it is influenced by the harmonic structure of the sound.

Timbre Explained

Timbre is determined by the spectral envelope, which represents all the frequencies that contribute to a single sound. For example, the same musical note played on a piano and a flute demonstrates different timbres due to the differing spectral content produced by each instrument's unique harmonic profile.

General Perceptions

  • Sounds of higher amplitude are perceived as louder, while higher frequency sounds are perceived as having a higher pitch. Understanding these distinctions is vital for fields such as music, acoustics, and sound engineering.

Types of Sounds

  • Periodic Sounds: These include musical notes, pure tones, and certain voicing types in speech (like vowels) and are characterized by their predictability and regular waveform patterns.

  • Aperiodic Sounds: Such as crashes and hisses, tend to lack a consistent pattern and are often associated with noise.

Time and Frequency Domains for Periodic Sounds

While pure tones are the simplest periodic sounds, periodic sounds found in nature often encompass complex waveforms that repeat over time, making them more intricate in structure.

Frequency Domain Representation

When illustrated in the frequency domain:

  • x-axis: indicates frequency (in Hertz).

  • y-axis: shows the relative power (amplitude) of the frequencies. A pure tone will appear as a single peak at its specific frequency with high amplitude on this graph.

Natural Sounds

Real-world sounds consist of a combination of numerous frequencies that generate harmonics. For instance, sounds like musical instruments produce a range of harmonics, like 100 Hz fundamental tones accompanied by harmonics at 200 Hz, 300 Hz, etc. In contrast, aperiodic sounds incorporate random frequencies, lacking clear periodicity.

Fourier Analysis

A pivotal mathematical technique, Fourier analysis, allows complex sounds to be broken down into their component frequencies. This process reveals that any complex sound, whether periodic or aperiodic, can be constructed by summing sine waves of varying frequencies and amplitudes.

Complex Periodic Sounds

These sounds exhibit more intricate waveforms in the time domain. For example, if a waveform repeats every 5 milliseconds, this corresponds to a frequency calculation of:
ext{Frequency} = rac{1}{5 ext{ ms}} = 200 ext{ Hertz}.

Frequency Domain of Complex Periodic Sounds

A complex periodic sound will display both a fundamental frequency and higher overtones (harmonics). For example, a sound with a fundamental frequency of 200 Hz typically includes harmonics at integer multiples, such as 400 Hz, 600 Hz, and higher, creating richer sound experiences. This is crucial for understanding how different instruments achieve their unique sounds.

Timbre and Harmonics

Although different instruments may play the same pitch (fundamental frequency), their perceived sound quality differs due to the distribution of harmonics; these differences define timbre.

Aperiodic Sounds
Transients

Transients are brief yet noticeable sounds, such as a rap on a desk or a clap, with waveforms in the time domain that show no regular periodicity.

Frequency Domain of Aperiodic Sounds

Aperiodic sounds do not contain regularly spaced harmonics but are made up of a blend of various frequencies, contributing to their unique timbres.

Non-Transient Aperiodic Sounds

For instance, white noise contains all possible frequencies and appears as a flat line of amplitude across the frequency domain, analogous to white light having equal intensity across the spectrum of colors.

Real-World Sounds

Unique characteristics of sounds, such as hissing (like the s sound), exhibit specific frequency components. Fricatives, as a class of sounds, show differences in both time and frequency domain representations, resulting in distinguishable auditory experiences.

Harmonics in Periodic Sounds

In periodic sounds, harmonics function as integer multiples of the fundamental frequency (F0), such as F1, F2, F3, up to FN. These harmonics have varying relative amplitudes, which determines the overall tone quality of a sound.

Factors Influencing Harmonics

Numerous factors affect harmonics, including the physical characteristics of the sound production device and resonance, which amplifies certain harmonics over others, affecting how sound is perceived.

Perceived Pitch

The perceived pitch of a sound generally corresponds to its fundamental frequency; however, there are exceptions based on the context of the sound and varying harmonic interactions.

Complex Periodic Sounds in Speech

Vowel sounds, such as those found in the English words "ah" (as in bat) and "ee" (as in heat), demonstrate distinct frequency domain patterns, marking their unique harmonic structures.

Harmonics in Speech

Speech sounds are produced at a particular pitch with harmonics appearing at multiples of the fundamental frequency, and the relative power of these harmonics can change, conveying different sounds in human speech.

Formants

Formants are specific regions of emphasized harmonics that define particular vowel sounds. For example, the "ah" sound has formants that accentuate harmonics just below and above 1 kHz, while the "ee" sound has a prominent formant in the lower range and additional peaks above 2 kHz, collectively shaping the sound's quality.

Spectrograms

Spectrograms are visual representations that combine both the time and frequency domains into a single comprehensive view, where:

  • x-axis: time.

  • y-axis: frequency.

  • Color or lightness codes: represent the amplitude or power of the sound, allowing for a nuanced analysis of sound characteristics.

Interpreting Spectrograms

By examining vertical slices of a spectrogram, one can access the frequency domain at any given moment. Formant patterns displayed in these spectrograms are crucial for distinguishing different vowel sounds; for instance, the characteristics of the "ah" and "ee" vowels can be easily identified through their unique formant alignments.

Characteristics of Spectrograms

Higher colors or intensity levels in spectrograms indicate greater power, thus correlating with increased loudness. Additionally, spectrograms illustrate how the frequency composition of sound evolves over time, providing insights into dynamic sound interactions.

Spectrograms of Speech

Spectrograms can illustrate the dynamic differences in spoken words, such as "hat," revealing variations based on individual voice characteristics.

Differences in Child vs. Adult Speech

Typically, a child’s voice features a higher fundamental frequency compared to an adult's voice, creating greater spacing between harmonics and making their speech sound qualitatively different.

Decibel Scale

The decibel scale is a logarithmic way to measure the intensity of sound, where:

  • Basic Unit of Intensity: Measured in watts per square meter.

  • Threshold of Hearing: A typical listener can barely detect a 1000 Hertz tone at 10^{-12} watts per square meter.

Logarithmic Scale

The logarithmic scale is employed due to the extensive range of sound intensities that humans can perceive. The decibel value is computed using the formula:
ext{dB} = 10 * log{10} rac{Im}{I_r} , where Im is the intensity of the sound being measured and Ir is a reference intensity (usually the threshold of hearing).

Decibel Values

Understanding decibel levels helps describe common sounds: 0 dB represents the absolute threshold of hearing, whispers around 40 dB, and typical speech exceeds 60 dB, whereas sounds exceeding 120 dB can cause pain or permanent hearing damage.

Human Sensitivity Across Audible Spectrum

Human sensitivity to sound is variable across different frequencies. Equal loudness curves illustrate that higher intensity is necessary to perceive lower frequencies as equally loud when compared to higher frequency sounds. Sensitivity peaks around 4000 Hz and drops significantly at frequencies above 20 kHz, which are inaudible to humans.