Click the following links to learn more about sound:
Sound:
dosits.orgSound Waves:
macrosonix.comWaveforms:
techtarget.comAuditory Perception:
wisegeek.comAmplitude:
howmusicworks.orgDecibels:
earq.comSound Digitization:
cs.cf.ac.ukSampling:
clara.netQuantization:
mediacollege.comSound Compression:
music.tutsplus.comDigital Audio Compression:
ncsu.eduMP3 Compression:
techradar.comThreshold of Hearing:
hyperphysics.phyMasking
vivid-acoustics.comThrough reading this page, you will learn about sound digitization, sampling and quantization. Sound complicated don’t they? Don’t worry, you’ll soon understand what they all mean.
What is Sound Digitization?
Digitizing sound is the process of transforming analogue audio signals into digital signals. Natural sounds are first transformed into electrical signals using a microphone, resulting in a time-varying waveform, known as an audio signal. These waveforms are usually quite complex, and map look similar to the one on the left.
Once turned into an electrical signal, in order to convert analogue audio to digital form, sampling and quantization must be carried out. Sampling and quantization can be carried out in any order and they are carried out by special hardware devices called analogue to digital converters (ADCs).
Sampling
Sampling is when a signal’s value is measured at discrete intervals. The image to the right shows how a signal is sampled (and then quantized). You can see, the lines in the image show that the signal is split into a sequence of values of equal space.
Sampling Rates
Sampling is measured in Hertz as a sampling rate, which is the number of samples in a fixed amount of time or space. Increasing the sampling rate will produce more sample data, but this increases the storage size of the audio file. Decreasing the sampling rate has the opposite effect of increasing it, but means that the quality of the output signal is affected.
Did you know...
- Animals use sound to detect danger, warning them of possible attacks?
- Dogs hear sound at a higher frequency than humans, meaning they can hear some sounds that we can’t?
- Sound travels four times faster in water than it does in air?
CDs use a sampling rate of 44.1 kHz, which has precisely been chosen to produce a desired playing time given the size of a CD. Due to the fact that the CD format dominated audio for over two decades, the same sampling rate is commonly used by computer sound cards, to provide compatibility.
When lower quality sound is necessary, sub-multiples of 44.1 kHz are used, for example, 22.05 kHz is usually used for audio that will only be delivered over the Internet, and 11.025 kHz is sometimes used for speech audio. Typically, modern computers support sampling rates of 44.1 kHz, 48 kHz and 96 kHz, meaning that the computer should be able to read digital audio in most formats to disk, without needing to resample it.
Quantization
Quantization is when the signal’s value is restricted to a fixed set of levels. By looking at the image above, you can see that some of the values from the sampled signal are trimmed from the top so that each value lies on a line defining the allowed levels.
Quantization Levels
Quantization levels are the levels to which a signal is quantized (the vertical lines in the image). These lines are equally spaced. For analogue-to-digital conversion, the number of quantization levels is typically chosen to fit a suitable number of bits. Sound usually has a sample size of 16 bits (used for CD audio), which gives 65,536 quantization levels.