Digital Audio

While digital audio has become a core component of modern music and telecommunications, digital audio technology is still very much in flux.

When sound is recorded on tape or transmitted via telephone, it is converted into an electrical signal. This is called “analog” audio, because the electrical signal is an analogue of the sound that created it. Upon playback, the electrical signal is converted back into sound. In digital audio, another step is added in which the electrical signal is converted into digital data—a series of zeros and ones—for the purposes of storage or transmission. There are several ways to do this, but the most common system in use, which was developed by Sony, is called Pulse Code Modulation (PCM).

Two variables are important for converting analog audio to digital data: bit depth and sampling rate. Bit depth determines the possible distance between maximum and minimum volume in a sound. For instance, in a one-bit system, the possible values are zero and one, so a sound is either off or on at full volume. With additional bits, possible levels double, so a 16-bit system (the standard for CDs) has over 65,000 possible levels, and a 24-bit system (the DVD standard) has over 16 million possible levels. Higher bit resolution is usually thought to correspond with higher sound quality. However, higher resolutions require more space to store and more bandwidth to transmit; a 24-bit file is 1.5 times as large as a 16-bit file. Specifications for digital audio formats reflect a compromise between audio quality and available storage space.

Sampling rate denotes how many thousands of times per second (kilohertz, or kHz) a digital audio system takes a “snapshot” of the analog signal. Think of sampling rate like the frames of a movie: By itself, each frame appears as a still image, but by projecting a series of still images, we create the illusion of motion. The same is true for digital audio. By playing back 44,100 samples a second, CDs provide the illusion of continuous sound. As with bit depth, higher sampling rates tend to sound better than lower rates, just as a higher number of frames per second provide a smoother image of motion. But the same tradeoffs are at issue with bandwidth and storage space; a 96 kHz file (the standard for DVD) is more than twice as large as a 44.1 kHz file.

Because digital audio converts sound into data, it has brought with it an unprecedented level of flexibility and portability in the production and consumption of music. Although much analog equipment is still used to record music, a whole new breed of microprocessor-based technologies has made recording cheaper and more powerful. Since digital audio is simply a series of digital data streams, it can be manipulated through computer software via a graphical user interface (GUI). This allows musicians and audio engineers to manipulate and edit any aspect of a sound, often through the creation and manipulation of images of sounds. Perhaps the most musically important innovation in this area is a device called the sampler (although sampling can also be done with a computer and software), which records a sound and can play it back in original or modified form. By stringing together a group of modified samples, it is possible to create an entirely new piece of music from other audio recordings.

...