Digital Audio is sound represented as digital data rather than as a continuously varying analogue electrical or physical signal. In the conventional process, an analogue sound waveform is captured by a microphone, converted into an electrical signal, sampled at discrete time intervals by an analogue-to-digital converter and quantised into numerical values. The resulting sequence of samples can be stored, processed, compressed and transmitted by digital systems.
During playback, a digital-to-analogue converter reconstructs an electrical representation of the waveform, which can then drive an amplifier and loudspeaker or headphones. Digital Audio can also be generated directly by software or electronic instruments without an analogue recording stage. The term therefore covers both the underlying representation of sound and the broader ecosystem of recording, editing, encoding, transmission and reproduction technologies built around that representation.
Sampling rate and bit depth are fundamental technical characteristics of Digital Audio. Sampling rate specifies how many measurements of the analogue waveform are taken per second and is expressed in hertz or samples per second. The Nyquist-Shannon sampling theorem establishes that a band-limited signal can be reconstructed from samples when the sampling rate is greater than twice its highest frequency component.
Common production and consumer sampling rates include 44.1 kHz and 48 kHz, while higher rates such as 88.2 kHz, 96 kHz or 192 kHz are used in some professional workflows. Bit depth describes the numerical resolution used for each sample. Greater bit depth allows more quantisation levels and increases theoretical dynamic range while reducing quantisation error. Digital Audio quality therefore depends on much more than a simple designation such as high or low quality, because recording equipment, analogue circuitry, conversion, microphone placement, signal processing and listening conditions also matter.
Digital Audio can be stored as uncompressed pulse-code modulation, compressed audio data or encoded within various container formats. Common file formats include WAV, AIFF, FLAC, MP3, AAC, Opus and Ogg-based containers, although the distinction between a container and a codec is technically important. A container organises streams and metadata, while a codec specifies how the audio data itself is encoded or compressed.
PCM is generally uncompressed and preserves the sampled representation, whereas FLAC provides lossless compression and therefore allows exact reconstruction of the encoded PCM data. MP3, AAC and Opus use lossy compression techniques that reduce file size by exploiting characteristics of human hearing and perceptual masking. Bit rate, channel configuration, sampling rate and codec settings affect storage requirements and perceived quality. Digital Audio may be monophonic, stereophonic or multichannel, with professional production potentially using many discrete channels before final distribution.
Digital Audio is fundamental to music recording and production, podcasts, radio, film, television, games, telecommunications, voice assistants, conferencing, accessibility technologies and scientific measurement. A digital audio workstation can record multiple tracks, edit waveforms, apply equalisation, compression, reverberation and other effects, automate parameters, mix channels and produce a final master. In music production, MIDI is frequently used alongside Digital Audio but is conceptually different: MIDI primarily represents musical performance instructions and control information rather than recorded sound itself. Digital Audio can also be used for speech recognition, acoustic analysis, hearing research and machine-learning applications. In communications, speech may be sampled, encoded and compressed for transmission through packet-switched networks, where latency, packet loss, jitter and codec behaviour become important engineering considerations.
Digital Audio developed from early computer sound and digital telecommunications research into practical recording and consumer technologies during the twentieth century. Pulse-code modulation was developed as a method for digitally representing analogue signals and became foundational to digital telephony and later audio systems. Digital recording gained commercial significance through formats such as compact discs and professional digital studio equipment, followed by computer-based editing, downloadable music and internet distribution.
Portable digital players and later smartphones shifted audio consumption away from physical media towards files and network delivery. Modern services increasingly use adaptive and efficient codecs designed for streaming, while production workflows commonly retain higher-quality source material and create distribution versions later. The history of Digital Audio is therefore closely connected to improvements in converters, storage density, processor performance, compression algorithms, networking and consumer electronics.
Digital Audio offers significant technical advantages, including repeatable copying, precise editing, non-destructive processing, sophisticated effects, automated analysis, compact storage through compression and efficient network transmission. Digital editing permits samples to be cut, moved, filtered, normalised or mathematically transformed without physically altering a tape or disc. Digital systems can also represent metadata and multiple versions of an asset, supporting professional archiving and distribution. Nevertheless, digital systems have practical constraints.
Poor sampling, clipping, excessive processing, inadequate dithering, clocking problems or inappropriate conversion can introduce audible or measurable artefacts. Lossy compression can remove information and introduce coding artefacts when used aggressively. Storage media and file formats also require management, and a technically perfect digital recording can still sound poor if the original microphone placement, room acoustics or analogue capture chain was inadequate.
Several Digital Audio concepts are frequently confused. Sample rate is not the same as bit rate, because sample rate describes temporal sampling frequency while bit rate describes the amount of encoded data transmitted or stored per unit of time. Bit depth is not equivalent to bit rate either, since bit depth concerns the resolution of individual samples while bit rate is affected by sampling rate, bit depth, channel count and compression.
Loudness is also different from peak level, and dynamic range is distinct from frequency response. Digital Audio is not synonymous with compressed audio, because uncompressed and losslessly compressed representations are also digital. Similarly, high-resolution audio is a term used in different ways across the industry and does not by itself guarantee perceptually superior results. These distinctions are important when evaluating recording specifications, codecs and delivery systems.
At a specialist level, Digital Audio can be understood as a discrete-time, quantised representation of an underlying acoustic or electrical signal. Signal-processing operations can be performed in the time domain or frequency domain using techniques such as finite impulse response filtering, fast Fourier transforms, spectral analysis, dynamic-range processing and resampling. Digital audio systems must account for aliasing, reconstruction filtering, quantisation noise, dithering, clock stability and numerical precision.
Professional workflows may use floating-point internal processing to provide substantial headroom during editing and mixing, while final distribution may employ integer PCM or perceptually coded formats. Digital Audio therefore combines acoustics, electrical engineering, information theory, signal processing, psychoacoustics and computer science, making it both a consumer media format and a substantial technical discipline.
