3 Digital Audio
Learn how digital audio is captured, recorded cleanly, edited, mixed, encoded, and prepared for different delivery needs.
How digital audio represents sound
Sound travels as changes in air pressure. A microphone converts those changes into an electrical signal, and an analog-to-digital converter (ADC) measures and encodes the signal as numbers. During playback, a digital-to-analog converter (DAC) converts those numbers back into an electrical signal that drives speakers or headphones.
In audio, measurements called samples are taken at regular time intervals, and each sample is assigned a numerical value. The describes how often the signal is measured; describes the precision used to represent each measurement.
and representable frequencies
is measured in hertz (Hz). A rate of means samples per second. The Nyquist principle says that representing a frequency requires a greater than twice that frequency. In practice, filters prevent frequencies above the representable range from causing distortion.
Audio CDs use a of , and is widely used in audio and video production. Higher sample rates can preserve a wider frequency range, but create larger files and do not automatically make a recording sound better.
and audio data
is the number of bits used to represent each sample's amplitude. More bits allow more possible values and a wider usable dynamic range, reducing quantization noise. Adequate and safe recording levels provide room for editing, but cannot fix noise or distortion already present in the signal.
Data rate and uncompressed file size depend on , , and channel count. At the same settings, uncompressed stereo PCM uses two channels and requires roughly twice the data of mono PCM. CD audio uses and -bit PCM; higher-resolution PCM is used when appropriate for production or preservation.
Recording clean audio
Good recording begins with the sound source and the room. Reduce background noise, turn off unnecessary equipment, and position the microphone to capture the desired sound while avoiding unwanted room reflections. For close speech, use a pop filter when needed. Listen through headphones before and during a take to catch hum, handling noise, or other problems.
Set input gain so the signal is strong enough to record clearly without reaching the digital maximum. When a signal exceeds that limit, it clips and produces harsh distortion that lowering the volume later cannot remove. Leave for unexpected peaks.
Make a short test recording, listen to it, and watch the recording meter. For multitrack work, label tracks and takes clearly, and keep a separate copy of the original recordings.
Choose a that fits the project and its destination; is often used for video work. Record to an uncompressed format such as WAV. A higher or can be useful when the workflow calls for it, but sensible microphone placement, low noise, and clean gain matter more than simply choosing the largest settings. Preservation guidance commonly favors uncompressed linear PCM and formats such as Broadcast WAV for archival masters.
Editing and mixing
A displays recorded sound as waveforms and lets users arrange, process, and combine tracks. A practical workflow moves from organizing material to editing, balancing, and checking the full mix.
Organize and select: Name tracks, keep useful takes, and remove only unwanted material. Preserve original recordings so edits can be revised.
Edit timing: Trim starts and ends, remove mistakes or long pauses, and use short fades or crossfades to avoid clicks at edits. Keep speech natural and music phrasing intact.
Balance and process: Adjust track levels first. Use panning to place sounds across the stereo field. Apply equalization (EQ) to change tonal balance, and dynamics processing such as compression to manage changing loudness. Processing should solve a specific problem rather than be added automatically.
Mix and check: Listen to the full mix on headphones and speakers when available. Check that important sounds are clear, transitions are smooth, and the final output does not clip. Compare at a moderate listening level and take breaks to reduce listening fatigue.
Many DAWs support : the software stores edit instructions while leaving the original audio unchanged. This makes cuts and effects easier to adjust later. Save project files and any external media they rely on; an audio export by itself usually does not preserve the editable track arrangement.
Formats and delivery
A organizes audio data and related information. WAV and AIFF commonly carry uncompressed PCM audio. Broadcast WAV (BWF) adds fields that can hold production metadata. FLAC uses , reducing file size without discarding audio information. MP3 and AAC use , discarding some information to create smaller files that can be convenient for streaming or sharing.
Repeatedly converting between lossy formats can further reduce quality. A format name may refer to a container or an encoding, so confirm the actual encoding and settings when exchanging files.
Match delivery to the intended use
Editable project or high-quality master: Keep the DAW project and a full-quality PCM export, commonly WAV or BWF, at the project's native and appropriate . Retain original recordings and backups.
Sharing or distribution: Export to the requested format and settings. Lossy MP3 or AAC can reduce file size; use the recipient's specifications when supplied. Converting a low-quality lossy file to a higher-resolution format does not restore lost detail.
Metadata and quality control: Include useful information such as title, creator, date, and track names when supported. Preservation guidance recommends uncompressed PCM for preferred audio submissions and emphasizes complete descriptive metadata.
Reopen or play the exported file before sharing it. Check its beginning and end, listen for clicks or missing sections, confirm channel count and duration, and verify that it meets delivery requirements.