Reference

Audio for video glossary

100+ terms with crisp definitions, aliases, and cross-references to deep-dive articles. From sample rate, LUFS, and loudness normalization to Opus, AAC, NetEQ, lip-sync, and Dolby Atmos — every codec, loudness metric, real-time primitive, and sync concept in one place. Search by name, filter by chapter, or jump straight to a letter.

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.
ALL
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

No terms match your filters. Try a different keyword or clear filters.



AAC

Advanced Audio Coding — the default audio codec for video on YouTube, Netflix, Apple, and most OTT platforms. Several profiles: AAC-LC, HE-AAC v1/v2, xHE-AAC.

A

AAC-LC

The baseline AAC profile — Low Complexity. The exact audio in most MP4 video, ~128-256 kbps stereo. When someone says 'AAC' without a profile, they mean this.

A



AC-3 (Dolby Digital)

The 1991 Dolby surround codec that brought 5.1 to DVD and ATSC television. Replaced for streaming by E-AC-3, but still in heavy use in legacy delivery.

A



AC-4

Dolby's next-generation broadcast codec. Object-based, immersive, with built-in dialogue enhancement. Adopted by ATSC 3.0 and DVB.

A

AC-4 IMS

AC-4 Immersive Stereo — a compact Dolby Atmos delivery mode for low-bitrate streaming and broadcast, rendering immersive audio from an efficient AC-4 stream.

A

AdaptationSet (DASH)

The DASH manifest grouping that holds interchangeable versions of one content component — e.g. all bitrates of the English audio track. The DASH peer of an HLS rendition group.

A

ADC (analog-to-digital converter)

The chip that turns a continuous microphone voltage into discrete PCM samples by measuring it at the sample rate and rounding to the bit depth.

A

AEC3

The third-generation acoustic echo canceller in libWebRTC — the default echo remover in Chrome-based calls, tuned for robustness across cheap laptop hardware.

A



AEC (Acoustic Echo Cancellation)

The algorithm that removes the speaker's voice (played by your speaker) from what your microphone hears. Without it, every call becomes a feedback loop.

A

AES67

The audio-over-IP interoperability standard that lets Dante, Ravenna and Livewire exchange uncompressed PCM over a network, clocked by PTP.

A

AGC2

The newer WebRTC automatic gain control: an adaptive digital design that replaced the analog-loop AGC1 for steadier voice levels across devices.

A



AGC (Automatic Gain Control)

A loop that adjusts microphone gain so the speaker's voice arrives at a consistent target loudness, regardless of how close they sit to the mic.

A

AIFF

Apple's uncompressed PCM container, the Mac counterpart to WAV. Same audio, big-endian byte order, common in macOS and pro-audio workflows.

A



ALAC

Apple Lossless Audio Codec — Apple's lossless format, equivalent to FLAC in quality. Native on iOS/macOS pipelines and Apple Music high-res tier.

A

Aliasing

The artefacts that appear when a signal contains frequencies above half the sample rate. An anti-aliasing filter removes them before the ADC.

A



Ambisonics

A scene-based spatial audio format that captures the full sound field around a point in space using a small number of channels. The default for 360° video and VR.

A

AMR (AMR-NB / AMR-WB)

The Adaptive Multi-Rate family from 2G/3G mobile. AMR-NB (narrowband) and AMR-WB (wideband, also G.722.2) still appear in VoLTE and legacy recordings.

A

ASR (automatic speech recognition)

Automatic Speech Recognition — turning the audio stream into text for live captions, transcripts and search. Fed by VAD-segmented, cleaned-up audio off the call.

A



ATSC A/85 (CALM Act)

The US broadcast loudness recommendation, enforced by the CALM Act. Target −24 LKFS for commercials and programs alike. Why ads can no longer be louder than the show.

A

Audio accessibility

The stack that makes audio usable for everyone — described video, clean-dialogue and dialogue-enhancement tracks, signed audio — much of it legally mandated.

A



Audio codec

Coder-Decoder — the algorithm that compresses audio to fewer bits and reconstructs it on playback. The unit of compression efficiency in any audio product.

A



Audio description

A narrated track that describes on-screen visual action for blind and low-vision viewers. Required by accessibility regulations in the US, EU, and Canada.

A

Audio frame

A fixed block of samples a codec encodes as one unit — 1024 samples for AAC, 20 ms for Opus. The atom of size, latency, and seeking in compressed audio.

A

Audio level (RFC 6464)

An RTP header extension carrying each packet's loudness. An SFU reads it to forward only the active speakers and drive 'who's talking' UI without decoding audio.

A

Audio mixing

Summing several participants' audio into one stream, with level management and limiting to avoid clipping. The core job of an MCU and of server-side recording.

A

Audio object

One sound plus its position metadata (x, y, z and size) in object-based audio. The renderer places it on the available speakers or headphones at playback time.

A



Audio rendition

One specific audio variant in an adaptive streaming manifest — a particular language, codec, bitrate, or channel layout. HLS calls it a rendition group; DASH calls it an AdaptationSet.

A

Auditory masking

The effect where a loud sound hides a quieter one nearby in frequency or time. Lossy codecs exploit it: masked detail can be removed and never heard.

A



Binaural rendering

Rendering spatial audio to a stereo signal designed for headphones, using HRTFs to recreate the directional cues a listener would perceive in real space.

B



Bit depth

How many bits represent each audio sample. Each bit adds roughly 6 dB of dynamic range; 24-bit gives 144 dB of headroom.

B