Video Editing Audio Post-Production Syllabus

What is the difference between voiceover (VO) and narration recorded for ADR?

VO/narration is non-sync speech from an unseen speaker recorded dry in a studio and laid over picture; ADR must lip-sync exactly to an on-screen performance. Both are studio-recorded, but only ADR requires sync matching.

What key steps make replaced ADR lines blend with production dialogue?

Match the microphone and distance used on set, match performance energy, then match EQ and add reverb/ambience (worldizing) so the studio recording sits in the same acoustic space as the location sound.

What is a 'dialogue split' or checkerboarding in dialogue editing?

Alternating dialogue clips across two or more tracks (A/B pattern) so each clip's head and tail can be cross-faded with room tone, enabling smooth, click-free transitions between takes and mics.

What does 'dialogue balancing' mean?

Adjusting clip gain, fades, EQ, and compression shot-by-shot so every line of dialogue in a scene sounds consistent in level and tone, regardless of which mic, take, or camera angle it came from.

Why is dialogue usually treated as the anchor of a mix?

Because intelligibility of speech is the top priority: dialogue level is set first, and music, effects, and ambience are balanced around it so they never mask the words.

What is the difference between clip gain and fader volume (track level)?

Clip gain adjusts a clip's level before any track processing (affecting what compressors/EQ receive); the fader adjusts the track's output level after inserts. Balance clips with clip gain first, then ride the fader for the mix.

What is a royalty-free music license?

A license where you pay once (or subscribe) for the right to use a track without paying ongoing royalties per use; it does not mean the music is free — copyright still applies and license terms restrict usage.

What are sync rights and master rights in music licensing?

Synchronization rights license the composition (from the songwriter/publisher) for use with picture; master rights license the specific recording (from the label/owner). Using a commercial song in a video requires both.

What is public domain music and does a modern recording of it need licensing?

Public domain compositions (copyright expired, e.g., most classical works) can be used freely — but a specific modern recording of them is still protected, so you must license the master or record your own performance.

What does a Creative Commons license permit for music in video, and what is a key caveat?

CC licenses allow free use under stated conditions (typically attribution); caveats include NC (non-commercial only) and ND (no derivatives) variants, and using CC-BY music still requires proper credit.

What is a needle-drop license?

A license fee charged per individual use ('drop') of a library music track in a production, as opposed to blanket or subscription licenses covering unlimited use of a catalog.

What criteria guide selecting music for a scene?

Emotional tone and genre matching the story, tempo/energy matching the pacing, instrumentation that doesn't mask dialogue frequencies, appropriate structure (builds, climaxes), and licensing that fits the budget and distribution.

What does 'editing music to picture' by cutting on the beat mean?

Placing music edits and picture cuts on the musical beat or phrase boundaries so transitions feel rhythmically natural; DAW/NLE markers on beats help align cuts and hits with visual action.

How do you shorten a music track cleanly to fit a scene's duration?

Cut at phrase or bar boundaries (typically every 4 or 8 bars), matching the waveform/beat grid, and crossfade the edit so the seam is inaudible; keep the intro and the natural ending/outro where possible.

What is a music 'stinger' and a 'sting/button' ending?

A stinger is a short musical accent that punctuates a moment (a reveal, a joke); a button/sting ending is a decisive final note or hit used to conclude a music cue cleanly rather than fading out.

What is a crossfade and what problem does it solve in audio editing?

An overlapping fade-out/fade-in between two clips. It hides edit seams, eliminates clicks caused by cutting a waveform at a non-zero-crossing point, and smooths transitions in music and ambience.

What is Foley and how does it differ from library sound effects?

Foley is custom performance of everyday sounds (footsteps, cloth, props) recorded in sync to picture by a Foley artist on a Foley stage; library SFX are pre-recorded effects pulled from collections and cut to picture.

What are the three classic categories of Foley?

Footsteps, moves (cloth/clothing rustle), and specifics/props (everything handled on screen: doors, cups, keys, punches).

What is the difference between hard effects, ambience/backgrounds, and designed sound effects?

Hard effects are sync sounds tied to on-screen events (door slam, gunshot); ambiences/BGs are continuous environmental beds (traffic, birds, room air); designed effects are created/layered sounds for things that don't exist naturally (sci-fi, monsters).

What is the standard track layout (stems) of a video mix?

Dialogue (DX), Music (MX), and Effects (FX) — the 'DME' stems. Keeping them separate allows independent balancing, foreign-language versions (M&E mix), and deliverable requirements.

What is panning and what is the convention for dialogue placement in a stereo mix?

Panning positions a sound in the stereo (or surround) field. Dialogue is conventionally panned to the center, while music and ambience are spread wider to create space around the voice.

What does an equalizer (EQ) do, and what are the three broad frequency regions for voice work?

EQ boosts or cuts specific frequency bands. For voice: lows below $\sim 200\ \text{Hz}$ carry body/boominess, mids $\sim 1\text{–}4\ \text{kHz}$ carry presence and intelligibility, highs above $\sim 6\ \text{kHz}$ carry air and sibilance.

Define the four main compressor parameters.

Threshold: level above which compression starts; Ratio: amount of gain reduction (e.g., $4{:}1$ means $4\ \text{dB}$ over threshold becomes $1\ \text{dB}$ out); Attack: how fast compression engages; Release: how fast it lets go. Makeup gain restores lost level.

With a compressor set to a $4{:}1$ ratio and a threshold of $-20\ \text{dBFS}$, what is the output when the input is $-8\ \text{dBFS}$?

The input is $12\ \text{dB}$ over threshold; output excess $= \frac{12}{4} = 3\ \text{dB}$, so output $= -20 + 3 = -17\ \text{dBFS}$ (before makeup gain).

What is the difference between a compressor and a limiter?

A limiter is an extreme compressor with a very high ratio ($\geq 10{:}1$, often $\infty{:}1$) and fast attack, used as a ceiling to stop peaks absolutely; a compressor with gentler ratios ($2{:}1$–$6{:}1$) smooths dynamics more transparently.

What is reverb and how does it differ from delay/echo?

Reverb is the dense wash of thousands of blended reflections that conveys a space's size and character; delay/echo is discrete, distinguishable repeats of the sound. Reverb is used to place sounds in a believable acoustic space.

What is LUFS and why did it replace simple peak metering for delivery standards?

Loudness Units relative to Full Scale — a measurement of perceived loudness integrated over time (per ITU-R BS.1770). Unlike peak levels, LUFS reflects how loud content actually sounds, enabling consistent loudness across programs and platforms.

What are the integrated loudness targets for EBU R128 and ATSC A/85 broadcast standards?

EBU R128 (Europe): $-23\ \text{LUFS}$ integrated ($\pm 0.5\ \text{LU}$); ATSC A/85 (USA): $-24\ \text{LKFS}$ ($\pm 2\ \text{dB}$). LKFS and LUFS are equivalent units.

What loudness does YouTube normalize to, and what true peak ceiling is commonly required for streaming delivery?

YouTube (and Spotify) normalize to approximately $-14\ \text{LUFS}$ integrated; deliverables commonly require true peaks no higher than $-1\ \text{dBTP}$ (broadcast often $-2\ \text{dBTP}$) to prevent inter-sample clipping after encoding.

What is true peak (dBTP) and how does it differ from sample peak?

True peak estimates the analog waveform's real maximum between digital samples (via oversampling); it can exceed the highest sample value, so a file peaking at $0\ \text{dBFS}$ sample peak may clip converters/codecs — hence true-peak limits like $-1\ \text{dBTP}$.