What is ADR? Dubbing, looping and lip-sync explained

• ADR replaces dialogue in the same language, usually performed by the original actor, to fix or change lines after filming.
• Lip-sync dubbing replaces dialogue with a translation performed by new actors, matched to mouth movements and timing.
• Voice-over translation leaves the original voice audible underneath and doesn't try to match lips.
• Studios cue actors with beeps and visual streamers so each line starts on the right frame.
• The core skill in ADR and dubbing is matching timing, breath and energy while still sounding spontaneous.
• You can practise the same skill at home by dubbing short scenes line by line in SceneDub's studio.

ADR stands for automated dialogue replacement, and the same letters are also read as additional dialogue recording. Older names for the process include looping and post-synchronisation. Whatever it's called, it means an actor returns to a studio after filming and re-records lines while watching the picture, so the new audio can replace the original.

Productions use ADR when the original sound can't be used or needs to change. Location recordings can be spoiled by aircraft, traffic, wind, generators or clothing rubbing against a body mic. Lines get rewritten in the edit to clarify a plot point. A performance might need a different emphasis. TV versions may replace profanity with cleaner lines, and new off-screen lines can be added over shots where the character's mouth isn't visible.

Most sound teams prefer usable production sound when they can get it, because it carries the energy of the moment on set. ADR that isn't matched well sounds detached from the scene, which is why the craft focuses so heavily on matching.

Before digital editing, the section of film containing a line was spliced into a physical loop so it could play over and over. The actor watched the loop and repeated the line until the timing matched the picture. That's where the name looping comes from.

Later systems automated the repeat-and-record cycle with synchronised machines, which is where the automated in ADR comes from. Today it's done in a digital audio workstation with software cueing, but the old vocabulary stuck. Groups of actors who record crowd chatter and background voices are still called loop groups, and that indistinct background chatter is called walla.

Before the session, the ADR editor or supervisor prepares a cue sheet listing each line with its character, timecode and the reason it's being replaced. On the day, the actor works at a microphone on an ADR stage or in a booth, often with a mic chosen and placed to resemble how the production boom captured the original, so the new line sits naturally in the scene.

The actor usually hears the original line in headphones as a guide, then gets a countdown. A common convention is three short beeps at even intervals, with the line starting where a fourth beep would fall. Many studios also use a visual cue: a streamer, a line that sweeps across the picture and ends in a flash, called a punch, at the moment the line begins.

The actor records several takes, and the editor picks the best and, if needed, nudges syllables to tighten the sync. The re-recording mixer then uses EQ and reverb to make the studio recording sound as if it was captured in the original location. Good ADR actors help by recreating the physical state of the scene: breathing hard after a run, turning their head for a shouted line, or moving as the character moved.

Dubbing usually means replacing a film or series' dialogue with a translated version performed by new actors. It's how many audiences watch foreign films and series, and it's standard for a lot of animation and anime released in other languages. Some countries, including Germany, Italy, Spain and France, have long traditions of dubbing; many others mainly use subtitles.

The script is the first challenge. A translator or dubbing adapter rewrites the dialogue so it keeps the meaning, fits the length of each line and, where the mouth is clearly visible, matches key mouth shapes. Visible lip closures on sounds like p, b and m are the hardest to cheat, so adapters try to put similar sounds where the original actor's lips close. Wide shots and characters facing away give more freedom.

Recording is led by a dubbing director, often with actors recording their parts separately, line by line, watching the picture with the original audio as a guide. In France and Quebec, many studios use the rythmo band (bande rythmo), a strip of text that scrolls across the bottom of the screen in sync with the picture, so the actor speaks each word as it crosses a fixed line.

Voice-over translation is a faster, cheaper alternative that doesn't attempt lip-sync. The original audio is lowered but left audible, and a translated voice is laid over it, typically starting a moment after the original speaker and finishing a little before. It's common for documentary interviews, news and reality TV, where it's sometimes called UN-style voice-over.

In some markets it's used for drama too. Poland is the best-known example: a single reader known as a lektor has traditionally voiced every character in TV broadcasts of foreign films, in a calm, mostly neutral tone. The aim is to inform rather than to replace the performance, which is the opposite of what a dubbing actor does.

The terms blur in casual use, and fans often call any replaced dialogue dubbing. In production, they describe different jobs.

ADR and lip-sync dubbing share the same core skill set, which is why many dubbing actors also do ADR and loop group work.

• ADR: same language, usually the original actor, replaces or fixes specific lines, and must match lips and performance closely.
• Lip-sync dubbing: new language, new actors, replaces the whole dialogue track, and matches timing and visible mouth movements while performing the role.
• Voice-over translation: new language, often one or a few voices, sits on top of the audible original, with no lip-sync.
• Narration voice-over: not replacing anyone, just a voice speaking over picture.

Timing comes first: starting on the right frame, matching the length of each line and landing visible consonants where the mouth closes. But timing alone produces technically accurate, lifeless lines. The harder part is matching energy, breath and emotional state, so you sound like you're in a car chase while standing still in a booth.

Dubbing actors also record everything the original performer does with their voice: breaths, sighs, laughs, gasps, grunts and small reactions. Leaving those out is one reason some dubs feel empty. And while you follow the shape of the original performance, the best dubs don't copy it note for note; they create a real performance inside those boundaries.

The most common failure is sounding read. Because you're watching a screen and often reading at the same time, it's easy to slip into a flat booth voice. Good dubbing actors learn each line quickly, then watch the actor's face rather than the text.

SceneDub's studio is built for this skill. Pick a prepared scene, read each line as it plays and record it with your own voice; SceneDub rebuilds the clip so you can watch and download your dub. You can also paste a public, embeddable YouTube clip of up to three minutes and record a synced take against it.

A useful routine: watch the scene twice with the original audio and note breaths, pauses and where mouths close. Record line by line, and on the first review watch only for sync. On the second pass, focus only on performance. Then combine them. For a harder drill, record one line five times, matching the start and end exactly each time, then change the emotion while keeping the timing. That's very close to what an ADR session asks of you.

If you're learning a language, try writing your own translation of a short scene and fitting it to the mouth movements. It's a quick way to understand why dubbing adapters make the choices they do.