How Subtitles and Dubbing Are Made, and Why They Don’t Match

You’re watching a Spanish thriller dubbed into English, with the subtitles on. A character says, “Let’s get out of here.” The subtitle reads, “We have to go now.” Same scene, same moment, different words.

It isn’t a mistake. Subtitles and dubbing are two different crafts with two different sets of rules, and each one changes the original dialogue in its own way.

How subtitles are made

A subtitler starts with the finished programme and a transcript of the dialogue. Their job is to translate it, then cut it into short pieces that appear and disappear in time with the speech.

The hard part isn’t the translation. It’s the space and the time. People speak faster than most of us can comfortably read, and the text has to share the screen with the picture. So professional subtitles follow strict limits:

• Two lines at most, placed low on the screen so they cover as little of the picture as possible.
• A limited number of characters per line. Netflix’s guidelines for English, for example, allow up to 42 characters per line.
• A maximum reading speed. Netflix allows up to 20 characters per second for adult programmes in English and 17 for children’s programmes. The BBC’s guidelines suggest a typical rate of about 160 to 180 words per minute.
• Timing that follows the picture. A subtitle shouldn’t hang over a cut to a new shot if it can be avoided, because viewers tend to re-read text when the picture changes.

To stay within those limits, subtitlers condense. Repetitions, hesitations and filler words disappear. Long sentences are simplified. Jokes and wordplay that don’t translate are replaced with something that works in the new language. A good subtitle keeps the meaning and the tone, not every word.

How dubbing is made

Dubbing replaces the original voices with new ones. Here the constraint isn’t reading speed but the actor’s face.

First, a translator produces a literal translation. Then a dubbing adapter, sometimes called a dialogue writer, rewrites it so that it can be spoken in the same time as the original line and roughly matches the movements of the actor’s mouth. Sounds made with closed lips, such as b, p and m, are especially noticeable, so adapters try to place similar sounds where the actor’s lips close on screen. Close-ups need the tightest matching; a character seen from behind or far away gives the writer more freedom.

The new lines are then recorded by voice actors in a studio, usually a few lines at a time while watching the scene, and mixed with the original music and sound effects.

Because the adapter is chasing lip movements and timing, the dubbed line can drift quite far from the original wording, even when the meaning stays the same.

A subtitle is shaped by how fast you can read. A dubbed line is shaped by how the actor’s mouth moves.

Why the two don’t match

Put the two processes side by side and the mismatch makes sense:

• They usually start from different scripts. Subtitles are normally translated from the original dialogue. The dub is a rewritten adaptation of it.
• They are usually made by different people, often at different companies and at different times.
• They solve different problems. One has to be short enough to read; the other has to fit a moving mouth.

There is one common exception. Streaming services often offer closed captions for the dubbed audio track, labelled something like “English [CC]”. These transcribe the dub itself, so they match what you hear, but they can differ from the standard subtitles translated from the original language. In 2021, viewers of the Korean series Squid Game debated whether the English captions they were reading had lost some of the original meaning, which drew wide attention to this difference.

Subtitles for deaf and hard-of-hearing viewers

Captions, or SDH (subtitles for the deaf and hard of hearing), do more than show dialogue. They also show who is speaking when it isn’t obvious, and describe important sounds, such as [phone buzzing], [tense music] or [door slams]. Some broadcasters use colours or position on screen to show different speakers.

Live programmes, such as news and sport, are usually captioned in real time, often by a trained respeaker who repeats the speech into voice-recognition software, or by a stenographer. That’s why live captions can lag behind and contain mistakes that are corrected moments later.

Why some countries dub and others subtitle

Habits vary widely around the world, largely for historical and economic reasons:

• Dubbing has long been the norm for films and TV in large language markets such as Germany, France, Italy and Spain, where there are enough viewers to cover the cost.
• Subtitles are the norm in many smaller markets, such as the Netherlands, the Nordic countries and Portugal, where dubbing every programme would be expensive. Children’s programmes are usually dubbed even there, because young children can’t read subtitles.
• Voice-over is a third option. In Poland, for example, many foreign TV programmes use a single narrator, called a lektor, who reads the translated lines over the quietly audible original soundtrack.

Streaming has blurred these lines. Major platforms now offer many programmes with several dubbed tracks and subtitle languages, and viewers in every country choose for themselves.

Which is better?

Neither, really. Subtitles keep the original actors’ voices and performances, but you have to read, and some of the dialogue is inevitably condensed. Dubbing lets you watch the picture freely, but you lose the original voices, and the script has been reshaped to fit.

If you want to stay closest to the original meaning, subtitles translated from the original language are usually the better choice. If you want to follow a fast-paced scene without reading, or you’re watching with children, a good dub can be just as enjoyable.