Wednesday, May 15, 2013

Criteria for Quality Subtitling


Subtitles must reproduce the intent of an original work that would otherwise be inaccessible or incomprehensible to a given audience.


Subtitles are defined as lines of text reproducing language content in screen media.

The process of subtitling involves controlling three key elements:
  • Time
  • Space
  • Content

The cueing of subtitles must follow the speech rhythm, taking cuts and sound bridges into consideration. Dialogue becomes incomprehensible if a subtitle enters at the wrong time.

The amount of text in a subtitle must be consistent with the viewers' reading speed. Achieving this usually requires compressing the dialogue; if the amount of text in the subtitle is too great in comparison with its exposure time on the screen, the audience won't have time to read it.

The translation must reproduce the message conveyed in the original work. Errors in translation or language will distract the viewer.

The subtitle's cueing, brevity, and content are factors that help to convey the intent of the original work. If any of these elements are inadequate, viewer perception will be impeded, and this may cause the audience to discontinue their viewing.


A lot of work has gone into all the aspects of any production. Subtitling alters the original, and consequently, certain rules must be observed in order to ensure that the subtitles blend in with the original work. High-quality subtitling enhances the overall experience for the target audience.

Time and Space
The viewer must be given enough time to take in both image, sound and subtitle.

Subtitles should be cued to match the dialogue, and adjusted according to the cuts. Sometimes it is best to cue in the subtitle at the cut instead of waiting for the corresponding dialogue. The subtitle should preferably be cued out at the cut. A subtitle may only linger after a cut if this is within the same scene, and must be cued out if a cut occurs from one scene to another.

Consecutive subtitles must be cued using a fixed interval, which can be set to 2, 3, or 4 frames and must be consistent throughout the entire translation. The fixed frames interval must be used if the pause between two subtitles is less than one second, unless several cuts occur during this time.

Context, language, audio and visual editing each have a bearing on the amount of text one subtitle may contain. A subtitle of two full lines requires a duration of approximately 6 seconds, while a subtitle of one full line requires approximately 3 seconds (i.e. 12 characters per second). Subtitles with a duration of less than one and a half seconds should be avoided.

Each subtitle should be semantically self-contained.

Idiomatic phrases and expressions should always be used. The subtitle content should not contradict the dialogue or image (e.g. body language). Replacing the names of people or places with the names of other people or places should be avoided; such names should be explained or, if necessary, translated.

The style of the target language must be equivalent to the style of the source language. Converted measurements and numbers must be rounded to an equivalent degree of precision.

National subtitling standards for punctuation, italics, text alignment, continuation and dialogue indicators must always be followed.

Click for criteria in Word

The criteria have so far been ratified by the following audio-visual translation unions:

FBO, Denmark
NAViO, Norway
TF section 108, Sweden