Localising Audiovisual Content:
Monolingual Subtitles

In the first post of our series on Localising Audiovisual Content, we talked about transcription. Today, we will look at monolingual subtitles, also called same-language subtitlescaptions or intralingual subtitles (not to be confused with ‘interlingual subtitles’! We’ll talk about those in the next blog). This is a type of subtitling that has become omnipresent due to the rise of social media videos and increased legislation regarding accessibility.

Monolingual subtitles don’t require translation per se, that is true, but they do require the skills of a linguist, as will become apparent.

Monolingual subtitles: when you want to improve the accessibility of your content

Monolingual subtitles are subtitles which are in the same language as the audio.

But, why do you need subtitles in the same language as the audio?

  • Because your audience might not turn on the sound: 85% of Facebook videos are watched on mute.
  • Because it includes language learners in your audience: subtitles in the same language as the audio help language learners access your content and also improve their language level.
  • Because the Deaf and Hard-of-Hearing audience needs them to access the content. According to Action On Hearing Loss, there are 11 million people with hearing loss across the UK – that’s around one in every six of us, and the number is growing.

Whether captions are “closed” (which means that displaying them is an option) or “open” (they are always displayed with the video), there will always be someone who listens to the audio or reads the speaker’s lips and notices that the subtitles are not “verbatim”, or identical to the audio.

But this is not the subtitler’s mistake! There are reasons to adapt the wording of the captions. A subtitler has to keep the following factors in mind:

  • Character limits: In Teletext, the service used to display subtitles on some broadcast platforms, line length is limited to 37 fixed-width (monospaced) characters (like the Courier New font). Other platforms use proportional fonts. The BBC guidelines say that a line should be 68% of the width of a 16:9 video and 90% of the width of a 4:3 video. Netflix, on the other hand, sets the limit at 42 characters for alphabetical languages.
  • Reading speed: The BBC guidelines state that for a general audience to keep up with the subtitles, they should be displayed at 160-180 words per minute (WPM). Note that Romero-Fresco’s analysis of BBC programmes shows that in interviews and weather reports, the speakers reach beyond 240 wpm. In English, 180 WPM corresponds to the standard of 15 characters per second (CPS), though some online streaming services use 17-20 CPS.
  • Clarity: Subtitlers strive to keep subtitles free from meaningless hesitations and repetitions to make them coherent grammatical units that are easy to understand and fast to read – unless the repetition and/or grammatical mistakes are part of the way a character is presented, of course.
  • Line breaks: Captions tend to occupy one to three lines. In a two or three-liner, there are three things to be kept in mind when splitting the text: the character limit per line, the need for clarity (i.e. not splitting an article and a noun, a preposition and following phrase, etc.), and that the lines in a subtitle should be more or less the same length so that the eye has less distance to travel (and bottom-heavy so that less of the image is obscured).
  • Shot changes: Another thing to take into account are scene changes: professional subtitles are timed to shot change, which means that, when there is a shot change, subtitles should not linger on screen. They should end right before the shot change and start with the shot change because if a subtitle hangs over a shot change, the audience tends to reread it. Of course, timing to shot affects the timing of the subtitles, and therefore, the reading speed, and therefore, how much text we can afford to use in a subtitle.
  • General timing rules: Generally, a subtitle should stay at least one second on screen, and no more than 6 seconds. Depending on the type of subtitle, a gap of one or two frames should be left between subtitles to avoid flickering. And finally, there is a maximum deviation between the time that speech starts and the subtitles appear: around seven frames, though this number varies if there are shot changes. And each broadcaster may have their own rules.

As you can imagine, it’s necessary to condense information in order to create a subtitle that fulfils character limits and reading speed requirements. But there is a type of subtitling that needs additional information on the subtitle:

These are closed captions, which in many cases are the same as monolingual subtitles for the Deaf and Hard-of-Hearing (SDH). What may vary between them is their appearance: closed captions are displayed using white text on a black band, and SDH with a proportional font that has a contrast outline.

The difference between these and simple monolingual subtitles is that closed captions/SDH include audio cues that a hearing audience doesn’t need, like “PHONE RINGING” or “♪ [band plays Latin jazz beat] ♪”, etc. On TV, captions also tend to have different colours for different speakers. Each country has its own conventions regarding colours, use of italics, capital letters, etc.

It is important to take into account that the Deaf and Hard-of-Hearing audience is not always a native speaker of the captions’ language, but of a sign language. This implies that reading speed should not be pushed to fit all the speech and cues in.

Another kind of monolingual subtitling is live subtitling: subtitling live programmes such as sports or news. In live subtitling, a linguist listens to the audio of a programme and, simultaneously, repeats or rephrases what they hear to a speech recognition software, turning these utterances into scrolling subtitles. This is called respeaking, and requires a lot of training – not only does the linguist have to become very good at respeaking, but they have to spend a long time training the speech recognition software to recognise the particularities of their own speech profile.

Next in this series of posts is interlingual subtitles, also called translated subtitles!