English SDH subtitling services

Three things, in the main. Sound that carries meaning gets a label, so a knock at a door or a car braking off camera reaches a viewer who cannot hear it. Speakers are named wherever the picture does not make it obvious who is talking. Music is marked, with lyrics written out where the words matter and a cue described where the mood is doing the work. The dialogue itself is the same text, spotted the same way.

By asking whether the material still makes sense without it. The DCMP Captioning Key sets exactly that test and we work to it: a sound goes in where understanding or enjoyment depends on it, and stays out where it is wallpaper. The label sits in brackets, names its source unless the source is on screen already, and picks a specific word rather than a vague one. Distant traffic rarely earns a line; a door slamming behind the speaker usually does.

The line moves. Ofcom's access services guidelines say subtitles come off the bottom of the frame when they would cover a speaker's mouth or vital information, and the FCC counts placement among its four caption quality standards, defining it as captions not blocking faces, featured text or anything else essential. Platform guides go further on titles, keeping on-screen text out of any subtitle that carries dialogue. We agree the available room at the brief.

No, and the guidance says not to. The English platform style guide asks for a speaker label only where the person talking cannot be worked out from the picture, which stops the brackets crowding out the dialogue. Ofcom allows the same job to be done by colour, by punctuation or by moving the line. Where a character has not been named yet, the label describes instead of naming: a man, a woman, a voice on the phone.

It runs from £17 for each minute of video: our English subtitling rate of £15, plus the £2 the rate card adds for SDH and closed captions. The audio moves the figure more than the running time does, because a drama with a dense sound design carries a judgement on every effect where a talking-head interview carries almost none. Transcription, burnt-in output and the number of formats do the rest. Every quote is itemised, with our Price Match Promise behind it.

Usually, and that is most of what we produce: English audio with English text laying the accessibility information over it. Accessibility teams ask for precisely that, and so does a social feed that plays without sound. SDH is also written for translated content, where the dialogue arrives in English and the sound labels, speaker names and music cues are written in English beside it. Tell us what the source audio is and we will confirm which of the two you want.

Yes, and it usually brings the cost down. We take your timings, check them against the picture, then add what the file is missing: sound labels, speaker identification, music cues, and the line moves that keep all of it clear of your graphics. Where the original spotting is too tight to hold the extra text, we re-cue rather than let a line vanish unread. With nothing to start from, we transcribe the audio and build from there.

The ones you are likely to be working in. SRT and VTT carry the brackets and the speaker labels into Premiere Pro, Final Cut Pro, DaVinci Resolve and the browser-based editors. TTML and IMSC hold position as well, which starts to matter once lines have been moved off the bottom of the frame, and EBU-STL covers broadcast delivery. If a sidecar file is more bother than it is worth, we burn the SDH into a finished MP4 or MOV.