{"course":{"slug":"visual-storytelling","title":"Making It Say Something","url":"https://addaly.com/learn/visual-storytelling","level":"beginner","track":"create"},"lesson":{"slug":"captions-and-subtitles","title":"Most of your audience has the sound off","url":"https://addaly.com/learn/visual-storytelling/captions-and-subtitles","position":36,"of":64,"minutes":10,"answer":"Captions are limited by reading speed rather than screen space (two lines, around 17 characters a second, broken at clause boundaries) and because a viewer reading text cannot study the picture at the same time.","keyIdea":"Captions are limited by reading speed rather than screen space (two lines, around 17 characters a second, broken at clause boundaries) and because a viewer reading text cannot study the picture at the same time, captioning is a structural decision rather than a final step."},"body":{"format":"markdown","text":"## Two different things with one name\n\n- **Subtitles** assume the viewer can hear but does not understand: they carry the dialogue, often translated.\n- **Captions** assume the viewer cannot hear: they carry the dialogue *and* the sound that matters: `[machine starts]`, `[laughs]`, the speaker's name when it is not obvious who is talking.\n\nThe distinction is not pedantry. A video whose meaning turns on a sound (a crack, a door, a change in a machine's note) is unintelligible with subtitles and intelligible with captions, and for a deaf viewer that is the difference between a piece that includes them and one that does not.\n\nThen there is the third group, which is most people: viewers with working ears watching in a queue, in bed, in an office, on a train, with the sound off because the platform started it muted and turning it on is a decision. Captions serve all three at once, which is why they have gone from an accessibility duty to the default.\n\n## The constraint is reading speed, not screen space\n\nThe mistake beginners make is fitting the words on screen. The real limit is whether they can be read in the time they are there, while the viewer is also watching the picture.\n\nThe broadcast conventions exist because of this:\n\n- **Two lines maximum.** Three lines covers too much picture and forces the eye to travel.\n- **Around 37 to 42 characters per line.** Wider than that and the eye has to sweep, which costs a fixation.\n- **Roughly 17 characters per second**, or about 160 to 180 words a minute, as a ceiling for comfortable reading. Faster is legible and exhausting.\n- **A minimum of about one second on screen**, even for one word, because anything shorter registers as a flash rather than as text.\n\nWhen the speech is faster than the reading limit, you edit the caption. Condensing speech into readable text is a real skill and it is not cheating; a verbatim caption that cannot be read has served nobody.\n\n> Figure: The caption limits, and what sets each one. Layers, top to bottom: Two lines maximum (Three covers too much picture); 37 to 42 characters a line (Wider costs the eye a fixation); About 17 characters a second (Faster is legible and exhausting); One second minimum on screen (Shorter registers as a flash); Break at clause boundaries (Reading grabs phrase-sized chunks). None of these limits is about fitting the words on screen. They are all about whether the words can be read in the time they are there, while the viewer is also watching the picture.\n\n## Break lines where the sense breaks\n\nThis is the single largest quality difference between amateur and professional captions, and it costs nothing.\n\nBad:\n\n```text\nthe machine jams on thick denim which is\nwhy we used to turn the work away\n```\n\nBetter:\n\n```text\nthe machine jams on thick denim\nwhich is why we used to turn the work away\n```\n\nBreak at clause boundaries: before a conjunction, before a preposition, between a subject and a long predicate. Never split an article from its noun or a name across two lines. The reason is mechanical: reading proceeds by grabbing phrase-sized chunks, and a break in the middle of a chunk forces a re-read of both halves.\n\n## The file, and why a sidecar beats burning in\n\nSubtitles live in a plain-text file, usually `.srt`, which is readable and editable in any text editor:\n\n```srt\n1\n00:00:04,120 --> 00:00:06,800\nthe machine jams on thick denim\n\n2\n00:00:06,800 --> 00:00:09,400\nwhich is why we used to turn the work away\n```\n\nA **sidecar** file like this can be switched off, translated, indexed by search engines and read by screen readers. **Burned-in** captions are painted into the picture and cannot be any of those things, but they are the only kind that survives being downloaded and re-uploaded, which is why social platforms are full of them.\n\nThe defensible answer for most people is both: burn in for the vertical social cut, supply an `.srt` everywhere that accepts one.\n\n## Automatic captions, and exactly where they fail\n\nAutomatic speech recognition is now good enough that starting from it and correcting is faster than typing from scratch. It fails in a predictable pattern, and knowing the pattern makes the correction pass quick:\n\n- **Proper nouns**: names of people, places, businesses, products. The model has never seen your client's surname.\n- **Numbers and units**, particularly currencies and measurements.\n- **Accents outside the training distribution**, which in practice means most speakers outside a narrow band of British and American English. Published error rates differ by accent and dialect by a factor of several, and the direction of that difference is consistent enough to plan around.\n- **Overlapping speech**, which it renders as a plausible single sentence that nobody said.\n\nThe last one is the dangerous failure, because it is fluent. It does not look like an error.\n\nFree tools for the whole job: YouTube's automatic captions with its editor, Whisper if you can run it locally, Subtitle Edit on Windows and Aegisub cross-platform for timing, CapCut and DaVinci Resolve's free version for burning in. None of this needs a subscription.\n\n## The limitation\n\nCaptions change what your piece is. A viewer reading captions is spending a share of their attention on text, and cannot look at the picture at the same time: the eye cannot be in two places. This means a caption over a shot that requires close looking is a genuine conflict, not a solved problem.\n\nThe consequence for the work is upstream: shots that need to be *studied* should not coincide with dense speech. That is an editing decision made at the structure stage, and it is one of the places where accessibility stops being a checklist item added at the end and becomes something that shapes the piece."},"check":{"question":"Why does breaking a caption line after \"which is\" rather than after \"denim\" measurably slow reading?","options":[{"text":"The second line starts with a lower-case word, which readers briefly interpret as a continuation error.","correct":false,"tempting":"Points at capitalisation, but the same slowdown occurs with sentence-case captions where both lines begin identically."},{"text":"It puts more characters on the first line, pushing it past the comfortable line-length limit.","correct":false,"tempting":"Treats it as a length violation, yet the break damages reading even when both arrangements sit well inside the character limit."},{"text":"Reading proceeds in phrase-sized chunks, so a break inside a chunk forces both halves to be re-processed.","correct":true},{"text":"Eye movement between lines is slower than movement within a line, so any two-line caption should be avoided where possible.","correct":false,"tempting":"Generalises a real cost of line changes into a rule against two-line captions, which broadcast practice contradicts and which does not distinguish good breaks from bad ones."}]},"prev":{"slug":"writing-for-the-ear","title":"A listener cannot go back","url":"https://addaly.com/learn/visual-storytelling/writing-for-the-ear"},"next":{"slug":"image-and-caption-together","title":"The caption decides what the photograph means","url":"https://addaly.com/learn/visual-storytelling/image-and-caption-together"},"authorship":"Founder and series editor: Dr Mustafa Mun. Written by AI models to a written standard. Exam keys checked blind by a second model: 3,665 of 3,666 agreed. Academic review invited.","review":{"status":"Draft, open for academic review"},"cite":"Addaly (2026) 'Most of your audience has the sound off', in Making It Say Something. Edited by M. Mun. Addaly. Available at: https://addaly.com/learn/visual-storytelling/captions-and-subtitles (Accessed: 13 October 2026).","licence":{"name":"CC BY-NC-SA 4.0","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"reader":{"schema":1,"title":"Most of your audience has the sound off","courseTitle":"Making It Say Something","module":{"index":5,"slug":"words-that-carry","title":"Words that carry the picture"},"answer":null,"keyIdea":"Captions are limited by reading speed rather than screen space (two lines, around 17 characters a second, broken at clause boundaries) and because a viewer reading text cannot study the picture at the same time, captioning is a structural decision rather than a final step.","hero":{"kind":"figure","figure":{"t":"figure","label":"Figure","figure":{"type":"stack","title":"The caption limits, and what sets each one","layers":[{"label":"Two lines maximum","detail":"Three covers too much picture"},{"label":"37 to 42 characters a line","detail":"Wider costs the eye a fixation"},{"label":"About 17 characters a second","detail":"Faster is legible and exhausting"},{"label":"One second minimum on screen","detail":"Shorter registers as a flash"},{"label":"Break at clause boundaries","detail":"Reading grabs phrase-sized chunks"}],"caption":"None of these limits is about fitting the words on screen. They are all about whether the words can be read in the time they are there, while the viewer is also watching the picture."},"alt":"Figure: The caption limits, and what sets each one. Layers, top to bottom: Two lines maximum (Three covers too much picture); 37 to 42 characters a line (Wider costs the eye a fixation); About 17 characters a second (Faster is legible and exhausting); One second minimum on screen (Shorter registers as a flash); Break at clause boundaries (Reading grabs phrase-sized chunks). None of these limits is about fitting the words on screen. They are all about whether the words can be read in the time they are there, while the viewer is also watching the picture."}},"blocks":[{"t":"h2","id":"two-different-things-with-one-name","runs":[{"text":"Two different things with one name"}]},{"t":"ul","items":[[{"text":"Subtitles","b":true},{"text":" assume the viewer can hear but does not understand: they carry the dialogue, often translated."}],[{"text":"Captions","b":true},{"text":" assume the viewer cannot hear: they carry the dialogue "},{"text":"and","i":true},{"text":" the sound that matters: "},{"text":"[machine starts]","code":true},{"text":", "},{"text":"[laughs]","code":true},{"text":", the speaker’s name when it is not obvious who is talking."}]]},{"t":"p","runs":[{"text":"The distinction is not pedantry. A video whose meaning turns on a sound (a crack, a door, a change in a machine’s note) is unintelligible with subtitles and intelligible with captions, and for a deaf viewer that is the difference between a piece that includes them and one that does not."}]},{"t":"p","runs":[{"text":"Then there is the third group, which is most people: viewers with working ears watching in a queue, in bed, in an office, on a train, with the sound off because the platform started it muted and turning it on is a decision. Captions serve all three at once, which is why they have gone from an accessibility duty to the default."}]},{"t":"h2","id":"the-constraint-is-reading-speed-not-screen-space","runs":[{"text":"The constraint is reading speed, not screen space"}]},{"t":"p","runs":[{"text":"The mistake beginners make is fitting the words on screen. The real limit is whether they can be read in the time they are there, while the viewer is also watching the picture."}]},{"t":"p","runs":[{"text":"The broadcast conventions exist because of this:"}]},{"t":"ul","items":[[{"text":"Two lines maximum.","b":true},{"text":" Three lines covers too much picture and forces the eye to travel."}],[{"text":"Around 37 to 42 characters per line.","b":true},{"text":" Wider than that and the eye has to sweep, which costs a fixation."}],[{"text":"Roughly 17 characters per second","b":true},{"text":", or about 160 to 180 words a minute, as a ceiling for comfortable reading. Faster is legible and exhausting."}],[{"text":"A minimum of about one second on screen","b":true},{"text":", even for one word, because anything shorter registers as a flash rather than as text."}]]},{"t":"p","runs":[{"text":"When the speech is faster than the reading limit, you edit the caption. Condensing speech into readable text is a real skill and it is not cheating; a verbatim caption that cannot be read has served nobody."}]},{"t":"h2","id":"break-lines-where-the-sense-breaks","runs":[{"text":"Break lines where the sense breaks"}]},{"t":"p","runs":[{"text":"This is the single largest quality difference between amateur and professional captions, and it costs nothing."}]},{"t":"p","runs":[{"text":"Bad:"}]},{"t":"code","lang":"text","label":"Text","text":"the machine jams on thick denim which is\nwhy we used to turn the work away"},{"t":"p","runs":[{"text":"Better:"}]},{"t":"code","lang":"text","label":"Text","text":"the machine jams on thick denim\nwhich is why we used to turn the work away"},{"t":"p","runs":[{"text":"Break at clause boundaries: before a conjunction, before a preposition, between a subject and a long predicate. Never split an article from its noun or a name across two lines. The reason is mechanical: reading proceeds by grabbing phrase-sized chunks, and a break in the middle of a chunk forces a re-read of both halves."}]},{"t":"h2","id":"the-file-and-why-a-sidecar-beats-burning-in","runs":[{"text":"The file, and why a sidecar beats burning in"}]},{"t":"p","runs":[{"text":"Subtitles live in a plain-text file, usually "},{"text":".srt","code":true},{"text":", which is readable and editable in any text editor:"}]},{"t":"code","lang":"srt","label":"Srt","text":"1\n00:00:04,120 --> 00:00:06,800\nthe machine jams on thick denim\n\n2\n00:00:06,800 --> 00:00:09,400\nwhich is why we used to turn the work away"},{"t":"p","runs":[{"text":"A "},{"text":"sidecar","b":true},{"text":" file like this can be switched off, translated, indexed by search engines and read by screen readers. "},{"text":"Burned-in","b":true},{"text":" captions are painted into the picture and cannot be any of those things, but they are the only kind that survives being downloaded and re-uploaded, which is why social platforms are full of them."}]},{"t":"p","runs":[{"text":"The defensible answer for most people is both: burn in for the vertical social cut, supply an "},{"text":".srt","code":true},{"text":" everywhere that accepts one."}]},{"t":"h2","id":"automatic-captions-and-exactly-where-they-fail","runs":[{"text":"Automatic captions, and exactly where they fail"}]},{"t":"p","runs":[{"text":"Automatic speech recognition is now good enough that starting from it and correcting is faster than typing from scratch. It fails in a predictable pattern, and knowing the pattern makes the correction pass quick:"}]},{"t":"ul","items":[[{"text":"Proper nouns","b":true},{"text":": names of people, places, businesses, products. The model has never seen your client’s surname."}],[{"text":"Numbers and units","b":true},{"text":", particularly currencies and measurements."}],[{"text":"Accents outside the training distribution","b":true},{"text":", which in practice means most speakers outside a narrow band of British and American English. Published error rates differ by accent and dialect by a factor of several, and the direction of that difference is consistent enough to plan around."}],[{"text":"Overlapping speech","b":true},{"text":", which it renders as a plausible single sentence that nobody said."}]]},{"t":"p","runs":[{"text":"The last one is the dangerous failure, because it is fluent. It does not look like an error."}]},{"t":"p","runs":[{"text":"Free tools for the whole job: YouTube’s automatic captions with its editor, Whisper if you can run it locally, Subtitle Edit on Windows and Aegisub cross-platform for timing, CapCut and DaVinci Resolve’s free version for burning in. None of this needs a subscription."}]},{"t":"h2","id":"the-limitation","runs":[{"text":"The limitation"}],"role":"limit"},{"t":"p","runs":[{"text":"Captions change what your piece is. A viewer reading captions is spending a share of their attention on text, and cannot look at the picture at the same time: the eye cannot be in two places. This means a caption over a shot that requires close looking is a genuine conflict, not a solved problem."}]},{"t":"p","runs":[{"text":"The consequence for the work is upstream: shots that need to be "},{"text":"studied","i":true},{"text":" should not coincide with dense speech. That is an editing decision made at the structure stage, and it is one of the places where accessibility stops being a checklist item added at the end and becomes something that shapes the piece."}]}],"check":{"question":[{"text":"Why does breaking a caption line after “which is” rather than after “denim” measurably slow reading?"}],"options":[{"text":[{"text":"The second line starts with a lower-case word, which readers briefly interpret as a continuation error."}],"correct":false,"tempting":[{"text":"Points at capitalisation, but the same slowdown occurs with sentence-case captions where both lines begin identically."}]},{"text":[{"text":"It puts more characters on the first line, pushing it past the comfortable line-length limit."}],"correct":false,"tempting":[{"text":"Treats it as a length violation, yet the break damages reading even when both arrangements sit well inside the character limit."}]},{"text":[{"text":"Reading proceeds in phrase-sized chunks, so a break inside a chunk forces both halves to be re-processed."}],"correct":true,"tempting":null},{"text":[{"text":"Eye movement between lines is slower than movement within a line, so any two-line caption should be avoided where possible."}],"correct":false,"tempting":[{"text":"Generalises a real cost of line changes into a rule against two-line captions, which broadcast practice contradicts and which does not distinguish good breaks from bad ones."}]}],"order":[1,0,3,2],"why":null},"prev":{"slug":"writing-for-the-ear","title":"A listener cannot go back","minutes":9},"next":{"slug":"image-and-caption-together","title":"The caption decides what the photograph means","minutes":8}}}