Reading time

Ability to estimate how long a piece would take to read (aloud) would be a handy feature.

Export to PDF in TNR 12 point on A4 page with 2 inch margins all round. An ordinary text leaded 14.4 (standard line spacing) gives 48 lines per page. Some at the end of paragraphs will be short lines, so take an estimate on how much that will reduce the lines per page to whole lines in your style of writing. Pages at the end of your chapters may be short pages so make an allowance for that in calculating the number of pages. Lines per page times number of pages gives you an estimate of the total number of lines. In my recording experience, a line of TNR 12 point on A4 with 2 inch margins left and right takes about 5 seconds to read at recording speed, so 12 lines = 1 minute. Divide the total number of lines by 12 and you have an estimate of how long in minutes it would take to read that text.

That’s my experience, on texts including an admixture of long words and numbers (which are very condensed on the line compared with the time they take to speak aloud … 77 is only two n-spaces but roughly .75 of a second, 727 is only 3 n-spaces but will take about 1.5 seconds).

I guess you could do it within Scrivener if you set your editor margins to give a line length close to that I give and set Scrivener to display line numbers … open a scrivening session to include all the text that you want to estimate. That would give you the number of lines with the caveat of short lines.

To me, asking Keith to program in a system that will do those calculations would be bloat and probably less accurate as people use different margin settings in their work and so it would only make sense to count words, which would not take into account the length of those words … “antidisestablishmentarianism” anyone?

I’ve given you my experience, as a fluent reader aloud having done many recordings. Your reading speed may vary from mine, as may everyone else’s, so it’s best to work it out for yourself by printing out an average page, recording it without hesitation, repetition or deviation and then see how long it takes you.


Edited to check and change calculation. 2 cm margins = 6 seconds.

That is my experience too, in occupations where getting it right (live TV) is important. Reading time depends to quite a significant degree on the reader (and, unless the reader consciously self-monitors, can vary from day to day and from time to time even then).

I’m genuinely very surprised at the suggestion, because I can’t see how it could be done, for a couple of reasons. First is the enormous variation between individuals, as the others have pointed out. I know a professor of psychology whose nickname is “Kalashnikov” because he speaks so fast, and I recall meeting someone from the north-west of England who told me that when he went to university in East Anglia, he thought all the locals were brain-damaged because they spoke so slowly.

But aside from that, there is an aspect of any single text that would be impossible to factor into the calculations – rhythm. Try reading a piece that has a poor rhythm, and it creates a kind of friction that slows everything down. A really rhythmic piece will flow off the tongue much more quickly (and I’m not talking about poetry here – I’m talking about rhythmic prose).

I also had a bizarre experience when reading a book that had been written by a Welshman – I found it quite hard until I tried reading it in a Welsh accent. It then became much easier.

However, thanks to Mark for his advice on the subject. Being from East Anglia, I will probably have to double the length of time …

Best, Martin.

Thanks Martin. That is my experience of recording hundreds of dubbing tracks over my time here, particularly at the TV Station. Since I had to fit my edited text and reading to the time taken for the recorded version of the equivalent Chinese paragraph, that was the rule I followed, and generally it worked very well. Spoken English takes up to one and a half times as long to say exactly the same thing as the Chinese, so it was important to edit down.

On one, hugely important recording, timing was so crucial — I had to record in parallel with the Chinese voice-over — that I went right down to counting syllables, so that my version was within one or two syllables of the Chinese original. Counting syllables in Chinese is a walkover, as each character is a single syllable; doing it for the English was a real pain. But I loved the expression on the editors’ faces when they played the resultant tape to check, and found that they weren’t going to have to make even a single change to the video to accommodate the English.

Chinese news-readers are something else, though … I think they’d make your “Kalashnikov” seem like an East Anglian! Makes news on Chinese TV incredibly hard to follow for us poor "lao wai"s. :slight_smile:


Edit: And I agree about rhythm, so I tell my students that whatever language they’re writing in, they should “write with their ears”! I try to make my texts rhythmic for that same reason, but it’s an uphill struggle with Chinese students, as Chinese, like so many languages is syllable-timed, where English is stress-timed, and they don’t hear the stresses any more than most of us hear their tones. My favourite piece of rhythmic English is Donne’s Meditation 17, I think it is … the “No man is an island” one — not for its sentiments but for the sheer beauty and rhythm of the language.

Donne is a wonderfully rhythmical writer. I also love the first sentence of Pride and Prejudice – a really brilliant piece of rhythmical writing. You can tell where the tone turns from serious to facetious just by listening to the change of rhythm. They don’t write them like that any more.

Best, Martin.

Ha! I wonder what Chinese into French is like? I used to find that French, especially in the mouths of French politicians, took as much as 25 per cent as long again to speak as the exact English translation. It would require some inventive expansion in order that the viewer wouldn’t feel short-changed. Going the other way would, I imagine, be difficult; going from French to Chinese even more so.

This is the reason why the various script formats are so demanding. One script page in standard format = about one minute of performance time.


So what your saying is that I would still need to cut things by 50% if I was providing english translations of french.

I haven’t thought about Chinese v French.

But I do have the suspicion that Gallic politicians have a prediliction for orotundity in their expostions to give an impression of intellectuality, consequentially overburdening their orations with a plethora of multisyllabic latinate verbosity.



No Jaysen, according to Hugh, you’d have to insert a further 25% of padding into your English, to make it match the time taken to say it in French. If that is the case, the amount of creative editing you’d have to do in French to match the timing of Chinese would be even more demanding than for English!


When I told a friend of mine that the draft of the book I was writing was too long, he suggested leaving out every other word …


Mr X,

The intention was a self-depricating comment that would lead to guffaws and general feelings of amusement as the readers who are familiar with my lengthy (and often parenthetical (aways done with the intent of creating sub thought within the main thought (except on those rare occasions where I am attempting to make a point))) demonstrations of fully “compliant” abuses of the english language often seen in “these here parts” - to use a colloquialism from my youth - in juxtaposition to the comment made by his Hugh-ness to show that the French may own the market on deep fried, julienned potatoes (what you call “chips” and we call “french fries”).

Or, as the chinese might have said that:

Score: Jaysen - 110, Chinese - 4

Odd that my word count per sentence is very close to my golf score. For nine holes. I bet there is a correlation there somewhere.


That’s german.

Brilliant word! My daughter’s got her ‘A’-level German exam coming up soon. I’m going to dare her to work that fabulous word into her answer paper. :smiley:

Ah, Jaysen, self-deprecation will get you everywhere :smiling_imp:

Actually, there are two issues: (i) your prolixity in verbiage and overenthusiasm for self-referential, self-embedding, parenthetical and convoluted structural chains when attempting the conveyance of your own cogitations; and (ii) the issue of when the res gestae are the re-interpretation and reformulation of conceptualisations expressed by others …

But your Chinese is not quite right … Chinese, in spite of its condensed nature still works on the principle of “saying it two ways is better than saying it one way; using two expressions is better than using one”, so your “translation” should read:


Ahh German! But is it germane? :laughing:

Whatever the answer, I am not going to get into research into the temporal relations between a Chinese text and its German translation … I leave that to others … Chinese and English is enough of a brain-twister. Mind you, I have at times had to act as a French-Chinese-French interpreter, in consecutive mode, not simultaneous, thank heavens.

I will say this, though: German and French, like Chinese, are syllable timed languages, so if you need to control timing by counting syllables, it would be easier for French and German than for English. I suppose someone is now going to come up with Danish, with its apparent habit of swallowing entire syllables … :unamused:


Sounds like you’re overcomplicating the swing thought. One should only have a simple swing thought – ideally just one word encapsulating a single idea, so that the swing is pure rhythm seamlessly transferring your vision of the shot into graceful, powerful accuracy.

Please feel free to borrow my swing thought. I’m sure you will find it tremendously useful.

It’s ‘onepiecetakeawayrotatetheclubfaceverticallyleftshoulderunderthechin-cupwristrotatehipkeeplagtakeballfirstthengrassohfuckwhere’sthatgone’.

You’re welcome.



That makes me all tingly. I’ll have to remember that.

In ENPS (system for controlling a lot of stuff, including the read script, in TV News broadcasting) there is a variable, reading speed, which we have to manually tune for each person reading. It is fairly accurate and is measured as characters per second.

We do a test recording, and then compute the variable for each.

For each entry in the script there is a total which takes the text written and divides it by the variable and computes the lenght of the text in seconds.

Some professionals can write scripts that will fit exactly inside the alloted time, even without referencing the computed time. They can come with a handwritten script, sit down in front of the mike, read it in one take, and it will be on the second.

It all comes down to experience.