Word count, and some other questions


Scrivener seems like a pretty good software for a writer, but I have few special needs I have not found in it… either they don’t exist, or I’m just can’t find them. I write scripts for Visual Novel game, and those scripts are also used for Voice Over production. Actors do their own recording(s), so those scripts have guidance for the desired tone of voice and mood. VO actors price their work using word count, so that’s something I need to calculate. So… can Scrivener:

  1. Have a word count per actor appearing in a script, with some easy way? And don’t say that select each line and check that… that’s worse than what I do now, deleting temporarily everything else than the lines I need to have a word count. There must a better way.

  2. Have some information (blocks of text) that I can easily toggle off/on for the export to a .PDF. I want to have the acting guidance for a specific character and corresponding VO actor, but I do not want to have the the characters guidance visible - that’s just clutter. Currently I’m do the script, and save different versions of it to all actors in it, deleting that other actor guidance manually.

So… help? Can I do these things in Scrivener? Or is there other writing software out there that could do this?

As to point (1):

Scrivener has no built-in way to give you word counts per character.

However, I wrote a pair of scripts to produce word counts for each character in a script. One works directly on script text copied from Scrivener. The other (which is more script-format agnostic) analyses a script compiled to Word.

I recently posted them both /in this thread/, though the user I did this for did not manage to get them working.

Unfortunately, both are written in Applescript, so not useable for you on Windows. Still, the algorithm behind the latter script could easily be worked as a Word-based script. So, if you are into scripting Word, that could be a solution for you.

As for your second question (which deserves a fuller answer), there would be various ways to get such a result. How you might approach depends on whether this per-actor-part guidance can be relied on to live in paragraphs separate from the dialog (and other) text. If, so, achieving your end is pretty straightforward and just involves these ideas: Scriv lets you slice your script up into separate documents seamessly, documents can be assigned keywords and labels, compiling your script can be restricted to just those component docs that have certain keywords or label.

To flesh that out a bit: Scrivener lets you split your script up into as many document pieces as you want and easily fuses them on the fly. Thus, your guidance chunks can live in their own docs while still remaining in their interstitial position in the script. Since docs in Scrivener can be assigned keywords you would then be able to do a simply search which would yield just script text plus one actor’s guidance. This could then be compiled to output, and so on for the other actors.

If your guidance is not always in its own paragraphs, but must sometimes occur in-line with dialog (or other) text, then your task would be trickier.


p.s. Sounds a little like hell to have a group of actors who don’t have a common pagination in their scripts. But, hey, it’s your crazy!

Thanks! Now splitting everything is good - to a point. I like that we would/could have the whole story in separate documents per chapter (like we do now, only totally separate word documents…) but still being easily reachable and compilable.

But… having the guidance also in separate pieces would split the script (one chapter or so) to dozens, and that’s not practical. Here’s brief example how we have our scripts (can’t use colors here… first character lines should be red, and second blue):


Style / mood:
[friendly] [warmly] [casual]

I knew I could do better, by investing the profits on my own business, making it grow and prosper.

I love my Dad, but he is a lousy businessman…

Style / mood:
[friendly] [warmly] [smilingly]

”Daddy’s little pumpkin”, that’s what he always said…

Style / mood:
[friendly] [embarrassed]

Oh no, don’t say that… I hate that, please don’t call me ”pumpkin”.

Now when we’re sending the script to be voiced, we want to have guidance for only the character that he or she is doing, rest of the dialogue is just reference and context. We would (and did) use also Character names instead of colors, but we thinks they pop-out better when you’re reading the lines and acting them out at the same time (like they do).

It’s pretty simple things that we would need, but there is no software out there that can do it… there’s bunch of screenwriting softs out there, but they are pretty inflexible and do things by imitating sheets of paper and typewriter, right down to almost mandatory Courier font :slight_smile: We would like to get past that, to this century…

I think Scrivener is your best shot, here.
As said earlier by @gr, give your actors and VO’s a Label each, find and filter by their Label. Optionally gather them in a Collection.
Keywords could be used for the moods you want to communicate, dragging them onto the documents from a floating pane.
Use small documents per paragraph and select them based on their Label and/or Keywords or their Collections for inclusion in a Compile action. Alternate Colors using Section Layouts.
Scrivener is flexibele software if you think creatively about the features it offers. It’s not beginners stuff, but opportunities certainly exist. You should give it a try.

For this kind of very frequent in-line guidance you could use Scrivener’s replacement capabilities. Consider this slight adaptation of the layout—

Charles {friendly, warmly, casual}:

I knew I could do better, by investing the profits on my own business, making it grow and prosper.

I love my Dad, but he is a lousy businessman…

Amelia {friendly, warmly, smilingly}:

”Daddy’s little pumpkin”, that’s what he always said…

The important feature of the above is that each guidance is enclosed in a single matching pair of (otherwise unused) marks and that each such span is fronted by a string that uniquely identifies the character (i.e. their name).

When compiling, regular expressions (regExp) can be used to target the bracketed guidance for a single character, say Amelia and transform it slightly to square brackets

Amelia [friendly, warmly, smilingly]:

And a second regExp replacement then targets all remaining curly bracketed spans and eliminates them.

So, you can achieve what you want with a couple of lines of regExp replacement. Someone here could help you with the regExp formulas you’d need, if that is foreign to you.

I think in practice this would be easy to work with. To compile different version all you would need to do is change the character name in a standing regExp in the compile dialog box.

Anyway, that is a quick idea to at least demonstrate how do-able what you want is with Scrivener.


p.s. Getting the focal character’s lines to automatically compile to a font color or other special styling would be a further challenge.

p.p.s. I should mention that the above suggestion presupposes you are working in Scrivener’s normal writing mode, not the special scriptwriting mode. Perhaps the idea can be adapted to suit sxriptwriting mode too, but as it stands I think putting the guidance right after the character name in scriptwriting mode would cause the Scrivener to mistake the whole string (name plus guidance) for a character name which would be undesirable.

That’s interesting - I should be able to get around regExp, I’ll take a look at that… thanks!