Goes too fast over proper nouns

[h3xx]

I was reading Azure Bonds, and the reader went waaay too fast over proper names not in the story before. The text was:

"This scroll was written by the hand of the Arch-cleric Mzentul himself. It should remove these horrors without delay."

On both "Arch-cleric" and "Mzentul" it went way too fast to read those words.

A possible compromise would be to do a dictionary check and increase the delay if it's not in the dictionary, and increase the delay for hyphenated words.

[kenakofer]

Thanks for using this project!

Yeah, this does seem like an area for improvement. Perhaps the simplest solution is when computing phrase display time, there should be a factor increase when it includes:

• capital letters (since missing a name while reading is more costly than missing other words)
• hyphens or other punctuation that aren't very common.

I don't think we want dictionary checks or back-searching, though I can see the appeal. It might be heavy to process, given how the parser "reads" the entire book every time WPM changes (why did I write it like that???).

I haven't touched this project in many years, but please feel free to submit a pull request along these lines if you make changes that work for you!