Auto-create lexical models

I really like the way that SAB allows us to download a Keyman keyboard package from the Keyman site (Keyboard Search) to be included in the app. This is now working really well, and predictably. Kudos to the SAB development team!

I’m wondering whether we can take this functionality to the next level in order to enhance the user experience (especially when they are searching for a word), but without putting an additional load on the App developer. Hear me out…

Pain point: Currently, when I click on the Search/Find icon, I have to type in the word that I’m looking for. But what if I don’t know the exact spelling of the word, or what it it is a really long word that is easy to misspell. I spend a lot of time searching in vain.

Opportunity: If we could automatically generate a Keyman lexical model [from the words in the text that is included in the App] then the user will be able to make use of (word-level) suggestions as they type, and won’t have to remember the exact spelling of words.

I understand that Keyman already allows a lexical model to be generated from the command line, so there’s no reason why we couldn’t do this for the App builder person and enhance the usability for their Scripture/Reader/Dictionary app(s).