The keyman lexical model can not differ letters with different tones

Hello keyman team.
I amd trying to do a vietnamese Chu Nom keyboard
However, I found that the keyman lexical model can not differ letters with different tonations, and let the whole keyboard can not work.

the second issue is I don’t know how to set the word break. To be honest, in many languages, the space is not the word breaker, including in the English, the words often come out in pairs. We should add those comibination in to our model, too. However the current lexical model will detect and space as a word breaker.

The third issue is I can not set specific punctuations as a letter instead of a word breaker, for example I want to set ` as a page down mark, however the keyman model detect it as a word breaker.

The custom icon in fact there is nothing inside, I can not set anything inside.

Could you told me how to solve those issues?

Hi @Cathaylab

The lexical-model guide for advanced topics may help answer some of your questions
https://help.keyman.com/developer/current-version/guides/lexical-models/advanced/

I was going to give the same link – I think your all three of your questions are answered by the topics there (refer to the ‘search key’ topic for handling different tonations)

I am sorry I don’t quite understand the tutorial
For example. I want to differ the vietnamese letters with different tones and not differ upper and lower cases.
How can I make it?
I am developing a vietnamese Chu Nom IME

I want to make a function that will automatically delete the letter parts in the output when generate the result. The vietnamese keyboard only contain hundred of latin letters. If I can set a function and automatically delete the latin letters part, it will be perfect.
If we can make it,
I think we have already made a IME based on the Keyman on the phone.

It is very great that our latest version keyman allows 8 selections on the lexical model.
It makes developing IME possible on the phone.
For example, can i store the data as ‘abc日’
then I can set a funtion to delete the ’ and those letters and remain the 日,
Can I realize it on the keyman lexical model? Or can I integrate it into the keyboard

Is it possible to have more selection on the keyman lexical model? for example can we have 30 selections. Because most of languages based on Chinese Characters will have 30 characters with same pronunciation at most.

Due to inactivity, this topic is now closed. Please feel free to open a new topic about a new issue.

Thank you!