Tait Hoyem

Hello Thomas, Yes this is indeed the case, Odilia does not currently respect associated labels yet. Thank you for reporting this.

@albertotirla can you see if this issue is fixed now?

Here is an example of how this should work: Instead of a function like `update_string_insert(start_pos, update_length, updated_text) -> impl Fn(&mut CacheItem) + '_` this would become: `update_string_insert(text_caret_moved_event) -> OdiliaEvent`. ```rust...

This effectively will turn the majority of the Odilia codebase into a bunch of mapping functions from `atspi` events to Odilia events. Then, state management would simply be the execution...

This would also make Odilia a very "accessible" codebase (pun not intended) for newcomers, since the implementation details of *how* to do something can quite easily be hidden from first-time...

> However, for disambiguating languages with the same writing system, it hogs a lot of memory, around 80mb extra, and should be opt in. There are ways to reduce this...

> If you don't load the model and just use lingua's rule-based disambiguation for easy languages comparisons like English vs Arabic vs Russian vs Chinese, it never has this upfront...

> changing the possible languages to be detected in a text would require an odilia restart, but that'll probably not be an issue, unless we're willing to load all the...

> [In my testing](https://github.com/C-Loftus/MultilingualSpeechDispatcherClient/blob/master/main.go#L80) at least lingua appears to have that 100ms latency on the first recognition if it has to lazy load the model, but after that, recognitions are...

I think this method severely limits what we can do with speech-dispatcher given that we are (slowly) moving to an SSIP-based library. This would restrict a lot of the functionality...