Love the feedback! Something to do with the language names. Each of those language names were decided rather carefully, since, for example, catalan was understood in a different word every time, but the word "catalonian" was transcribed with perfect accuracy. One possible solution would be to test for a variety of different words that sound like the keyword, similar to what BTC did with the new "Ok Wiki" functionality, and that would require you awesome folks to test the words, and report back what they are transcribed as.
In addition to this, I forgot to add, there are multiple variants of several languages, such as english US, UK, Australian, South African; spanish Spain, etc. etc. I plan to add all of this functionality sometime soon, as well. For all of the languages the API supports that my little wrapper program could easily be changed to modify, they are listed in this StackOverflow post: https://stackoverflow.com/a/23733589/14719982. Feel free to speak the translated language names into the plugin, and see what they are transcribed as. (Since the default lang I have set is English-US, the language names must sadly be translated into English to be understood, such as Espanol to Spanish. This also accounts for maybe "improper" languages names, as pointed out.) On Wednesday, August 11, 2021 at 4:41:54 AM UTC-4 BurningTreeC wrote: > Hello all, I've now changed the *Command *command to *Ok Wiki* > > I've made it also detect Okay Vicky, Ok WC, and more keywords ... If those > of you who are testing this would report any other keywords we should add, > that would be a great help > > What should work right now is *Ok Wiki change language to <<language>> *and > *Ok Wiki stop talking* > > > > TiddlyTweeter schrieb am Mittwoch, 11. August 2021 um 10:28:45 UTC+2: > >> Ciao Flanc... & BTC >> >> *SIDE NOTES ON API BEHAVIOR* >> >> This does not effect anything YOU two are doing. Merely observations on >> the "behaviours" of the recognitions. >> But I do think it is useful to understand the end-product the API issues >> forth. >> >> *1 -* There are occasional QUIRKY things that happen that have no clear >> rationale. For instance if you dictate "cat dog mouse horse" you get back >> ... "cat dog Mouse horse". Maybe mice, or mouses, actually rule the >> universe as "The Hitchhiker's Guide To The Galaxy" contended"? >> <https://www.bbc.co.uk/programmes/profiles/1MmQjXrkwp98jNdc17j3tr3/mice> >> >> *2 -* Sometimes words that are not understood by the api are simply >> DISCARDED. (For instance in Italian mode most dictated English words are >> just ignored). >> >> *3 - * Sometimes words that are not understood by the api are >> capitalized as Proper Names. There is some very clever thing going on >> inside the recog. machine that decides when to do this. >> >> None of these minor points change anything other than the need to be >> slightly aware the API does stuff and awareness of that is helpful I think. >> >> I'll may add more comments about these behaviors after I used it more, if >> there any other obvious ones. >> >> Best wishes >> TT >> > -- You received this message because you are subscribed to the Google Groups "TiddlyWiki" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/tiddlywiki/9fb8bd77-c8db-4c18-9d24-f7197157b45an%40googlegroups.com.
