Thanks for the insights! I'll certainly play with it - once I find some
time to do so...
However, this touches on a problem with my script in its current form.
Siri's speech recognition attempts to apply its own sense of context to
what it hears... which the script then supplies to LMS as the search:
parameter of the genres, artists, albums, and/or titles commands. Unless
Siri ended up with precisely the same spelling the server obviously
won't provide a match. It works most of the time for me, but there are a
few artist and album names that rarely, if ever, work.
It's an issue I've seen with many voice control systems: while we would
like to talk in German (or whatever other language) to those systems,
many of our artists' names are English. Even worse so album or track
titles. Those systems are still struggling dealing with mixed languages
("spiel mir Them Changes von Thundercat"). This is particularly true
when the voice recognition system doesn't know the context of the
utterance. I'll see.
--
Michael
_______________________________________________
plugins mailing list
[email protected]
http://lists.slimdevices.com/mailman/listinfo/plugins
