Rejecto with dynamic grammar

Home Forums OpenEars plugins Rejecto with dynamic grammar

Viewing 5 posts - 1 through 5 (of 5 total)

  • Author
    Posts
  • #1030006
    Tapestes
    Participant

    We’re using OpenEars dynamic grammar for phrase recognition. The environment includes constant conversation, most of which is not directed at the app. OpenEars is working really well for utterances directed to the application, but we’re finding OpenEars routinely matches unrelated conversation to phrases in the grammar as well. To address this issue, we’re wondering if Rejecto would be appropriate for dynamic grammar applications. I understand it’s a tall order to expose the app to constant conversation and expect it to properly reject all the unrelated utterances. Just want to get a sense for what’s feasible.

    An ideal solution would be one that allows us to use Rejecto for a keyphrase at the start of each grammar, not unlike “Hey Siri” or “OK, Google”, but retains the existing aggressive attempts at matching for the rest of the phrase. I’m guessing that’s not possible, but I figured better to ask the expert and get an authoritative answer than to assume it’s impossible based on my limited knowledge.

    Thanks for your time and help.

    #1030010
    Halle Winkler
    Politepix

    Hi,

    I’m pretty sure I’ve heard of other projects taking this approach (starting listening with a probabilistic language model plus Rejecto and then switching to a grammar when a wake-up word has been invoked). I haven’t personally tried it yet but there shouldn’t be a technical or speed barrier to this working. It is also possible that RuleORama could handle both parts, maybe worth giving it a try as well.

    #1030355
    evilliam
    Participant

    I do this and it does work pretty well. I do find that calling a switch on language model isn’t without cost and that it’s not an entirely predictable amount of time – but roughly 0.7s or so which can be quite a tricky one for a grammar as the grammar can miss what the person is saying during the switch.

    I kinda fixed it through UI but if there’s a way of making sure it’s seamless, or processing speech that was uttered while the language model was switching then that would be awesome!

    #1030356
    Halle Winkler
    Politepix

    That duration sounds a little long, are there logs showing it?

    #1030472
    Halle Winkler
    Politepix

    The 2.502 version of OpenEars that came out yesterday (http://changelogs.politepix.com) has some optimizations to model generation time and will be followed up shortly by a version of RuleORama with further model generation speed optimizations (will be announced at the same site).

Viewing 5 posts - 1 through 5 (of 5 total)
  • You must be logged in to reply to this topic.