Category Archives: offline

Thank you to all… Jasper v1.6 Image News!

First, I would like to say thank you to the community members for helping out. I have gotten a lot of input on the new image.  This has also given me better feel for what the community wanted in an image. Which leads me to the next bit of news.


Some of you are undoubtedly wondering why I have not released my Jasper v1.6 image as of yet.  Well, I can assure you there are very good reasons.

Firstly,  I have had some severe family/health issues come up as of late which take priority.  I am also in the middle of moving a family of 5.  Combine that with the other reasons below, and I feel that most will feel the wait to be worth it.

This leads me to the next bit of news, however shocking it may be.

  1.  There will be no version 1.6 image.
    1. There will be no image at all…
  2.  Instead, there will be a full installer.
    1. This is one of the reasons for the delay.
  3.   The installer will work with a full GUI, and on the desktop.
    1. Examples:
      1. Full ubuntu Install.
      2. Fresh RPI install with GUI.
  4. The installer will be iterated on by feature going forward.
    1. We can add community approved features one at a time.
      1. Example:
        1. Change Jaspers Name
        2. Change STT after setup
        3. Etc.

FAQ (COMING SOON)

  • So whats that mean for me?
  • What do I do if I am running Jasper v.15?
  • Where do I get the installer?
  • Do I need to upgrade?

 

How AI listens in Jasper (and others)!

 There are two basic kinds of listeners, an active and a passive.  You will not get good offline (passive listener) results for a wide variety of speech currently.  Most offline listeners are used as ‘passive’ listeners. They usually just listen for the wake word or just a few small commands.  They are very accurate but can’t sustain a large vocabulary without CPU issues.  Some are also closed source such as Kitt.ai (snowboy), and require that you train each word on their site.

The other kind, aka the “Active listener”; or the one always listening.  Would be used to listen once activated by the passive listener, and has a much more broad vocabulary.  However, these are online only. Due to the way they work (gathering metadata to increase accuracy), its just not possible currently on a small device. Examples would be standard APIs such as Google, Wit.ai, Ivona/Polly, etc.

There are projects to fix these offline issues, but they are all in the works. See my site for info on some of them. mattcurry.com