NAOMI-HEAD

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

20161002_155225

NEW Jasper Documentation!

Hello All,

First, I want to thank everyone that has been patient with me so far in getting these things done one at a time…

I know it is not easy to get some of these things completed without good documentation.  So, with the help of a few community members, I have started what is a very rough draft of our documentation.  It is neither complete, or ready….  However, we are adding information as we go, so keep a close eye.  Even feel free to clone it, and contribute.  There will be documentation on how to do both of those coming as well.  Feel free to watch here, the repository, or the Support Forums/Slack for updates.  I update them all.

If you have items you would like to see added to the documentation, I would run it by myself, or d34dman first.  If we have not already written it (there are some not committed yet); then we will just let you know to put in a PR (aka Pull Request).  Once we have both reviewed it and give it a thumbs up.  We will merge it.

Alright, all that said… Here is the link to the documentation!

 

Thanks,

Matthew Curry

Jasper Image v1.5 Released

Major Updates

NOTE:  Be sure to do a “git pull” in the “Jasper-RPI-Tools” directory to get the updated scripts.

  • Installed Home-Assistant.io
    • Script to enable Home-Assistant.io Daemon
      • It is located /home/pi/Jasper-RPI-Tools/enable-hass-service.sh
    • Config under /home/hass/.homeassistant
    • Installed in Virtual Environment
    • Installed under the “hass” user.
  • Removed clutter from rc.local
    • Was causing harmless error at boot.
  • Installed needed dependencies for the development branch
    • libmad0 * libmad0-dev
  • Created/Enabled Service for jasper “jasper-daemon”
  • Created/Disable Service for Home-Assistant.io (user needs to enable)
  • Created Script to download a specified version of Jasper/Jasper-dev
    • It is located /home/pi/Jasper-RPI-Tools/installers/jasper-repo-installer.sh
      • You can pick what branch you would like to use at clone-time
  • Ensure phonetisaurus pre-compiled/installed
  • Removed Erroneous cron entries
  • Updated the OS and all related pkgs
naomi-v1

Pre-Built Jasper Kits, Ready to go out of the box!

These kits will arrive and be ready to use the moment you power them up. They will be customized to the user, and an image will be provided so they can restore any time they like.

They will include the following:

  • RPI3 B+
  • 32GB Class 10 SD Card
  • US Robotics 9610 [USB Speaker/Mic]
  • Customized/Installed/Setup of Jasper by me personally
    • Install of Jasper and all dependencies
    • STT Setup
    • TTS Setup
  • Testing
    • Running overnight
    • With your settings if you wish
  • Copy of your personalized image for backup
    • In case you wish to restore to a “Known Good” state
  • Updates
    • Such as customized integrations and modules provided by me.
    • Updated Installer / Setup script COMING SOON.


If you are interested in one of these kits, feel free to contact me or start the process by purchasing one below:




New Community TTS Server! – Thanks Mycroft.ai!

It has been a little while; but the guys and gals at Mycroft.ai helped us (The Jasper Project) with a donated TTS (Text to Speech) server.  This has helped many of our users to get their projects working virtually out of the box.  Keep in mind Mycroft.ai does almost the same thing. We would almost be a competitor.  However, I do not see it that way.  We are open source and here to help each other out.

 

Thank you again to the guys at Mycroft.ai!

To those not familiar with them; here is a link to their Indigogo campaign.

Jasper Images Updated version 1.4 (aka Revision 4)

They are updating now; this includes a few fixes and features.

 

  • Jasper Tools Cron Installed (keeps them updated)
  • TTS Server Setup Out of the box!
  • STT Setup Out of the Box! (Just add key in profile.yml)
  • Jasper Start at Boot Cron Installed (allows for Jasper to start at boot)
  • Tested with several different variations of USB devices
    • USB Mic /Speaker Combinations
  • A ‘paired’ installer will be released soon!
    • This will allow the user to answer a few simple questions and have it running in seconds.

I apologize for the delays recently, I have had some personal issues come up.

jasper-icon

Jasper Client Links

These are several links I have compiled, and/or created in the last few months. They can help you get Jasper going. I am also working on a full how-to for both RPI2 and RPI3.

  • Jasper-Support-Forums
    • Support forums, I help run these if you have any questions. Contact Me.
  • SLACK SIGN UP
    • Chat room for helping with Jasper
  • Jasper-Client
    • Official Jasper-Client
  • Jasper-RPI-Tools:
    • Set of tools I created in helping setup Jasper/RPI. They are a work in progress, so be sure to do a ‘git pull’ often.
  • RPI-Jasper-Dependencies:
    • Set of pre-compiled dependencies, that would normally take hours to compile. Tested on RPI2/3. Usually used for a non-Jasper image; as the images I created have them all installed.
  • ALSA Setup Wiki
    • Help with ‘.asoundrc’, and setting your default audio device
    • Usually the most common road block.
  • Home-Assistant.io: Home automation for the RPI, works very well. Working on integration on the same RPI; however it has dependency conflicts. Testing in pyenv soon…
  • How TO: Burning SD Card

NAOMI gets an Upgrade… A head…

NAOMI-HEAD

 

Hardware Shown:

  • High Resolution Raspberry PI Camera
    • Eventually I would like to try stereoscopic vision
  • PIR Motion Sensor
    • Used to ‘Wake’ NAOMI to motion in the room.
  • UltraSonic Range Sensor
    • She will track any motion
    • Once an Object/Person/Face is being tracked the Range Sensor will take measurement readings.
  • PWM HAT
    • Used to remove ‘Jitter’, and allow for other Servos
    • May try an “Arm”
  • Server Mount Allowing Pan/Tilt