I could build almost all OpenCog components under my RPi3, with the
exception of MOSES cause it took too much RAM, but configuring gcc to
use less RAM should work. I used Ubuntu MATE as distro, I see no reason
it wouldn't work under Raspbian.
Nil
On 03/01/2017 12:39 AM, Dave Xanatos wrote:
In light of the significant enhancements to the Raspberry Pi in version 3 now, I'd like to re-visit
this thread. I am currently running a RPi 3, with Raspbian Jessie, Python 3.4, OpenCV 3.2.0 and
support code such as NumPy, SciPy, SKLearn, Swig, etc., doing face detection and recognition, and
responding in real-time to visually recognized instances via espeak, and controlling servos via SPI
through the GPIOs to control eye movement to maintain "eye contact" with the detected
face's eyes. In addition, I am running CMUSphinx/PocketSphinx for speech recognition (still needs
significant training). None of this is taxing my Pi significantly. I just discovered OpenCog
while searching for an AI framework to pull all my cognitives together and create a
"presence", with the goal of creating a reasonably functional conversational
interactivity to go along with the CV, hearing and Speech components. Interested in hearing from
folks who may have done, or be in process with this. Thanks.
--
You received this message because you are subscribed to the Google Groups
"opencog" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/opencog.
To view this discussion on the web visit
https://groups.google.com/d/msgid/opencog/4753b87d-09dd-6a07-69a7-2086d4ef2251%40gmail.com.
For more options, visit https://groups.google.com/d/optout.
