I created a speech recognition system that combines face detection as well, and integrates with VERA.
I thought I would share my progress on this forum and see if anyone in the community is working on something similar or would like to share their comments.
It appears to be very cool. I’m curious about how the speech recognition is handled and if its apparent intelligence is preprogrammed(canned) responses or if it is free form.
Hi Z-Waver, In its current form, it only replies to a preset grammar set. But you can load more than 200 phrases into it and speech recognition accuracy will still be as you expect it. I put in all possible ways you can make a command and also put it a lot of fun phrases like “tell me a joke” or “what do you think of Siri” for when I show it to my friends.
I am actually working on another project (home robot) where I have implemented an natural language engine (based in AIML) that can conversate with you in free form. Much like Siri or cortana but much more human like and would never of course read a bunch of Google results rather than answering your question as a human would. My only issue is that speech transcription is not very good and almost non-existent for open dictation. I wish I could implement my natural language engine into a project like jarvis, but speech transcription is the problem. It will take a few years until a reliable engine is available to the public.
hehehe, good question. Even though I run a home automation company (installer), this specific work so far has been just a hobby project, I do not plan on making it public or commercializing.
thats really cool how you built that. It is really fast and on point for it to do so manythings as well I think thats great.I feel like everyday we are moving forward with technology. Congrats
Best Home Automation shopping experience. Shop at Ezlo!