[Naran] was intrigued with the Amazon Echo’s ability to control home electronics, but decided to roll his own. By using a Raspberry Pi with the beta Prota OS, he managed to control some Phillips Hue bulbs and a homebrew smart outlet.
Prota has a speech application, which made the job simpler. He does point out though, that his project doesn’t replace the Echo’s ability to answer questions by searching the Internet. The advantage, though, is it is easily tailored to your specific application. Also, if you have a Raspberry Pi hanging around, you can’t beat the price.
We couldn’t help but wonder if this was the easiest way to go though. We always have a couple of old Android phones with no service around. Connected to WiFi, they do a great job of answering questions with Google Now and have excellent speech recognition. A simple voice app could do wonders here, even it wound up still sending electronic control signals to the Pi. On the other hand, it is hard to write an Android app that always listens and doesn’t require you to tap a button (or, at least, say “OK Google”) to make it start (or restart) listening.
If you wanted another way to make the Raspberry Pi listen, try Jasper. Or you could always just do it all with–what else–an Arduino and uSpeech.
ive always wanted to use voice commands on my devices but I am not comfortable with sending (potentially) every sound in my environment to google and the NSA for permanent archival
agreed; we know that if it can be abused, it will (in general). the fact that voice control means that a mic is always-on, that’s pretty much a show-stopper for me.
we need to have an admin switch, so to speak, that is trustable and reliable and does not impede use of voice input. if control is fully in the user’s hands, THEN I feel comfortable embracing such technologies, but not until then.
Same issue, always on listening but please, not always on NSA. I just want to catch phrases that are uttered when people are uncomfortable with the climate control “it’s hot” “it’s humid” “I’m freezing” so I can snapshot the temperature/humidity levels and see if it’s a real issue or imagined one.
I hear the folks behind Prota have left in the ability to “target” ads; which means it’s phoning home. I understand everyone “needs to eat”; but it since this is literally visceral raw data – who’s to say who it’s shared with or purchased by anywhere in the future? The general public does not understand as a whole how dangerous the whole approach is to them.
Neat but you could implement it in Rasbian as well though it’ll be more work but there you can be sure it will not spy on you.
I use the EasyVR board for hardware speech recognition. This only listens for the word “computer”, which then triggers the raspberry pi to record speech for 5 seconds after the trigger word. Was then sending that to google for translation, but since they screwed with their API access, I need to find a different way.
I have been wondering about getting some EPS8266s and adding an I2S audio chip to them so they can send audio back to a server for home control. They are cheap and you could add one in every room in the house.