Computer, Tea, Earl Grey, Hot!

Not related to the GF, but there are alot of geeks very happy right now


I just changed my wake word for our primary echo :laughing:


If only you could create your own wake word. I think Jarvis would be toward the top of the list for me.


I changed the main word too. The dot in my bedroom my wife vetoed it. At least I have one to play with.


I just saw this myself, I’m a little geaked out.


The future is now

Wow. I’ll brag a little (but it’s so geeky is it really bragging?)… I did this back in the 1990’s with my Commodore Amiga 1200 and condenser mics in the drop ceiling. I think I used a variant of the Rexx language called ARexx. It couldn’t do much (although I guess I could have made it do whatever I wanted). I had it so one could say variations of “Computer, what time is it?” or “Computer, time?” I also had family members’ birthdays in there. I may have had current weather as well. I forget now. It would play the TNG computer thinking beeping sound then speak the answer. Quite a lot of fun. I’ve thought about recreating it on PC. Most don’t know, but Windows has built-in speech recognition. Yep, built-in… no Dragon software or anything else needed.


Going to have to try this

too bad the Kindle Fire doesn’t have “voice wake” only a button wake for Alexa, so it doesn’t work with what I have… I will have to look into upgrading.

So do Macs. Plus desktops now have Siri. I never use it though as it’s a bit rubbish to be honest!


“too bad the Kindle Fire doesn’t have “voice wake” only a button wake for Alexa, so it doesn’t work with what I have… I will have to look into upgrading.”

You’ll just have to make a retro “enterprise” or original series version port!

-sorry, still haven’t figured out how to quote from mobile Firefox browser version.


Most voice recognition of the past have been rubbish, primarily because of the poor rate of success in understanding (without extensive training). While far from perfect, we are on the cusp of having speech recognition to the point where its good enough. The AI behind the recognition still needs a lot of work, but long are the days where you needed to train the system to understand you.


We started out with one Echo about 1.5 years ago. We now have four Echos, two Fire TVs (with Alexa) and various home automation. The voice recognition is shockingly good (we also have Siri, OK Google and Cortana on various devices, none of which are near as good) . Even the kids with their new orthodontic work have no issues communicating with it/her.


I haven’t had to train anything in a while. My Kinect, for example, understands every member of my family even from other rooms! And my phone, man, Google has about 98% accuracy with me. The only time it seems to screw up is when I query something bizarre and out of context.

Which I do on purpose all the time just to see how well it works, and it works amazingly well. My most-recent vague question to Google was “Who’s the guy who played Desmond?” Without any context at all it gave me the exact answer I was looking for. Which was “Desmond Hume was played by Henry Ian Cusick.” Now c’mon… That’s awesome. I asked a random question about a character’s first name from a show that went off the air 7 years ago and got the answer!


Just remember to turn off ordering and keep your windows closed…

I switched the kitchen dot so that I can pick which one I’m talking to. I hadn’t thought of this before but this is one way I can solve my home audio dilemma.

I don’t like that you can’t specify the playback device. “Alexa, play radio on living room dot.” I can just setup a whole home system with a dot set to a specific wake word. That should be accessible from about 90% of the house with a shout if needed.


That is a great solution! I hope that at some point they offer more control over which device you associate with which command. They do a wonderful job of deciding which you are closest to, but that is not always what you want.


Exactly. My living room and kitchen can be covered by a single device but recognition suffers in some cases. That’s why I went with two. It works great with the occasional “why did that one answer?” Not typically a problem, unless I’m trying to get a specific one to stop playing music. Now I can just add a third that responds to “computer” that is hooked up to a whole home audio system.

Not perfect but a whole lot better than the way it is now. That just means I have to build the whole home audio system. :stuck_out_tongue_winking_eye:

1 Like

I found that out about my Mac a long time ago…that I could turn on speech recognition and you had to press a certain key to use it. I tried it out and it was cool and tech and all that, but it ended up seeming so silly to me. I don’t use Siri on my Mac…and I seldom use it on my phone either, except that it’s what facilitates using my hands free bluetooth speaker in the car. I can certainly see it being useful for some…just not to me.


OMG, I made the change and then on a whim said "computer, start self-destruct sequence.

The response had me dying and the neighbors could hear me cackling down the street.

Try it, you won’t be sorry.