Read the statement by Michael Teeuw here.
Hello-Lucy
-
Hey Mykle,
I already Wrote a while ago that I Plan to use snips As Voice recognition for MM2.
Well finally got Started with The project And The Speech recognition itself Works Fine. However I still Need To create a Module for it. Since I Really like The features of Hello Lucy My Plan Is To use The Main.js And just replace The nodehelper
However since I am a absolute noob both with JavaScript And Modules in MM2 I still have a few questions towards Hello Lucy- did I See that right: everything Voice related is Done in The nodehelper And When something is recognised it just Gets Send To The Main.js over a socketnotification?
- I did Not Understand How The nodehelper Gets Called By The Main.js
- where is The hotword recognition? Failed To find what exactly in The Code does this
Thank You for The help
Max -
Hello Max,
I did write the Hello-Lucy enhancement to MMM-voice (by Strawberry). However, it did not require me to fully understand the workings of MMM-voice. I only needed to know how to send and receive notifications that would trigger hide and show commands to MM itself. I’m not being evasive. I simply cannot explain as Strawberry himself would, being the author of MMM-voice. If you direct your question toward him I am sure you will get a timely response, if he doesn’t happen to look at this topic
-
Ok ty, I will contact him. Btw when you talked to your MagicMirror I recognised you always only say the keywords like for example “Show Page 1”. Is this just because thats the way you talk to the mirror or would it fail when you say for example “Show me page one please” or even “Show me the first Page”?
-
@maxbachmann said in Hello-Lucy:
Ok ty, I will contact him. Btw when you talked to your MagicMirror I recognised you always only say the keywords like for example “Show Page 1”. Is this just because thats the way you talk to the mirror or would it fail when you say for example “Show me page one please” or even “Show me the first Page”?
If you look at the coding you’ll notice that it listens for words or sentences that you have created. So, you could create sentences such as “Show me page one please”. It is my feeling that the longer the sentence, the more room for error. Thus, I kept mine short and simple. I’m quite satisfied with the success rate.
-
@Mykle1 thats what I thought :)
Well I hope I get a first test version up and running next week. Have to finally get to the building part -
I’ll keep an eye out. I’m interested to see what you have done
-
@Mykle1
Might be able to show a first video soon (Hopefully next week) -
Excellent! :-)
-
@Mykle1
Well I have the code finished I guess, but have to setup my raspberry pi again before trying ^^
Update failed in the middle and now the Operating System is absolutely broken. I really have a talent to destroy stuff
I hate when I am hyped to try my stuff out but can’t :disappointed_face: -
Well that sucks. Sorry dude. Don’t give up.
-
@mykle1
I don’t. Pretty normal that a lot of stuff goes wrong when a noob creates a big project lol. Nothing that could stop me -
Well for anyone interested I created a seperate thread
https://forum.magicmirror.builders/topic/7162/german-voice-recognition -
Hello Lucy!
(My granddaughter)
-
@Mykle1 what are the specs of the laptop mainboard your using?
-
ubuntu 16.04 LTS
AMD C-50 1 GHz
Dual-Core
Cache L2 - 1 MB
64-bit
DDR3 SDRAM 2 GB
Graphics Processor AMD Radeon HD 6250This one is a Compaq CQ57. Runs MM amazingly well
-
@Mykle1 still have a windows 10 development board with 4gb ram and a Intel Atom x5-Z8350. Hope I have the time to try MM with that this we :)
-
That’ll do. For me, I don’t like having to run Windows for a MagicMirror, especially if the resources of the board I am using are limited (Ex. 2GB RAM). Also, at some point, you may run into a problem installing dependencies under a Windows system (Ex MMM-voice). Running Ubuntu and MM eliminates both of these drawbacks.
-
@Mykle1 hm I will see drivers on there work better with windows. However if it sucks with windows I can still put Ubuntu on it ;)
Don’t need mmm-voice since I use a voice recognition on a different raspberry pie -
You will have better luck with the Windows Developer than Windows 10.
But, you may have problems with the Atom CPU.
Which is from my experiences…
I have not tried it on Ubuntu or any other OS besides Windows. I did try Widows Developer, But the computer I was using just wasn’t strong enough to run it.I have tried Windows 7 and windows 10 on several different types of laptops and desktops. And it really does make a difference making sure you always have the latest updates, and you will need a beefy CPU and more than 4gb RAM. I have an older SFF Lenovo desktop and it’s maxed out on the CPU and RAN it can have, and the more I work on it, the more I have to reboot to clear memory as it locks up. But, if I have MM running, and absolutely nothing else, it runs pretty smooth. Same experiences with a couple laptops, with So-So cpu’s and 4 gb max RAM. I also attempted to run MM on a Surface RT, it doesn’t allow any applications other than what it came with, so, as expected, wouldn’t work. I do know of a person running MM on a Surface Pro 15.6" Tablet PC, don’t know the specs, but he says he doesn’t have any problems with it. I also know of a person trying it on an Ubuntu Laptop, but he is having problems getting it to run at all. I would help him, but he knows more about Ubuntu and the Linux-Based OS’s than I do.
I also had MM on an HP All-In-One, with maxed out CPU and 16gb RAM, and an Asus laptop with the latest Core-i7 CPU and 16gb RAM, and they run very smooth, even if I’m running other applications, surfing the web, making changes to MM and restarting it several times, I had no problems with it. I also had no issues using it on a Lenovo 14" 360 laptop with a core-i3 CPU and 8gb RAM.
I had it going on a Nexxus 10.1" tablet running Windows 10, but it was extremely slow… I had nothing on this 32gb tablet with a 32gb micro SD except a stripped down version of Win 10 and the Mirror. and it was just too slow. if I remember right, it only had 2gb RAM.
And yes, Mykle is very correct about the voice issue. I’m am still researching on what may be available for running voice and facial recognition with Windows. but as of right now, I have tried several different types of modules and applications. have yet to find anything that works, YET!!!
When you get it installed on your Atom device, I would definitely like to know your outcome!
"Life's Too Short To Dance With Ugly People"
Jim Hallock - 1995 -
@Mykle1
Just tried my Monitor out. Thats a sick black. I can’t see any difference between black and Monitor off.
And there is no message when unplugging or plugging in the HDMI cable :grinning_face:Hope I get the walnutwood for my frame next week
Hello! It looks like you're interested in this conversation, but you don't have an account yet.
Getting fed up of having to scroll through the same posts each visit? When you register for an account, you'll always come back to exactly where you were before, and choose to be notified of new replies (either via email, or push notification). You'll also be able to save bookmarks and upvote posts to show your appreciation to other community members.
With your input, this post could be even better 💗
Register Login