Read the statement by Michael Teeuw here.
Voice/motion control
-
I think using https://snowboy.kitt.ai is the best approach right now for detecting a hot word and then continuing with the google one (annyang). The same way the guy at the http://docs.smart-mirror.io has done it.
-
@alexyak yea, i have seen the smart-mirror code. But i don’t pref 2 voice modules… the thing is if your playing around with the map function (zoom in, zoom in, zoom in, go to, zoom out, enz) then you run out of api calls fast.
I give up on the voice control module… i hope there will be one soon ^^ -
Actually I’ve started looking at this: https://syl22-00.github.io/pocketsphinx.js/ Need to test the performance on RaspPi
-
I have to second @alexyak and @tyho. Relying on chrome keys is not a good solution, especially when the goal is to interact with mirror through voice completely. Pocketsphinx is a good solution and I am getting it to work on RPi as well.
-
Yea i have finaly installed node-pocketsphinx. It needs a lot of dependeties. I start find out now how get a voice command out of it. Are u using node-pocketsphinx as well? Or are u using it standalone? Or maybe the pocketsphinx.js?
I like to hear if you get something working. -
Ok, I find out that https://syl22-00.github.io/pocketsphinx.js/ is the best bet.
I have run pocketsphinx before and it’s really fast. So i have no doubt the js port is also good.
But with my lack of skill i don’t know how to create a module. Who can?
-
I have tried standalone, but I am currently trying the node version. I’ll let you know once something gets going! Thanks.
-
@tyho, I have not, but then I’m not the one working on voice control. :)
-
@ronny3050 i guess you can better try pocketsphinx.js
-
OK, I’ve tested the pocketsphinx.js. It’s very unreliable with the minimal setting - too many false positives. The native pocketsphinx installation and configuration is too convoluted with a lot of dependencies.
Actually I was able to achieve the best and consistent results with https://snowboy.kitt.ai. There’s a way to create multiple keyword models on their web site and include it with the app. I’ve got this working pretty well in the MM. I should be able to publish the voice control module based on snowboy pretty soon.
