VoiceControl module

gismo2006

Maybe you can help me again?

If I put only one comand like Play everything ist fine. But if I put more than one comand, like Play and Stop the Player is completly confused. Then only a handclap or a any other word is need to start a track or to beginn from beginning again…

greets gismo

simonsays

This post is deleted!

simonsays

@alexyak @tyho I was able to get my sleep and wake up function to work using tyho’s code as well as able to find a way to turn the LED on. However, I cannot seem to turn it off once on. I am driving the switch with a gpio pin and transistor. It seems that after the pi executes the action, it wont listen for anything else. Example is if I tell it to turn the LED on, it does so. Then i say go to sleep, and it won’t. I go for the reboot and try again and start with the sleep command first. Then I try to turn the LED on and it won’t do that. Is there a way to have the function in tyho’s node_helper.js loop back again to the beginning? I feel like it get’s stuck once it executes a function. I also have an issue with the screen being yellow and zoomed in when i wake up the mirror after sleep. Don’t think it is a hardware issue, but I haven’t swapped HDMI cables yet.

For reference:

Code for modified node_helper.js:

socketNotificationReceived: function(notification) {
if (notification === “CONNECT”) {
clearTimeout(timer);
this.autosleep();
}
if (notification === ‘go_to_sleep’) {
exec(“/opt/vc/bin/tvservice -o”, null);
clearTimeout(timer);
=}
if (notification === ‘wake_up’) {
exec(“/opt/vc/bin/tvservice -p”, null);
exec(“/bin/fbset -depth 8 && /bin/fbset -depth 16 “, null);
exec(”/usr/bin/xrefresh”,null);
clearTimeout(timer);
this.autosleep();

}
if (notification === ‘lights_on’) {
exec(“echo ‘18’ > /sys/class/gpio/export”, null);
exec(“echo ‘out’ > /sys/class/gpio/gpio18/direction”, null);
gpio.setup(‘sys’);
gpio.digitalWrite(18,1);
clearTimeout(timer);
}
if (notification === ‘lights_off’) {
exec(“echo ‘18’ > /sys/class/gpio/export”, null);
exec(“echo ‘out’ > /sys/class/gpio/gpio18/direction”, null);
gpio.setup(‘sys’);
gpio.digitalWrite(18,0);
clearTimeout(timer);
}
},
autosleep: function(){
var self = this;
timer=setTimeout(function(){
exec(“/opt/vc/bin/tvservice -o”, null);
self.sendSocketNotification(“autosleep”, {});
}, 10601000);
}
});

Any help is greatly appreciated.

gismo2006

Hi,

maybe someone can help me… :-)

So far the musiccontroll works fine. Now I spent my time for a new project. I Copy the code from @tyho and save it as notification.js and node_helper.js. The I copy the code into my config file like he posted. Th two files I coppied into an new folder called notification, that I took under Modules. Then I train two Hotword for wake up and go sleeping an copy and paste the two pmdl files correctly. The Mirror stats fine and everything is working But the two Hotwords didn’t take their job.

Greets gismo

tgl

Hello :-)

At first - great job!

And the second - my English is not the best :-/

But I hope of Help.

I´ve allready installt the module.
Then I become the Error of Alsa.

Can anybody help me please?

Here my log:

[0_1476519121142_mm-error-0.log](Uploading 100%)

Thank you for help

MichMich

It would be nice if you can extend the module with hide/show options:

 {
    keyword: "simpleMode",   // keyword 
    description: "Say 'Simple Mode' to switch to the simple UI.",
    file: "simpleMode.pmdl", // trained model file name
    message: false,   // Unused in this command. Can be omitted.
    show: "simple foo bar",  // Show these modules.
    hide: "advanced" // Hide these modules.
}

When activating this command, it shows all modules with de classes defined in the show variable. And it hides all modules defined in the hide variable.

When implementing this, make sure you use the Visibility Locking mechanism that will be introduced in version 2.1.0: https://forum.magicmirror.builders/topic/241/2-1-0-api-revising-the-show-hide-mechanism/9

tgl

Hello again,

is it possible to post my log here?

gismo2006

Good evenning, morning,

I have now spent many hours to create @tyho notification module. And now it works.

Maybe the following can help someone, to…

If I say wake up the monito wake up but the the monitor picture was terrible and i had to reboot.

Now I have edit one line in the node_helper.js and everthing works fine. Here my changes:

if (notification === ‘wake_up’) {
exec(“/opt/vc/bin/tvservice -p”, null);
exec(“sudo /bin/chvt 6 && sudo /bin/chvt 7”, null);
clearTimeout(timer);
this.autosleep();

greets gismo

carteblanche

@alexyak great work on creating this module! This is the only voice module that works for me on MM2.

I was wondering if its possible to customize the sensitivity settings? On the snowboy website when training voice models it gives you this option when testing so wasn’t sure if that feature could also be added into the module itself. Right now with the default settings I need to speak very close to the mic for it to trigger the wake word.

A Former User

Think i got mixed up with Amazon Echo, can someone tell me or is there a showcase of how advanced the voice module actually is?

Thanks.

DevilRev

Hi,
I’m quite new to this and trying hard to get voice control to work but upon starting and after properly installing I just get the following really long error. I hope you can help me fix this because I really like the module and snowboy as a speech recognition tool instead of some online offers.
http://pastebin.com/13tY0kiP
Thanks in advance

DevilRev

@DevilRev
Well I “fixed” it by making the microphone (which is USB) devine nr 0 instead of 1 in ALSA, now the hotword detection works flawless BUT my speaker connected to the normal 3.5mm jack does not work anymore.
Pygame which I am using for audio output at another location of my Magic Mirror says there are no audio devices.

Is it possible for someone to make the audio used for the microphone with this module selectable? I really need that and can’t work around it.

GinSeng

Good morning, really cool module, thanks for this :)

@gismo2006
I want to do the same think as you - control the mm-music-player with this module.
I followed the help from @Jopyth but when i start the Mirror now it displays “mm-music-player” as a text instead of the module, the Voice commands are shown right so i guess i messed up something in the mm-music-player.js.
I copied the pdml files in /MagicMirror/modules/voicecontrol
Do i have have to changed the node_helper.js as well?

thanks in advance for your help and greetings

GinSeng
€: Okay it shows me my music player now … was a little ) missing :)
but still it don’t work my microphone works i’ve tested it with another voice control module. I guess the pdml files are now the problem. i copied them with : nano mv machmusik.pdml /root int the root folder

gismo2006

Hey, @GinSeng later today if I have some time I will post my code, and look if I can help you.

Greets gismo

GinSeng

@gismo2006

Hey Gismo, i would really appreciate that :) Thank you.
Maybe you can also help me with the pdml files. I’m not sure if i copied them into the right folder ( example: sudo mv machmusik.pdml / root).

Greets GinSeng

gismo2006

Hi, sorry but yesterday here was a big … (in German we say “chaos”) :-) So I haven’t any time. But today I will have free time to help you.

Anhand dem Namen deiner Pmdl File schließe ich du bist auch deutsch. Ich schreibe dich später an, und erklär dir mein vorgehen. Das ist für mich, auf deutsch doch einfacher als auf englisch. :-)

GinSeng

@gismo2006

ger: Kein Problem :) Ja bin auch deutsch. Okay wäre super und auch einfacher für mich :D

eng: No Problem :) Yep also german. That would be great and also easier for me

GinSeng

I want to say a big thank your @gismo2006 for helping me at the voice control via message. I want to share my solution and gismos help just in case someones also struggles with this.

It is important to record your voice with the raspberry pi and the connected microphone ( Was my first mistake i record them at my laptop with the internal microphone)
Copy the .pdml files in the home/pi/MagicMirror folder and just in case also in the voicecontroll module.
Copy the name of the pdml files and don’t write them manually in the config.js

Again a big thank you at @gismo2006 for the help and @alexyak for the module :).

Greetings GinSeng

gismo2006

Hi, sorry at the moment I haven’t got any free time. So I have read your post first now. Because you post your solution, I think it work. You don’t have to thank me, I am happy I can help. Like other helps me if I have a problem.

Maybe we should write the following, too.

You shouldn’t use some words that are to simple, like Play or Stop. At my experience this words will be to simple and your mirror will not regonize the correct order and will do every order at the same time. Because the system can not decide which order you have speak out.

Greets gismo

frodrd00

Hi, I have a problem with the module. The microphone and speaker work well. I do not understand why this happens. The error log:

ALSA lib pcm_dmix.c:1022:(snd_pcm_dmix_open) unable to open slave
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.rear
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.center_lfe
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.side
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.hdmi
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.hdmi
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.modem
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.modem
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.phoneline
ALSA lib pcm.c:2239:(snd_pcm_open_noupdate) Unknown PCM cards.pcm.phoneline
ALSA lib pulse.c:243:(pulse_connect) PulseAudio: Unable to connect: Connection refused
ALSA lib pulse.c:243:(pulse_connect) PulseAudio: Unable to connect: Connection refused
ALSA lib pcm_dmix.c:1022:(snd_pcm_dmix_open) unable to open slave
Cannot connect to server socket err = No such file or directory
Cannot connect to server request channel
jack server is not running or cannot be started
Expression ‘paInvalidSampleRate’ failed in ‘src/hostapi/alsa/pa_linux_alsa.c’, line: 2048
Expression ‘PaAlsaStreamComponent_InitialConfigure( &self->capture, inParams, self->primeBuffers, hwParamsCapture, &realSr )’ failed in ‘src/hostapi/alsa/pa_linux_alsa.c’, line: 2719
Expression ‘PaAlsaStream_Configure( stream, inputParameters, outputParameters, sampleRate, framesPerBuffer, &inputLatency, &outputLatency, &hostBufferSizeMode )’ failed in ‘src/hostapi/alsa/pa_linux_alsa.c’, line: 2843
Traceback (most recent call last):
File “./modules/voicecontrol/snowboy/kws-multiple.py”, line 34, in
detector = snowboydecoder.HotwordDetector(models, sensitivity=sensitivity)
File “/home/pi/MagicMirror/modules/voicecontrol/snowboy/snowboydecoder.py”, line 115, in init
stream_callback=audio_callback)
File “/usr/lib/python2.7/dist-packages/pyaudio.py”, line 747, in open
stream = Stream(self, *args, **kwargs)
File “/usr/lib/python2.7/dist-packages/pyaudio.py”, line 442, in init
self._stream = pa.open(**arguments)
IOError: [Errno Invalid sample rate] -9997
connected