MMM-AssistantMk2

teitlebot

@sean I’ll give it a try. How do I install something that isnt in the master branch?

A Former User

sudo apt-get install mpg321
cd ~/MagicMirror/modules/MMM-AssistantMk2
git pull
git checkout 1.1.0
npm install --save wav

A Former User

UPDATE

[1.1.0] - 2018.10.04

MP3 Output is supported. Now you can get more unchunky sound result.(Set audio.encodingOut to MP3 // OGG is not yet supported.)
ding.wav will be played when Assistant is ready to hear your voice.
How to update from prior version

sudo apt-get install mpg321
cd ~/MagicMirror/modules/MMM-AssistantMk2
git pull
npm install --save wav

teitlebot

@sean Thank You
Ding is working just fine. Thank You. A little annoying sound and I was thinking of replacing it with another sound file. What if I wanted google to say something instead of ding, can I give it text to say easily?

A Former User

@teitlebot
change other wav file and rename it as ding.wav,
Anyway, if you mean Text-To-Speech, nope. (My prior had that functionality but I abandoned it) I don’t want to back again.

cola0927

@teitlebot totally agree!

teitlebot

@sean ok. Can I ask why you abandoned it?

A Former User

@teitlebot
Complexity. And most TTS solutions ate not so good to hear. The best Google solution is not free.

teitlebot

@sean I see. I manually created a sound file and replaced it and it works fine, although a bit choppy.

teitlebot

@sean Next Question
Why doesnt google allow me to use the MM to play music. I can get it to play out of my google home mini in my kitchen using the MM but there doesnt seem a way to use the MM speaker to play. not even spotify.
Although youtube does work as you seem to have added that functionality.

A Former User

@teitlebot
Idk. Because this sdk might be just on beta stage?

vijayrraut

@sean Thank you very much. The changes for “ding” sound are working like a charm. Initially I got error while git pull as I had my own changes to my .js files. I did ‘git stash’ and merged the new changes. Thank you so much.

git stash
git pull
git stash pop
npm install --save wav

Stabber

Hi. I have a problem at this point:
In your SBC, you can run auth-tool for authentification. (not via SSH)
cd ~/MagicMirror/modules/MMM-AssistantMk2
node auth_and_test.js

and i see this:

That’s an error.

Error: disabled_client

A native application: project-1019057327317

You can email the developer of this application at: unconfiguredapp@google.com

This app is not yet configured to make OAuth requests. To do that, set up the app’s OAuth consent screen in the Google Cloud Console.

Probably SSH is the problem, but I don’t understand this: “(So you’d better to execute this script in your RPI shell, not via SSH)”

Could you help me with this?

A Former User

@stabber
I think your Google project(project-1019057327317) is not setup completely. There could be many reasons to cause 4XX errors. But I think you missed making consent screen. Go To Google Cloud Platform > YOUR PROJECT > APIs & Services > Credentials > OAuth Consent Screen Tab and put some information.
I think Application name and Support email are enough. You don’t need to verify, just save it. Then try again.

Cr4z33

Before I am going to try to use these modules can someone please tell me whether they can be used with the Google AIY Voice Kit (v1) or not? :grimacing_face:

A Former User

@cr4z33
If you can record and play your voice with that device, it could be.

citizendevpi

@cr4z33 I am in the process of setting up now on voice kit v1. did you start yet?

citizendevpi

ok so i have installed hotword (working i guess), assistantmk2 i went all the way tru and got a response after typing hello what time is it . and get api interactions in my google cloud. also installed notifaction trigger the mirror starts up and if i say jarvis,smartmirror i get hotword dectected but nothing from assistant im at a loss on how to get it going/response. i will upload my terminal output if i can figure out how if any one can help im pretty sure im missing somthing in my config.EDIT ok i input the test config and BAM it is alive the audio is super crunchy but it could be a hardware issue so any one interested yes i am running pi3b+ with aiy kit v 1 and it works.

A Former User

@citizendevpi @Cr4z33
Still got problem?
I’m making whole new update. It will be more stable and easy to install & use. I’ll release this new update in this week. Plz wait a few days.

citizendevpi

@sean no i did some tweeking with mp3 for audio.encoding out and that cleared up the sound issue i do get a frequent js error like every time i use the hotword:Error: open() failed: -1
at Speaker._open (/home/pi/MagicMirror/modules/MMM-AssistantMk2/node_modules/speaker/index.js:106:13)
at Speaker._write (/home/pi/MagicMirror/modules/MMM-AssistantMk2/node_modules/speaker/index.js:183:23)
at doWrite (/home/pi/MagicMirror/modules/MMM-AssistantMk2/node_modules/readable-stream/lib/_stream_writable.js:428:64)
at writeOrBuffer (/home/pi/MagicMirror/modules/MMM-AssistantMk2/node_modules/readable-stream/lib/_stream_writable.js:417:5)
at Speaker.Writable.write (/home/pi/MagicMirror/modules/MMM-AssistantMk2/node_modules/readable-stream/lib/_stream_writable.js:334:11)
at Reader. (/home/pi/MagicMirror/modules/MMM-AssistantMk2/node_helper.js:287:17)
at emitNone (events.js:111:20)
at Reader.emit (events.js:208:7)
at /home/pi/MagicMirror/modules/MMM-AssistantMk2/node_modules/wav/node_modules/readable-stream/lib/_stream_readable.js:934:16
at _combinedTickCallback (internal/process/next_tick.js:131:7)
Some error happens. Try again.