Sean, thanks for the help and patience, finally managed to work out the issue.

it was mainly with improper setting on .asoundrc and config.js

This is hands down the best voice recognition with visual there is.

One small thing I noticed is the youtube video opened sometimes throw “invalid parameter”. Could it be default youtube parameter on config.js?

Thanks @cruunnerr.

In my case, no LED involved hence should be simpler

I run PIR.PY externally (via PM2) to call monitor on/off shell script and everything seems to work fine i.e.: turn screen blank and back on does work. No error reported. Its just the adding your script does not seems to make any difference, the turn on still instant.

Here is my PIR.py file

#!/usr/bin/env python

import sys
import time
import RPi.GPIO as io
import subprocess
 
io.setmode(io.BCM)
SHUTOFF_DELAY = 15  # seconds delay to turn off
PIR_PIN = 17        # Pin 11 on the board
TURNON_DELAY = 10   # seconds delay to turn on
 
def main():
    io.setup(PIR_PIN, io.IN)
    turned_off = False
    last_motion_time = time.time()
 
    while True:
        if io.input(PIR_PIN):
            last_motion_time = time.time()
            sys.stdout.flush()
            if turned_off and time.time() > TURNON_DELAY:
                turned_off = False
                turn_on()
        else:
            if not turned_off and time.time() > (last_motion_time + SHUTOFF_DELAY):
                turned_off = True
                turn_off()
        time.sleep(.1)
 
def turn_on():
    subprocess.call("sh monitor_on.sh", shell=True)
 
def turn_off():
    subprocess.call("sh monitor_off.sh", shell=True)
 
if __name__ == '__main__':
    try:
        main()
    except KeyboardInterrupt:
        io.cleanup()

Here is the monitor_off.sh

#! /bin/bash
export DISPLAY=:0
xset dpms force off

And Monitor_on.sh

export DISPLAY=:0
xset dpms force on

@cruunnerr

I actually use option 2.2 - external script and already set both sensitivity and time delay to minimal (all the way to left).

would you mind to send me your pir.py?

Regards,

@cruunnerr

I tried your script but unfortunately it doesn’t seems to work yet. Like @blebbens , my MM is on the floor and activated a little bit too easy.

Any other suggestion?

Regards,

Sean, thanks for the help and patience, finally managed to work out the issue.

it was mainly with improper setting on .asoundrc and config.js

This is hands down the best voice recognition with visual there is.

One small thing I noticed is the youtube video opened sometimes throw “invalid parameter”. Could it be default youtube parameter on config.js?

Apparently my microphone was not setup correctly. without MMM-Hotword and MMM-NotificationTrigger, it works but i have manually click the mic icon on every interaction with GA.

All i need is to use the hotword module to trigger the GA. The sample config file is too complicated for me and i don’t use IFTTT, GAction, etc.

Please help.

@sean

Hi,

I followed installation step by step until configuration part. credentials works fine when test.js ran.

Once reboot, i initially had blank screen but once electron build completed, i have mic icon on MM but unable to activate google assistant. I am using default config.js with single default profile.

I am using HDMI as output from TV which confirmed working after forcing HDMI output via raspi-config.

As input, I use USB Mini Microphone which confirmed after command “arecord -l” ran.

Do i need to install MMM-Hotword & MMM-NotificationTrigger too?

@magno99900

Make sure you run chmod - chmod +x pir.py

Then add it to startup:

1. run - sudo nano /etc/rc.local
2. Add this (above the “exit 0”): python /home/pi/pir.py &
3. Save

Reboot to test

Hello,

This is my 1st mirror build and appreciate any feedback. The goal is to create motion activated vertical smart mirror integrated with Google assistance.

The setup is RPI 3B+, Acrylic see through mirror on 32" display (with built in speaker so RPI3 can drive it via HDMI), USB Microphone (like this), and PIR motion sensor.

1.Will USB mic as input works fine with TV Speaker output via HDMI out of the box?

2. What aspect in particular should i consider to pic ideal display i.e.: LED/LCD, black level, port location, easy to detach housing, HDMI CEC, etc.

3. Voice Assistant (Google vs Alexa). Which one easier to install, use or more practical. Which MM module should I use?

Much appreciate any input in advance.

@cruunnerr Hello, My PIR works completely fine (with method #6, just turn black) but I also plan to build my mirror as floor mirror and wondering the same as @blebbens

I tried your suggestion but unfortunately doesn’t seems to work. the screen is on instantly if motion detected. Any other suggestion?