Raspberry Pi 3 Personal Assistant

Basic voice activated app created for Raspberry Pi 3 with NodeJS + Electron. Uses Annyang and Responsive Voice JavaScript libraries.

See it in action

Check this forum post with video.

Setup

1. Download Raspbian Jessie

2. Prepare the SD Card

3. Boot!

4. Use a monitor & keyboard & mouse to connect to the wifi

5. Expand SD Card - Otherwise will be only 3.4 GB of space

  sudo raspi-config
  sudo reboot
  df -h

6. Update & Upgrade

  sudo apt-get update
  sudo apt-get dist-upgrade

7. Install NodeJS

  wget https://nodejs.org/dist/v4.3.2/node-v4.3.2-linux-armv6l.tar.gz 
  tar -xvf node-v4.3.2-linux-armv6l.tar.gz 
  cd node-v4.3.2-linux-armv6l
  sudo cp -R * /usr/local/

8. Install audio settings

  sudo aptitude install pulseaudio
  sudo nano /etc/modules   
Add line: 
  snd_bcm2835
CTRL + O (save)  CTRL + X (exit)
  sudo apt-get install mplayer
  sudo amixer -c 0 cset numid=3 1

9. Copy the ‘personal-assistant’ code (easier with a flash-drive).

10. Install packages (basically just electron package, aprox. 60MB)

npm install

11. Setup the personal-assistant as autostart

Create a new file
  sudo nano /home/pi/personal-assistant.sh
  
Add the following to this file (the folder should be the same as step 9)
  !#/bin/bash
  cd /home/pi/Documents/personal-assistant && npm start
CTRL + O (save)  CTRL + X (exit)

Set system as owner
  sudo chown pi:pi /home/pi /personal-assistant.sh
  sudo chmod +x /home/pi/personal-assistant.sh

add this .sh (script) to startup
  sudo nano /home/pi/.config/lxsession/LXDE-pi/autostart.sh

add the following line:
  /home/pi/personal-assistant.sh &