juliebee1024/smart-mirror-alexa

An SDK for commercial device makers to integrate Alexa directly into connected products.

Smart Mirror

Introduction

Alexa is ran in the background of the Smart Mirror.

• alexa-client-sdk v1.10

Getting Started

Create Smart Mirror Profile on Amazon Developer

1. Follow https://developer.amazon.com/docs/alexa-voice-service/register-a-product.html to create your product profile on Amazon Developer

Required Hardware

• Raspberry Pi 3B+ Kit
• USB Microphone
• 3.5mm Audio Cable
• Monitor
• Keyboard
• Mouse

Referenced from: https://developer.amazon.com/docs/alexa-voice-service/required-hardware.html

Required Software

• Raspbian OS
• AVS Device SDK

1. Follow https://developer.amazon.com/docs/alexa-voice-service/set-up-raspberry-pi.html to setup Raspbian OS and the Raspberry Pi
2. With Raspbian OS loaded, run sudo apt-get upgrade in terminal to get the necessary updates in the Raspberry Pi
3. Get necessary files:

wget https://raw.githubusercontent.com/juliebee1024/smart-mirror-alexa/master/pi.sh \
wget https://raw.githubusercontent.com/juliebee1024/smart-mirror-alexa/master/setup.sh

4. Create a config file:

sudo nano config.txt

5. Copy and paste into the config file:

#NOTE: The Device Serial Number can be any unique number
DEVICE_SERIAL_NUMBER=""
CLIENT_ID=""
PRODUCT_ID=""

6. Open a web browser and login to Amazon Developer (https://developer.amazon.com/avs/home.html#/avs/home) and click on your product
7. Scroll down and copy & paste the Product ID into config.txt
8. Click on "Security Profile" under the "Product Details" tab
9. Scroll down and click on "Other devices and platforms"
10. Copy & paste the Client ID into config.txt
11. Enter any number (i.e. 123456) as the Device Serial Number
12. Ctrl-o and enter to save
13. Ctrl-x to exit
14. Follow https://developer.amazon.com/docs/alexa-voice-service/input-avs-credentials.html starting at "Download your credentials" to get the config.json file
15. In terminal, run the install script

cd /home/pi
sudo bash setup.sh config.txt

16. Follow https://developer.amazon.com/docs/alexa-voice-service/build-the-avs-device-sdk.html for the rest of the installation process

Referenced from https://github.com/carolinedunn/Alexa-RPi-AutoStart

To Run Alexa

If you do not have the Refresh token yet:

1. Follow https://developer.amazon.com/docs/alexa-voice-service/get-a-refresh-token.html to get your refresh token and start the Alexa application

If you already have the Refresh token:

1. Run the Alexa application by:

sudo bash /home/pi/startsample.sh

Reference https://developer.amazon.com/docs/alexa-voice-service/talk-with-alexa.html should you have any issues communicating with Alexa.

To Add Wake Sound

1. For wake sound, follow https://developer.amazon.com/docs/alexa-voice-service/indicate-device-state-with-sounds.html

To Add Notification of Boot Up

1. Stop Alexa if it is running with Ctrl-c and download sound file from terminal:

wget https://raw.githubusercontent.com/juliebee1024/smart-mirror-alexa/master/hello.wav

2. Open File Manager and move hello.wav from /home/pi/Downloads folder to /home/pi/sounds folder
3. If you do not have permission to move files into the sounds folder, enter in terminal sudo chown -R pi:pi /home/pi and try step 2 again
4. Navigate to cd /home/pi/avs-device-sdk/SampleApp/src/ and nano UIManager.cpp
5. Double check to make sure #include <cstdlib> is included in the beginning of UIManager.cpp -- if not, add it in
6. Scroll down to void UIManager::printState() function
7. Find

} else if (m_connectionStatus == avsCommon::sdkInterfaces::ConnectionStatusObserverInterface::Status::PENDING) {
ConsolePrinter::prettyPrint("Connecting...");

and add in system("play /home/pi/sounds/hello.wav"); such that it looks like this:

} else if (m_connectionStatus == avsCommon::sdkInterfaces::ConnectionStatusObserverInterface::Status::PENDING) {
ConsolePrinter::prettyPrint("Connecting...");
system("play /home/pi/sounds/hello.wav"); //sound file for notification of bootup

8. Save the file (Ctrl-o) and close (Ctrl-x)
9. Rebuild the Sample App for changes to take effect:

cd /home/pi/build/SampleApp
sudo make

10. Once the Sample App rebuilds, restart Alexa: sudo bash /home/pi/startsample.sh

Referenced from https://github.com/carolinedunn/Alexa-RPi-AutoStart

To Autoboot Alexa

1. Stop Alexa if it is running with Ctrl-c and make a copy of the original .bashrc just in case: cp /home/pi.bashrc /home/pi/.bashrc-original
2. Edit the .bashrc file with sudo nano /home/pi/.bashrc
3. Scroll all the way to the bottom of the file and add in sudo bash /home/pi/startsample.sh on a new line after the final fi
4. Save (Ctrl-o) and exit (Ctrl-x)
5. Enter sudo raspi-config in terminal -> select 3 Boot Options -> select B1 Desktop/ CLI -> select B2 Console Autologin
6. <Finish> and select <yes> to reboot the Raspberry Pi

Referenced from https://github.com/carolinedunn/Alexa-RPi-AutoStart

Customizing Alexa

1. To set your location for weather and traffic data, follow https://developer.amazon.com/docs/alexa-voice-service/set-device-location.html
2. To adjust sound sensitivity of the mic, type alsamixer in terminal
3. To force sound through 3.5mm audio jack or HDMI, type sudo raspi-config in terminal -> select 7 Advanced Options -> select A4 Audio -> choose either 1 or 2 depending on what you want
4. To train Alexa to your voice for a better sensitivity say "Learn my voice" and follow Alexa's instructions to create your voice profile (more information found: https://www.amazon.com/gp/help/customer/display.html/ref=hp_bc_nav?ie=UTF8&nodeId=202199440)