These steps need to be followed in order to setup a raspberry pi image with the figure speaker environment.
The project is only tested with a raspbian image. However, other raspberry pi images should work as well.
Follow the instructions at Raspberry Pi Setup Instructions in order to flash an image. Download the lite image from Raspberry Pi Downloads. There is no need for the Desktop version, because the figure-speaker won't have a display attached.
Connect the SD card of your raspberry pi to your computer and flash the image via etcher.
Create an empty file named ssh at root level of the sd card in order to access the raspberry pi later on.
In order to directly access your raspberry pi over via ssh, you can provide the wifi credentials on sd card level. Without doing this you need to plug in your raspberry pi to your network via ethernet.
Create a file called wpa_supplicant.conf at root level of the sd card. Edit the file and place the following content with your modifications regarding the credentials:
country=DE
ctrl_interface=DIR=/var/run/wpa_supplicant GROUP=netdev
update_config=1
network={
ssid="<SSID>"
psk="<PASSWORD>"
}
Put the sd card into your raspberry pi and start it. It should connect to the configured wifi automatically. Connect to the raspberry pi via ssh:
ssh pi@raspberrypi.local
Use the default password raspberry for authentication.
You should directly change the default password by entering the command passwd
From now on, you have two options: Either download and execute a setup script, which will execute every step necessary automatically, or do each step manually. If you decide for the automatic script, all you have to do is to execute the following command and then restart the raspberry pi:
curl https://raw.githubusercontent.com/Kkoile/figure-speaker/master/setup.sh -o ~/setup.sh | ~/setup.sh
If you want to execute each step manually, follow the following commands on this page.
The volume will be handled later by our application. In order to be able to get the maximum volume you should set it on the raspberry pi:
amixer set PCM -- 100%
Type sudo raspi-config and go to Interfacing Options. Select SPI and make sure to enable it.
Mopidy is the audio server which connects to spotify and other providers. You can find more about mopidy here. In order to install mopidy on the raspberry pi follow the instructions.
Depending on the mopidy extensions you want to include you should install them. For example the spotify extension:
sudo apt-get install mopidy-spotify
Modify the file ~/.config/mopidy/mopidy.conf and add the following lines to it:
[local]
media_dir = ~/.config/figure-speaker/files
Run mopidy local scan in order for mopidy to scan the files in there.
In order to test if mopidy is working, you can install a mopidy client like mpc:
sudo apt-get install mpc
This allows you to play mp3 files on your raspberry pi, or if already configured, extensions like spotify.
For the full documentation type man mpc. If you only want to test the audio output you can use the following code snippet:
curl https://www.sample-videos.com/audio/mp3/crowd-cheering.mp3 -o test.mp3
mpc add file:///home/pi/test.mp3
mpc play
Use mpc stop to stop playing the current audio file.
Node.js is the environment in which figure-speaker is written. Hence, to be able to execute the application you need to install it.
curl -sL https://deb.nodesource.com/setup_8.x | sudo -E bash -
sudo apt-get install -y nodejs
Change the directory of installing global npm modules, in order to reduce permission errors. Follow this link to do so.
Figure-Speaker is written as a global npm module. So we can install it by executing the following command:
sudo npm install -g figure-speaker
Don't mind the vast amount of error logs...
You can test the installation by typing figure-speaker in the command line. This should start the server. The frontend should now be available at http://figure-speaker:3000 or http://figure-speaker.local:3000 (MacOsX).
We want our application to always start when booting raspberry pi. To achieve this, we create a service file which we can then manage with systemctl.
Create a file figure-speaker.service:
sudo touch /etc/systemd/system/figure-speaker.service
sudo chmod 664 /etc/systemd/system/figure-speaker.service
And paste the following content in it. You should change the spotify credentials according to your spotify app. //TODO: link to another chapter where this is explained
[Service]
WorkingDirectory=/home/pi/
ExecStart=/home/pi/.npm-global/bin/figure-speaker
Restart=always
StandardOutput=syslog
StandardError=syslog
SyslogIdentifier=figure-speaker
User=pi
Group=pi
Environment=NODE_ENV=production SPOTIFY_CLIENT_ID=<SPOTIFY_CLIENT_ID> SPOTIFY_CLIENT_SECRET=<SPOTIFY_CLIENT_SECRET> GPIO_INCREASE_VOLUME_BUTTON=27 GPIO_DECREASE_VOLUME_BUTTON=22
[Install]
WantedBy=multi-user.target
Execute the following command:
sudo systemctl enable figure-speaker
Now, either start the service directly with sudo systemctl start figure-speaker, or reboot the raspberry pi.
You can always see the current logs by executing journalctl -u figure-speaker -f
sudo npm install -g figure-speaker-update-manager
Create a file figure-speaker-update-manager.service:
sudo touch /etc/systemd/system/figure-speaker-update-manager.service
sudo chmod 664 /etc/systemd/system/figure-speaker-update-manager.service
And paste the following content in it.
[Service]
WorkingDirectory=/home/pi/
ExecStart=/home/pi/.npm-global/bin/figure-speaker-update-manager
Restart=always
StandardOutput=syslog
StandardError=syslog
SyslogIdentifier=figure-speaker-update-manager
User=pi
Group=pi
Environment=NODE_ENV=production
[Install]
WantedBy=multi-user.target
Execute the following command:
sudo systemctl enable figure-speaker-update-manager
Now, either start the service directly with sudo systemctl start figure-speaker-update-manager, or reboot the raspberry pi.
You can always see the current logs by executing journalctl -u figure-speaker-update-manager -f
sudo npm install -g figure-speaker-wifi-connector
Create a file figure-speaker-wifi-connector.service:
sudo touch /etc/systemd/system/figure-speaker-wifi-connector.service
sudo chmod 664 /etc/systemd/system/figure-speaker-wifi-connector.service
And paste the following content in it.
[Service]
WorkingDirectory=/home/pi/
ExecStart=/home/pi/.npm-global/bin/figure-speaker-wifi-connector
Restart=always
StandardOutput=syslog
StandardError=syslog
SyslogIdentifier=figure-speaker-wifi-connector
User=root
Group=root
Environment=NODE_ENV=production
[Install]
WantedBy=multi-user.target
Execute the following command:
sudo systemctl enable figure-speaker-wifi-connector
Now, either start the service directly with sudo systemctl start figure-speaker-wifi-connector, or reboot the raspberry pi.
You can always see the current logs by executing journalctl -u figure-speaker-wifi-connector -f
//TODO document how to configure ad hoc network and expose figure-speaker-wifi-connector