Using voice-controlled interfaces via Amazon Alexa

Control Program

Since the running Ngrok server is now blocking the current terminal, it makes sense to open a second terminal window. In it, create a text file named house.py with the contents from Listing 2. This Python program calls the Alexa skill later on. The program provides a server that controls three devices: a TV set, a lamp, and a socket outlet.

Listing 2

house.py

 

For each of the three devices, the program has what is known as an intent, which controls exactly this one function. The corresponding decorators @ask.intent() define the functions that the intent then executes as an action when called.

Say Hello!

The @ask.launch() decorator displays a friendly greeting text. In this example, the Australian "Gooday" is used, a greeting that would be appropriate at any time of day.

In the functions, the commands necessary for controlling the GPIOs are executed. The Flask framework abstracts the complexity behind a skill very well. Further data can be found in the templates.yaml resource file. The file must reside in the same directory as the Python program that communicates with the skill. This file contains texts that the skill will use. This makes it easier to adjust the speech output if necessary, for example to change the greeting or to enter other devices. The quite compact resource file for the example from this article is shown in Listing 3.

Listing 3

templates.yaml

§§number
01 greeting: Gooday
02 TV: TV is {{status}}
03 light: Light is {{status}}
04 socket: Socket is {{status}}

The shell python house.py command starts the Python server. Ngrok and the Python program simply keep running after that. Of course, you cannot close any of the terminal windows afterwards. You need to make sure that Ngrok is running when the Python server enters the scene, otherwise the connection between the Echo skill and the server program cannot be established.

Setting up an Alexa Skill

For the next step when programming an Alexa skill, the Echo device you use must be activated. You first need an Amazon Web Services (AWS) account, which can be obtained free of charge on the AWS homepage. All you need is a credit card and a mobile phone connection.

You can now create a new Alexa skill via the AWS account. The easiest way to get there is via the list of current Alexa skills, which is empty for a new user account (Figure 4). There the user clicks on the Create Skill button to create a new skill.

Figure 4: The (still empty) list of Alexa skills on an Amazon Web Services account.

The skill should then be given a descriptive name and a desired language. A mouse click on the Next button takes you directly to the next step of choosing the Custom model: it offers the highest degree of flexibility.

Another click on the button Create Skill closes the wizard. Now the homepage belonging to the new skill appears (Figure 5). Here you can set further parameters of the skill. The video under "How to get started" at the top of the page helps you to get started with the subtleties of skill creation.

Figure 5: The web page for the newly created skill.

Buy this article as PDF

Express-Checkout as PDF
Price $2.95
(incl. VAT)

Buy Linux Magazine

SINGLE ISSUES
 
SUBSCRIPTIONS
 
TABLET & SMARTPHONE APPS
Get it on Google Play

US / Canada

Get it on Google Play

UK / Australia

Related content

  • Programming Snapshot – Alexa

    Asking Alexa only for built-in functions like the weather report gets old quickly, and add-on skills from the skills store only go so far. With a few lines of code, Mike teaches this digital pet some new tricks.

  • Mycroft

    Voice-activated assistants like Mycroft bring online, hands-free help to users, but with more transparency and less spying.

comments powered by Disqus

Direct Download

Read full article as PDF:

Price $2.95

News