Marco Milani

Android & IoT Developer

Home Projects Blog About Contact

Alexa Skill Kit: Voice Apps for Everyone

Written on:

How Alexa Skills work, workflow, and Video Skill Kit explained.

What is a skill?

Ease of Access

  • Voice user interfaces (VUIs) are natural, conversational, and user-centric.
  • VUIs are rich and flexible.
  • The easier a skill is to use, the faster and more efficient the experience.

Speed and Efficiency

  • Voice commands can be faster than using a keyboard or touchscreen.

Skill Monetization

  • Skills can support existing businesses.

Alexa Skill Kit Workflow

Alexa Skill Kit Workflow Diagram
  1. Step 1: User says a command using natural language. The Alexa device records it and sends it to the cloud.
    Example: “Alexa, open android dev skills”
    Step 1
  2. Step 2: In the cloud, the voice command is processed through Automatic Speech Recognition to convert it into text. Then, Natural Language Understanding converts the text into JSON.
    Step 2
  3. Step 3: The JSON is sent to the appropriate AWS host, which contains the skill (Lambda Function) that manages the request.
    Step 3
  4. Step 4: The skill (Lambda Function) processes the request and produces a response in JSON format, which is sent back to the Alexa cloud.
    Step 3
  5. Step 5: In the cloud, the response can be converted into text and then into a voice reply, or sent to the Alexa device for a visual response.
    Step 5

Prebuilt and Custom Skills

Video Skill Kit

Video Skill Kit

Conclusion

The Video Skill Kit interfaces use a pre-built voice interaction model and are more robust than custom skills. Alexa is aware of devices and services through these interfaces, while custom skills must be invoked by name.

It is always better to use a pre-built skill when available and develop a custom skill only if needed.

Whenever possible, use a pre-built skill. Develop a custom skill only when necessary.