Voice chatbots | 4 MIN READ

Demystifying Alexa

With the increasing democratization of chatbots and the AI tech that powers them, more and more businesses are looking to automate operations and improve customer engagement using chatbots. 2017 was touted the year of the chatbot, and all evidence points to 2018 becoming the year of the voice bot.

This blog is dedicated to Alexa, Amazon’s sleek little cylinder which lights up when you call her by name.

Let us start with differentiating between Alexa & Echo. Echo is the hardware from Amazon, which is powered by the Voice Control Software – Alexa. Without Alexa, Echo is just another speaker.  So, Alexa is what makes Echo the Smart one.

Alexa has certain built-in capabilities like set timers , alarms, read headlines , answer general questions, track weather etc. The ability of Alexa to do these things is called its “Skill”.  

Alexa can also learn to do more things. Amazon has provided the “ASK – Alexa Skills Kit”, which enables developers to teach Alexa new skills. ASK is a collection of self-service APIs, tools, documentation and code samples. When you add a skill to Alexa, you are making that service available to millions of customers through Alexa enabled devices. There are thousands and thousands of Alexa Skills that are available in the market now. Not all are developed by Amazon.

What type of Skills can you add to Alexa ?

  • Smart Home Skills let users do Home Automation using Voice Commands. For example, Alexa on Hue lets you control the light intensity based on moods.  Alexa can also be integrated with Thermostats and lets you switch on the room heater remotely and keep your house warm by the time you reach home. You can even query a lock to see if it is locked!!
  • Video Skills allow video providers to control video content using Alexa. This enables companies to teach Alexa how to interface with their video content and services. It enables Channel Navigation, Playing Content, Search for a content & Playback controls.

Alexa representation

(Courtesy: Amazon Developer Blogs)

This means, users can search for a specific show without specifying a provider or device. Through this skill, Alexa is aware of the video devices and services the customer has subscribed to. For example, if I command Alexa to find play the 100th episode of GOT, then without lifting a finger, I will be able to start viewing the show in my TV.

  • Flash Briefing Skills – This is probably the finest and most widely used skill. Most of us start our day with a quick overview of news and other content like comedy, Interviews and lists. This is achieved by the Flash Briefing Skills. Alexa either reads out the original text content that is provided by the developer or plays the audio content that is provided by each enabled skill.Alexa skill set representation
  • Custom Skills are skills that can handle any type of request. It can be a look up for an information from a webservice OR any thing you can think of!!  

One can combine these skills to come up with most innovating solutions. Here is couple of examples from Amazon Blog ((Meet the Women Shaping the Future of Voice with Alexa).

  • Andrea Bianco – A Home automation consultant, she designs and implements personalized Smart Home Solutions which combines Smart Home Devices, Creates Custom Alexa Skills And provide a Single Interface customized for that customer, like a mobile App to control them. Imagine having a mobile app that lets you query for the status of various home devices -lights, coffee pots, humidifiers through Alexa.
  • Amy Stapleton has turned her love for story telling into business. She has started a company called Telleables, which creates story-based voice experiences for Alexa. These skills are designed to keep children engaged while developing critical- thinking skills.

Alexa Voice Services

The other side of the story is that one can put Alexa on anything, with AVS – Alexa Voice Service, thereby providing a Voice Interface to the product. AVS is essentially a suite of resources to quickly and easily build Alexa enabled products.  It includes the following

  • Hardware – AVS requires a microphone , speaker and an internet connection.
  • AVS APIs  – Using these APIs developers can enable the Voice Service from the connected device using a Speaker and a Microphone. Once this is integrated, the product (it can be a speaker or a thermostat or any other device) will have access to  built in capabilities of Alexa (like music playback, timers and alarms, package tracking, movie listing, calendar management, and more) and third part Skills developed using ASK
  • Device SDKs – Provides C++ Libraries that enable your device to process Audio inputs , Wake word Detection , Alexa Communication Library, Alexa Directive Sequencer Library, Capability Agents and Activity Focus Manager Library.

For accelerated AVS integration, Amazon has recently launched Single-On Chip (SoC) Solutions, with built in hardware and the required SDKs .These are complete end to end systems for building  AVS products. They include the chipsets, Voice Processing technologies and client software that leverages the AVS APIs. These are qualified development kits from Qualcomm, Amlogic & AllWinner.

It is evident that Alexa is getting smarter and smarter with added skills and natural language understanding and accuracy. And with the launch of AVS, you can put Alexa virtually on anything. Opportunities with Innovative Voice Design has increased in manifolds which is going to make our lives easier and exciting.

Leave a Reply

Your email address will not be published. Required fields are marked *

SHARE linkedin twitter