Last week I attended the Digital Future Summit at European School for Management and Technology in Berlin. Two days fully packed with inspiring keynotes, hands-on workshops and participants from all over the world.
One of my personal highlights was the workshop Amazon Alexa Skills – Building Voice Experiences. We get an overview on Amazon Alexa and how everyone can participate in the “voice revolution”.
Use Cases for Alexa
But first things first: Amazon Alexa is a voice-based virtual assistant released in 2014, first used in the smart speaker Amazon Echo. By now, Amazon Alexa is supported by thousands of devices built by Amazon and other companies including smart speakers, phones and tablets, laptops, wearables and even cars (we’ve been waiting for this since K.I.T.T. from Knight Rider in the the 80ies…)
Voice assistants can interpret human speech and respond via synthesized voices. Next to Amazon Alexa, you may be familiar with Apple’s voice assistant Siri or Mircrosoft’s Cortana. Basic functionality includes asking the voice assistants questions or to play music, to call someone or to operate the smart home system. Right now, there are more than 80.000 Amazon Alexa Skills available.
Many companies started to build their own skills and enrich their customer interaction with Amazon Alexa. You can book an Uber by using Alexa or play Jeopardy!. Assassin’s Creed fans can enable their Amazon Alexa to interact in a spartan way with Amazon Alexios. It can greet in a true spartan attitude: “I am Alexios, the great Spartan warrior, here to bring honor to your suburban lives.”.
How Alexa works
All voice assistants work in similar ways. The software is constantly listening for a key word – such as “Alexa” – to wake it up. Once the key word was mentioned the systems starts recording the voice of the user. The voice input gets processed and interpreted as a command to search for information needed to be played back to the user.
Compared to early voice-based devices which had a small set of pre-defined commands, Alexa and co are constantly connected to the internet. With that, they can respond to a (theoretically) infinite set of commands and questions.
Moreover, recent developments in Natural Language Processing improve the abilities of voice assistants. This is basically because of increased computing power and tons of linguistic data that can be applied to powerful machine learning methods. With that, the interaction between computer and human gets more natural.
But why are we talking to our devices anyhow? Well, voice user interfaces have some important advantages from a Human Factors perspective.
First, it’s faster to talk than to type (especially for people who are not sending text messages 24/7 😊). Next, it’s hand-free which can be a major security upside, e.g. while driving.
Moreover, voice-based user interfaces are intuitive and with that inclusive. Even the least tech savvy person knows how to have a conversation. There’s no need for training which lessens adaption barriers.
Lastly, voice interfaces may be able to transmit more information than written words. Sarcasm is hard to detect in an e-mail. In a voice-based interface we have not the written word alone but also information of voice, rate of speech, volume and intonation. With those upsides, we are likely to see more voice enabled devices soon.
Alexa is my new BFF
Following the empathy argument, I want to present you a study by Cornell University with the title “Alexa is my new best friend”. The scientists analyzed 851 user reviews of the Echo posted to Amazon.com and looked for personification of the device and if there is a connection to user satisfaction.
As we learned in previous posts, people tend to anthropomorphize computers, machines and robots, e.g. by ascribing personalities or even race to them. The design choices by Amazon to give Alexa a name and a gender even amplifies this effect.
Based on the star ratings and written reviews on Amazon.com, they found out that users who tend to personify Amazon Echo, e.g. by naming it “Alexa” and using the pronoun “her” are more satisfied with the device.
“Simply put, people who love her, love the Echo.”– Quote from the study by Purington et al. (2017)
Still, it’s unclear “which came first” – if satisfaction leads to personification or personification to satisfaction. But this exploratory study is a first step in the direction of understanding the specifics of human-computer interaction via voice.
If you want to build your own Amazon Alexa Skill, check out the numerous learning documents provided by Amazon.
General summary on voice assistants: Hoy, M. B. (2018). Alexa, siri, cortana, and more: An introduction to voice assistants. Medical reference services quarterly, 37(1), 81-88.
Study at Cornell University: Purington, A., Taft, J. G., Sannon, S., Bazarova, N. N., & Taylor, S. H. (2017, May). Alexa is my new BFF: social roles, user satisfaction, and personification of the amazon echo. In Proceedings of the 2017 CHI Conference Extended Abstracts on Human Factors in Computing Systems (pp. 2853-2859). ACM.