History of Digital Assistants
Computers with audio interfaces are not new. They even entertained us for some time. Sci-Fi stories about a voice in the room, like Hal 9000, or about a voice in a suit, like Jarvis from Iron Man, seemed plausible and very useful. However, in practice, earlier implementation of such technologies seemed to say: “I'm sorry Dave, I'm afraid I can't do that.”
The history of robotic helpers takes us back to automatons and Al-Jazari, who designed a prototype that would play music or serve drinks. At the start of the 13th century this was an amazing undertaking. The book “Wonderland: How Play Made the Modern World”, by Steven Johnson, covers the modern era of automatons, as well as other major inventions. It demonstrates the role of playful thinking in innovation and technology.
In 1952 Bell Laboratories introduced Audrey as a speech-recognition system. Later it was followed by a chatbot Eliza from MIT, and in 1997 by Jeeves a natural language search engine. Siri was introduced in 2011. It was founded by a DARPA project for military artificial intelligence (AI). Since then, we see Amazon Alexa service with Echo devices, Microsoft Kortana, Google Assistant, and many others.
From CLI, through GUI, to AI
The command line interface (CLI) was a very productive way to type in commands in the day of Unix and it is still used by many technologists. However, the graphical user interface (GUI) took over the end-user world with Apple and Microsoft operating systems. The GUI was of course developed by Xerox. Just as GUI never completely replaced CLI the audio interfaces likely will not replace the command line or the GUI, but they fit a niche area, which likely promises a long and successful future
Audio interfaces are very useful in connection with tasks traditionally fulfilled by radio devices. Playing music, listening to news, and other basic audio services can be accomplished on a PC, but the quality of speakers and availability of a PC in various locations is unreliable. Therefore, a market opened up for specialized, low power device like the Amazon Echo. The Echo uses WIFI and can play music, deliver news, and perform many other functions.
The emergence of powerful devices with audio interfaces require AI, because there is no time during a conversation to explain what can be said and how. The user can ask Alexa to “turn off the light”, or “turn the light off”. The AI is able to figure out what you meant, even though you didn’t use exact, expected statements. The AI market is very competitive with IBM Watson and other major AI systems in play. Likely as digital assistant systems compete, the quality of AI will determine their success.
Want to learn how to develop an Amazon Alexa skill? Or would you like to see how you can check Blackboard Learn grades with Alexa? Visit the Innovation Center in the Exhibit Hall on Wed at 4pm.
More about Alexa: