Voice as a User Interface?
In the world of information management most are concerned with one important thing... how to store information so users can find it!! Most companies are getting good at this, not great, but good... and they are getting better each year. Unfortunately, they are dealing with even more information than every before:
Microsoft, Google, Amazon and Apple have all demonstrated that people can interface with their services using only their Voice. Of course, us humans are only coming around to fully adopting this as part of our daily lives.... but that is because there are still some barriers. Eventually, those barriers will be removed by some innovation so why not plan for the future and start looking at a Voice UI for Employees to help them get the information they need to do their job better.
Creating a Voice UI for your Enterprise?
Through the use of Microsoft's Bot Framework and some innovation we believe companies can train "Bots" to help employees find information more efficiently and mobile than ever before! If you don't believe, me have a look at this case study by Microsoft where they created a chatbot to assist educators to find training and teaching resources.
When you think of Bots, you may be thinking of "chatbots"... you're right, those are the same bots that would power your Voice UI. Bots are essentially software programs that automate tasks we would otherwise ourselves. There have always been simple bots that provide a very specific service; However, most recently, we are seeing the emergence of intelligent bots that can serve more complex use-cases.
All that is needed is a voice layer for these bots and we're in business, right? Perhaps, but once you layer on Voice, you need to be much more accurate in your responses for it to have utility... 99% accurate according to Google.. for that you need machine learning, combined with a strong foundation in information architecture and data science.
Machine Learning is a must...
You need some machine learning because information may need to be summarized or analyzed to provide an answer before it's sent back to the person asking for information. Without it, you'll just get a list of search results displayed on the screen.... this might be ok in some cases, but to be relavent a number of factors from the user's profile must be taken into account before coming up with the recommended response such as:
Did you catch, that I said "Recommended Response"?
This is very important in the workplace because we need to ensure accountability and that people are making decisions... not machines. The machine should present a response as a recommendation in almost every case unless it's a know fact (e.g. what time is it?)
Types of Bots
Consumer focused bots like Amazon Echo, Google Now, Cortana and Siri are not really used that often in the real-world. Don't get me wrong, they are very great at what they do but they have yet to make things any easier for people... mostly because they are trying to do too much in our opinion. That said, they are paving the way for the emergence of this type of service and Voice UI in the workplace.
Business focused bots are more widely used and everywhere... These bots are available through platforms and interfaces like Slack, Skype, Microsoft Teams, website chat windows, email assistants, etc. Today, these bots are focused on solving specific collaboration, replacing/augmenting emails, information assistants, support, and speeding up decision-making/communications.
Here are a few examples:
Shane Quigley is an expert in data warehousing, business intelligence, systems analysis, and solution architecture.