Conversational AI signals a huge advancement in the way we interact with computers. Unlike menus, touchscreens, or mouse clicks, using our voices to have conversations is one of the most natural ways to use a computer; it requires no learning curve.
This new method of human-computer interaction makes powerful computer applications even easier to use and accessible to more people. For example, instead of having to make several swipes and clicks to play music, you can simply say, “Alexa, play the top songs in Seattle.”
Conversational AI systems are computers that people can interact with simply by having a conversation. With conversational AI, voice-enabled devices like Amazon Echo are enabling the sort of magical interactions we’ve dreamed of for decades. Through a voice user interface (VUI), voice services like Alexa can communicate with people in ways that feel effortless, solve problems, and get smarter over time.
Conversations can be conceptually and emotionally complex. When we talk to each other, how we say something matters just as much as what we say.
Computers can’t grasp these nuances, and that’s where voice design comes in. A well-designed VUI is flexible, and takes into account these unwritten rules of conversation.
Here are five key elements to remember when designing conversational experiences.
Conversations are more than simply understanding words and sentences. People expect to interact in natural and unconstrained ways, using the phrases they prefer in the order they prefer. They will correct themselves and change direction during the course of a dialog.
Everyday conversations take into account more just the words we speak. Delightful conversation design involves giving context about why, when, and where. For example, “Alexa, remind me to buy groceries tomorrow,” is a simple request during the daytime, but at 1:00 in the morning, “tomorrow” could mean a different thing.
Conversations are dynamic and inconsistent. They require both parties to understand, respond, and remember what the other participant is saying. Design the VUI so that Alexa captures the important parts of the conversation no matter how it is presented. For example, if a user provides both a destination and a date when requesting to book a flight, Alexa should avoid asking the follow-up question: “What’s your departure date?”
Face-to-face conversations are filled with personality, emotion, and even surprises. When appropriate, incorporating these quirks can make interacting with Alexa feel more like having a conversation.
Conversations are not isolated occurrences. They build on previous questions and answers, whether from five minutes ago or five days ago. When appropriate, voice-first experiences should account for prior context to make the conversation more engaging and relevant to the user.
There are many other layers to creating natural conversational experiences. For example, you should present only the most relevant information needed in a concise and simple way. Then, gradually build on the user’s response with follow-up questions and information. Also, remember that we often use many combinations of words and phrases to mean the same thing. Sometimes we ask imprecise questions. Sometimes we tell jokes. A well designed voice experience becomes a conversation when it takes all of these elements into account.
An AI-driven approach to dialog management that lets customer use the phrases they prefer in the order they prefer and requires less code.
There are many elements to voice design, but you don’t need to be an expert to start designing and building voice experiences. The Alexa Skills Kit (ASK) is a collection of self-service APIs and tools for making Alexa skills. Skills are like apps for Alexa, enabling customers to engage with your content or services naturally with voice.
Join hundreds of thousands of developers who are building Alexa skills to engage and delight customers on hundreds of millions of Alexa devices.