7 Key Tips for Designing Voice-User Interfaces Keep these essential recommendations in mind to build effective voice-user interfaces.
By Nick Babich Edited by Amanda Breen
Our biggest sale — Get unlimited access to Entrepreneur.com at an unbeatable price. Use code SAVE50 at checkout.*
Claim Offer*Offer only available to new subscribers
Opinions expressed by Entrepreneur contributors are their own.
Voice-user interfaces (VUIs) allow users to interact with machines using voice alone. They use speech recognition and natural-language processing to enable users to complete some tasks. VUIs have been a part of sci-fi movies for a long time, but recent progress in this field has made those interfaces a daily reality.
Here are seven essential design recommendations that the Milkinside team follows when designing voice-user interfaces.
1. Define key user tasks
Why will people want to use your product? What tasks do they want to accomplish using it? Those are two fundamental questions that you need to answer prior to creating your product.
The ultimate goal of VUI design is to create a product that will be integrated meaningfully into users' lives. That's why you need to invest in user research early on in the design process to learn the needs of your target audience. You can conduct user interviews and participant observation to learn user behavior — understand how users interact with your product right now and identify a set of tasks that can be good for voice interactions.
Related: Creating a Visual Language for Innovative Products
2. Understand the context of use
Not only how but also where users will interact with your product is important. Will it be a private space like a living room or a public space like a hotel hall? The factors like sound quality (which is a combination of quality of the microphone, distance from a device and background noise) and user-privacy aspects (users are less likely to share private information when they speak in public places) will affect your product-design decisions.
3. Design a dialog flow
A dialog flow is a script that illustrates the conversation between the user and the voice-based product in the context of the user's particular task. A dialog flow is typically visualized in a format of a tree that outlines all branches that represent all situations where the conversation could go. Dialog flow will likely affect your development, so you should specify it right at the early stages of product design.
The design of a dialog flow should always start with outlining a sample dialog. This dialog typically visualizes a happy path interaction with a system (when everything goes as planned and users achieve their goal successfully) and doesn't contain any branches or back-and-forth cases. Introduce more branches as you learn more conditions for your dialog.
4. Design for eyes-free interactions
Voice-user interfaces are primarily designed for people whose eyes and hands are busy doing other things. Imagine a man who's cooking dinner and has his hands dirty. He activates voice UI to learn a recipe. Or a woman driving on a busy street who wants to know where the nearest parking spot is. Even when a voice-based product has a screen, you should try to design an app that doesn't require a glance at the screen.
Related: The 10 Obstacles Keeping You From Great Product Design
5. Design for how people actually talk
Interaction with a voice-user interface shouldn't feel like a conversation with a robot. The more natural conversation is, the better the user experience will be because users don't need to learn anything new to interact with a system.
Here are a few things that can help you design more human conversations:
Design for user vocabulary. Learn as much as you can about the language your users use in their daily conversation and use this language in your product.
Consider all possible variations of utterance (how the user phrases a command). The number of different ways people can ask for something can be impressive. For example, when we want to know the weather, some users ask, "How is the weather today?" while another user might ask, "Is it cold outside?"
Make pauses between questions and options. When a system asks users a question and offers some options to choose from, it should pause for half a second after asking a question and only after that vocalize options to choose from. This helps users comprehend information.
6. Be clear and concise
Comparing graphical-user interfaces (GUI) and voice-user interfaces (VUI), you will see a vast difference — GUI supports multitasking while VUI is strictly linear. When users interact with VUI, they can't skip some UI sections; they have to listen to the information the system provides. Plus, users have to keep information they hear from VUI in their memory, and the more information the system offers, the more cognitive load a user will have. But designers can make users' life better by optimizing the information users receive from the system.
It's essential to reduce the amount of information the system provides to the user without sacrificing the meaning. For example, when a user asks, "What is the weather today?" The system can respond with "The weather in your region, San Francisco, California, today, 20 Dec 2021, is 12 Celsius, sunny, wind 3 km/h," or say "12 Celsius, it's a bit cold and windy outside, wear a coat." The second response is more concise and clear for users.
Related: What the Future of Logo Design Will Look Like
7. Design brand persona
The tone of voice you use in voice-user interfaces has a tremendous impact on user experience. When users start interacting with VUI, they intuit likeability in the first seconds of hearing a voice. It's something people do unconsciously. A brand persona is a combination of personality traits and attitudes that your brand conveys to users. If you don't design a brand persona, your users will do it for you, and the persona they create might not be the same as the one you want to portray.
