AI technology is experiencing a boom in popularity and adoption, and Automatic Speech Recognition (ASR) is an exciting part of that field. More and more devices and platforms incorporate ASR. As data scientists rapidly improve the technology, UX & product designers need to make sure we don’t fall behind. You need to be prepared to create voice interfaces that are intuitive, user-friendly, and accessible. In addition to staying on top of the latest developments in ASR, you need design inspiration. But where do you look for inspiration when the field is still fairly new? Well, you look to other domains that have successfully created engaging experiences with new technology: Like game design.
One of the key principles of game design is the idea of "playability," or the extent to which a game is enjoyable and engaging to play. Perhaps even more relevant here is the “replayability” of such a game—the extent to which the interaction was so enjoyable that it’s worth retrying even if the outcomes become expected as a result of repeated playthroughs. These principles can be applied to voice interfaces as well. You should strive to create interfaces that are not only simple and functional, but also fun and engaging for users, no matter how frequently those interactions occur.
There are many principles that game designers use that can help you improve voice interfaces. Here are our top 5:
Define the environment
Within the game world, you map out different paths the user might want to take, even if most of those paths lead to predetermined outcomes. Extrapolating out from the typical software metaphors we could consider “the user” as something more akin to “the player” of the game, one tasked with navigating nuance to ultimately arrive at the desired outcome. For the player it might feel like a whole world of opportunities, while in reality it’s often a few paths that can come together in multiple ways to arrive at (mostly) the same conclusions. You can think of voice interfaces in the same way: Make sure the user feels in control and capable of maneuvering the paths you have designed for them.
The game’s environment is where the action happens. It can be a fantasy world, a real-world location, or an abstract space. The environment should be designed to have the desired effect on players. The same can be done for voice interfaces. In your design process, consider what you want the user (e.g. the player) to experience.
What tone of voice does your brand have? What are the different tactics the player may pursue to achieve their goals, and how can your best support them on those different journeys? How can you make an immersive experience for the person using your software, that makes them want to engage beyond the surface level of your offering, instead of zoning out in a state of analysis paralysis?
Understand the user's intent
It is important to understand and define clear goals for the interaction. In a game setting, the player’s goal is usually to win. When designing your voice interface: think about what the player wants to win, even if it’s as simple as the correct answer to the question, “How many teaspoons are in a tablespoon?” (Trust us: we know the answer is 3 teaspoons, but we still ask every time.)
As the (metaphorical) game designer here, it’s your responsibility to ensure that the system that you’re creating remains engaging whether it’s a player’s first or fortieth playthrough. Don’t burden players with repeated or extraneous information requests, especially if they’re returning customers.
Let the user explore at their own pace
People will want to interact with your interface in different ways and at their own pace. Regardless of whether they’re a newcomer or a seasoned veteran, every player appreciates knowing they’re on the right path to accomplish their goal.. Folks will give up quickly if it seems difficult or impossible to get what they need, so it’s important to provide regular feedback and rewards. And remember, just as in gaming, longtime players value the easter eggs and shortcuts built into software experiences.
Design meaningful sonic feedback
Games are full of sound effects. As is the case IRL. Whether we’re conscious of it or not, these real-world sounds provide feedback and guidance to us in response to our actions and decisions. I believe there is a lot that can be done in voice interfaces by including good sound effects - as long as it’s done in a way that feels natural. Ask yourself: "What sound does it make when XYZ happens?" Then have fun with it, but only so much
Build in little wins along the way
One of the more important things to take away from the design of great games is to consider the start of the interaction to be an onboarding moment.
Remember how Mario starts? It has a long stretch you can use to learn your movements, then a simple box to interact with before one simple predator arrives. That’s a tutorial right here. Provide your users with easy-to-achieve “Ah-ha!” moment as early as possible in their experience. Ensure that folks achieve that early win which builds the skills they need to eventually beat the Final Boss.
Overall, building a voice interface is completely different from basically any other hardware or software design task. One of the most important things you can do is to develop a clear interaction model, one that’s simple and relevant for your users. By incorporating some of the principles listed above, you can help shape the future of voice interfaces and create more engaging and enjoyable AI experiences for your users.
Level unlocked! 👾
If you have any feedback about this post, or anything else around Deepgram, we'd love to hear from you. Please let us know in our GitHub discussions .