3 Pillars of UX For Voice User Interfaces.

Principles to make the experience of Voice User Interfaces desirable for the users And give you an edge over your competitors.

Originally published at rohanmishra.design on November 6, 2018.

Are you a Designer, Developer, or a product person making or enhancing the experience of a Voice User Interface? Or curious to learn about UX for Voice User Interfaces (VUI)?

If you answered yes to any of the previous questions, then this article is a must read for you. So, you don’t end up making a VUI that nobody needs.

Three Pillars of UX for Voice User Interfaces.
That will make the experience with your System desirable for the users, And provide you with an edge over your competitors.

Today, more than ever, companies are investing in making Voice based products. And it makes sense, as for the human voice is the primary way to communicate.

With more no. of voice-based product companies want to attract users on their side. So, they can’t entirely rely on better technologies to provide a pleasurable experience.

In this article, we will explore the things that make an awesome experience of Voice User Interfaces. And how you can use the same three principles to make your VUIs experience more desirable.

Our Echo understands us better now.

I can feel that by spending the last week with my Echo after a long time. Sending Echo in search of the toughest questions, I found.

It didn’t let me down (most of the times :P).

“But how come is it getting so good at understanding my queries and having engaging conversations?” I Wondered.

Last I remember me having a discussion with my colleagues a few months back, about the snappy comebacks and rude responses to our queries.

It has come so far. So, I thought “What makes UX of User Interfaces so conversational, effective?”

An exploration and playing several hours with this Echo Plus and Google assistant on my phone I observed certain patterns that added so much to this wonderful experience.

And as I think of these patterns and the principles derived I have a firm believe this can add so much to the devices that are coming these days.

We are surrounded by Voice User Interface (VUI) devices from Google Assistant on mobile, Cortana on windows to Siri on our Apple devices.

We ask them everything useful and everything weird as well. Right?

Sometimes just to check what comes back. Right again?

But only if these devices can understand us better, respond better (or just talk), and get us relevant information. In contrast, most of them leave us frustrated and annoyed.

So, the User Experience of these devices needs to be way better. If our aspired future is with most of the conversations in verbal form.

To get the answer of most fundamental questions of designing a voice user interface. It makes sense that we understand whom we are designing it for.

How hard could it be? You asked.

Do you remember this viral video on Youtube?

You might remember this viral video, where this Italian old woman is constantly trying to use Google Home and fails.

It seems funny in the beginning, but is it the human error, or it has deeper hidden problems with the experience?

For a better understanding of the functionality of any product, we need to first understand the users.

Know thy users:

Does this device understand It is having a conversation with 5 years old or a woman in mid-eighties?

It comes down to whether the device can understand what the user wants and help them do that or not. To give a relevant answer the device needs to understand the query first.

So, knowing the basic principles of User Experience will help you make a better UX for Voice User Interfaces.

In past talking to a box was considered so unnatural. Fast forward to today, you can hear the voice coming from homes like Ok Google and Alexa more than often. And according to many reports, the no. Of such devices are increasing at a rapid scale.

These are just going to increase every year. And, if we really need to develop these Voice User Interfaces we should know some of the fundamentals of User Experience Design in these devices to deliver a pleasurable and desirable experience.

You might be thinking Voice User Interface (VUI) is just the Google Assistant or Siri on a Mac so let’s first address the question “What is a Voice User Interface?”.

What is a Voice User Interface?

Voice User Interface (VUI) enables users to communicate with the computer using the voice/speech. VUI doesn’t always necessarily be a device like Google home or Alexa. It can be used in any speech application.

For a long time, we have been communicating with the Computer with the help of text.

But the most natural we communicate today is by voice. So, it makes sense investing in Voice based Devices.

For the longest possible time, human needs haven’t changed much, but the way to access and the context has changed entirely. Voice User Interfaces (VUI) use speech recognition technology to enable users to interact with devices and applications using just their voices.

Virtual assistants like Google Assistant, Amazon Alexa, Apple Siri all brought Voice User Interfaces in the foreground. Talking to these metal boxes doesn’t feel like a cliche anymore.

VUIs allows for hands-free, efficient communication which is a very good thing in making the systems bit more Universal. As I explained in my article Universal Design: Design for everyone about the fact that designs all around us in real life are made for mass adoption.

UX Principles are still relevant.

Most of our communication with these Voice based devices ends with frustration due to their lack of understanding and empathy.

now I tried asking Cortana to find My Visiting Card, it ended up doing a search for: And if VUIs is going to be a mainstream technology it needs to get better.

UX for Voice User Interfaces matters when we think about all those experiences. Most of the companies out there are fighting on the data and more relevant response so, the technology with a better user experience will definitely have a competitive edge on others.

AI: At the Heart of Voice User Interfaces

AI (Artificial Intelligence) plays a major role in these voice-based devices, parses the user’s request. Helps understand the query, decide the relevant output and give it in more understandable form to the users.

Without the advancements in AI, Google Home is more of a pretty piece that lies on your Tabletop. AI is used to determine what is the best possible answer to handle this kind of query from the data it has been fed from the data engineers as well as previous interaction with the users.

Artificial Intelligence (AI) with the help of Machine Learning (ML) provides more relevant answers to user needs and desires. AI enables the Alexa for Playing a song, or deciding normal query from a web search and making a calculation.

Conversational Design: Skill of future Voice User Interfaces

We like to talk more than anything. It is the way we get around our life every day. And not surprisingly, we want to talk to these interfaces as we talk to humans.

So, to act conversational as humans do the interfaces should have mainly three characteristics at the core.

• The Conversation should Sound natural, not superficial by any means. Keep the answers relevant, short but descriptive.
• Interfaces should be cooperative, after all, they are there to help. An example could be remembering the last request and using it in context to a new one.
• No one like to cut in the middle of speaking, Take turns while speaking. Listen to the whole query or statement of the user before processing the results or answers.

With all that stated now, it’s the time to dive straight into the three pillar principles of UX for Voice User Interfaces.

Three Pillars of UX for Voice Interfaces

The UX for Voice User Interfaces system depends on 3 characteristics.

Comprehension: Did the System understand the query well?

While designing voice user interfaces, always remember the first step in providing relevant results is to understand the query and the context in which it is asked. The question could be a follow-up or a new one altogether.

With that being said also the voice of the user as we saw earlier in the video that the accent was the main concern for that Italian lady, not the query. Understanding our query, and giving us a chance to correct if it made a mistake understanding the query counts.

Understanding the query: Also means giving feedback, like we humans nod when we understand something.

It could repeat your query for example If you say “Call Gaurav” it should repeat “Ok, Calling Gaurav jr.”. This also gives the user feedback that the command he gave is properly understood by the device. Also the ability to modify or cancel.

If it doesn’t understand us, it will give irrelevant results. We might get frustrated. If the system doesn’t understand the query and keep getting useless results, it will cause the lower lifetime cost of the users. And increase the chances that they might switch to competitors.

Now, as you understood the importance of understanding, now it’s time to discuss, How to keep the conversation engaging? So that user come back to have one again.

Personality: Do user want to have a conversation again?

Imagine you are chatting to a real human being, you are sending paragraphs. and all other person is doing is replying with Yes/ No, okay, hmm…

How would you feel?

I bet it would be a big Turn-off for you. So the same things will happen to these if they keep responding like yes, no and stuff.

So, the users will get a feeling this thing is not interested in talking. And we want to humanise the experience as much as we can.

Do you remember the first tagline of Google Assistant? It was a promise, “Your personal Google Assistant”. Don’t you feel the personal human experience in that?

The problem with most of the Voice User Interface system I have seen is they are rude. And that needs to change.

someone has a conversation with these devices, the feeling should be of Enjoyment. And it all depends on personality (Style and tone) of the assistant.

So, firstly let’s understand what personality is… Personality is basically a set of habitual behaviours, thoughts, and emotional patterns that evolve from biological and environmental factors.

The factors in personality:

The system should be concerned about me (personalized results) and more precisely only about me. After all, it’s my personal assistant. I observed that use of humour, compassion makes the experience engaging. It feels like a friend.

Companies are shifting from robots to human sounds as we do not like to talk to robots. We are humans, and we like humans.

The point is the systems should not sound robotic. While we are discussing personality in having a conversation. The experience could be even better by picking better words. Also, complimenting user's choice once in a while won’t hurt.

And last, but without this nothing else matters is Results.

Results: Is this answer relevant to the query, and the context?

The system should help to get the desired results. What good a luxurious car is if it doesn’t have an engine. Until unless you are looking for a big tin box that remains in a showcase.

We should not forget the use of Voice UX is to give users what they need it could be a simple conversion thing, playing music or turning on lights.

Start building for humans:

While building a UX for Voice User Interfaces we should think about the understanding the query of the user, having a positive conversation, personality, and the ability to provide a relevant result for the query. We should invest in technologies, but to provide the best experience we need to invest in experience.

These three things separate the box with answers to an experience that is engaging and enjoyable. Always remember you are making it for humans.

Thanks for reading. 👏

Liked this article? Please tell me some of your learnings in the comments or drop me a Mail.

Want to learn more about Product and UX Design? Register here: https://rohanmishra.design/blog/process/ux-voice-user-interface/#register

Originally published at rohanmishra.design on November 6, 2018.