Over the next ten years, a new, huge market will open for designers that until now have been extremely niche: Voice Designers.
Voice design? How does it vary from conventional UX design?
Actually, there are many similarities. User research, persona creation, prototyping, user flow creation, usability testing, and iterative design will all be necessary steps in creating enjoyable, useful voice interactions.
But there will be some huge differences too.
In this article, I am going to look at some of the things that make voice interaction differ from regular interactions carried out on graphical user interfaces. I will proceed with suggesting some tips on how to think differently when designing for voice.
1. Throw your User Flows out the Window and Start Over
Why? Users may well use voice interactions to achieve different goals.
Imagine you have a music streaming website. Let us say you are jumping on board with voice now, in an easy way: through creating an Alexa skill.
You should think carefully about which goals users will want to achieve using voice interaction. You may initially offer a stripped back version, for example not allowing to create playlists, but rather start playing a playlist.
You may carry out user research and find that actually when controlling music by voice, people do not want playlists – they want something different. They might want to spontaneously call out the name of a song of their choice and listen to that!
2. Make Sure Your Use Cases Make Sense
Naturally, companies will get excited about voice and want to jump on the bandwagon.
“Redesign this feature for voice”, the executives will cry!
However, voice UIs should be user-led, not company-led. User research will be crucial. UX designers will have the exciting task of understanding which situations people feel comfortable using voice command in, for example in a car and definitely not on a train.
It might make sense to order groceries via voice, but for example shopping for online clothing will, we can assume, remain a visual affair!
While this example is obvious, other ones will be a bit more challenging. As we get used to voice, we will find it easier to achieve more and more goals using it. For example, following a recipe may seem more visually intuitive at the moment, but when you consider having your hands dirty, a recipe delivered via voice over time (thus also serving as a timer) could be a significant improvement to the UX of cooking.
3. Communication Beyond Talking – Show System Status
When we use a graphical interface such as a laptop or smartphone, we are used to seeing what it is doing. Is it loading a web page? Is it 10% of the way through a video?
With voice, we do not have these visual clues that tell us the system status. How will I know if my voice device has crashed?
There are already new conventions emerging, for example, Amazon Alexa devices use subtle flashing lights or non-disruptive noises to give you clues about its status.
4. Make Personality Your Unique Selling Point
This is one of the exciting things about voice. Where previously, designers might have helped determine the tone of copy, colour schemes, or image guidelines, now they can create a whole personality.
Having a system persona that is delightful to interact with will give technology companies a huge boost in the race to create the best voice interactions.
5. Make Sure that Commands are Understood to Prevent Terrible UX
Previous voice products have failed to catch on due to the system being unable to understand different accents or unexpected commands. As a result, users would often get frustrated with voice technology. Despite being around since the 90s, it is only just now gaining traction.
This is because cloud computing, artificial intelligence, and machine learning are making the systems behind voice much smarter. They can instantaneously access vast quantities of data, meaning they can understand lots of different variations of speech and, of course, respond accordingly.
Teams working on developing voice recognition software from scratch will usually have a team of specialist scientists, applying statistical modelling to make sure a wide variety of speech input is recognised.
However, if you are designing for an existing voice service, for example creating an Amazon Alexa skill, you may well find yourself having to create a list of different commands. For example, order me a pizza, send me a pizza, deliver me a pizza, bring me a pizza, get me a pizza, I need pizza, I want pizza… I think you get the idea!
There will be a lot of design and development hours required to get us there, but it will be fascinating to see how voice will transform human-computer interaction over the next decade.
Machine learning has created the potential for exponential improvement, and it would be no surprise if soon machines are more capable than humans when it comes to understanding a drunken answer machine message.
If you are interested in voice, now is a great moment to get involved. From voice recognition scientists, to voice artists, to programmers, and of course UX designers, there will be a lot of demand for talented people to lead this exciting shift.
If you are a designer and want to specialize in voice, check out the 8-week online course by the company I represent, CareerFoundry in conjunction with Amazon Alexa.
(Lead image: Depositphotos – affiliate link)