Want to keep learning?

This content is taken from the UAL Creative Computing Institute & Institute of Coding's online course, Introduction to Conversational Interfaces. Join the course to learn more.

Skip to 0 minutes and 22 seconds I’m Nicky Birch, and I’m currently a commissioning editor or commissioning exec at BBC Voice and AI. I have a background in voice design, having created my own company, RosinaSound, where we created a couple of long form voice experiences on Amazon Alexa and for Google Assistant. And then I spent some time in R&D really exploring rich voice experiences with R&D, and now, I’ve moved over into the BBC Proper, and we’re looking at how we can take all the work I’ve done in the past and sort of bring it to audiences on a larger scale. I’m Kane Sims. I’m the co-founder of VUX World and VUX World Studios.

Skip to 1 minute and 6 seconds We are a consultancy and design studio that specialise in digital assistants and conversational interfaces. We started out as a podcast. We’ve still got the podcast, VUX World, every week. We interview an industry thought leader and share their knowledge with the voice community. And we also work with agencies and brands to help them navigate this territory and figure out where the value is within the voice assistant platforms and voice interfaces and conversational interfaces in general.

Skip to 1 minute and 39 seconds Essentially, conversational interface is anything that a user can communicate with using their own expression, their own language, be that typed or spoken. And people often confuse what a conversational interface is in thinking that a conversational interface is an interface that you have to be able to have a conversation with. Whereas, in actual fact, it doesn’t necessarily have to do that. It doesn’t have to communicate back to you conversationally. You can still use a conversational interface to do something like turn on and off a light. As long as you can use natural conversational language to initiate that interaction, it’s still a conversational interface.

Skip to 2 minutes and 15 seconds There are obviously interactions that do require more of a back and forth and more of a conversation as you would typically call it, but it doesn’t necessarily have to do that for it to be conversational interface.

Skip to 2 minutes and 29 seconds I think the things that make it a good conversational interface are when you don’t notice it. You’re just having a conversation, and you don’t have to think, how do I say the next thing in order to bring up or take this conversation forward. And unfortunately, that’s most of them for various reasons, whether it’s platform limitation or bad design, but those that are kind of smooth and end up just being like a natural conversation are those that work best. A good conversational interface is one that is typically grounded in a user need. What is a user trying to accomplish? That could be turning on and off the heating. It could be ordering a takeaway. A need could be an entertainment need.

Skip to 3 minutes and 14 seconds It could be that we want to be entertained, want to kill some time because we’re seeing a lot of really good games and interactive stories on platforms like Amazon Alexa. So there is a company called Labworks and another one called Matchbox. They’re making some really interesting games. There’s one called Would You Rather and Question of the Day, and they’ve got real good traction, and users seem to be enjoying them. The goods ones there are when you have actually limited control. So things like Audible, which has always had a very good integration with Alexa, and you can just say, can I have this chapter of this book? Can you skip forward 30 seconds? Can you pause? Can you rewind?

Skip to 3 minutes and 51 seconds And it always understands me, and therefore, it’s successful. A good conversational interface should enable users to speak to it naturally. How you speak is different to how I speak, and how John speaks is different to how Shirley speaks. If it is a conversational style interaction like booking a train ticket for argument’s sake, being able to respond in the user’s language as well. So those are two of the main concerns - having it rooted in a user need, start with designing and creating the right thing, and then, do it in the right way, which is allowing the user to speak freely in their language and responding in natural language as well.

Skip to 4 minutes and 29 seconds Often, like any kind of new technology, it’s good when it saves you time, or it’s more efficient, and also allows you to do something at the same time. The beauty of radio, for example, is that it’s a hands-free medium, and often, it’s a secondary medium. You may be doing the washing up. You may be driving a car. You may be walking down the street. You’re doing something at the same time as you’re listening to audio, and voice enables that plus computational ability, right? So not only can you have a hands-free moment, you can also interact. You can get information.

Skip to 5 minutes and 2 seconds That’s why it’s so exciting, and that’s why lots of organisations and platforms, and there are lots of use cases for it. It’s never going to replace touch. It’s never going to replace your eyes. It’s just another way of interacting with computers. They’ve increased in popularity largely because for generations or decades, we’ve been training ourselves and trained by the devices that we have access to how to use computers. There’s a keyboard here and a screen there. We had to learn how to use this stuff. Same thing happened with mobile. All of the sudden, we’ve got these little mini computers in our pockets, and we need to get our head around how that works. What is an app?

Skip to 5 minutes and 41 seconds How do we navigate through these devices? And what we did is we took the keyboard, and we just squashed that down onto a tiny small screen with a mobile, but still we’re needing to learn how to use the device. Most people, not everyone, but most people learn to talk, and most people have conversations their whole life. It certainly requires less thought, requires less training, and requires less cognitive load on people to be able to use a conversational interface, specifically or particularly a voice interface. We can speak 120 words a minute on average. An average typer will type 70 words a minute, and on your phone, it’s like 35 words a minute.

Skip to 6 minutes and 20 seconds And so voice is, by its inherent nature, a lot quicker as an input device, and as history and technology have shown over the generations that people gravitate towards things that are either more intuitive or less friction or quicker. And so those are the main reason why conversational interfaces now are starting to become the kind of things where we’re now starting to say that conversational interfaces are the interface of the future.

Meet the experts: Nicky Birch & Kane Simms

We learned in the previous step about this history of conversational interfaces. Here we speak to two current experts working in the field of voice.

In this video you will hear Nicky Birch and Kane Simms define conversational interfaces, describe what makes for a good experience and explain the benefits of using conversational interfaces.

Have your say

What kind of conversational interfaces do you interact with in your daily lives? What have your experiences been and do you agree with our experts on the potential benefits of using them?

Share your thoughts with the other learners in the comments section.

Share this video:

This video is from the free online course:

Introduction to Conversational Interfaces

UAL Creative Computing Institute