Ilarna Nche: Creating Voice Skills, Actions, and Capsules – Episode 72

Podcast: Play in new window | Download

Subscribe: Apple Podcasts | Google Podcasts | Stitcher | TuneIn | RSS

Ilarna Nche is an award-winning expert on the technical aspects of building voice applications for Amazon Alexa, Google Assistant, and Samsung Bixby.

She designs and develops voice apps for clients who want to capitalize on the branding power that voice products bring to the digital content mix.

Ilarna and I talked about:

• her path into the voice design industry
• how easy it is, even for someone new to the field, to develop voice applications
• how her team at Matchbox works together and with their clients
• how voice creates new customer experience opportunities
• some of the uses and situations that are well-suited to voice apps: games, wellness, cooking, car travel, etc.
• some of the difference between voice platforms like Amazon Alexa and Samsung Bixby
• how less-technical development tools like Jovo and Voiceflow permit folks with voice-design skills to build apps on their own
• the wide range of platforms where voice apps can show up – Bixby for example in Samsung appliances – and the proliferation of devices designed for voice interaction
• the importance of staying on top of trends in the fast-moving world of voice
• the basics of the technical implementation of voice recognition, natural language understanding, and generating query responses
• the future of voice technologies and applications
• how to get started designing your own voice applications

Ilarna’s Bio

Ilarna Nche is an award winning Alexa Champion and Bixby Premier Developer working as a Senior Software Engineer at matchbox.io. She has a huge portfolio of voice applications across Google, Bixby, and Alexa and is an expert and thought leader in the voice & conversation design field, from both a development and design perspective.

Links Mentioned in the Podcast

• Jovo
• Voiceflow
• Question of the Day
• Find My Phone
• Music Bop Adventures
• echo frames
• echo loop

Video

Here’s the video version of our conversation:

Podcast Intro Transcript

When she left university just a couple of years ago, Ilarna Nche started experimenting with voice technologies like Alexa and Bixby. She learned very quickly that her technology and design skills were a great fit for this new medium. Voice tech also plays to her preference for voice interactions over graphical interfaces. In just two years, Ilarna has already earned several industry awards for her work. But she’s quick to point out that you don’t have to be a technical expert like her to build engaging voice experiences.

Interview Transcript

Larry:
Hi, everyone. Welcome to episode number 72 of the Content Strategy Insights podcast. I’m really happy today to have with us, Ilarna Nche. Ilarna is a senior software engineer at Matchbox.io. She’s also won a number of awards for her work as a voice interaction designer. So welcome to Ilarna. Can you tell the folks a little bit more about yourself, your background?

Ilarna:
Hi everyone. Yes. So yeah, my name’s Ilarna and as Larry says, I am a senior software engineer at Matchbox. But I started off my voice journey, building apps after university, where I just wanted to get to know the voice industry. It was completely new and I really enjoyed it, mostly because I didn’t have to design anything visually because that’s where I lack in my skills. So just to have this application where you could basically create an application without any visual interaction, was something that really attracted me to the voice industry. And so I’ve made over 50, 70 applications on Amazon Alexa, and that’s where it started off for me.

Ilarna:
And I managed to create an app called Music Bop Adventures, which won an award for being a kids app and that helped me with the money, the prize money I won there, it made me realize, there might be a future in this. So I spent the year after university, basically building all these voice applications, and then I managed to become a Bixby partner with Samsung Bixby, and also an Alexa champion and then also Bixby Developer of the Year, and also join the company Matchbook, who are responsible for applications such as Question of the Day and Find My Phone. So yeah, it’s been a very good journey for me, all over the space of two years and yeah, I’m looking forward to-

Larry:
All that’s happened in only two years, that’s great. We were talking a little bit before we went on the air about, and most of this has been a one-person show, that’s one of the things that fascinates me about the voice world, is it’s this super powerful technology, but the way you operate goes back. I’m old enough to have been a webmaster back in the day when that was a thing, when you wrote the copy, you built the website, you did the graphics, you managed the server, and all that stuff. It seems like that’s how it is in the voice design world, that you can operate up and down the stack of skills that are needed to create. Well, actually an interesting, because Alexa calls them skills, Google calls them, they have a name for it, right? Oh, actions. And so you can design all these pretty much by yourself.

Ilarna:
Yeah. I mean the entry level really was building a fact skill for example, and they had templates and loads of supportive documentation, that one person could essentially create Alexa skills in less than an hour. And it’s so incredible to see how many people have built on that and have now stretched out and expanded into being their own company or business or agency and hire multiple people to really bring and deliver more experiences to the voice platform.

Larry:
Because you’re the senior technical person at Matchbox, right? But there’s a bunch of other, there’s content people and managers and executives as well. Is it like an agency model, is that how you operate?

Ilarna:
Well, Matchbox, we’re different because it’s more about voice as the brand, whereas an agency will be delivering experiences for other clients, Matchbox is more like the customer and the voice is our brand. So essentially we still have a team behind, someone focusing on content, someone focusing on the analytics, marketing, and the development, but yeah, in a sense we’re a team that’s creating these voice experiences.

Larry:
That’s so interesting. And I can see how you were drawn to it. Like you were saying your, not aversion to, but just you’re not a visual person and great, there’s this voice thing, I’ll go and do that. In terms of the technical implementation of the work, do you work with a product manager or do you fill that role yourself as the designer of the skill or the actions that you’re doing?

Ilarna:
Yeah, so we work as a small team together, pitching ideas and also trying to figure out what would be listening to customer feedback, which is number one, because at the end of the day, they’re the ones using your product. And we just all come together and focus on that, on what we can improve really

Larry:
I want to back up just a little bit because you mentioned before, there’s this whole new world of voice apps. With other kinds of content there’s publications or blogs or podcasts or social media streams of feeds. This is a little different ecosystem. Like you said, you’re developing branded voices or voice brands and then are you then deploying them as specific skills and actions in the voice world, you’re augmenting brands that already exist or are you creating new stuff as well?

Ilarna:
Yeah, I mean, a lot of the stuff with voice, creates new opportunities to create a new experience. But on top of that, you can essentially take something from a mobile app and convert it into a voice app if the opportunity is there. So most of the time, take question of the day, the concept is very similar to other trivia questions. It’s a trivia game and a lot of Alexa skills and apps. There she goes.

Larry:
That’s right, I always have to be careful in a voice …

Ilarna:
Yeah, one of the things. Yeah, but a lot of the skills or applications actions that you see was mostly games because I think, especially with it being so new, the driving force is games are very sociable, they provide a way to interact with longer retention periods with the user. And so I think that it’s been very reassuring to see that on a voice platform, the most used situations would be, or user cases is, listening to music, using if for smart home purposes or essentially playing a game. And we can tell with Question of the Day, with it being a daily use application, they’re coming back for more. So I definitely think that it’s important when looking at these different applications and experiences, is that you look at the app store, you look at what’s being done, you look at what’s not being done and see if you can create a voice experience that would actually benefit the user.

Larry:
And it benefits the user, but there’s also a benefit to the business when you can engage people like that. I think it’s so interesting that like in other parts of the digital world, gamification is a thing because that’s a way to keep people engaged and it’s naturally built in to what you’re doing out of the gate here. The fact that games are such a … they’re the ones that the games get traction quicker than other kinds of skills that you do, is that the …

Ilarna:
Also I’d say wellness skills as well. You’ve got an applications and actions, like helping you to sleep, I think that’s one of the very big factor is helping people to go to sleep where they can literally, before they go to bed, they just say Alexa or Google, and then launch an app that will help them go to sleep by playing rain sounds or sleep noises, anything to help them. And it’s become a part of the routine for many people.

Larry:
Well, those two examples are really interesting because I think for a lot of people who are coming to voice new, what is this best at? And those two examples are great: games because you can almost picture playing a trivia game with your car phone or your voice assistant or wherever you are and the wellness, that that makes perfect sense, the sleep stuff. And you can also picture in workout routines, you don’t want to look at an interface, you just want to have your … Are there other applications that are emerging as particularly well suited to voice?

Ilarna:
Yeah. I mean, I’ve noticed that we’re starting to see more brands in big companies come into the fold of voice simply because they see an opportunity for marketing their own kind of, if they have a movie coming out, for example, they’re looking at another marketing channel to go through with that, whether that is make it more interactive and give you more special access behind the scenes of whatever they’re trying to sell. I mean, cooking as well has always been something that people, especially because depending on where … we’re at the stage where we have devices with the screens where the best place to put them, I would say, would be in the kitchen and so with that, gives you opportunities to follow along, whether that’s with recipes and cooking. So I definitely think as well, I’ve always said in the car is a new opportunity that will be coming to voice platforms and assistance soon. So I definitely think there’s an opportunity there as well.

Larry:
Yeah. The thing about all those environments you just described, they’re all hands free. I’ve seen flour covered iPads and kitchens and car wrecks caused by people trying to text. So those seem like really good safety and convenience features. So you’re native to voice as a designer, so I wanted to ask you, I don’t know if you’ve ever talked to people who come out of more the GUI world, as opposed to the VUI world, the graphical world, even though you’re not a visual person, I’m just wondering, do you have a feel for what the unique challenges of developing in this environment are? The reason I ask is that so much of the GUI stuff comes out of millennia old technologies of publishing and reading and things like that. I mean, in some sense, voice is even older, in that that’s probably how we’ve always communicated, but in terms of the technical implementation of it, it seems newer. Are there ongoing challenges that you face as you develop apps?

Ilarna:
Yeah, definitely. I think when obviously Amazon came out because to me, that was the first thing I was introduced to. They had obviously their base languages, like Node.JS, Python, which I had familiarity with, which was a big help in creating the app, but they also had their own SDK, which obviously learning an SDK, there’s different stuff that you have to learn with them. And that was a struggle at first because it was new rules to get used to, new syntax and stuff. But you do realize once you get used to one platform and then you move to another one, it’s completely different and you have to ignore everything you’ve learned from the previous platform, because in some ways or another there’s new terms and new ways of doing stuff and a new approach, I think, especially with Samsung Bixby that came out, was introduced last year.

Ilarna:
Their way of building a Bixby capsule is completely different to the way that Amazon Alexa, you build a skill. Like I say, Samsung Bixby, are more model focused when it comes to building, so you could create the same application in Amazon and Bixby and come up with so much less code in Bixby compared to Amazon, because Bixby have come up with their own way of meeting some of your demands on the platform. So I definitely say it’s so different when you do that, but there are solutions such as Jovo, which I talk about a lot, because it eliminates that issue where it’s cross platforming, so you can literally have one code base and write for all platforms, whether it’s Samsung, Amazon, Google, and just do it on one code base, which I know a lot of people use because it’s so well spoken about, so well maintained as well.

Larry:
Well that kind of tool. I want to back up just a little bit for people who are less technical, you used the acronym SDK earlier, which is a software developer kit, I believe, because not too many of my listeners are programmers I don’t think. So I just wanted to clarify that that’s … But basically any software development environment comes with IDEs or SDKs, things that help you operate within that environment. Anyhow, that’s the technical underpinning to it. So it’s actually, and you also mentioned Node.JS and Python, so to be this one person show, do you have to have those technical skills or would a tool like Jovo, for example, would a nontechnical person be able to develop apps with something like that?

Ilarna:
So yeah, good question because what I’ve realized from this whole voice industry, is that developers may not be well suited to this because VUI is so crucial in terms of how you can interact with the user, because it literally is voice that’s driving the application. So a lot of people like content creator, story makers, they would benefit the most from voice. And so there’s tools out there that you can use to actually build an application without any technical knowledge. One good example is Voiceflow, so they literally allow you to build applications if you don’t have any technical knowledge because at the end of the day, it’s so important that people who specialize in this, can create it as well. And I know Amazon and Google, they all have these ways to help people who are not as technical, build their product easier.

Larry:
I guess, are those products that Amazon and Google create themselves or are there third party developers who create things in these development ecosystems?

Ilarna:
No. Voiceflow for example, was a very big example of a team that made this software tool where people can build the application and have the backing of these big organizations, such as Amazon, to be like, “Look, you can make your own.” Because it benefits everyone at the end of the day. So if you can encourage people to come and build for your platform and give them tools to help them out, then yeah, it’s a win, win. There’s a lot of funding backed by these tools.

Larry:
Yeah. Because that’s how it always works. There’s this whole underpinning to any industry where the big platform players are trying to get mind share and market share. But for you, it’s interesting, a lot of people who profile like you, would settle on one platform or another, but you’re just in Amazon, Google, Bixby. Are you typical in that way? Is that a common path for somebody like you or do some people settle into one or the other?

Ilarna:
Yeah, I’d say it’s mostly because of experimenting. I want to experiment what everyone’s doing and as well, because it was so new, I like to get in there and experience what everything is all about. So because Amazon came out first, I mean, I had no idea that people were making third-party applications when I first heard about Alexa, it was something where I was like, “Oh, wow, there’s these applications and people making them.” And I wanted to jump on that. And then when Bixby came, I wanted to be a part of that as well, because this is Samsung who has millions of products available, and you could have Bixby potentially on those platforms, all those fridges and washing machines, that’s a big user base where they could be using your app. And it still amazes me to this day that you’ve got free platforms, well known platforms and there’s many other platforms that will come into play as well with this whole of voice industry.

Larry:
Right. You just mentioned that because I don’t know that much about the Samsung Bixby and I’ll put links to this stuff in the show notes so that if people want to explore more. But you mentioned, of course Samsung makes appliances, so a lot of this is like internet of things, IOT kinds of stuff, where you’re talking to your refrigerator. That’s so interesting. This is a big and ever growing environment, it sounds like, of possible places.

Ilarna:
Exactly. I think right now, my mindset, when I think of Alexa, Samsung and Bixby, Samsung, and Google, are so different. When I think of Alexa, my first thought goes to the echo devices. And I think that hopefully we’ll get to a point where Alexa will be everywhere, where you won’t refer to it as something that’s a device and you’ll refer to it as something you can access everywhere. Whereas with Google, you literally have your Google Assistant, which is available on your phone and you could carry that anywhere for Google and I think Samsung is the same thing. So yeah, I think you can tell that even though Alexa is available as an assistant on your phone, I still refer to it as the devices that you’ve got around your house and hopefully with the car and everything and all the new devices, like the glasses, the echo frames and the echo loop ring, I think they’re trying to make a point that they’re going to be everywhere, you can call Alexa anywhere.

Larry:
Right. This is going to be a weird new immersive world where we’re just surrounded by gadgets.

Ilarna:
Exactly.

Larry:
Let me ask you, I know you’re mostly solo in your implementation and execution of the projects, but tell me a little bit about the team around a voice product. Do you generate your own ideas or do clients come to you with ideas? How do these ideas for these, like the trivia game, I forget the name of it, the Question of the Day, are those kinds of products, were they existing things in other media that you bring to voice or how do these ideas generally come to voice?

Ilarna:
Yeah. I mean, before I worked for Matchbox, I had loads of ideas I’d written down, stuff I wanted to see and be able to work on and use generally. And also they were very novelty stuff, but now working with Matchbox and having something like Question of the Day, which is something that was made years ago when Alexa first came out, to just having an idea, a simple idea that has now attracted millions and millions of users. It’s very important that you either look at your portfolio of applications and think, “Right, which ones, now that years have passed, which ones are effective, which ones need work, which ones should we look to do? And the world changes really quickly, so something that was popular last year might not be popular this year. So you just got to keep up with the trends and think what would benefit people now.

Ilarna:
Obviously we saw with everything that’s going on, is there something that people can do to adapt to that, to change that and we know that voice assistants have had increased usage during this time. So this is the time to look at your applications and think what is missing, what can I do, what would be a benefit here?

Larry:
Yeah. Nice. I think that’s my hope and asking these questions, is that how can people figure out the best things to do with voice, what it’s best at. Because a lot of times, like that Question of the Day seems like a perfect, wow, it’s even better in voice than it was in its prior iteration as a other kind of digital product. And so I guess that gets, the title of the podcast is Content Strategy Insights and one things about content strategy is like, what is that content and where does it come from? I think in voice interactions. In publishing, digital publishing models, there’s an underlying CMS, where the data is structured in some way and then presented out to various channels.

Larry:
This is mostly a single, well, I mean, there’s a variety of devices, but it’s a single channel in the sense that it’s a person talking to a machine and getting answers back. I guess I’m just trying to get my head around the implications of that ecosystem, again, back to the team. I continue to have this incredulity that you can do so much yourself with this, because there’s so much going on under the hood. Well, I guess maybe let me back up from that a little bit. So the basic interaction that’s happening is a person is saying something to a machine, it’s doing some voice recognition, which is like … like you and I, I have this Western messy American accent, you have a little bit of a British accent. We’re negotiating that just fine, no problem at all in this conversation, but to a computer that whole voice recognition, that’s a whole thing.

Larry:
And then there’s the natural language understanding that has to happen to turn that into something you can ask a machine and then that machine has to spit out the right answer. And that’s where we get into the content, is that level, that layer in the machine, where it’s like, “Okay, I’ve understood your question, here’s the answer.” Where does that content come from? What is the source of the words that answer the questions?

Ilarna:
Yeah. So like you said, depending on the platform, it’s different, but the main underlining use case is that you would grab whatever the user has said and then it’s up to you to decide where that goes. So for example, if you’re asking for the weather, that weather part, where you’ve asked for the weather will then go back to us, we would then be like, “Okay, what do we want to do with this query? Do we want to tell them the weather, or do we have a fall back in case that is not our intention?” So it’s all about setting up and modeling a perfect model so that you can grab every intention, and every situation where you can catch what they’re saying. And yes, voice recognition is something that is very highly, we’re reliant on.

Ilarna:
If it doesn’t work, then it’s out of our control really, but it’s getting better and the more and more people are using these voice assistants, we’re getting to understand that there’s different dialects and different accents that need understanding. It will slowly get better to get the queries. I know that one thing that is a big, big misunderstanding is, words that sound alike. So obviously if you say, for example, C, it could be the letter C, or it could be S-E-E or S-E-A, and you have to understand what context would that apply to, because it’s something out of your control, but you have to make sure that you can attend to it really.

Larry:
Well, I guess for now, that’ll just be the magic thing that happens in the background that we’re happy is there, because there are these machines that can understand those semantic differences. And there’s a whole world of the semantically structured content folks who can help put that together. Hey, I notice we’re already coming close to time, these conversations always go so quickly, but I always like to give my guests a chance. Is there anything last, anything that’s come up in the conversation, or that’s just on your mind about voice design or voice interactions that you want to make sure we share with the folks?

Ilarna:
Yeah, definitely. I think that if you have the chance, definitely take a look into voice because just like mobile phones, I think voice is now going to be something that will be there forever in terms of interaction, whether that is complimented by a screen, which we’re slowly starting to see. But I’ve always said that if there’s a technology, we’ve one upped before with the smartphones, with touch, and now we’re going to voice and maybe even a multitude of the two. So I definitely think that people should have a look, virtual reality as well is something that I think will definitely come up with voice and they will compliment very well together. So yeah, I’m very excited to see where in five years time, or even in three years time, where everything will lead to.

Larry:
And it sounds like, that’s one of the things that’s so interesting to me about this, is that it is a pretty democratized realm that somebody could just get, I think you said Voiceflow, Would that maybe be one of the easier, if somebody just wanted to try this and play with it on their own, that would be-

Ilarna:
Understandably, yeah. If you have no technical experience, Voiceflow is something to look into. If you have some experience, like beginner’s experiences, theres’s so many courses, I didn’t have any experience in any voice before I started. So just learning through the documentations and the helpful courses, it was really handy and really helpful.

Larry:
Nice, well that’s good. Well, thanks so much, Ilarna, this has been great. Oh, one last thing. How can people, if people want to follow you on social media, on the interwebs, what’s the best way to keep in touch?

Ilarna:
Yeah. I’m most active on Twitter. I’m not really that good at social media, but I try, so Twitter, @Ilarna_preach is my Twitter handle. That’s where you’ll find me most.

Larry:
I’ll put that in the show notes as well. Well, thanks again, Ilarna, this was a really fun conversation. I appreciate you taking the time to talk with me.

Ilarna:
Thank you. Thank you for listening.