Voice This! Podcast: Episode 4

Concept Generation with Brooke Hawkins

In this episode, we look into concept generation with Brooke Hawkins — Conversation Designer at MyPlanet, and voice tech ethics researcher with the Open Voice Network. Brooke leverages her unique perspective as we explore how to transform research insights into early concepts. as well as how to stay mindful of your audience and the potential blind spots in your data throughout that process. Other topics involve inclusivity in the design process, and how the value of automation may be shifting in the aftermath of the COVID-19 pandemic.

Highlights:

• [05:44] Brooke’s definition of concept generation
• [07:13] Brook’s approach to initiating concept generation
• [13:03] Brook’s perspective on conversation design tools
• [17:06] Designing with consideration for those impacted by your work
• [21:56] Incorporate input from representative users/consumers in research
• [22:54] Brook’s take on the ethical considerations of conversational AI
• [26:18] Brooke’s perspective on the future of conversational AI

[05:44] Brooke’s definition of concept generation

I think concept generation post research really looks like taking all that excellent research you did and synthesizing it into clear user needs. … it’s in an ideal world and as much as possible, I think just trying to incorporate those real user insights into meaningful, you know, principles or approaches to solving something that you think will really solve that user need, rather than kind of putting your own bias on a problem and thinking that you know exactly we’ll solve this.

[07:13] Brook’s approach to initiating concept generation

Q: So once you understand their problem and you get all the research and the amazing information, what are the steps after that with all the research that you have? Are there specific frameworks or workshops?

So first, persona for the system that I’m designing. Now that I know what audience I’m designing for and what they need, what persona for the system I’m designing would speak to that audience. …what are their needs? What are their potentially emotional needs or like systemic needs in that environment, and how can my persona address some of that stuff?

Initial scripting: doing some initial scripting. I think for every kind of conversational or voice project I’ve ever been on, just getting some initial scripting done really helps take it from thought to, uh, reality.And, and you can do a lot of initial testing once you have a little bit of scripting done, but again, kind of starting to think at a high level. Mm-hmm. Okay. These, these are my user needs. Here’s what they, uh, are expecting or could be benefited from, from the system, and how could I maybe design a couple of high level user flow maps or scripts that kind of address that need and, and start to really break it down and get into the nitty gritty of, of that problem.

[13:03] Brook’s perspective on conversation design tools

Q: Are there tools that you use now? Are there tools that you see now where this is built for ideation or script building?

I will say for designers that are working at companies, and you know, I, I swear this happens to every company I move to. Sometimes the most robust tools still are excel and flow charts. So if that’s what you’re using, don’t be intimidated. That’s. Where we’re at as an industry.

But I will say, there are a couple of tools that I’ve really been excited. Voice flow just launched another iteration of their tool and they have an excellent tool for ideating and iterating and building conversational experiences. And thing that they focus on in particular, which I think is really cool, is building something you haven’t talked about yet, context into the conversation….I think Botmock is another cool tool…you can just go in and test your idea.

[17:06] Designing with consideration for those impacted by your work

Q: Is there a specific approach in generating these ideas? Is there a specific method or something personal for you?

I roughly understand their problem areas and what I could do to kind of solve those problems, but I also really try to ask and make sure that I’m understanding what are the potential consequences of the tool that I’m building. If I make a voice assistant that’s replacing workers in some kind of environment, what does that mean? Or why? Do I feel like that’s the best solution? So kind of asking yourself additional questions once you’ve come up with solutions to determine why you’ve come to that conclusion. … Again, just incorporate people who are gonna be impacted by the things that you’re designing into that decision process.

[21:56] Incorporate input from representative users in research

Q: Do you find that these myths and misconceptions do come into these conversations? And if so, how do you tackle that to be able to get good research out of it? To be able to generate concepts?

For example, I think if you were putting together your dream Tesla car, and you were testing that out with a bunch of. People who really love Teslas, they would think this is the coolest voice app ever. And I will pay a lot of money for it and I’m so excited about it. But again, if you come up to Northern Michigan, it it where I am and, and ask that to potentially a group of farmers, I think they’d say different things.

They’d say, well, that’s not very important to me. So it’s, yeah, I think it’s really about, and not only asking the right questions, but making sure you’re thinking a little outside the box when you’re assembling the people you’re asking and, and kind of who’s gonna be impacted and who you’re trying to market to as well.

[22:54] Brook’s take on the ethical considerations of conversational AI

Q: So why don’t you tell me one thing that’s true that nobody agrees with you upon when it comes to conversation AI over the years?

I think more specifically, and this is a point of contention, although I, I think more and more people agree with this now, that having some ethical principles or best practices that we can adhere to as a burgeoning industry over the voice industry would be really helpful.

[26:18] Brooke’s perspective on the future of conversational AI

Q: What do you think is gonna come next? What’s the future of conversational AI?

So over the last couple of years, I think when the Alexas and the Google Homes all jumped on the scene and all of these other voice assistants, we saw a lot of skills-based interaction. So you’d walk up to your voice assistant and you’d say, Hey, Alexa, open X scale.

And a big problem for years, and I think which is still a problem, is discoverability for those skills. So often you have a voice assistant at home, and it’s kind of like our cell phones. You use the same six apps over and over again. How often do you actually go to the app store in peruse all the apps you can use?

And I, I’ve seen, uh, Amazon talk about this too. I think these companies are gonna kinda move toward a future where things are a little bit more intelligent or maybe suggest a skill or suggest some kind of interaction based on maybe the routine that you’re doing within your day or where you are if you’re talking to a voice assistant in your car. Kind of just surfacing those things more intelligently, like providing you a skill in the moment rather than needing to go invoke it.

About Voice This! Podcast

Conversations with the people who make conversational AI 🎙️Join Millani Jayasingkam and guests as they discuss voice technology, conversation design, AI trends, and the strategy of creating effective conversational experiences. Tune in for first-hand learnings, insights, anecdotes, and sometimes jokes! Say hi and send us your questions at: voicethispod@gmail.com.

Find us wherever you get your podcasts or Listen on Spotify!