Video: How to Make Your VUI Inclusive
Learn more about VUI design at the next SpeechTEK conference.
Read the complete transcript of this clip:
Diana Deibel: Let's talk about background, which is really the inclusivity part about this. So we're thinking about inclusive access. We're really thinking about the cultural components and the social components of speech and conversation. Word and slang choices are a big part of that. We've all thought about personas: "Who is this character that is speaking to my users? What do I want them to say? Why do I want them to say it?"
But it's important to understand that no matter what you do in terms of a persona, if you commit to a strong character--and I'm not saying that's a bad thing--but if you do it, you're gonna be inclusive for some people and exclusive for other people. Because those word choices that you make are going to be referential and inclusive and point people in for the people who know what you're talking about. But for the people who don't, it's gonna feel like, "Oh, okay I guess I don't know that part of the conversation, all right." And the more that you do that, the more it feels like this conversation is not for you.
So you can get away with it once or twice. You can get away with it all the time if you're doing the pregnant app and all you have are is pregnancy lingo, you could do that, but you have to know why you're making those choices and make sure that it's something that everybody in that group would know.
Names are another big one because I know a lot of us do introductions to bots where you say something like, "Hi, I'm pizza bot, what's your name?" You don't actually have the ability to parse everybody's name on the other side of that. That's immediately slamming the door in somebody's face on that conversation. And I have worked on a lot of different platforms and I will tell you that none of them is very inclusive because of machine learning. We've all trained these machines to recognize Anglo-American names. Those are really the ones that get recognized right away. But anytime you put in something that's not that, you start to run into trouble. And I know that I've had to do a lot of heavy lifting of training and I'm not going to get every single name. So then you start to have to ask yourself, "Do I need to build this from scratch? Is there more training that I can do? Who could I get to help me with this? Do I even need to have a name?"
Just know that if you're gonna set yourself up for having names in there, you'd better be able to recognize that, Because nothing is more exclusive than saying, I'm sorry, I didn't understand your name. That's just terrible.
Humor is another one. I love humor and I always want to include it and I try so hard. There is like nothing that's really safe other than dad jokes, unfortunately. Humor is so reliant on the context, on the relationship, and on references. If you don't have complete security and knowledge of who your users are and what their contexts are, if you haven't built that relationship with them, it's very hard to use humor effectively.
I'm not saying don't use it, I'm just saying be cautious and be thorough on it. And then there's cultural references, which is similar to words or slang choices. When you choose a reference point that your going to be like "Well, everybody knows..."
I had a slide of Jerry Seinfeld up there. Everybody knows Seinfeld, rigth? Untrue. I definitely showed my age around my colleagues at work when I showed that and there were like, "Who's that?"
Just be aware that any time that you think--"Well, this was such a cultural touchpoint, how could people not know this?"--it was a cultural touchpoint for you, for your group of people, the people that are around you. But as we've seen, time and again, as the years have gone by, that we all kind of live in bubbles and even though it might feel like that resonates with you and the people around you, it might not resonate for your user base.
So it's really important to be sure you're doing research around that to make sure that if you're going to use those, that those are things that actually resonate with the people who are going to be using your product. So again, include or avoid slang appropriately. Get outside your perspective. Go and do some research, be around people. Humor is not all-encompassing, unfortunately. Use cultural touchpoints that function for everybody.
Here's a nice little checklist. I ask myself these questions: Is this really accessible? It's probably okay if you're product is built for a specific user group, for their particular experience. Again, thinking about the pregnancy app. It's probably okay if your product categorizes users by access needs, in order to give them a more personalized experience. This means, if it asks the user, "Hey, do you have any visual needs that we should know about?"
Once it knows that about the user, they can put them in to the faster speed version of it. Great, you've just made them not have to choose that, or not have to ask for that, just by knowing something about them. Probably needs questioning if you've never even discussed accessibility. And it probably needs questioning if your reasons for excluding populations are based on business or resource rationale and not on user rationale. And I know that we've all run into this at some point or another, where somebody says, "We don't have time for that. We don't have money for that. That's just... We'll get to it in the next release."
I would really encourage everybody to push back on that, if you know that part of your user group is going to need something in terms of access, to say, "No, this really has to be part of that first release if we want to have something successful here." Because you only get one shot. Once people try it and it's not inclusive for them or they can't get access to it for whatever reason, they're not going to come back and try it again. People's patience is very thin. and their threshold for this is very low. So we need to make sure that we're really hitting that mark every single time.
Allstate Conversational Designer Katie Lower outlines working models for assessing the viability of a conversational interface with multiple teams within an organization in this clip from her presentation at SpeechTEK 2019.
Allstate Conversational Designer Katie Lower defines the customer journey map as a visualization of the customer's process and explains why it's valuable in this clip from her presentation at SpeechTEK 2019.
Grand Studio Lead Designer Diana Deibel discusses the ethical implications of speech UIs and remaining cognizant of the inherent human elements of speech and conversation in this clip from her presentation at SpeechTEK 2019.
Grand Studio Lead Designer Diana Deibel discusses multiple approaches to making VUI design transparent--the Google vs. Alexa, system-initiated vs. user-initiated--in this clip from her presentation at SpeechTEK 2019.
Pindrop Director of Product Marketing Ben Cunningham discusses best practices for voice authentication in IVR design in this clip from his panel at SpeechTEK 2019.
Much as developers rushed to put new apps onto Apple's App Store in 2009, we're seeing a bit of a gold rush to develop new skills for voice assistants. But developing for a VUI is an entirely different challenge.
Gridspace Co-Founder and Co-Head of Engineering Anthony Scodary demonstrates Grace, Gridspace's new automonous call center agent, in this clip from his keynote at SpeechTEK 2019.
AI allows companies to retrieve 100% of the audio from contact center calls without compromising quality and accuracy. With this knowledge, companies can improve CX, reduce effort and increase brand loyalty.
Orion Labs Head of Product Ellen Juhlin and Voicea CMO Cory Treffiletti discuss persisting challenges in speech-to-text, AI identifying intent, user expectations, and more in enterprise speech tech applications in this clip from their panel at SpeechTEK 2019.
451 Research Senior Analyst Raul Castanon discusses new findings of a recent survey on speech technology adoption in the enterprise and how adoption of devices in the consumer space have impacted enterprise adoption in this clip from his panel at SpeechTEK 2019.
Gridspace Co-Founder and Co-Head of Engineering Anthony Scodary discusses the transactional nature of speech and how that understanding impacts effective, AI-driven call center analytics in this clip from his keynote at SpeechTEK 2019.
Conversational Technologies Principal Deborah Dahl lays out a plan for making more virtual assistants more effective in this clip from her keynote at SpeechTEK 2019.
Conversational Technologies Principal Deborah Dahl discusses the state of the art for the three pillars of conversational systems in this clip from her keynote at SpeechTEK 2019.
Conversational Technologies Principal Deborah Dahl explains how more targeted enterprise knowledge could make VAs more effective in organizations in this clip from her keynote at SpeechTEK 2019.
Nuance Communications' Roanne Levitt delineates the differences between text-dependent and text-independent biometrics and what the advent of text-independent means for IVR applications in this clip from SpeechTEK 2019.
Companies and Suppliers Mentioned