Design 01: Text to Speech

I began my project wishing to interview administrators on campus. Administrators, I felt, deal with a plethora of tasks and would probably benefit from the added multitasking that a smartphone might allow them. However, most administrators did not use their smartphones for work purposes, and our conversations usually steered to more mundane tasks like checking email or texts: things they did in their every day life. And so I expanded my search to administrators and receptionists and the occasional wild card that I hoped would give me a completely different spin. They did not disappoint.

Interviews

note that all names are made up and almost all quotes are paraphrased.

Karl was my first interviewee. A self-described "recently graduated smart person," Karl works in Geo-spacial Information Systems.

"What do you normally use your smart-phone for?" I asked him.

"Nothing." Noting my quirked eyebrow he clarified: "I don't have a smart-phone."

I had prepared for this. We discussed his main use for his computer (printing our directions to places) and times when he wished he had a smart phone. Karl had recently been on a trip abroad and felt that something to look up street signs and directions would be helpful.

"So I was in Prague," Karl begins. "I was at the train station and I had no idea which line I was supposed to take. It would have been nice if I could have looked that up on the internet."

"Let's assume you have a smart-watch," I saw. "It's a screen about yay big and on your wrist. How would you like to interact with it to get the information you need?"

"Well first I'd go to the mobile site. Because of course there's a Praha mobile website development team."

"Of course."

"It's staffed entirely by dragons." I look up in surprise and confusion. "But they're midget dragons and thus the size of normal humans."

In the midst of our joined laughter, I took comfort in the fact that Karl is currently employed. Clearly my sense of humor would not handicap my post-graduation employment options.

I then asked Karl about possible GIS applications for a smart-watch. He said it was possible to use the screen to trace paths or select points, but at such a small scale it was bound to be inaccurate. I suggested perhaps a joining of the watch and the computer: a context-aware trackpad a la Duo, but Karl was not impressed. But I think that still might be an option worth exploring.  

SHANTI: The Haas Administrator

I interviewed Shanti next. As an administrator and coordinator at Haas, I expected that her job would involve a lot of tedious computer work that perhaps could be improved by a smart watch app. This did not prove to be the case.

"What did I want to do when I was looking for my phone? Probably to look something up on the internet or text a friend." She could not think of anything further.

"Now let's say you could text or Google from your smart phone interface. Can you walk through how you'd want that interaction to go?"

"I'd definitely want to be able to type into it..." she trailed off. "Yeah."

Shanti did not wish to be photographed.  

Mick is, without doubt, the coolest interviewee that I came across. He stood alone in his security booth, fingers littered with more rings than Prince Ali Fabulous-He Ali Ababwa had golden monkeys. He's missing about a third of his teeth.

"Whadduay use my smart phone for? Tumblr. I have a tumblr. You see some really cool stuff."

"Can you think of a time when you needed your smart phone but couldn't reach it? Either it was charging in another room or in your bag or--" Mick pats the leather smart-phone case that I now see is strapped across his chest above the blue uniform.

"It's always right here."

"What about a time when you didn't want to take your smart-phone out of the case?"

Mick's face lights up. "Pictures. Sometimes you see the coolest stuff around here. Have you seen the guy on the unicycle? There are also these skateboards with six wheels. I wish I could take a picture of that." As I pause to scribble this down he yells out another one. "One time! One time this woman parks over here and comes to me to get her parking permit. And she goes: 'I'll be back.' She sounded EXACTLY like Arnold Schwarzenegger. The accent and everything! I wish I could have recorded that."

I ask him how he would like to interact with the interface, but he had difficulty describing things. "I'm in no way computer literate," he noted. "So it would have to be easy to use. Also, it should auto focus. So I don't have to fiddle with things. I just wanna take a quick snap-shot."  

LEE and JANET: WYMCA Receptionists

Lee used her phone mainly for texting and social media. When was she unhappily without her phone? "When I'm working out. It would be nice if I could use it to play music or track my activity and monitor my heart rate."

"So let's say you have a smart watch. How would you want to interact with it when you're working out?"

"I'd want it to tell me what mile I was on, my heart rate. How many calories I burned, maybe, and my pace. Plus any notifications on my phone."

Once Lee had gone Janet noted that it would be great if smart watches could have "one of those zombie chase running apps."

Janet used her phone mainly to access her email and calendar, as well as to text friends and "Google". She usually missed her phone when it would be rude to use it: in class or during a meeting or while out with friends. "Sometimes you just want to know when you have a text." We discussed various options for non-intrusive text notification.

"I'd probably want it to be a small light or something. Something unintrusive."

"Do you want the text to just appear on the screen or do you want to click on something?"

"I'd rather click on something. But before I do that it should tell me who it's from. And give me a little snippet of the text: like email notifications do on your phone."  

MARTA: Avant Card Cashier

Marta already had a smart-watch, which made the interview a little more interesting. In addition to talking about what she currently uses her phone most for (facebook), I asked her what she currently used her smart-watch for.

"Notifications for messages and for snapchat. There's also a little pedometer on there, which is eh. I've also got bluetooth in this so I can press play on spotify here and play music in my car."

I then asked her what she wishes she could do with her smart-watch.

"I wish I could talk on it, so I could answer the phone. Sometimes my mom calls me at random times and panics when I don't pick up. I also wish it were waterproof. I'd also like to be able to do simple things like catch up with emails or go on the internet."

Marta's watch was a smaller, more notification-based device.  

FATHER AND CHILD: Wild Card

While walking back home from all my interviews, I caught site of an older gentleman playing with a small baby in some grass. "This man," I thought. "This man will have a different opinion." On approaching him, I was disappointed to find out that he was a student. But his perspective was indeed different, as I had hoped.

"It would be nice to look for places nearby," said the "Father" portion of "Father and Child." He gestures to the little girl. "Like maybe while walking here we would have found a park or something."

"How would you want to interact with this device?"

"Talk to it. You know, say 'Where's the nearest Indian restaurant?' And it'll tell me."

"Would you prefer turn by turn direction or--"

"No, no. Just a list. And then if I need to check back, I can. You look like an idiot just blindly staring at a screen while walking. It's better if you read, understand, and check back if necessary."

"What if it spoke directions to you?"

"That would be cool. It should be able to speak, yeah."

There was a bit of a pause as Child threw her "Three Little Kittens" book at me.

"You know what else would be cool?" Father noted. "Translation." He was in the South East Asian Studies department. "You know, translate words into different languages. Just casually as you're working. Or if you're traveling. That would be cool."

This hearkened back to Karl's suggestion, and combined with Mick's desire for a camera, an interesting product idea began to form.  

Things Learned

• Most people wanted to be able to speak to their watch and wanted to be able to hear their watch. For me this raised a lot of questions of how this could be done well.
• Only one person wanted to be able to type into her watch.
• Only one person thought of how they wanted to select things in the watch, but I think Karl's idea was very good.
• Context-awareness was fairly important. Navigation apps, suggestions apps, and translation apps were all discussed.

Brainstorming

Final App List

1. A hardware/software package system that allows you to take pictures with a smart-watch.

2. App that takes in a speech input and translates it.

3. App that uses a camera to scan text and translates it.

4. App that does text to speech conversion for the visually impaired.

5. In Cortana, you can tell it to remind you to do something once you are near a certain place. This is an app which either ties in with Cortana or you can add a list of chores via a mobile phone app or vocal input. Whenever you are near a place where that chore can be done, it will notify you. Possibility for a yelp tie in here as well.

6. Workout coach app that keeps track of exactly what your workout should be and walks you through each step of it. Possible addition of a workout mat so it can count repetition of push-ups, etc. based on movement of watch.

7. Gamify workouts with a zombie chase app. It keeps track of how far you've run while also pretending a zombie is chasing you. You get points for speeding up, dodging, etc.

8. Workout monitor: keeps track of heart rate, calories burned, etc.

9. A good system for making a receiving calls on a smart watch. Either a hardware system that makes it easy to speak into the phone (Can we create a bluetooth headset that fits into the phone somehow?)

10. discreet notification system

11. enjoyable goof-off site browsing: tumblr, reddit, etc.

12. possible geo-spacial software trackpad tie-in

13. sign-language to text conversion. Or sign-language to speech. Or speech to sign-language.

While idea #5 was one of my favorites, it didn't have enough user interface to be interesting. Much of the actual user input would have to be done on a smartphone. The watch would simply notify you when you were near a good location.

Instead, I went with option 13 as I was very much drawn to the idea of translation. Text to text translation was out, because there wasn't much room on the screen for translating an entire sentence, but text to speech could definitely be a thing. There are three main reasons why I went with this option:

1. An interesting design problem ("Where can you put a camera on a smartwatch?")
   
2. Helping the disabled: a group that designers historically forget about
   
3. The watch design is important here: if you were to use a smartphone, you would have to hold the camera up above the words. Here the fact that it's a watch is important to being able to use the app.
   

Design and Creation

There are three hardware design features to my phone. The first is the camera along the width of it. I sketched out many possible places for the camera to be and liked this the best. I tested to see how this would fare against books by holding my phone camera up against my wrist and placing my wrist on the book. It seemed to work reasonably well. A real prototype would require a camera on the end to get user feedback on the vision aspect. The second is the gear around the edge used to navigate, as per Karl's idea. I found that to be intuitive and it solved of the small-screen problem in terms of navigation and swiping. The fourth is a small Bluetooth headphone. The idea is either for it to fit into the side of the phone and it can be detached or to assume the user has one. I created one using a headphone and a piece of tape and curled it into the body of the watch.
In order to make a gear that moved, I used a paperclip to connect the gear to the watch face. The paperclip then wrapped around to create part of the band. I found this to be handy in my prototype because I was able to bend it to fit the wrist of the individual user.
In designing my app, I wanted the workflow to be as simple as possible. The user should touch the screen as little as possible. This led to a few problems down the line.
My proposed workflow was so:

1. app launch screen

   1. SOUND: “Text to Speech”

2. prompt user to align screen to words on page

   1. SOUND: “Align watch to word on screen.”

   2. SOUND: ding when word has been identified

3. Displays which language was detected with the option to change the language if the app is wrong.

   1. SOUND: “Reading in [Language]. Scroll to select different language. Tap to confirm language [Language].”

4. High-contrast text of word appears. User can scroll down to see livestream of word with higher contrast if analysis seems wrong.

   1. SOUND: “[word]”

Testing this ran me up against a large amount of interesting user scenarios. By assuming the user preferred to read as little as possible, I didn't add robust text instructions on how to use the app. Instead, since the user was guaranteed to be listening, I provided instructions where necessary verbally. I also provided a small down arrow in the event that the user could scroll down for more functionality. 90% of the time this worked, but 10% of the time, it left my user slightly confused. Particularly on the "language detected" screen where the down arrow seemed to indicate that they had to scroll, not that it was simply an option. In future iterations I would remove that arrow.

User Feedback

• the scroll gear was positively received

• the round shape was also positively received

• some felt that the positioning of the camera would be a little weird if you were translating a computer screen or a sign or something that was held up.

  • I agree that a phone app would work better for those scenarios, but I think for a book or a piece of paper on a desk, the wrist propping would be helpful and useful.

• that the wrist movement along a page wasn't natural

  • I agree with this. It would work better if the camera was actually a small ring on the person's finger. They can trace the lines and it'll capture the words. Hm...

• a GREAT suggestion was speech to speech translation instead. That's an idea I love and would like to run with later. It would also work as a increased functionality of this app.

• What happens when a visually impaired user reaches the end of a page and doesn't notice? Currently the app does nothing about that.

  • We considered a “you've reached blank space” or returning back to the “position camera to view text.” I would like to flesh out those user scenarios a little more.

• How does the user know that scrolling will take them to the live-feed version of the word? Will that live-feed version be helpful at all angles?

  • This is a great point. I don't know how to tell the user that scrolling is an option without giving them that information every time they open the app. I also don't know if there are angles that'll naturally occur but will be useless to a user when viewed on the screen. This would require testing with an actual camera. With my phone I didn't see such angles, but I could have missed something.