•UX pitch on solutions
•Flow and wireframe for speech-enabled interactions across various contexts
•Implementation with engineers
•User testing/ iteration
Weekly debug
zenbo
About zenbo
A humanoid robot
Zenbo is an Android powered robot, it can control smart home devices, respond to voice queries, take pictures, act as a home security system,sing songs, make a video call, dance to music, and tell stories, play songs, roam around independently, even promising to work as a tutor to children. It is designed to interact with Children, Adults, and elders as well.
Examples of the features:
-
Zenbo Robot can be programmed to control your home appliances and can also work as a camera, it can also take videos and photos time to time so that you will never miss a moment
-
Zenbo Can tell children their favorite stories while displaying accompanying pictures
-
Sensors of the Robot can even monitor if a person falls — and call for help
Zenbo Design Guideline
Zenbo is a lovely family companion. He is also one platform of immense opportunities for all designers and developers. Before developing apps for Zenbo, there are several rules and tips that you should follow in order to create a smooth and pleasing interaction experience with Zenbo.
Emotions
There are 25+ expressions to tell users its status and emotions
For different contexts, Zenbo shows the corresponding expressions. Example: When a task is accomplished, Zenbo's facial expression is confident and proud. When in low battery, Zenbo's expression is sleepy.
There are about 60+ body languages
The body languages collaborate with the facial expression and its status..
Designing the dialogue
As Zenbo is given a characteristic, the dialogue and tone design should follow its character: helpful, lively, polite, warm-hearted, kind and positive.
Example:
When user asks something that Zenbo can't answer, Zenbo should:
Don’t: "I can't help you"
Do: "i can;t find the information, can I help tou search online?"
Key insights when designing VUI
Using speech recognition to understand spoken commands and answer questions.Google has developed an open source operating system called Android, which allows a user to perform voice commands such as: send text messages, listen to music, get directions, call businesses, call contacts, send email, view a map, go to websites, write a note, and search Google.
Trigger
Zenbo is in low power but still on
Intuitiveness
Empathy
Zenbo's reaction should be intuitive,emotional like human
Zenbo's every reaction can deepen the emotion convey. The dialogue design should let users feel empathetic, understood and affirmed.
VUI and action design
Interaction
Zenbo's feedback should be humanistic and induce user to interact more
"I'm hungry" (red light turn on)
*growling stomach*(red light flashing)
*I'm ...hungry...* (slow down the speech, red light flashing quickly)
the Challenge
Language, OS, dialogue emotions
Zenbo is only available in Mandarin (was selling in China, Taiwan area). When designing the speech flow, need to understand the contexts and popular ways of slang in the certain areas. The operating System is Android. Knowledge in Android development is required. When designing the voice enabled conversation, need to think of what are the triggered words and what are the answers can be found in the database(similar to chatbot).
Make of the team
Communicate with different PMs & engineers to make out different apps. Bridged co-work mode with the designers and engineers in Taiwan.
MY ROLE
Designed and developed speech-enabled interaction flows across various contexts.
I am the person in charge of these apps in Zenbo: Browser, weather, memo, recorder, calculator, timer. Also, I took part in business strategy meetings to brainstorm more business opportunies for Zenbo. The photos are the internal presentation to out sales and customer care people. I presented the function development to them and help our internal staff to understand our first Humanoid robot.
Ideate and pitch what we needed to refine.
Create the function structure to make the feature clear
Flow
Make flow to define the decision making and the various corresponding results feedback to users. Then sent for engineers to develop
Left: Zenbo default expression
Right: zenbo answers the results in voice, the interface will show the result as well
Presented the design journey and the features in Zenbo to our internal staff, customer care and sales
Google analytics is applied
Understand what are the most popular apps and find out the timings when users use them. This helps to get understand more on the target group's behavioural usage.
The most popular apps are: educational video, music and browser.
