SlideShare a Scribd company logo
AI/UX
Designing applications for
voice interface platforms
Philip Man, 2018.04.19
5
How are they able to have a conversation? How does she know what he
is saying and how does she produce speech to talk back? How did her
designer create that?
Currently a product manager at Starbucks Digital Ventures.
Before that product manager at smaller internet companies
and startups.
Occasional film maker, written and directed some short
films and documentaries.
Born and raised in the Netherlands.
Hello!
The voice interface is the biggest revolution for UX and UI designers
since the birth of the graphical user interface.
Voice is the next large platform
Speech is the most natural form
of interaction
Write command lines
Point with a mouse!
Pinch with your fingers!
Command with your voice!
The Voice User Interface
35 million users of voice-enabled
speakers in 2017
(+129% over 2016)
55 million households will have a
home assistant device by 2020.
Smart devices will be up to 175 million
Voice is on the rise
Building applications for voice
25,000 skills
40million devices
for Alexa
What’s out there
Cooking experience; guide through a recipe while having dirty hands
Order coffee ahead in the car; could even do pick up in drive through
Voice controlled home appliances; start laundry while still in bed
Whiskey tips and info; don’t have to put your glass down anymore
Order a ride; start queuing up while still doing your make up
Where is your brand present?
.COM
Thinking through what VX will mean to your brand
• Purpose and user stories
• Design and create
• Write a script
• Create a personality
• Identity intents
• New opportunities and new roles
A VX design primer
Product design for screens Product design for voice
From screen to speech… …the words become your UI
Speech
Subscribe me to the email news letter
Sign me up for the news letter
I want updates from this blog
I want to subscribe to this service
Design for utterances and intent
Screen
Click to subscribe
S U B S C R I B E
Machine learning is at the heart of voice applications
Automatic Speech Recognition
Transcribe speech into textASR
NLU
Natural Language Understanding
Extract meaning from the text
=
=“Black Car”
“Black Car”
Text to Speech
Transform text to natural speechTTS =“Audi A8”
Building the right product;
where does my brand need to be heard?
Purpose
Eyes and hands free
Home device
Personal
Proactive
Use case: what tasks are easier to complete? What things are more fun?
Context: where is the user? Create for when eyes and hands are occupied.
It’s not a personal device anymore, it’s a family and home device.
You’re an assistant: design a proactive experience instead of reactive
You’re a person now: learn about the customer from each conversation
Create the conversation flow
Write a script and possible flows. Consider them your conversation wireframes.
Just start writing in notepad!
Break down the speech
What is the user trying to do? Design for variations.
Utterances
Intent
Your app is a collection of intents
Cooking Skill Intents
o GetRecipe
o OrderIngredients
o ShareDinnerRequest
o StartRecipe
o SetAlarmClock
o PreheatOven
o PlayDinnerMusic
The output
goes in here
VI: Visual Identity
The meta information from the voice
Who’s talking?
Gender, age, accent, tone, pace, etc.
Personality:
What does this person answer?
Is the person optimistic or pessimistic?
Emotion:
In what state is this person during the
interaction? Happy? Disappointed? Surprised?
VI: Voice Identity
Be as demanding about words and sounds as you are about being pixel perfect.
Pushing words and sounds; what is on brand?
"Sorry, you don't have enough
money”
"Sorry, it seems like there's not
enough money”
Timeline markers Acknowledgements Pointers Transitions
“First”
”Then”
“Finally”
Help set expectations
about duration and
next steps
“Got it”
”Thanks”
”Okay”
Feedback to show it’s
understood and
confirmed
“This”
“That”
“Here is”
Help identify subjects
previously mentioned
and referenced
“Now”
“So”
“Next”
Help introduce
change when moving
to different topic
Help the user with speech navigation patterns
DON’T
“I can help you get a ride, tipping your driver, receipts for your last ride, or surge check.”
DO
“I can help you get a ride, tip your driver, check your surge pricing, or get a copy of a receipt.”
Clear UI; Use parallel language
• Whisper
• Break
• Emphasis
• Phonemic
• Prosody
and more…
This is your personal assistant, Google Home.
This is your personal assistant Google Home.
The HTML for speech
Speech synthesis markup language (SSML)
<speak> I already told you, <emphasis level=“strong”>the Tomorrow event is a
private event.</emphasis> Why don’t you believe me? </speak>
A framework to understand personalities (MBTI model)
Favorite world - Extraversion or Introversion: focus on inner or outer world?
Information - Sensing or Intuition: take in information or interpret and add meaning?
Decisions - Thinking or Feeling: make decisions by logic or people and feeling?
Structure - Judging or Perceiving: get things decided or stay open and optional?
Create a character; who is your brand voice?
Define its personality in the script
• How does this person respond
to critique?
• How does this person think
about gay marriage?
• How does this person process
new information?
• How does he or she make
decisions?
• How does he or she prefers life
to be structured?
And in different parts of the world
What is your voice identity?
And in different parts of the world
What is your voice identity?
A young boy
Or an uncle
New kind of teams
VX designer
Casting director
SSML front end developer
Sound engineer
AI product manager
Audio designer
Script writer
Linguist
Production designer
Semantics designer
Character developer
New kind of companies
New business models
Analytics company
Advertising
Gaming
Social media
Legal and trademarking
Security and accessibility
New design patterns
A new paradigm
Migrating your
application to voice
www.alpine.ai
Try it out yourself
Write your experience
www.getstoryline.com
Try it out yourself
Alexa simulator
www.echosim.io
Try it out yourself
Design with your eyes closed
Philip Man, philipmanx@gmail.com, www.phlp.mn

More Related Content

PDF
Personality For Voice by Ozlem Williams
PPTX
Principles of symbol design
PDF
Make Your Mobile Brand Iconic
PPTX
BabelBuds Marketing
KEY
M3 conf
PDF
CAVE Language Presentation for AI Camp
PDF
HealthCare corporate site - Design Evaluation by Mani hk
PPTX
Multi Channel Publishing
Personality For Voice by Ozlem Williams
Principles of symbol design
Make Your Mobile Brand Iconic
BabelBuds Marketing
M3 conf
CAVE Language Presentation for AI Camp
HealthCare corporate site - Design Evaluation by Mani hk
Multi Channel Publishing

Similar to Designing applications for voice interface platforms (20)

PDF
Introduction to Voice Design
 
PPTX
Great Voice Experiences Start with Listening: Best Practices in Research and ...
PPTX
Getting Started with Voice UI
PDF
Voice Tech TO #1
PDF
Stratis Valachis, Designing for Voice Interfaces
PPTX
Designing for Voice UI: Planning and Writing for Voice Interaction
PDF
Getting ready for voice
PDF
Designing for Voice
PDF
HAMBURG Voice MEETUP #4 LEARN voice user interface design!
PDF
Conversational user interfaces (by Jochem Grietens)
PDF
Making a More Human Interface with Voice UI & UX
PDF
Dagan "'Alexa, get me the articles': user experience and voice interfaces in ...
PPTX
Planning and Writing for Voice UI
PDF
Personality for Voice
PDF
Introduction to the Elements of Conversational Design with Lauren Golembiewski
PDF
Speakeasy 04 2017
PDF
Designing the Future, with Voice
PDF
Several bots are typing - Talk given at Nashville UX
PPTX
What do you understand by voice user interface design (VUI).pptx
PDF
[UX London 2018] The Future of Voice
Introduction to Voice Design
 
Great Voice Experiences Start with Listening: Best Practices in Research and ...
Getting Started with Voice UI
Voice Tech TO #1
Stratis Valachis, Designing for Voice Interfaces
Designing for Voice UI: Planning and Writing for Voice Interaction
Getting ready for voice
Designing for Voice
HAMBURG Voice MEETUP #4 LEARN voice user interface design!
Conversational user interfaces (by Jochem Grietens)
Making a More Human Interface with Voice UI & UX
Dagan "'Alexa, get me the articles': user experience and voice interfaces in ...
Planning and Writing for Voice UI
Personality for Voice
Introduction to the Elements of Conversational Design with Lauren Golembiewski
Speakeasy 04 2017
Designing the Future, with Voice
Several bots are typing - Talk given at Nashville UX
What do you understand by voice user interface design (VUI).pptx
[UX London 2018] The Future of Voice
Ad

Recently uploaded (20)

PDF
Quality Control Management for RMG, Level- 4, Certificate
PDF
UNIT 1 Introduction fnfbbfhfhfbdhdbdto Java.pptx.pdf
PDF
Facade & Landscape Lighting Techniques and Trends.pptx.pdf
PPTX
Special finishes, classification and types, explanation
PPTX
12. Community Pharmacy and How to organize it
PDF
Phone away, tabs closed: No multitasking
PPT
UNIT I- Yarn, types, explanation, process
PDF
GREEN BUILDING MATERIALS FOR SUISTAINABLE ARCHITECTURE AND BUILDING STUDY
PPTX
Wisp Textiles: Where Comfort Meets Everyday Style
PPTX
YV PROFILE PROJECTS PROFILE PRES. DESIGN
PPTX
mahatma gandhi bus terminal in india Case Study.pptx
PDF
High-frequency high-voltage transformer outline drawing
PDF
Urban Design Final Project-Context
PPT
unit 1 ppt.ppthhhhhhhhhhhhhhhhhhhhhhhhhh
PPTX
Tenders & Contracts Works _ Services Afzal.pptx
PDF
Key Trends in Website Development 2025 | B3AITS - Bow & 3 Arrows IT Solutions
PPTX
Fundamental Principles of Visual Graphic Design.pptx
PPTX
Complete Guide to Microsoft PowerPoint 2019 – Features, Tools, and Tips"
PPTX
AC-Unit1.pptx CRYPTOGRAPHIC NNNNFOR ALL
PPTX
ANATOMY OF ANTERIOR CHAMBER ANGLE AND GONIOSCOPY.pptx
Quality Control Management for RMG, Level- 4, Certificate
UNIT 1 Introduction fnfbbfhfhfbdhdbdto Java.pptx.pdf
Facade & Landscape Lighting Techniques and Trends.pptx.pdf
Special finishes, classification and types, explanation
12. Community Pharmacy and How to organize it
Phone away, tabs closed: No multitasking
UNIT I- Yarn, types, explanation, process
GREEN BUILDING MATERIALS FOR SUISTAINABLE ARCHITECTURE AND BUILDING STUDY
Wisp Textiles: Where Comfort Meets Everyday Style
YV PROFILE PROJECTS PROFILE PRES. DESIGN
mahatma gandhi bus terminal in india Case Study.pptx
High-frequency high-voltage transformer outline drawing
Urban Design Final Project-Context
unit 1 ppt.ppthhhhhhhhhhhhhhhhhhhhhhhhhh
Tenders & Contracts Works _ Services Afzal.pptx
Key Trends in Website Development 2025 | B3AITS - Bow & 3 Arrows IT Solutions
Fundamental Principles of Visual Graphic Design.pptx
Complete Guide to Microsoft PowerPoint 2019 – Features, Tools, and Tips"
AC-Unit1.pptx CRYPTOGRAPHIC NNNNFOR ALL
ANATOMY OF ANTERIOR CHAMBER ANGLE AND GONIOSCOPY.pptx
Ad

Designing applications for voice interface platforms

  • 1. AI/UX Designing applications for voice interface platforms Philip Man, 2018.04.19
  • 2. 5
  • 3. How are they able to have a conversation? How does she know what he is saying and how does she produce speech to talk back? How did her designer create that?
  • 4. Currently a product manager at Starbucks Digital Ventures. Before that product manager at smaller internet companies and startups. Occasional film maker, written and directed some short films and documentaries. Born and raised in the Netherlands. Hello!
  • 5. The voice interface is the biggest revolution for UX and UI designers since the birth of the graphical user interface. Voice is the next large platform
  • 6. Speech is the most natural form of interaction Write command lines Point with a mouse! Pinch with your fingers! Command with your voice! The Voice User Interface
  • 7. 35 million users of voice-enabled speakers in 2017 (+129% over 2016) 55 million households will have a home assistant device by 2020. Smart devices will be up to 175 million Voice is on the rise
  • 8. Building applications for voice 25,000 skills 40million devices for Alexa
  • 9. What’s out there Cooking experience; guide through a recipe while having dirty hands Order coffee ahead in the car; could even do pick up in drive through Voice controlled home appliances; start laundry while still in bed Whiskey tips and info; don’t have to put your glass down anymore Order a ride; start queuing up while still doing your make up
  • 10. Where is your brand present? .COM
  • 11. Thinking through what VX will mean to your brand • Purpose and user stories • Design and create • Write a script • Create a personality • Identity intents • New opportunities and new roles A VX design primer
  • 12. Product design for screens Product design for voice From screen to speech… …the words become your UI
  • 13. Speech Subscribe me to the email news letter Sign me up for the news letter I want updates from this blog I want to subscribe to this service Design for utterances and intent Screen Click to subscribe S U B S C R I B E
  • 14. Machine learning is at the heart of voice applications Automatic Speech Recognition Transcribe speech into textASR NLU Natural Language Understanding Extract meaning from the text = =“Black Car” “Black Car” Text to Speech Transform text to natural speechTTS =“Audi A8”
  • 15. Building the right product; where does my brand need to be heard? Purpose Eyes and hands free Home device Personal Proactive Use case: what tasks are easier to complete? What things are more fun? Context: where is the user? Create for when eyes and hands are occupied. It’s not a personal device anymore, it’s a family and home device. You’re an assistant: design a proactive experience instead of reactive You’re a person now: learn about the customer from each conversation
  • 16. Create the conversation flow Write a script and possible flows. Consider them your conversation wireframes. Just start writing in notepad!
  • 17. Break down the speech What is the user trying to do? Design for variations. Utterances Intent
  • 18. Your app is a collection of intents Cooking Skill Intents o GetRecipe o OrderIngredients o ShareDinnerRequest o StartRecipe o SetAlarmClock o PreheatOven o PlayDinnerMusic
  • 21. The meta information from the voice Who’s talking? Gender, age, accent, tone, pace, etc. Personality: What does this person answer? Is the person optimistic or pessimistic? Emotion: In what state is this person during the interaction? Happy? Disappointed? Surprised? VI: Voice Identity
  • 22. Be as demanding about words and sounds as you are about being pixel perfect. Pushing words and sounds; what is on brand? "Sorry, you don't have enough money” "Sorry, it seems like there's not enough money”
  • 23. Timeline markers Acknowledgements Pointers Transitions “First” ”Then” “Finally” Help set expectations about duration and next steps “Got it” ”Thanks” ”Okay” Feedback to show it’s understood and confirmed “This” “That” “Here is” Help identify subjects previously mentioned and referenced “Now” “So” “Next” Help introduce change when moving to different topic Help the user with speech navigation patterns
  • 24. DON’T “I can help you get a ride, tipping your driver, receipts for your last ride, or surge check.” DO “I can help you get a ride, tip your driver, check your surge pricing, or get a copy of a receipt.” Clear UI; Use parallel language
  • 25. • Whisper • Break • Emphasis • Phonemic • Prosody and more… This is your personal assistant, Google Home. This is your personal assistant Google Home. The HTML for speech Speech synthesis markup language (SSML) <speak> I already told you, <emphasis level=“strong”>the Tomorrow event is a private event.</emphasis> Why don’t you believe me? </speak>
  • 26. A framework to understand personalities (MBTI model) Favorite world - Extraversion or Introversion: focus on inner or outer world? Information - Sensing or Intuition: take in information or interpret and add meaning? Decisions - Thinking or Feeling: make decisions by logic or people and feeling? Structure - Judging or Perceiving: get things decided or stay open and optional? Create a character; who is your brand voice?
  • 27. Define its personality in the script • How does this person respond to critique? • How does this person think about gay marriage? • How does this person process new information? • How does he or she make decisions? • How does he or she prefers life to be structured?
  • 28. And in different parts of the world What is your voice identity?
  • 29. And in different parts of the world What is your voice identity? A young boy Or an uncle
  • 30. New kind of teams VX designer Casting director SSML front end developer Sound engineer AI product manager Audio designer Script writer Linguist Production designer Semantics designer Character developer New kind of companies New business models Analytics company Advertising Gaming Social media Legal and trademarking Security and accessibility New design patterns A new paradigm
  • 31. Migrating your application to voice www.alpine.ai Try it out yourself
  • 34. Design with your eyes closed Philip Man, philipmanx@gmail.com, www.phlp.mn