We are all experts at human-to-human conversation. But conversing only seems easy because it’s familiar, you’ve been doing it since you were born.
The key to building a good voice interface is to not fall into the trap of simply converting a GUI into a VUI.
In these slides we will cover the best practices to design our Action on Google (and any other Voice UI).
7. Persona is
conveyed
through:
● Tone
● Word and phrase choices
● Functional design
● Style
● Technique
● Voice
And based on:
● Your user population
● Their needs
● The imagery & qualities
associated with your brand
17. Sample dialogs
What to include:
1. Canonical “Happy path”
2. First time experience
3. Tapered experience (Return user)
4. Repair
Journey 5: Conversation repair
for timeouts
USER Ok Google, let me talk to Number Genie.
GOOGLE
ASSISTANT
Sure, Here's Number Genie.
NUMBER GENIE Welcome to Number Genie! I'm thinking of a
number from 0 to 100. What's your first guess?
USER (says nothing)
NUMBER GENIE I didn't hear a number.
USER (silent or muffled)
NUMBER GENIE If you're still there, what's your guess?
USER (silent)
NUMBER GENIE We can stop here. Let's play again soon.
18. Sample dialogs
What to include:
1. Canonical “Happy path”
2. First time experience
3. Tapered experience (Return user)
4. Repair
Journey 7: Quit game. User gives
up and ends the game
USER Hey Google, let me talk to Number Genie.
GOOGLE
ASSISTANT
Sure, Here's Number Genie.
NUMBER GENIE Welcome to Number Genie! I'm thinking of a
number from 0 to 100. What's your first guess?
USER 21
NUMBER GENIE It’s higher than 21. What’s your next guess?
USER I give up. [LOGIC: catch ending intent]
NUMBER GENIE Sure, I’ll tell you the number anyway. It was 90.
19. Give users credit
They know how to talk
Avoid force feeding commands.
(such as “to reply to a message,
say ‘reply’” or “to hear that again,
say ‘repeat’”.)
People hardly ever say nonsense.
They just weren’t parsed
correctly. There was a
misalignment in meaning.
They know what they want
There’s no such thing as a query with
no intent. The user wanted to do
something, even if they didn’t say so.
They want to help
Don’t create dead ends when
something doesn’t work.
If you ask for them to say something
again, they will, but rarely exactly the
same way.
Honor that they’re invested in
outcome of a repair attempt.
20. Confidential + Proprietary
● Error “events” ≠ users misbehaving
● Different types of error events deserve different strategies
● Users are cooperative; they initiate repair on their own
● Users have intent, whether spoken aloud or not
● It’s our job to figure out their intent, not just point out something went wrong
Things to remember
21. Use the context
Environmental
Where is the user?
What are they doing?
What type of device
are they using?
Situational
What’s their intent?
Where is the user’s
frame of mind?
Temporal
What happened
right before?
What’s happening next?
How is the experience
influenced over time?
Behavioral
What do users
know going in?
What data can you use
to enrich the experience
predictively or to cut out
unnecessary steps?
22. Example strategies to have ready
Rapid reprompt ● “What was that?”
● “Say that again?”
Reframe the
question
● “What time is this for?” → “Sorry, what time?”
● “For when?” → “What time would you like to book this for?”
Ask another way ● “If it helps, we can do this one piece a time.”
Be ready for
questions about the
question
● “I have your name and email from your account, so now all I need is
your phone number.”
● “You can give me the day, the time, or both.”
Be proactive ● “I could put you down for 6 for now, does that work?”
● “Do you want to finish this later?”
23. A VUI is inextricably linked with time
● When a user is navigating a GUI they can
skip ahead visually to the part they are
interested in, or dismiss as they go.
● With a VUI, the user can feel like a
“prisoner” of the VUI design. They must
listen with (or without) patience to each
word before they can hear the one that
follows it.
24. 1. Avoid written language and use spoken one
2. Kick off the conversation: introduce your own brand,
provide a list of high level actions available by Action.
3. Kick off the conversation: if given the context -
example: Talk to <agent_name> about <subject_one>
avoid menu listening.
VUI best practises
Source: www.google.com
25. 4. Guide user through the conversation
a. if you expect users to ask make sure to ask a clear
question.
b. Don’t leave mic open without a prompt
5. Keep our TTS short and clear
a. Analyse if your TTS can be split up in a logical way.
b. Ask first before giving long TTS descriptions
VUI best practises
Source: www.google.com
26. 1. Avoid data points unrelated to user query.
2. Follow natural turn taking in the conversation
a. If you ask a question leave the space for the user to answer it.
b. Avoid asking a question with a follow up sentence like: “Would you like to hear
that again? You can say ‘yes’, ‘no’ or ‘repeat’.
3. Use conversational makers
Adding the “glue” to the conversation and making it more engaging, examples:
a. Timelines - “first”, “halfway there”, “finally”
b. Acknowledgements - “thanks”, “got it”, “alright”, “sorry, about that”
c. Positive feedback - “good job”, “nice to hear that”
VUI best practises
Source: www.google.com
27. Recovering conversation
Step 1 Error cannot be eliminated
Step 2 Use broader explanation when re-prompting
Step 3 Don’t blame the user
Step 4 Don’t repeat prompts in error cases
29. Google Confidential and Proprietary
Generate TTS
Using Terminal
curl
"http://www.google.com/speech-api/v1
/synthesize?lang=en-us&text=actions+
on+google+rock" -o aog-rock-hack.mp3