In interaction design, the voice interaction design in the car is also the scene we want to understand the application. For example, how to deal with the input before and after input, and how to feedback to the user in the car. The author of this article has analyzed the design of in-vehicle voice interaction, let's take a look at it.
I have talked about HMI interaction design before, and today we will talk about the interaction design of voice, first of all, we will clarify a concept.
Everything without talking about the needs of the scene is pseudo-demand!
The voice interaction design in the car also needs to understand the application scenario!
For example, what we need to do before input, how to deal with it after input, and how to feedback to users in the car after processing and accepting (remember!). The owner of the car is not the only user ^_^).
1. Cases
Before input: Standby.
- Wake up: Input: HI, Seven Sauces – Output: Master, I'm here
- Command + Listen + Process + Feedback: Open the sunroof ——— listen to the emoticon animation - open the sunroof in the car
How is the weather today——— listen to the expression effect——- today's Chaoyang District is cloudy, the maximum temperature is 22 to 33 degrees:
2. Case summary
Therefore, the entire voice interaction is divided into two parts, the first part is the standby state before input; The second part is after the wake-up, there are four steps: wake-up - listening - analysis processing - feedback.
Generally, the two parts of listening and processing are not perceived by the user, but when we design the product, we must take into account the processing time, and take the processing time into account when designing the listening expression.
The above is our daily voice interaction scene, as an HMI designer also need to understand the voice interaction, I popularize the voice interaction under the cockpit scene!
1. Voice interaction
Definition: Voice interaction is short for VUI (Voice User Interface Voice User Interface), the most common voice system now, such as Apple's Siri, Tmall Genie, etc.
Second, the advantages and disadvantages of voice interaction?
1. Let's start with the pros
1) Efficiency! Efficiency! Or efficiency!
Say something important three times, and voice input is at least three times more efficient than touch input.
2) Voice interaction is not limited by space
Whether in the life scene or the car scene, voice interaction can be operated at a distance of 1 meter to 5 meters, in this scene, touch interaction seems to be incomparable.
In the retrieval scenario, voice is more efficient than manual input, especially in complex retrieval problems, because voice supports combined command input, for example, "small P, help me navigate the cheapest gas station recently", and the recent + cheapest + gas station three instructions combined together. Such a sentence consists of multiple levels of instructions in a combined way passed, and then combined to determine the execution of these instructions separately. In this scenario, the efficiency is much higher than that of touch input.
2. Let's talk about the disadvantages
1) Reception efficiency
In fact, the biggest problem with voice input is the reception and processing of information.
The first is the complexity of the environment, in a noisy environment, when various sounds are superimposed, it will have a great impact on voice interaction, but in the car scene, this situation will be less.
Especially in the car scene, some voice commands are issued to the car, if it is not in line with the voice interaction logic, there will usually be a second confirmation, or the instructions will not be processed, which and our mobile end, will have serious consequences and irreversible results to evoke the secondary pop-up window.
What's more, the first priority of the vehicle terminal is safety, so the point that must be considered in the design of voice interaction now is the fault tolerance mechanism.
For example, the fault-tolerant mechanism of "can't understand your instructions" will obviously affect the user's experience, so when the user cannot identify and process the instructions issued by the user, the result should be explained + input suggestions, for example, "Sorry did not hear your instructions, can you say it again" or "Master, do you want to play Guo Degang's crosstalk".
In general, it is necessary to have "everything has a response, everything has an account". This can be divided into two steps, one is to guide the user to rearrange the instructions, and the second is to give the direction of the instructions, do not let the user go
3. Extracurricular trivia: secondary pop-up window
To sum up the advantages of voice interaction is still greater than the disadvantages, the so-called low efficiency or the emergence of fault tolerance mechanism is still a small scene after all, and the interaction direction after that is still combined in a variety of interactive forms, in line with a variety of scenarios and mutual strengths.
Third, the voice interaction process
Having said all that voice interaction we bring into the car scene as an example.
1. Wake up
The driver enters the car and says "wake up word" to the center console to wake up the voice system.
2. Voice input
Enter the command: "Navigate to Daxing Airport"
In this instruction, several processes are: determining the end of speech – extracting valid information – recognizing the translated text – understanding semantics according to the algorithm.
3. Identify feedback
There are two cases here:
- Identification done: "Find Beijing Daxing International Airport for you, whether to navigate." ”
- Identify the error: "Sorry, didn't hear clearly, please say it again." ”
4. Execution Feedback
When performing the feedback operation, it is divided into three scenarios, confirming the instruction before execution, real-time feedback during execution, and hibernation after execution.
1) Before execution: Confirm the instruction
- The user replies "Start Navigation", explains that the feedback is correct, and starts navigating.
- The user replies to the "negative statement", indicating that the feedback is wrong, at this time, it is necessary to take the initiative to continue to ask the user "Master, where do you want to go?" ”
- The user replied "Other place names", explained the recognition error, and changed the destination again "For you to find XXX, located in XXXX, whether to start sending navigation".
If the above instructions are confirmed, start navigating with slot feedback.
2) Execution in progress
The introduction of voice interaction rules in execution accounts for the majority, which allows users to reduce the number of wrong instructions and reduce misoperation. Try to avoid interaction, if the speed of 65km/h, 3S interaction principle, is equivalent to blind opening 54m, the probability of accidents is 23 times the usual.
The specific voice commands are still combined with the navigation scene to do different designs.
3) Execution completed
There are two cases after the execution is completed, the first is the end of the task, and the second is that the user actively exits.
After the navigation is over, it automatically goes to sleep and waits for the next time it is woken up.
Above we listed the most basic navigation scenarios, voice interaction of course not only to meet the basic needs, users also expect to modify the destination and re-plan the route, and then the development direction is to actively introduce the surrounding locations to the user, so that the user can issue instructions, from passive to active.
Fourth, the voice emotional design
In Luo Zhenyu's New Year's Eve speech, he mentioned a sentence, "product value = functional value + emotional value + asset value".
In addition to fault tolerance, fast processing of user instructions and timely feedback in the process of voice interaction, it is a functional value. Emotional value can make the product bring short-term pleasant experience to users, improve the emotional impression of the product, and thus produce emotional experience, trust, and security.
1. Emotional classification of voice images
Voice image emotional action design, from the emotional is divided into negative and positive, from the state is divided into large actions and small actions.
Negative emotions have anxiety and apathy; Positive emotional atmosphere of excitement and satisfaction.
The movements are divided from no action to large movements to acquiescence, thinking, listening, and communication.
2. Emotional case presentation
(1) BMW iDrive smart cockpit
(2) Zebra system
(3) Niolai
(4) Ideal one
(5) Xiaopeng
(6) Huawei
(7) Baidu
(8) Concept
V. Case Summary
The design of the car voice image is generally divided into three types:
The first: anthropomorphism
Features: IPH, vivid, robot shape, facial features are obvious.
The second: flat graphics
Features: flat lines, simple graphics, monotonous colors, two-dimensional graphics.
(1) NIO-nomi
(2) Baidu - small degree
(3) Ideal - Ideal classmate
(4) Gaohe-hiphigo
The third kind: technology, cool, the future
Features: Irregular graphics, light-sensitive light effects and flowing, colorful three-dimensional graphics.
Zebra System - Zebra:
Sixth, the voice image design thinking dimension
The process of voice image design needs to be considered from both practical point of view and artistry. Specific refinement is divided into four dimensions: functionality, brand communication, affinity, and sense of technology.
(1) Functionality: Or to put it again, "product value = functional value + emotional value + asset value". The design of the voice image is first of all to meet the functional requirements.
(2) Brand communication: In the context of market competition, an emotional voice image is the best transmission of the brand, so differentiation + combination of brand tonality is particularly important.
(3) Affinity: I feel that the voice image in the car is more like a butler, affinity is the basic requirement of a butler, and it is also a good expression that can quickly bring users and avatars closer.
(4) Sense of technology: the voice image itself is an AI image, in the process of design, the sense of science and technology is the first to consider, in the cockpit environment, the interior chrome, wood decoration, atmosphere lights, are the means to improve the texture of the car, so in the voice image design, the sense of technology and quality is to be reflected.
Columnist
Hao Xiaoqi, WeChat public account: Seven sauce design notes, everyone is a product manager columnist. Dragonfly FM Senior UI Designer, 5 years of work experience. Focusing on experience design, we welcome all students to communicate together.
This article was originally published by everyone who is a product manager, and reproduction without the permission of the author is prohibited.
The title image is from Unsplash and is based on the CC0 protocol.
The views of this article only represent the author himself, everyone is a product manager platform only provides information storage space services.