+353-1-416-8900REST OF WORLD
+44-20-3973-8888REST OF WORLD
1-917-300-0470EAST COAST U.S
1-800-526-8630U.S. (TOLL FREE)


China Automotive Multimodal Interaction Development Research Report, 2021

  • ID: 5463681
  • Report
  • November 2021
  • Region: China
  • 110 Pages
  • Research In China


  • Audi
  • BYD
  • GAC
  • Honda
  • NETA
  • Toyota

Our China Automotive Multimodal Interaction Development Research Report, 2021 combs through and summarizes the interaction modes of mainstream auto brands.

The current automotive human-computer interaction (HMI) is characterized by the following:

1. In terms of interaction modes, in addition to conventional interaction methods such as touch control, button and knob, voice and mobile phone have become favored options in automotive HMI. Moreover, face, gesture, and car lamp have also become popular interaction options. Some brands even have begun to try new interaction ways, such as eye tracking, voiceprint, and fingerprint.

2. As concerns layout, joint-venture brands remain conservative in interaction modes. They seldom use new interaction methods except for voice and mobile phone. Yet luxury brands are pioneers of multimodal interaction. One example is Mercedes-Benz which has equipped its new-generation MBUX system with multiple new interactions like face, gesture, eye, and fingerprint. By comparison, Chinese leading brands are more active in new interaction modes, for example, Changan Auto has eyesight wake-up and Great Wall Motor offers voiceprint recognition.

Voice as a conventional interaction mode offers improvements in functions.

Besides the original capabilities of voice interaction, features like continuous dialogue, voice area lock, tone switch, and user-defined voice have become next orientations of mainstream brands.

1. Continuous dialogue and voice area lock have been used by several brands.

After a wake-up, a voice interaction system with the continuous dialogue capability needs no repeated wake-up within a certain period of time, which makes the response to instructions more coherent; the voice area lock feature with the ability to locate the voice command issuer, executes commands like “open window” and “adjust air conditioning temperature” more accurately. The brands using the two features include Geely, Changan, Great Wall, Chery, SAIC Roewe, GAC Trumpchi, Xpeng, NIO and Lixiang.

Geely Xingyue L - the “driver geek mode” of Galaxy OS frees the driver of wake-up in 24 hours, and enables lock of four voice areas. 

Xpeng Motors - the voice assistant Little P delivers voice functions in all scenarios and allows 30s continuous dialogue and dual-voice area lock.

2. Tone switch refers to the availability of other speech options the voice system provides for users in addition to default voices and sounds, such as, voices of celebrities and game characters.

The brands with this feature include Geely, Changan, Great Wall, BYD and Roewe, as well as emerging carmakers like NIO, Xpeng and Lixiang. For instance, Banma Zhixing Venus System mounted on SAIC Roewe RX5 offers five tone options: affinity female voice, intellectual female voice, magnetic male voice, innocent child voice, and passionate female voice.

3. User-defined voice feature allows customization of voice tones and voice functions. 

“Voice tone customization” means that the system applies the uploaded voices of the user’s own or his/her families’ in scenarios such as voice response and navigation broadcast. Models carrying this capability include Geely, Great Wall, SAIC Roewe, GAC Trumpchi and R Car.

“Voice function customization” means that users can define a voice command as well as corresponding operation, which brings the voice system into full play and meets the personalized needs of users. The typical brand using this capability is Xpeng Motors.

Geely Xingyue L - Galaxy OS allows users to define voice. The customized human voice is available to all scenarios including navigation and response.

Xpeng Motors - Little P Customization allows Xpeng car owners to define voice commands and corresponding operations via their smartphone APP.

Gesture has become the third new interaction mode behind voice and face.

Gesture interaction uses the in-car camera to judge a gesture and send the data to the vehicle system control unit which will then call out the function corresponding to the gesture. Gesture control is more accurate and quicker, and easier to operate than voice control. 

Brands using gesture interaction include BMW, Mercedes-Benz, Changan, Great Wall, SAIC Roewe, GAC Trumpchi, VOYAH, R Car, BAIC ARCFOX, and Xpeng, of which Great Wall and Xpeng support gesture recognition outside car.

WEY Mocha - enable gesture control over ignition, forward/backward movement, suspension, and flameout outside the car.

Changan Ford EVOS - the new-generation SYNC+ 2.0 allows gesture control over muting, music play, interface split, and back to the home interface, etc.

Light interaction is available to brands like Mercedes-Benz, HiPhi and IM Motors.

Light interaction sends a message out by way of flashing, light projection and combination headlight. Light interaction provides channels for communication with traffic participants outside the car, and enables autonomous vehicles to communicate with the surroundings at a time of implementing autonomous driving. Now brands including Mercedes-Benz, Audi, Volkswagen, Great Wall, BYD, HiPhi, IM and Xpeng have packed this interaction mode.

Mercedes-Benz - Projector headlights project warning signs, assisted markings and other information related to driving assistance systems onto the road ahead. 

IM L7 - Projection headlights enable the projection of interactive alerts, navigation, vehicle information among others, offering lane-level guidance with HD maps.; combination headlights provide several patterns to show the status of users in the cockpit; interactive taillights enable display of words and emoticons, as well as information, for example, thanks, pedestrian warning and state of charge.

As well as combination headlights, some brands like BAIC and Samsung are trying to add displays at the front and rear of the body to send messages to pedestrians. 

BAIC Lite 300 - LED displays at the front and rear of the body show words, emoticons, animations, state of charge and other content.

Multimodal fusion holds a future trend. 

Single-modal interaction has a limited role to play in improvement of interaction experience. Multimodal fusion is obviously a feasible and efficient option for better interaction experience in future. Brands that are developing this feature are Mercedes-Benz, BMW, Huawei and Changan.

1. Voice + eye sight

Voice interaction shows little ability to recognize fuzzy pronouns like “this” and “that”. In this case, the addition of eye tracking for locking direction, plus voice commands, can make the response to commands much quicker.

Mercedes-Benz MBUX Travel Knowledge - a driver who fixes his/her gaze on a building says “I want to know the purpose of this building”, and he/she will acquire related information. Cerence Look provides technical support.

2. Voice + lip movement recognition

Accurate acquisition of commands is a key step in the voice interaction process. Complex cockpit environment, noise and high personnel density make it harder to recognize voice. The new voice area lock feature added recently comes as a solution to the problem. Yet for its limitations, the single-model voice interaction technology has a limited effect in improving interaction experience.

The combination with lip movement recognition renders voice recognition far more accurate in voice interaction. Lip movements vary over language. In a noisy cockpit environment, combining lip movements, the system can ensure high voice recognition accuracy, even as the voice sounds low.


  • Audi
  • BYD
  • GAC
  • Honda
  • NETA
  • Toyota

1 Overview of Multimodal Interaction
1.1 Definition of Multimodal Interaction
1.2 Development History of Multimodal Interaction in Vehicle Scenarios
1.3 Application of Interaction Modes of Mainstream Brands
1.4 Touch Interaction
1.5 Voice Interaction
1.6 Development Trends of Voice Interaction
1.7 Face Interaction
1.8 Gesture Interaction
1.9 Light Interaction
1.10 Biometrics
1.11 Other Interaction Modes
1.12 Development Trends
2 Multimodal Interaction of Joint Venture Auto Brands
2.1 BMW
2.1.1 Development History of Interaction
2.1.2 Examples of Models with iDrive 7.0
2.1.3 Examples of Models with iDrive 8.0
2.1.4 BMW Digital Key
2.2 Mercedes-Benz
2.2.1 Development History of Interaction
2.2.2 Examples of Models with MY MBUX System
2.2.3 Mercedes-Benz EQS
2.2.4 Highlights of MBUX Hyperscreen
2.2.5 Interaction Highlights of Vision AVTR Concept Car
2.3 Audi
2.3.1 Development History of Interaction
2.3.2 Examples of Models with MMI System
2.4 Volkswagen
2.4.1 Development History of Interaction
2.4.2 Examples of Models with MOS 4.0
2.5 Ford
2.5.1 Development History of Interaction
2.5.2 Examples of Models with SYNC +
2.6 Buick
2.6.1 Development History of Interaction
2.6.2 Examples of Models with eConnect 3.0
2.7 Toyota
2.7.1 Development History of Interaction
2.7.2 Examples of Models with Toyota Connect
2.8 Nissan
2.8.1 Development History of Interaction
2.8.2 Examples of Models with Nissan Connect
2.9 Honda
2.9.1 Development History of Interaction
2.9.2 Examples of Models with Honda Connect
2.10 Volvo
2.10.1 Development History of Interaction
2.10.2 Examples of Models with the New Generation Intelligent In-vehicle Interaction System
2 Multimodal Interaction of Chinese Auto Brands
3.1 Geely
3.1.1 Development History of Interaction
3.1.2 Typical Model with Galaxy OS - Xingyue L
3.1.3 Interaction Highlights of Galaxy OS
3.1.4 Typical Model with GKUI 19 - Xingyue
3.1.5 Other Interaction Modes
3.2 Changan Auto
3.2.1 Development History of Interaction
3.2.2 Typical Models with Interactive Monitoring System (IMS)
3.2.3 Interaction Highlights of IMS
3.2.4 Other Interaction Modes
3.3 Great Wall Motor
3.3.1 Development History of Interaction
3.3.2 Typical Models with Fun-Life 2.0
3.3.3 Other Interaction Modes
3.4 BYD
3.4.1 Development History of Interaction
3.4.2 Typical Models with Dlink 4.0
3.5 Chery
3.5.1 Development History of Interaction
3.5.2 Typical Models with i-Connect@Lion 4.1
3.6 SAIC
3.6.1 Development History of Interaction
3.6.2 Typical Models with Banma Zhixing 4.0
3.7 GAC
3.7.1 Development History of Interaction
3.7.2 Typical Models with ADiGO 4.0
3.8 Interaction Highlights of Voyah FREE
3.9 Interaction Highlights of IM L7
3.10 Interaction Highlights of R Car
3.11.1 Interaction Highlights of BAIC ARCFOX
3.11.2 Typical Models with HarmonyOS
4 Multimodal Interaction of Emerging Auto Brands
4.1 NIO
4.1.1 Development History of Interaction
4.1.2 NIO OS: Mainly Button, Knob and Touch, Supplemented by Voice and Robot
4.1.3 NIO ET7
4.1.4 Smart Key Interaction
4.2 Xpeng Motors
4.2.1 Development History of Interaction
4.2.2 Interaction Highlights of Xmart OS
4.2.3 Typical Models with Xmart OS
4.2.4 Interaction Details
4.3 Lixiang
4.3.1 Development History of Interaction
4.3.2 Interaction Highlights of Lixiang ONE
4.3.3 Other Interaction Modes
4.4 Interaction Highlights of Weltmeister W6
4.5 Interaction Highlights of HiPhi X
4.6 Interaction Highlights of NETA U Pro
  • BMW
  • Mercedes-Benz
  • Audi
  • Volkswagen
  • Ford
  • Buick
  • Toyota
  • Nissan
  • Honda
  • Volvo
  • Geely
  • Changan Auto
  • Great Wall Motor
  • BYD
  • Chery
  • SAIC
  • GAC
  • Voyah
  • IM Motor
  • BAIC
  • NIO
  • Xpeng Motors
  • Lixiang
  • Weltmeister
  • HiPhi
  • NETA