Method and system for recognizing and relieving road rage based on personalized intelligent voice interaction

Abstract

The invention provides a method and a system for recognizing and relieving road rage based on personalized intelligent voice interaction, and relates to the technical field of driving safety. Comprising the following steps: acquiring road condition information, and detecting whether a driving scene causing road rage exists in the road condition information; real-time voice and language feature information of the driver are obtained, and whether the driver is in the negative emotion of road rage or not is recognized; acquiring voice feature information of a driver in a daily driving process, and determining personality features of the driver; the method comprises the following steps: matching personalized sound according to personality characteristics of a driver, matching voice materials according to road condition information, generating personalized voice interaction content based on the personalized sound and the voice materials, and relieving road rage negative emotion of the driver. According to the method, the road condition information and the personality characteristics of the driver are combined, the personalized sound and voice materials are matched, the personalized voice interaction content is generated, and a better and more targeted relieving effect can be achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of automobile intelligent driving and driving safety, and in particular relates to a road rage identification and relief method and system based on personalized intelligent voice interaction. Background Art

[0002] Family cars have entered thousands of households and become the basic means of transportation for people. In modern society, people are under increasing pressure in life, work and other aspects. Emotions that have nowhere to vent can easily erupt at a certain moment. In particular, drivers are easily irritated while driving and thus behave irrationally. "Road rage" is one of them. "Road rage" refers to the aggressive or angry behavior of motor vehicle drivers, which is mainly manifested in the use of insulting words or gestures, deliberately driving in a way that threatens safety, and carrying out behaviors that threaten the lives and health of others.

[0003] Road traffic safety is an important component of social public safety. With the increase in the number of vehicles on the road and the increasingly complex road traffic conditions, the impact of "road rage" on road traffic safety has become increasingly prominent. Therefore, it is particularly important to detect and intervene in drivers' road rage.

[0004] In the prior art, there is a Chinese invention patent application with publication number CN 117549902 A and titled A driver road rage detection device and detection method thereof. The device obtains the driver's grip pressure and steering wheel slapping pressure status information through a pressure detection module, obtains the angular velocity status information of the steering wheel through a direction detection module, obtains the driver's heart rate status information through a heart rate detection module, and obtains the driver's sound decibel status information through a sound pressure level detection module; the processing module processes and analyzes the status information obtained by the pressure detection module, the direction detection module, the heart rate detection module and the sound pressure level detection module, and integrates and determines whether the driver is in a road rage state; the alarm module receives the information from the processing module and sends corresponding warning signals, including voice signals, seat shaking and forced parking.

[0005] Although the above technical solution can detect road rage situations to a certain extent, it does not take into account the potential factors that may cause road rage in road condition information during detection, and the warning signal does not take into account the driver's personality. Simple voice signals, seat shaking and forced parking may aggravate the manifestations of road rage and further cause negative stimulation to the driver. Summary of the invention

[0006] In order to overcome the deficiencies of the above-mentioned prior art, the present invention provides a road rage identification and relief method and system based on personalized intelligent voice interaction, which combines road condition information to identify whether there are driving scenarios that lead to road rage, and combines the driver's personalized factors, taking into account the driver's personality characteristics, matching personalized sounds and voice materials, and generating personalized voice interaction content, so as to alleviate the driver's negative emotions of road rage, and can achieve better and more targeted relief effects.

[0007] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0008] A first aspect of the present invention provides a road rage identification and mitigation method based on personalized intelligent voice interaction.

[0009] The road rage recognition and mitigation method based on personalized intelligent voice interaction includes the following steps:

[0010] Obtaining road condition information and detecting whether there is a driving scenario that may cause road rage in the road condition information;

[0011] When there is a driving scenario that may cause road rage in the road condition information, the real-time voice and language feature information of the driver is obtained to identify whether the driver is in a negative emotion of road rage;

[0012] Obtain the driver's voice feature information during daily driving and determine the driver's personality characteristics;

[0013] Personalized voice is matched according to the driver's personality characteristics, and voice materials are matched according to road conditions information. When the driver is under the influence of negative emotions, personalized voice interaction content is generated based on the personalized voice and voice materials to alleviate the driver's negative emotions of road rage.

[0014] A second aspect of the present invention provides a road rage identification and mitigation system based on personalized intelligent voice interaction.

[0015] Road rage recognition and mitigation system based on personalized intelligent voice interaction, including:

[0016] The road rage factor detection module is configured to: obtain road condition information, and detect whether there is a driving scenario that causes road rage in the road condition information;

[0017] The road rage emotion recognition module is configured to: when there is a driving scene that causes road rage in the road condition information, obtain the driver's real-time voice and language feature information to identify whether the driver is in a negative emotion of road rage;

[0018] The personality characteristic determination module is configured to: obtain voice characteristic information of the driver during daily driving to determine the personality characteristics of the driver;

[0019] The personalized voice interaction module is configured to: match personalized voice according to the driver's personality characteristics, match voice materials according to road condition information, and when the driver is under the influence of negative emotions, generate personalized voice interaction content based on the personalized voice and voice materials to alleviate the driver's negative emotions of road rage.

[0020] A third aspect of the present invention provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps in the method for identifying and alleviating road rage based on personalized intelligent voice interaction as described in the first aspect of the present invention.

[0021] The fourth aspect of the present invention provides an electronic device, including a memory, a processor, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for identifying and alleviating road rage based on personalized intelligent voice interaction as described in the first aspect of the present invention are implemented.

[0022] One or more of the above technical solutions have the following beneficial effects:

[0023] The present invention provides a method and system for identifying and alleviating road rage based on personalized intelligent voice interaction. When detecting the road rage emotion of the driver, the road condition information is combined to identify whether there is a driving scene that causes road rage. When there is a driving scene that causes road rage, the method further identifies whether the driver is in a negative emotion of road rage based on the driver's real-time voice and language feature information. The above technical means can identify the road rage emotion in a more targeted manner.

[0024] When soothing and alleviating the driver's road rage, the driver's personalized factors are combined, the driver's personality characteristics are taken into account, personalized sound and voice materials are matched, and personalized voice interaction content is generated to alleviate the driver's negative emotions of road rage, which can achieve better and more targeted relief effects.

[0025] The method of the present invention combines scenarios that are likely to cause negative driving, real-time fusion perception of the driver, and long-term fusion perception to determine the driver's road rage risk and effectively alleviate it, thereby reducing the probability of road rage occurring, and enhancing and improving the driving experience while ensuring the driver's safe driving.

[0026] The present invention directly obtains surrounding road condition information through the autonomous driving sensor ADAS system, which is more accurate and real-time compared with other methods of obtaining road information using the GPS system; the voice system can provide different road rage emotion relief feedback based on different personality traits.

[0027] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0029] Figure 1 This is an overall flow chart of Embodiment 1 of the present invention.

[0030] Figure 2 This is a flow chart of obtaining surrounding road condition information through ADAS in Example 1 of the present invention.

[0031] Figure 3 This is a flow chart of determining the current emotion of the driver in accordance with the first embodiment of the present invention.

[0032] Figure 4 This is a flow chart of determining the personality characteristics of a driver in Embodiment 1 of the present invention. DETAILED DESCRIPTION

[0033] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0034] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0035] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0036] Embodiment 1

[0037] In order to solve the driving safety problems caused by the driver's road rage, take into account the comprehensiveness and accuracy of road rage recognition, and be able to use personalized voice and language according to the driver's personality characteristics to effectively soothe the negative emotions of road rage and avoid further adverse stimulation to the driver, the embodiment of the present invention designs a method for alleviating road rage based on intelligent voice interaction and ADAS information. The method combines scenes that are likely to cause negative driving, real-time fusion perception of the driver, and long-term fusion perception to judge the driver's road rage risk, and effectively alleviate it, thereby effectively reducing the probability of road rage.

[0038] In general, this embodiment first obtains surrounding road condition information through ADAS; then determines the driver's current emotion based on intelligent fusion perception; determines the driver's personality characteristics based on the results of long-term intelligent fusion perception; and determines a personalized driver emotion relief plan based on the current road condition and driver's personality traits obtained by the ADAS system. Thus, the probability of road rage is reduced in a targeted manner.

[0039] The technical solution of this embodiment will be further explained below in conjunction with the accompanying drawings.

[0040] like Figure 1 As shown, the road rage recognition and mitigation method based on personalized intelligent voice interaction includes the following steps:

[0041] Obtaining road condition information and detecting whether there is a driving scenario that may cause road rage in the road condition information;

[0042] When there is a driving scenario that may cause road rage in the road condition information, the real-time voice and language feature information of the driver is obtained to identify whether the driver is in a negative emotion of road rage;

[0043] Obtain the driver's voice feature information during daily driving and determine the driver's personality characteristics;

[0044] Personalized voice is matched according to the driver's personality characteristics, and voice materials are matched according to road conditions information. When the driver is under the influence of negative emotions, personalized voice interaction content is generated based on the personalized voice and voice materials to alleviate the driver's negative emotions of road rage.

[0045] In some embodiments, obtaining road condition information and detecting whether there is a driving scenario that causes road rage in the road condition information specifically includes:

[0046] Obtaining road condition information through the ADAS system, including traffic congestion, speed of surrounding vehicles, whether other vehicles have cut in, distance to surrounding vehicles, and whether surrounding vehicles frequently change lanes;

[0047] It is detected whether there is a driving scenario that causes road rage in the road condition information, where the driving scenario that causes road rage includes being cut in by another vehicle, the preceding vehicle suddenly braking, and the traffic jam time exceeding a set time.

[0048] In some embodiments, obtaining road condition information through the ADAS system includes collecting information around the vehicle through a variety of sensors and predicting potential dangerous conditions, wherein the information around the vehicle is collected through a variety of sensors, specifically including:

[0049] Capture visual information around the vehicle through cameras, including road signs, lane lines, pedestrians, and obstacles;

[0050] Detect the distance, speed and azimuth of the target object through millimeter wave radar;

[0051] Obtain depth information of the surrounding environment through LiDAR;

[0052] Close range detection via ultrasonic radar;

[0053] Predicting potentially dangerous situations, including:

[0054] Detect obstacles around the vehicle and track them based on the information collected by the sensors;

[0055] Through image processing technology and algorithms, the system can identify the vehicle's current driving status and lane position, and give a warning when the vehicle deviates from the lane;

[0056] Detect obstacles in front of the vehicle and predict collision risks to issue early warnings;

[0057] The vehicle in front of the vehicle is detected by sensors and the speed is automatically adjusted to maintain a safe distance from the vehicle in front.

[0058] In some embodiments, obtaining the real-time voice and language feature information of the driver to identify whether the driver is in a negative emotion of road rage specifically includes:

[0059] Obtain the driver's voice feature information, including volume, speaking speed, pitch, intensity, formant, and tone; obtain the driver's language feature information, including swear words and uncivilized language; obtain other information related to the driver, including interjections, sighs, sounds of slapping objects, and sounds played by mobile phones;

[0060] While considering the driver's personality and age voice characteristics, the above-mentioned voice feature information, language feature information and other driver-related information are comprehensively considered and matched with the corresponding feature information of typical anger / irritability / tension to determine whether the driver is already in the negative emotion of road rage.

[0061] It can be understood that in order to facilitate matching, it is necessary to pre-set an information database of anger / irritability / tension corresponding features, and comprehensively consider the similarity between the above-mentioned voice feature information, language feature information and other driver-related information and the corresponding features in the information database. When the similarity is greater than the set threshold, it can be determined that the current driver is in the negative emotion of road rage.

[0062] In some embodiments, obtaining voice feature information of the driver during daily driving to determine the personality characteristics of the driver specifically includes:

[0063] Acquiring speech feature information of the driver during daily driving, inputting the speech feature information of the driver during daily driving into a personality trait model, and obtaining personality trait prediction results, wherein the personality trait prediction results include five personality traits: openness, responsibility, extroversion, agreeableness, and neuroticism;

[0064] The personality trait model is used to determine personality trait information based on speech feature information.

[0065] In some embodiments, matching personalized voice according to the personality characteristics of the driver and matching voice material according to road condition information specifically include:

[0066] Consider the personality characteristics of the driver and select a voice that matches the driver's personality and emotional preferences from the existing voice library as a matching personalized voice. The personalized voice needs to be polite and friendly and consistent with the vehicle's product image.

[0067] According to the road condition information, voice materials that match the current road condition information are selected from the existing voice material library as matching voice materials. The matching voice materials contain both positive and negative content and use environmental description as the subject, wherein the positive content is praise-like and the negative content is prompt-like, warning-like and imperative-like.

[0068] In some embodiments, personalized voice interaction content is generated based on personalized sound and voice material to alleviate the negative emotions of road rage for the driver, specifically including:

[0069] Considering the specific classification of the driver's five personality traits of openness, responsibility, extroversion, agreeableness and neuroticism, personalized voice interaction content is generated based on personalized sound and voice materials, including:

[0070] For drivers with open personality traits, use more direct language to remind the driver;

[0071] For drivers with responsible personality traits, from the perspective of driving safety, use encouraging language to remind the driver;

[0072] For drivers with extraversion personality traits, the language used to remind the driver is more direct or clear than that for drivers with openness personality traits.

[0073] More specifically, Figure 1-Figure 4 As shown, this embodiment may include the following steps:

[0074] Step S1:

[0075] The ADAS system is used to obtain road and road condition information, mainly to detect potential factors that may lead to road rage in the scene. These potential factors include traffic congestion, the speed of surrounding vehicles, whether there is any other car cutting in, the distance to surrounding vehicles, whether surrounding vehicles frequently change lanes, etc.

[0076] The ADAS system, which stands for Advanced Driver Assistance System, is a vehicle-mounted system that obtains environmental data through sensors such as cameras and radars, and then processes and analyzes the data through hardware and software systems to identify and judge traffic and driving conditions, thereby providing information and driving assistance to the driver. When potential danger is predicted, the driver will be warned and even active intervention measures such as emergency braking will be taken.

[0077] In this embodiment, the road condition information mainly refers to driving scenes that are likely to cause negative emotions in people, which are generally identified by sensors of the autonomous driving system. The road condition information includes but is not limited to:

[0078] 1) Being cut in line by another car;

[0079] 2) The car in front braked suddenly and almost rear-ended me;

[0080] 3) Long traffic jams.

[0081] Furthermore, the specific steps of using ADAS to obtain road and road condition information mainly include three stages: environmental perception, calculation analysis, and control execution. Figure 2 shown.

[0082] (1) Environmental Perception

[0083] The main function of environmental perception is to perceive the environment around the vehicle. At this stage, different types of vehicle sensors are used to collect information around the vehicle, such as traffic congestion, the speed of surrounding vehicles, whether there are other vehicles cutting in, the distance to surrounding vehicles, whether surrounding vehicles frequently change lanes, etc.

[0084] These sensors include: cameras, millimeter-wave radars, lidars, and ultrasonic radars, etc.

[0085] Further:

[0086] The camera captures visual information around the vehicle, including road signs, lane lines, pedestrians, obstacles, etc., and converts this information into digital signals for system processing;

[0087] Millimeter-wave radar detects the distance, speed, and azimuth of target objects by emitting and receiving millimeter-level electromagnetic waves;

[0088] LiDAR obtains depth information of the surrounding environment by emitting laser beams and measuring the reflection time;

[0089] Ultrasonic radar is mainly used for close-range detection, such as measuring the distance between a vehicle and an obstacle in a parking assistance system.

[0090] During the process of environmental perception, the data collected by the sensors will be transmitted to the electronic control unit or domain controller for processing.

[0091] (2) Operational analysis

[0092] The main function of computational analysis is to analyze the information obtained from environmental perception to predict potential dangerous situations. At this stage, the electronic control unit or domain controller will process and analyze the data collected by the sensor, and use various algorithms and models to identify obstacles around the vehicle, determine the vehicle's driving status and road conditions. Based on this information, the system will make corresponding decisions and issue instructions.

[0093] The operational analysis phase includes:

[0094] Obstacle Detection and Tracking: Based on the information collected by the sensors, the system can detect obstacles around the vehicle and track them.

[0095] Lane departure detection and warning: Through image processing technology and algorithms, the system can identify the vehicle's current driving status and lane position, and give a warning when the vehicle deviates from the lane.

[0096] Forward Collision Detection and Warning: Through sensors such as radar or cameras, the system can detect obstacles in front of the vehicle and predict the risk of collision to issue a warning in advance.

[0097] Adaptive cruise control: The system can detect the vehicle in front of the vehicle through sensors and automatically adjust the vehicle speed to maintain a safe distance from the vehicle in front.

[0098] (3) Control execution

[0099] The main function of control execution is to control the vehicle through the actuator according to the results of the operation analysis. The actuator can control the acceleration, braking, steering, etc. of the vehicle to ensure the safe driving of the vehicle. At this stage, the electronic control unit or domain controller will send corresponding control instructions to the actuator to realize various control functions.

[0100] The control execution phase includes:

[0101] Automatic parking: The system can automatically control the vehicle's steering, acceleration, braking, etc. to achieve automatic parking function.

[0102] Automatic braking: When sensors detect obstacles ahead, the system can control the vehicle to brake automatically to avoid the risk of collision.

[0103] Automatic acceleration: The system can automatically control the acceleration and deceleration of the vehicle according to the driving status of the vehicle in front to maintain a safe distance from the vehicle in front.

[0104] Automatic steering: By detecting the vehicle's driving status and road conditions through sensors, the system can automatically control the vehicle's steering to ensure that the vehicle is driving on the correct road.

[0105] Step S2:

[0106] In this embodiment, determining the current emotion of the driver based on intelligent fusion perception mainly includes identifying the current emotional state based on voice and language feature information.

[0107] The intelligent fusion perception includes but is not limited to:

[0108] 1) Driver’s facial expression and head movement information captured by the camera;

[0109] 2) Driver language and voice information extracted by microphone;

[0110] 3) Driving information extracted by motion sensors such as steering, braking, and acceleration;

[0111] 4) Information about the driver’s daily use of the car, such as the force and angle of opening and closing the door.

[0112] Further, such as Figure 3 As shown, if it is determined that there are many potential factors causing road rage in the current road conditions, it can be mainly identified whether the driver is currently in an angry, nervous or irritable mood through the current driver's voice and language feature information provided by the microphone.

[0113] The voice feature information may include volume, speaking speed, pitch, intensity, formant, tone, etc. While taking into account voice features such as personality and age, the driver's feature information is matched with typical angry / irritable / nervous voice feature information to determine whether the driver is currently in road rage.

[0114] The language feature information mainly includes the detection of swear words. The voice and language detection may also include other information related to the driver, such as interjections, sighs, patting objects, and whether there is a sound of a mobile phone playing anything. This type of information can be used to determine whether the driver is under the influence of negative emotions.

[0115] Step S3:

[0116] like Figure 4As shown, determining the personality characteristics of the driver based on the results of long-term intelligent fusion perception mainly includes identifying the driver's personality traits based on voice feature information and a personality trait model, wherein the personality trait model is used to determine the personality trait information based on the voice feature information.

[0117] A microphone is provided in the car, and the computer device obtains multiple frames of voice data of the driver during the process of driving the car through the microphone.

[0118] In order to obtain the correlation between the driver's voice characteristics and personality when driving, and because the car may include other people in the co-pilot seat or the back seat in addition to the driver in the main driving seat; therefore, it is necessary to determine whether the acquired voice data is the driver's voice data; if the voice data is not the driver's voice data, the voice data is reacquired until the driver's voice data is acquired; in addition, if the voice data includes the driver's voice data but also includes the voice data of other people, the voice data still needs to be reacquired until the voice data including only the driver's voice data is acquired, thereby improving the quality of the acquired voice data.

[0119] The driver's voiceprint features are stored in advance in the computer device; accordingly, the step of the computer device determining whether the acquired voice data is the driver's voice data may be:

[0120] Extract voiceprint features from the voice data, and determine whether the extracted voiceprint features match the voiceprint features of the driver; if the extracted voiceprint features match the voiceprint features of the driver, determine that the voice data is the driver's voice data; if the extracted voiceprint features do not match the driver's voiceprint features, determine that the voice data is not the driver's voice data.

[0121] Based on multiple frames of speech data, multiple groups of sound features of the driver while driving the car are determined, each group of sound features including volume, pitch, intensity, tone, formant, speaking speed, effective speech length and number of pauses.

[0122] Because it is necessary to find the correlation between the driver's voice characteristics when driving and his personality, it is necessary to solve the characteristic values ​​of the driver's voice when driving, such as pitch, sound intensity, resonance peak, effective speech length, speaking speed and number of pauses, and then solve the maximum value, minimum value, variance, mean and median of these characteristics to obtain the driver's voice characteristics.

[0123] The sound characteristics include pitch; this is achieved by the following steps, including:

[0124] For each frame of speech data, determining a fundamental frequency value of the speech data;

[0125] The pitch is determined by the vibration frequency of the sound source; in this step, the computer device performs autocorrelation calculation on each frame of speech data, and estimates the fundamental frequency value of each frame of speech data through the result of the autocorrelation calculation.

[0126] Based on the fundamental frequency value of the voice data, the pitch of the driver determined based on the voice data is determined by the following formula:

[0127]

[0128] Here, h represents the pitch of the driver determined based on the voice data, and f represents the fundamental frequency value of the voice data.

[0129] Sound features include sound intensity; sound intensity is the energy carried by sound propagation and is determined by the amplitude of the sound. This embodiment is calculated by Fourier transform and energy spectrum in the frequency domain. Fourier transform is performed on each frame of data to convert the time domain signal into a frequency domain signal. The amplitude of the frequency domain signal is squared to calculate the energy spectrum of each frame. The energy spectrum represents the energy of each frequency component. The sound intensity of each frame is calculated based on the energy spectrum.

[0130] Accordingly, this step may be:

[0131] For each frame of speech data, the speech data is Fourier transformed to obtain a frequency domain signal of the speech data, and the frequency domain signal includes multiple frequencies; for each frequency in the frequency domain signal, the square of the amplitude of the frequency is determined to obtain the energy spectrum of the frequency; based on the energy spectrum of each frequency, the sound pressure value of the speech data is determined; based on the sound pressure value of the speech data, the sound intensity of the driver determined based on the speech data is determined by the following formula three:

[0132]

[0133] Wherein, Lp represents the sound intensity of the driver determined based on the voice data, p represents the sound pressure value, and p0 represents the reference sound pressure value.

[0134] Sound features include resonance peaks; resonance peaks: broad peaks or local maximum values ​​in the spectrum. Resonance peaks are areas where energy is relatively concentrated in the sound spectrum. They reflect the physical characteristics of the vocal tract. The key to extracting resonance peaks is to estimate the maximum value in the spectrum envelope of natural speech. Resonance peak information is contained in the frequency envelope, so the methods for extracting resonance peak parameters include the cepstrum method, the LPC (linear predictive coding) method, and the HHT (Huang transform) method. In the embodiment of the present application, the cepstrum method is used as an example to determine the resonance peak; accordingly, it can be achieved by the following steps, including:

[0135] Perform pre-emphasis processing on multiple frames of voice data;

[0136] Performing windowing and framing processing on the multiple frames of voice data after pre-emphasis processing to obtain multiple frames of second voice data;

[0137] The frame length of each frame of the second voice data in the multiple frames of the second voice data is not greater than the frame length of the frame processing; and the frame length of the frame processing can be set and changed as needed.

[0138] By performing Fourier transform on multiple frames of second voice data using the following formula, frequency domain signals of multiple frames of second voice data are obtained:

[0139]

[0140] Wherein, Xi(n) represents the frequency domain signal of the second voice data of the i-th frame, xi(n) represents the second voice data of the i-th frame, N represents the frame length of the frame processing, and n represents the number of frames of the second voice data;

[0141] Performing inverse Fourier transformation on frequency domain signals of multiple frames of second voice data to obtain inverse Fourier transform sequences of multiple frames of second voice data;

[0142] Based on the dop sequence of the multi-frame second voice data, the envelope of the frequency domain signal of the multi-frame second voice data is determined by the following formula:

[0143]

[0144] Wherein, Hi(k) represents the envelope of Xi(n), hi(n) represents the product of the window function and the cepstrum sequence, N represents the frame length of the frame processing, and n represents the number of frames of the second voice data;

[0145] For each second voice data, a maximum value is determined on the envelope of the frequency domain signal of the second voice data, and a resonance peak of the driver determined based on the second voice data is obtained.

[0146] The sound feature includes the effective speech length; accordingly, the effective speech length can be obtained by determining the difference between the duration of multiple frames of speech data and the silence duration.

[0147] The sound features include speech rate, which is the vocabulary capacity contained in a unit time; accordingly, the speech rate can be obtained by determining the ratio of the total duration of multiple frames of voice data to the number of words.

[0148] The sound features include the number of pauses; accordingly, it can be: removing the time period at the beginning of silence and the time period at the end of silence from multiple frames of voice data, and determining the number of times the silence duration in the remaining voice segments is greater than the preset duration as the number of pauses.

[0149] After completing the above steps, based on the multiple groups of sound features, the maximum value, minimum value, variance, mean and median of each sound feature included in the multiple groups of sound features are determined to obtain the sound features of the driver in the process of driving the car.

[0150] A group of sound features includes pitch, sound intensity, resonance peak, effective speech length, speaking speed and number of pauses; in this step, the computer device determines the maximum value, minimum value, variance, mean and median of the pitch included in multiple groups of sound features, determines the maximum value, minimum value, variance, mean and median of the sound intensity included in multiple groups of sound features, determines the maximum value, minimum value, variance, mean and median of the effective speech length included in multiple groups of sound features, determines the maximum value, minimum value, variance, mean and median of the speaking speed included in multiple groups of sound features, determines the maximum value, minimum value, variance, mean and median of the number of pauses included in multiple groups of sound features, and obtains the sound features of the driver in the process of driving the car.

[0151] Afterwards, sample data is generated based on the voice features, and the label of the sample data is the driver's personality features obtained in advance. The computer device generates multiple sample data based on the voice features of multiple drivers, and then performs model training. Based on the sample data, the personality trait model is trained through a machine learning algorithm, and the personality trait model is used to identify the user's personality features.

[0152] The computer device constructs multiple machine learning models based on multiple machine learning algorithms, determines quality parameter values ​​of the multiple machine learning models, selects a target machine learning model with the best quality from the multiple machine learning models based on the quality parameter values ​​of the multiple machine learning models, and then trains the target machine learning model based on the sample data. The model obtained after the training is the personality trait model.

[0153] The multiple machine learning algorithms may be multiple algorithms such as random forest, naive Bayes, support vector machine, decision tree, and neural network.

[0154] The step of determining the quality parameter values ​​of the plurality of machine learning models by the computer device may be:

[0155] For each machine learning model, the computer device determines the accuracy, precision, recall rate, and comprehensive evaluation score of the machine learning model, and performs weighted summation of the accuracy, precision, recall rate, and comprehensive evaluation score of the machine learning model to obtain the quality parameter value of the machine learning model.

[0156] In the embodiment of the present application, since the machine learning model with the best quality is selected for training, the accuracy of the personality trait model obtained based on the machine learning model training can be improved.

[0157] Furthermore, the personality trait model needs to obtain a training data set, and model training is performed based on the training data set to obtain the personality trait model.

[0158] Furthermore, the driver's voice feature information is input into the personality trait model to obtain the corresponding personality trait prediction result of the driver. The personality trait prediction result should include five major personality traits: openness, conscientiousness, extraversion, agreeableness and neuroticism, which can be used as a factor in selecting the sound and voice material for voice interaction.

[0159] Furthermore, appropriate sounds are selected from an existing sound library, and corresponding materials are selected from existing voice materials according to different road scenes.

[0160] Furthermore, the in-vehicle terminal equipment should have a pre-set default voice language setting, which can be set to female, with a polite and friendly voice tone that is consistent with the vehicle's product image.

[0161] Furthermore, based on the driver's personality traits and the road condition information obtained by the ADAS system, the sound and voice materials that suit the current situation are selected.

[0162] The voice material contains both positive and negative content, and both contents are based on environmental description as much as possible, and the two can be used in combination.

[0163] Furthermore, the definition of positive content is praise-based, for example:

[0164] 1. "It is right not to cut in line at will. You are an excellent driver who obeys the traffic rules."

[0165] 2. “You helped to keep the road driving environment safe”;

[0166] 3. “Your driving makes people feel at ease”;

[0167] 4. "You are the guardian and practitioner of road safety."

[0168] Furthermore, the definition of negative content is prompt, warning and command, and the driver's full name (XXX) can be used as a reminder. For example:

[0169] 1. "XXX, please slow down as there is a turn ahead";

[0170] 2. "XXX, the road is slippery in heavy rain, and you need to keep a distance between vehicles";

[0171] 3. "XXX, you should stay focused and drive carefully now. Being angry will only lead to a car accident."

[0172] 4. "XXX, the drivers and passengers of the surrounding vehicles can see your bad driving behavior and will condemn your driving behavior," etc.

[0173] Furthermore, the voice aspect takes into account the personality traits of the driver, selects the voice that suits their character and emotional preferences, and allows the driver to switch the voice of voice interaction at any time.

[0174] Step S4:

[0175] The driver's emotion relief program is determined based on the current road conditions and driver personality traits obtained by the ADAS system. It mainly includes alleviating the driver's road rage and negative emotions during driving through voice interaction. Through voice interaction, the following purposes are achieved:

[0176] 1) Alleviate road rage and prevent drivers from engaging in bad and dangerous driving behaviors;

[0177] 2) Divert the driver's attention, so that he can focus on driving instead of venting his emotions. In the long run, the driver can learn to look at bad road conditions more objectively and rationally, and better adjust his mentality;

[0178] 3) Suppress swear words and prevent them from reinforcing the negative effects of driving anger.

[0179] In one embodiment, based on the situation where another car cuts in on the road, in the same negative driving emotional atmosphere:

[0180] For drivers with open personality traits, voice interaction will take into account the thinking style of open personality and remind the driver in a more direct language, such as: "You are currently angry or impatient, please calm down in time and keep a good mood while driving";

[0181] For drivers with responsible personality traits, voice interaction will be considered from the perspective of driving safety, and will interact with encouraging language, such as: "You are a very responsible driver, please pay attention to the safety of yourself and others, and avoid cutting in in time";

[0182] For drivers with extroverted personality traits, voice interaction will use more direct or clear language to ease the driver's current emotions, such as: "Beautiful lady / handsome guy, please don't get angry because of the vehicle behind you cutting in, as anger will ruin your happy mood."

[0183] For drivers with agreeable personality traits, voice interaction will ease the driver's current emotions with friendly, collaborative and supportive language. For example: "Thank you for always thinking about others and being patient. Your friendly behavior not only makes the road more harmonious, but also ensures driving safety. Please continue to influence the people around you with your good behavior."

[0184] For drivers with neurotic personality traits, voice interaction will use soothing language to help drivers manage their current emotions. For example: "You can try to take a deep breath and slow down, give yourself some space to adjust your mood. We understand that your current emotions are normal, but focusing on safe driving can make the journey smoother and more comfortable."

[0185] When detecting the driver's road rage, the present invention combines road condition information to identify whether there is a driving scene that causes road rage. When there is a driving scene that causes road rage, it further identifies whether the driver is in a negative emotion of road rage based on the driver's real-time voice and language feature information. The above technical means can identify road rage in a more targeted manner.

[0186] When soothing and alleviating the driver's road rage, the driver's personalized factors are combined, the driver's personality characteristics are taken into account, personalized sound and voice materials are matched, and personalized voice interaction content is generated to alleviate the driver's negative emotions of road rage, which can achieve better and more targeted relief effects.

[0187] The method of the present invention combines scenarios that are likely to cause negative driving, real-time fusion perception of the driver, and long-term fusion perception to determine the driver's road rage risk and effectively alleviate it, reducing the probability of road rage, while ensuring the driver's safe driving and improving the driving experience. The ADAS system directly obtains surrounding road condition information through the automatic driving sensor, which is more accurate and real-time than other road information obtained using the GPS system; the voice system can provide different road rage emotion relief feedback for different personality traits.

[0188] Embodiment 2

[0189] This embodiment discloses a road rage recognition and mitigation system based on personalized intelligent voice interaction.

[0190] Road rage recognition and mitigation system based on personalized intelligent voice interaction, including:

[0191] The road rage factor detection module is configured to: obtain road condition information, and detect whether there is a driving scenario that causes road rage in the road condition information;

[0192] The road rage emotion recognition module is configured to: when there is a driving scene that causes road rage in the road condition information, obtain the driver's real-time voice and language feature information to identify whether the driver is in a negative emotion of road rage;

[0193] The personality characteristic determination module is configured to: obtain voice characteristic information of the driver during daily driving to determine the personality characteristics of the driver;

[0194] The personalized voice interaction module is configured to: match personalized voice according to the driver's personality characteristics, match voice materials according to road condition information, and when the driver is under the influence of negative emotions, generate personalized voice interaction content based on the personalized voice and voice materials to alleviate the driver's negative emotions of road rage.

[0195] Embodiment 3

[0196] The purpose of this embodiment is to provide a computer-readable storage medium.

[0197] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the method for identifying and alleviating road rage based on personalized intelligent voice interaction as described in Example 1 of the present disclosure.

[0198] Embodiment 4

[0199] The purpose of this embodiment is to provide an electronic device.

[0200] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for identifying and alleviating road rage based on personalized intelligent voice interaction as described in Example 1 of the present disclosure are implemented.

[0201] The steps involved in the apparatuses of the above embodiments 2, 3 and 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0202] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0203] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A road rage identification and mitigation method based on personalized intelligent voice interaction, characterized in that: The following steps are involved: Obtaining road condition information and detecting whether there is a driving scenario that may cause road rage in the road condition information; When there is a driving scenario that may cause road rage in the road condition information, the real-time voice and language feature information of the driver is obtained to identify whether the driver is in a negative emotion of road rage; Obtain the driver's voice feature information during daily driving and determine the driver's personality characteristics; Personalized voice is matched according to the driver's personality characteristics, and voice materials are matched according to road conditions information. When the driver is under the influence of negative emotions, personalized voice interaction content is generated based on the personalized voice and voice materials to alleviate the driver's negative emotions of road rage.

2. The method for road rage identification and mitigation based on personalized intelligent voice interaction according to claim 1, characterized in that: Obtain road condition information and detect whether there is a driving scenario that causes road rage in the road condition information, including: Obtaining road condition information through the ADAS system, including traffic congestion, speed of surrounding vehicles, whether other vehicles have cut in, distance to surrounding vehicles, and whether surrounding vehicles frequently change lanes; It is detected whether there are driving scenarios that may cause road rage in the road condition information, where the driving scenarios that may cause road rage include being cut in by another vehicle, sudden braking of the vehicle ahead, and traffic jam time exceeding a set time.

3. The method for road rage identification and mitigation based on personalized intelligent voice interaction according to claim 2, characterized in that: The ADAS system obtains road condition information, including collecting information around the vehicle through various sensors and predicting potential dangerous conditions. The information around the vehicle is collected through various sensors, including: Capture visual information around the vehicle through cameras, including road signs, lane lines, pedestrians, and obstacles; Detect the distance, speed and azimuth of the target object through millimeter wave radar; Obtain depth information of the surrounding environment through LiDAR; Close range detection via ultrasonic radar; or, Predicting potentially dangerous situations, including: Detect obstacles around the vehicle and track them based on the information collected by the sensors; Identify the vehicle's current driving status and lane position, and give a warning when the vehicle deviates from the lane; Detect obstacles in front of the vehicle and predict collision risks, giving early warnings; Detects the vehicle in front of you and automatically adjusts your speed to maintain a safe distance from the vehicle in front.

4. The method for road rage identification and mitigation based on personalized intelligent voice interaction according to claim 1, characterized in that: Obtain the driver's real-time voice and language feature information to identify whether the driver is in a negative emotion of road rage, including: Obtain the driver's voice feature information, including volume, speaking speed, pitch, intensity, formant, and tone; obtain the driver's language feature information, including swear words and uncivilized language; obtain other information related to the driver, including interjections, sighs, sounds of slapping objects, and sounds played by mobile phones; While considering the driver's personality and age voice characteristics, the above-mentioned voice feature information, language feature information and other driver-related information are comprehensively considered and matched with the corresponding feature information of typical anger / irritability / tension to determine whether the driver is already in the negative emotion of road rage.

5. The method for road rage identification and mitigation based on personalized intelligent voice interaction according to claim 1, characterized in that: Acquire the driver's voice feature information during daily driving and determine the driver's personality characteristics, including: Acquiring speech feature information of the driver during daily driving, inputting the speech feature information of the driver during daily driving into a personality trait model, and obtaining personality trait prediction results, wherein the personality trait prediction results include five personality traits: openness, responsibility, extroversion, agreeableness, and neuroticism; The personality trait model is used to determine personality trait information based on speech feature information.

6. The method for road rage identification and mitigation based on personalized intelligent voice interaction according to claim 1, characterized in that: Match personalized voice according to the driver's personality characteristics, and match voice materials according to road conditions, including: Consider the personality characteristics of the driver and select a voice that matches the driver's personality and emotional preferences from the existing voice library as a matching personalized voice. The personalized voice needs to be polite and friendly and consistent with the vehicle's product image. According to the road condition information, voice materials that match the current road condition information are selected from the existing voice material library as matching voice materials. The matching voice materials contain both positive and negative content and use environmental description as the subject, wherein the positive content is praise-like and the negative content is prompt-like, warning-like and imperative-like.

7. The method for road rage identification and mitigation based on personalized intelligent voice interaction according to claim 5, characterized in that: Generate personalized voice interaction content based on personalized sound and voice materials to alleviate the negative emotions of road rage for drivers, including: Considering the specific classification of the driver's five personality traits of openness, responsibility, extroversion, agreeableness and neuroticism, personalized voice interaction content is generated based on personalized sound and voice materials, including: For drivers with open personality traits, use more direct language to remind the driver; For drivers with responsible personality traits, from the perspective of driving safety, use encouraging language to remind the driver; For drivers with extraversion personality traits, the language used to remind the driver is more direct or clear than that for drivers with openness personality traits.

8. A road rage recognition and mitigation system based on personalized intelligent voice interaction, characterized in that: include: The road rage factor detection module is configured to: obtain road condition information, and detect whether there is a driving scenario that causes road rage in the road condition information; The road rage emotion recognition module is configured to: when there is a driving scene that causes road rage in the road condition information, obtain the driver's real-time voice and language feature information to identify whether the driver is in a negative emotion of road rage; The personality characteristic determination module is configured to: obtain voice characteristic information of the driver during daily driving to determine the personality characteristics of the driver; The personalized voice interaction module is configured to: match personalized voice according to the driver's personality characteristics, match voice materials according to road condition information, and when the driver is under the influence of negative emotions, generate personalized voice interaction content based on the personalized voice and voice materials to alleviate the driver's negative emotions of road rage.

9. A computer-readable storage medium having a program stored thereon, characterized in that: When the program is executed by a processor, the steps in the method for identifying and alleviating road rage based on personalized intelligent voice interaction as described in any one of claims 1 to 7 are implemented.

10. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the road rage identification and mitigation method based on personalized intelligent voice interaction as described in any one of claims 1 to 7 are implemented.

Images