Method for automatically classifying journeys of a motor vehicle into categories of an electronic logbook, device and computer program product

Abstract

Method for automatically classifying journeys of a motor vehicle (10) into categories of an electronic logbook, wherein: - during the respective journey of the motor vehicle (10) whose interior is continuously monitored by means of at least one microphone device (14) of the motor vehicle and the microphone device (14) provides at least one continuous audio data stream (16a-d) which characterizes at least one sound event in the interior detected by the monitoring of the interior by means of the microphone device (14); - by means of an electronic computing device (20) the audio data stream (16a-d) is analyzed and it is thereby determined whether the sound event includes at least one human voice; - if it is determined that the sound event includes at least one human voice, the respective journey is automatically classified into one of the predefined categories of the electronic logbook depending on the voice identified; and - by means of the electronic computing device (20) the determined voice is assigned to one of several seats (12a-d) arranged in the interior of the motor vehicle, whereby the respective journey is automatically classified into one of the predefined categories of the electronic logbook depending on the assigned seat.

Description

[0001] The invention relates to a method for automatically classifying journeys of a motor vehicle into categories of an electronic logbook. Furthermore, the invention relates to a data processing device. The invention also relates to a computer program product.

[0002] German patent DE 200 11 139 U1 discloses a navigation system as known in which a mobile communication transmitter is integrated. This transmitter in the navigation system is automatically addressed by software and transmits specific data for a logbook. Furthermore, WO 2023 / 036403 A1 discloses a method for providing an automatic logbook for a driver of a vehicle.

[0003] The object of the present invention is to create a method, a device and a computer program product such that journeys of a motor vehicle can be automatically classified and thus categorized in a particularly advantageous manner into categories of an electronic logbook.

[0004] This problem is solved by a method with the features of claim 1, by a device with the features of claim 8, and by a computer program product with the features of claim 9. Advantageous embodiments with expedient further developments of the invention are specified in the remaining claims.

[0005] A first aspect of the invention relates to a method for automatically classifying journeys of a motor vehicle, also referred to simply as a vehicle and preferably designed as a motor car, in particular as a passenger car, into, in particular different, categories of an electronic logbook. The automatic classification of the journeys of the motor vehicle into the categories of the electronic logbook is also referred to as automatic categorization or automatic categorization of journeys.

[0006] In this method, the interior of the motor vehicle, also referred to as the passenger compartment or passenger space, is continuously monitored during each journey using at least one microphone device, in particular an electrical or electronic one. The microphone device provides a continuous audio data stream. This audio data stream, also simply referred to as the data stream, characterizes at least one sound event in the interior, detected by the microphone device. In other words, the audio data stream characterizes at least one sound event occurring in the interior, which is detected by the microphone device by continuously monitoring the interior of the motor vehicle, also referred to as the passenger compartment or passenger space, during each journey.The feature that the microphone device continuously listens to the interior of the motor vehicle during the journey means that the microphone device listens to the interior without interruption, i.e., continuously, for a period of time, whereby this period extends over at least part of the journey, in particular over a predominant part and thus over at least half of the journey, and preferably over the entire journey, specifically from the beginning of the journey (also referred to as the start of the journey) to the end of the journey (also referred to as the end of the journey). Since the microphone device continuously listens to the interior during the journey, i.e., continuously, it captures the sound event occurring during the journey.Within the scope of this disclosure, the term "sound event" refers to at least one sound wave, and in particular several sound waves, which is or are detected by the microphone device by continuously monitoring the interior of the motor vehicle. In other words, the sound event is or comprises at least one sound wave, and in particular several sound waves, which is or are detected by the microphone device by continuously monitoring the interior while the vehicle is in motion.

[0007] In this process, an electronic computing device analyzes the audio data stream to determine whether the sound event includes at least one human voice from a person inside the vehicle. In other words, people such as the driver may be inside the vehicle, particularly while it is in motion. If a person inside the vehicle speaks while it is moving, for example, by uttering words, then the sound event includes this voice and these words. This voice and these words are captured by the microphone system, which continuously monitors the vehicle's interior while it is in motion. Since the sound event includes the voice, the voice is characterized by the audio data stream.

[0008] The electronic computing device used to analyze the audio data stream in order to determine whether the sound event includes at least one human voice is, for example, either a component of the motor vehicle or an external device located outside the vehicle, which is not a component of the vehicle. For example, the microphone provides the audio data stream, which is received by the electronic computing device, either via a wired or wireless connection.

[0009] If the electronic computing device determines that the sound event includes at least one human voice, the respective journey is automatically classified, specifically by means of the electronic computing device or another electronic computing device, into one of the predefined categories of the electronic logbook, depending on the voice detected. The other electronic computing device can, for example, be a component of the vehicle, or it can be an external device, provided in addition to the vehicle, which is not a component of the vehicle.For example, the categories are stored in a data storage device, particularly electrical or electronic, also simply referred to as storage, of the electronic computing device or other electronic computing device. Furthermore, it is conceivable that the electronic logbook is stored in the data storage device, also simply referred to as storage.

[0010] The invention thus utilizes, for example, speech recognition, which is carried out by means of the electronic computing device, to determine whether the sound event includes at least one human voice. The method, in particular the electronic computing device and the further electronic computing device, then performs an automatic categorization in order to classify the respective journey into, and in particular precisely, one of the categories, depending on the detected voice.

[0011] The first category, for example, characterizes a commute, which is understood to be a journey between a person's home and their place of work. If a journey is classified into the first category, the procedure automatically determines that the journey is a commute between home and work. The second category characterizes, for example, a business trip, which is understood to be a journey undertaken by a person using a motor vehicle for professional or business reasons, and not between home and work.If the procedure automatically classifies a trip into the second category, i.e., assigns it to or divides it into the second category, then the procedure automatically determines that the trip in question is a business trip. A third category, for example, characterizes a private trip undertaken by a person for purely private reasons using a motor vehicle. If the procedure automatically classifies a trip into the third category, i.e., assigns it to or divides it into the third category, then the procedure automatically determines that the trip in question is a private trip.

[0012] The invention is based in particular on the following findings, considerations, and background information: A problem identified in the automated creation of an electronic logbook is the correct categorization of automatically recognized trips. For the electronic logbook, also known as a digital logbook, to be recognized as valid by tax authorities and, in particular, to be financially advantageous for the user, the categorization of each trip should be correct and plausible. The aforementioned three categories are available for selection.In other words, for example, classifying the respective trip into one of the specified categories means that, depending in particular on the vote determined, one of the specified categories is selected, and the respective trip is assigned to the selected category.

[0013] A logbook typically serves as evidence for a number of business trips, also known as company journeys. Due to the importance of correct trip categorization, conventional solutions usually require manual categorization in the electronic logbook by a single user. This user often has only a limited timeframe, such as seven days, to complete the task, particularly after the trip has ended. If the user fails to categorize the trip within this timeframe, conventional solutions automatically classify it as a private trip, even if it was actually a business trip. This can lead to significant disadvantages for the user, especially financial ones. Automatic categorization can address this issue, and this invention enables precise trip categorization.

[0014] Furthermore, it was found that trip recognition based solely on movement and / or position data of the vehicle, and thus the categorization of a vehicle's trips, does not allow for sufficient conclusions to be drawn about the actual category of the respective trip. For example, additional environmental characteristics that can be derived during the respective trip should be used to categorize the trips.

[0015] Preferably, the inventive method automatically classifies each journey into one of the predefined categories of the electronic logbook, based on movement and / or position data. The movement and / or position data are determined, for example, particularly during the journey, by means of a vehicle navigation system, and especially by satellite-based means. The movement and / or position data characterize, for example, the vehicle's positions on Earth, which the vehicle has assumed or reached during the journey.

[0016] In the invention, the electronic computing device assigns the detected voice to one of several seats arranged in the interior of the motor vehicle, with the respective trip being automatically classified into one of the predefined categories of the electronic logbook depending on the assigned seat. Thus, the invention utilizes the fact that modern speech recognition systems in motor vehicles are capable of distinguishing between different voices, particularly when the respective speech recognition system detects different voices, and of assigning each voice, especially precisely, to a specific seat arranged in the interior of the motor vehicle, particularly during a journey. This avoids misinterpretations of voice commands or advantageously minimizes their number.

[0017] In order to categorize journeys in a particularly advantageous, and especially precise, way, one embodiment of the invention provides that, by means of the electronic computing device, at least one voice characteristic of the identified voice is determined by analyzing the audio data stream and based on the identified voice, wherein the respective journey is automatically classified, depending on the identified voice characteristic, into, and in particular precisely, one of the predefined categories of the electronic logbook.

[0018] Another embodiment is characterized in that, by means of the electronic computing device, at least one personal characteristic of a person who is present in the interior of the vehicle, particularly during the journey, and whose voice is detected, is determined by analyzing the audio data stream and based on the detected voice. The respective journey is then automatically classified, depending on the detected personal characteristic, into one of the predefined categories of the electronic logbook. This allows incorrect categorizations to be avoided or their number to be advantageously kept to a minimum.

[0019] The voice attribute can comprise exactly one individual voice attribute or several individual voice attributes. Furthermore, it is conceivable that the person attribute comprises exactly one individual person attribute or several individual person attributes.

[0020] To enable particularly precise automatic categorization of journeys, a further embodiment of the invention provides that the person characteristic includes the person's age and / or gender and / or accent and / or language. The person characteristic, gender, accent, and language are thus individual characteristics of the person characteristic. In particular, it is conceivable that individual characteristics of the voice and / or the person, determined by analyzing the audio data stream and the identified voice using the electronic computing device, are combined or summarized into at least one characteristic vector. One of the individual characteristics could, for example, be the aforementioned individual voice characteristic. Alternatively or additionally, one of the individual characteristics could, for example, be the individual person characteristic.Alternatively or additionally, at least some of the individual characteristics could be the aforementioned personal characteristics, such as age, gender, accent, and / or language. Depending on the characteristic vector, each trip can then be categorized particularly precisely and automatically.

[0021] In a further, particularly advantageous embodiment of the invention, it is provided that at least one correction input, effected by a person, preferably manually or by voice command, is received by means of a first input device, in particular electrical or electronic. Preferably, the input device is an input device, i.e., a component of the motor vehicle. It is also conceivable that the input device is not a component of the motor vehicle and is thus provided in addition to the motor vehicle, wherein the input device can, for example, be an external device with respect to the vehicle. In this case, the respective trip is automatically classified, depending on the received correction input, into, in particular, precisely one of the predefined categories of the electronic logbook.This means that a person can correct the automatic categorization of a trip performed by the system. This allows for precise and particularly advantageous categorization of trips for users of the electronic logbook. For example, if a trip is automatically assigned to one category, the person can correct this automatic classification and, for instance, assign the trip to a different category.

[0022] To enable particularly advantageous automatic categorization of journeys, a further embodiment of the invention provides that each journey is automatically classified, based on the detected voice and a set of rules, into one of the predefined categories of the electronic logbook. For example, the set of rules is stored in the aforementioned data storage device. Depending on the set of rules, it is checked, for example, by means of the electronic computing device or by means of a further electronic computing device, whether the detected voice, in particular at least one value characterizing the detected voice, fulfills at least one or more predefined criteria. Depending on whether the respective criterion is fulfilled, the respective journey is then automatically classified, in particular, into one of the predefined categories of the electronic logbook.

[0023] It has proven particularly advantageous to modify, or rather, vary and adapt, the rule set based on the correction input. This allows the quality of the automatic categorization of each trip to improve over time, resulting in a particularly effective automatic categorization of trips. By modifying the rule set based on the correction input, the process learns or is trained, thus enabling a particularly effective automatic categorization of trips.

[0024] The invention allows at least the following advantages to be realized: - Compared to conventional solutions, less manual user intervention is required to ensure the correct maintenance of the digital logbook, resulting in greater user convenience compared to conventional solutions; - Avoidance of incorrectly categorized journeys after the aforementioned period of, for example, seven days, and thus avoidance of monetary disadvantages; - Use of functionality already available in the motor vehicle through voice control, i to advantageously carry out the automatic categorization of journeys.

[0025] A second aspect of the invention relates to a device for data processing, also referred to as a system, wherein the device comprises means for carrying out the method according to the first aspect of the invention. Advantages and advantageous embodiments of the first aspect of the invention are to be regarded as advantages and advantageous embodiments of the second aspect of the invention and vice versa.

[0026] A third aspect of the invention relates to a computer program product comprising instructions that, when executed by a computer, cause the computer to execute the method according to the first aspect of the invention. Advantages and advantageous embodiments of the first and second aspects of the invention are to be regarded as advantages and advantageous embodiments of the third aspect of the invention, and vice versa.

[0027] A further, fourth aspect of the disclosure relates to a computer-readable storage medium comprising instructions which, when executed by a computer, cause the computer to execute the method according to the first aspect of the invention. Advantages and advantageous embodiments of the first, second, and third aspects of the invention are to be regarded as advantages and advantageous embodiments of the fourth aspect, and vice versa.

[0028] Further advantages, features, and details of the invention will become apparent from the following description of a preferred embodiment and from the drawing. The features and combinations of features mentioned above in the description, as well as those mentioned below in the figure description and / or shown in the single figure alone, can be used not only in the combinations specified, but also in other combinations or individually, without departing from the scope of the invention.

[0029] The drawing shows, in its single figure, a block diagram to illustrate a procedure for the automatic categorization of journeys of a motor vehicle in an electronic logbook.

[0030] The single figure shows a block diagram, which is used below to describe a method for automatically classifying journeys of a motor vehicle 10, depicted schematically in the figure, into categories of an electronic logbook. The automatic classification of journeys of the motor vehicle, also referred to simply as the vehicle, into the aforementioned categories of the electronic logbook (for example, at least or exactly three) is also referred to as automatic categorization of journeys. For example, the motor vehicle 10, whose interior (also referred to as the passenger compartment) is bounded by a structure of the motor vehicle 10, such as a self-supporting body, is classified as a motor vehicle, in particular a passenger car.In the embodiment shown in the figure, at least or exactly four seats 12a-d are arranged in the interior of the motor vehicle 10, which is also simply referred to as a vehicle. Each seat 12a-d is provided, for example, by a respective seating arrangement of the motor vehicle 10, the respective seating arrangement being located in the interior.

[0031] In this method, the interior of the motor vehicle 10 is continuously monitored during each journey by means of at least one microphone device 14 of the motor vehicle 10, which is shown schematically in the figure. In particular, the microphone device 14 provides a continuous audio data stream 16a-d for each seat 12a-d, especially during the journey. The respective audio data stream 16a-d characterizes at least one sound event in the interior, which is detected by the microphone device 14 during the monitoring of the interior. In other words, if, for example, a first person located in seat 12a, and thus in the interior, speaks during the journey, the sound event characterized by the audio data stream 16a assigned to seat 12a is or includes the first voice.For example, if a second person, located in seat 12b and thus in the passenger compartment, speaks with their second voice while the vehicle is in motion, then the sound event characterized by audio data stream 16b, assigned to seat 12b, is or includes the second voice. Similarly, if a third person, located in seat 12c and thus in the passenger compartment, speaks with their third voice while the vehicle is in motion, then the sound event characterized by audio data stream 16c, assigned to seat 12c, is or includes the third voice. Finally, if a third person, located in seat 12d and thus in the passenger compartment, speaks with their third voice while the vehicle is in motion, then the sound event characterized by audio data stream 16d, assigned to seat 12d, is or includes the fourth voice.

[0032] Naturally, the method is applicable to more or fewer than the four seats 12a-d shown in the figure. The figure schematically depicts speaker recognitions 18a-d. The speaker recognitions 18a-d are, for example, components of an electronic computing device 20 or are performed with this electronic computing device 20. In the embodiment shown in the figure, the electronic computing device 20 is a first electronic computing device and a component of the motor vehicle 10, thus an electronic computing device of the motor vehicle 10. It can be seen that speaker recognition 18a is assigned to the audio data stream 16a, speaker recognition 18b to the audio data stream 16b, speaker recognition 18c to the audio data stream 16c, and speaker recognition 18d to the audio data stream 16d, and vice versa.Using the respective speaker recognition functions 18a-d, the corresponding audio data stream 16a-d is analyzed to determine whether the sound event characterized by the respective audio data stream 16a-d includes at least one human voice. If, as described above, the four people are speaking during the journey, such that the sound events include the four voices, then the first voice is identified using speaker recognition 18a, the second voice using speaker recognition 18b, the third voice using speaker recognition 18c, and the fourth voice using speaker recognition 18d. In other words, speaker recognition functions 18a-d determine, or rather, detect, that the people are speaking during the journey.

[0033] Furthermore, speaker classifications 22a-d are shown schematically in the figure. In the embodiment shown in the figure, the speaker classifications 22a-d are, for example, components of the first electronic computing unit 20 or are executed by the first electronic computing unit 20. It can be seen that during the journey, for each seat 12a-d, in particular an occupied one, the respective assigned and thus associated audio data stream 16a-d is forwarded, i.e., transmitted, to the respective assigned speaker classification 22a-d, whereby the respective speaker classification 22a-d receives the forwarded and assigned audio data stream 16a-d.It can be seen that in the embodiment shown in the figure, speaker classification 22a is assigned to audio data stream 16a, speaker classification 22b to audio data stream 16b, speaker classification 22c to audio data stream 16c, and speaker classification 22d to audio data stream 16d, and vice versa. Each speaker classification 22a-d determines at least one feature vector from its respective assigned audio data stream 16a-d, each vector comprising several individual features.The feature vector determined from audio data stream 16a by speaker classification 22a is also referred to as the first feature vector; the feature vector determined from audio data stream 16b by speaker classification 22b is also referred to as the second feature vector; the feature vector determined from audio data stream 16c by speaker classification 22c is also referred to as the third feature vector; and the feature vector determined from audio data stream 16d by speaker classification 22d is also referred to as the fourth feature vector. The first feature vector is assigned to the first person and thus the first voice, the second feature vector to the second person and thus the second voice, the third feature vector to the third person and thus the third voice, and the fourth feature vector to the fourth person and thus the fourth voice, and vice versa.The individual feature of each feature vector is, for example, a specific characteristic of the person to whom the respective feature vector is assigned, and / or the individual feature of each feature vector is, for example, a specific characteristic of the voice to which the respective feature vector is assigned. The individual features of each feature vector can thus be, for example, the age of the person, the gender of the person, the language spoken by the person, the accent of the person, etc. By determining the respective feature vector, the person assigned to the respective feature vector and / or their voice is classified.

[0034] As indicated by arrows in the figure, the feature vectors, also referred to as voice feature vectors, are transmitted to a central processing unit 24, which receives them. This central processing unit 24 is, for example, a component of the electronic computing device 20 and is also referred to as a head unit or head unit control unit. The processing unit 24 is, in particular, an electronic computing unit. The processing unit 24 collects the feature vectors and transmits them, for example, wirelessly via a wireless data network 26, to another electronic computing device 28, which receives the feature vectors.The electronic computing device 28 is also referred to as the second electronic computing device and is not a component of the motor vehicle 10. It is an external device with respect to the motor vehicle 10, i.e., located outside of the motor vehicle 10. A data cloud 30, also referred to as a cloud, is shown particularly schematically in the figure and is, for example, a component of the data network 26. For example, the feature vectors are provided by the computing unit 24 and thereby stored, in particular temporarily stored, in the data cloud 30. The electronic computing device 28 is also referred to as the backend or logbook backend and receives, for example, the feature vectors from the data cloud 30.The backend (electronic computing unit 28) temporarily stores the received feature vectors until they can be assigned to a recognized trip of the motor vehicle 10 from which the received feature vectors originate. In particular, the logbook backend (electronic computing unit 28) performs the aforementioned automatic categorization of the trips. If, for example, the electronic computing unit 20 determines that the respective sound event comprises the respective human voice, the respective trip of the motor vehicle 10 is automatically classified, in particular by the electronic computing unit 28, into one of the predefined categories of the electronic logbook, depending on the respective voice detected.This means that the backend performs the automatic categorization of each detected trip, automatically assigning it to one of the categories based on the respective voice input. The automatic categorization of trips follows, for example, an extended set of rules that takes into account the characteristic vectors assigned to the trip. This means that the automatic categorization of each trip of motor vehicle 10 is based on the characteristic vectors and the rule set. The automatic categorization of each trip is performed by the electronic computing unit 28 and is illustrated in the figure by a block 32.

[0035] In the figure, block 34 illustrates a manual input by a person into a playback device, particularly an electrical or electronic one. Specifically, block 34 illustrates that the playback device captures the aforementioned input, for example, a manual input from the person. This input is a correction input received by the input device. The correction input is a trigger 36, shown schematically in the figure, which initiates a calibration 38 of the rule set. The calibration 38 of the rule set means that, depending on the correction input, the rule set is modified and thus calibrated. In the figure, block 40 illustrates machine learning based on the correction input, that is, based on manual correction inputs made by people.This includes, in particular, the following: The logbook backend processes correction entries, especially those made manually by users, which correct the automatic categorization of trips performed by the backend. This involves a change to the rule set, also known as calibration or recalibration, and specifically a modification of the processing of the characteristic vectors. For example, if a first trip of vehicle 10 is automatically categorized in a certain way based on the rule set, and a correction entry is subsequently made by a user that alters the automatic categorization of the first trip, the rule set is then changed so that, for example, a second trip of vehicle 10 is automatically categorized in a different way than the first.To calibrate the rule set, machine learning methods, for example, can be used. A subsystem for automatic categorization in the backend is designed such that the quality of the automatic categorization continuously improves through the consideration of feature vectors over a sequence of multiple rule set calibrations. Reference symbol list 10 motor vehicle 12a-d Seat 14 Microphone setup 16a-d Audio data stream 18a-d Speaker Recognition 20 electronic computing equipment 22a-d Speaker Classification 24 computing units 26 Data network 28 electronic computing equipment 30 Data cloud 32 blocks Block 34 36 triggers 38 Calibration 40 blocks

Claims

Method for automatically classifying journeys of a motor vehicle (10) into categories of an electronic logbook, in which: - during the respective journey of the motor vehicle (10), its interior is continuously monitored by means of at least one microphone device (14) of the motor vehicle and the microphone device (14) provides at least one continuous audio data stream (16a-d) which characterizes at least one sound event in the interior detected by the microphone device (14); - by means of an electronic computing device (20), the audio data stream (16a-d) is analyzed and it is thereby determined whether the sound event includes at least one human voice; - if it is determined that the sound event includes at least one human voice, the respective journey is automatically classified into one of the predefined categories of the electronic logbook depending on the voice detected;and- by ​​means of the electronic computing device (20) the determined voice is assigned to one of several seats (12a-d) arranged in the interior of the motor vehicle, whereby the respective journey is automatically classified into one of the predefined categories of the electronic logbook depending on the assigned seat.; Method according to claim 1, characterized in that, by means of the electronic computing device (20), at least one voice characteristic of the determined voice is identified by analyzing the audio data stream (16a-d) and on the basis of the identified voice, wherein the respective journey is automatically classified into one of the predefined categories of the electronic logbook depending on the identified voice characteristic. Method according to claim 1 or 2, characterized in that at least one personal characteristic of a person is determined by means of the electronic computing device (20) by analyzing the audio data stream (16a-d) and on the basis of the determined voice, wherein the respective journey is automatically classified into one of the predefined categories of the electronic logbook depending on the determined personal characteristic. The method according to claim 3, characterized in that the person characteristic comprises the person's age and / or gender and / or accent and / or language. Method according to one of the preceding claims, characterized in that at least one correction input (34) effected by a person is received by means of an input device, wherein at least one of the journeys is classified into one of the predefined categories of the electronic logbook depending on the received correction input (34). Method according to one of the preceding claims, characterized in that the respective journey is automatically classified into one of the predefined categories of the electronic logbook depending on the determined voice and on the basis of a set of rules. Method according to claims 5 and 6, characterized in that the rule set is changed depending on the correction input (34). Device for data processing, comprising means for carrying out the method according to one of the preceding claims. Computer program product comprising instructions which, when the computer program product is executed by a computer, cause the computer to execute the method according to any one of claims 1 to 7.

Images