Vehicular noise data storage method and noise data storage apparatus

Abstract

To grasp detail of a vehicular state when noise occurs, thereby, shortening the time required for taking a countermeasure against the noise.SOLUTION: A reference sound recording unit 1 manually records, as reference sound, sound felt by a user subjectively as a noise. A sound data acquisition unit 2 automatically records, as sound data, sound generated in a vehicle during driving of the vehicle. A vehicle signal acquisition unit 3 continuously acquires vehicle signals during driving of the vehicle. A similarity determination unit 4 compares the sound data acquired by the sound data acquisition unit 2 with the reference sound recorded by the reference sound recording unit 1 and determines a similarity between them. In a case where the similarity determination unit 4 determines that there is a similarity between the reference sound and the sound data, a sound-vehicle signal database 8 stores, in association with each other: the reference sound; the sound data determined to be similar to the reference sound; and the vehicle signal acquired by the vehicle signal acquisition unit 3 at a time when the sound data is acquired.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a method and device for storing abnormal sound data of a vehicle. [Background technology]

[0002] Patent Document 1 discloses a method for determining whether an abnormal sound is generated from a device consisting of multiple moving parts, such as an elevator. In this method, sound data including the sound generated from the device is acquired, and any sensing data related to the device, such as temperature, air pressure, or humidity, is acquired by each sensor. Then, a reference sound is generated based on the sensing data, and this reference sound is compared with the sound data to determine whether the device is generating an abnormal sound. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2013-200143 A Summary of the Invention [Problem to be solved by the invention]

[0004] Patent Document 1 discloses determining whether an abnormal sound from an equipment is generated by comparing the sound data with a reference sound based on sensing data from a sensor, but does not take into consideration sounds that a human user subjectively perceives as abnormal vehicle sounds. [Means for solving the problem]

[0005] The present invention relates to a method for storing abnormal sound data of a vehicle, in which a sound that a vehicle user subjectively feels to be an abnormal sound of the vehicle is stored as a reference sound, sound data and a vehicle signal are acquired while the vehicle is being driven, and the acquired sound data is compared with the reference sound to obtain a similarity, and if the similarity is equal to or greater than a predetermined threshold, the sound data and the corresponding vehicle signal are stored as abnormal sound data of the vehicle. Effect of the Invention

[0006] According to the present invention, abnormal sound data including sound data and corresponding vehicle signals is automatically saved as data when a user subjectively feels that an abnormal sound is coming from the vehicle, making it possible to grasp the condition of the vehicle in detail when an abnormal sound occurs and shortening the time it takes to take measures against the abnormal sound. [Brief description of the drawings]

[0007] [Figure 1] 1 is a functional block diagram of a vehicle abnormal sound data storage device in a first embodiment. FIG. [Diagram 2] 4 is a flowchart showing a process flow in the first embodiment. [Diagram 3] FIG. 11 is a functional block diagram of a vehicle abnormal sound data storage device in a second embodiment. [Figure 4] 13 is a flowchart showing a process flow according to a second embodiment. [Diagram 5] FIG. 11 is a functional block diagram of a vehicle abnormal sound data storage device in a third embodiment. [Figure 6] FIG. 13 is an explanatory diagram showing the internal configuration of a reference sound generating unit according to a third embodiment. [Figure 7] 13 is a flowchart showing a process flow according to a third embodiment. [Figure 8] 10 is an explanatory diagram showing an example of a user interface used for generating a reference sound in a reference sound generating unit; FIG. [Figure 9] FIG. 13 is a functional block diagram of a vehicle abnormal sound data storage device in a fourth embodiment. [Figure 10] 13 is a flowchart showing a process flow according to a fourth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0008] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The entire abnormal sound data storage device for a vehicle of the embodiment is configured as a cloud system including a cloud server managed by, for example, an automobile company or a dealership to which the vehicle's related parties belong, in addition to a computer system installed in the vehicle itself. The vehicle is configured as a so-called connected car so that necessary information can be exchanged between the vehicle's user and the related parties. In addition, as a prerequisite for using the abnormal sound data storage device for a vehicle of the embodiment, a user is informed in advance, for example, at the time of purchasing a vehicle, so that the user can actively obtain a reference sound, which will be described later and serves as a criterion for determining whether an abnormal sound is generated when the user subjectively feels that the abnormal sound is generated, in order to share abnormal sound data and quickly respond to the occurrence of an abnormal sound.

[0009] 1 is a functional block diagram of an abnormal sound data storage device for a vehicle in the first embodiment. This abnormal sound data storage device includes a reference sound recording unit (reference sound acquisition unit) 1, a sound data acquisition unit 2, a vehicle signal acquisition unit 3, a similarity determination unit 4, an acquired data output unit 5, an automobile company database 6, a dealer database 7, and a sound / vehicle signal database 8. Here, the reference sound recording unit 1, the sound data acquisition unit 2, the vehicle signal acquisition unit 3, the similarity determination unit 4, and the acquired data output unit 5 constitute parts of a computer system provided in the vehicle.

[0010] The reference sound recording unit 1 manually records a sound that is bothersome to a human user while driving a vehicle, that is, a sound that the user subjectively feels is an abnormal sound of the vehicle, as a reference sound. The reference sound recording unit 1 has a sound collection device provided in the vehicle, for example, an in-vehicle microphone or a sound collection function of a drive recorder, and a trigger such as a button that allows the user to manually switch the start and stop of recording. The sound collection device may not be provided in the vehicle, but may be a sound collection function of a PC (personal computer) or a smartphone. In this embodiment, the sound recorded by the reference sound recording unit 1 is a sound that continues periodically for a relatively long time, for example, several seconds. The recorded sound is stored in the automobile company database 6, and is also stored in the dealer database 7 as necessary.

[0011] The sound data acquisition unit 2 automatically records sounds generated in the vehicle while the vehicle is being driven as sound data. This sound data is recorded at all times while the vehicle is being driven. The sound data acquisition unit 2 has a sound collection device provided in the vehicle, for example, an in-vehicle microphone or a sound collection function of a drive recorder. The sound collection device may not be provided in the vehicle, but may be a sound collection function of a PC (personal computer) or a smartphone. The sound collection device of the sound data acquisition unit 2 may be the same as the sound collection device of the reference sound recording unit 1.

[0012] The vehicle signal acquisition unit 3 constantly acquires vehicle signals while the vehicle is in operation. Here, the vehicle signal is a signal related to one or more of the operations or states of the engine, brake, motor, muffler, oil, connector, and belt. In other words, the vehicle signal is a signal related to one of the operations or states of the engine, etc., or a signal related to two or more of the operations or states of the engine, etc. When the vehicle signal is a signal related to the operation or state of the engine, the vehicle signal is, for example, the engine speed. Note that the engine speed is related to the vehicle speed and the accelerator opening, so the vehicle speed and the accelerator opening may be used as the vehicle signal. Also, when the vehicle signal is a signal related to the operation or state of the brake, the vehicle signal is, for example, the brake depression force.

[0013] The similarity determination unit 4 compares the reference sound recorded by the reference sound recording unit 1 with the sound data acquired by the sound data acquisition unit 2 to determine the similarity between them. Here, the similarity determination can be performed using a well-known method of visualizing sound data using a frequency spectrogram or mel spectrogram to determine the similarity, or a well-known method of determining the similarity of time-series data, such as cosine similarity, CCF (Cross-Correlation Function), or DTW (Dynamic Time Warping).

[0014] When using DTW as an example of a method for determining similarity, first, the two time series data (waveforms) for which you want to calculate the similarity are expressed as x = x1, x2, . . ., x m, y=y1,y2,···,y n Let us consider a grid in which each time series data is arranged vertically and horizontally. Point (i,j) in the grid corresponds to the alignment between element si and element tj. Note that element si is the set of x1, x2, . . . , x before it is arranged on the grid. m , while element tj corresponds to any one of y1, y2, . . . , y n The distance between two elements, element si and element tj, at point (i,j) in the grid is

[0015]

number

[0016] Then, the DTW is defined as follows:

[0017]

number

[0018] Here, "w" (warping path) represents the distance between each element of s and t, and the shortest distance is the DTW. Therefore, when the similarity between the reference sound recorded by the reference sound recording unit 1 and the sound data acquired by the sound data acquisition unit 2 is expressed using DTW, if the similarity is equal to or greater than a predetermined threshold, the two are determined to be similar. When determining the similarity, the similarity between the reference sound and each of the multiple sound data acquired by the sound data acquisition unit 2 is determined for each length of time equal to the arbitrary time (any arbitrary number of seconds in this embodiment) that the reference sound continues. In addition, the overlap indicating the time overlapping portion of the two sound data is set arbitrarily.

[0019] The acquired data output unit 5 outputs sound data determined to be similar to the reference sound from among sound data acquired every several seconds during which the reference sound continues for the purpose of judgment by the similarity judgment unit 4. The acquired data output unit 5 has, for example, a display and a speaker provided in the vehicle. The acquired data output unit 5 outputs information to the user of the vehicle by one or more methods from among reproducing the sound data, displaying a waveform, displaying a result of visualizing the sound data using a frequency spectrogram, etc. At the time of this output, in order to facilitate comparison between the sound data determined to be similar and the reference sound, it is also possible to alternately reproduce the reference sound and the sound data, or simultaneously display the waveform, frequency spectrogram, etc. of the reference sound and the sound data.

[0020] The automobile company database 6 is a database (storage unit) owned by a company or department responsible for analysis and countermeasures regarding abnormal noises at the automobile company to which the vehicle-related personnel belong, and is accessible by the personnel of the company or department. The automobile company database 6 stores the reference sounds recorded by the reference sound recording unit 1 and data on abnormal noises obtained during the design, testing, and development of the vehicle.

[0021] The dealer database 7 is a database owned by a company or department responsible for maintenance and countermeasures regarding abnormal noises at the automobile dealership to which the vehicle-related personnel belong, and can be accessed by personnel of the company or department. The dealer database 7 stores abnormal noise data and countermeasure flows when abnormal noises are reported or repair requests are received.

[0022] The sound / vehicle signal database 8 is a database (storage unit) that stores the reference sound, the sound data determined to be similar to the reference sound, and the vehicle signal acquired by the vehicle signal acquisition unit 3 at the time the sound data was acquired, in association with each other, when the similarity determination unit 4 determines that the reference sound and the sound data are similar. The sound / vehicle signal database 8 is provided outside the vehicle, for example, on a cloud server, and stores not only the reference sound, sound data, and vehicle signal information related to one vehicle in this embodiment, but also the reference sound, sound data, and vehicle signal information related to all of the various vehicles managed by the automobile company and the dealer. The sound / vehicle signal database 8 shares the associated information of the reference sound, sound data, and vehicle signal with the automobile company database 6 and the dealer database 7 as necessary.

[0023] FIG. 2 is a flowchart showing the flow of processing in the first embodiment.

[0024] First, in step S101, when a sound that bothers the user while driving the vehicle is heard, the sound inside the vehicle is manually recorded and used as a reference sound. In other words, a sound that the user subjectively feels to be an abnormal sound while driving the vehicle is saved as a reference sound.

[0025] Next, in step S102, the reference sound is stored in the automobile manufacturer database 6. At this time, the reference sound is also stored in the dealer database 7 as necessary.

[0026] Then, in step S103, the vehicle sound is recorded as sound data. As described above, this recording is always performed by the sound data acquisition unit 2 while the vehicle is being driven, so in reality, the recording is also performed during the execution of steps S101 and S102, and further continues after step S103.

[0027] Next, in step S104, vehicle signals are acquired, such as vehicle speed, engine speed, brake pedal force, accelerator opening, and the like.

[0028] After acquiring the vehicle signal, in step S105, the similarity between the sound data and the reference sound is calculated by a known method every several seconds during which the reference sound continues. In this embodiment, the similarity between the sound data and the reference sound is calculated by using the above-mentioned DTW (dynamic time warping method).

[0029] Then, the process proceeds to step S106 to determine whether the similarity is equal to or greater than a threshold value. If the similarity is smaller than the threshold value, the flow ends.

[0030] Moreover, if it is determined in step S106 that the similarity is equal to or greater than the threshold, the process proceeds to step S107, where the reference sound, the sound data, and the corresponding vehicle signal are stored in the sound / vehicle signal database 8. In other words, the reference sound, the sound data determined to be similar to the reference sound, and the vehicle signal acquired by the vehicle signal acquisition unit 3 at the time the sound data was acquired are linked together and stored in the sound / vehicle signal database 8.

[0031] Next, in step S108, the reference sound, the sound data, and the corresponding vehicle signal are sent to the acquired data output unit 5.

[0032] Then, in step S109, the sound data determined to be similar to the reference sound is output. When outputting, the sound data is played back at a timing selected by the user. The sound data may be output by visualizing the sound data using a frequency spectrogram or the like and displaying the visualized data on a display.

[0033] After the sound data is output, in step S110, the reference sound, the sound data, and the corresponding vehicle signal are stored in the automobile company database 6 and the dealer database 7 as necessary, and information including the linked reference sound, etc. is shared.

[0034] As described above, in the first embodiment, the reference sound recording unit 1 manually records a sound that the user subjectively feels as an abnormal sound of the vehicle as the reference sound. In addition, the sound data of the sound data acquisition unit 2 and the vehicle signal of the vehicle signal acquisition unit 3 are linked. The similarity determination unit 4 compares the reference sound recorded by the reference sound recording unit 1 with the sound data acquired by the sound data acquisition unit 2 to determine the similarity between them, and if the similarity is equal to or greater than a predetermined threshold, the sound data and the corresponding vehicle signal are stored in the sound / vehicle signal database 8 as abnormal sound data of the vehicle. Therefore, as data when the user subjectively feels that the vehicle is abnormal, the sound data and the corresponding vehicle signal, that is, the abnormal sound data including the linked sound data and the vehicle signal, are automatically stored in the sound / vehicle signal database 8. Therefore, when an abnormal sound occurs, the vehicle state can be grasped in detail based on the vehicle signal, and measures can be efficiently taken for the vehicle state, thereby shortening the time required for measures against the abnormal sound. Furthermore, abnormal sounds that can be subjectively sensed by a user may include small abnormal sounds that cannot be detected by a detection device such as a sensor, so that a wide range of abnormal sound data can be collected.

[0035] Furthermore, in this embodiment, the abnormal sound data is stored in an automobile company database 6 and a dealer database 7, which are databases accessible by persons involved with the vehicle. Therefore, when an abnormal sound occurs, the linked sound data and vehicle signal are quickly transmitted to persons involved with the vehicle who are in remote locations, shortening the phase in which the persons involved and the user come to the same understanding, and preventing erroneous understanding of the abnormal sound and its cause.

[0036] FIG. 3 is a functional block diagram of the vehicle abnormal sound data storage device according to the second embodiment.

[0037] The vehicle abnormal sound data storage device of the second embodiment is configured by eliminating the reference sound recording unit 1 from the abnormal sound data storage device of the first embodiment and adding a reference sound selection unit (reference sound acquisition unit) 9 and an abnormal sound sample database 10.

[0038] The reference sound selection unit 9 selects an arbitrary sample sound from an allophone sample database 10, which stores a plurality of sample sounds created in advance, based on the conditions of the sound that the user wants to use as the reference sound. When making this selection, the user listens to the playback sound of each sample sound and selects the sample sound to be used as the reference sound.

[0039] The abnormal sound sample database 10 is a database (storage unit) that stores sample sounds based on past abnormal sound data provided by the automobile manufacturer database 6 and the dealer database 7, as well as sample sounds obtained through experiments, evaluations, and the like. The multiple sample sounds in the abnormal sound sample database 10 are grouped by vehicle model and abnormal sound occurrence part, and also contain information on the frequency of occurrence or the frequency of reporting by general users. The multiple sample sounds are sorted based on the vehicle model, abnormal sound occurrence part, and frequency of occurrence or reporting, and are presented to the user when selecting a reference sound in the reference sound selection unit 9.

[0040] Fig. 4 is a flowchart showing the flow of processing in the second embodiment. Note that in Fig. 4, the same steps as in the first embodiment are given the same numbers.

[0041] First, in step S201, the data in the abnormal sound sample database 10 is grouped by vehicle type and abnormal sound generating portion. For example, when grouping the data by abnormal sound generating portion, the exhaust system, braking system, etc. are used as the abnormal sound generating portion.

[0042] Next, in step S202, the user inputs the vehicle model and the assumed part where the sound to be searched for as the reference sound occurs and selects a group. That is, the same group as the vehicle model and the abnormal sound occurrence part specified as conditions by the user is selected.

[0043] Then, in step S203, the sample sounds in the selected group are sorted in order of frequency of occurrence. Also, instead of sorting the sample sounds in order of frequency of occurrence, the sample sounds may be sorted in order of frequency of indications from general users.

[0044] Next, in step S204, the user listens to the sample sounds arranged in order in step S203, selects the sound that is similar to the sound that the user finds interesting, and sets it as the reference sound.

[0045] After the reference sound is selected, the reference sound is stored in the automobile manufacturer database 6 in step S102, and the sound of the vehicle is recorded as sound data in step S103. Note that this recording is performed by the sound data acquisition unit 2 at all times while the vehicle is being driven, so in reality, it is performed even while steps S201 to S102 are being executed, and further continues after step S103.

[0046] Then, in step S104, a vehicle signal is acquired, and in step S105, the similarity between the sound data and a reference sound is calculated using a known method every several seconds that the reference sound continues, and in step S106, it is determined whether the similarity is equal to or greater than a threshold value. If the similarity is smaller than the threshold value, the flow ends.

[0047] If it is determined in step S106 that the similarity is equal to or greater than the threshold, then in step S107, the reference sound, the sound data, and the corresponding vehicle signal are stored in the sound / vehicle signal database 8, and in step S108, the reference sound, the sound data, and the corresponding vehicle signal are sent to the acquired data output unit 5. Then, in step S109, the sound data determined to be similar to the reference sound is output, and in step S110, as necessary, the reference sound, the sound data, and the corresponding vehicle signal are stored in the automobile company database 6 and the dealer database 7, and information including the linked reference sound, etc. is shared.

[0048] As described above, in the second embodiment, the reference sound selection unit 9 selects an arbitrary sample sound from the abnormal sound sample database 10 in which a plurality of sample sounds created in advance are stored, based on the conditions of the sound that the user wants to use as the reference sound. When making this selection, the user listens to the playback sound of each sample sound and selects it as the reference sound. Therefore, the user can easily select a sound similar to the annoying sound from a plurality of sample sounds created in advance, without manually recording the annoying sound. In addition, the user does not necessarily need to complete the selection of the reference sound and the subsequent output of sound data within the vehicle, and can take flexible measures such as selecting the reference sound after arriving at a dealer, repair shop, etc.

[0049] Furthermore, in the second embodiment, the multiple sample sounds are sorted based on the vehicle type, the part where the abnormal sound occurs, and the frequency of occurrence or the frequency of notification, and are then presented to the user when selecting a reference sound in the reference sound selection unit 9. For this reason, the user can easily select a reference sound by actively searching for a sample sound with a high priority from among the multiple sample sounds based on the vehicle type, the part where the abnormal sound occurs, and the frequency of occurrence or the frequency of notification, and selecting any sample sound.

[0050] FIG. 5 is a functional block diagram of a vehicle abnormal sound data storage device according to the third embodiment.

[0051] The vehicle abnormal sound data storage device of the third embodiment is configured by eliminating the reference sound recording unit 1 from the abnormal sound data storage device of the first embodiment and adding a new reference sound generation unit (reference sound acquisition unit) 11.

[0052] The reference sound generating unit 11 manually generates a sound that bothers the user while driving the vehicle, that is, a sound that the user subjectively feels to be an abnormal sound of the vehicle.

[0053] As shown in FIG. 6, the reference sound generation unit 11 includes an onomatopoeia selection unit 11a, an onomatopoeia-car sound database (onomatopoeia database) 11b, a variable sound presence / absence selection unit 11c, a variable period selection unit 11d, a variable period adjustment unit 11e, a pitch (pitch) selection unit 11f, a pitch (pitch) adjustment unit 11g, a generated sound playback unit 11h, and a user-generated sound database 11i.

[0054] The onomatopoeia selection unit 11a selects a sound (hereinafter referred to as a "base sound") that will be the base for generating a reference sound from the onomatopoeia-car sound database 11b. More specifically, the onomatopoeia selection unit 11a acquires a sample sound and an onomatopoeia (an onomatopoeic word) that corresponds to the sample sound from the onomatopoeia-car sound database 11b, presents them to the user, and sets the sample sound that corresponds to the onomatopoeia selected by the user as the base sound. When selecting an onomatopoeia, the onomatopoeia selection unit 11a asks the user to select the onomatopoeia that is most similar to the sound the user wants to generate from a plurality of onomatopoeias. All onomatopoeias are written in Japanese kana, such as "katakata" and "gatagata."

[0055] The onomatopoeia-car sound database 11b is a database (storage unit) used by the onomatopoeia selection unit 11a to select a base sound. The onomatopoeia-car sound database 11b is a database that stores in advance sample sounds related to sounds that often occur in vehicles and onomatopoeias (text information that onomatopoeically translates the sample sounds) that correspond to the sample sounds. Note that one sample sound does not necessarily correspond to one onomatopoeia, and may correspond to multiple onomatopoeias.

[0056] The variable sound presence / absence selection unit 11c asks the user to select whether the sound to be generated has a sense of fluctuation or not. The variable sound presence / absence selection unit 11c asks the user to select whether the sound to be generated is a stationary sound (a sound that continues) or a variable sound (a periodic sound).

[0057] When the user answers that there is a sense of fluctuation in the fluctuation sound presence / absence selection unit 11c, the fluctuation period selection unit 11d asks the user to select how many times the fluctuation period of the sound to be generated is. The fluctuation period is adjusted by decreasing or increasing the period of the base sound as a reference (1 time), for example, to 0.8 times, 0.9 times, 1.1 times, 1.2 times, etc., so that the user can select an arbitrary fluctuation period.

[0058] The fluctuation period adjustment unit 11e reflects the fluctuation period selected by the user in the fluctuation period selection unit 11d to the bass note selected by the user in the onomatopoeia selection unit 11a. The fluctuation period is adjusted by expanding or contracting the time axis of the sound data. For example, if the magnification of the fluctuation period specified by the user is 2 times, the original data is contracted by half to adjust the fluctuation period. Also, if the magnification of the fluctuation period specified by the user is 0.5 times, the original data is stretched by 2 times to adjust the fluctuation period. When adjusting the fluctuation period, simply expanding or contracting the time axis of the sound data will also change the pitch of the sound, so the pitch of the sound is returned to the original pitch by using a well-known pitch shift method or the like.

[0059] The pitch (pitch) selection unit 11f asks the user to select the pitch (pitch) of the sound that he or she wishes to generate.

[0060] The pitch (height) adjustment unit 11g reflects the pitch of the sound selected by the user in the pitch (height) selection unit 11f in the bass sound after the adjustment of the fluctuation period. When reflecting the pitch, the pitch of the bass sound is changed to a pitch of the user's choice by using a well-known pitch shifting method or the like.

[0061] The generated sound reproducing unit 11h outputs a sound (hereinafter referred to as "generated sound") generated by performing processing on the bass sound selected by the onomatopoeia selecting unit 11a in the fluctuation period adjusting unit 11e and the pitch (height) adjusting unit 11g. The generated sound reproducing unit 11h outputs information about the sound to the user by one or more methods of reproducing the sound, displaying a waveform, visualizing and displaying the sound using a frequency spectrogram or the like, and the like. When outputting, in order to facilitate comparison between the generated sound and the bass sound, the generated sound and the bass sound may be reproduced alternately, or the waveforms, frequency spectrograms, etc. of the generated sound and the bass sound may be simultaneously displayed.

[0062] The user-generated sound database 11i is a database (storage unit) that stores generated sounds. In addition to the generated sounds, the user-generated sound database 11i may store data on the bass sound, a history of adjustments to the fluctuation period and pitch (height), and the like.

[0063] Fig. 7 is a flowchart showing the flow of processing in Example 3. In Fig. 7, the same steps as in Example 1 are denoted by the same numbers.

[0064] First, in step S301, an onomatopoeia that is most similar to the sound that the user wants to generate is selected from the onomatopoeia-car sound database 11b. In this selection, the user selects the onomatopoeia that is most similar to the sound that the user wants to generate from a plurality of onomatopoeias stored in the onomatopoeia-car sound database 11b in response to a question posed by the onomatopoeia selection unit 11a.

[0065] Next, in step S302, the user selects whether or not the sound to be generated has a sense of fluctuation. When making this selection, the user selects whether or not there is a sense of fluctuation in response to a question posed by the fluctuation sound presence / absence selection unit 11c. If there is no sense of fluctuation, the process proceeds to step 305.

[0066] If the user selects that there is a sense of fluctuation in step S302, the user selects the fluctuation period in step S303. In this selection, the user selects how many times the fluctuation period of the sound to be generated is in response to a question posed by the fluctuation period selection unit 11d.

[0067] Next, in step S304, the fluctuation period is changed to generate the sound. The fluctuation period is changed by expanding or contracting the time axis of the sound data, as described above.

[0068] Then, in step S305, the user selects the pitch of the sound. When making this selection, the user selects the pitch of the sound in response to a question posed in the pitch (pitch) selection section 11f.

[0069] After the pitch of the sound is selected, in step S306, the pitch of the sound whose fluctuation period has been changed in step S304 is changed based on the pitch of the sound selected in step S305, and a sound is generated. At this time, the generated sound becomes a generated sound.

[0070] Next, in step S307, the generated sound is output. When outputting, as described above, one or more of the following methods are used: playing back the sound data, displaying a waveform, displaying a result of visualizing the sound data using a frequency spectrogram or the like.

[0071] Then, in step S308, it is determined whether the generated sound is appropriate. In other words, it is determined whether the generated sound corresponds to the sound desired to be generated (the sound desired to be generated at the time of step S301). This determination is made by the user based on one or more of the above methods used in step S307. If the generated sound is not appropriate, the process returns to step S303 and the bass sound is corrected again.

[0072] If it is determined in step S308 that the generated sound is appropriate, the process proceeds to step S309, where the generated sound is stored in the user-generated sound database 11i and is used as a reference sound.

[0073] Then, in step S102, the reference sound is stored in the automobile manufacturer database 6, and in step S103, the sound of the vehicle is recorded as sound data. As described above, this recording is performed by the sound data acquisition unit 2 at all times while the vehicle is being driven, so in reality, it is performed even during the execution of steps S301 to S309, and further continues after step S103.

[0074] Next, in step S104, a vehicle signal is acquired, and in step S105, the similarity between the sound data and a reference sound is calculated using a known method every several seconds that the reference sound continues, and in step S106, it is determined whether the similarity is equal to or greater than a threshold value. If the similarity is smaller than the threshold value, the flow ends.

[0075] If it is determined in step S106 that the similarity is equal to or greater than the threshold, then in step S107, the reference sound, the sound data, and the corresponding vehicle signal are stored in the sound / vehicle signal database 8, and in step S108, the reference sound, the sound data, and the corresponding vehicle signal are sent to the acquired data output unit 5. Then, in step S109, the sound data determined to be similar to the reference sound is output, and in step S110, as necessary, the reference sound, the sound data, and the corresponding vehicle signal are stored in the automobile company database 6 and the dealer database 7, and information including the linked reference sound, etc. is shared.

[0076] Next, an example of a user interface used for generating a reference sound in the reference sound generating unit 11 will be described with reference to FIG.

[0077] First, on screen A, the process performed by onomatopoeia selection unit 11a involves the user answering question Q1, "What kind of sound was made?" In this example, the user selects the onomatopoeia "gatagata" by, for example, touching screen A with a finger. When making this selection, a sample sound corresponding to "gatagata" is played, and by listening to this sample sound, the user can easily select the base sound. After selecting the onomatopoeia, the user moves to screen B by touching "next" button b1 with a finger.

[0078] Next, on screen B, as a process implemented by the fluctuating sound presence / absence selection unit 11c, the user answers the question Q2, "Was it a steady sound (a sound that continues continuously) or was it a fluctuating sound (a periodic sound)?" In this example, the user selects "fluctuating sound" by touching screen B with his finger. After selecting fluctuating sound, the user moves to screen C by pressing the "Next" button b2.

[0079] Then, on screen C, as a process performed by the fluctuation period selection unit 11d, the user answers the question Q3 "What kind of period was it?". In this example, the user selects the fluctuation period by moving the selection bar 12 (shown as a gray rectangle on screen C) provided on screen C up and down with his finger to adjust it. When selecting this fluctuation period, a sound in which the fluctuation period is reflected in the base sound is played every time the selection bar 12 is moved and stopped, and the user adjusts the fluctuation period while listening to this sound. In this example, a "rattling" sound is played with the fluctuation period adjusted. After selecting the fluctuation period, the user moves to screen D by pressing the "Next" button b3.

[0080] Next, on screen D, as a process performed by the pitch (pitch) selection unit 11f, the user answers the question Q4, "What kind of pitch was it?". In this example, the user selects the fluctuation period by moving the selection bar 13 (shown as a gray rectangle on screen D) on screen D up and down with a finger. When making this selection, every time the selection bar 13 is moved and stopped, a sound in which the pitch of the sound after the fluctuation of the fluctuation period is reflected is played, and the user selects the pitch of the sound while listening to this sound. In addition, in the area to the right of the movable range of the selection bar 13, information such as "the sound of an airplane" is displayed as an example to show which pitch of sounds existing in daily life corresponds to the pitch of the selected sound. After selecting the pitch of the sound, the user moves to screen E by pressing the "Next" button b4.

[0081] Then, on screen E, the user answers the question Q, "Was the generated sound like this?" as a process implemented by generated sound replay unit 11h. When answering, the generated sound is replayed. If the generated sound corresponds to the generated sound, the user saves the generated sound by pressing "Save" button b5 as a process implemented by user-generated sound database 11i. If the generated sound does not correspond to the generated sound, the user returns to screen C, for example, by pressing "Back" button b6, and corrects the bass sound again. When correcting the bass sound, the user may return to any of screens A, B, and D, instead of screen C.

[0082] As described above, in the third embodiment, the onomatopoeia-car sound database 11b stores sample sounds related to sounds that often occur in vehicles and onomatopoeias corresponding to the sample sounds in advance. Then, the user selects one of the onomatopoeias and modifies the sample sound corresponding to the selected onomatopoeia to generate a reference sound. Therefore, the user can intuitively select a sound close to the sound of concern using the onomatopoeia, and further modify the sense of variation, pitch, etc. of the sample sound corresponding to the onomatopoeia as appropriate, thereby easily generating the reference sound by the user himself without manually recording the reference sound. Also, as in the second embodiment, in the third embodiment, the user does not necessarily need to complete the generation of the reference sound and the subsequent output of sound data within the vehicle, and can take flexible measures such as selecting the reference sound after arriving at a dealer, repair shop, etc.

[0083] FIG. 9 is a functional block diagram of a vehicle abnormal sound data storage device according to the fourth embodiment.

[0084] The abnormal sound data storage device of the vehicle of the fourth embodiment is configured by abolishing the reference sound recording unit 1 and the acquired data output unit 5 from the abnormal sound data storage device of the first embodiment, and newly adding an abnormal sound level / abnormal sound cause registration unit 14 and an abnormal sound level / estimated cause display unit 15. Furthermore, in the fourth embodiment, abnormal sound data including sound data and vehicle signals acquired in a linked state in the first to third embodiments are accumulated in the sound / vehicle signal database 8, and this abnormal sound data is used for a user other than the user of the first to third embodiments and the vehicle of this other user (a vehicle other than the vehicle of the first to third embodiments). For this reason, the sound data acquisition unit 2 acquires sound data related to the vehicle of the other user. Furthermore, in the fourth embodiment, the similarity determination unit 4 determines the similarity between the sound data of the vehicle acquired from the sound data acquisition unit 2 and the sound data (sound data similar to the abnormal sound) in the abnormal sound data accumulated in the sound / vehicle signal database 8. The method used for this determination is the same as that in the first to third embodiments. It may take a long time to determine the similarity of all sound data in the sound / vehicle signal database 8. However, the determination time can be shortened by arranging the sound data in order of occurrence frequency or frequency of user indication and comparing the sound data in order of frequency.

[0085] The abnormal sound level / abnormal sound cause registration unit 14 registers each abnormal sound in the abnormal sound data based on the first to third embodiments and the corresponding repair history using the automobile company database 6 and the dealer database 7 through the sound / vehicle signal database 8. Each abnormal sound and the corresponding repair history are used to classify a plurality of abnormal sound levels (including repair and replacement responses) for abnormal sounds of vehicles different from the vehicles of the first to third embodiments. Various abnormal sound levels can be cited, for example, "replacement of large parts such as engine required," "replacement of small parts required," "simple repair possible," "not an abnormal sound," etc.

[0086] When the similarity determination unit 4 determines that the sound data acquired from the sound data acquisition unit 2 is similar to the sound data in the abnormal sound data acquired from the sound / vehicle signal database 8, the abnormal sound level / estimated cause display unit 15 presents the sound data determined to be similar and the abnormal sound level accompanying this sound data to a user of a vehicle other than the vehicles of the first to third embodiments.

[0087] Fig. 10 is a flowchart showing the flow of processing in the fourth embodiment. Note that in Fig. 10, the same steps as in the first embodiment are given the same numbers.

[0088] First, in step S401, the sound of a vehicle other than the vehicles in the first to third embodiments is recorded.

[0089] Next, in step S402, vehicle signals of the other vehicle are acquired, such as vehicle speed, engine speed, brake pedal force, accelerator opening, and the like.

[0090] Then, in step S403, the vehicle sound data acquired from the sound data acquisition unit 2 is judged for its similarity to sound data in the sound / vehicle signal database 8 that is similar to the abnormal sound.

[0091] Then, in step S106, it is determined whether the similarity is equal to or greater than a threshold value. If the similarity is smaller than the threshold value, the flow ends.

[0092] If it is determined in step S106 that the similarity is equal to or greater than the threshold value, the process proceeds to step S404, where the sound data and the abnormal sound level are sent to the abnormal sound level / abnormal sound cause registration unit 14, which is a display device.

[0093] Next, in step S405, the sound data and the abnormal sound level are output to a user different from those in the first to third embodiments. At the time of this output, sound data in the abnormal sound data acquired from the sound / vehicle signal database 8 is reproduced, and the abnormal sound level, for example, "Large parts such as the engine need to be replaced", etc. are displayed.

[0094] As described above, in the fourth embodiment, a plurality of abnormal sound data including sound data and vehicle signals acquired in a linked state in the first to third embodiments are accumulated in the sound and vehicle signal database 8. Then, the abnormal sound level and abnormal sound cause registration unit 14 registers each abnormal sound in the abnormal sound data and the corresponding repair history, and each abnormal sound and the corresponding repair history are used to classify a plurality of abnormal sound levels for abnormal sounds of a vehicle different from the vehicle of the first to third embodiments. Then, when the similarity determination unit 4 determines that the abnormal sounds are similar, the abnormal sound level and estimated cause display unit 15 presents the sound data determined to be similar and the abnormal sound level associated with this sound data to a user of a vehicle different from the vehicle of the first to third embodiments. In this way, by providing the abnormal sound data of the first to third embodiments to a user different from the first to third embodiments, this user can appropriately deal with the abnormal sound of his or her vehicle. [Explanation of symbols]

[0095] 1. Reference tone recording section 2. Sound data acquisition section 3. Vehicle signal acquisition unit 4...Similarity determination section 5. Acquired data output section 6. Automobile Company Database 7. Dealer database 8. Sound and vehicle signal database 9. Reference tone selection section 10. Abnormal noise sample database 11...Reference sound generation section 11a Onomatopoeia selection section 11b Onomatopoeia - Car Sound Database 11c...Fluctuation sound on / off selection section 11d... Variable period selection section 11e Fluctuation period adjustment section 11f Pitch (height) selection section 11g Pitch (height) adjustment part 11h...Generated sound playback section 11i User-Generated Sound Database 14. Abnormal noise level and abnormal noise cause registration section 15 Abnormal noise level / probable cause display section

Claims

1. A sound that a user of the vehicle subjectively feels to be an abnormal sound of the vehicle is stored as a reference sound; Acquiring sound data and vehicle signals while driving the vehicle, and associating the sound data and vehicle signals with each other; The acquired sound data is compared with the reference sound to determine a similarity. If the degree of similarity is equal to or greater than a predetermined threshold, the sound data and the corresponding vehicle signal are stored as abnormal sound data of the vehicle. How to save vehicle abnormal noise data.

2. 2. The method for storing abnormal sound data of a vehicle according to claim 1, wherein the abnormal sound data is stored in a database accessible to persons related to the vehicle.

3. 2. The method of claim 1, wherein the vehicle signal is a signal related to the operation or state of one or more of an engine, a brake, a motor, a muffler, oil, a connector, and a belt.

4. 2. The method for storing abnormal sound data of a vehicle according to claim 1, further comprising: storing a plurality of sample sounds created in advance in an abnormal sound sample database; and the user listening to a playback sound of each sample sound and selecting it as the reference sound.

5. The plurality of sample sounds in the abnormal sound sample database are grouped by vehicle type and abnormal sound generating portion, and include information on the frequency of occurrence or the frequency of abnormal sound being pointed out by general users, 5. The method for storing abnormal sound data of a vehicle according to claim 4, further comprising: sorting the plurality of sample sounds based on the vehicle model, the abnormal sound occurrence location, and the occurrence frequency or the reporting frequency, and then presenting the sample sounds to the user.

6. A plurality of sample sounds and onomatopoeias corresponding to each sample sound are stored in an onomatopoeia database in advance, 2. The method for storing abnormal sound data of a vehicle according to claim 1, further comprising: generating the reference sound by modifying a sample sound corresponding to the selected onomatopoeia, while the user selects any one of the onomatopoeias.

7. A plurality of abnormal sound data are stored, Registering each abnormal sound in the abnormal sound data and the corresponding repair record, 2. The method for storing abnormal noise data for a vehicle according to claim 1, further comprising classifying abnormal noises of another vehicle different from the said vehicle into a plurality of levels based on each abnormal noise and the corresponding repair history, and presenting the classification to a user of the other vehicle.

8. a reference sound acquisition unit that acquires a reference sound that a user of the vehicle subjectively feels to be an abnormal sound of the vehicle; a sound data acquisition unit that acquires sound data related to the vehicle while the vehicle is being driven; a vehicle signal acquisition unit that acquires a vehicle signal associated with the sound data; a similarity determination unit that compares the acquired sound data with a reference sound to determine a similarity; a storage unit that stores the sound data and a corresponding vehicle signal as abnormal sound data of the vehicle when the degree of similarity is equal to or greater than a predetermined threshold value; The abnormal noise data storage device for a vehicle is provided with the abnormal noise data storage device.

Images