JP2007107902A5 - - Google Patents

Description

Acoustic monitoring support device and acoustic monitoring support method

  The present invention relates to an acoustic monitoring support apparatus and an acoustic monitoring support method used when monitoring the state of a device based on sound generated from the device.

  In the engine room of the ship, many generators and auxiliary machines including the main engine are operated, and engine personnel are required to perform maintenance work on these devices. Here, the abnormal sound of the device is an indispensable information source in monitoring the device. If the engineer can recognize the abnormal sound at an early stage, it is possible to avoid an engine failure and a marine accident beforehand. Therefore, it is very important to monitor the equipment by the sensory organs (hearing) of engine personnel. In other words, the engine staff performs equipment maintenance by monitoring whether or not abnormal noise is generated from the equipment. As described above, it is generally performed in various fields to determine whether or not a device is good by detecting abnormal sounds generated from the device during operation.

For example, in a manufacturing process of an engine for a vehicle such as an automobile, a test operation of the engine is performed, and the engine sound at that time is determined by human hearing, so that the determination of engine pass / fail may be performed (for example, , See Patent Document 1).
JP-A-8-278191

  However, since the noise level in the engine room of a ship is very high, it is often difficult for engine personnel to hear abnormal sounds of equipment. Here, it is known that in a situation where the noise level is high, a masking phenomenon occurs in which a target sound (for example, an abnormal sound of a device) is difficult to hear due to noise. Therefore, it is necessary to reduce the influence of this masking phenomenon in order to reliably hear the abnormal sound of the device even under a high noise level.

Means for Solving the Problems and Effects of the Invention

  Here, there are two cases for the masking phenomenon in which the target sound is difficult to hear due to the interference sound. First, the first case is a case where it is difficult to hear the target sound generated at the time when the disturbing sound is generated. And the 2nd case is a case where it becomes difficult to hear the target sound generated before and after the time when the disturbing sound is generated. The masking phenomenon at this time is called the successive masking phenomenon, and the forward masking phenomenon that makes it difficult to hear the target sound after the interfering sound, and the backward masking phenomenon that makes it difficult to hear the target sound before the interfering sound. There is.

  Therefore, the present inventor has conducted research to solve the above problems, and as a result, in the first and second cases, the sound pressure level of the interfering sound is reduced or the interfering sound is deleted. As a result, it has been found that the time at which the interference sound is generated and the target sound generated before and after the interference sound can be easily heard.

  The acoustic monitoring support apparatus according to the present invention includes a data acquisition unit that acquires sound waveform data, a data division unit that divides sound waveform data obtained by the data acquisition unit into a plurality of divided waveform data, and the data division Frequency analysis means for analyzing the sound pressure level for each frequency for each of the plurality of divided waveform data divided by the means, and each of the plurality of divided waveform data is reduced based on the frequency analysis result of the frequency analysis means. Determining means for determining whether or not the data corresponding to the disturbing sound that is the sound processing target is included, and the divided waveform that is determined by the determining means to include the data corresponding to the disturbing sound that is the target of sound reduction processing A sound reduction processing means for reducing at least a part of the sound pressure level for each frequency included in the data, and an interference sound to be reduced by the determination means. The divided waveform data determined not to include the processed data, and the determination unit determines that the data corresponding to the interference sound to be reduced is included, and the sound reduction processing by the sound reduction processing unit is performed. Data connecting means for connecting the divided waveform data after being performed, and output means for outputting a sound corresponding to the waveform data to which a plurality of divided waveform data are connected by the data connecting means to the outside.

  The acoustic monitoring support method of the present invention includes a data acquisition step of acquiring sound waveform data, a data division step of dividing the sound waveform data obtained in the data acquisition step into a plurality of divided waveform data, and the data division A frequency analysis step for analyzing the sound pressure level for each frequency for each of the plurality of divided waveform data divided in the step, and each of the plurality of divided waveform data is reduced based on the frequency analysis result in the frequency analysis step. A determination step for determining whether or not the data corresponding to the disturbing sound to be processed includes sound, and a divided waveform that is determined to include the data corresponding to the disturbing sound to be reduced in the determining step A sound reduction processing step for reducing at least a part of a sound pressure level for each frequency included in the data; and the determination step. The divided waveform data determined not to include the data corresponding to the interference sound to be reduced in the process and the data corresponding to the interference sound to be reduced in the determination step are determined to be included. And a data connection step for connecting the divided waveform data after the sound reduction processing in the sound reduction processing step is performed, and a sound corresponding to the waveform data to which the plurality of divided waveform data are connected by the data connection step. And an output step for outputting to the outside.

  In this specification, “interfering sound subject to sound reduction processing” means an interfering sound having a sound pressure level higher than a set sound pressure level corresponding to a magnitude that makes it difficult to hear the target sound due to the influence of the masking phenomenon. Yes.

According to this configuration, the sound reduction processing is performed on the divided waveform data including data corresponding to the interference sound interference sound to be subjected to the sound reduction processing among the plurality of divided waveform data. Thereafter, the divided waveform data subjected to the sound reduction processing is connected to other divided waveform data, and a sound corresponding to the connected waveform data is output to the outside. For this reason, the portion of the sound output from the sound monitoring support device that includes the disturbing sound that has undergone the sound reduction processing is lower than the original sound pressure level. Accordingly, since the influence of the masking phenomenon due to the interference sound is reduced, the target sound such as an abnormal sound of the device can be easily heard.

  In the acoustic monitoring support device of the present invention, it is used to determine whether or not the divided waveform data includes data corresponding to the disturbing sound to be reduced based on the frequency analysis result of the frequency analysis means. A representative value calculating means for calculating a representative value for each divided waveform data, and a comparing means for comparing the representative value calculated by the representative value calculating means with a threshold value. If the comparison result that the representative value of the divided waveform data is larger than the threshold value is obtained, it may be determined that the divided waveform data includes data corresponding to the disturbing sound to be reduced.

  According to this configuration, since the determination is only a comparison between the representative value and the threshold value, if only the representative value is calculated, it is a binary comparison, and whether the divided waveform data includes data corresponding to the interference sound to be reduced. It can be easily determined for each divided waveform data. Moreover, since the representative value and the threshold value are compared, it is suitable for determining whether or not the disturbing sound to be subjected to the sound reduction process is emitted over a wide frequency range.

  In the acoustic monitoring support device of the present invention, the sound reduction processing unit is configured such that when the determination unit determines that the divided waveform data includes data corresponding to the disturbing sound to be reduced, the divided waveform You may reduce the sound pressure level of all the frequencies contained in data.

  According to this configuration, it is possible to easily perform the sound reduction process as compared with the case of reducing the sound pressure level of some frequencies. In addition, the processing time of the program for performing the sound reduction process can be shortened.

  The acoustic monitoring support device of the present invention further includes a comparison unit that compares a sound pressure level for each frequency in the divided waveform data with a threshold value, and the determination unit determines the sound pressure level for each frequency in the comparison unit. If the comparison result that at least one is larger than the threshold value is obtained, it may be determined that the divided waveform data includes data corresponding to the disturbing sound to be reduced.

  According to this configuration, since the sound pressure level and the corresponding threshold value are compared for each frequency, it is possible to accurately perform the sound reduction process for the disturbing sound generated in the frequency band of the local range. .

  In the acoustic monitoring support device of the present invention, the sound reduction processing unit is configured such that when the determination unit determines that the divided waveform data includes data corresponding to the disturbing sound to be reduced, the divided waveform You may reduce the sound pressure level of all the frequencies contained in data.

  According to this configuration, it is possible to easily perform the sound reduction process as compared with the case of reducing the sound pressure level of some frequencies. In addition, it is possible to reduce temporal discontinuity in the analysis result when performing frequency analysis, and to reduce missing information.

  In the acoustic monitoring support device of the present invention, the sound reduction processing unit is configured such that when the determination unit determines that the divided waveform data includes data corresponding to the disturbing sound to be reduced, the divided waveform You may reduce only the sound pressure level of the frequency which has a sound pressure level larger than the threshold value contained in data.

  According to this configuration, it is possible to reduce the target sound while reducing the interfering sound. Therefore, since the sound pressure level of the target sound generated at the time when the disturbing sound is generated cannot be lowered, it becomes possible to hear the target sound generated at that time. Further, compared to the case where the sound pressure levels of all the frequencies included in the divided waveform data are lowered, information loss is reduced.

  Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a schematic configuration of an acoustic monitoring support apparatus according to the first embodiment of the present invention.

  As shown in FIG. 1, the acoustic monitoring support device 1 includes a main unit 10, and an AD converter 20, a DA converter 30, and an operation unit 40 are connected to the main unit 10. Yes. A microphone 21 is connected to the AD converter 20, and an acoustic amplifier 31 and an acoustic speaker 32 are sequentially connected to the DA converter 30. In the present embodiment, since the acoustic monitoring support apparatus 1 used when monitoring the state of the equipment in the engine room of the ship will be described, the microphone 21 is arranged around the equipment to be monitored.

  The main unit 10 includes a data storage unit 11, a data division unit 12, a frequency analysis unit 13, a representative value calculation unit 14, a threshold storage unit 15, a determination unit 16, a sound reduction processing unit 17, and an improvement. A processing unit 18 and a waveform connecting unit 19 are formed.

  The data storage unit 11 stores digital sound data corresponding to the analog sound data acquired by the microphone 21. That is, the digital sound data stored in the data storage unit 11 is obtained by digitizing analog sound data acquired by the microphone 21 by the AD converter 20. In the present embodiment, sampling data whose sampling time per point is 0.0000195 seconds is stored in the data storage unit 11 as waveform data. Here, FIG. 3 shows an example of the waveform data stored in the data storage unit 11.

  The data dividing unit 12 divides the waveform data stored in the data storage unit 11 into a plurality of divided waveform data having a length necessary for frequency analysis. The data dividing unit 12 divides the data into a plurality of divided waveform data at predetermined time intervals (for example, every 0.1 second). Here, two consecutive divided waveform data among the plurality of divided waveform data respectively include waveform data at the same time (overlapping). FIG. 3 shows respective time widths D1 to Dn when the waveform data is divided into a plurality of divided waveform data. That is, the waveform data is divided into n pieces of divided waveform data corresponding to the times D1, D2, D3, D4,.

  In this embodiment, the time widths D1 to Dn are all 0.1 seconds, and the data dividing unit 12 divides the data into sampling data every 0.1 seconds, that is, divided waveform data every 512 points. Here, the sampling data for every 512 points is the number of points when the FFT is executed by this apparatus. When the FFT resolution is increased, the point is increased, and when the FFT resolution is not so much required, the point is decreased. Also good.

  The frequency analysis unit 13 performs frequency analysis for analyzing the sound pressure level for each frequency with respect to each of the plurality of divided waveform data divided by the data dividing unit 12. In the present embodiment, a window function is used for frequency analysis, and sampling data of divided waveform data every predetermined time, that is, sampling data of 256 points at the center of sampling data of 512 points every 0.1 second. Frequency analysis is performed. Here, the sampling data of 256 points corresponds to data per 0.05 second. In FIG. 3, frequency analysis is performed on n pieces of divided waveform data corresponding to each of the times D1, D2, D3, D4,..., Dn. As shown in the figure, the waveform data subjected to the data processing from the frequency analysis result is connected as data of the central portions d1, d2, d3,..., Dn of each divided waveform data and output as sound. become. In addition, the frequency analysis unit 13 stores data indicating the sound pressure level for each frequency obtained by frequency analysis (spectrum data corresponding to each frequency) for each divided waveform data.

  FIGS. 4A to 4C show the frequency analysis results regarding the divided waveform data corresponding to the times D2 to D4. In FIG. 4, the horizontal axis indicates the frequency X, the vertical axis indicates the sound pressure level of each frequency, and the solid line in FIG. 4 indicates the frequency analysis result. As will be described later, the broken lines in FIG. 4 indicate the average value of the sound pressure levels for each frequency regarding a plurality of extracted waveform data extracted from a plurality of divided waveform data.

  The representative value calculation unit 14 calculates, for each divided waveform data, a representative value used when determining whether or not the plurality of divided waveform data includes waveform data corresponding to the disturbing sound to be subjected to the sound reduction process. The representative value is a value indicating how far the sound pressure at a certain moment is away from the average value.

  Here, the procedure for calculating the representative value in the representative value calculating unit 14 will be described with reference to FIG. 5, taking the divided waveform data corresponding to the time D3 as an example. FIG. 5 is the same as FIG. 4B and shows the frequency analysis result regarding the divided waveform data corresponding to time D3, and is a diagram for explaining the procedure for calculating the representative value of each divided waveform data. First, the representative value calculation unit 14 calculates an average value of sound pressure levels for each frequency related to a plurality of divided waveform data based on the frequency analysis result in the frequency analysis unit 13. In the following description, since the frequency is indicated by the variable Xn on the frequency axis, the average value of the sound pressure levels for each variable Xn is calculated.

When the sound pressure level in a certain variable Xn is Fs (Xn) and the average value of the sound pressure level in the variable Xn is Fa (Xn), as shown in (Equation 1), the sound in a certain variable Xn A value S is calculated by subtracting the average value Fa (Xn) from the pressure level Fs (Xn) and adding the value from X1 to Xz. Next, when the value S calculated above is divided by the number of variables Xn as shown in (Expression 2), a representative value Sf is calculated as a difference per variable.

  The threshold value storage unit 15 stores a threshold value used when determining whether or not the plurality of divided waveform data includes waveform data corresponding to the disturbing sound to be subjected to the sound reduction process. In the present embodiment, since it is determined whether or not the waveform data corresponding to the disturbing sound to be reduced is included based on the magnitude of the representative value calculated by the representative value calculation unit 14, the threshold value is stored. The unit 15 stores a threshold value related to the representative value. Here, the threshold value stored in the threshold value storage unit 15 can be changed by operating the operation unit 40, and can be set as appropriate within a range of, for example, −10 dB to +20 dB. In the present embodiment, the threshold storage unit 15 stores +10 dB as a threshold.

  The determination unit 16 includes a comparison unit 16a, and the comparison unit 16a compares the representative value Sf related to each of the plurality of divided waveform data with the threshold value stored in the threshold value storage unit 15. Then, the determination unit 16 determines whether or not the plurality of divided waveform data includes waveform data corresponding to the disturbing sound to be subjected to the sound reduction process, based on the comparison result in the comparison unit 16a. That is, when the comparison unit 16a obtains a comparison result that the representative value related to each divided waveform data is larger than the threshold value, the determination unit 16 uses the waveform corresponding to the disturbing sound to be subjected to the sound reduction process. Determine that it contains data.

  The sound reduction processing unit 17 performs sound reduction processing on the divided waveform data that is determined by the determination unit 16 to include waveform data corresponding to the disturbing sound that is the target of the sound reduction processing. In the present embodiment, sound reduction processing is performed on the waveform data at the central portions d1, d2,... Of each divided waveform data. FIG. 6 is a diagram illustrating waveform data before and after the sound reduction process is performed. In the sound reduction processing unit 17 of the present embodiment, as shown in FIG. 6, the sound pressure levels of all frequencies included in the waveform data in the central part where the sound reduction processing is performed are reduced to the original value of 1 by the sound reduction processing. The value is changed to / 2.

  The waveform connecting unit 18 reconnects the waveform data at the central portions d1, d2,... Of the plurality of divided waveform data divided by the data dividing unit 12 to create one waveform data. Here, the waveform data that has not been subjected to the sound reduction processing can be smoothly connected without correcting the connection portions thereof. On the other hand, waveform data that has undergone sound reduction processing and waveform data that has not undergone sound reduction processing often cannot be smoothly connected as they are. Therefore, it is preferable that the connection portions are smoothly connected by appropriately correcting the connection portions.

  The improvement processing unit 19 deletes the component in the low frequency region included in the waveform data created by the waveform connecting unit 18 using a high-pass filter. Therefore, it is possible to suppress the low frequency sound from having a broad masking effect on the higher frequency sound. In the present embodiment, the improvement processing unit 19 normally deletes components in the low frequency region of about 500 Hz or less. Note that the upper limit value of the low frequency region deleted by the improvement processing unit 19 can be changed, and is changed as appropriate while listening to the sound. Then, the waveform data from which the components in the low frequency region are deleted by the improvement processing unit 19 is supplied to the DA converter 30.

  In the present embodiment, since the plurality of divided waveform data is subjected to data processing in groups of four, after the improvement processing is performed on the central waveform data of the four divided waveform data, Four pieces of waveform data are connected, and the waveform data is supplied to the DA converter 30. For example, regarding the divided waveform data corresponding to each of the times D1 to D4, when the waveform data of the central portions d1 to d4 are connected and improvement processing is performed on the waveform data, the connected waveform data is converted to a DA converter. 30.

  Next, the procedure of the acoustic monitoring support method in the acoustic monitoring support apparatus 1 will be described with reference to FIG. FIG. 2 is a flowchart showing the procedure of the acoustic monitoring support method in the acoustic monitoring support apparatus of FIG.

  First, sound waveform data is acquired by the microphone 21 (step S101). This sound includes an operation sound of the device to be monitored and may include an abnormal sound generated when the device has an abnormality. The sound acquired by the microphone 21 is digitized by the AD converter 20 and then stored as waveform data in the data storage unit 11 (step S102). Then, the data dividing unit 12 divides the waveform data into a plurality of divided waveform data (step S103).

  Thereafter, frequency analysis is performed for each of the plurality of divided waveform data (step S104). Based on the frequency analysis result, a representative value used for determining whether or not each divided waveform data includes waveform data corresponding to the disturbance sound to be reduced is calculated (step S105). ).

  Then, it is determined whether or not the representative value of the divided waveform data is greater than or equal to a threshold value (step S106). Here, if the representative value of the divided waveform data is equal to or greater than the threshold value, a sound reduction process is performed on the waveform data at the center of each divided waveform data (step S107). On the other hand, if the representative value of the divided waveform data is not greater than or equal to the threshold value, the sound reduction processing is not performed on the waveform data at the center of the divided waveform data.

  When the determination as to whether or not the waveform data corresponding to the disturbing sound to be reduced is included for one piece of divided waveform data, the determination as to whether or not the representative value of the divided waveform data is greater than or equal to the threshold value. It is determined whether or not has been executed a predetermined number of times (step S108). Here, the same processing is performed on the next divided waveform data until it is determined that the predetermined number of times has been executed. In this way, with respect to all the divided waveform data, the comparison between each representative value and the threshold value is performed, so that the determination as to whether or not the waveform data corresponding to the disturbing sound to be subjected to the sound reduction process is completed. Waveform data in a range corresponding to the central portion of the plurality of divided waveform data is connected (step S109).

  Here, in the divided waveform data corresponding to each of the times D1, D2, and D4 in FIG. 3, it is assumed that only the representative value Sf related to the divided waveform data corresponding to the time D3 is larger than the threshold value. In this case, since it is determined in the determination unit 16 that only the divided waveform data corresponding to the time D3 includes data corresponding to the disturbing sound to be subjected to the sound reduction process, each of the times D1, D2, and D4 is determined. The sound reduction processing is not performed on the waveform data of the central portions d1, d2, and d4 of the divided waveform data corresponding to, but the sound reduction processing is performed on the waveform data of the central portion d3 of the divided waveform data corresponding to time D3. FIG. 7 shows a state in which after the sound reduction processing is performed on the waveform data at the center d3 of the divided waveform data corresponding to the time D3, the waveform data is connected to other divided waveform data. In FIG. 7, the connection portion A between the waveform data subjected to the sound reduction process and the waveform data not subjected to the sound reduction process becomes discontinuous as shown in FIG. When connected as indicated by the broken line, a high frequency component is generated. Therefore, in order to prevent the generation of high frequency components, the connection portions are smoothed as shown by thick lines by performing appropriate correction as shown in FIG. 7B. Connected to. The same correction is performed also at the connection portion B in FIG.

  Thereafter, the low frequency component is deleted from the waveform data to which the plurality of divided waveform data are connected (step S110). Then, after the waveform data is supplied to the DA converter 31 and converted into an analog signal, the sound corresponding to the waveform data is output to the outside from the acoustic speaker 32 (step S111).

As described above, in the acoustic monitoring support device 1 according to the present embodiment, among the plurality of divided waveform data, the waveform data whose representative value is larger than the threshold value, that is, the interfering sound that is the target of the sound reduction process. A sound reduction process is performed for the waveform data including the waveform data. Thereafter, the waveform data subjected to the sound reduction process is connected to other waveform data, and a sound corresponding to the connected waveform data is output to the outside. For this reason, the portion of the sound output from the sound monitoring support device 1 that includes the disturbing sound that has undergone the sound reduction processing is lower than the original sound pressure level. Accordingly, since the influence of the masking phenomenon due to the interference sound is reduced, the target sound such as an abnormal sound of the device can be easily heard.

  Here, in the present embodiment, the abnormal sound generated in the central portion d3 of the divided waveform data corresponding to the time D3 when the sound reduction processing is performed is combined with the disturbing sound to be subjected to the sound reduction processing in the central portion d3. Since the sound is reduced, it is considered that the abnormal sound generated in the central portion d3 is difficult to hear. However, the disturbing sound that is the target of the sound reduction process in the center part d3 also affects the masking phenomenon to other sounds in time (other than the center part d3). That is, it becomes difficult to hear abnormal sounds generated during other times due to the interference sound to be reduced in the central portion d3. Therefore, by reducing the influence of the masking phenomenon caused by the interference sound to be reduced in the central portion d3, it is possible to obtain an effect that it becomes easy to hear the abnormal sound generated in other time. Therefore, when the target sound is always generated and the interfering sound is generated intermittently, by reducing the sound pressure level for a short time including the time when the interfering sound is generated, the time other than the time when the interfering sound is reduced This makes it easier to hear the target sound generated within the time. Therefore, it is effective when there is combustion noise such as engine noise.

  Further, by comparing the representative value calculated for each divided waveform data with a threshold value, it is determined whether or not each divided waveform data includes waveform data corresponding to the disturbing sound to be reduced. Therefore, since the determination is only a comparison between the representative value and the threshold value, if only the representative value is calculated, it is a binary comparison, and it is determined whether or not the divided waveform data includes data corresponding to the disturbing sound to be reduced. It can be easily determined for each divided waveform data. Moreover, since the representative value and the threshold value are compared, it is suitable for determining whether or not the disturbing sound to be subjected to the sound reduction process is emitted over a wide frequency range.

  Further, when the divided waveform data includes waveform data corresponding to the disturbing sound to be reduced, the sound pressure levels of all the frequencies included in the divided waveform data are reduced. As compared with the case of reducing the sound pressure level of some frequencies, it is possible to easily perform the sound reduction process.

  Moreover, since the threshold value memorize | stored in the threshold value memory | storage part 15 can be changed by the operation part 40, according to the kind etc. of target sound, a threshold value can be set suitably.

  Further, since the waveform data is divided into a plurality of divided waveform data for each predetermined time, division into a plurality of divided waveform data becomes easy.

  In addition, since two divided waveform data that are continuous among a plurality of divided waveform data include waveform data at the same time, it is possible to reduce temporal discontinuity in the analysis result when performing frequency analysis. , Information loss can be reduced.

Next, an acoustic monitoring support device according to a reference example of the present invention will be described with reference to the drawings. FIG. 8 is a diagram showing a schematic configuration of an acoustic monitoring support apparatus according to a reference example of the present invention.

The sound monitoring support device 101 of the present reference example is different from the sound monitoring support device 1 of the first embodiment in that the sound monitoring support device 1 includes waveform data corresponding to the interference sound to be reduced. In the sound monitoring support apparatus 101, the waveform data at the center of the divided waveform data is the waveform at the center of the other divided waveform data. It is a point that is changed to data. Since the other structure of the acoustic monitoring support apparatus 101 is the same as that of the acoustic monitoring support apparatus 1, the same reference numerals are given and detailed description thereof is omitted.

  The main unit 110 includes a data storage unit 11, a data division unit 12, a frequency analysis unit 13, a representative value calculation unit 14, a threshold storage unit 15, a determination unit 16, a determination unit 117, and an improvement processing unit. 18 and a waveform connecting portion 19 are formed.

  The determination unit 117 can change the waveform data at the center of the divided waveform data determined by the determination unit 16 to include data corresponding to the interference sound to be reduced, and the determination unit 16 can change the waveform data. The waveform data in the center of the divided waveform data determined to contain no data corresponding to the disturbing sound to be reduced is determined. In the present embodiment, the determination unit 117 applies the waveform data in the central portion of the divided waveform data immediately before the divided waveform data determined by the determining unit 16 to include data corresponding to the disturbing sound to be reduced. decide.

  The waveform connecting unit 18 reconnects the waveform data at the central portions d1, d2,... Of the plurality of divided waveform data divided by the data dividing unit 12 to create one waveform data. At this time, the waveform data at the center of the divided waveform data determined to include data corresponding to the interference sound to be reduced is determined by the determination unit 117 and becomes the interference sound to be reduced. The waveform data is changed to the waveform data at the center of the divided waveform data not including the corresponding data. Here, except for the waveform data at the center of the divided waveform data determined to contain the data corresponding to the disturbance sound to be reduced, and the waveform data changed, Can be smoothly connected without correcting the connected portions. On the other hand, the central waveform data, the changed waveform data, and the other central waveform data determined to contain data corresponding to the interference sound to be reduced are smoothly connected as they are. Often it is not possible. Therefore, it is preferable that the connection portions are smoothly connected by appropriately correcting the connection portions.

  Next, the acoustic monitoring support method in the present embodiment will be described with reference to FIG. FIG. 9 is a flowchart showing the procedure of the acoustic monitoring support method in the acoustic monitoring support apparatus.

  First, sound waveform data is acquired by the microphone 21 (step S201). The sound acquired by the microphone 21 is digitized by the AD converter 20 and then stored as waveform data in the data storage unit 11 (step S202). Then, the data dividing unit 12 divides the waveform data into a plurality of divided waveform data (step S203).

  Thereafter, frequency analysis is performed for each of the plurality of divided waveform data (step S204). Based on the frequency analysis result, a representative value used for determining whether or not each divided waveform data includes waveform data corresponding to the disturbing sound to be reduced is calculated (step S205). ).

  Then, it is determined whether or not the representative value of the divided waveform data is greater than or equal to a threshold value (step S206). Here, when the representative value of each divided waveform data is greater than or equal to the threshold value, the waveform data at the center is changed to the waveform data at the center just before that (step S207). On the other hand, when the representative value related to the waveform data at each central portion of the divided waveform data is not equal to or greater than the threshold value, the waveform data at the central portion is not changed.

  When the determination as to whether or not the waveform data corresponding to the disturbing sound to be reduced is included for one piece of divided waveform data, the representative value for the waveform data at each central portion of the divided waveform data is greater than or equal to the threshold value. It is determined whether or not the determination as to whether or not is executed a predetermined number of times (step S208). In this way, with respect to all the divided waveform data, the comparison between each representative value and the threshold value is performed, so that the determination as to whether or not the waveform data corresponding to the disturbing sound to be subjected to the sound reduction process is completed. The waveform data at the center of the plurality of divided waveform data are connected (step S209).

  Here, as described above, in the divided waveform data corresponding to each of the times D1, D2, and D4 in FIG. 3, it is assumed that only the representative value Sf related to the divided waveform data corresponding to the time D3 is larger than the threshold value. In this case, since the determination unit 16 determines that only the waveform data of the central portion d3 includes data corresponding to the disturbing sound to be subjected to the sound reduction process, the waveform data of the central portion d3 is immediately before it. To the waveform data of the central part d2. Therefore, in the waveform connecting unit 18, the waveform data of the central part d1, the waveform data of the central part d2, the waveform data of the central part d2, and the waveform data of the central part d4 are connected in this order.

  Thereafter, the low frequency region component is deleted from the waveform data to which the plurality of divided waveform data are connected (step S210). Then, after the waveform data is supplied to the DA converter 31 and converted into an analog signal, the sound corresponding to the waveform data is output to the outside from the acoustic speaker 32 (step S211).

As described above, in the acoustic monitoring support device 101 according to this reference example , the divided waveform data including the data corresponding to the disturbing sound interference sound to be reduced among the plurality of divided waveform data is reduced. The data is changed to divided waveform data not including data corresponding to the disturbing sound to be processed. Thereafter, a plurality of divided waveform data are connected, and a sound corresponding to the connected waveform data is output to the outside. Therefore, the same effect as that of the first embodiment can be obtained.

  The preferred embodiments of the present invention have been described above. However, the present invention is not limited to the above-described embodiments, and various design changes can be made as long as they are described in the claims. For example, in the above-described embodiment, by comparing the representative value Sf calculated for each divided waveform data with a threshold value, each divided waveform data includes waveform data corresponding to the disturbing sound to be subjected to the sound reduction process. However, it may be determined by other methods whether each divided waveform data includes waveform data corresponding to the disturbing sound to be subjected to the sound reduction process. Therefore, the representative value calculated for each divided waveform data may be a representative value calculated by a method different from the above-described representative value Sf.

  Further, by comparing the sound pressure level for each frequency in the divided waveform data with a threshold value without calculating a representative value for each divided waveform data, each divided waveform data is converted into an interference sound to be reduced. It may be determined whether corresponding waveform data is included. In this case, the threshold value storage unit stores a threshold value relating to the sound pressure level (for example, 70 dB), and when a comparison result is obtained that at least one of the sound pressure levels for each frequency is greater than the threshold value, It is determined that the divided waveform data includes data corresponding to the disturbing sound to be subjected to the sound reduction process.

  Further, in the above-described embodiment, when the divided waveform data includes waveform data corresponding to the interference sound to be reduced, the sound pressure levels of all the frequencies included in the divided waveform data are reduced. However, it is not necessary to reduce the sound pressure levels of all frequencies included in the divided waveform data.

  Further, as described above, by comparing the sound pressure level for each frequency in the divided waveform data with the threshold value, whether or not each divided waveform data includes waveform data corresponding to the disturbing sound to be reduced. If the divided waveform data includes waveform data corresponding to the disturbance sound to be reduced, the frequency of the sound pressure level that is greater than the threshold value included in the divided waveform data is determined. Only the sound pressure level may be reduced. In other words, when only the sound pressure level at the frequency Xm is higher than 70 dB when the threshold is set to 70 dB, the sound pressure level at a frequency other than the frequency Xm is not decreased, but only the sound pressure level at the frequency Xm is decreased. May be. In this case, since the sound pressure level and the corresponding threshold value are compared for each frequency, it is possible to accurately perform the sound reduction processing for the disturbing sound generated in the frequency band of the local range. . Therefore, it is possible to suppress the target sound from being reduced at the same time as reducing the interference sound.

  Further, in the above-described embodiment, the divided waveform data is divided into a plurality of divided waveform data for each predetermined time, and two consecutive divided waveform data among the plurality of divided waveform data respectively include waveform data at the same time. However, it may be divided into a plurality of divided waveform data corresponding to different times, or two consecutive divided waveform data among the plurality of divided waveform data may not include waveform data at the same time. Good (does not need to overlap). Further, when the frequency analysis can be performed without dividing the waveform data stored in the waveform data storage unit, the waveform data may not be divided into a plurality of divided waveform data.

It is a figure which shows schematic structure of the acoustic monitoring assistance apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart which shows the procedure of the acoustic monitoring assistance method in the acoustic monitoring assistance apparatus of FIG. It is a figure which shows an example of the waveform data memorize | stored in a data storage part. It is a figure which shows the frequency analysis result regarding each division | segmentation waveform data. It is a figure for demonstrating the calculation procedure of the representative value of each division | segmentation waveform data. It is a figure which shows the waveform data before and behind a sound reduction process. It is a figure which shows a mode that several division | segmentation waveform data are connected. It is a figure which shows schematic structure of the acoustic monitoring assistance apparatus which concerns on the reference example of this invention. It is a flowchart which shows the procedure of the acoustic monitoring assistance method in the acoustic monitoring assistance apparatus of FIG.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1,101 Sound monitoring assistance apparatus 11 Data storage part 12 Data division part (data division means)
13 Frequency analysis unit (frequency analysis means)
14 representative value calculation unit (representative value calculation means)
15 Threshold storage unit (threshold storage means)
16 Determination part (determination means)
16a comparison part (comparison means)
17 Sound reduction processing part (sound reduction processing means)
18 Improvement processing part 19 Waveform connection part (data connection means)
21 Microphone (data acquisition means)
32 Acoustic speaker (output means)
40 Operation part (change means)
117 determination part (determination means)

Claims (7)

Data acquisition means for acquiring sound waveform data;
Data dividing means for dividing the waveform data of the sound obtained by the data acquiring means into a plurality of divided waveform data;
Frequency analysis means for analyzing the sound pressure level for each frequency for each of a plurality of divided waveform data divided by the data dividing means;
Based on the frequency analysis result in the frequency analysis means, determination means for determining whether each of the plurality of divided waveform data includes data corresponding to the interference sound to be subjected to sound reduction processing;
Sound reduction processing means for reducing at least a part of the sound pressure level for each frequency included in the divided waveform data determined to include data corresponding to the interference sound to be subjected to sound reduction processing in the determination means;
The determination means includes divided waveform data determined not to include data corresponding to the disturbance sound to be reduced, and data corresponding to the disturbance sound to be reduced by the determination means. And a data connection means for connecting the divided waveform data after the sound reduction processing by the sound reduction processing means is performed,
An acoustic monitoring support apparatus comprising: output means for outputting a sound corresponding to waveform data to which a plurality of divided waveform data is connected by the data connection means. Based on the frequency analysis result in the frequency analysis means, a representative value used for determining whether or not the divided waveform data includes data corresponding to the interference sound to be subjected to the sound reduction process is determined for each divided waveform data. Representative value calculating means for calculating;
A comparison means for comparing the representative value calculated by the representative value calculation means with a threshold value;
When the comparison means obtains a comparison result that the representative value of the divided waveform data is larger than the threshold value in the comparison means, the divided waveform data includes data corresponding to the disturbing sound to be reduced. The sound monitoring support apparatus according to claim 1, wherein the determination is performed.   The sound reduction processing means, when the determination means determines that the divided waveform data includes data corresponding to the disturbing sound to be reduced, the sound of all frequencies included in the divided waveform data The acoustic monitoring support apparatus according to claim 1, wherein the pressure level is reduced. Comparing means for comparing the sound pressure level and the threshold value for each frequency in the divided waveform data,
When the comparison means obtains a comparison result that at least one of the sound pressure levels for each frequency is greater than the threshold, the division waveform data includes data corresponding to the disturbing sound to be reduced. The sound monitoring support apparatus according to claim 1, wherein the sound monitoring support apparatus is determined to contain the sound.   The sound reduction processing means, when the determination means determines that the divided waveform data includes data corresponding to the disturbing sound to be reduced, the sound of all frequencies included in the divided waveform data The sound monitoring support apparatus according to claim 4, wherein the pressure level is lowered.   The sound reduction processing means, when the determination means determines that the divided waveform data includes data corresponding to the disturbing sound to be reduced, the sound pressure greater than the threshold included in the divided waveform data. The sound monitoring support apparatus according to claim 4, wherein only the sound pressure level of a frequency having a level is lowered. A data acquisition step for acquiring sound waveform data;
A data division step for dividing the waveform data of the sound obtained in the data acquisition step into a plurality of divided waveform data;
A frequency analysis step of analyzing a sound pressure level for each frequency for each of a plurality of divided waveform data divided in the data division step;
Based on the frequency analysis result in the frequency analysis step, a determination step for determining whether each of the plurality of divided waveform data includes data corresponding to the interference sound to be subjected to the sound reduction process;
A sound reduction processing step for reducing at least a part of the sound pressure level for each frequency included in the divided waveform data determined to include data corresponding to the interference sound to be subjected to the sound reduction processing in the determination step;
The divided waveform data determined not to include the data corresponding to the disturbance sound to be reduced in the determination step, and the data corresponding to the disturbance sound to be reduced in the determination step are included. And a data connection step for connecting the divided waveform data after the sound reduction processing in the sound reduction processing step is performed,
An acoustic monitoring support method comprising: an output step of outputting to the outside a sound corresponding to the waveform data to which a plurality of divided waveform data is connected in the data connection step.