Sound production control device

Abstract

When the input waveform (Win) includes an exceeding portion (EX) in which the absolute value of the level (Lv) exceeds the limiter threshold value (Th), the absolute value waveform generation unit (S03) generates an absolute value waveform (Wab) obtained by converting the level into the absolute value of the level in the input waveform. A peak holding waveform generation unit (S04) generates a peak holding waveform (Wp) on the basis of the absolute value waveform. A gain curve generation unit (S05) generates a gain curve (Wgn) indicating the relationship between a gain (Gn) obtained on the basis of the level in the peak-held waveform and an input waveform time (Tm). A smoothing processing unit (S06) obtains a smoothed gain curve (Wsgn) based on the gain curve by means of smoothing processing. Sound generation data generation units (S07, S08) generate sound generation data by multiplying a delayed input waveform (Wind), which is obtained by shifting the input waveform to the time forward side (Dtp), by the smooth gain curve.

Description

[0001] Cross-reference to related applications

[0002] This application claims priority based on Japanese Patent Application No. 2023-44827 filed on March 21, 2023, the description of which is incorporated herein by reference. Technical Field

[0003] The present disclosure relates to a sound emission control device for causing a sound emitting body to emit sound. Background Art

[0004] For example, the vehicle approach notification device described in Patent Document 1 is conventionally known and includes a function for generating sound data for causing a sound-emitting device, such as a speaker, to emit sound. The vehicle approach notification device described in Patent Document 1 is installed in a vehicle, such as a hybrid vehicle or an electric vehicle. For example, the vehicle approach notification device may combine an approach notification sound, which notifies pedestrians around the vehicle of its approach, with a reverse warning sound, which notifies pedestrians of the vehicle's backing up, during low-speed driving, and emit these sounds simultaneously from a single sound-emitting device.

[0005] Prior art literature

[0006] Patent Literature

[0007] Patent Document 1: Japanese Patent Application Laid-Open No. 2013-49312 Summary of the Invention

[0008] A sound-generating system (e.g., the vehicle approach notification device disclosed in Patent Document 1) in which a sound-generating object generates sound based on sound data includes a digital-to-analog converter and a power amplifier for processing the sound signal input to a sound-generating object such as a speaker. Allowable input / output settings are set for each of these digital-to-analog converters and the power amplifier. Furthermore, the sound-generating object also has allowable input / output settings.

[0009] Therefore, in a sound generation system, a sound generation control device that generates sound generation data needs to suppress input and output signals of audio devices such as digital-to-analog converters, power amplifiers, and sound generation bodies to below allowable input / output and effectively utilize the allowable input / output.

[0010] In contrast, for example, simply limiting the peak values ​​of the waveforms could be used to suppress the input and output signals of an audio device to below the permissible input / output. However, simply limiting the peak values ​​of the waveforms in this way generates harmonic distortion due to waveform distortion, which in turn produces unusual sounds. The inventors have discovered this through detailed research.

[0011] In view of the above problems, an object of the present disclosure is to provide a sound emission control device that can limit the amplitude of a waveform in sound emission data while suppressing abnormal sounds such as harmonic distortion caused by the amplitude limitation of the waveform.

[0012] To achieve the above objectives, according to one aspect of the present disclosure, a sound control device is included in a sound system in which a sound-emitting body generates sound according to a sound signal based on sound data representing a sound waveform, and generates sound data. The sound control device comprises:

[0013] an excess determination unit that acquires an input waveform representing a relationship between a voltage level and time corresponding to a speech signal and serving as a basis for speech data, and determines whether the input waveform includes an excess portion whose absolute value of the voltage level exceeds a predetermined limiter threshold;

[0014] an absolute value waveform generating unit that generates an absolute value waveform by converting a level in the input waveform to an absolute value of the level when the excess determining unit determines that the input waveform includes an excess portion;

[0015] a peak hold waveform generating section that generates a peak hold waveform representing a relationship between a level and a time in an input waveform, that is, an input waveform time, based on the absolute value waveform;

[0016] a gain curve generating unit that generates a gain curve representing a relationship between a gain obtained based on a level in a peak hold waveform and a time period of an input waveform;

[0017] a smoothing processing unit that performs a smoothing process on the gain curve to obtain a smoothed gain curve based on the gain curve; and

[0018] The speech data generating unit generates speech data by multiplying the delayed input waveform by a predetermined waveform delay time, which is shifted toward the time-passing side of the input waveform, i.e., the time-positive side, in the time axis direction, by the smoothing gain curve.

[0019] The peak hold waveform has a waveform shape obtained by extracting the outer shape formed by the side with a higher level in the waveform aggregate. The waveform aggregate includes: an absolute value waveform; a peak hold portion, which is provided for each maximum point of the level in the absolute value waveform and extends linearly from the maximum point to the positive side in time while showing the same level as the maximum point, with a limit of continuing the peak hold time from the maximum point until it reaches the absolute value waveform; and a release portion, which is provided for each peak hold portion extending for the peak hold time and extends from the end of the peak hold portion on the positive side in time, moving toward the lower side of the level as it moves toward the positive side in time, until it reaches the absolute value waveform.

[0020] The gain of the gain curve is set to 1 within the range in the time axis direction where the level in the peak hold waveform is below the limiter threshold, and the gain of the gain curve is set to a value obtained by "Gn=Th / Lv" within the range in the time axis direction where the level in the peak hold waveform exceeds the limiter threshold, where the gain is Gn, the level is Lv, and the limiter threshold is Th.

[0021] The smooth gain curve is set to shift the gain delay time toward the positive side in time by comparing the peak parts where the gain becomes maximum or minimum with respect to the gain curve.

[0022] The waveform delay time is longer than the gain delay time, and the peak hold time is longer than the waveform delay time.

[0023] In this way, sound data is generated based on the input waveform. Simultaneously, by multiplying the delayed input waveform by the smoothed gain curve, the amplitude of the waveform represented by the sound data is limited to below the limiter threshold. Furthermore, since the waveform of this amplitude-limited sound data is smoothed, abnormal sounds such as harmonic distortion caused by amplitude limiting can be suppressed.

[0024] It should be noted that in various columns of the application documents, each element may be annotated with a parenthetical reference numeral. In such cases, the reference numeral merely represents an example of the correspondence between the element and the specific structure described in the embodiments described below. Therefore, this disclosure is not limited in any way by the description of the reference numerals. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 This is a block diagram showing a schematic configuration of a sound generation system in the first embodiment.

[0026] Figure 2 This is a flowchart showing a control process executed by the limiter unit included in the sound emission system according to the first embodiment.

[0027] Figure 3 The graph shows an input waveform to the limiter section in (a), an absolute value waveform and a peak hold waveform in (b), and a gain curve in (c), while aligning the time axes of the orthogonal coordinate systems.

[0028] Figure 4 This diagram shows the waveform aggregate and absolute value waveform that are the basis of the peak hold waveform.

[0029] Figure 5 The graphs show a gain curve in (a) and a smoothed gain curve obtained by smoothing the gain curve in (b) while aligning the time axes of the respective orthogonal coordinate systems.

[0030] Figure 6 These are diagrams showing a smoothed gain curve in (a), a delayed input waveform in (b), and a waveform represented by speech data in (c) while aligning the time axes of the respective orthogonal coordinate systems. DETAILED DESCRIPTION

[0031] Hereinafter, embodiments will be described with reference to the accompanying drawings. It should be noted that, in the following embodiments including other embodiments described later, identical or equivalent parts are denoted by the same reference numerals in the drawings.

[0032] (First embodiment)

[0033] In this embodiment, Figure 1 The illustrated sound system 8 is a vehicle approach notification device installed in, for example, a hybrid vehicle or electric vehicle. For example, the sound system 8 is electrically connected to a sound-emitting device 80, such as a speaker or buzzer. Furthermore, the sound system 8 outputs various warning sounds, including a vehicle approach notification sound that notifies the surrounding area of ​​the approaching vehicle equipped with the sound system 8, as well as other notification sounds, from the sound-emitting device 80.

[0034] The sound generating system 8 includes a microcomputer 10, a digital-to-analog converter 12, and a power amplifier 14. In the description of this embodiment, the digital-to-analog converter 12 is also referred to as DAC 12, and the power amplifier 14 is also referred to as AMP 14.

[0035] The microcomputer 10 outputs the sound data Dsd generated by the microcomputer 10 to the DAC 12. The DAC 12 converts the sound data Dsd input from the microcomputer 10 into a sound signal Ssa, which is an analog signal, and outputs the sound signal Ssa to the AMP 14. It should be noted that the sound data Dsd, the output waveform data DW described later, the first synthesized waveform data D1m, and the second synthesized waveform data D2m are digital data representing a waveform (e.g., a PCM waveform) of sound.

[0036] Based on the application of voltage from a constant voltage power supply (not shown), the AMP 14 causes a current corresponding to the sound signal Ssa input from the DAC 12 to flow through the sound-generating body 80. The sound-generating body 80 generates sound in response to the current supplied by the AMP 14. Specifically, the sound-generating body 80 generates sound in response to the sound signal Ssa based on the sound data Dsd. It should be noted that in this embodiment, the circuit gain of the sound circuit between the microcomputer 10 and the sound-generating body 80, that is, the circuit gain of the sound circuit formed by the DAC 12 and the AMP 14, is set to be constant, for example.

[0037] The microcomputer 10 is an electronic control device, and has a structure as an on-vehicle microcomputer equipped with a CPU, RAM, ROM, non-volatile rewritable memory, etc. (not shown). That is, the microcomputer 10 reads and executes a computer program stored in a ROM or non-volatile rewritable memory as a non-transitory physical storage medium. By executing the computer program, the method corresponding to the computer program is executed. That is, in the microcomputer 10, according to the computer program, for example, the limiter unit 26 included in the microcomputer 10, which will be described later, is executed. Figure 2 Various control processing such as control processing.

[0038] like Figure 1 As shown, the microcomputer 10 functionally includes a plurality of sound source control units 16 , a mixing unit 22 , an equalizer unit 24 , and a limiter unit 26 .

[0039] The multiple sound source control units 16 each select one or more sound source data from a plurality of sound source data pre-stored in a storage medium such as a ROM, based on the vehicle status signal. The sound source control units 16 then perform pitch and volume control on each selected sound source data to generate output waveform data DW based on the selected sound source data, and output this output waveform data DW to the mixing unit 22. The vehicle status signal is a signal indicating the status of the vehicle equipped with the sound generation system 8 (e.g., the vehicle's driving conditions, the remaining fuel level, the vehicle's surrounding conditions, etc.), and is input to the microcomputer 10 from, for example, various sensors included in the vehicle.

[0040] Among the plurality of sound source control units 16, a sound source control unit 16 that selects sound source data to be played as, for example, a vehicle approach notification sound performs pitch control and volume control on the selected sound source data. On the other hand, a sound source control unit 16 that selects sound source data to be played as, for example, a notification sound other than the vehicle approach notification sound performs volume control instead of pitch control on the selected sound source data.

[0041] The mixing section 22 mixes the plurality of output waveform data DW inputted to the mixing section 22 from the plurality of sound source control sections 16. The mixing section 22 then outputs the mixed waveform data, i.e., the first synthesized waveform data D1m, to the equalizer section 24. In other words, the mixing section 22 outputs the first synthesized waveform data D1m synthesized based on the plurality of output waveform data DW to the equalizer section 24.

[0042] The equalizer section 24 modifies the acoustic characteristics of the first synthesized waveform data D1m to generate second synthesized waveform data D2m based on the first synthesized waveform data D1m, and outputs the second synthesized waveform data D2m to the limiter section 26. For example, the filter characteristics of the equalizer section 24 are determined by specifying the filter type, center frequency, gain, and filter width. The degree of change in the second synthesized waveform data D2m relative to the first synthesized waveform data D1m varies greatly depending on the combination of the first synthesized waveform data D1m and the filter characteristics of the equalizer section 24, making it difficult to predict the peak amplitude of the digital waveform represented by the second synthesized waveform data D2m.

[0043] It should be noted that the filter types are selected, for example, by known techniques and may include low-pass filters, high-pass filters, band-pass filters, notch filters, low-shelf filters, high-shelf filters, and peak filters.

[0044] The limiter unit 26 limits the amplitude of the digital waveform represented by the sound data Dsd to below a predetermined limiter threshold Th, and outputs the amplitude-limited sound data Dsd to the DAC 12. The limiter threshold Th corresponds to the allowable input of the sound-emitting body 80 and the AMP 14, and the full range of the DAC 12. For example, the limiter threshold Th is set in advance through experiments so that the limiter threshold Th is as large as possible while keeping the voltage of the input signal below the allowable input at the sound-emitting body 80 and the AMP 14, respectively, and keeping the waveform amplitude of the sound data Dsd input to the DAC 12 below the full range of the DAC 12.

[0045] In order to limit the amplitude of the sound data Dsd in this manner, the limiter section 26 pre-reads the second synthesized waveform data D2m input from the equalizer section 24 for outputting the sound data Dsd to the DAC 12, and performs Figure 2 For example, if the second synthesized waveform data D2m is input to the limiter section 26, the Figure 2 The control process is repeatedly executed. It should be noted that the limiter unit 26 corresponds to the sound control device of the present disclosure.

[0046] like Figure 2 As shown, the limiter section 26 first obtains an input waveform Win input to the limiter section 26 in step S01. This input waveform Win is a digital waveform represented by the second synthesized waveform data D2m input from the equalizer section 24 to the limiter section 26, and forms the basis for the sound data Dsd output from the limiter section 26 to the DAC 12. It should be noted that in addition to the input waveform Win, the waveforms and curves generated based on this input waveform Win in steps S03 to S09 described below are also digital waveforms.

[0047] In addition, if Figure 3 As shown in (a) of FIG. 1 , the input waveform Win is a waveform showing the relationship between the level Lv and the time Tm. The level Lv corresponds to the voltage of the sound signal Ssa, which is an analog signal, and the amplitude of the level Lv corresponds to the voltage amplitude of the sound signal Ssa. In the description of this embodiment, the time Tm on the horizontal axis of the input waveform Win is sometimes referred to as the input waveform time Tm. Figure 2 After step S01, the process goes to step S02.

[0048] In step S02, as Figure 3 As shown in (a) in FIG. 1 , the limiter section 26 determines whether the input waveform Win includes an excess portion EX in which the absolute value of the level Lv exceeds the limiter threshold Th. Figure 3 The input waveform Win illustrated in (a) includes three excess portions EX.

[0049] exist Figure 2 In step S02, if it is determined that the input waveform Win includes the excess portion EX, the process proceeds to step S03. On the other hand, if it is determined that the input waveform Win does not include the excess portion EX, the process proceeds to step S09.

[0050] In step S03, Figure 3 As shown in (b) in FIG. 1 , the limiter section 26 generates an absolute value waveform Wab obtained by converting the level Lv on the vertical axis in the input waveform Win into the absolute value of the level Lv. In other words, the absolute value waveform Wab is a waveform obtained by inverting the negative portion of the input waveform Win where the level Lv is a negative value toward the positive side of the level Lv with the zero position of the level Lv as a reference, and a waveform obtained by merging the positive portion of the input waveform Win where the level Lv is a positive value. Figure 2 After step S03, the process goes to step S04.

[0051] In step S04, Figure 3 (b) Figure 4 As shown, the limiter section 26 generates a peak hold waveform Wp indicating the relationship between the level Lv and the input waveform time Tm based on the absolute value waveform Wab.

[0052] It should be noted that in the description of this embodiment, in an orthogonal coordinate system with the level Lv as the vertical axis and the input waveform time Tm as the horizontal axis, the side that passes through the input waveform time Tm in the time axis direction Dt as the horizontal axis is referred to as the time positive side Dtp, and the side that traces back to the input waveform time Tm is referred to as the time negative side Dtm. In addition, in this orthogonal coordinate system, the side with a higher level Lv in the level axis direction DL as the vertical axis is referred to as the high level side DLp, and the side with a lower level Lv is referred to as the low level side DLm. In addition, for confirmation, it should be noted that Figure 3 The absolute value waveform Wab and Figure 4 The displayed absolute value waveform Wab is the same.

[0053] Specifically, in order to obtain the peak hold waveform Wp, the limiter section 26 first performs the following operation: Figure 4 As shown, a waveform aggregate 30 consisting of an absolute value waveform Wab, one or more peak hold portions 32, and one or more release portions 33 is generated.

[0054] The peak holding portion 32 included in the waveform aggregate 30 is provided for each local maximum point 28 of the level Lv in the absolute value waveform Wab. That is, the peak holding portion 32 is provided as many as the local maximum points 28 of the level Lv in the absolute value waveform Wab.

[0055] Peak hold portion 32 extends linearly from maximum point 28 of level Lv to the positive temporal side Dtp in parallel with time axis direction Dt. Specifically, peak hold portion 32 extends linearly from maximum point 28 as the starting point to the positive temporal side Dtp while maintaining the same level Lv as that at maximum point 28 for a predetermined peak hold time Ht until reaching absolute value waveform Wab.

[0056] For example, if peak hold portion 32 reaches any portion of absolute value waveform Wab before extending from maximum point 28 and reaching the length of peak hold time Ht, the length of peak hold portion 32 is less than the length of peak hold time Ht. On the other hand, if peak hold portion 32 extends from maximum point 28 and reaches the length of peak hold time Ht without intersecting absolute value waveform Wab, the length of peak hold portion 32 becomes the length of peak hold time Ht. Peak hold time Ht is, for example, a constant value and is pre-set to be as short as possible within a range that can limit the amplitude of the waveform represented by speech data Dsd to below limiter threshold Th.

[0057] The release portion 33 included in the waveform aggregate 30 is provided for each peak hold portion 32 that extends for the peak hold time Ht. Specifically, the release portion 33 is provided for each peak hold portion 32 whose length in the time axis direction Dt reaches the length of the peak hold time Ht, but is not provided for each peak hold portion 32 whose length in the time axis direction Dt is less than the length of the peak hold time Ht. Therefore, the number of release portions 33 provided is the same as the number of peak hold portions 32 whose length in the time axis direction Dt reaches the length of the peak hold time Ht.

[0058] The release portion 33 begins at the extended end 32a, the end of the peak hold portion 32 on the positive temporal side Dtp, which lasts for the peak hold time Ht. From this extended end 32a, it progresses toward the lower level side DLm as it approaches the positive temporal side Dtp, extending until it reaches any position in the absolute value waveform Wab. The gradient of the release portion 33 in the orthogonal coordinate system of the level Lv and the input waveform time Tm can be constant or can become steeper as it moves away from the extended end 32a, the starting point. The gradient of the release portion 33 is preset to avoid the unnatural sound produced by limiting the amplitude of the waveform represented by the sound data Dsd to below the limiter threshold Th.

[0059] like Figure 3 (b) Figure 4 As shown, the limiter section 26 extracts the shape of the high-level side DLp formed in the waveform aggregate 30 formed in the above-mentioned manner, that is, the high-level side shape 301, and sets the shape of the extracted high-level side shape 301 as the waveform shape of the peak hold waveform Wp. In other words, the limiter section 26 determines the peak hold waveform Wp so that the waveform shape of the peak hold waveform Wp is the same as the shape of the high-level side shape 301 of the waveform aggregate 30. In short, the peak hold waveform Wp generated in step S04 is determined to have a waveform shape that is the same as the shape of the high-level side shape 301 of the waveform aggregate 30. It should be noted that the time width occupied by the release section 33 in the time axis direction Dt in the peak hold waveform Wp is called the release time Rt. Figure 2 After step S04, the process proceeds to step S05.

[0060] In step S05, Figure 3 (c) Figure 5 As shown in (a) in FIG. 1 , the limiter section 26 generates a gain curve Wgn showing the relationship between the gain Gn obtained based on the level Lv in the peak hold waveform Wp and the input waveform time Tm.

[0061] It should be noted that in the description of this embodiment, in an orthogonal coordinate system with gain Gn as the vertical axis and input waveform time Tm as the horizontal axis, the side with greater gain Gn in the gain axis direction Dgn, which is the vertical axis, is referred to as the high-gain side Dgnp, and the side with less gain Gn is referred to as the low-gain side Dgnm. Furthermore, in this orthogonal coordinate system of gain Gn and input waveform time Tm, the direction of the time axis Dt is represented by the positive time side Dtp and the negative time side Dtm, similar to the orthogonal coordinate system of level Lv and input waveform time Tm described above. Furthermore, in the description and various figures of this embodiment, gain Gn is sometimes displayed as a raw value and sometimes as a percentage.

[0062] Specifically, the gain Gn of the gain curve Wgn is set to 1 in the ranges R1t and R2t in the time axis direction Dt (ie, within the threshold ranges R1t and R2t) where the level Lv in the peak hold waveform Wp is equal to or lower than the limiter threshold Th.

[0063] On the other hand, the gain Gn of the gain curve Wgn is set to a value obtained by the following formula F1 in the range R3t and R4t in the time axis direction Dt where the level Lv in the peak hold waveform Wp exceeds the limiter threshold Th (i.e., the over-threshold range R3t and R4t). The over-threshold range R3t and R4t is the range excluding the in-threshold range R1t and R2t from the width of the peak hold waveform Wp in the time axis direction Dt. In the following formula F1, Gn is the gain Gn of the gain curve Wgn, Th is the limiter threshold Th, and Lv is the level Lv in the peak hold waveform Wp. Figure 2 After step S05, the process proceeds to step S06.

[0064] Gn=Th / Lv…(F1)

[0065] In step S06, Figure 5 As shown in (a) and (b) of FIG. 1 , the limiter unit 26 performs a smoothing process to smooth the gain curve Wgn, thereby obtaining a smoothed gain curve Wsgn based on the gain curve Wgn. For example, the smoothing process is performed by passing the gain curve Wgn through a double moving average filter, a Bessel filter, or a Thiran low-pass filter.

[0066] For confirmation, it is necessary to explain that Figure 3 The gain curve Wgn and Figure 5 The gain curve Wgn is the same as that shown in (a) in FIG. Figure 6 The smooth gain curve Wsgn shown in (a) is Figure 5 The smooth gain curve Wsgn is the same as that shown in (b) in FIG.

[0067] Specifically, the smoothed gain curve Wsgn obtained in step S06 is set to be shifted toward the temporal positive side Dtp by a gain delay time St relative to the gain curve Wgn by comparing the peak portions 36a and 37a where the gain Gn reaches a maximum or minimum. This gain delay time St is a constant value.

[0068] For example, in the time axis direction Dt, the center time T1 of a peak portion 36a included in the gain curve Wgn is compared with the center time T2 of the peak portion 37a corresponding to the peak portion 36a of the smooth gain curve Wsgn. In this case, the center time T2 of the peak portion 37a of the smooth gain curve Wsgn is shifted toward the positive temporal side Dtp by the gain delay time St relative to the center time T1 of the peak portion 36a of the gain curve Wgn. This is also the same as when comparing the other peak portions 36a and 37a between the gain curve Wgn and the smooth gain curve Wsgn. Figure 2 After step S06, the process proceeds to step S07.

[0069] In step S07, Figure 6 As shown in (b) of FIG. 1 , the limiter section 26 generates a delayed input waveform Wind, which is a waveform obtained by shifting the input waveform Win in the time axis direction Dt toward the positive temporal direction Dtp by a predetermined waveform delay time D. This waveform delay time D is, for example, a constant value. In other words, the delayed input waveform Wind is a waveform obtained by parallel shifting the input waveform Win toward the positive temporal direction Dtp by the waveform delay time D.

[0070] It should be noted that the magnitude relationship between the gain delay time St, the waveform delay time D, and the peak hold time Ht is "St≤D≤Ht". For example, in this embodiment, the gain delay time St, the waveform delay time D, and the peak hold time Ht are "St=D=Ht=1msec". Figure 2 After step S07, the process proceeds to step S08.

[0071] In step S08, Figure 6 As shown in steps (a) through (c) of FIG5 , the limiter unit 26 generates speech data Dsd by multiplying the delayed input waveform Wind by the smoothed gain curve Wsgn. This multiplication of the delayed input waveform Wind by the smoothed gain curve Wsgn generates speech data Dsd. In other words, the product of the level Lv of the delayed input waveform Wind and the gain Gn of the smoothed gain curve Wsgn is calculated, and this calculated product is used as the level Lv of the waveform Wsd represented by the speech data Dsd. The limiter 26 outputs the speech data Dsd generated in step S08 to the DAC 12.

[0072] Through the processing of steps S03 to S08 described above, the limiter unit 26 controls the amplitude of the waveform Wsd represented by the speech data Dsd to below the limiter threshold Th. Furthermore, by continuously adjusting the gain Gn, the limiter unit 26 can produce speech data Dsd with suppressed distortion of the waveform Wsd. However, the waveform Wsd represented by the speech data Dsd output by the limiter unit 26 is delayed relative to the input waveform Win by a waveform delay time D.

[0073] exist Figure 2 In step S09 , the limiter unit 26 does not suppress the level Lv of the waveform Wsd represented by the speech data Dsd with respect to the level Lv of the input waveform Win acquired in step S01 , but outputs the speech data Dsd at the original level Lv.

[0074] above Figure 2 The processing in each step constitutes a functional unit that implements its respective function. Specifically, steps S01 and S02 correspond to the excess determination unit, step S03 corresponds to the absolute value waveform generation unit, step S04 corresponds to the peak hold waveform generation unit, and step S05 corresponds to the gain curve generation unit. Furthermore, step S06 corresponds to the smoothing unit, steps S07 and S08 correspond to the speech data generation unit, and step S09 corresponds to the level non-suppression unit. The limiter 26 includes these excess determination units, absolute value waveform generation units, peak hold waveform generation units, gain curve generation units, smoothing units, speech data generation units, and level non-suppression units.

[0075] As described above, according to this embodiment, Figure 3 、 Figure 6 As shown, when the input waveform Win includes the excess portion EX of the level Lv, the limiter section 26 generates the utterance data Dsd by multiplying the delayed input waveform Wind by the smoothed gain curve Wsgn.

[0076] Therefore, the sound data Dsd is generated based on the input waveform Win. At the same time, by multiplying the delayed input waveform Wind by the smooth gain curve Wsgn, the waveform Wsd (see Figure 6 The amplitude of (c) is limited to be below the limiter threshold Th. Furthermore, since the waveform Wsd of the amplitude-limited speech data Dsd is smoothly formed, abnormal sounds such as harmonic distortion caused by the amplitude limitation can be suppressed.

[0077] In particular, the input waveform Win input to the limiter section 26 becomes Figure 1 The waveform is synthesized in the mixing section 22 and the frequency characteristics are corrected in the equalizer section 24. Therefore, it is difficult to predict the excess level Lv included in the input waveform Win (see Figure 3 (a) in FIG. 1 ). In such a case, the limiter unit 26 of this embodiment can also control the amplitude of the waveform Wsd to be below the limiter threshold Th as described above, and generate the sound data Dsd in which abnormal sounds such as harmonic distortion are suppressed.

[0078] Furthermore, in this embodiment, the reduction in sound pressure can be achieved by setting the duration for which the amplitude of the waveform Wsd of the sound output data Dsd is limited—that is, the sum of the gain delay time St, the peak hold time Ht, and the release time Rt—to a value sufficiently shorter than a time corresponding to human sensitivity to volume changes, such as the 125 msec time constant of the Fast characteristic. For example, the sum of the gain delay time St, the peak hold time Ht, and the release time Rt can be set to 20 msec or less. This prevents a reduction in auditory sound pressure, or a decrease in perception, by limiting the amplitude only for a short period near the portion EX where the level Lv exceeds the input waveform Win.

[0079] By executing the above Figure 2 The control processing structure can provide a sound system 8 that does not impair the perception of the notification sound, has the freedom of simultaneous sound emission, the freedom of correction, and suppresses the product size and cost.

[0080] Furthermore, according to this embodiment, when the input waveform Win does not include a portion EX exceeding the level Lv, the limiter unit 26 outputs the sound data Dsd without suppressing the level Lv of the waveform Wsd represented by the sound data Dsd, relative to the level Lv of the input waveform Win. Therefore, the amplitude of the waveform Wsd represented by the sound data Dsd is not unnecessarily suppressed.

[0081] (Other embodiments)

[0082] (1) In the above embodiment, the sound system 8 is, for example, a vehicle approach notification device mounted on an automobile. However, this is merely an example. The sound system 8 does not need to be a vehicle approach notification device, and various uses of the sound system 8 are contemplated. Furthermore, the sound system 8 may not be mounted on an automobile.

[0083] (2) In the above embodiment, Figure 2 In step S04, the limiter section 26 generates Figure 4 The peak hold waveform Wp is generated based on the waveform aggregate 30, but this is an example. As a result, the peak hold waveform Wp is formed as long as it is consistent with the high level side profile 301 of the waveform aggregate 30 (refer to Figure 3 The waveform shape may be the same as the shape of (b) in FIG. 1 , and the limiter unit 26 may not generate the waveform aggregate 30 .

[0084] (3) In the above embodiment, Figure 2 Each process executed by the microcomputer 10, such as the process of each step shown in the flowchart, is implemented by a computer program, but may also be implemented by hardware.

[0085] (4) It should be noted that the present disclosure is not limited to the above-described embodiments and can be implemented in various modifications. In addition, it goes without saying that the elements constituting the embodiments in the above-described embodiments are not essential, except for cases where they are clearly indicated as essential or where they are obviously essential in principle.

[0086] Furthermore, in the above-described embodiments, when reference is made to numerical values ​​such as the number, value, amount, and range of components of the embodiments, these values ​​are not limited to specific values, except where specifically indicated as essential or where, in principle, they are clearly limited to specific values. Furthermore, in the above-described embodiments, when reference is made to materials, shapes, positional relationships, and the like of components, these values ​​are not limited to specific materials, shapes, positional relationships, and the like, except where specifically indicated as essential or where, in principle, they are clearly limited to specific materials, shapes, positional relationships, and the like.

Claims

1. A sound control device (26) included in a sound system (8) in which a sound-generating body (80) generates sound based on a sound signal (Ssa) based on sound data (Dsd) representing a sound waveform, the sound control device generating the sound data, the sound control device comprising: an excess determination unit (S01, S02) that acquires an input waveform (Win) representing a relationship between a level (Lv) and time (Tm) of a voltage corresponding to the speech signal and serving as a basis for the speech data, and determines whether the input waveform includes an excess portion (EX) in which an absolute value of the level exceeds a predetermined limiter threshold value (Th); an absolute value waveform generating unit (S03) for generating an absolute value waveform (Wab) by converting the level in the input waveform into an absolute value of the level when the excess determining unit determines that the input waveform includes the excess portion; a peak hold waveform generating section (S04) for generating a peak hold waveform (Wp) representing a relationship between the level and the time in the input waveform, that is, an input waveform time (Tm), based on the absolute value waveform; a gain curve generating unit (S05) for generating a gain curve (Wgn) indicating a relationship between a gain (Gn) obtained based on the level in the peak hold waveform and a time period of the input waveform; a smoothing processing unit (S06) for performing a smoothing process on the gain curve to obtain a smoothed gain curve (Wsgn) based on the gain curve; and a speech data generating unit (S07, S08) for generating the speech data by multiplying the delayed input waveform (Wind) shifted by a predetermined waveform delay time (D) in the time axis direction (Dt) toward the time passing side of the input waveform, i.e., the time positive side (Dtp), by the smoothed gain curve, The peak hold waveform has a waveform shape obtained by extracting an outline (301) formed on the side (DLp) with a high level in the waveform aggregate (30), and the waveform aggregate includes: the absolute value waveform; a peak hold portion (32), which is provided at each maximum point of the level in the absolute value waveform, and extends linearly from the maximum point to the positive time side while showing the same level as the maximum point for a predetermined peak hold time (Ht) until reaching the absolute value waveform; and a release portion (33), which is provided at each peak hold portion extending for the peak hold time, and extends from an end (32a) on the positive time side of the peak hold portion toward the lower level side (DLm) as it moves toward the positive time side until reaching the absolute value waveform. The gain of the gain curve is set to 1 within a range (R1t, R2t) in the time axis direction in which the level in the peak hold waveform is below the limiter threshold, and the gain of the gain curve is set to a value obtained by "Gn=Th / Lv" within a range (R3t, R4t) in the time axis direction in which the level in the peak hold waveform exceeds the limiter threshold, where the gain is Gn, the level is Lv, and the limiter threshold is Th. The smooth gain curve is set to shift the gain delay time (St) toward the positive side of the time by comparing peak portions (36a, 37a) where the gain becomes maximum or minimum with respect to the gain curve. The waveform delay time is longer than the gain delay time, and the peak hold time is longer than the waveform delay time.

2. The sound control device according to claim 1, wherein: have: The level non-suppression unit (S09) outputs the speech data without suppressing the level of the waveform represented by the speech data with respect to the level of the input waveform when the excess determination unit determines that the input waveform does not include the excess portion.

Images