JP2012054863A - Sound reproducing apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correct a transfer characteristic of a sound signal to the optimal transfer characteristic which is free from an excess input into a loudspeaker or causing signal distortion.SOLUTION: A target transfer characteristics generation unit 4, in generating a target transfer characteristic with an amplitude for every frequency planarized by correcting a transfer characteristic from a loudspeaker 8 to a listening position 11, rolls off a low region and high region, and while planarizing an intermediate part, makes a correction of a peak and dip part with a large correction amount restrictive. A correction coefficient generation unit 3 derives a correction coefficient by identifying an inverse characteristic of this target transfer characteristic, and a non-circulation type digital filter unit 2 corrects the transfer characteristic of the sound signal using this correction coefficient to be supplied to the loudspeaker 8.

Description

  The present invention relates to a sound reproducing device that corrects an audio signal with a non-recursive digital filter to obtain a substantial sound quality in the original audio signal.

  Conventionally, there are various factors that hinder reproduction faithful to an original audio signal in an audio reproduction system of various AV (Audio Visual) devices such as a television receiver. For example, when a structure such as a sound conduit and a punching plate is present on the front surface of the speaker, the sound quality deteriorates due to loss of fidelity such as generation of a peak or dip due to sound conduit resonance and high-frequency attenuation. In addition, the environment in which the audio reproduction system exists may cause a loss of this fidelity. For example, when there is a dominant reflected wave comparable to a direct wave as a sound transmission path from the AV device to the listening position, or when there is an object or person that attenuates the sound wave.

  On the other hand, conventionally, the digital filter is used to correct the transfer characteristic from the speaker to the listening position, and more specifically, the sound pressure and group delay characteristic of the speaker system including the transfer space is flattened. Attempts have been made to obtain sound quality that is faithful to audio signals.

  For example, in the invention according to Patent Document 1, a convolver that performs a convolution operation processing on a signal from a sound source according to a set filter coefficient, and an impulse response obtained by actual measurement, only the mid-range is flattened and the low-range An acoustic reproduction apparatus is disclosed that includes a coefficient supply unit that obtains a calculated inverse filter coefficient that is processed to be the same as the original transfer characteristic in a high frequency range. There is also disclosed an acoustic reproduction device including a calculation means for obtaining a coefficient for an inverse filter that converges by targeting a characteristic that is flat only in the middle range and is the same as the original transfer characteristic in the low and high ranges.

Japanese Patent Laid-Open No. 8-110783

  Since the sound reproduction device of Patent Document 1 is configured as described above, when the difference between the actual transfer characteristic and the flat characteristic when the speaker output is faithful to the original audio signal is large, the sound reproduction apparatus is flat. There is a possibility that an excessive correction amount is generated due to the conversion. However, there is no mention of this, and there is a concern that excessive input may be supplied to the speaker when the difference between characteristics is large, or distortion on the signal may occur due to correction was there.

  The present invention has been made to solve the above-described problems. Each speaker system does not cause excessive input or signal distortion regardless of the characteristics of the actual transfer function. The inverse filter coefficient that can be corrected to the optimal transfer characteristics can be easily realized, which is always appropriate to the characteristics, that is, does not increase the distortion to the extent that it can be perceived acoustically and signalally. An object is to provide a sound reproducing device.

  The sound reproduction apparatus according to the present invention includes a correction coefficient that is a reverse characteristic of the target transfer characteristic generated by correcting the transfer characteristic from the speaker to the listening position by a correction amount within a predetermined limit range by the non-circular digital filter unit. The audio signal filtered using is supplied to a speaker for acoustic emission.

  According to the present invention, the target transfer characteristic used in generating the correction coefficient used in the acyclic digital filter unit is always appropriate in accordance with the characteristics of each speaker system regardless of the characteristics of the actual transfer function, that is, the sound. And optimal transmission characteristics that do not increase distortion to a level that can be perceived in terms of sound and signal, so that optimal correction can be made at all times without causing excessive input to the speaker or causing signal distortion. it can.

It is a block diagram which shows the structure of the sound reproduction apparatus which concerns on Embodiment 1 of this invention. It is a graph which shows an example of the reproduction | regeneration characteristic in the listening position of the sound reproduction apparatus shown in FIG. 3 is a graph showing a reverse characteristic of the reproduction characteristic shown in FIG. It is a graph which shows the ideal listening characteristic flattened in all the bands. It is a graph which shows the limited reverse characteristic which limitedly corrected the low region and the high region among the reverse properties of the reproduction characteristic shown in FIG. It is a graph which shows a listening characteristic at the time of carrying out filter correction by the correction coefficient based on the limited inverse characteristic shown in FIG. It is a figure which shows the structure of the calculation block which calculates the correction coefficient of the conventional acyclic digital filter part. FIG. 2 is a diagram illustrating a configuration of a calculation block that calculates a correction coefficient of an acyclic digital filter unit in the sound reproduction device illustrated in FIG. 1. FIG. 6 is a graph showing a limited inverse characteristic in which a region having a larger correction amount is limitedly corrected with respect to the limited inverse characteristic shown in FIG. 5. It is a graph which shows a listening characteristic at the time of carrying out filter correction by the correction coefficient based on the limited inverse characteristic shown in FIG. It is a figure which shows an example of the characteristic setting method of the target characteristic setting part of the sound reproduction apparatus shown in FIG. It is a figure which shows another example of the characteristic setting method of the target characteristic setting part of the sound reproduction apparatus shown in FIG. It is a block diagram which shows the structure of the sound reproduction apparatus which concerns on Embodiment 2 of this invention. It is a figure which shows an example of the characteristic setting method of the target characteristic setting part of the sound reproduction apparatus shown in FIG. It is a block diagram which shows the modification of the sound reproduction apparatus which concerns on Embodiment 2 of this invention. It is a block diagram which shows the structure of the sound reproduction apparatus which concerns on Embodiment 3 of this invention. It is a block diagram which shows the modification of the sound reproduction apparatus which concerns on Embodiment 3 of this invention.

Embodiment 1 FIG.
As shown in FIG. 1, the sound reproduction apparatus according to Embodiment 1 has an audio signal input terminal 1 for inputting an audio signal (audio signal) to an acyclic digital filter unit 2 and transfer characteristics of the input audio signal. An acyclic digital filter unit 2 that is variable by filter correction, a correction coefficient generation unit 3 that generates a correction coefficient that is used by being mounted on the acyclic digital filter unit 2, and a target transmission that is used to generate the correction coefficient A target transfer characteristic generating unit 4 for generating characteristics, an impulse response input terminal 5 for inputting an impulse response to the correction coefficient generating unit 3 and the target transfer characteristic generating unit 4, and setting information of the target transfer characteristic for the target transfer characteristic generating unit 4 A setting input terminal 6 for input to the power supply, a power amplifier 7 for amplifying the filtered audio signal, a speaker 8 for reproducing and outputting the amplified audio signal, and an output sound (sound wave) in a predetermined direction. A sound guide tube 9 for transmission to, and a punching plate 10 for protecting the speaker 8.

As described above, the sound conduit 9 and the punching plate 10 existing in front of the speaker 8 are factors that obstruct reproduction faithful to the audio signal input to the audio signal input terminal 1. Here, when a television receiver is considered as an application example of the speaker system including the configuration from the speaker 8 to the punching plate 10, the punching plate 10 is a porous acoustic resistance portion that hits the front portion of the speaker on the front panel. Reference numeral 9 denotes a resin-molded portion that joins the front edge of the speaker 8 and the back surface of the punching plate 10, and forms a predetermined volume of a front chamber between the speaker 8 and the punching plate 10.
The above speaker system is an example, but the design design and acoustic structure vary widely depending on the system, and accordingly, the acoustic characteristics, that is, how the peak dip appears are also varied. Even if design uniformity is achieved, the acoustic characteristics will change if the rear structure and the speaker unit are different. Therefore, acoustic correction is required in accordance with the characteristics specific to each object.

  Also, the transfer function H1 shown in FIG. 1 corresponds to the inverse characteristic of the frequency amplitude characteristic of the speaker 8 only. The transfer function H2 corresponds to the inverse characteristic of the frequency amplitude characteristic of only the sound conduit 9. The transfer function H3 corresponds to the inverse characteristic of the frequency amplitude characteristic of the punching plate 10 only. The transfer function H4 corresponds to the inverse characteristic of the frequency amplitude characteristic of only the acoustic space from the output end of the punching plate 10 to the listening position 11. The transfer function H0 corresponds to an inverse characteristic of the total frequency amplitude characteristic from the speaker 8 to the listening position 11, and the transfer function H0a corresponds to a limited inverse characteristic (details will be described later) of the frequency amplitude characteristic.

  Each transfer function H0, H0a, H1 to H4 is calculated by a calculation block as necessary, and the calculation block is constituted by a calculation device such as a computer. In FIG. 1, the H0 calculation block 100 calculates the transfer function H0, the H0a calculation block 100a calculates the transfer function H0a, the H1 calculation block 101 calculates the transfer function H1, and the H2 calculation block 102 calculates the transfer function H2. Then, the H3 calculation block 103 calculates the transfer function H3, and the H4 calculation block 104 calculates the transfer function H4. In the first embodiment, it is assumed that the correction coefficient generation unit 3 corresponds to the H0a calculation block 100a, and calculates the transfer function H0a.

  The target transfer characteristic generation unit 4 receives a numerical sequence from the impulse response input terminal 5 and converts it from a time axis to a frequency axis, an FFT (Fast Fourier Transform) unit 4 a, an output from the FFT unit 4 a, and a setting input terminal 6. The target characteristic setting unit 4b that receives the setting information and generates the target transfer characteristic, and the inverse FFT unit 4c that returns the output of the target characteristic setting unit 4b from the frequency axis to the numerical value sequence on the time axis.

Next, the operation of the sound reproducing device will be described.
FIG. 2 is an example of a reproduction characteristic of a sound reproduction device using the speaker 8 as an electroacoustic transducer, and shows a radiated acoustic characteristic (transfer characteristic) at the listening position 11 of the speaker 8 including the sound conduit 9 and the punching plate 10. It is a graph to show. In the configuration of FIG. 1, the solid line represents the listening position 11 when a correction coefficient with a transfer characteristic of 1 is implemented in the acyclic digital filter unit 2 or when the acyclic digital filter unit 2 is bypassed. Corresponds to radiation acoustic characteristics. In FIG. 2, the horizontal axis of the graph indicates frequency [Hz], and the vertical axis indicates level [10 dB / div]. A characteristic indicated by a broken line as a target (hereinafter referred to as a target flat characteristic or a target flat position) indicates the case of the transfer characteristic 1, and indicates that a radiated acoustic characteristic faithful to the audio signal is obtained at the listening position 11. If the filter correction by the non-recursive digital filter unit 2 is not performed, a peak, a dip, or the like that is far from the target flat characteristic occurs.

  FIG. 3 is a graph showing the inverse characteristic of the radiated acoustic characteristic at the listening position 11 shown in FIG. 1 and is a transfer characteristic represented by the transfer function H0 obtained by the H0 calculation block 100. In the graph of FIG. 2, the greater the distance between the radiated acoustic characteristics indicated by the solid line and the target flat characteristic indicated by the broken line, the greater the correction amount in FIG. 3. In particular, when the correction amount on the plus side (upward direction of the graph) is large, it is conceivable that an excessive input is supplied to the speaker 8 or a distortion on the signal occurs. Conversely, when the amount of correction on the negative side (downward direction of the graph) is large, the quantization distortion increases.

  4 is flattened over the entire band, ideal when an audio signal that has been subjected to filter correction by the acyclic digital filter unit 2 using a correction coefficient based on the inverse characteristic of the radiated acoustic characteristic shown in FIG. 3 is reproduced. It is a graph which shows the listening characteristic (solid line). In the graph of FIG. 4, the broken line indicates the listening characteristic before filter correction.

  However, there is a limit to giving an excessive correction amount to the reproduction capability of the speaker 8 for the audio signal, and particularly in the low range, the limit of the diaphragm operating range of the speaker 8 is reached near the lowest resonance frequency. There may be a case where the speaker operation is in a broken state, for example, an abnormal sound is generated due to a beating phenomenon (operation limit range). That is, in the first place, it is physically difficult to completely flatten the low frequency characteristics by the signal correction of the acyclic digital filter unit 2. In addition, if the amplitude is increased for flattening the high frequency characteristics, the distortion increases. Therefore, it can be said that it is appropriate to have a roll-off characteristic in the low range and the high range. This is also taken into consideration in Patent Document 1.

Hereinafter, a case where the low-frequency and high-frequency characteristics are rolled off will be described.
FIG. 5 is a graph showing an inverse characteristic (hereinafter referred to as a limited inverse characteristic) when the low frequency and the high frequency of the inverse characteristics of the radiated acoustic characteristic shown in FIG. 3 are limitedly corrected. Compared to the reverse characteristic of FIG. 3 indicated by the broken line, the correction amount of the low range and the high range is small in the limited reverse characteristic indicated by the solid line.

  FIG. 6 is a graph showing a listening characteristic (solid line) when an audio signal that has been filter-corrected by the acyclic digital filter unit 2 using a correction coefficient based on the limited inverse characteristic of the radiated acoustic characteristic shown in FIG. 5 is reproduced. is there. The low range and high range roll off, and the middle part (mid range) is flattened. In the graph of FIG. 6, the broken line indicates the listening characteristic before correction.

However, even in the case of limited correction as shown in FIG. 5 and FIG. 6, if the peak and dip of the intermediate part are large, the correction amount of the peak and dip part becomes large at the time of flattening. There may be an excessive input or signal distortion.
Therefore, in the sound reproduction device according to the first embodiment, when the transfer characteristic from the speaker 8 to the listening position 11 is corrected to approach the target flat characteristic, a correction amount is set for the transfer characteristic having a large deviation from the target flat characteristic. Limited correction to be limited is performed. Hereinafter, a characteristic in which the transfer characteristic from the speaker 8 to the listening position 11 is limited and corrected to be close to the target flat characteristic is referred to as a target transfer characteristic. Then, the correction coefficient generation unit 3 derives a correction coefficient having an inverse characteristic with respect to the target transfer characteristic generated by the target transfer characteristic generation unit 4 and outputs the correction coefficient to the acyclic digital filter unit 2.

Here, a method for deriving the transfer function H0 will be described.
FIG. 7 is a block diagram showing the configuration of the H0 calculation block 100 for deriving the transfer function H0. Assume that there is a system in which when an impulse signal is input from the input terminal 200 as an acoustic input signal, the delay signal is delayed by Δt time by a delay unit 204 and the same impulse signal as the input is output. The output signal in this system has a characteristic that the amplitude characteristic is constant at all frequencies and the group delay characteristic is constant (the phase characteristic is linear with respect to the frequency). An acoustic characteristic in which the amplitude characteristic is constant, that is, the sound pressure frequency characteristic is flat and the phase characteristic is linear, is one ideal of an audio reproduction system, and a method for realizing this characteristic with an acyclic digital filter (FIR filter). Indicates.

  In FIG. 7, a signal 205 obtained by multiplying a transfer function Hs 202 of the target (speaker system) to be corrected by a coefficient of the adaptive filter 203 is an error from the output signal 206 of the delay unit 204. By performing the adaptive signal processing so that 201 is minimized, the transfer function H0 having the inverse characteristic of the target transfer function Hs is identified, and the coefficient of the adaptive filter 203 that realizes the transfer function H0 is determined. In practice, a microphone is installed at the listening position 11 shown in FIG. 1, and the collected acoustic characteristics are given to the system as a target transfer function Hs202, and the adaptive filter 203 uses the least mean square (LMS) algorithm to reverse the characteristics. Is identified. The H0 calculation block 100 shown in FIG. 1 shows the above steps for deriving the transfer function H0. As the adaptive signal processing, any method may be used as long as it can identify the inverse characteristic of the target transfer function in addition to the LMS algorithm.

  Further, in FIG. 7, the frequency characteristic of the output signal 206 of the delay device 204 is set to a flat characteristic indicated by a broken line in FIG.

  In contrast, FIG. 8 illustrates a case where the target transfer characteristic is not a complete impulse signal, that is, the reverse characteristic is identified from the limited corrected target transfer characteristic instead of the completely flat target flat characteristic. It is a block diagram. When a time signal having a target transfer characteristic is input from the input terminal 307 as an acoustic input signal, it is assumed that there is a system that outputs the same time signal 306 as the input after being delayed by Δt time by a delay unit 304. The output signal in this system has a characteristic that the amplitude characteristic is constant except for a specific frequency range, and the group delay characteristic is constant (the phase characteristic is linear with respect to the frequency). On the other hand, when a complete impulse signal is input as an acoustic input signal from the input terminal 300, a signal obtained by multiplying the transfer function Hs302 of the target (speaker system) to be corrected by the coefficient of the adaptive filter 303 305 performs adaptive signal processing so that the error 301 from the output signal 306 by the delay unit 304 is minimized, whereby a transfer function H0a having a limited inverse characteristic of the target transfer function Hs is identified, and The coefficient of the adaptive filter 303 that realizes the transfer function H0a is determined. Other steps are the same as those described with reference to FIG.

  When the configuration of FIG. 8 is applied to FIG. 1, the impulse transfer input terminal 5 corresponds to the target transfer function Hs302, and the output of the target transfer characteristic generation unit 4 is output to the acoustic input signal input to the input terminal 307. Equivalent to. Further, the coefficient identified by the adaptive filter 303 corresponds to the output of the correction coefficient generation unit 3, and this output is mounted on the acyclic digital filter unit 2. Therefore, the operation of the correction coefficient generation unit 3 is nothing but the process of identifying the transfer function H0a having the limited inverse characteristic of the target transfer function shown in FIG.

  FIG. 9 shows the limited inverse characteristic when a region with a large correction amount is further added as a limited correction region to the limited inverse property shown in FIG. FIG. 10 is a graph showing a listening characteristic (solid line) when an audio signal that has been filter-corrected by the acyclic digital filter unit 2 using a correction coefficient based on the limited inverse characteristic of the radiated acoustic characteristic shown in FIG. 9 is reproduced. is there. The low range and the high range are rolled off in the same manner as in FIG. 6, and in addition, in FIG. 9, the correction of the region (peak and dip) having a large correction amount is limited while the intermediate portion is flattened. Therefore, in the dip portion, it is possible to prevent a situation in which the correction amount for the acoustic signal becomes large during flattening, and thus it is possible to avoid a situation in which excessive input or signal distortion occurs in the speaker 8. Further, since the correction amount is similarly reduced at the peak portion, it is possible to avoid a situation in which the bit accuracy is unnecessarily lowered to cause sound quality deterioration and quantization distortion.

A specific operation will be described below.
FIG. 11 is a diagram illustrating an example of a characteristic setting method of the target characteristic setting unit 4b, which is a screen display example. In this example, it is assumed that the target transfer characteristic generation unit 4 includes a display device and an input device such as a keyboard or a mouse, and the output of the input device is input to the setting input terminal 6. The target transfer characteristic generated by the target transfer characteristic generation unit 4 is displayed on the screen of the display device, and the user interactively inputs the target transfer characteristic setting information using the input device.

  The impulse signal data string input from the impulse response input terminal 5 is converted into a frequency axis data string by the FFT unit 4a and input to the target characteristic setting unit 4b. When the frequency axis data string is input, the target characteristic setting unit 4b displays a frequency characteristic graph simulated based on only the amplitude information. This is the state shown in FIG. 11, and the broken line shows the initial characteristics simulated based on the amplitude information. Then, the target characteristic setting unit 4b selects the component point of the frequency characteristic based on the setting information received from the setting input terminal 6, moves it in the amplitude direction, and adjusts the target transfer characteristic to obtain the desired characteristic. Generate. In the drawing, white circles indicate frequency characteristic component points, and when each component point is moved and adjusted as indicated by an arrow, the characteristic-adjusted characteristic indicated by a solid line is obtained. That is, the setting information supplied from the setting input terminal 6 is the movement information of the characteristic component point.

  However, instead of moving each component point so as to completely match the target flat position, limited correction is performed so that the movement amount is within a predetermined limit range. In addition, the frequency characteristics of the high and low frequencies are not corrected or limitedly corrected.

In the illustrated example, the constituent points of the dip portion are mainly moved, but each constituent point indicates a target portion for adjustment, and the constituent points may be provided in other portions. In the illustrated example, the concept of smoothly changing the characteristics is shown in which not only the respective constituent points are moved but also the peripheral constituent points are accompanied, but the present invention is not limited to this form.
The characteristic may be adjusted in any way. For example, as described above, the user may input setting information about each component point interactively while operating the input device, or all the component points may be input. You may input the setting information about at once.

  The data string of the target transfer characteristic generated by the target characteristic setting unit 4b is input to the inverse FFT unit 4c, and the inverse FFT unit 4c converts it into a data string on the time axis and outputs it to the correction coefficient generation unit 3. Then, as described with reference to FIG. 8, the correction coefficient generation unit 3 receives the limitedly corrected target transfer characteristic from the input terminal 300, and generates a transfer function H0a that is a limited inverse characteristic of the frequency amplitude of the target transfer characteristic. Will be identified.

FIG. 12 is a diagram illustrating another example of the characteristic setting method of the target characteristic setting unit 4b. In the example of FIG. 12, the setting information received by the setting input terminal 6 is a frequency range in which correction is performed and a maximum correction amount (hereinafter, maximum correction amount). Based on the setting information, the target characteristic setting unit 4b determines the target transfer characteristic so that the amplitude approaches the target flat characteristic within the limit range of the maximum correction amount in the frequency range to be corrected. Note that high and low frequencies outside the frequency range are not corrected.
Specifically, for example, a difference is obtained by comparing the amplitude level at a frequency having an initial characteristic indicated by a broken line in the figure with the level of the target flat position, and the difference is compared with the maximum correction amount so that the difference is maximum. If the correction amount is exceeded, the value is shifted toward the target flat position by the maximum correction amount from the amplitude level of the initial characteristic. In FIG. 12, not only the amplitude level of the initial characteristic is simply shifted to the target flat position, but also the point adjacent to the shifted point is adjusted so as to be smoothly coupled to the target flat position. It is not limited.

  As described above, in the sound reproduction device according to the first embodiment, the acyclic digital filter unit 2 that performs filter correction on the transfer characteristics of the input audio signal using the correction coefficient, and the acyclic digital filter unit 2 output. A speaker system having a power amplifier 7 for amplifying an audio signal, a speaker 8 for acoustically radiating an audio signal output from the power amplifier 7, a sound conduit 9 and a punching plate 10, and transfer characteristics from the speaker 8 to the listening position 11. A target transfer characteristic generation unit 4 that generates a target transfer characteristic by correcting with a correction amount within a predetermined limit range, and a correction coefficient generation unit 3 that derives a reverse characteristic of the target transfer characteristic and generates a correction coefficient. Configured. For this reason, the target characteristic used in generating the correction coefficient used in the acyclic digital filter unit 2 is always appropriate in accordance with the characteristics of each speaker system regardless of the characteristics of the actual transfer function, that is, acoustically. Optimal transfer characteristics that do not increase the distortion to such an extent that it can be recognized in terms of signal audibility can be obtained, so that it is possible to always perform optimal correction without causing excessive input to the speaker 8 or causing signal distortion.

  Further, according to the first embodiment, the target transfer characteristic generation unit 4 receives an impulse response indicating the transfer characteristic from the speaker 8 to the listening position 11 as an input, and sets a numerical sequence of the impulse response from the time axis to the frequency axis. The amplitude is corrected by the correction amount specified by the setting information input from the setting input terminal 6 with respect to the transfer characteristic based on the amplitude information in the FFT unit 4a to be converted and the numerical value sequence of the frequency axis converted by the FFT unit 4a. The target characteristic setting unit 4b that generates the target transfer characteristic and the inverse FFT unit 4c that converts the target transfer characteristic generated by the target characteristic setting unit 4b from the frequency axis to the time axis are configured. For this reason, by discarding the phase information from the actual transfer characteristics and using only the amplitude information in the processing of the FFT unit 4a, the target transfer characteristics matching the characteristics of the speaker system can be easily set.

  In addition, according to the first embodiment, the target characteristic setting unit 4b determines the frequency range of the transfer characteristics based on the frequency range specified by the setting information input from the setting input terminal 6 and the correction amount limit range. Is corrected by a correction amount within the limit range. For this reason, it is possible to prevent overcorrection over the entire frequency range of the transfer characteristic.

Embodiment 2. FIG.
FIG. 13 is a block diagram showing a configuration of the sound reproduction device according to the second embodiment. In the second embodiment, instead of the setting input terminal 6 of the first embodiment shown in FIG. 1, the measurement signal generation unit 12, the measurement input terminal 13, the distortion determination unit 14, the input switching unit 15, Is newly provided. Accordingly, the operation of the target characteristic setting unit 4b is also changed. In addition, about the component other than this, since it is the same as that of FIG. 1, or equivalent, it attaches | subjects the same code | symbol and abbreviate | omits description.

  In the sound reproducing apparatus shown in FIG. 13, when an audio signal is input from the audio signal input terminal 1, the radiated acoustic characteristic at the listening position 11 is as shown in FIG. Similar to the above, the correction coefficient generation unit 3 generates a correction coefficient based on the impulse response input from the impulse response input terminal 5 according to the principle described with reference to FIG. 8 with the output of the target transfer characteristic generation unit 4 as a target. And mounted on the acyclic digital filter unit 2.

  FIG. 14 is a diagram illustrating a characteristic setting method of the target characteristic setting unit 4b according to the second embodiment. The graph of FIG. 14A represents an initial characteristic and a target flat characteristic extracted from the dip portion in an example of the radiated acoustic characteristic at the listening position 11. The graph of FIG. 14B represents the harmonic distortion generated by amplifying the dip portion when the initial characteristics are flattened. This dip is a part that occurs based on the acoustic structure and the speaker reproduction capability. If this dip is corrected to the target flat characteristic by the acyclic digital filter unit 2, signal amplification will occur. When the operation reaches the saturation region, harmonic distortion is caused, which is a problem.

  Therefore, the distortion determination unit 14 issues an instruction to the target characteristic setting unit 4b, determines a certain allowable value for harmonic distortion, and performs signal amplification exceeding the allowable value, as shown in the graph of FIG. Do not do it. As a result, flattening is limited as shown in the graph of FIG. 14D, and the correction is not performed sufficiently in the deep dip portion (that is, limited correction for correction with a correction amount within the limit range). )

  Specifically, the measurement signal generator 12 generates a sine wave signal (measurement signal). Then, the input switching unit 15 switches the input to the acyclic digital filter unit 2 from the audio signal input terminal 1 to the measurement signal generation unit 12 and inputs the sine wave signal to the acyclic digital filter unit 2. The radiated acoustic characteristic (distortion determination transfer characteristic) of the speaker 8 based on the sine wave signal is measured at the listening position 11 and input to the distortion determination unit 14 from the measurement input terminal 13.

Then, the distortion determination unit 14 compares the sine wave signal input from the measurement signal generation unit 12 with the radiation acoustic characteristic input from the measurement input terminal 13 to calculate the harmonic distortion. In general, harmonic distortion is expressed as the ratio of the power levels of the second and third harmonics to the power level of the fundamental signal. Therefore, the distortion determination unit 14 may extract the basic signal derived from the measurement signal generation unit 12 input from the measurement input terminal 13 and calculate it as the ratio between the second and third harmonics and the power level.
Further, the measurement signal generator 12 calculates the harmonic distortion in the distortion determination unit 14 while sequentially changing the frequency of the sine wave signal. Then, the distortion determination unit 14 outputs setting information to the target characteristic setting unit 4b so that the calculation result of the harmonic distortion for each measurement frequency falls within the preset allowable value range, and the target transfer characteristic Control the operation to match the level setting.

  When the target characteristic setting unit 4b receives the setting information from the distortion determination unit 14, for example, the target characteristic setting unit 4b generates a target transfer characteristic by moving the system in a direction to weaken the flattening based on the distortion determination from a state in which the initial characteristic is completely flattened. Alternatively, conversely, the target transfer characteristic may be generated by moving the system in the direction of flattening from the initial characteristic.

  As described above, the distortion determination unit 14 performs distortion determination so that the harmonic distortion falls within the allowable value, and the target characteristic setting unit 4b performs target transmission as illustrated in FIG. 14D based on the determination result. Deriving properties. In FIG. 13, the measurement signal generation unit 12 is used in a configuration in which the input switching unit 15 is used to switch to the audio signal input terminal 1. However, the measurement signal generation unit 12 is not limited to this configuration. The measurement signal may be input and supplied to the distortion determination unit 14 as well.

  Alternatively, the sound reproducing device may be configured as shown in FIG. The sound reproducing device having the configuration shown in FIG. 15 is configured to estimate instead of the configuration in which the measurement signal of the measurement signal generating unit 12 is input to the acyclic digital filter unit 2 and the radiated acoustic characteristics at the listening position 11 are measured. is there. Specifically, a virtual coefficient is calculated by convolving a correction coefficient used in the non-recursive digital filter unit 2 with a transfer characteristic from the speaker 8 to the listening position 11 by a calculation device (not shown). The measurement result corresponding to each measurement signal input is calculated as an estimation result. The estimation result is input from the estimation result input terminal 16 to the distortion determination unit 14 and the corresponding measurement signals are input from the measurement signal generation unit 12, thereby correcting the deep dip as described with reference to FIG. You can also work to suppress it. In the case of this configuration, the distortion determination unit 14 can perform distortion determination based on the calculated value without performing measurement at the listening position 11 for each frequency of the measurement signal. There is an advantage that can be set.

  As described above, the sound reproduction device according to Embodiment 2 generates a predetermined measurement signal and outputs the measurement signal to the acyclic digital filter unit 2, and the measurement signal generated by the measurement signal generation unit 12. And the measurement result of the distortion determination transfer characteristic up to the listening position 11 when the measurement signal output by filtering the non-recursive digital filter unit 2 is radiated by the speaker 8 is compared, and the distortion determination transmission And a distortion determination unit 14 that calculates the distortion amount of the characteristic. The target transfer characteristic generation unit 4 is configured to adjust the correction amount based on the distortion amount calculated by the distortion determination unit 14. For this reason, it is possible to determine the degree of distortion for each measurement signal using the measurement result at the listening position 11 related to the measurement signal as an input, and adjust the target transfer characteristic used to generate the correction coefficient based on the determination result. it can. Therefore, optimal transfer characteristics that are always appropriate to the characteristics of each speaker system regardless of the characteristics of the actual transfer function, i.e., do not increase distortion to the extent that they can be perceived acoustically and signalally. Therefore, it is possible to reliably perform the optimal correction without causing excessive input to the speaker 8 or causing signal distortion.

  Further, according to the second embodiment, the target characteristic setting unit 4b estimates the distortion determination transmission characteristic based on the correction characteristic of the non-circular digital filter unit 2 and the transmission characteristic from the speaker 8 to the listening position 11. Therefore, the target transfer characteristic can be adjusted quickly and easily. Even if the estimation result is used instead of the measurement result, it is always appropriate according to the characteristics of the speaker system regardless of the actual characteristics of the transfer function, that is, in terms of audibility both acoustically and signalally. Since the optimum transfer function does not increase the distortion to such an extent that it can be recognized, the optimum correction without causing excessive input to the speaker 8 or causing signal distortion can be reliably performed.

Embodiment 3 FIG.
FIG. 16 is a block diagram illustrating a configuration of the sound reproduction device according to the third embodiment. In the third embodiment, a setting input terminal 17 is newly provided in addition to the configuration of the second embodiment shown in FIG. Accordingly, the operation of the target characteristic setting unit 4b is also changed. In addition, about the component other than this, since it is the same as that of FIG. 13, or equivalent, the same code | symbol is attached | subjected and description is abbreviate | omitted.

  In the sound reproduction device shown in FIG. 16, the target characteristic setting unit 4 b is configured to receive the determination result of the harmonic distortion from the distortion determination unit 14 and the setting information from the setting input terminal 17. From this setting input terminal 17, for example, a frequency range for setting a target transfer characteristic is designated based on distortion determination by the distortion determination unit 14. Specifically, when setting information designating a frequency range as shown as “correction frequency range” in FIG. 12 is input from the setting input terminal 17, the target characteristic setting unit 4 b The low frequency range is not corrected, and the target transfer characteristics are set based on the distortion determination for the frequency band excluding those frequency regions. As a result, it is possible to limit the region where the characteristic setting based on the distortion determination is performed, leading to a reduction in the time required for the target transfer characteristic setting.

  Alternatively, the configuration shown in FIG. 17 may be used. The sound reproducing device having the configuration shown in FIG. 17 has a configuration in which a setting input terminal 17 is newly added to the modification of the second embodiment shown in FIG. Also in the case of this configuration, similarly to the configuration of FIG. 16, it is possible to limit the region where the characteristic setting based on the distortion determination by the target characteristic setting unit 4 b can be limited, and there is an advantage that the time required for setting the target transfer characteristic is shortened. In addition, since the distortion determination unit 14 can perform distortion determination based on the calculated value without performing measurement at the listening position 11 for each frequency of the measurement signal, the target transfer characteristic is set with an easy configuration. There are also advantages that can be made.

  Alternatively, in FIG. 16 and FIG. 17, the frequency range is designated by the setting information so that the measurement signal generation unit 12 and the distortion determination unit 14 are also designated. The time can be further shortened by generating a signal and allowing the distortion determination unit 14 to perform only the distortion determination in the frequency range.

  As described above, according to the third embodiment, the target transfer characteristic generation unit 4 corrects the frequency range specified by the setting information input from the setting input terminal 17 in the transfer characteristic with the correction amount based on the distortion amount. Configured to do. For this reason, it is possible to easily and quickly adjust the target transfer characteristic used for generating the correction coefficient. Even in this configuration, as in the second embodiment, it is always appropriate according to the characteristics of each speaker system regardless of the characteristics of the actual transfer function, that is, in terms of audibility in terms of acoustics and signals. Since the optimum transfer function does not increase the distortion to such an extent that it can be recognized, the optimum correction without causing excessive input to the speaker 8 or causing signal distortion can be reliably performed.

In the first to third embodiments, the audio reproduction system including the audio signal input terminal 1, the acyclic digital filter unit 2, the power amplifier 7, the speaker 8, the sound conduit 9, and the punching plate 10, and other configurations. Although the sound reproduction device is configured including the calculation system composed of elements, the calculation system may be configured by a different device.
For example, in the case of the first embodiment, a calculation system including the correction coefficient generation unit 3, the target transfer characteristic generation unit 4, the impulse response input terminal 5, and the setting input terminal 6 is configured by a separate calculation device. What is necessary is just to obtain | require a correction coefficient with an arithmetic unit, and to mount in the acyclic digital filter part 2 of a sound reproduction apparatus. Further, when the arithmetic unit is configured by a computer, an arithmetic program describing the processing contents of the correction coefficient generation unit 3 and the target transfer characteristic generation unit 4 is stored in the memory of the computer, and the CPU of the computer stores the memory in the memory. What is necessary is just to run the operation program currently performed.
Similarly, in the case of the second and third embodiments, the measurement signal generation unit 12, the measurement input terminal 13 (or the estimation result input terminal 16), the distortion determination unit 14, and the setting input terminal 17 are included as the calculation system. A body arithmetic unit may be configured.

In the first to third embodiments, the speaker 8 includes the sound conduit 9 and the punching plate 10 as an example. However, the speaker system may have other configurations, for example, the sound conduit 9. And the structure provided with one of the punching boards 10 may be sufficient. In the case of this configuration, the transfer function H0a is a transfer characteristic from the speaker 8 to the listening position 11 via the sound conduit 9, or a transfer characteristic from the speaker 8 to the listening position 11 via the punching plate 10.
It should be noted that the embodiments can be appropriately combined, changed, omitted, etc. within the scope of the present invention.

  DESCRIPTION OF SYMBOLS 1 Audio | voice signal input terminal, 2 Acyclic digital filter part, 3 Correction coefficient production | generation part, 4 Target transfer characteristic production | generation part, 4a FFT part, 4b Target characteristic setting part, 4c Inverse FFT part, 5 Impulse response input terminal, 6 setting Input terminal, 7 Power amplifier, 8 Speaker, 9 Sound conduit, 10 Punching plate, 11 Listen position, 12 Measurement signal generator, 13 Measurement input terminal, 14 Distortion judgment unit, 15 Input switching unit, 16 Estimation result input terminal, 17 Setting input terminal, 100 H0 operation block, 100a H0a operation block, 101 H1 operation block, 102 H2 operation block, 103 H3 operation block, 104 H4 operation block.

Claims (9)

An audio reproduction device that radiates sound by supplying an audio signal filter-corrected using a correction coefficient to a speaker by an acyclic digital filter unit,
The sound reproduction apparatus according to claim 1, wherein the correction coefficient is a reverse characteristic of the target transfer characteristic generated by correcting the transfer characteristic from the speaker to the listening position with a correction amount within a predetermined limit range. The correction factor is
A target transfer characteristic generating unit that generates a target transfer characteristic by correcting the transfer characteristic from the speaker to the listening position with a correction amount within a predetermined limit range;
The sound reproduction device according to claim 1, wherein the sound reproduction device is generated by a correction coefficient generation unit that derives a reverse characteristic of the target transfer characteristic and generates a correction coefficient. The target transfer characteristic generator is
An FFT unit that receives an impulse response indicating a transfer characteristic from the speaker to the listening position as input, and converts a numerical sequence of the impulse response from a time axis to a frequency axis;
A target characteristic setting unit that generates a target transfer characteristic by correcting the amplitude with a correction amount specified by the setting information with respect to the transfer characteristic based on the amplitude information in the numerical value sequence of the frequency axis converted by the FFT unit;
The sound reproducing apparatus according to claim 2, further comprising: an inverse FFT unit that converts the target transfer characteristic generated by the target characteristic setting unit from a frequency axis to a time axis.   The target characteristic setting unit corrects the amplitude of the frequency range of the transfer characteristic with the correction amount within the limit range based on the frequency range specified by the setting information and the limit range of the correction amount. The sound reproducing device according to claim 3.   The target transfer characteristic generation unit is for determining a distortion to a listening position when a predetermined measurement signal and a measurement signal output by filtering the measurement signal by the acyclic digital filter unit are emitted by the speaker. 3. The sound reproducing apparatus according to claim 2, wherein the correction amount is adjusted based on a distortion amount obtained by comparing the transfer characteristic.   6. The sound reproducing apparatus according to claim 5, wherein the distortion determining transfer characteristic is a result of measurement at a listening position.   6. The sound reproduction apparatus according to claim 5, wherein the distortion determination transfer characteristic is a result estimated based on a correction characteristic of the non-circular digital filter unit and a transfer characteristic from the speaker to a listening position. . The target transfer characteristic generator is
An FFT unit that receives an impulse response indicating a transfer characteristic from the speaker to the listening position as input, and converts a numerical sequence of the impulse response from a time axis to a frequency axis;
The frequency range to be corrected based on the distortion amount calculated by the distortion determination unit and the correction amount are obtained, and the frequency range is determined with respect to the transfer characteristic based on the amplitude information in the numerical value sequence of the frequency axis converted by the FFT unit. A target characteristic setting unit that corrects the amplitude of the signal with the correction amount and generates a target transfer characteristic;
6. The sound reproducing apparatus according to claim 5, further comprising an inverse FFT unit that converts the target transfer characteristic generated by the target characteristic setting unit from a frequency axis to a time axis.   The sound reproduction device according to claim 8, wherein the target characteristic setting unit performs correction with a correction amount based on a distortion amount in a frequency range specified by setting information of transfer characteristics.

Images