JP2013102327A - Deflection angle calculating device and sound processing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily calculate a complex number deflection angle.SOLUTION: A deflection angle calculating section 20 converts a real part Re[X] and an imaginary part Im[X] of a complex number X into a non-negative value by code inversion, and calculates a sum M and a difference S of the real part Re[X] and the imaginary part Im[X]. The deflection angle calculating section 20 calculates a ratio R of the difference S to the sum M, and calculates from the ratio R a provisional deflection angle φ of the complex number X. For calculating the provisional deflection angle φ, for example, an approximation formula "φ=(π/4)x(-R+1)" is applied. The deflection angle calculating section 20 converts the provisional deflection angle φ into a deflection angle θ corresponding to quadrant to which the complex number X before code inversion belongs.

Description

  The present invention relates to a technique for calculating an argument of a complex number.

  In order to calculate the argument of a complex number, an arc tangent calculation is required. Since direct calculation of arc tangent is difficult, a storage circuit (table) such as a ROM that stores arc tangent values in advance is generally used. However, problems such as enlargement of the memory circuit may occur. Therefore, Patent Document 1 discloses a technique for approximately calculating an arctangent.

JP 2001-285246 A

  In the technique of Patent Document 1, it is necessary to perform division between the I signal and the Q signal. However, since division between signals can be division by zero, special measures are needed to deal with the case of division by zero. In view of the above circumstances, an object of the present invention is to easily calculate the argument of a complex number.

  The declination calculating apparatus of the present invention is an apparatus for calculating the declination of a complex number, and includes a first conversion unit that converts a real part and an imaginary part of a complex number to a non-negative value by sign inversion, and after processing by the first conversion unit. Sum-difference calculating means for calculating the sum and difference between the real part and imaginary part, ratio calculating means for calculating the ratio of the difference to the sum, and declination for calculating the provisional argument of the complex number from the ratio calculated by the calculating means Provisional means, and second conversion means for converting the provisional deviation angle into a deviation angle corresponding to the quadrant to which the complex number before sign inversion belongs. In the above configuration, the argument of the complex number is calculated according to the ratio of the difference between the real part and the imaginary part of the complex number. Since the real part and the imaginary part are converted into non-negative values by sign inversion, the sum located in the denominator (divisor) of the ratio never becomes zero (that is, division by zero does not occur). Therefore, there is an advantage that the complex argument can be easily calculated as compared with a configuration that requires special measures to deal with division by zero.

In a preferred aspect of the present invention, the declination provisional means calculates the provisional declination φ from the ratio R by the following approximate expression (A).
φ = (π / 4) ・ (−R + 1) (A)
The above aspect has an advantage that the provisional declination can be easily calculated by the calculation of the approximate expression (A). However, in the present invention, a method for specifying the provisional deviation angle φ from the ratio R is arbitrary.

  The sound processing apparatus according to a preferred aspect of the present invention has a configuration to which the deflection angle calculating apparatus of each form exemplified above is applied. The acoustic processing apparatus according to the first aspect of the present invention includes a frequency analysis unit that generates a complex spectrum of an acoustic signal for each of the left and right channels, and a declination calculation unit that calculates a declination for each frequency of the complex spectrum for each of the left and right channels. And a phase adjustment means for reducing the difference in the deflection angle for each frequency of the left and right channels calculated by the deviation angle calculation means, and signal generation for generating left and right channel acoustic signals from the complex spectrum of each channel processed by the phase adjustment means Means for converting the real part and the imaginary part of the complex spectrum into non-negative values by sign inversion, and the real part and the imaginary part after processing by the first conversion means. A sum difference calculating means for calculating a sum and a difference; a ratio calculating means for calculating a ratio of the difference to the sum; a declination provisional means for calculating a temporary declination of a complex number from the ratio calculated by the ratio calculating means; The deflection angle complex number No. inversion before correspond to quadrants belonging and a second converting means for converting the provisional declination. In the above aspect, the left and right channel acoustic signals are generated from the complex spectrum of each channel after the difference in the deflection angle for each frequency of the left and right channels calculated by the deflection angle calculating means is reduced. Therefore, there is an advantage that the left and right channel acoustic signals with reduced reverberation components can be generated. In addition, the specific example of the above aspect is later mentioned as 2nd Embodiment, for example.

  The sound processing apparatus according to the second aspect of the present invention includes a frequency analysis unit that generates a complex spectrum of an acoustic signal for each unit period, and a declination calculation unit that calculates a declination for each frequency of the complex spectrum of each unit period. , A declination analysis means for calculating the change amount of the declination in each unit period before and after each frequency, and an acoustic evaluation means for specifying the frequency of the sine wave component included in the acoustic signal according to the time variation of the change amount The declination calculating means includes a first conversion means for converting the real part and the imaginary part of the complex spectrum into a non-negative value by sign inversion, and the sum of the real part and the imaginary part after processing by the first conversion means. And a difference calculating means for calculating the difference, a ratio calculating means for calculating the ratio of the difference to the sum, a declining provisional means for calculating a temporary declination of a complex number from the ratio calculated by the ratio calculating means, and before sign inversion Temporary deviation in the deviation corresponding to the quadrant to which the complex number belongs And a second converting means for converting the. In the above aspect, it is possible to easily specify the frequency of the sine wave component in accordance with the temporal variation of the variation amount of the declination in the unit periods that follow each other. In addition, the specific example of the above aspect is later mentioned as 3rd Embodiment, for example.

  The localization analysis apparatus according to each of the above aspects is realized by hardware (electronic circuit) such as DSP (Digital Signal Processor) dedicated to calculation of complex argument, and also by general-purpose such as CPU (Central Processing Unit). This is also realized by cooperation between the arithmetic processing unit and the program. The program according to the present invention calculates a first conversion process (SA1) for converting a real part and an imaginary part of a complex number into a non-negative value by sign inversion, and a sum and a difference between the real part and the imaginary part after the first conversion process. Sum difference calculation processing (SA1), ratio calculation processing (SA2) for calculating the ratio of the difference to the sum, declination provisional processing (SA3) for calculating the temporary declination of the complex number from the ratio calculated in the ratio calculation processing, And causing the computer to execute a second conversion process (SA4) for converting the provisional argument to the argument corresponding to the quadrant to which the complex number before sign inversion belongs. According to the above program, the same operation and effect as the declination calculation apparatus according to the present invention are exhibited. The program of the present invention is provided in a form stored in a computer-readable recording medium and installed in the computer, or is provided in a form distributed via a communication network and installed in the computer.

It is a block diagram of a declination calculating apparatus according to the first embodiment. It is a flowchart of operation | movement of a declination calculation part. It is explanatory drawing of operation | movement of a declination calculation part. It is a graph which shows the relationship between ratio and provisional deflection angle. It is a graph which shows the relationship between the actual value of a declination, and a square error. It is a block diagram of the sound processing apparatus which concerns on 2nd Embodiment. It is a block diagram of the sound processing apparatus which concerns on 3rd Embodiment.

<First Embodiment>
FIG. 1 is a block diagram of a declination calculation apparatus 100 according to the first embodiment of the present invention. The declination calculation apparatus 100 is an apparatus that calculates the declination θ of the complex number X, and is realized by a computer system that includes the arithmetic processing unit 12 and the storage device 14. The arithmetic processing unit 12 functions as the declination calculating unit 20 that calculates the declination θ of the complex number X by executing the program PGM stored in the storage device 14. The storage device 14 stores a program PGM executed by the arithmetic processing device 12 and various data (for example, complex number X) used by the arithmetic processing device 12. A known recording medium such as a semiconductor recording medium or a magnetic recording medium or a combination of a plurality of types of recording media is arbitrarily used as the storage device 14.

  FIG. 2 is a flowchart of processing in which the deflection angle calculation unit 20 calculates the deflection angle θ. For example, the process of FIG. 2 is started in response to an instruction from the user. When the processing is started, the declination calculating unit 20 converts the real part Re [X] and the imaginary part Im [X] of the complex number X to non-negative values (zero or positive number) by sign inversion, and then the real part Re [ The sum M and the difference S between X] and the imaginary part Im [X] are calculated (SA1). The process of converting the real part Re [X] and the imaginary part Im [X] into non-negative values by sign inversion is performed by converting the complex numbers X belonging to each of the second to fourth quadrants of the complex plane as shown in FIG. This corresponds to a process (mapping) for converting to a numerical value in one quadrant.

First, when both the real part Re [X] and the imaginary part Im [X] are non-negative values (that is, when the complex number X is a numerical value in the first quadrant of the complex plane), the declination calculating unit 20 As expressed by the following formulas (A1) and (B1), the sum M and the real part Re [X] and the imaginary part Im [X] without performing sign inversion (that is, movement between quadrants) The difference S is calculated.
M = Re [X] + Im [X] (A1)
S = Re [X] -Im [X] (B1)

Second, when the real part Re [X] is a negative number and the imaginary part Im [X] is a non-negative value (that is, when the complex number X is a numerical value in the second quadrant of the complex plane), the declination calculating unit 20 The sum M and the difference S are calculated after inverting the sign of the real part Re [X] as expressed by the following formulas (A2) and (B2).
M = (-Re [X]) + Im [X] (A2)
S = (-Re [X])-Im [X] (B2)

Third, when both the real part Re [X] and the imaginary part Im [X] are negative numbers (that is, when the complex number X is a numerical value in the third quadrant of the complex plane), the declination calculating unit 20 As expressed by the following equations (A3) and (B3), the sum M and the difference S are calculated after inverting the signs of both the real part Re [X] and the imaginary part Im [X].
M = (-Re [X]) + (-Im [X]) (A3)
S = (-Re [X])-(-Im [X]) (B3)

Fourth, when the real part Re [X] is a non-negative value and the imaginary part Im [X] is a negative number (that is, when the complex number X is a numerical value in the fourth quadrant of the complex plane), the declination calculating unit 20 The sum M and the difference S are calculated after inverting the sign of the imaginary part Im [X] as expressed by the following mathematical formulas (A4) and (B4).
M = Re [X] + (-Im [X]) (A4)
S = Re [X] − (− Im [X]) (B4)

  As can be understood from the above description, the arithmetic processing unit 12 converts the real part Re [X] and the imaginary part Im [X] of the complex number X into non-negative values by sign inversion (first conversion means), and , Function as an element (sum difference calculation means) for calculating the sum M and difference S of the converted real part Re [X] and imaginary part Im [X].

  When the sum M and the difference S of the real part Re [X] and the imaginary part Im [X] are calculated, the declination calculation unit 20 calculates a ratio R (R = S / M) of the difference S with respect to the sum M ( SA2). Then, the declination calculation unit 20 calculates a temporary declination (hereinafter referred to as “provisional declination”) φ according to the ratio R calculated in step SA2 (SA3). The provisional declination φ is a virtual declination with the sign inversion in step SA1 taken into account, and is different from the actual declination θ when the complex number X belongs to the second quadrant to the fourth quadrant. Therefore, the notation is distinguished from the argument θ. The relationship between the ratio R and the provisional deflection angle φ will be described in detail below.

The argument θ of the complex number X is the arctangent of the ratio of the imaginary part Im [X] to the real part Re [X] (Im [X] / Re [X]), as expressed by the following equation (1). (Arc tangent).
θ = arctan (Im [X] / Re [X]) (1)
The real part Re [X] and the imaginary part Im [X] are expressed by the following formulas (2a) and (2b). The symbol r is the absolute value of the complex number X.
Re [X] = r · cosθ (2a)
Im [X] = r · sinθ (2b)

Therefore, the sum M and the difference S of the real part Re [X] and the imaginary part Im [X] when the complex number X belongs to the first quadrant are expressed by the following formulas (3a) and (3b).
M = Re [X] + Im [X] = r (cos θ + sin θ)
= (√2) r · cos (θ−π / 4) (3a)
S = Re [X] −Im [X] = r (cos θ−sin θ)
= (√2) r · cos (θ-3π / 4)
=-(√2) r · sin (θ-π / 4) (3b)

From the equations (3a) and (3b), the following equation (4) expressing the ratio R of the difference S to the sum M is derived.
R = S / M = −tan (θ−π / 4) (4)
By transforming Equation (4), the following Equation (5) expressing the relationship between the deflection angle θ and the ratio R is derived.
θ = -arctan (R) + π / 4 (5)
In the first embodiment, considering that code conversion is performed in step SA1 (the provisional deviation angle φ may be different from the actual deviation angle θ), the deviation angle θ in the equation (5) is changed to the provisional deviation angle φ. The following mathematical formula (6) is established.
φ = −arctan (R) + π / 4 (6)

Further, when the ratio R is a numerical value within a range of, for example, −1 or more and 1 or less, the following approximation is established.
arctan (R) = (π / 4) · R
Therefore, the following formula (7) that approximates (first-order approximation) the formula (6) is derived.
φ = (π / 4) (− R + 1) (7)
In FIG. 4, the relationship between the ratio R and the provisional deflection angle φ in Expression (6) is illustrated by a broken line, and the relationship between the ratio R and the provisional deviation angle φ in Expression (7) is illustrated by a solid line. It can be understood from FIG. 4 that the approximation by the equation (7) is appropriate.

  The declination calculating unit 20 in FIG. 1 calculates the temporary declination φ by executing the calculation of Equation (7) for the ratio R calculated in step SA2 in the above-described step SA3. As understood from the above description, the arithmetic processing unit 12 calculates the ratio R of the difference S with respect to the sum M (ratio calculation means) and the element that calculates the provisional declination φ of the complex number X from the ratio R. It functions as (a declination provisional means).

When the provisional argument φ is calculated by the above procedure, the argument calculation unit 20 converts the provisional argument φ to an actual argument θ corresponding to the quadrant to which the complex number X before sign inversion in step SA1 belongs ( SA4). First, when both the real part Re [X] and the imaginary part Im [X] of the complex number X are non-negative values, the declination calculation unit 20 performs step SA3 as shown in the following equation (C1). The calculated provisional deflection angle φ is determined as the deflection angle θ.
θ = φ ...... (C1)

Second, when the real part Re [X] of the complex number X is negative and the imaginary part Im [X] is non-negative (that is, when the complex number X is converted from the second quadrant to the first quadrant in step SA1), The angle calculation unit 20 calculates the deflection angle θ corresponding to the complex number X in the second quadrant by the calculation of the following mathematical formula (C2).
θ = π−φ (C2)

Third, when both the real part Re [X] and the imaginary part Im [X] of the complex number X are negative numbers (when the complex number X is converted from the third quadrant to the first quadrant), the declination calculation unit 20 The declination angle θ corresponding to the complex number X in the third quadrant is calculated by the following equation (C3).
θ = φ + π (C3)

Fourth, when the real part Re [X] of the complex number X is a non-negative value and the imaginary part Im [X] is a negative number (when the complex number X is converted from the third quadrant to the first quadrant), the declination calculation unit 20 Calculates the declination θ corresponding to the complex number X in the third quadrant by the calculation of the following equation (C4).
θ = 2π−φ (C4)

That is, the arithmetic processing unit 12 functions as an element (second conversion unit) that converts the temporary deviation angle φ into the actual deviation angle θ. The above is a specific example of the process for calculating the deflection angle θ of the complex number X. FIG. 5 is a graph showing the relationship between the actual deviation angle θR (horizontal axis) of the complex number X and the square error (θR−θ) ² of the deviation angle θ calculated by the deviation angle calculation unit 20 of the first embodiment. It is. Although an error due to approximation or the like of Equation (6) occurs, it can be understood from FIG. 5 that according to the first embodiment, the deviation angle θ of the complex number X can be calculated with sufficiently high accuracy.

  As described above, in the first embodiment, the deflection angle θ of the complex number X is calculated according to the ratio R of the difference S to the sum M of the real part Re [X] and the imaginary part Im [X]. Since the real part Re [X] and the imaginary part Im [X] are converted to non-negative values by sign inversion, the sum M located in the denominator (divisor) of the ratio R never becomes zero. That is, no division by zero occurs in the calculation of the ratio R in step SA3. Therefore, there is an advantage that the declination angle θ of the complex number X can be easily calculated as compared with a configuration that requires special measures for dealing with division by zero.

  In the above description, when both the real part Re [X] and the imaginary part Im [X] of the complex number X are negative numbers, the signs of both the real part Re [X] and the imaginary part Im [X] are inverted. Therefore, the sum M and the difference S are calculated (formula (A3), formula (B3)), but the sign inversion is canceled out at the stage of calculating the ratio R of the difference S to the sum M. If both Re [X] and imaginary part Im [X] are negative numbers, the sign inversion can be omitted (that is, the sum M and the difference S are calculated by the formulas (A1) and (B1)). It is.

  In the above description, Equation (6) is approximated by Equation (7), but approximation is not essential for calculating the provisional deviation angle φ. For example, a table that associates each numerical value of the ratio R with each numerical value of the provisional deflection angle φ so as to satisfy the relationship of Equation (6) is stored in the storage device 14, and the deflection angle calculation unit 20 calculates in step SA2. A configuration in which the provisional deviation angle φ corresponding to the ratio R is retrieved from the table in step SA3 may be employed.

  When both the real part Re [X] and the imaginary part Im [X] are zero, the declination angle θ is indefinite (the declination angle θ cannot be defined). Therefore, when both the real part Re [X] and the imaginary part Im [X] are zero, the declination calculation unit 20 sets the temporary declination φ to a sign (NaN: Not a Number) meaning indefinite. (SA3), and the deviation angle θ corresponding to the provisional deviation angle φ is similarly set to an indefinite code (SA4). In addition, when both the real part Re [X] and the imaginary part Im [X] are zero, the declination calculation unit 20 notifies the user that the declination angle θ cannot be calculated by, for example, an image or sound (error Notification).

Second Embodiment
A second embodiment of the present invention will be described below. In addition, about the element which an effect | action and a function are equivalent to 1st Embodiment in each form illustrated below, each reference detailed in the above description is diverted and each detailed description is abbreviate | omitted suitably.

  FIG. 6 is a block diagram of the sound processing apparatus 110A of the second embodiment. The audio processing device 110A is supplied with a stereo-type audio signal xL (t) and an audio signal xR (t) from the signal supply device 200. The left-channel acoustic signal xL (t) and the right-channel acoustic signal xR (t) are time domain signals of sound (for example, voice or musical sound) to which reverberation is added. A playback device that acquires the acoustic signal xL (t) and the acoustic signal xR (t) from the portable or built-in recording medium and supplies the acoustic signal to the acoustic processing device 110A, and the acoustic signal xL (t) and the acoustic signal xR (t) Can be adopted as the signal supply device 200.

  The acoustic processing device 110A is a signal processing device that generates the acoustic signal yL (t) and the acoustic signal yR (t) in the time domain in which the reverberant sound of the acoustic signal xL (t) and the acoustic signal xR (t) is reduced. Similar to the declination calculation apparatus 100 of the first embodiment, it is realized by a computer system including the arithmetic processing unit 12 and the storage unit 14. The arithmetic processing unit 12 realizes a plurality of functions (frequency analysis unit 32, declination calculation unit 20, phase adjustment unit 34, and signal generation unit 36) by executing the program PGM stored in the storage device 14. A configuration in which each function of the arithmetic processing unit 12 is distributed over a plurality of integrated circuits, or a configuration in which a dedicated electronic circuit (DSP) realizes each function may be employed. The acoustic signal xL (t) and the acoustic signal xR (t) can also be stored in the storage device 14.

  The frequency analysis unit 32 calculates the frequency spectrum XL (k, m) of the acoustic signal xL (t) and the frequency spectrum XR (k, m) of the acoustic signal xR (t) for each unit period (frame) on the time axis. Generate sequentially. Each of the frequency spectrum XL (k, m) and the frequency spectrum XR (k, m) is a complex spectrum expressed by a complex number. Symbol k is a variable that indicates an arbitrary frequency (band) on the frequency axis, and symbol m is a variable that indicates an arbitrary unit period on the time axis. For the processing of the frequency analysis unit 32, for example, a known frequency analysis such as a short-time Fourier transform can be arbitrarily employed.

The declination calculating unit 20 includes the declination θL (k, m) of each frequency of the frequency spectrum XL (k, m) of the left channel and the declination θR () of each frequency of the frequency spectrum XR (k, m) of the right channel. k, m) are calculated sequentially for each unit period. As can be understood from the following equations (8a) and (8b), the deflection angle θL (k, m) and the deflection angle θR (k, m) mean the phase (phase spectrum) for each channel frequency. To do. The method for calculating the argument θL (k, m) and the argument θR (k, m) by the argument calculation unit 20 is the same as the method for calculating the argument θ of the complex number X in the first embodiment. Therefore, the second embodiment can achieve the same effect as the first embodiment.
XL (k, m) = | XL (k, m) | e- ^{jθL (k, m)} (8a)
XR (k, m) = | XR (k, m) | e- ^{jθR (k, m)} (8b)

The phase adjustment unit 34 adjusts the argument θL (k, m) and the argument θR (k, m) calculated by the argument calculation unit 20. Specifically, the phase adjustment unit 34 reduces the difference between the declination angle θL (k, m) and the declination angle θR (k, m) that have a common frequency and unit period (ideally in-phase). One or both of the deflection angle θL (k, m) and the deflection angle θR (k, m). For example, the phase adjustment unit 34 calculates the difference (absolute value) Δθ (Δθ = | θL (k, m) −θR (k, m) between the deviation angle θL (k, m) and the deviation angle θR (k, m). When the difference Δθ exceeds a predetermined threshold value TH (when the deviation angle θL (k, m) and the deviation angle θR (k, m) are different), the following equation (9a) and As expressed by Equation (9b), the deviation angle (phase) θR (k, m) of the frequency spectrum XR (k, m) of the right channel is replaced with the deviation angle θL (k, m) of the left channel.
XL (k, m) = | XL (k, m) | e- ^{jθL (k, m)} (9a)
XR (k, m) = | XR (k, m) | e- ^{jθL (k, m)} (9b)
The amplitude | XL (k, m) | and the angle θL (k, m) of the left channel frequency spectrum XL (k, m) and the amplitude | XR (k, m) of the right channel frequency spectrum XR (k, m) ) | Is not changed. On the other hand, when the difference Δθ between the left and right channels is less than the threshold value TH (when the deflection angle θL (k, m) and the deflection angle θR (k, m) are substantially equal), the frequency spectrum XL (k, m) And the frequency spectrum XR (k, m) is not changed.

  The signal generation unit 36 generates the time-domain acoustic signal yL (t) and the acoustic signal yR (t) from the frequency spectrum XL (k, m) and the frequency spectrum XR (k, m) processed by the phase adjustment unit 34. To do. Specifically, the signal generation unit 36 converts the frequency spectrum XL (k, m) of each unit period into a time domain signal by, for example, a short-time inverse Fourier transform, and connects the previous and subsequent unit periods to each other. A channel acoustic signal yL (t) is generated. The right channel acoustic signal yR (t) is also generated from the frequency spectrum XR (k, m) of each unit period in the same manner. The acoustic signal yL (t) and the acoustic signal yR (t) are supplied to a sound emitting device (not shown) such as a speaker and reproduced as a sound wave.

  The reverberant sound having different transfer characteristics (amplitude characteristics and phase characteristics) is added to the recorded sounds of the left and right channels in the reverberation space. Therefore, the reverberant sound tends to be perceived as dominant (high intensity) as the difference Δθ between the declination angles (θL (k, m), θR (k, m)) between the left and right channels increases. In the second embodiment, the deflection angle θL (k, m) and the deflection angle θR (k, m) are maintained while maintaining the left and right channel amplitudes (| XL (k, m) |, | XR (k, m) |). Is reduced (that is, the phase difference is reduced or eliminated), it is possible to generate the stereo acoustic signal yL (t) and the acoustic signal yR (t) with reduced reverberation.

In the above description, the right channel argument θR (k, m) is replaced with the left channel argument θL (k, m), but the left channel argument θL (k, m) is replaced with the right channel argument θL (k, m). A configuration in which the angle θR (k, m) is substituted can also be adopted. Further, as expressed by the following equations (10a) and (10b), both the right channel deviation angle θR (k, m) and the left channel deviation angle θL (k, m) are average values of both. It is also possible to substitute θave (k, m) (thus reducing the difference in declination).
XL (k, m) = | XL (k, m) | e- ^{jθave (k, m)} (10a)
XR (k, m) = | XR (k, m) | e- ^{jθave (k, m)} (10b)

<Third Embodiment>
FIG. 7 is a block diagram of the sound processing apparatus 110B of the third embodiment. A monaural sound signal x (t) is supplied from the signal supply device 200 to the sound processing device 110B. The acoustic signal x (t) is an acoustic signal of a mixed sound of a plurality of acoustic components including a sine wave component and a noise component. The acoustic processing device 110B of the third embodiment functions as an acoustic analysis device that specifies the frequency FS in which the sine wave component in the acoustic signal x (t) exists, and, similarly to the acoustic processing device 110A of the second embodiment, This is realized by a computer system including an arithmetic processing unit 12 and a storage device 14.

  The arithmetic processing unit 12 realizes a plurality of functions (frequency analysis unit 42, declination calculation unit 20, declination analysis unit 44, and acoustic evaluation unit 46) by executing the program PGM stored in the storage device 14. A configuration in which each function of the arithmetic processing unit 12 is distributed over a plurality of integrated circuits, or a configuration in which a dedicated electronic circuit (DSP) realizes each function may be employed. The acoustic signal x (t) can also be stored in the storage device 14.

  Similar to the frequency analysis unit 32 of the second embodiment, the frequency analysis unit 42 generates a frequency spectrum (complex spectrum) X (k, m) of the acoustic signal x (t) for each unit period. The declination calculator 20 calculates the declination θ (k, m) of each frequency of the frequency spectrum X (k, m) for each unit period. The method by which the declination calculating unit 20 calculates the declination θ (k, m) is the same as in the first embodiment. Therefore, the third embodiment can achieve the same effect as the first embodiment.

  The declination analysis unit 44 determines the difference between the declination θ (k, m) of each unit period calculated by the declination calculation unit 20 and the declination θ (k, m−1) of the previous unit period (hereinafter referred to as “change”). Δ (k, m) is calculated for each frequency. That is, the change amount Δ (k, m) means the degree to which the deviation angle θ (k, m) fluctuates between successive unit periods. Note that the unit period for which the change amount δ (k, m) is calculated is the entire unit period or a specific part (for example, about five) of unit periods of the acoustic signal x (t).

  The temporal variation δ (k, m) of the deviation angle θ (k, m) of the sine wave component is kept constant. On the other hand, for example, the temporal variation δ (k, m) of the deviation angle θ (k, m) of the noise component (for example, white noise) due to the circuit noise of the sound collecting device, etc. changes every moment and remains constant. Not maintained. In consideration of the difference in characteristics described above, the acoustic evaluation unit 46 of the third embodiment determines the acoustic signal x (() according to the time variation of the variation δ (k, m) calculated for each frequency by the declination analysis unit 44. The frequency FS of the sine wave component in t) is specified.

  Specifically, the acoustic evaluation unit 46 calculates the variance value V (k) and average value M (k) of the variation δ (k, m) of the deviation angle θ (k, m) in each unit period for each frequency. Calculated and the frequency at which the variance V (k) of the variation δ (k, m) is below a predetermined threshold among a plurality of frequencies on the frequency axis (that is, the frequency at which the variation δ (k, m) is small) Is selected as the frequency FS at which the sine wave component exists, and the average value M (k) of the variation δ (k, m) is determined as the instantaneous frequency of the sine wave component. The evaluation result (the frequency FS and the instantaneous frequency at which the sine wave component is detected) by the acoustic evaluation unit 46 is notified to the user by, for example, an image or sound.

DESCRIPTION OF SYMBOLS 100 ... Declination calculation apparatus, 110A, 110B ... Sound processing apparatus, 12 ... Arithmetic processing apparatus, 14 ... Memory | storage device, 20 ... Declination calculation part, 32, 42 ... Frequency analysis part, 34 ... Phase adjustment unit 36... Signal generation unit 44... Declination analysis unit 46... Acoustic evaluation unit 200.

Claims (4)

A device for calculating the argument of a complex number,
First conversion means for converting a real part and an imaginary part of the complex number to a non-negative value by sign inversion;
Sum-difference calculating means for calculating the sum and difference of the real part and the imaginary part after processing by the first conversion means;
A ratio calculating means for calculating a ratio of the difference to the sum;
Declination provisional means for calculating a provisional declination of the complex number from the ratio calculated by the ratio calculation means;
A declination calculating apparatus comprising: second conversion means for converting the temporary declination into a declination corresponding to a quadrant to which the complex number before sign inversion belongs. The temporary declination means calculates the temporary declination φ from the ratio R by the following approximate expression (A): φ = (π / 4) · (−R + 1) (A)
The deflection angle calculating apparatus according to claim 1. Frequency analysis means for generating a complex spectrum of the acoustic signal for each of the left and right channels;
Declination calculating means for calculating the declination for each frequency of the complex spectrum for each of the left and right channels;
Phase adjustment means for reducing the difference in deflection angle for each frequency of the left and right channels calculated by the deflection angle calculation means;
Signal generating means for generating left and right channel acoustic signals from the complex spectrum of each channel after processing by the phase adjusting means;
The declination calculating means includes
First conversion means for converting a real part and an imaginary part of the complex spectrum to a non-negative value by sign inversion;
Sum-difference calculating means for calculating the sum and difference of the real part and the imaginary part after processing by the first conversion means;
A ratio calculating means for calculating a ratio of the difference to the sum;
Declination provisional means for calculating a provisional declination of the complex number from the ratio calculated by the ratio calculation means;
And a second conversion means for converting the provisional deflection angle into a deflection angle corresponding to a quadrant to which the complex number before sign inversion belongs. Frequency analysis means for generating a complex spectrum of an acoustic signal for each unit period;
Declination calculating means for calculating the declination for each frequency of the complex spectrum of each unit period;
Declination analysis means for calculating, for each frequency, the amount of change of the declination in each unit period before and after,
Acoustic evaluation means for specifying the frequency of the sine wave component included in the acoustic signal according to the temporal variation of the amount of change,
The declination calculating means includes
First conversion means for converting a real part and an imaginary part of the complex spectrum to a non-negative value by sign inversion;
Sum-difference calculating means for calculating the sum and difference of the real part and the imaginary part after processing by the first conversion means;
A ratio calculating means for calculating a ratio of the difference to the sum;
Declination provisional means for calculating a provisional declination of the complex number from the ratio calculated by the ratio calculation means;
And a second conversion means for converting the provisional deflection angle into a deflection angle corresponding to a quadrant to which the complex number before sign inversion belongs.

Images