JP6566692B2 - Signal processing apparatus and method for sound signal - Google Patents

Description

  The present invention relates to a signal processing apparatus and method for sound signals.

  Conventionally, processing for obtaining a frequency spectrum from a sound signal whose waveform changes with time has been performed.

  The sound signal usually includes components of various frequencies, and the characteristics of the sound signal can be grasped by examining the intensity distribution of the frequency spectrum.

  A frequency spectrum is obtained by converting a sound signal represented in the time domain into a sound signal represented in the frequency domain. As this sound signal conversion method, for example, Fourier transform is used.

  FIG. 1 is a diagram illustrating an example of a two-dimensional graph of a frequency spectrum.

  FIG. 1 shows a two-dimensional graph of a frequency spectrum obtained by obtaining a frequency spectrum from a sound signal that changes over time in a sound waveform, using time and frequency as coordinate axes, and expressing the magnitude of the frequency spectrum intensity in shades. is there.

  A dark part indicates a region where a frequency having a high frequency spectrum intensity is distributed, and a light color part indicates a region where a frequency having a low frequency spectrum intensity is distributed. Using such a two-dimensional graph of the frequency spectrum, it is possible to analyze the time evolution of the frequency spectrum included in the sound signal.

JP 2009-210888 A JP2011-209592A JP 2014-56181 A

  However, as shown in FIG. 1, when the magnitude of the frequency spectrum intensity is expressed by shading, it is difficult to visually recognize the frequency characteristics of the sound signal because the difference in the region is not visually obvious.

  In particular, the frequency spectrum of a sound signal including a transient sound whose temporal change in the sound waveform is not in a steady state may include various frequency components, and the sound signal is expressed using a two-dimensional graph of the frequency spectrum. It was sometimes difficult to grasp the characteristics of

  Further, there is a case where it is desired to clearly grasp the relationship between two different sound signals including transient sounds.

  This specification makes it a subject to provide the signal processing apparatus and method of a sound signal which can recognize the frequency spectrum of the sound signal containing a transient sound clearly.

  It is another object of the present specification to provide a signal processing apparatus and method for sound signals that clarify the relationship between two different sound signals including transient sounds.

  According to the signal processing apparatus for a sound signal disclosed in the present specification, an input unit that inputs a sound signal including a transient sound whose temporal change in sound waveform is not in a steady state, and a plurality of times from the sound signal. A frequency spectrum for the section is obtained, and an arithmetic unit that generates a three-dimensional graph with time, frequency, and magnitude of the frequency spectrum as coordinate axes, and a display unit that displays the three-dimensional graph are provided.

  According to another sound signal signal processing device disclosed in the present specification, the time domain representation of the first sound signal and the time domain representation including a transient sound whose temporal change in the sound waveform is not in a steady state. An input unit for inputting a second sound signal; converting the first sound signal in the time domain representation into a frequency domain representation; converting the second sound signal in the time domain representation into a frequency domain representation; A calculation unit that obtains a transfer function that is a ratio between the first sound signal and the second sound signal expressed in the frequency domain.

  Furthermore, according to the signal processing method of a sound signal disclosed in the present specification, a sound signal including a transient sound whose sound amplitude change is not in a steady state with respect to time is input, and a plurality of time intervals are obtained from the sound signal. A frequency spectrum is obtained, a three-dimensional graph is generated using time, frequency, and magnitude of the frequency spectrum as coordinate axes, and the three-dimensional graph is displayed.

  Furthermore, according to another signal processing method of a sound signal disclosed in the present specification, a first sound signal and a time domain representation in a time domain representation including a transient sound in which a temporal change in the sound waveform is not in a steady state. The second sound signal is input, the first sound signal in the time domain representation is converted into the frequency domain representation, the second sound signal in the time domain representation is converted into the frequency domain representation, and the first sound signal in the frequency domain representation is converted. A transfer function which is a ratio between the sound signal and the second sound signal expressed in the frequency domain is obtained.

  According to the sound signal processing device disclosed in this specification, the frequency spectrum of a sound signal including a transient sound can be clearly recognized.

  Further, according to another sound signal processing device disclosed in the present specification, the relationship between two different sound signals including transient sound can be clarified.

  Furthermore, according to the sound signal processing method disclosed in this specification, it is possible to clearly recognize the frequency spectrum of a sound signal including a transient sound.

  Furthermore, according to another sound signal processing method disclosed in this specification, the relationship between two different sound signals including a transient sound can be clarified.

It is a figure which shows the example of the two-dimensional graph of a frequency spectrum. 1 is a diagram illustrating an embodiment of an underwater simulated sound signal processing system disclosed in this specification. FIG. It is a figure which shows one Embodiment of the signal processing apparatus of the sound signal disclosed by this specification. It is a flowchart (the 1) which shows operation | movement of the signal processing apparatus of the sound signal disclosed to this specification. It is a flowchart (the 2) which shows operation | movement of the signal processing apparatus of the sound signal disclosed to this specification. It is a flowchart (the 3) which shows operation | movement of the signal processing apparatus of the sound signal disclosed to this specification. It is FIG. (1) explaining operation | movement of the signal processing apparatus of the sound signal disclosed in this specification. It is FIG. (2) explaining operation | movement of the signal processing apparatus of the sound signal disclosed to this specification. It is a figure which shows an example of the three-dimensional graph of the frequency spectrum which the signal processing apparatus of the sound signal disclosed in this specification produced | generated. (A) is a figure which shows an example of the two-dimensional graph of the frequency spectrum which the signal processing apparatus of the sound signal disclosed in this specification produced | generated, (B) is a figure which shows the input sound signal. (A) is a figure which shows the outline of the three-dimensional graph in predetermined time, (B) is a figure which shows the outline of the three-dimensional graph in predetermined frequency, (C) is on a frequency-time plane. It is a figure which shows the outline of the three-dimensional graph in a predetermined direction. It is a figure which shows the modification of operation | movement of the signal processing apparatus of the sound signal disclosed to this specification. It is a flowchart which shows other operation | movement of the signal processing apparatus of the sound signal disclosed in this specification. It is a figure explaining other operation | movement of the signal processing apparatus of the sound signal disclosed in this specification.

  Hereinafter, a preferred embodiment of the underwater simulated sound signal processing system disclosed in this specification will be described with reference to the drawings. However, the technical scope of the present invention is not limited to these embodiments, but extends to the invention described in the claims and equivalents thereof.

  FIG. 2 is a diagram illustrating an embodiment of the underwater simulated sound signal processing system disclosed in this specification.

  The underwater simulated sound signal processing system 30 of this embodiment includes an underwater simulated sound signal generation device 20 and a signal processing device 10.

  The underwater simulated sound signal generation device 20 inputs parameters for generating a simulated sound signal, generates a simulated sound signal that propagates in water, and outputs the generated simulated sound signal. The simulated sound signal generated by the underwater simulated sound signal generation device 20 generates a steady sound in which the temporal change of the sound waveform is in a steady state. Examples of the simulated sound signal that propagates in water include the engine sound of a ship that moves at a constant speed.

  In this specification, the stationary sound means a sound whose frequency component does not change with time.

  The signal processing apparatus 10 receives a time domain representation sound signal including a transient sound whose temporal change in sound waveform is not in a steady state, converts the time domain representation sound signal into a frequency domain representation, and a frequency domain representation. The processed sound signal is converted into a time domain representation and output to the underwater simulated sound signal generation device 20. The sound signal input to the signal processing device 10 is a signal recorded in water. As a sound signal including a transient sound, for example, a sound signal generated when a marine engine is started can be cited.

  In this specification, a transient sound means a sound whose frequency component changes with time.

  The underwater simulated sound signal generation device 20 generates and outputs a simulated sound signal that includes transient sound and propagates in water based on the sound signal input from the signal processing device 10.

  Further, the signal processing apparatus 10 inputs the first sound signal and the second sound signal expressed in the time domain including the transient sound whose temporal change in the sound waveform is not in a steady state, and the first sound signal expressed in the time domain. Is converted into a frequency domain representation and a second sound signal in a time domain representation is converted into a frequency domain representation, and a transfer function that is a ratio between the first sound signal in the frequency domain representation and the second sound signal in the frequency domain representation is The obtained transfer function is output to the underwater simulated sound signal generation device 20.

  The underwater simulated sound signal generation device 20 converts the simulated sound signal generated by itself using the input transfer function, and outputs the converted simulated sound signal.

  FIG. 3 is a diagram illustrating an embodiment of a sound signal processing apparatus disclosed in the present specification.

  The signal processing device 10 includes a calculation unit 11, a storage unit 12, a display unit 13, an input unit 14, and a communication unit 15.

  The calculation unit 11 performs each process of the signal processing device 10 according to a predetermined program stored in advance in the storage unit 12, and uses the storage unit 12 to temporarily store data generated during the process. The storage unit 12 may include a primary storage device and a secondary storage device. The calculation unit 11 controls operations of the storage unit 12, the display unit 13, the input unit 14, and the communication unit 15.

  The input unit 14 is operated by a user or the like of the signal processing device 10 and inputs various types of information. As the input unit 14, for example, a touch panel, a keyboard, a mouse, or the like can be used. The calculation unit 11 performs each process using information input from the input unit 14.

  The display unit 13 is controlled by the calculation unit 11 and displays various types of information. For example, a liquid crystal panel can be used as the display unit 13.

  The communication unit 15 is controlled by the calculation unit 11 and can communicate with the underwater simulated sound signal generation device 20 or a network (not shown). The calculation unit 11 communicates with the underwater simulated sound signal generation device 20 using the communication unit 15.

  Next, an example of the operation of the signal processing apparatus 10 described above will be described below with reference to the drawings. 4 to 6 are flowcharts showing the operation of the signal processing apparatus for sound signals disclosed in this specification. 7 and 8 are diagrams for explaining the operation of the signal processing device in each step of the flowchart. The operation of the signal processing apparatus 10 described below includes a transient sound, inputs a sound signal in a time domain expression, converts the sound signal into a frequency domain expression, processes the sound signal, and converts the processed sound signal to This is converted again into the time domain representation, and the processed sound signal of the time domain representation is generated.

  First, in step S <b> 10, the signal processing apparatus 10 uses the communication unit 15 to input a time domain representation sound signal including transient sound recorded in water. FIG. 7A shows a sound signal S (t) whose waveform (amplitude) changes with time. The sound signal S (t) is recorded over a predetermined time area. The sound signal S (t) is a function of time t, and the amplitude of the sound signal at time t is a function value.

  Next, in step S12, the signal processing apparatus 10 divides the input sound signal S (t) at predetermined time intervals as shown in FIG. 7B. The sound signal S (t) is divided into a time interval between time T1 and time T2, a time interval between time T2 and time T3, and a time interval between time T3 and time T4. . In the example shown in FIG. 7B, each time interval is equal, but it may not be equal. Hereinafter, the portion of the sound signal S (t) in the time interval between the time T1 and the time T2 is referred to as a sound signal S1 (t).

  Next, in step S14, the signal processing apparatus 10 calculates a product of the divided sound signal and the window function, and generates a divided sound signal for each time interval. As shown in FIG. 7C, the product of the sound signal S1 (t) and the window function W (t) in the time interval between time T1 and time T2 is taken to generate a divided sound signal SW1 (t). Is done. The divided sound signal SW1 (t) has the same value as the sound signal S (t) in the time interval between the time T1 and the time T2, and is a function of the time t in which the value is zero in the other time intervals. It is. The divided sound signal SW1 (t) may be generated by taking the product of the sound signal S (t) and the window function W (t) shown in FIG. Further, a frequency domain representation of the sound signal may be generated using a wavelet method.

  Next, in step S <b> 16, the signal processing apparatus 10 performs a Fourier transform on each divided sound signal to generate a divided sound signal that has been Fourier-transformed for each time interval. FIG. 8D illustrates a process of performing Fourier transform on the divided sound signal SW1 (t) to generate a Fourier-converted divided sound signal FS1 (f) in the frequency domain representation. The divided sound signal FS1 (f) subjected to the Fourier transform is a function of the frequency f, and the frequency component of the frequency f is a function value.

  The value of the Fourier-transformed divided sound signal FS1 (f) is a real number. As shown in the equation (1) in FIG. 8D, the real part a (f) and the imaginary part b (f) are Have. Here, j means an imaginary unit.

  The magnitude (hereinafter also referred to as amplitude) A of the divided sound signal FS1 (f) subjected to Fourier transform is represented by Expression (2) in FIG. Here, N is the time length of the time interval between time T1 and time T2. The magnitude (amplitude) of the divided sound signal FS1 (f) subjected to Fourier transform indicates the frequency spectrum intensity in the time interval between the time T1 and the time T2.

  Further, the phase θ of the Fourier-transformed divided sound signal FS1 (f) is expressed by the equation (3) in FIG. 8D. The phase θ of the Fourier-transformed divided sound signal FS1 (f) was later processed into a time-domain representation based on a processed three-dimensional graph represented in frequency space using an inverse Fourier transform method. Used when generating sound signals.

  Next, in step S18, as shown in FIG. 8E, the signal processing apparatus 10 arranges the magnitudes (amplitudes) of the divided sound signals FS (f) subjected to the Fourier transform in order of time. Here, the time at which each frequency spectrum is arranged is defined as the start time of the corresponding time interval. That is, the magnitude of the divided sound signal FS1 (f) subjected to Fourier transform (hereinafter also referred to as the frequency spectrum magnitude or spectrum intensity) is arranged at time T1.

  Next, in step S20, the signal processing apparatus 10 generates a three-dimensional graph with time, frequency, and frequency spectrum size (spectrum intensity) as coordinate axes. The signal processing device 10 displays the generated three-dimensional graph on the display unit 13.

  FIG. 9 is a diagram illustrating an example of a three-dimensional graph of a frequency spectrum generated by the signal processing apparatus for sound signals disclosed in this specification.

  The signal processing device 10 changes the color based on the value of the frequency spectrum and generates a three-dimensional graph. Thus, the three-dimensional graph is designed so that the user can easily grasp the characteristics of the frequency spectrum visually.

  Further, the signal processing device 10 generates a two-dimensional graph of the frequency spectrum and displays it on the display unit 13.

  FIG. 10A is a diagram illustrating an example of a two-dimensional graph of a frequency spectrum generated by the sound signal signal processing device disclosed in the present specification, and FIG. 10B is a diagram illustrating an input sound signal. It is.

  The signal processing apparatus 10 also creates a two-dimensional graph of a frequency spectrum in which time and frequency are coordinate axes and the magnitude of the spectrum intensity is expressed by shading. The characteristics of the sound signal can be analyzed by comparing the three-dimensional graph and the two-dimensional graph of the frequency domain expression together with the sound signal of the time domain expression before the Fourier transform.

  Next, in step S <b> 22 (see FIG. 5), the signal processing apparatus 10 inputs information for processing the three-dimensional graph using the input unit 14. The signal processing device 10 processes the three-dimensional graph based on information input from the user. Although the content which processes a three-dimensional graph is not specifically limited, For example, the following processes are mentioned. Increase or decrease the spectral intensity in a predetermined frequency domain and a predetermined time domain. Filtering the spectral intensity in a predetermined frequency domain, filtering the spectral intensity in a predetermined time domain, advancing or delaying the phase of the spectral intensity.

  As shown in the equation (4) in the block of S22, the divided sound signal FS (f) subjected to the Fourier transform before the processing is expressed as FS ′ (f) after the processing. Further, the post-processing Fourier-transformed divided sound signal FS ′ (f) is expressed by Expression (4) using the size A ′ of the frequency component after processing and the phase θ before processing. In addition, when the process which advances or delays a phase is performed, the phase after a process is used.

  Further, the signal processing device 10 generates an outline obtained by cutting the three-dimensional graph on a plane passing through an arbitrary line on a plane having time and frequency as coordinate axes in accordance with an instruction from the user. The arbitrary line is a line extending in a direction that coincides with the direction of the time axis, a line extending in a direction that coincides with the direction of the frequency axis, or a line extending in a direction not coincident with the directions of the time axis and the frequency axis . The plane for cutting the three-dimensional graph is a plane perpendicular to a plane having time and frequency as coordinate axes.

  The user can grasp the frequency characteristics of the sound signal based on the contour cut from the three-dimensional graph. In addition, the user can determine how to process the three-dimensional graph based on the contour cut from the three-dimensional graph.

  An outline obtained by cutting the 3D graph on a plane passing through a line extending in a direction that coincides with the direction of the time axis generates an outline of the 3D graph at a predetermined time. FIG. 11A shows a diagram illustrating the outline of a three-dimensional graph at a predetermined time. Here, the line extending in the direction that coincides with the direction of the time axis coincides with the time axis or is a line parallel to the time axis.

  A contour obtained by cutting the three-dimensional graph on a plane passing through a line extending in a direction that coincides with the direction of the frequency axis generates a contour of the three-dimensional graph at a predetermined frequency. FIG. 11B shows a diagram showing the outline of a three-dimensional graph at a predetermined frequency. Here, a line extending in a direction that coincides with the direction of the frequency axis is a line that coincides with the frequency axis or is parallel to the frequency axis.

  An outline obtained by cutting the three-dimensional graph on a plane passing through a line extending in a direction not coincident with the direction of the time axis and the frequency axis generates an outline of the three-dimensional graph in a predetermined direction on the frequency-time plane. FIG. 11C is a diagram showing the outline of a three-dimensional graph in a predetermined direction on the frequency-time plane. Each spectrum intensity is discretely arranged in a lattice pattern on the time axis and the frequency axis. When the graph of FIG. 11C is generated, the spectrum intensity that is not located on the grid can be obtained as a weighted average value weighted by distance based on the spectrum intensity of neighboring grid points in the vicinity.

  The signal processing apparatus 10 has a graphical interface using an input unit 14 such as a mouse. In the above-described processing of the three-dimensional graph, the user uses the input unit 14 such as a mouse to select a frequency region or A graph can be processed by designating a time domain or the like.

  Next, in step S <b> 24, the signal processing device 10 displays the processed three-dimensional graph on the display unit 13. The user visually confirms the processed three-dimensional graph and confirms that the desired processing is performed. While confirming the three-dimensional graph or outline displayed on the display unit 13, the user inputs information to be further processed from the input unit 14 and processes the graph if necessary. Information on the processed three-dimensional graph is stored in the storage unit 12.

  Next, in step S26 (see FIG. 6), the signal processing apparatus 10 divides the processed three-dimensional graph at predetermined time intervals. The signal processing device 10 arranges the frequency spectrum in the order of time based on the divided three-dimensional graph. By this processing, information corresponding to FIG. 8E is obtained for the processed three-dimensional graph.

  Next, in step S28, the signal processing apparatus 10 performs inverse Fourier transform on the frequency spectrum for each time interval, and generates a processed divided sound signal. Referring to FIG. 8D, the divided sound signal FS′1 (f) processed in the frequency domain representation is subjected to inverse Fourier transform, and the divided sound signal SW′1 (t) processed in the time domain representation. ) Is required.

  Next, in step S30, the signal processing apparatus 10 arranges the divided sound signals in order of time, and generates a processed sound signal. By this processing, information corresponding to FIG. 7A is obtained for the processed three-dimensional graph.

  Next, in step S <b> 34, the signal processing device 10 outputs the processed sound signal of the time domain expression to the underwater simulated sound signal generation device 20 using the communication unit 15. The underwater simulated sound signal generation device 20 generates and outputs a simulated sound signal that propagates in water including transient sounds based on the sound signal input from the signal processing device 10.

  According to the above-described operation example, the signal processor 10 can be used to clearly recognize the frequency spectrum of the sound signal including the transient sound. Therefore, it is possible to easily process a sound signal including a transient sound. In addition, since it is easy to visually grasp the characteristics of the sound signal, it becomes easy for a user with little processing experience to grasp the characteristics of the sound signal.

  Next, a modified example of the operation of the signal processing apparatus 10 described above will be described below with reference to FIG.

  FIG. 12 is a diagram illustrating a modification of the operation of the signal processing apparatus for sound signals disclosed in this specification.

  This modification improves the time resolution while maintaining the frequency resolution of the frequency spectrum.

  First, as illustrated in FIG. 12A, the signal processing apparatus 10 inputs a time domain representation sound signal including transient sound recorded in water using the communication unit 15.

  Next, as shown in FIG. 12B, the signal processing apparatus 10 divides the input sound signal S (t) at predetermined time intervals so that a part of the divided time sections overlap. . The sound signal S (t) includes a time interval R1 between time T1 and time T2, a time interval R2 between time T1.5 and time T2.5, and a time between time T2 and time T3. Time interval R3, time interval R4 between time T2.5 and time T3.5, time interval R5 between time T3 and time T4, and time between time T3.5 and time T4.5 It is divided into section R6. The time intervals R1 to R6 have the same time length L.

  The time interval R1 and the time interval R2 overlap the length L / 2 of the latter half of the time interval R1. The time interval R2 and the time interval R3 overlap with the length L / 2 of the latter half of the time interval R2. The time interval R3 and the time interval R4 overlap the length L / 2 of the latter half of the time interval R3. In the time interval R4 and the time interval R5, the length L / 2 in the latter half of the time interval R4 overlaps. The time interval R5 and the time interval R6 overlap with the length L / 2 of the latter half of the time interval R5.

  In this modified example, the time resolution of the frequency spectrum is improved twice as compared with the above-described embodiment. On the other hand, the frequency resolution of the frequency spectrum of this modified example is the same as that of the above-described embodiment.

  In the embodiment described above, if the length of time for dividing the input sound signal S (t) is L / 2, the frequency resolution of the frequency spectrum is halved.

  According to the modified example of the operation described above, the time change of the frequency spectrum can be analyzed with higher accuracy. Further, in the above-described modified example of the operation, adjacent time intervals overlap with half of the interval, but the overlapping time is not particularly limited, so a part of the divided time interval overlaps. It only has to be.

  Next, an example of another operation of the signal processing apparatus 10 described above will be described below with reference to the drawings. FIG. 13 is a flowchart showing the operation of the sound signal processing apparatus disclosed in this specification. FIG. 14 is a diagram for explaining another example of the operation of the signal processing apparatus 10. The operation of the signal processing apparatus 10 to be described below is to input a first sound signal and a second sound signal in a time domain representation including a transient sound, convert the first sound signal in the time domain representation into a frequency domain representation, and The second sound signal in the domain representation is converted into the frequency domain representation, and a transfer function that is a ratio between the first sound signal in the frequency domain representation and the second sound signal in the frequency domain representation is obtained.

  First, in step S <b> 40, the signal processing apparatus 10 uses the communication unit 15 to input a first sound signal and a second sound signal expressed in time domain including transient sounds recorded in water. FIG. 13A shows the first sound signal SA (t) and the second sound signal SB (t) whose waveforms change with time. The first sound signal SA (t) and the second sound signal SB (t) are functions of time t, and the amplitude of the sound signal at time t is a function value.

  As shown in FIG. 13 (B), the first sound signal SA (t) and the second sound signal SB (t) have the same sound source but different sound propagation paths in water. The first sound signal SA (t) is a direct sound signal that is output from a sound source, propagates in water, and is recorded using a receiver. On the other hand, the second sound signal SB (t) is a reflected sound signal that is output from the sound source, propagates in water, reflects on the bottom of the water, propagates in water again, and is recorded using the receiver. Therefore, the propagation path of the second sound signal SB (t) in water is longer than that of the first sound signal SA (t).

  Next, in step S42, the signal processing device 10 detects a phase difference between the first sound signal SA (t) and the second sound signal SB (t). The phase of the second sound signal SB (t) is delayed with respect to the first sound signal SA (t).

  Next, in step S44, the signal processing device 10 corrects the phase difference so as to advance the phase of the second sound signal SB (t), and matches the phase of the first sound signal SA (t). Let

  Next, in step S46, the signal processing apparatus 10 performs a short-time Fourier transform on the first sound signal SA (t) and a short-time Fourier transform on the second sound signal SB (t), and each of them is expressed in the frequency domain representation. Convert. The process of performing a short-time Fourier transform on each sound signal is the same as the above-described operation example. As a result of performing the same processing, a divided first sound signal FSA (f) and a divided second sound signal FSB (f) subjected to Fourier transform for each time interval are generated.

  Next, in step S48, the signal processing apparatus 10, as shown in FIG. 14C, the divided first sound signal FSA (f) of the frequency domain expression and the divided second sound signal of the frequency domain expression for each time interval. A transfer function H (f) that is a ratio to FSB (f) is obtained.

  Next, in step S <b> 50, the signal processing device 10 outputs the transfer function H (f) for each time interval to the underwater simulated sound signal generation device 20. When there is no change in the propagation path, the transfer function H (f) is usually the same in each time interval. Further, the signal processing device 10 may generate a transfer function that averages the transfer function H (f) for each time interval and outputs the generated transfer function to the underwater simulated sound signal generating device 20.

  The underwater simulated sound signal generation device 20 can convert the sound signal using the input transfer function. For example, the underwater simulated sound signal generation device 20 can convert a stationary sound simulated sound signal, which is a direct sound signal, into a reflected sound signal using a transfer function. In addition, the underwater simulated sound signal generation device 20 can convert a simulated sound signal including a transient sound into a reflected sound signal using a transfer function. Further, the underwater simulated sound signal generation device 20 may convert the reflected sound signal into a direct sound signal using a transfer function.

  According to the other operation example described above, the signal processing apparatus 10 can obtain the transfer function of two sound signals including transient sounds, and thus can clarify the relationship between two different sound signals including transient sounds. The underwater simulated sound signal generation device 20 can convert a direct sound signal into a reflected sound signal or convert a reflected sound signal into a direct sound signal using this transfer function.

  As described above, the underwater simulated sound signal generation device can generate a simulated sound signal of another environment different from that input by the parameter by converting the simulated sound signal using the transfer function.

  In the other operation examples described above, the transfer function has a relationship between the direct sound signal and the reflected sound signal, but the first sound signal in the time domain expression including the transient sound and the second sound signal in the time domain expression. The environment in which is recorded is not particularly limited.

  In the present invention, the sound signal processing apparatus and method of the above-described embodiments can be modified as appropriate without departing from the spirit of the present invention.

DESCRIPTION OF SYMBOLS 10 Processing apparatus 11 Computation part 12 Storage part 13 Display part 14 Input part 15 Communication part 20 Underwater simulated sound signal generation apparatus 30 Underwater simulated sound signal processing system

Claims (9)

An input unit for inputting a sound signal including a transient sound in which a temporal change in a sound waveform is not in a steady state;
A calculation unit for obtaining a frequency spectrum for a plurality of time intervals from the sound signal, and generating a three-dimensional graph with time, frequency, and magnitude of the frequency spectrum as coordinate axes;
A display unit for displaying the three-dimensional graph;
Bei to give a,
The arithmetic unit processes the three-dimensional graph based on information input from the user by the input unit, obtains the sound signal for a plurality of time intervals based on the processed three-dimensional graph, A signal processing apparatus for sound signals, wherein the sound signals of the sections are arranged in order of time to generate the processed sound signals.   The signal processing apparatus according to claim 1, wherein the sound signal is a signal recorded in water. The arithmetic unit, based on information which the input unit is inputted from the user, a plane passing through the arbitrary line on a plane whose coordinate axes of time and frequency, to generate a contour cutting the 3-dimensional graph The signal processing apparatus according to claim 1, wherein the three-dimensional graph is processed .   The arbitrary line is a line that extends in a direction that coincides with the direction of the time axis, a line that extends in a direction that coincides with the direction of the frequency axis, or a line that extends in a direction that does not coincide with the directions of the time axis and the frequency axis. The signal processing device according to claim 3. The signal processing apparatus as described in any one of Claims 1-4 provided with the output part which outputs the processed said sound signal. The arithmetic unit, the sound signal inputted divided at predetermined time intervals, for each divided time interval, the signal processing according to any one of claim 1 to 5, determining the frequency spectrum of the sound signal apparatus. The signal processing device according to claim 6 , wherein the arithmetic unit divides the input sound signal at a predetermined time interval so that a part of the divided time interval overlaps. The arithmetic unit, by changing the color based on the value of the frequency spectrum, the signal processing apparatus according to any one of claim 1 to 7 for generating the three-dimensional graph. Input a sound signal containing a transient sound whose sound amplitude change is not in a steady state over time,
A frequency spectrum for a plurality of time intervals is obtained from the sound signal, and a three-dimensional graph is generated using time, frequency, and the magnitude of the frequency spectrum as coordinate axes,
Displaying the three-dimensional graph;
Based on information input from the user, the three-dimensional graph is processed,
A method of generating the processed sound signal by obtaining the sound signals for a plurality of time intervals based on the processed three-dimensional graph and arranging the sound signals of each time interval in order of time .