JP7552034B2 - SOUND PROCESSING DEVICE, SOUND PROCESSING PROGRAM, AND SOUND PROCESSING METHOD - Google Patents

Description

The present invention relates to an audio processing device, an audio processing program, and an audio processing method, and can be applied, for example, to sound masking processing used as a method for preventing the contents of a conversation from being leaked to third parties around a speaker.

In recent years, there has been a problem with conversations being leaked to third parties when a speaker talks with another person at a reception counter, window, meeting space, etc. in a facility where a large number of people are present (e.g., hospitals, pharmacies, banks, etc.).

Preventing conversation content from being leaked to third parties is called speech privacy, and the masking effect of sound is used to achieve speech privacy.

The sound masking effect is a phenomenon in which a certain sound (hereafter referred to as the target sound) becomes difficult to hear (is masked) when another sound with acoustic characteristics (e.g. frequency characteristics, pitch, formants, etc.) similar to the target sound is present. In general, the masking sound is called the masker, and the masked sound is called the maskee.

Patent Document 1 proposes a sound masking device that uses this sound masking effect to prevent the contents of a conversation from being leaked to a third party (protecting speech privacy).

JP 2006-243178 A

In the audio processing method described in Patent Document 1, the spectral envelope of the extracted audio signal is transformed to generate a transformed spectral envelope, and the transformed spectral envelope is then synthesized with the spectral fine structure of the extracted audio signal for use in generating a masker signal. However, because the generated masker signal is output without taking into account the frequency characteristics of the speaker, the frequency characteristics of the masker signal change when it is output from the speaker, reducing the masking effect.

Furthermore, in the audio processing method described in Patent Document 1, when masking audio output from a speaker (for example, audio guidance from a transaction device such as a ticket vending machine, the voice of the communication partner in a remote conference system, the voice of an operator in a remote location, etc.), if the frequency characteristics of the speaker that outputs the audio differ from the frequency characteristics of the speaker that outputs the masker signal, the frequency characteristics of the audio output from the speaker and the frequency characteristics of the masker signal differ, and the masking effect is reduced.

In view of the above problems, there is a demand for an audio processing device, an audio processing program, and an audio processing method that can output a masker signal that provides a higher masking effect for sounds output from a speaker.

A first sound processing device of the present invention includes: (1) frame division means for dividing an input signal supplied from a terminal for inputting a sound from a target sound source that outputs a sound to be masked into a predetermined length suitable for sound analysis ; (2) pitch estimation means for estimating the pitch of the input signal divided into frames by the frame division means; (3) input signal storage means for storing the input signal divided into frames by the frame division means and the pitch estimation result estimated by the pitch estimation means for each frame; (4) signal selection means for selecting a plurality of masker segment signals to be used for generating a masker signal from the pitch estimation result estimated by the pitch estimation means and past pitch estimation results stored in the input signal storage means; and (5) a signal selection means for selecting a plurality of masker segment signals to be used for generating a masker signal from the pitch estimation result estimated by the pitch estimation means and past pitch estimation results stored in the input signal storage means. (5) a masker signal generating means for generating and outputting the masker signal that makes at least the sound output from the target sound source difficult to hear, using a signal used in (1) above; and (6) a frequency characteristic correcting means for correcting the frequency characteristic of the masker signal generated by the masker signal generating means in accordance with the frequency characteristic of the target sound source to generate a corrected masker signal, and outputting the corrected masker signal to a speaker, (7) the input signal storing means stores the input signal that has been divided into frames by the frame dividing means without processing it, and (8) the frequency characteristic correcting means corrects the masker signal using a correction filter that corrects the difference between the frequency characteristic of the speaker and the frequency characteristic of the target sound source.

The second acoustic processing program of the present invention includes a computer including: (1) frame division means for dividing a microphone input signal supplied from a terminal for inputting a sound from a target sound source that outputs a sound to be masked into a predetermined length suitable for speech analysis ; (2) pitch estimation means for estimating the pitch of the microphone input signal divided into frames by the frame division means; (3) input signal storage means for storing the microphone input signal divided into frames by the frame division means and a pitch estimation result estimated by the pitch estimation means for each frame; (4) signal selection means for selecting a plurality of masker segment signals to be used for generating a masker signal from the pitch estimation result estimated by the pitch estimation means and past pitch estimation results stored in the input signal storage means; and (5) a signal selection means for selecting a plurality of masker segment signals to be used for generating a masker signal from the pitch estimation result estimated by the pitch estimation means and past pitch estimation results stored in the input signal storage means. a masker signal generating means for generating and outputting a masker signal that makes at least the sound output from the target sound source difficult to hear, using a signal used for generating the masker signal; (6) a frequency characteristic correcting means for correcting the frequency characteristics of the masker signal generated by the masker signal generating means in accordance with the frequency characteristics of the target sound source to generate a corrected masker signal and outputting the corrected masker signal to a speaker; (7) the input signal storing means for storing the input signal that has been divided into frames by the frame dividing means without processing it; and (8) the frequency characteristic correcting means for correcting the masker signal using a correction filter that corrects the difference between the frequency characteristics of the speaker and the frequency characteristics of the target sound source.

The third aspect of the present invention is a sound processing method performed by a sound processing device, the sound processing device comprising: (1) a frame division means, a pitch estimation means, an input signal storage means, a signal selection means, a masker signal generation means, and a frequency characteristic correction means; (2) the frame division means divides an input signal supplied from a terminal for inputting a sound from a target sound source that outputs a sound to be masked into a predetermined length suitable for sound analysis ; (3) the pitch estimation means estimates the pitch of the input signal divided into frames by the frame division means; (4) the input signal storage means stores the input signal divided into frames by the frame division means and the pitch estimation result estimated by the pitch estimation means for each frame; and (5) the signal selection means selects a masker signal from the pitch estimation result estimated by the pitch estimation means and past pitch estimation results stored in the input signal storage means. (6) the masker signal generating means uses a signal used to generate the masker signal to generate and output the masker signal that makes at least the sound output from the target sound source difficult to hear; (7) the frequency characteristic correction means corrects the frequency characteristic of the masker signal generated by the masker signal generating means in accordance with the frequency characteristic of the target sound source to generate a corrected masker signal and outputs the corrected masker signal to a speaker; (8) the input signal storage means stores the input signal that has been divided into frames by the frame division means without processing it; and (9) the frequency characteristic correction means corrects the masker signal using a correction filter that corrects the difference between the frequency characteristic of the speaker and the frequency characteristic of the target sound source.

The present invention provides an audio processing device, an audio processing program, and an audio processing method that output a masker signal that provides a higher masking effect for sounds output from a speaker.

1 is a block diagram showing a functional configuration of a sound masking device according to a first embodiment. 1 is a block diagram showing an example of a hardware configuration of a sound masking device according to a first embodiment. FIG. 2 is a conceptual diagram of a means for outputting a masker signal generated by the sound masking device according to the first embodiment. FIG. 11 is a block diagram showing a functional configuration of a sound masking device according to a second embodiment. FIG. 11 is a block diagram showing a functional configuration of a sound masking device according to a third embodiment. FIG. 13 is a block diagram showing a functional configuration of a sound masking device according to a fourth embodiment.

(A) First embodiment A first embodiment of the sound processing device, sound processing, and sound processing method of the present invention will be described in detail below with reference to the drawings. In the first embodiment, an example in which the sound processing device, sound processing program, and sound processing method of the present invention are applied to a sound masking device will be described.

(A-1) Configuration of the First Embodiment FIG. 1 is a block diagram showing the functional configuration of a sound masking device 100 according to the first embodiment.

The sound masking device 100 has a device output sound input terminal 101, an amplifier 102, an AD converter 103, a DA converter 104, a speaker amplifier 105, a speaker output terminal 106, and a sound masking processing unit 200.

The device output sound input terminal 101 is an interface (audio interface) that inputs, as an electrical signal, a sound signal such as human voice or sound (for example, the voice guidance of a transaction device such as a ticket vending machine, the voice of the communication partner of a remote conference system, the voice of an operator in a remote location, etc.) output from a speaker of a terminal or device that outputs a masking signal.

The amplifier 102 amplifies the electrical signal input to the device output sound input terminal 101.

The AD converter 103 converts the electrical signal (analog signal) amplified by the amplifier 102 into a digital signal. Hereinafter, the signal converted by the AD converter 103 will be referred to as the "input signal."

As described above, in the first embodiment, the device output sound input terminal 101, the amplifier 102, and the AD converter 103 constitute a sound signal input means for inputting a masking signal as an electrical signal. Note that the detailed configuration of the sound signal input means is not limited to the configuration in FIG. 1, and various audio device configurations can be applied. For example, a configuration without the amplifier 102 may be used so that an amplifier can be attached externally, or a configuration without the amplifier 102 and the AD converter 103 may be used so that a digital signal is output.

The sound masking processing unit 200 generates and outputs a masker signal from the input signal.

The DA converter 104 converts the digital signal output from the sound masking processing unit 200 into an electrical signal (analog signal).

The speaker amplifier 105 amplifies the electrical signal converted by the DA converter 104.

The speaker output terminal 106 is an interface (audio interface) that outputs an amplified electrical signal.

Next, we will explain the detailed configuration of the sound masking processing unit 200.

The sound masking processing unit 200 has a frame division unit 201, a DB (database) writing unit 202, an input signal DB 203, a frame signal DB 204, a frame signal selection unit 205, a masker signal generation unit 206, a masker signal frequency characteristic correction unit 207, a sound input terminal IN, and a sound output terminal OUT.

The sound input terminal IN is an interface (audio interface) that inputs an input signal to the sound masking processing unit 200.

The frame division unit 201 divides the input signal input to the sound masking processing unit 200 into frames (hereinafter referred to as "divided frames") of a predetermined length (hereinafter referred to as "frame length L") and outputs the frames. It is desirable to apply a length to the frame length L that is generally suitable for analyzing audio. For example, in the frame division unit 201, the frame length L may be 100 to 200 msec. The frame division unit 201 then outputs the divided frame signals (hereinafter referred to as "divided frame signals").

The DB writing unit 202 writes the divided frame signal to the frame signal DB 204 of the input signal DB 203.

The input signal DB203 is a storage means for storing (holding) past divided frame signals for each divided frame. The data format in the input signal DB203 is not limited, but here, the input signal DB203 is assumed to be composed of a frame signal DB203 that stores past divided frame signals.

The frame signal selection unit 205 selects multiple frames of signals (hereinafter referred to as "masker segment signals") to be used to generate a masker signal from the past divided frame signals stored in the frame signal DB 204 of the input signal DB 203, and outputs the selection result.

The masker signal generation unit 206 reads out multiple frames of the selected masker fragment signal from the frame signal DB 204 of the input signal DB 203 based on the selection result of the frame signal selection unit 205, and generates and outputs a masker signal using the read masker fragment signals of the multiple frames.

The masker signal frequency characteristic correction unit 207 performs a process to correct the frequency characteristics of the generated masker signal (hereinafter referred to as "frequency characteristic correction process") and outputs the masker signal after the frequency characteristic correction process. The frequency characteristic correction process will be described in detail later.

The sound output terminal OUT is an interface (audio interface) that outputs the masker signal that has been subjected to frequency characteristic correction processing to the DA converter 104.

The sound masking processing unit 200 may be configured entirely as hardware (for example, constructed using a dedicated board or DSP (Digital Signal Processor)), or may be configured entirely as software using a computer. The sound masking processing unit 200 may be configured, for example, by installing a program (including the acoustic processing program according to the embodiment) on a computer having a memory and a processor.

In this embodiment, the AD converter 103 and the DA converter 104 are disposed outside the sound masking processing unit 200, but the AD converter 103 and the DA converter 104 may be mounted on the sound masking processing unit 200.

Figure 2 shows the configuration when the sound masking processing unit 200 is realized in software (using a computer).

The sound masking processing unit 200 shown in FIG. 2 is configured in software using a computer 300. Programs (including the sound processing program of the embodiment) are installed in the computer 300. Note that the computer 300 may be a computer dedicated to the sound processing program, or may be configured to be shared with programs of other functions.

The computer 300 shown in FIG. 2 has a processor 301, a primary memory unit 302, a secondary memory unit 303, a sound input terminal IN, and a sound output terminal OUT. The sound input terminal IN and the sound output terminal OUT are the same as the elements shown in FIG. 1.

The primary memory unit 302 is a storage means that functions as a working memory (work memory) for the processor 301, and is, for example, a high-speed memory such as a DRAM (Dynamic Random Access Memory).

The secondary storage unit 303 is a storage means for recording various data such as the OS (Operating System) and program data (including data of the sound processing program according to the embodiment), and may be, for example, a non-volatile memory such as a flash memory, HDD (Hard Disk Drive), or SSD (Solid State Drive).

In the computer 300 of this embodiment, when the processor 301 starts up, it reads the OS and programs (including the acoustic processing program according to the embodiment) recorded in the secondary storage unit 303, and deploys them on the primary storage unit 302 for execution. Note that the specific configuration of the computer 300 is not limited to the configuration in FIG. 2, and various configurations can be applied. For example, if the primary storage unit 302 is a non-volatile memory (e.g., a flash memory, etc.), the secondary storage unit 303 may be excluded from the configuration.

Figure 3 shows an example of the layout relationship between a transaction device 400 such as a terminal or device that outputs (plays) a masker signal, a speaker 401 of the transaction device 400, a masker signal playback speaker 501 that outputs (plays) a masker signal generated by the sound masking device 100, a user U1 using the transaction device 400, and a person U2 behind or around the user U1 other than the user U1 (a person who is the target of making it difficult to hear (masking) the sound output from the speaker of the terminal or device that outputs the masker signal; hereinafter referred to as the "masking target"). In Figure 3, only the speaker 401 and the transaction device sound output terminal 402 of the transaction device 400 are illustrated as the configuration of the transaction device 400, but the configuration of the transaction device 400 is not limited, and various device configurations (for example, configurations including various computers, devices that perform transactions, DACs, amplifiers, etc.) can be applied.

Here, as shown in FIG. 3, the speaker 401 of the transaction device 400 is described as being mounted on the transaction device 400, but the device on which the maskee signal is reproduced is not limited.

The transaction device 400 is a device or terminal that can communicate with a user (in this embodiment, user U1) or a person in a remote location via a network using voice and video to conduct transactions such as ticket vending machines that sell tickets (e.g., train tickets) for transportation, vending machines that sell goods (e.g., food and drink), and ATMs (Automatic Teller Machines) of financial institutions. The transaction device 400 outputs audio (e.g., audio guidance or the voice of a person in a remote location) from the speaker 401 of the transaction device 400 as necessary during the course of the transaction.

As described above, in the first embodiment, the DA converter 104, the speaker amplifier 105, the speaker output terminal 106, and the speaker connected to the speaker output terminal 106 output the masker signal as an electrical signal, and output (play) it from the masker signal playback speaker 501 connected to the speaker output terminal 106 (hereinafter referred to as the masker signal output means). The detailed configuration of the masker signal output means is not limited to the configuration of FIG. 3, and various acoustic device configurations can be applied. For example, the sound masking device 100 may be configured without the speaker amplifier 105, and the speaker amplifier may be externally attached, or the sound masking device 100 may be configured without the DA converter 104 and the speaker amplifier 105, and output as a digital signal. In addition, the transaction device 400 outputs (plays) the masker signal to the speaker 401 of the transaction device (hereinafter referred to as the masker signal output means). The detailed configuration of the masking signal output means is not limited to the configuration shown in FIG. 3, and various audio device configurations can be applied. For example, the configuration may be such that the speaker transaction device 400401 does not have to be provided, and a speaker can be attached externally.

In this embodiment, user U1 in front of the transaction device 400 (speaker 401 of the transaction device 400) can only hear the sound (masker signal) output from the speaker 401 of the transaction device 400, and the masking target U2 can hear the sound output from the speaker 401 of the transaction device 400 and the masker signal generated by the sound masking device 100.

In FIG. 3, the directivity of the direct sound DS (Direct Sound) output from the masker signal playback speaker 501 is shown by a dotted line. Also, in FIG. 3(a), the directivity of the reflected sound RS (Reflected Sound) generated by the direct sound being reflected by the floor FR is shown by a dashed line.

In FIG. 3(a), the masker signal reproduction speaker 501 is installed in front of the user U1 at knee height with the vibration surface of the masker signal reproduction speaker 501 facing downward and at an angle to the surface of the floor FR, so that direct sound is reflected by the floor FR and the reflected sound RS is directed to be transmitted to the masking target U2 behind the user U1. The masker signal emitted from the masker signal reproduction speaker 501 is output toward the surface of the floor FR and is reflected when it reaches the floor FR. As a result, the masker signal reflected by the floor FR is transmitted to the masking target U2 behind the user U1. At this time, the direct sound of the voice uttered by the user U1 is also transmitted to the masking target U2, but is masked by the masker signal.

As described above, the masker signal reproduction speaker 501 can be installed in a wide variety of ways as long as it is installed so that the masker signal cannot be heard by the user U1 and can be heard by the masking subject U2. For example, as shown in FIG. 3B, if there is space to install the speaker behind the user U1, the vibration surface of the masker signal reproduction speaker 501 may be directly directed to the masking subject U2 to output the masker signal, as shown in FIG. 3C, the masker signal reproduction speaker 501 may be embedded in the floor FR and the vibration surface of the masker signal reproduction speaker 501 may be directly directed to the masking subject U2 to output the masker signal, or as shown in FIG. 3D, the masker signal reproduction speaker 501 may be installed on the ceiling CE and the vibration surface of the masker signal reproduction speaker 501 may be directly directed to the masking subject U2 to output the masker signal.

(A-2) Operation of the First Embodiment Next, the operation of the sound masking device 100 of the first embodiment having the above-mentioned configuration (the sound processing method of the embodiment) will be described.

The sound masking device 100 starts operating, a transaction is started between user U1 and the transaction device 400, sound (e.g., audio guidance) is output from the speaker 401 of the transaction device 400, and an audio signal is input to the device output sound input terminal 101.

The audio signal input to the device output sound input terminal 101 is converted to an electrical signal (analog signal), amplified by the amplifier 102, converted from the analog signal to a digital signal by the AD converter 103, and input to the sound input terminal IN of the sound masking processing unit 200 as the input signal x(n). In the input signal x(n), n is a parameter indicating a discrete time series of the input signal.

When the input signal x(n) begins to be input to the sound input terminal IN of the sound masking processing unit 200, it is input to the frame division unit 201.

The frame dividing unit 201 divides the input signal x(n) into divided frame signals having a frame length L. The frame dividing unit 201 may divide the input signal x(n) into divided frames according to, for example, equation (1).
x_frame(l;m)=x(l・L+m)...(1)

In equation (1), x_frame(l;m) is a divided frame signal, l is a frame number, and m is a time within the frame (m=0, 1, 2, ..., L-1). The frame dividing unit 201 outputs the divided frame signal x_frame(l;m) to the DB writing unit 202.

The DB writing unit 2024 writes the frame-divided input signal x(n) for each frame into the frame signal DB of the input signal DB 203. The DB writing unit 202 writes the divided frame signal x_frame(l;m) for each divided frame into the frame signal DB 204 of the input signal DB 203, for example, in accordance with the equations (2) and (3).

In equation (2), j is the database index number (j = 0, 1, 2, ..., DB_LEN-1; primary key; split frame identifier), and DB_LEN is the database length.

The input signal DB203 is a storage means that accumulates (holds) each past divided frame signal.

As described above, in this embodiment, the input signal DB203 includes the frame signal DB204. Here, the signal x_frame(l;m) of each divided frame is stored in the frame signal DB204.

The frame signal selection unit 205 selects multiple frames as masker segment signals from past divided frame signals stored in the frame signal DB 204 of the input signal DB 203. The method of selecting the masker segment signal is not limited, and various methods can be applied. For example, the frame signal selection unit 205 selects a frame number T(p) according to equation (4).

In equation (4), p (p = 0, 1, ..., SEL_NUM-1) is the index of the selected frame number T(p), SEL_NUM (SEL_NUM <= DB_LEN-1) is the number of selected masker segment signals, and j is the index of the database in equation (3). Equation (4) selects the divided frame signals stored in frame signal DB204 of input signal DB203 in chronological order, and assigns the index number of the database in which the selected divided frame signal is stored to T(p).

Alternatively, the frame signal selection unit 205 may randomly select a frame number T(p) from the divided frame signals stored in the frame signal DB 204 of the input signal DB 203 as shown in, for example, equation (5).
T(p)=MOD(rand(p), DE-LEN)...(5)

In formula (5), rand(p) is a function that generates a non-negative random integer number for a natural number p. Formula (5) uses the MOD function to return the remainder when the random number generated by rand(k) is divided by DB_LEN, and the selection result T(p) is assigned a random value between 0 and DB_LEN-1.

As described above, the frame signal selection unit 205 selects multiple frames from the divided frame signals stored in the frame signal DB 204, and outputs the selection result T(p) to the masker signal generation unit 206.

The masker signal generation unit 206 reads out a masker segment signal from the frame signal DB 204 of the input signal DB 203 based on the selection result T(p) of the frame signal selection unit 205, and generates and outputs a masker signal. The masker signal generation unit 206 generates a masker signal h(l;m) according to, for example, equation (6).

Equation (6) is an equation for generating a masker signal h(l;m) by reading multiple frames of multiple masker segment signals selected by the frame signal selection unit 205 from the frame signal DB204DB_singal(j;m) of the input signal DB203 and superimposing the read masker segment signals.

Then, the masker signal generation unit 206 outputs the generated masker signal h(l;m) to the masker signal frequency characteristic correction unit 207.

The masker signal frequency characteristic correction unit 207 corrects the generated masker signal h(l;m).

In this embodiment, the masker signal frequency characteristic correction unit 207 corrects the difference between the frequency characteristic (hereinafter referred to as "frequency characteristic FC1") of the signal output from the masker signal output means (in this embodiment, the DA converter 104, the speaker amplifier 105, the speaker output terminal 106, and the masker signal playback speaker 501) and the frequency characteristic (hereinafter referred to as "frequency characteristic FC2") of the signal output from the masker signal output means (in this embodiment, the transaction device 400 and the speaker 401 of the transaction device 400).

For example, the frequency characteristic FC1 of the signal output from the masker signal output means of the sound masking device 100 and the frequency characteristic FC2 of the signal output from the masker signal output means may be measured in advance, and the measurement results may be used to calculate a correction filter filter(m) that corrects the difference between the two frequency characteristics, and the correction filter filter(m) may be set in the masker signal frequency characteristic correction unit 207.

When the masker signal h(l;m) is supplied to the masker signal frequency characteristic correction unit 207, the masker signal frequency characteristic correction unit 207 calculates and outputs a corrected masker signal h'(l;m) using a preset correction filter. For example, the masker signal frequency characteristic correction unit 207 may calculate the corrected masker signal h'(l;m) according to equation (7).
h'(l;m)=filer(m)*h(l;m)...(7)

In equation (7), filter(m) is a correction filter that corrects the frequency characteristics. Also, the "*" in equation (7) represents a convolution operation.

The method for obtaining the correction filter filer(m) that corrects the difference between the frequency characteristics FC1 and FC2 can be a wide variety of methods.

For example, the correction filter filer(m) can be obtained by performing the following process in advance. First, the sound produced by outputting (playing back) the same sample sound signal from the masker signal output means and the maskee signal output means is recorded (for example, by connecting a microphone and a microphone amplifier to the sound signal input means of the sound masking device 100 and recording). Then, the recorded sound from each signal output means is transformed into the frequency domain (for example, by using a fast Fourier transform (FTT) or the like) to calculate the amplitude spectrum of the recorded sound from each speaker. Then, for each frequency, the amplitude spectrum of the sound recorded by the masker signal output means is divided from the amplitude spectrum of the sound recorded by the masker signal output means, the phase of the spectrum of the sound recorded by the masker signal output means is added, and the value converted to the time domain (for example, by inverse fast Fourier transform (IFTT) or the like) may be set as the amplitude spectrum of the correction filter filer(m) and set in the masker signal frequency characteristic correction unit 207.

In addition, various methods can be widely applied to the process of correcting the masker signal h(l;m) in the masker signal frequency characteristic correction unit 207.

For example, the masker signal frequency characteristic correction unit 207 may convert the correction filter filer(m) and the masker signal h(l;m) into the frequency domain, multiply the amplitude spectrum of the correction filter by the amplitude spectrum of the masker signal, add the phase of the spectrum of the masker signal h(l;m) to the amplitude spectrum, and perform an inverse fast Fourier transform (IFFT).

As described above, the masker signal frequency characteristic correction unit 207 corrects the frequency characteristics of the masker signal h(l;m).

Then, the masker signal frequency characteristic correction unit 207 outputs the corrected masker signal h'(l;m) according to equation (8) as an output signal y(n) to the sound output terminal OUT.
y(l・M+m)=h'(l;m)...(8)

The output signal y(n) output from the sound output terminal OUT of the sound masking processing unit 200 is converted from a digital signal to an analog signal by the DA converter 104, amplified by the speaker amplifier 105, and then output from the speaker output terminal 106.

(A-3) Advantages of the First Embodiment According to the first embodiment, the following advantages can be achieved.

In the sound masking device 100 of the first embodiment, the frequency characteristics of the generated masker signal are corrected by a correction filter that corrects the difference between the frequency characteristics of the masker signal reproducing means and the frequency characteristics of the masker signal output means. As a result, in the sound masking device 100 of the first embodiment, the frequency characteristics of the masker signal reproducing speaker 501 are corrected to the frequency characteristics of the speaker 401 of the transaction device 400, and the generated masker signal has the same frequency characteristics when output from the speaker 401 of the transaction device 400, resulting in a higher masking effect.

(B) Second embodiment A second embodiment of the sound processing device, sound processing program, and sound processing method according to the present invention will be described below in detail with reference to the drawings. In this embodiment, an example in which the sound processing device, sound processing program, and sound processing method according to the present invention are applied to a sound masking device will be described.

(B-1) Configuration of the Second Embodiment Fig. 4 is a block diagram showing the functional configuration of a sound masking device 100A according to the second embodiment. In Fig. 4, the same or corresponding parts as those in Fig. 1 are denoted by the same or corresponding reference numerals.

The following describes the second embodiment, focusing on the differences from the first embodiment, and omitting explanations of parts that overlap with the first embodiment.

The sound masking device 100A of the second embodiment differs from the first embodiment in that the sound masking processing unit 200 is replaced with a sound masking processing unit 200A.

The sound masking processing unit 200A differs from the first embodiment in that a voice section determination unit 208, a DB accumulation determination unit 209, and a masker signal generation determination unit 210 are added, and the DB writing unit 202, the frame signal selection unit 205, and the masker signal generation unit 206 are replaced with a DB writing unit 202A, a frame signal selection unit 205A, and a masker signal generation unit 206A.

The sound masking device 100A of the second embodiment differs from the sound masking device 100 of the first embodiment in that a speech segment determination unit 208 and a DB accumulation determination unit 209 are added and replaced with a DB writing unit 202A, resulting in a different accumulation method for divided frame signals accumulated in the frame signal DB 204 of the input signal DB 203; a masker signal generation determination unit 210 is added and replaced with a frame signal selection unit 205A, resulting in a different method for selecting a masker edge signal; and a masker signal generation unit 206A is replaced with a different method for generating a masker signal.

Next, we will explain the detailed configuration of the sound masking processing unit 200A.

The sound masking processing unit 200A has a frame division unit 201, a voice section determination unit 208, a DB accumulation determination unit 209, a DB writing unit 202A, an input signal DB 203, a frame signal DB 204A, a masker signal generation determination unit 210, a frame signal selection unit 205A, a masker signal generation unit 206A, a masker signal frequency characteristic correction unit 207, a sound input terminal IN, and a sound output terminal OUT.

The voice section determination unit 208 determines whether the divided frame signal is a voice section or a non-voice section (a section other than a voice section) and outputs the determination result.

The DB accumulation determination unit 209 determines whether to write and accumulate the divided frame signals in the DB based on the determination result of the voice section determination unit 208, and outputs the determination result.

The DB writing unit 202A writes the divided frame signal to the frame signal DB 204 of the input signal DB 203 based on the result of the DB accumulation determination unit 209's determination as to whether or not to write and accumulate the divided frame signal in the DB.

The masker signal generation determination unit 210 determines whether or not to generate a masker signal based on the determination result of the voice section determination unit 208, and outputs the determination result.

Based on the judgment result of the masker signal generation judgment unit 210, the frame signal selection unit 205A selects masker edge signals of multiple frames from past divided frame signals stored in the frame signal DB 204 of the input signal DB 203, and outputs the selection result.

In the second embodiment, the masker signal generation determination unit 210 may be omitted, as in the first embodiment.

(B-2) Operation of the Second Embodiment Next, the operation of the sound masking device 100A in the second embodiment having the above-mentioned configuration (the sound processing method according to the embodiment) will be described in detail.

The basic operation of the sound masking process in the sound masking device 100A according to the second embodiment is similar to the sound masking process described in the first embodiment.

The following will explain in detail the differences from the first embodiment, focusing on the operations of the speech segment determination unit 208, DB accumulation determination unit 209, DB writing unit 202A, masker signal generation determination unit 210, frame signal selection unit 205A, and masker signal generation unit 206A.

The frame division unit 201 divides the input signal x(n) into divided frame signals of frame length L1, and sequentially outputs the divided frame signals x_frame(l;m) to the speech segment determination unit 208 and the masker signal generation determination unit 210.

The voice section determination unit 208 uses the divided frame signal x_frame(l;m) to determine whether it is a voice section or a non-voice section. The voice section or non-voice section determination means performs the determination according to, for example, equations (9) and (10).

In equations (9) and (10), x_frame_amp(l) is the average amplitude value of the divided frame signal x_frame(l;m), VAD(l) is the voice section determination result, and TH is the threshold value used to determine the voice section. Equation (9) is an equation for finding the average amplitude value x_frame_amp(l) of the divided frame signal x_frame(l;m), and equation (10) is an equation for finding that if the average amplitude value x_frame_amp(l) of the divided frame signal x_frame(l;m) found by equation (9) is greater than the threshold value TH, it is determined to be a voice section and 1 is substituted for the voice section determination result VAD(l), and if it is less than the threshold value TH, it is determined to be a non-voice section and 0 is substituted for the voice section determination result VAD(l).

The threshold value TH in equation (10) only needs to be able to determine the presence or absence of speech, and various methods can be widely applied. For example, as shown in equation (11), a fixed threshold value TH may be used in which a frame of a predetermined length (hereinafter referred to as "frame length L2") after the sound masking device 100A begins operating is treated as a silent section (hereinafter referred to as the "initial frame"), and the average amplitude value of that initial frame is used as the threshold value TH; or, as shown in equation (12), a threshold value TH(l) that varies for each frame by using a time constant filter for x_frame_amp(l) may be used.

In equation (12), a is the coefficient of the time constant filter, and is a value between 0 and 1. In equation (12), if you want to slow down the updating of the threshold, a value close to 1 is desirable (for example, a = 0.9, etc.), and if you want to speed up the updating of the threshold, a value close to 0 is desirable (for example, a = 0.1, etc.).

The means for determining whether it is a voice section or a non-voice section can be a wide variety of methods, for example, the method of determining whether it is a voice section or a non-voice section by calculating the autocorrelation of the divided frame signal x_frame(l;m) may be used. The voice section determination unit 208 outputs the voice section determination result VAD(l) to the DB accumulation determination unit 209 and the masker signal generation determination unit 210.

The DB accumulation determination unit 209 determines whether or not to accumulate the divided frame signal x_frame(l;m) in the frame signal DB 204 of the input signal DB 203 based on the voice activity determination result VAD(l) of the voice activity determination unit 208. The determination means performs the determination according to, for example, equation (13).

In equation (13), DB_flag(l) is the result of the decision as to whether to store or not. In equation (13), when the speech segment decision result VAD(l) is 1, it is decided to store in the DB, and 1 is substituted for the decision result DB_flag(l). On the other hand, when the speech segment decision result VAD(l) is 0, it is decided not to store in the DB, and 0 is substituted for the decision result DB_flag(l).

The DB storage determination unit 209 outputs the determination result DB_flag(l) of whether to store in the DB to the DB writing unit 202A.

The DB writing unit 202A writes the divided frame signal x_frame(l;m) to the frame signal DB204 of the input signal DB203 according to, for example, equations (2) and (3) only when the judgment result DB_flag(l) output from the DB accumulation judgment unit 209 is 1. On the other hand, when the judgment result DB_flag(l) of the DB accumulation judgment unit 209 is 0, the DB writing unit 202A does not write the divided frame signal x_frame(l;m) to the frame signal DB204 of the input signal DB203.

The masker signal generation determination unit 210 determines whether or not to generate a masker signal based on the voice activity determination result VAD(l) from the voice activity determination unit 208. The determination means performs the determination according to, for example, equation (14).

In equation (14), mask_flag(l) is the result of the decision as to whether or not to generate a masker signal. When the speech segment decision result VAD(l) is 1, equation (14) decides that a masker signal is to be generated, and assigns 1 to the decision result mask_flag(l). On the other hand, when the speech segment decision result VAD(l) is 0, it decides that a masker signal is not to be generated, and assigns 0 to the decision result mask_flag(l).

The masker signal generation determination unit 210 outputs the determination result mask_flag(l) of whether to generate a masker signal to the frame signal selection unit 205A and the masker signal generation unit 206A.

The frame signal selection unit 205A selects frame number T(p) according to, for example, equation (4) or (5) only when the judgment result mask_flag(l) of whether to generate a masker signal output from the masker signal generation judgment unit 210 is 1. On the other hand, when the judgment result mask_flag(l) of whether to generate a masker signal is 0, multiple frames are not selected.

As described above, the frame signal selection unit 205A selects multiple frames from the divided frame signals stored in the frame signal DB 204 only when the determination result mask_flag(l) of whether or not to generate a masker signal is 1, and outputs the selection result T(p) to the masker signal generation unit 206A.

The masker signal generating unit 206A reads out a masker segment signal from the frame signal DB 204 of the input signal DB 203 based on the decision result mask_flag(l) of whether or not to generate a masker signal output from the masker signal generation deciding unit 210 and the selection result T(p) of the frame signal selecting unit 205A, and generates and outputs a masker signal. The masker signal generating unit 206A generates a masker signal ha(l;m) according to, for example, equation (15).

In equation (15), only when the judgment result mask_flag(l) of whether or not to generate a masker signal output from the masker signal generation judgment unit 210 is 1, multiple frames of masker segment signals selected by the frame signal selection unit 205A are read from the frame signal DB 204 of the input signal DB 203, and the read masker segment signals are superimposed to generate a masker signal h(l;t) and assigned to ha(l;t). On the other hand, when the judgment result mask_flag(l) of whether or not to generate a masker signal is 0, 0 is assigned to ha(l;t).

As described above, the masker signal generating unit 206A generates a masker signal only when the determination result of whether or not to generate a masker signal, mask_flag(l), is 1.

Then, the masker signal generation unit 206A outputs the generated masker signal ha(l;m) to the masker signal frequency characteristic correction unit 207.

(B-3) Advantages of the Second Embodiment According to the second embodiment, the following advantages can be obtained.

In the sound masking device 100A of the second embodiment, by storing divided frame signals in the frame signal DB 204 of the input signal DB 203 only when the input signal is determined to be a voice section (for example, when voice is output from the speaker transaction device 400401), only the voice section is stored in the DB, so that a masker signal can be generated from only voice and a high masking effect can be maintained. Furthermore, since the masker signal is generated only when it is determined to be a voice section, a masker signal can be generated and output only when voice is input, and it can be configured to output the masker signal only when voice is input.

(C) Third embodiment Hereinafter, a third embodiment of the sound processing device, sound processing program, and sound processing method of the present invention will be described in detail with reference to the drawings. In this embodiment, an example in which the sound processing device, sound processing program, and sound processing method of the present invention are applied to a sound masking device will be described.

(C-1) Configuration of the Third Embodiment Fig. 5 is a block diagram showing the functional configuration of a sound masking device 100B according to a third embodiment. In Fig. 5, the same or corresponding parts as those in Figs. 1 and 4 are denoted by the same or corresponding reference numerals.

The following describes the third embodiment, focusing on the differences from the second embodiment, and omitting a description of the parts that overlap with the second embodiment.

The sound masking device 100B of the third embodiment differs from the second embodiment in that the sound masking processing unit 200A is replaced with a sound masking processing unit 200B.

The sound masking processing unit 200B differs from the second embodiment in that a pitch estimation unit 211 is added, and the DB accumulation determination unit 209 and the masker signal generation determination unit 210 are replaced with a DB accumulation determination unit 209B and a masker signal generation determination unit 210B, respectively.

The sound masking device 100B of the third embodiment differs from the sound masking device 100A of the second embodiment in that the pitch of the divided frame signal is estimated by adding a pitch estimation unit 211, and the method of storing the divided frame signal stored in the frame signal DB204 of the input signal DB203 is different by replacing it with a DB storage determination unit 209B.

The detailed configuration of the sound masking processing unit 200B is explained below.

The sound masking processing unit 200B has a frame division unit 201, a voice section determination unit 208, a pitch estimation unit 211, a DB accumulation determination unit 209B, a DB writing unit 202A, an input signal DB 203, a frame signal DB 204A, a masker signal generation determination unit 210B, a frame signal selection unit 205A, a masker signal generation unit 206A, a masker signal frequency characteristic correction unit 207, a sound input terminal IN, and a sound output terminal OUT.

The pitch estimation unit 211 estimates the pitch (sound pitch) of the divided frame signal only when a voice section is determined based on the voice section determination result output from the voice section determination unit 208, and outputs the estimated pitch value.

The DB accumulation determination unit 209B determines whether or not to accumulate the divided frame signal in the input signal DB based on the pitch estimate value of the pitch estimation unit 211, and outputs the determination result.

The masker signal generation determination unit 210B determines whether or not to generate a masker signal based on the pitch estimate from the pitch estimation unit 211, and outputs the determination result.

In the third embodiment, the masker signal generation determination unit 210 may be omitted, as in the first embodiment.

(C-2) Operation of the Third Embodiment Next, the operation of the sound masking device 100B in the third embodiment having the above-mentioned configuration (the sound processing method according to the embodiment) will be described in detail.

The basic operation of the sound masking process in the sound masking device 100B according to the third embodiment is similar to the sound masking process described in the second embodiment.

The following provides a detailed explanation of the processing operations of the pitch estimation unit 211, DB accumulation determination unit 209B, and masker signal generation determination unit 210B, which are different from the second embodiment.

The speech section determination unit 208 determines whether the divided frame signal x_frame(l;m) is a speech section or a non-speech section, and outputs the speech section determination result VAD(l) to the pitch estimation unit 211.

The pitch estimation unit 211 estimates the pitch of the divided frame signal x_frame(l;m). The pitch estimation method is determined according to, for example, equations (16) to (18).

In equation (16), τ2 (τ2 = 0, 1, ..., L1-1) is the autocorrelation delay, in equation (17), fs is the sampling frequency, and in equation (18), pitch(l) is an estimate of the pitch. Equation (16) is an equation for finding the autocorrelation function x_frame_corr(l; τ2) of the divided frame signal x_frame(l; m). Equation (17) is an equation for finding the delay τ2 at which the autocorrelation function x_frame_corr(l; τ2) is maximized, and dividing this by the sampling frequency to calculate the pitch, which is then substituted for tmp_pitch(l). And, (18) is an equation that substitutes the calculated pitch estimate value pitch(l) when the speech section determination result VAD(l) is 1 (when there is a speech section), and substitutes 0 for the pitch estimate value pitch(l) when the speech section determination result VAD(l) is 0 (when there is a non-speech section).

The pitch can be estimated using a wide variety of methods. For example, the divided frame signal x_frame(l;m) can be subjected to a discrete Fourier transform or a fast Fourier transform, and then a cepstrum analysis can be performed to calculate the pitch.

The pitch estimation unit 211 outputs the pitch estimate pitch(l) to the DB accumulation determination unit 209B and the masker signal generation determination unit 210B.

DB accumulation determination section 209B determines whether or not to accumulate divided frame signal x_frame(l;m) in frame signal DB 204 of input signal DB 203 based on pitch estimate value pitch(l) of pitch estimation section 211. The method of judging whether or not to accumulate is, for example, according to equation (19).

In equation (19), DB_flag(l) is the result of the decision as to whether to store, and TH_PITCH is the threshold value used to decide whether to store in the DB. In equation (19), if the pitch estimate value picth(l) is greater than the threshold value TH_PITCH, it is decided to store in the DB, and 1 is substituted for the decision result DB_flag(l). On the other hand, if the value is smaller than the threshold value TH_PITCH, it is decided not to store in the DB, and 0 is substituted for the decision result DB_flag(l).

The threshold value TH_PITCH only needs to be able to determine whether to store in the DB, and various methods can be widely applied. For example, when the pitch estimation unit 211 estimates the pitch value picth(l) as a value other than 0, TH_PITCH may be set to 0 to store in the DB, or when the value is equal to or greater than the lower limit of the fundamental frequency of human speech (e.g., approximately 100 Hz), TH_PITCH may be set to 100 to store in the DB.

The DB storage determination unit 209B outputs the determination result DB_flag(l) of whether to store in the DB to the DB writing unit 202A.

The masker signal generation decision section 210B decides whether or not to generate a masker signal based on the pitch estimate pitch(l) from the pitch estimation section 211. The decision means makes the decision according to, for example, equation (20).

In equation (20), mask_flag(l) is the result of the decision as to whether or not to generate a masker signal, and TH2_PITCH is the threshold value used to decide whether or not to store. In equation (20), if the pitch estimate picth(l) is greater than the threshold value TH2_PITCH, it is decided to store in DB, and 1 is substituted for the decision result DB_flag(l). On the other hand, if the value is smaller than the threshold value TH2_PITCH, it is decided not to generate a masker signal, and 0 is substituted for the decision result mask_flag(l).

The threshold value TH2_PITCH may be set to 0 so that it is possible to determine whether or not to store in the DB, and various methods may be widely applied. For example, when the pitch estimation unit 211 determines that picth(l) is not 0, it may store in the DB, and TH2_PITCH=0, or when it is equal to or greater than the lower limit of the fundamental frequency of the voice (e.g., about 100 Hz), it may store in the DB, and TH2_PITCH=100. Also, in order to generate a masker signal when a split frame signal is written, it may be set to TH2_PITCH=TH_PITCHDB, which is the same value as the threshold value TH_PITCH in equation (19) used in the DB storage determination unit 209B.

The masker signal generation determination unit 210 outputs the determination result mask_flag(l) of whether to generate a masker signal to the frame signal selection unit 205A.

(C-3) Advantages of the Third Embodiment According to the third embodiment, the following advantages can be obtained.

In the sound masking device 100B of the third embodiment, the pitch is estimated only when the input signal is determined to be a voice section (for example, when a voice is output from the speaker 401 of the transaction device 400), and the pitch estimate is used in the DB accumulation determination unit 209B and the masker signal generation determination unit 210B. By accumulating the divided frame signal in the frame signal DB204 of the input signal DB203, only voiced sounds in the voice section are accumulated in the DB, so that the masker signal can be generated only from the voiced sounds of the voice, and a high masking effect can be maintained. Furthermore, in the sound masking device 100B of the third embodiment, the masker signal is generated only when the voice section is determined to be a voiced sound, so the masker signal is generated and output only when a voice is input. As a result, the sound masking device 100B of the third embodiment can be configured to output the masker signal only when a voice is input.

(D) Fourth embodiment A fourth embodiment of the sound processing method will be described in detail with reference to the drawings. In this embodiment, an example in which the sound processing device, the sound processing program, and the sound processing method of the present invention are applied to a sound masking device will be described.

(D-1) Configuration of the Fourth Embodiment Fig. 6 is a block diagram showing the functional configuration of a sound masking device 100C according to a fourth embodiment. In Fig. 7, the same or corresponding parts as those in Fig. 5 above are denoted by the same or corresponding reference numerals.

The following describes the fourth embodiment, focusing on the differences from the third embodiment, and omitting explanations of parts that overlap with the third embodiment.

The fourth embodiment of the sound masking device 100C differs from the third embodiment in that the sound masking processing unit 200B is replaced with a sound masking processing unit 200C.

The sound masking processing unit 200C differs from the third embodiment in that a pitch estimation information DB 212 has been added, and that the DB writing unit 202A, the input signal DB 203, and the frame signal selection unit 205A have been replaced with a DB writing unit 202C, an input signal DB 203C, and a frame signal selection unit 205C.

The fourth embodiment of the sound masking device 100C differs from the third embodiment of the sound masking device 100B in that a pitch estimation information DB 212 has been added and replaced with an input signal DB 203C, so that pitch estimates are stored in the input signal DB, the DB writing unit 202C has been replaced with a different method of storing the input signal DB 203C, and the frame signal selection unit 205C has been replaced with a different method of selecting the masker edge signal.

The detailed configuration of the sound masking processing unit 200C is explained below.

The sound masking processing unit 200C has a frame division unit 201, a voice section determination unit 208, a pitch estimation unit 211, a DB accumulation determination unit 209B, a DB writing unit 202C, an input signal DB 203C, a frame signal DB 204, a pitch estimation information DB 212, a masker signal generation determination unit 210B, a frame signal selection unit 205C, a masker signal generation unit 206A, a masker signal frequency characteristic correction unit 207, a sound input terminal IN, and a sound output terminal OUT.

Based on the result of the determination by DB accumulation determination unit 209B as to whether or not to write and accumulate the divided frame signal in the DB, DB writing unit 202C writes the divided frame signal to frame signal DB204 of input signal DB203C and writes the pitch estimate to pitch estimation information DB212 of input signal DB203C.

The input signal DB203C is a storage means that stores (holds) past divided frame signals in association with past pitch estimates. There are no limitations on the data format in the input signal DB203C, but here, the input signal DB203C is assumed to be composed of data including a frame signal DB204 that stores past divided frame signals, and a pitch estimation information DB212 that stores past pitch estimates.

Based on the judgment result of the masker signal generation judgment unit, the frame signal selection unit 205C compares the pitch estimate value estimated by the pitch estimation unit 211 with the past pitch estimate values stored in the pitch estimation information DB212 of the input signal DB203C, and based on the comparison result, selects a masker segment signal from the frame signal DB204 of the input signal DB203C, and outputs the selection result.

In the fourth embodiment, the masker signal generation determination unit 210B may be omitted, as in the first embodiment.

(D-2) Operation of the Fourth Embodiment Next, the operation of the sound masking device 100C in the fourth embodiment having the above-mentioned configuration (the sound processing method according to the embodiment) will be described in detail.

The basic operation of the sound masking process in the sound masking device 100C according to the fourth embodiment is similar to the sound masking process described in the third embodiment.

The following will explain in detail the processing operations of the DB writing unit 202C, the input signal DB 203C, and the frame signal selection unit 205C, which are differences from the third embodiment.

The pitch estimation unit 211 estimates the pitch of the divided frame signal x_frame(l;m) and outputs the pitch estimate to the DB accumulation determination unit 209B, the DB writing unit 202C, the masker signal generation determination unit 210B, and the frame signal selection unit 205C.

Based on the judgment result DB_flag(l) output from the DB accumulation judgment unit 209B, the DB writing unit 202C writes the divided frame x_frame(l;m) and the pitch estimate value pitch(l) in association with the frame signal DB204 of the input signal DB203 and the pitch estimation information DB212. Only when the judgment result DB_flag(l) of the DB accumulation judgment unit 209 is 1, the DB writing unit 202C writes the divided frame signal x_frame(l;m) to DB_singal(j;m) of the frame signal DB204 and simultaneously writes the pitch estimate value pitch(l) to DB_pitch(j) of the pitch estimation information DB212 according to, for example, equations (2), (21), and (3). On the other hand, when the determination result DB_flag(l) of DB accumulation determination section 209 is 0, divided frame signal x_frame(l;m) and pitch estimate value pitch(l) are not written to frame signal DB 204 and pitch estimation information DB 212 of input signal DB 203C.
DB_pitch(j)=x_frame_long(l)...(21)

The input signal DB203C stores (holds) the divided frame x_frame(l;m) and the pitch estimate pitch(l) in association with the frame signal DB204 of the input signal DB203 and the pitch estimation information DB212.

As described above, in this embodiment, the input signal DB203C includes a frame signal DB204 and a pitch estimation information DB212. Here, the signal x_frame(l;m) of each divided frame is recorded in the frame signal DB204, and the pitch estimation information DB212 records the pitch estimate value pitch(l).

The frame signal selection unit 205C compares the pitch estimate value (the pitch estimate value based on the current divided frame) of the pitch estimation unit 211 with the past pitch estimate value stored in the pitch estimation information DB 212 based on the decision result mask_flag(l) of whether to generate a masker signal output from the masker signal generation decision unit 210B, and selects a divided frame signal having a pitch estimate value (close pitch value) close to the pitch estimate value of the pitch estimation unit 211 as a masker segment signal. The frame signal selection unit 205C selects a frame number T(p) according to, for example, equation (22) or (23) only when the decision result mask_flag(l) of whether to generate a masker signal output from the masker signal generation decision unit 210B is 1. On the other hand, when the decision result mask_flag(l) of whether to generate a masker signal is 0, multiple frames are not selected.
Sub_pitch(j)
=｜picth(l)−“DB_pitch”(j)｜…(22)
Tc(p)
= small (Sub_Pitch(j), p+1))...(23)

In equation (22), Sub_pitch(j) is the absolute value of the difference between the pitch estimate pitch(l) of the pitch estimation unit 211 and the past pitch estimate DB_pitch(i), and in equation (23), Tc(p) is the selected frame number, and p (p=0, 1, ..., SEL_NUM-1) is the number of selected frames. Furthermore, small(x(k), p) in equation (23) is a function that outputs the index k_p of the p-th smallest x(k_p) in array x(k). Equation (22) is an equation for selecting, as a masker segment signal, multiple frames of index j (divided frame) having the smallest absolute value of the difference between the pitch estimation result pitch(l) of the pitch estimation unit 211 and the past pitch estimate stored in the pitch estimation information DB 212 for each index j.

As described above, the frame signal selection unit 205C selects the signal (index) of the divided frame having pitch information (close pitch value) close to the pitch estimate value of the pitch estimation unit 211 as the masker segment signal, and outputs the selected frame number Tc(p) to the masker signal generation unit 206A.

(D-3) Advantages of the Fourth Embodiment According to the fourth embodiment, the following advantages can be obtained.

In the sound masking device 100C of the fourth embodiment, pitch is estimated only when the input signal is determined to be a voice section (for example, when sound is output from the speaker 401 of the transaction device 400), and the pitch estimate is used in the DB accumulation determination unit 209B and the masker signal generation determination unit 210B, and at the same time, the divided frame signal and the pitch estimate are accumulated in the input signal DB203C and used in the frame signal selection unit 205C, so that a divided frame signal with a similar pitch can be selected from the frame signal DB204 of the input signal DB203C, and the frequency characteristics of the masker signal become close to the pitch of the input voice, making it possible to generate a masker signal with a high masking effect.

(E) Other Embodiments The present invention is not limited to the above-described embodiments, and may include modified embodiments such as those exemplified below.

(E-1) For example, the sound masking device of the present invention may be installed in a device that prevents the contents of a conversation between user U1 and other people around him or her during a conference call.

(E-2) In each of the above embodiments, the sound masking section of the sound masking device may be configured to be processed by a processing device (e.g., a server, etc.) on the network.

(E-3) In each of the above embodiments, the sound masking device has been described as including an audio device (device output sound input terminal, amplifier, AD converter, speaker output terminal, speaker amplifier, and DA converter), but the sound masking device may be manufactured without the audio device, and the audio device may be connected separately at the site where it is actually used. In other words, the sound masking device may be configured to include at least a sound masking processing unit.

100...sound masking device, 101...device output sound input terminal, 102...amplifier, 103...AD converter, 104...DA converter, 105...speaker amplifier, 106...speaker output terminal, 113...target speaker, 114...person other than the corresponding speaker, 115...floor, 116...masker signal radiation area, 117...ceiling, 200...sound masking device, 201...frame division unit, 202...DB writing unit, 203...input signal DB, 204...frame signal DB, 205...frame signal Selection unit, 206... delay time calculation unit, 207... masker signal generation unit, 208... voice section determination unit, 209... DB accumulation determination unit, 210... masker signal generation unit, 211... pitch estimation unit, 212... pitch estimation information DB, 300... computer, 301... processor, 302... primary memory unit, 303... secondary memory unit, 400... trading device, 401... trading device speaker, 402... trading device sound output terminal, 501... masker signal playback speaker, IN... sound input terminal, OUT... sound output terminal.

Claims (3)

a frame division means for dividing an input signal supplied from a terminal for inputting a sound from a target sound source that outputs a sound to be masked into a predetermined length suitable for speech analysis ;
a pitch estimation means for estimating the pitch of the input signal divided into frames by the frame division means;
an input signal storage means for storing, for each frame, the input signal divided into frames by said frame division means and the pitch estimation result estimated by said pitch estimation means;
a signal selection means for selecting a plurality of masker segment signals to be used for generating a masker signal from the pitch estimation result estimated by said pitch estimation means and the past pitch estimation results stored in said input signal storage means;
a masker signal generating means for generating and outputting a masker signal that makes it difficult to hear at least the sound output from the target sound source, using a signal used for generating the masker signal;
a frequency characteristic correction means for correcting a frequency characteristic of the masker signal generated by the masker signal generation means in accordance with a frequency characteristic of the target sound source to generate a corrected masker signal, and outputting the corrected masker signal to a speaker,
the input signal storage means stores the input signal divided into frames by the frame division means without processing it,
The sound processing device, wherein the frequency characteristic correction means corrects the masker signal using a correction filter that corrects a difference between the frequency characteristic of the speaker and the frequency characteristic of the target sound source. Computer,
a frame division means for dividing an input signal supplied from a terminal for inputting a sound from a target sound source that outputs a sound to be masked into a predetermined length suitable for speech analysis ;
a pitch estimation means for estimating the pitch of the input signal divided into frames by the frame division means;
an input signal storage means for storing, for each frame, the input signal divided into frames by said frame division means and the pitch estimation result estimated by said pitch estimation means;
a signal selection means for selecting a plurality of masker segment signals to be used for generating a masker signal from the pitch estimation result estimated by said pitch estimation means and the past pitch estimation results stored in said input signal storage means;
a masker signal generating means for generating and outputting a masker signal that makes it difficult to hear at least the sound output from the target sound source, using a signal used for generating the masker signal;
a frequency characteristic correction means for correcting a frequency characteristic of the masker signal generated by the masker signal generation means in accordance with a frequency characteristic of the target sound source to generate a corrected masker signal, and outputting the corrected masker signal to a speaker;
the input signal storage means stores the input signal divided into frames by the frame division means without processing it,
the frequency characteristic correction means corrects the masker signal using a correction filter that corrects a difference between the frequency characteristic of the speaker and the frequency characteristic of the target sound source. In a sound processing method performed by a sound processing device,
the sound processing device includes a frame division means, a pitch estimation means, an input signal storage means, a signal selection means, a masker signal generation means, and a frequency characteristic correction means;
The frame division means divides an input signal supplied from a terminal for inputting a sound from a target sound source that outputs a sound to be masked into a predetermined length suitable for speech analysis ;
The pitch estimation means estimates the pitch of the input signal divided into frames by the frame division means,
the input signal storage means stores the input signal divided into frames by the frame division means and the pitch estimation result estimated by the pitch estimation means for each frame;
the signal selection means selects a plurality of masker segment signals to be used for generating a masker signal from the pitch estimation result estimated by the pitch estimation means and the past pitch estimation results stored in the input signal storage means;
the masker signal generating means generates and outputs the masker signal, using a signal used for generating the masker signal, the masker signal making at least the sound output from the target sound source difficult to hear;
the frequency characteristic correction means corrects the frequency characteristic of the masker signal generated by the masker signal generation means in accordance with the frequency characteristic of the target sound source to generate a corrected masker signal, and outputs the corrected masker signal to a speaker;
the input signal storage means stores the input signal divided into frames by the frame division means without processing it,
The acoustic processing method, wherein the frequency characteristic correction means corrects the masker signal using a correction filter that corrects a difference between the frequency characteristic of the speaker and the frequency characteristic of the target sound source.

Images