JPH1168573A - Sound data processing method, sound data processor and record medium, recorded with sound data processing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To easily confirm and grasp the existence of sound data which has been copied illegally by embedding one of bit data of security data into plural Fourier transform coefficient units respectively. SOLUTION: A blocked Fourier transform coefficient generating means 5 executes Fourier transform of blocked sound data in each block and calculates a Fourier transform coefficient. A unit-forming means 6 configures plural units, which are constituted of two Fourier transform coefficients that become a pair from plural Fourier transform coefficients that constitute a block. A security data embedding means 7 allocates one of bit data of security data to plural units respectively that have been configured by the unit-forming means. An inverse Fourier transform processing means 8 performs inverse Fourier transform processing of a blocked Fourier transform coefficient group, in which the security data has been embedded in each block and restores them to sound data based on the result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio data processing method and an audio data processing apparatus, and more particularly, to an audio data processing method and an audio data processing method for effectively preventing unauthorized copying of audio data. The present invention relates to an audio data processing device.

[0002]

2. Description of the Related Art Conventionally, audio data such as digital audio data and musical sound data recorded on an arbitrary recording medium such as a tape or a disk is transmitted once the recording medium is sold in the market. Without the consent of the copyright owner, the unauthorized copy is sold in large quantities at a low price, so the rights of the copyright owner are unjustly infringed and the copyright owner It is currently suffering damage.

[0003] At present, there is practically no means for effectively preventing such illegal copying, and it is difficult to confirm the evidence from the law, so it is difficult to prosecute such illegal copying or pirated copies. In state. Many methods have been proposed for preventing such unauthorized copying of audio data. However, there are problems such as deterioration in sound quality of reproduced sound, and at present it is not practically used.

Therefore, in order to improve this problem, the copyright holder of the digital audio data, or
The licensee audibly adds security data, such as encryption and signature data composed of digital bit data, indicating that the audio data is genuine audio data that has been authorized for sale, to the audio data itself. Research on "data hiding", which embeds data so that it has no influence and prevents infringement of copyright and ownership of the audio data, has been actively conducted recently. Data hiding technology has not been developed.

[0005]

SUMMARY OF THE INVENTION It is an object of the present invention to improve the above-mentioned drawbacks of the prior art, and to provide a person who has a copyright on predetermined audio data in advance based on a simple technical configuration or a license thereof. Embeds security data that can be checked later to confirm that the product is its own, so that it is possible to determine whether or not it is its own product, whether or not it has been copied illegally, etc. An object of the present invention is to provide an audio data processing method and an audio data processing device that can be easily performed.

[0006]

The present invention basically employs the following technical configuration in order to achieve the above object. That is, the first of the present invention
As a mode of embedding, when embedding the security data in the specific audio data, the specific audio data is divided into a plurality of continuous blocks that are at least twice the number of digital bits of the security data, and Fourier transform is performed on the audio data of the block for each block to obtain a Fourier transform coefficient for each block, and the furthest away from the plurality of Fourier transform coefficients in the block. A first unit is formed with two Fourier transform coefficients located at the same distance as a first set, and then located farthest apart from each other except for the two Fourier transform coefficients that constituted the first unit. A second unit is formed in which two Fourier transform coefficients are used as a second set, and a unit is sequentially formed in the same manner. Audio data processing method configured to form a number of Fourier transform coefficient units corresponding to the number of digital bits of security data, and to embed one of the bit data of the security data in each of the plurality of Fourier transform coefficient units. Also, as a second aspect of the present invention, a voice data storage means, a security data storage means, a voice data blocking means, and a Fourier transform is executed for each block of the blocked voice data. Block Fourier transform coefficient generating means for obtaining a Fourier transform coefficient by means of: a unit forming means for forming a plurality of units composed of a pair of two Fourier transform coefficients from the plurality of Fourier transform coefficients in the block; For each of the plurality of units formed by the unit forming means, Security data embedding means for allocating one of bit data of security data, inverse Fourier transform for performing inverse Fourier transform processing for each block on a block of Fourier transform coefficients in which the security data is embedded, and returning the result to original voice data The audio data processing device includes a processing unit and a control unit that comprehensively controls the above units.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The audio data processing method and audio data processing apparatus according to the present invention employ the above-described technical configuration, so that predetermined audio data does not affect human hearing. It is possible to embed specific security data, for example, encryption data, signature data, and the like in a format, and it is possible to easily reproduce it.

That is, in the present invention, one of the human auditory characteristics is cited as follows: "It is insensitive to the absolute phase of the audio signal, but sensitive to the relative phase." This is a good use of the characteristic. That is, in general, one-dimensional audio data is expressed in the form of a linear combination of trigonometric functions having different periods as a basis function group. After taking out some basis functions from the basis functions and shifting their phases by some degree, the voice data expressed in the form of their linear combination again is completely comparable to the original voice data and the human ear. There is a relationship that can not do.

Therefore, in the present invention,
For audio data using human auditory characteristics, security data such as appropriate signature data and encryption data are partially changed in absolute phase without changing the relative phase of each audio signal. This makes it possible to realize the effect of inserting security data and preventing the deterioration of sound quality.

[0010]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of a voice data processing method and a voice data processing apparatus according to the present invention. That is, FIG. 1 is a block diagram showing the configuration of one specific example of the audio data processing apparatus according to the present invention. In FIG. 1, audio data storage means 2, security data storage means 3, audio data blocking means 4,
Blocked Fourier transform coefficient generation means 5 for performing a Fourier transform on each block of the speech data for each block to obtain a Fourier transform coefficient, a pair of two Fourier transform coefficients constituting the block, Unit forming means 6 forming a plurality of units formed by Fourier transform coefficients; security data for allocating one of the bit data of the security data to each of the plurality of units formed by the unit forming means; Embedding means 7, inverse Fourier transform processing means 8 for performing an inverse Fourier transform process for each block on the block of Fourier transform coefficients in which the security data is embedded, and returning the result to the original audio data, and the above means. Processing device 1 for audio data, comprising It is shown.

Further, in the audio data processing apparatus 1 according to the present invention, as is apparent from the above description, the original audio data in which the security data is embedded is Fourier-transformed. Means 11; voice data storage means 12 for temporarily storing voice data to be tested for which it is unknown whether or not security data is embedded; means 13 for applying the same Fourier transform to the voice data to be tested 13 And comparing means 14 for comparing the data after the Fourier transform processing with respect to both voice data, and determining means 15 for determining whether or not the voice data to be inspected includes predetermined security data. is there.

If the audio data processing method according to the present invention is described using the audio data processing apparatus shown in FIG. 1, in the present invention, first, it is necessary to embed predetermined security data. Select some audio data,
It is stored in a predetermined voice data recording means 2. On the other hand, security data to be embedded in the audio data is stored in the security data storage unit 3.

As the security data used in the present invention, there can be used security data composed of a data string in which a plurality of binary bit data taking either 1 or 0 are serially arranged. That is, when the security data is set to ABC, each code is set to “11101”, for example.
It is decided to say “10011” or “00110”.

Of course, the security data used in the present invention can be used not only for the above-described character strings but also for image data. Next, the audio data in which the security data selected as described above is to be embedded is read from the audio data recording means 2, and the audio data blocking means 4 performs block processing for dividing the data into a plurality of blocks.

The number of blocks in such blocking is as follows:
As will be described later, the number of bits of the bit data constituting the security data must be at least twice, preferably twice or more, and more preferably 2n + 2. In the present invention, the block is, as shown in FIG.
It is desirable that the above-mentioned number of blocks be configured so as to be continuously arranged.

Further, in the present invention, the position where the plurality of continuous blocks are formed in the audio data is
It is not particularly limited, and can be arranged and formed at an arbitrary position. The plurality of block groups are not limited to one, and can be repeatedly arranged in the audio data. It is possible. Thereby, the third party
Even if only a part of the audio data is cut out and used, it is possible to confirm the illegal copy.

In the present invention, the size of each block is not particularly limited, but is preferably a power of 2 (2 ⁿ ). For example, each block has one block of data. It is desirable to block the number so that it becomes 256. FIG. 2 is a diagram for explaining the principle of the above-described audio data blocking. More specifically, audio data to be embedded has two samples per block.
It is shown as a set of blocks that are powers of.

That is, now, a one-dimensional real-valued function Sound (i
) (i is an integer greater than or equal to 0) is the audio data to be embedded. Also, the number of samples in the block is 2, k, j ∈
When Z = i = 2k + j (0 ≦ j < ²ⁿ ), the one-dimensional real-valued function B _k (j) is expressed by B _k (j) = Sound (i) (Equation 1). However, Z represents the whole integer.

Next, in the present invention, the Fourier transform of the audio data of the block is performed for each of the blocks of the audio data in the block by the Fourier transform coefficient generation means 5 for blocking. Then, a plurality of Fourier transform coefficients F (k) are obtained for each block. In the present invention, the purpose of blocking audio data is to perform a Fourier transform on the audio data as described above, and to reduce the time required for the Fourier transform processing operation. .

Since it is common to perform Fourier transform on such audio data to obtain Fourier transform coefficients, detailed description thereof will be omitted here, but only the basic concept will be described. I will do it. That is, since the audio data is digital data, the discrete Fourier transform will be described here. However, in the present invention, the fast Fourier transform is adopted, and this fast Fourier transform is a speedup of the discrete Fourier transform.
The calculation result is exactly the same as that obtained by performing the discrete Fourier transform.

Definition 1 (Discrete Fourier Transform) Discrete one-dimensional real-valued function f (n) (n∈Z, 0 ≦ n <N)
Is given, a discrete one-dimensional complex-valued function F (k) (k∈Z, 0 ≦ k <N) represented by (Equation 2) is called a discrete Fourier transform of f (n).

[0022]

(Equation 1)

The discrete Fourier transform defined in Definition 1 is applied to all blocks, and a security data bit string is embedded in the data of the Fourier transform coefficients. Next, in the present invention, after a plurality of Fourier transform coefficients F (k) are obtained for each block by the above operation, the unit forming means 6 sets the plurality of Fourier transform coefficients F (k) of the block. From the group of transform coefficients, as shown in FIG. 2, first, two Fourier transform coefficients F (0) located farthest apart from each other
And a Fourier transform coefficient F (n) as a first unit to form a first unit U1.

That is, in the block B0 shown in FIG. 2, since the Fourier transform coefficient F (0) and the Fourier transform coefficient F (n) satisfy the above condition, the Fourier transform coefficient F (0) and the Fourier transform The first unit U1 is formed with the coefficient F (n).
Next, in the two blocks constituting the first unit, that is, in the remaining Fourier transform coefficient group except for the Fourier transform coefficient F (0) and the Fourier transform coefficient F (n), the positions farthest apart from each other are set. A second unit is formed in which the two Fourier transform coefficients are used as a second set.

That is, in the block B0 shown in FIG. 2, since the Fourier transform coefficient F (1) and the Fourier transform coefficient F (n-1) satisfy the above condition, the Fourier transform coefficient F (1) A second unit U2 is formed with the Fourier transform coefficient F (n-1). Next, four Fourier transform coefficients constituting the first unit and the second unit, that is, Fourier transform coefficients F
(0), F (1), F (n-1) and F (n), except for the remaining Fourier transform coefficient groups apart from each other, the two Fourier transform coefficients located farthest apart from each other are set to the third To form a third unit U3.

That is, in the Fourier transform coefficient group shown in FIG. 2, the Fourier transform coefficient F (2) and the Fourier transform coefficient F
Since (n-2) satisfies the above condition, since the block B2 and the block B _N-2 satisfy the above condition, the Fourier transform coefficient F (2) and the Fourier transform coefficient F (n-2) A third unit U3 is formed. Similarly, units are sequentially formed from 2n blocks to form n Fourier transform coefficient units corresponding to the digital bit number n of the security data.

In the present invention, when performing the above-mentioned block operation, it is sometimes desirable from the Fourier transformation operation not to use the data of the first and last blocks of the block group. In forming the plurality of block groups from the audio data, 2n + 2 blocks are formed with respect to the number n of constituent bits of the security data, and each block at both ends of the block group is used for Fourier transform. It is desirable not to do it.

After performing such an operation, in the present invention, in the security data embedding means 7, a plurality of units U1 to U1 formed by the unit forming means 6 are formed.
An operation of assigning one of the bit data of the security data to each of Un is performed. As an operation of assigning one of the bit data of the security data to each unit in the present invention, for example, as described above, the security data to be embedded is A
And the digital bit data is “1110”
1 ", the first unit U1 has
The data of “1” is embedded, the data of “1” is embedded in the second unit U2, and the third unit U3 is embedded in the second unit U2.
Is embedded with data of “1”, and the fourth unit U
The operation is performed such that the data of “0” is embedded in 4 and the data of “1” is embedded in the fifth unit U5.

As an example of the security data embedding method in the present invention, for example, when the bit data of the security data is “1”, the phase of the Fourier transform coefficient of the block is shifted by π. When the bit data of the security data is “0”, the phase of the Fourier transform coefficient of the block is not shifted by π.

More specifically, when the bit data of the security data is “1”, processing to shift the phase of the Fourier transform coefficient of the block by π is performed by changing the Fourier transform coefficient of the block. Is converted to an imaginary part and its sign is reversed, while the imaginary part of the Fourier transform coefficients of the block is
It means to convert the sign to the real part without changing the sign.
If the bit data of the security data is “0”, it means that neither the real part nor the imaginary part of the Fourier transform coefficient of the block is changed.

The principle of the security data embedding processing operation will be briefly described below. Any k (k ∈
Z, k ∈ [0, the total number of blocks]), and F _k is the result of discrete Fourier transform of B _k . Also, the embedded data bits are converted to a one-dimensional discrete integer value function U (d) (where d, d
_n Ｚ Z, d _n <2 ^n-1 , U (d) = 1 or 0, d ∈ [1, d
_n ], U (d) = 0), and F _k (m) (m∈Z, m∈ [1, 2
ⁿ ]), ^assuming that a function obtained as a result of embedding the bit data U (d) is F ′ _k (m), F ′ _k (m) becomes

[0032]

(Equation 2)

[0033] In the present invention, after the predetermined security data is embedded in the desired audio data by the above-described method, the inverse Fourier transform processing means 8 sets the blocked Fourier transform coefficient group in which the security data is embedded. Then, an operation of performing an inverse Fourier transform process for each block and incorporating the embedded conversion result into the original audio data is performed.

Here, the principle of the inverse Fourier transform processing according to the present invention will be briefly described. Definition 1 (Discrete Fourier inverse transform) Discrete one-dimensional real-valued function f (n) (n∈Z, 0 ≦ n <N)
And a discrete one-dimensional complex-valued function F (k) (k ∈
If Z, 0 ≦ k <N) is a discrete Fourier transform of f (n), Equation 7 holds.

[0035]

(Equation 3)

[0036] When 'discrete data values obtained by performing inverse Fourier transform on the _k f' above F and _k, corresponding to the f _'k is by embedding the security data bit sequence U to the digital audio data B _k . Hereinafter, a specific example of the security data embedding processing operation according to the present invention will be described with reference to a simple model example.

That is, the sound data to be embedded with security data is Sound (k), and the embedded security bit string is s.
yomei [j], the result of embedding the security data bit string in F [Sound] (p), F [Sound] (p) of the data obtained by performing a discrete Fourier transform of the embedded sound data Sound, and F '[Sound] (p) I do.

Here, Sound (k) is a function that returns an integer value defined on the integer space, and k = 0, 1,..., N. Syomei [j] is a function defined on the integer space, and 0,
A function that returns only 1 and j = 0,1. Here, syomei [j] is defined by the following equation.

[0039]

(Equation 4)

At this time, F [Sound] (p) is a function defined on an integer space, and is a function that returns an imaginary number, and is expressed by the following equation 8 using equation 2. Here, p = 0, 1,..., N. F is an imaginary number, and its real part is represented by Re [F] (p) and its imaginary part is represented by Im [F] (p). Equation 8 If F '[Sound] (p) is represented by Equations 3, 4, 5, 6 , And is represented by the following equation. From Syomei [0] = 1,

[0041]

(Equation 5)

Here, Re and Im respectively represent a real part and an imaginary part in {}. Also, from syomei [1] = 0,

[0043]

(Equation 6)

By performing an inverse discrete Fourier transform on F ′, embedded audio data Sound ′ (k) is obtained. Also, to retrieve security data,

[0045]

(Equation 7)

Sound (k) and Sound '(k) are Fourier-transformed, respectively, and their Fourier-transformed data are compared. If the values are different, signature bit 1 is taken out. Extract bit 0. The algorithm is briefly described below. ｛N is n
Moves from 1 to N in order.

If F [Sound] (n) = F [Sound '] (n), syomei [n-1] = 1; F [Sound] (n) = F [Sound'] (n) If so, syomei [n-1] = 0} An example of the procedure of the above-described audio processing method according to the present invention will be described with reference to the flowchart of FIG. That is,
As a procedure of a method of embedding specific security data in specific audio data according to the present invention, first, in step (1), the specific audio data is selected, and in step (2), the security data of the security data is selected. Check the digital bit number n.

Thereafter, the process proceeds to step (3) to execute a process of dividing the selected specific audio data into a plurality of continuous blocks having at least twice the number of digital bits n of the security data. In step (4), a Fourier transform is performed on the audio data included in the block for each block, and a Fourier transform coefficient is obtained for each block.

Next, proceeding to step (5), from among the plurality of Fourier transform coefficients in the one block, first, two Fourier transform coefficients located farthest from each other are combined with each other as a first one. Selecting a set to form a first Fourier transform coefficient unit, and wherein the four Fourier transform coefficients except for the two Fourier transform coefficients constituting the first unit are located farthest apart from each other 2
Selecting two Fourier transform coefficients as a second set to form a second Fourier transform coefficient unit, and then sequentially forming units in the same manner to form a third, fourth,. Forming a set and forming a number of Fourier transform coefficient units corresponding to the number of digital bits of the security data is continuously performed to form n built-in Fourier transform coefficient units.

Thereafter, in step (6), an operation of applying one of the bit data of the security data to each of the plurality of Fourier transform coefficient units and embedding the security data in the voice data is performed, In step (7), an operation of incorporating the group of blocked Fourier transform coefficients of the audio data in which the security data is embedded into the original audio data is performed.

As an example of the operation of embedding the voice data in the step (6) in this specific example, there is a method shown by a subroutine shown in FIG. That is, step (6)
In step (20) after the start, a predetermined position, for example, the head, that is, unit Ui (i = 1) is selected, and in step (21), the first bit data of the security data is selected. Is determined to be 1 or not.

If NO in step (21), the Fourier transform coefficients F (1) and F (n) in the unit U1 execute the phase change of π in step (22). If the answer is YES in step (21), the Fourier transform coefficient F in the unit U1 is used.
(1) and F (n) execute a phase change of π in step (23).

Thereafter, the process proceeds to a step (24), where it is determined whether or not i of the unit Ui is i = n.
The operation becomes END, and in step (24), N
If O, the process proceeds to step (25), in which the value of i is incremented by 1 and the process returns to step (21), and the operations of step (21), step (22), and step (23) are performed. Repeated.

When it is determined that each bit data of the security data is embedded in all the n units Ui, the subroutine ends, and the process returns to the step (6). Also, after the above step (7),
As an example of the inspection processing operation for the voice data to be inspected or the extraction processing operation of the embedded security data, as shown in the flowchart of FIG. 5, after the start, the security data is embedded in step (30). An operation of dividing the obtained audio data into blocks and performing a Fourier transform on each of the blocks in step (31) is executed.
In step 2), the voice data before embedding the security data is divided into blocks, and in step (33), an operation of performing a Fourier transform on each block is performed.

Thereafter, the flow advances to step (34) to compare the Fourier transform coefficients of the respective blocks with each other.
That is, it is determined whether or not the two coefficients are the same. If YES, it is determined in step (35) that the operation of changing the Fourier transform coefficient by embedding did not exist, and the corresponding bit of the security data is set to 0. And if NO in step (34), step (3)
In 6), it is determined that the bit of the security data is 1.

In another aspect of the present invention, a process for embedding security data in specific audio data is performed by a computer, and the step of selecting the specific audio data, the digitalization of the security data, Checking the number of bits n, dividing the selected specific audio data into a plurality of continuous blocks at least twice the number of digital bits n of the security data, A step of performing a Fourier transform on the audio data contained in the block to obtain a Fourier transform coefficient for each block; first, from among the plurality of Fourier transform coefficients in the block, The two Fourier transform coefficients located apart from each other are selected as a first set, and the first Fourier transform coefficient unit is selected. Forming a first unit, and then, among the respective Fourier transform coefficients excluding the two Fourier transform coefficients constituting the first unit, the two Fourier transform coefficients located farthest from each other are referred to as a second Fourier transform coefficient. To form a second Fourier transform coefficient unit by selecting the set of .times.
Forming an n-th set and forming a number of Fourier transform coefficient units corresponding to the number of digital bits of the security data, wherein each of the plurality of Fourier transform coefficient units is one of bit data of the security data; Applying the security data to the audio data, performing an inverse Fourier transform process for each block on the block of Fourier transform coefficients in which the security data is embedded, and returning the result to the original audio data. A recording medium on which a program for causing a computer to execute a configured audio data processing method is recorded, and further, the recording medium performs a Fourier transform on the original audio data in which the security data is embedded. It is unknown whether the process or the security data is embedded A step of performing the same Fourier transform on the audio data to be inspected, and comparing the data after the Fourier transform processing for both audio data to determine whether or not the audio data to be inspected includes predetermined security data. And a discriminating step.

[0057]

Since the audio processing method according to the present invention has the above-described configuration, the audio data is divided into a plurality of blocks, and a Fourier transform is performed for each block. Since the coefficient is obtained and the value of each bit data of the security data is embedded in the Fourier transform coefficient, predetermined security data can be easily embedded in the voice data under a condition unrelated to human hearing. An audio processing method and an audio processing apparatus which can easily extract the audio data and can easily confirm and grasp the presence or absence of illegally copied audio data are provided.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a specific example of an audio processing device according to the present invention.

FIG. 2 is a diagram illustrating an example of blocking audio data used in the present invention.

FIG. 3 is a flowchart illustrating a procedure of a specific example of an audio data processing method according to the present invention.

FIG. 4 is a flowchart showing a subroutine used in a part of the audio data processing method according to the present invention.

FIG. 5 is a flowchart illustrating a procedure of another specific example of the audio data processing method according to the present invention.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Audio data apparatus 2 ... Audio data storage means 3 ... Security data storage means 4 ... Audio data blocking means 5 ... Blocked Fourier transform coefficient generation means 6 ... Unit formation means 7 ... Security data embedding means 8 ... Inverse Fourier transform processing Means 9 ... Control means 10 ... Security data embedded voice data storage means 11 ... Security data embedded voice data Fourier transform means 12 ... Original voice data or test object voice data storage means 13 ... Fourier transform means 14 ... Comparison means 15 ... Judgment means

Claims (11)

[Claims] When embedding security data in specific audio data, the specific audio data is divided into a plurality of continuous blocks each having at least twice the number of digital bits of the security data. For each block, the audio data of the block is subjected to a Fourier transform to obtain a Fourier transform coefficient for each block, and the Fourier transform coefficients within the block are located at the farthest distance from each other. To form a first unit having two Fourier transform coefficients as a first set, and then the two units located farthest apart from each other except for the two Fourier transform coefficients that constituted the first unit. Forming a second unit having the Fourier transform coefficients as a second set, forming units sequentially in the same manner, and A method of processing voice data, comprising forming a number of Fourier transform coefficient units corresponding to the number of digital bits of data, and embedding one of the bit data of the security data in each of the plurality of Fourier transform coefficient units. . 2. When embedding one of the bit data of the security data in each of the plurality of Fourier transform coefficient units, if the bit data of the security data is 1, the embedding of the Fourier transform coefficient unit is performed. 2. The audio data processing method according to claim 1, wherein the phase of the coefficient value is shifted by π. 3. Embedding one of the bit data of the security data in each of the plurality of Fourier transform coefficient units, if the bit data of the security data is 1, 2. The audio data processing method according to claim 1, wherein the imaginary part and the real part in the coefficient value are exchanged, and the sign of the real part is inverted. 4. A security data-embedded data obtained by performing an inverse Fourier transform process on a series of blocked Fourier transform coefficients including a plurality of Fourier transform coefficients in which the security data is embedded is converted into original voice data. 4. The method for processing audio data according to claim 1, wherein the audio data is incorporated. 5. An object to be inspected which performs a Fourier transform for each block of audio data including a Fourier transform coefficient in which the security data is embedded and which is blocked, and in which whether or not the security data is embedded is unknown. A method for processing audio data, wherein the audio data is subjected to a similar Fourier transform, and the two are compared to inspect the audio data to be inspected. 6. An audio data storage unit, a security data storage unit, an audio data blocking unit, and a blocked Fourier transform coefficient generation unit for performing a Fourier transform on each of the blocked audio data to obtain a Fourier transform coefficient. Means, a plurality of Fourier transform coefficients in the one block, a unit forming means constituting a plurality of units composed of a pair of two Fourier transform coefficients, a plurality of units formed by the unit forming means Security data embedding means for allocating one of the bit data of the security data to each of the units; performing an inverse Fourier transform process for each block on a group of blocked Fourier transform coefficients in which the security data is embedded; Inverse Fourier transform processing means incorporated in the audio data of And a control means for comprehensively controlling each of the above means. 7. A means for performing a Fourier transform on the original voice data in which security data is embedded, wherein the voice data to be inspected is unknown whether or not the security data is embedded. Means for performing a similar Fourier transform to the voice data, and discriminating means for determining whether or not the voice data to be inspected includes predetermined security data by comparing the data after the Fourier transform processing for both voice data. 7. The audio data processing device according to claim 6, further comprising: 8. When embedding the security data in the specific audio data, selecting the specific audio data, confirming the number of digital bits n of the security data, at least transmitting the selected specific audio data to the specific audio data. Dividing the security data into a plurality of continuous blocks each having twice the number of digital bits n; performing a Fourier transform on the audio data included in each block for each block; Obtaining a Fourier transform coefficient; from among the plurality of Fourier transform coefficients in the block, first, two Fourier transform coefficients located farthest apart from each other are selected as a first set and a first set is selected. Forming a Fourier transform coefficient unit of the following: Next, the two Fourier transform coefficients constituting the first unit are removed. Of the Fourier transform coefficients that were located, the two Fourier transform coefficients located farthest apart from each other are referred to as the second.
Forming a second Fourier transform coefficient unit by selecting a set of
... forming an n-th set and forming a number of Fourier transform coefficient units corresponding to the number of digital bits of the security data; Embedding the security data in the audio data by applying one, performing an inverse Fourier transform process for each block on the block of Fourier transform coefficients in which the security data is embedded, and incorporating the result in the original audio data; and And a method for processing audio data. 9. performing a Fourier transform on the original voice data in which the security data is embedded;
A step of performing the same Fourier transform on the audio data to be inspected for which it is unknown whether or not the security data is embedded.By comparing the data after the Fourier transform processing for both the audio data, the audio data to be inspected is 9. The method for processing audio data according to claim 8, further comprising a determining step of determining whether the security data includes predetermined security data. 10. When embedding security data in specific audio data, a step of selecting the specific audio data, a step of confirming a digital bit number n of the security data, Dividing the security data into a plurality of continuous blocks each having twice the number of digital bits n; performing a Fourier transform on the audio data included in each block for each block; Obtaining a Fourier transform coefficient; from among the plurality of Fourier transform coefficients in the block, first, two Fourier transform coefficients located farthest apart from each other are selected as a first set and a first set is selected. Forming a Fourier transform coefficient unit of the following, where the two Fourier transform coefficients forming the first unit are: Among the Fourier transform coefficients except for the two, the two Fourier transform coefficients located farthest apart from each other are defined as the second
Forming a second Fourier transform coefficient unit by selecting a set of
... forming an n-th set and forming a number of Fourier transform coefficient units corresponding to the number of digital bits of the security data; Applying one of the security data to the audio data, performing an inverse Fourier transform process for each block on the block of Fourier transform coefficients in which the security data is embedded, and returning the result to the original audio data; The recording medium which recorded the program which makes a computer perform the processing method of the audio | voice data comprised from this. 11. A step of performing a Fourier transform on the original voice data in which the security data is embedded, and performing a similar Fourier transform on the voice data to be inspected for which it is unknown whether or not the security data is embedded. And a discriminating step of judging whether or not the voice data to be inspected includes predetermined security data by comparing the data after the Fourier transform processing for both voice data. A recording medium storing a program for causing a computer to execute the method for processing audio data according to claim 10.