JP3321767B2 - Apparatus and method for embedding watermark information in audio data, apparatus and method for detecting watermark information from audio data, and recording medium therefor - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and method for embedding watermark information in audio data, an apparatus and method for detecting watermark information from audio data, and a recording medium thereof. In addition, the present invention relates to a digital watermarking technique suitable for performing copyright protection and the like by adding copyright information and the like to a human ear so as to be indistinguishable.

[0002]

2. Description of the Related Art As a method of embedding watermark information in audio data, a quantization difference at the time of quantizing an analog into a digital waveform, or a quantization noise utilizing type for embedding watermark information in quantization noise has been conventionally used. There are known a phase-based type using a phase difference of an audio signal, a masking effect-based type in which watermark information is embedded using a masking effect of a human ear, and the like. The quantization noise utilization type embeds watermark information in lower bits, so that a large amount of bit information can be embedded.
For example, there is a drawback that the noise is erased by the noise generated by the audience like a sports event, and is heard in the case of classical music. Furthermore, since the data is embedded in the lower bits, there is a disadvantage that the resistance due to noise or the like is weak. On the other hand, in the watermark information embedding method of the phase use type or the masking effect use type, there is a drawback that the sound quality is remarkably deteriorated by embedding the watermark information because the waveform is distorted.

[0003]

SUMMARY OF THE INVENTION An object of the present invention is to improve the above-mentioned drawbacks of the prior art, and in particular, to determine the rate of change of the amplitude of speech and to place watermark information where the rate of change is greater than a predetermined threshold. By embedding, the effect on the auditory characteristics is eliminated, and by embedding the watermark information in a portion where the change rate of the amplitude of the voice is relatively high, the watermark information can be periodically embedded in any waveform. An object of the present invention is to provide an apparatus and method for embedding watermark information into new audio data, and a recording medium for the method. Another object of the present invention is to
It is an object of the present invention to provide a novel watermark information detecting apparatus and method which can detect the watermark information even with a small amount of information, that is, short-time reproduced audio data. Another object of the present invention is to provide a new digital watermarking technique in which watermark information can be detected in an arbitrary section by periodically embedding watermark information.

[0004]

SUMMARY OF THE INVENTION The present invention basically employs the following technical configuration to achieve the above object. That is, a first aspect of an apparatus for embedding watermark information in audio data according to the present invention is an apparatus for embedding watermark information in audio data, and means for dividing pulse-code modulated audio data into blocks of a predetermined length. Means for calculating a rate of change in the amplitude of audio at the embedding position for embedding the watermark information in the blocked audio data; means for determining whether watermark information can be embedded based on the rate of change; Means for deciding whether or not to embed the watermark information is characterized in that the means for deciding whether or not to embed the watermark information is constituted by embedding means for embedding the watermark information in the portion where the embedding of the watermark information is decided. Means for determining whether the rate of change at the embedding position for embedding watermark information exceeds a predetermined threshold value. And a step of embedding the watermark information when the comparing means detects that the rate of change at the position to be embedded exceeds the predetermined threshold. In a third aspect, the means for determining whether or not to embed watermark information includes a calculating means for calculating the threshold value.
Further, a fourth aspect is characterized in that the calculating means obtains the threshold based on a moving average value of a change rate, and a fifth aspect is that the watermark information embedding means A header for identifying watermark information and watermark information following the header are embedded in the block with a predetermined length,
In a sixth aspect, the watermark information is embedded in bits other than the least significant bit of the audio data. In a seventh aspect, the watermark information is embedded in the least significant bit of the audio data. In an eighth aspect, the watermark information is periodically embedded.

[0005] In addition, state-like device for detecting the watermark information from the audio data according to the present invention is an apparatus for detecting watermark information from the audio data, the pulse code modulation audio data of a predetermined length of the block means for extracting means for dividing, the bit information from the predetermined bits of each audio data of each block said block into the bit information
Rate to find the rate of change of voice amplitude at the embedding position
Calculating means and the change rate detected by the change rate calculating means
The weight according to the magnitude of the change rate of the extracted bit information
At the same position of each block
Audio amplitude at the position where the bit data of the watermark information is 0
The sum of each weight according to the magnitude of the rate of change of
Value as well as the bit data at the same position
According to the magnitude of the change rate of the voice amplitude at the position where
The second value is obtained by calculating the sum of the weights, and the first value and the second value are obtained.
By comparing with the second value, the embedded watermark information
Characterized by comprising means for detecting
You.

A first aspect of the method of embedding watermark information in audio data according to the present invention is a method of embedding watermark information in audio data, wherein pulse-code modulated audio data is divided into blocks of a predetermined length. A first step of dividing, a second step of obtaining a change rate of audio data at an embedding position in order to embed the watermark information in the amplitude of the blocked audio, and calculating a threshold value from the change rate A third step of embedding the watermark information at the position when the rate of change of the embedding position for embedding the watermark information exceeds the threshold value. And the second
An aspect is characterized in that, in the fourth step, the audio data is embedded in bits excluding a least significant bit. In a third aspect, the first to fourth steps are repeated. , Embedding watermark information of a predetermined length in the block,
In a fourth aspect, the watermark information is periodically embedded.

[0007] A first aspect of the method for detecting watermark information from audio data according to the present invention is a method for detecting watermark information from audio data. first step and a second step of extracting the bit information from the predetermined bits of each audio data of each block the blocked, the bit data of the watermark information at the same position of each block is divided into blocks Of the rate of change of the voice amplitude at the position where is 0
The first value is obtained by calculating the sum of the respective weights according to the above, and the sum of the respective weights corresponding to the magnitude of the change rate of the voice amplitude at the position where the bit data at the same position is 1 is similarly calculated. A third value for detecting embedded watermark information by calculating a second value by comparing the first value with the second value .
It is characterized by including the step of
In two modes, the third step is a step of calculating a change rate of the amplitude of the sound at the embedding position of each block;
Determining the weight from the obtained change rate, and accumulating the weight corresponding to the bit data of the watermark information in each detected block for each of the bit data 0 and 1. is there.

A first aspect of the recording medium according to the present invention in which the watermark information is embedded in the audio data is a storage medium in which the watermark information is embedded in the audio data. When the change rate is equal to or more than a predetermined threshold value, the watermark information is embedded, and the threshold value is determined based on a change rate of the amplitude of the sound near the position where the watermark information is embedded, and the watermark information is substantially periodically It is characterized by being embedded.
A first aspect of the recording medium on which the computer program of the present invention is recorded is a recording medium on which a computer program for embedding watermark information in audio data is recorded. A first step of dividing the audio data into blocks, a second step of obtaining a rate of change in the amplitude of the audio at the embedding position in order to embed the watermark information in the blocked audio data, and calculating a threshold from the rate of change. A third step to seek;
The rate of change of the embedding position for embedding watermark information is
And a fourth step of embedding the watermark information at the position when the threshold value is exceeded, and recording a program for causing a computer to execute a series of processes including the fourth step. According to a second aspect, a fifth step of dividing pulse code modulated audio data into blocks of a predetermined length, and a sixth step of extracting bit information from predetermined bits of each audio data of each of the divided blocks. And a seventh step of detecting watermark information embedded from the bit data extracted in the sixth step. Things.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS According to the present invention, there is provided an apparatus for embedding watermark information in audio data, comprising: means for dividing pulse-code-modulated audio data into blocks of a predetermined length; Means for calculating the rate of change of the amplitude of the sound at the embedding position for embedding information; means for determining whether or not to embed watermark information based on the rate of change; and means for determining whether or not to embed watermark information. And embedding means for embedding the watermark information in the portion where is determined. Therefore, it is possible to embed the watermark information periodically. In particular, since it is a method of embedding in a portion where the change rate of the amplitude of the sound is high, the sound quality is not deteriorated,
Can be embedded. In addition, since the embedding is performed in a portion having a relatively high change rate, the watermark information can be periodically embedded regardless of the waveform of the recorded audio data.

According to the present invention, there is provided an apparatus for detecting watermark information from audio data, comprising: means for dividing pulse-code modulated audio data into blocks of a predetermined length; Means for extracting bit information from predetermined bits of data, and means for detecting watermark information embedded from the bit data extracted by the bit information extracting means. Thus, it is possible to always detect embedded watermark information from audio data arbitrarily extracted.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An apparatus and method for embedding watermark information in audio data according to the present invention and a specific example of an apparatus and method for detecting watermark information from audio data will be described in detail with reference to the drawings. First, an apparatus and method for embedding watermark information in audio data according to the present invention will be described. The watermark information embedding method according to the present invention detects a rate of change of audio data at a temporal position of audio data to be embedded, that is, a difference in audio amplitude at a predetermined time. The watermark information is embedded only when the rate of change is higher than a predetermined threshold, whereas the watermark information is not embedded when the rate of change is lower than the predetermined threshold, so that the watermark information cannot be perceived in terms of audibility. .

In particular, in the present invention, since the threshold is determined based on a moving average value of a predetermined section before a position where watermark information is to be embedded, the recorded audio waveform is Regardless of the waveform, watermark information can be embedded periodically.

FIG. 1 is a functional block diagram of a specific example of an apparatus for embedding watermark information of audio data according to the present invention, FIG. 2 is a flowchart for explaining the operation of this apparatus, and FIG. 3 is an example of a temporal change of the audio data. FIG. 4 is a diagram for explaining a state in which watermark information is embedded in a block. In these figures, there is shown a device for embedding watermark information in audio data, wherein pulse code modulated audio data DT is shown. Means for dividing the audio data DT into blocks of a predetermined length; means 2 for calculating a rate of change s of the amplitude of the audio DT at an embedding expected position for embedding the watermark information in the audio data DT divided into blocks; Means 3 for determining whether to embed watermark information based on the rate s, and means 3 for deciding whether or not to embed watermark information embed the watermark information at the position where the embedding of watermark information is determined. 100, characterized by being configured by the embedding means 4 are shown.

The means 3 for determining whether or not the watermark information can be embedded includes a comparing means 32 for comparing whether or not the rate of change s at the embedding position for embedding the watermark information exceeds a predetermined threshold value ha. And the comparing means 32
However, when it is detected that the change rate s at the position to be embedded exceeds the predetermined threshold value ha, it indicates that the watermark information is to be embedded at that position.
Further, it is shown that the means 3 for determining whether or not the watermark information can be embedded includes a calculating means 31 for calculating the threshold value ha.

Next, this device will be described in more detail. First, to embed watermark information periodically,
The entire audio data DT is divided into blocks so as to include a fixed number of sample points.
The following watermark information embedding processing is performed. In order to reduce the influence of auditory distortion caused by embedding the watermark information, the change rate s of the music amplitude is obtained, and the watermark information is embedded in the data of the sample point having the large change rate s.

As an example of calculation of the change rate, the change rate s [i] of the amplitude of the sample point i is obtained as a difference between the amplitudes w [i + 5] and w [i-5] of the music data between the 10 points before and after i. , Simplified and calculated as in equation (1). Further, here, in order to detect the information, in order to prevent the change rate from changing before and after embedding the watermark information, before the change rate is calculated, the bit below the watermark information embedding is set to 0, and then the change rate is calculated. .

[0017]

(Equation 1)

In the present invention, the watermark information is embedded only at the position where the rate of change obtained by the equation (1) is larger than a certain threshold. In this case, the threshold is fixed for all blocks. If the value is set to a value, the amount of watermark information that can be embedded in blocks having many points with a low rate of change becomes very small. Make it possible to embed watermark information.

For example, specifically explaining the music data DT shown in FIG. 3, the moving average value a of the change rate is calculated from the change rates at 11 points.
If s [i] is obtained as in equation (2), and 1/3 of the value is the threshold ha [i] of the sample point i, the moving average value as
[I] and the threshold value ha [i] can be obtained from the following equation.

As [i] = ｛s [i-10] + s [i-9] + s [i-8] + s [i-7] + s [i-6] + s [i-5] + s [i-4] ] + S [i-3] + s [i-2] + s [i-1] + s [i] / 11 (2) ha [i] = as [i] / 3 (3)

Next, the embedding of the header and the watermark information will be described. Ib [i] is assigned to the bit information of the watermark information in order from the leading sample point of the block. If the information of the bit string in which the watermark information of the audio data is embedded is wb [i], the watermark information bit corresponding to the sample point is replaced with the music data bit only when the change rate is greater than the threshold value. Embed bits of information. For example, bits of music data larger than a threshold are defined as wb [i-5], wb
When [i-3] and wb [i] are set, the bits of the watermark information to be embedded are Ib [i-5], Ib [i-3], Ib
[I], and Ib [i-4], Ib [i-2], Ib
[I-1] is not embedded. By performing the above processing on all the blocks divided into blocks,
The bits of the watermark information are periodically embedded.

FIG. 4 shows blocks A0, A1, A2,.
.. Is a diagram schematically showing a state in which watermark information Ib [0] to Ib [9] are embedded based on a threshold value for An, and watermark information is embedded in bits where Ib is not written in each block. It is not shown.

FIG. 2 is a flowchart showing the flow of embedding the watermark information. As described above, the first step S1 of dividing the pulse code modulated audio data into blocks of a predetermined length, A second step S2 for obtaining a rate of change in the amplitude of the voice at the embedding position in order to embed the watermark information in the converted voice data, a third step for calculating a threshold value from the rate of change, and
The rate of change of the embedding position for embedding watermark information is
If the threshold value is exceeded, the watermark information is embedded in the fourth steps S3, S4, and S5 of embedding the watermark information at that position.

Next, an apparatus and a method for detecting embedded watermark information will be described. The watermark information according to the present invention also includes bits in which the watermark information is not embedded, depending on the threshold value. For this reason, it is configured that the embedded bits are estimated and specified by using a statistical method as described below.

FIG. 5 is a functional block diagram of a specific example of a device for detecting watermark information according to the present invention, and FIGS. 6 and 7 are flow charts for explaining the operation of this device.
An apparatus for detecting watermark information from audio data, comprising: means 5 for dividing pulse-code modulated audio data into blocks of a predetermined length; and means for dividing predetermined bits of each audio data of each of the blocks into blocks. A watermark information detecting device 110 comprising means 6 for extracting bit information and means 7 for detecting embedded watermark information from the bit data extracted by the means 6 for extracting bit information.
It is shown.

Further, the means 7 for detecting the embedded watermark information is provided such that the bit data of the watermark information at the same position of each block in which each watermark information is embedded is 0.
And the first value is obtained by calculating the sum of the weights of the positions, and the second value is similarly calculated by calculating the sum of the weights of the positions where the bit data of the watermark information at the same position in each block is 1. A first calculating means for obtaining a value; and a comparing means for comparing the first value and the second value, wherein the embedded bits of the watermark information are estimated based on the comparison result of the comparing means. 2 shows a watermark information detecting device configured as described above. Further, the means 7 for detecting the embedded watermark information includes a second calculating means 73 for calculating the change rate s of the amplitude of the sound at the position where each watermark information is embedded.
A watermark information detection device is shown, comprising a third calculation means 74 for obtaining the weight from the change rate s detected by the second calculation means 73.

Next, this device will be described more specifically. First, music data is divided into blocks using the same number of data points as in the embedding process. Next, for each block, bit information of a bit string in which watermark information is embedded is extracted. By the way, in the present invention, since the watermark information bit is embedded in a portion where the rate of change s of the audio data is large, at the same sample point of each block, the extracted bit is “0”, , A weight corresponding to the magnitude of the rate of change at the sample point of each block is added over all the blocks, the sum thereof is obtained, and finally, in the magnitude judgment, whether the watermark information bit is 0 or 1 is obtained. Is estimated.

This is represented by the following equation in the actual calculation. i is the sample point, j is the order of the block, and the weight of the block at the sample point is k [i] [j], and ib
It is assumed that 0 [i] [j] is obtained by adding a weight over each block when the bit is “0”, and ib1 [i]
Assuming that [j] is obtained by adding the weight over each block when the bit is “1”, it can be expressed by the following equation. ib0 [i] [j] = ib0 [i] [j-1] + k [i] [j] (4) ib1 [i] [j] = ib1 [i] [j-1] + k [i ] [J] (5)

As the weight k [i] [j], the rate of change at the j-th block and the i-th sample point is s.
[I] [j], the maximum amplitude of the audio data
If it is l, it can be simplified and expressed as follows. k [i] [j] = s [i] [j] / maxlevel (6)

Then, from the results of equations (4) and (5), the bits of the watermark information are estimated from the magnitude judgment. Assuming that the sum in the case of "0" is Ib0 and the sum in the case of "1" is Ib1, step S13 in FIG. 6 can be shown as a flowchart in FIG.

FIG. 8 specifically shows the process of calculating ib0 and ib1. FIG. 8 (a) shows block 0 (j = 0).
8 (b) shows the result of adding the result of block 1 (j = 1) to the result of block 0.

Next, as shown in FIG. 8, finally, from the estimation result of the watermark embedding bits, as shown in FIG. 9, header information indicating the head position of the watermark information is detected. Detect incoming watermark information bits. As described above, the method of detecting watermark information from audio data according to the present invention includes the first step S11 of dividing pulse-code modulated audio data into blocks of a predetermined length; Second step S12 of extracting bit information from predetermined bits of each audio data
And a third step S13 for detecting embedded watermark information from the bit data extracted by the bit information extracting means. The third step S13 includes the steps of: The first value is obtained by calculating the sum of the weights of the positions where the bit data of the watermark information at the same position of the block is 0, and similarly, the sum of the weights of the positions where the bit data of the same position is 1 is calculated. Calculating a second value, and comparing the first value with the second value to detect an embedded bit of the watermark information. The third step S13 includes:
Calculating the rate of change of the amplitude of the sound at the embedding position of each block; obtaining the weight from the obtained rate of change; and determining the weight corresponding to the bit data of the watermark information in the detected block by bit data. 0 and 1 respectively. The embedding and detection of the watermark information according to the present invention may be achieved by software or may be configured by hardware.

Therefore, in the recording medium according to the present invention in which the watermark information is embedded in the audio data, the watermark information is embedded when the change rate of the amplitude of the audio at the position where the watermark information is embedded is not less than a predetermined threshold value. The threshold value is determined based on the change rate of the amplitude of the sound near the position where the watermark information is embedded, and the watermark information is embedded substantially periodically. A first aspect of the recording medium on which the computer program of the present invention is recorded is a recording medium on which a computer program for embedding watermark information in audio data is recorded. A first step of dividing the audio data into blocks, a second step of obtaining a rate of change in the amplitude of the audio at the embedding position in order to embed the watermark information in the blocked audio data, and calculating a threshold from the rate of change. A series of processes including a third step of obtaining the watermark information and a fourth step of embedding the watermark information at the position when the rate of change of the embedding position for embedding the watermark information exceeds the threshold value is performed by the computer. According to another aspect of the present invention, a program for executing the program is recorded. A fifth step of dividing over data into blocks of a predetermined length,
A sixth step of extracting bit information from predetermined bits of each audio data of each of the block blocks;
And a seventh step of detecting embedded watermark information from the bit data extracted in the sixth step. The program records a program for causing a computer to execute a series of processes.

[0034]

The present invention is constructed as described above.
The following effects are obtained. (1) Since the watermark information is embedded in a place where the amplitude of the voice changes greatly, the influence is less perceptual. Further, since the positions of the sample points to be embedded are dispersed, the watermark information is hard to be found, and the distortion given to the voice is also dispersed, so that it has almost no auditory effect. (2) Since the audio data is blocked and the watermark information is embedded in each of the blocked data, the watermark information is periodically embedded in all the audio data, and the watermark information is detected from the arbitrarily cut data. It is possible. (3) Since the embedded watermark bit information is specified by a weighted majority decision, it is resistant to bit errors.

[Brief description of the drawings]

FIG. 1 is a functional block diagram of an apparatus for embedding watermark information in audio data according to the present invention.

FIG. 2 is a flowchart illustrating the operation of FIG. 1;

FIG. 3 is a waveform diagram showing an example of a temporal change of audio data.

FIG. 4 is a diagram for explaining a state in which watermark information is embedded in a cut-out block.

FIG. 5 is a functional block diagram of an apparatus for detecting embedded watermark information.

FIG. 6 is a flowchart illustrating the operation of FIG. 5;

FIG. 7 is a flowchart illustrating the operation of FIG. 5;

FIG. 8 is a diagram illustrating a specific example of calculation for estimating a detected bit.

FIG. 9 is a diagram illustrating a specific example of bits that are detected watermark information.

[Explanation of symbols]

1, 5 means for dividing audio data into blocks of a predetermined length 2 means for calculating the rate of change of the amplitude of audio 3 means for determining whether or not embedding is possible 4 embedding means for embedding watermark information 6 means for extracting bit information 7 Means for estimating embedded watermark information 31 Calculation means for obtaining a threshold value 32, 72 Comparison means 71 First calculation means 73 Second calculation means 74 Third calculation means

──────────────────────────────────────────────────続 き Continued on the front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G10L 11 / 00,19 / 00 G09C 5/00

Claims (18)

(57) [Claims] 1. An apparatus for embedding watermark information in audio data, comprising: means for dividing pulse-code-modulated audio data into blocks of a predetermined length; and embedding the watermark information in the blocked audio data. Means for calculating the rate of change of the amplitude of the sound at the embedding position; means for determining whether or not watermark information can be embedded based on the rate of change; and a part where the means for determining whether or not embedding has determined embedding of watermark information. And an embedding unit for embedding the watermark information in the audio data. 2. The means for determining whether watermark information can be embedded includes a comparing means for comparing whether the rate of change at an embedding position for embedding watermark information exceeds a predetermined threshold value, 2. The audio according to claim 1, wherein the comparing means embeds the watermark information when detecting that the rate of change at the position to be embedded exceeds the predetermined threshold. A device that embeds watermark information in data. 3. The apparatus for embedding watermark information in audio data according to claim 1, wherein the means for determining whether or not to embed the watermark information includes an arithmetic means for calculating the threshold value. 4. The apparatus according to claim 3, wherein said calculating means obtains said threshold value based on a moving average value of a change rate. 5. The watermark information embedding means embeds a header for identifying watermark information and watermark information following the header to a predetermined length in the block. An apparatus for embedding watermark information in the audio data according to any one of 1 to 4. 6. The apparatus according to claim 1, wherein the watermark information is embedded in bits other than the least significant bit of the audio data. 7. The apparatus according to claim 1, wherein the watermark information is embedded in a least significant bit of the audio data. 8. The apparatus according to claim 1, wherein the watermark information is periodically embedded. 9. An apparatus for detecting watermark information from audio data, comprising: means for dividing pulse code modulated audio data into blocks of a predetermined length; Means for extracting bit information from predetermined bits, and a rate of change in the amplitude of audio at the position where the bit information is embedded.
From the change rate calculating means for calculating the change rate and the change rate detected by the change rate calculating means.
Weighting to find weights according to the rate of change
Calculating means, and bit data of the watermark information at the same position of each block.
According to the magnitude of the change rate of the voice amplitude at the position where the
The first value is obtained by calculating the sum of each weight, and similarly,
The sound amplitude at the position where the bit data at the same position is 1
The sum of the weights according to the magnitude of the rate of change of the width is obtained and the second
And comparing the first value with the second value.
And a means for detecting embedded watermark information . An apparatus for detecting watermark information from audio data. 10. A method for embedding watermark information in audio data.
Is a method of transmitting pulse code modulated audio data to a block of a predetermined length.
A first step of dividing the audio data into blocks, and embedding the watermark information in the blocked audio data.
To determine the rate of change in audio amplitude at the embedding position.
A second step, and a third step of calculating a threshold from the change rate.
And the rate of change of the embedding position for embedding watermark information
Is greater than the threshold, the watermark information is
Embedding watermark information in audio data, characterized by including a fourth step of embedding in the position.
How to fit. 11. The method according to claim 11, wherein in the fourth step, the audio data
Data is embedded in bits except the least significant bit
Watermark information is embedded in the audio data according to claim 10.
How to fit. 12. The first to fourth steps are repeated.
By doing so, a predetermined length of watermark information is stored in the block.
The sound according to claim 10, wherein the sound is embedded.
A method of embedding watermark information in voice data. 13. The watermark information is periodically embedded.
The method according to any one of claims 10 to 12, wherein
How to embed watermark information in the audio data. 14. Detecting watermark information from audio data.
A pulse code modulated audio data of a predetermined length.
A first step of dividing the audio data of each of the divided blocks into
A second step of extracting bit information from the bits of the watermark information at the same position of each block.
According to the magnitude of the change rate of the voice amplitude at the position where the
The first value is obtained by calculating the sum of each weight, and similarly,
The sound amplitude at the position where the bit data at the same position is 1
The sum of the weights according to the magnitude of the rate of change of the width is obtained and the second
And comparing the first value with the second value.
A third step of detecting embedded watermark information
It detects the watermark information from the audio data comprising a and
How to get out. 15. The third step comprises the steps of:
A step for calculating the rate of change of the audio amplitude at the embedding position
Determining the weight from the determined change rate;
The bit data of the watermark information in each detected block is
The corresponding weights are accumulated for bit data 0 and 1 respectively.
The sound of claim 14, comprising:
A method for detecting watermark information from voice data. 16. Watermark information embedded in audio data
The storage medium, wherein the rate of change of the amplitude of the sound at the position where the watermark information is embedded is
When the watermark information is embedded when it is equal to or greater than a predetermined threshold,
In both cases, the threshold is near the position where the watermark information is embedded.
Determined based on the rate of change of the amplitude of the
The feature is that the scarf information is embedded almost periodically.
Recording medium. 17. A method for embedding watermark information in audio data.
A computer-readable recording medium on which a pulse code modulated audio data is stored in a block of a predetermined length.
A first step of dividing the audio data into blocks, and embedding the watermark information in the blocked audio data.
To determine the rate of change in audio amplitude at the embedding position.
A second step, and a third step of calculating a threshold from the change rate.
And the rate of change of the embedding position for embedding watermark information,
If the threshold value is exceeded, the watermark information
And a fourth step of embedding in a computer.
A recording medium on which a program is recorded. 18. A method for storing pulse code modulated audio data in a location
A fifth step of dividing into blocks of constant length, each predetermined audio data of each block the blocked
A sixth step of extracting bit information from the bits of the data, and embedding the bit information from the bit data extracted in the sixth step.
And a seventh step of detecting the inserted watermark information .
18. The program according to claim 17, wherein the program is recorded.
Recording medium.