JP4520922B2 - Data format determination method, apparatus, program, and recording medium - Google Patents

Description

The method if there is any sound data determine Teisu Ru in the digital data, equipment, a program and a recording medium.

There are many proposals for compression coding without distortion. For example, a general-purpose compression encoding method called ZIP is often used for text data such as sentences and programs. On the other hand, as a compression encoding method without distortion such as a music signal, there is, for example, Patent Document 1. This method can compress much more effectively than ZIP.
If the file storing the digital data to be compressed has a uniform extension, it can be determined by looking at the extension whether it is text data or audio data. However, the digital data to be compressed may not have an extension or may not have a correct extension, and it is often unclear whether the data is text data or audio data. In such a case, an efficient compression encoding method cannot be selected and compressed, and therefore data including acoustic data cannot be particularly efficiently compressed.

In other words, if it can be determined whether or not the digital data includes acoustic data, it can be efficiently compressed. However, there is no such determination technique in the prior art.
JP 2005-115267 A

In the prior art, it is possible to determine whether or not sound data is included by looking at the extension of a file in which digital data to be compressed is stored. However, if no extension is added, it cannot be determined whether the digital data includes acoustic data. As a result, since the encoding using the characteristics possessed by the acoustic data cannot be performed, the encoding efficiency is greatly deteriorated. This also applies to digital waveform data obtained by sampling analog waveform data. An object of the present invention, the digital data input is in the child determines whether or not containing waveform data such as sound data.

In the present invention, assuming that the input digital data is a sample value sequence of waveform data in one or more known data formats, the likelihood of the waveform data is obtained using the relationship between the sample values in the sample value sequence. . If one or more of the obtained waveform data characteristics satisfy a preset condition, it is determined that the input digital data includes waveform data, and the waveform data satisfies the condition. If not, it is determined that the input digital data does not include waveform data .

  According to the present invention, even if it is not known whether waveform data is included in the input digital data, the input digital data includes the waveform data by determining the likelihood of the waveform data of the input digital data. It can be determined whether or not. Furthermore, when it is determined that the waveform data is included, the data format as the waveform data can also be predicted, so it is also possible to determine what waveform data format should be used for encoding. Therefore, efficient encoding can be expected.

In the following, an embodiment in which the present invention is applied to acoustic data, which is the most representative analog waveform data, will be described. In order to avoid duplication of description, the same number is assigned to a component having the same function or a process step performing the same process, and the description is omitted.
[First Embodiment]
An example of the functional configuration of the acoustic data format discrimination device of the present invention is shown in FIG. Moreover, the processing flow of the acoustic data format discrimination device 100 is shown in FIG. The data format determination device 100 includes a data recording unit 110, an index determination unit 120, and a data format determination unit 130. The index determination unit 120 includes a storage format recording unit 121 and an index calculation unit 124. The data format determination unit 130 includes an index comparison unit 131.

  The input digital data is first recorded in the data recording unit 110 (S110). As described below, the index determination unit 120 assumes that the input digital data is acoustic data in a preset format, performs acoustic data likelihood determination, and indicates an acoustic data likelihood index in each data format. Is output (S124). First, the index calculation unit 124 of the index determination unit 120 sequentially selects a data format from one or a plurality of data formats for acoustic data set in the storage format recording unit 121 in advance. Next, using the relationship between the sample values in the sample value sequence when each data format is assumed, the input digital data is likely to be the acoustic data when it is assumed that each data format is used. To evaluate. Then, the index value of the sound data likelihood in each data format is output (S124). Here, the preset data format is the number of bits per sample of a plurality of types, whether it is an integer representation or a floating point representation, the byte storage order for a plurality of types of samples, the number of channels of a plurality of types, etc. It is. The digital data used for evaluation may be all input data or a part thereof. In the case of using a part, for example, the first thousands of samples to tens of thousands of samples of digital data may be used. For the evaluation of the likelihood of acoustic data, it is performed for each frame composed of several thousand samples.

Specifically, the following expression is used as an index of the likelihood of sound data.

Here, p is a predetermined positive integer, and k _i (i = 1 to p) is a PARCOR coefficient.
F approximates the ratio of the energy of the input signal to the energy of the prediction residual in the case of p-th order prediction. It can be said that a larger F value is more likely to be acoustic data, and when compressing input digital data, a high compression effect can be expected by using an encoding means for acoustic data. Note that, for example, when calculating an index of the likelihood of sound data in units of frames, if there are a plurality of sample value sequences in an assumed data format, the values of a plurality of indexes for each data format and each frame are obtained. It will be. Thus, when there are a plurality of index values in the same data format, the maximum value among them, that is, the index value that seems to be the most acoustic data, is set as the index value for the sample value string in the data format. .

The data format discriminating unit 130 compares the value of the index (F) of the acoustic data likelihood corresponding to each data format, and provides the data format that gives the maximum index, that is, the data format determined to be the most acoustic data, and the format The value of the index (F) of the likelihood of sound data at is obtained. (S130). Further, the index comparison unit 131 of the data format determination unit 130 compares the acoustic data-like index (F) with a threshold value, and determines whether or not the data is likely to be acoustic data (S131). As will be described in an experimental example described later, for example, a case where the value of F is 1000 or more is determined to be acoustic data. When step S131 is Yes, the data format discrimination | determination part 130 discriminate | determines that it contains acoustic data, and outputs that and the data format of acoustic data (S141). If step S131 is No, the data format determination unit 130 determines that no acoustic data is included, and outputs that effect (S142). Note that when one data format is set in advance, the data format is not selected sequentially in step S124, but it is only necessary to obtain the likelihood of acoustic data using only one data format. Further, there is no need to perform step S130.
By discriminating the data format in this way, it is possible to estimate whether the input digital data includes acoustic data and, if acoustic data is included, the data format.

[Modification 1]
In the first embodiment, F is used as an index indicating the likelihood of sound data. However, in the case of encoding to be converted into the frequency domain, evaluation can be performed using the following expression.

Here, M is a predetermined positive integer, and Y _j (j = 0 to M−1) is a square value of the j-th frequency domain coefficient.
E is an arithmetic mean ratio of square values of coefficients corresponding to the power spectrum. If the coefficient value is constant regardless of j, the minimum value is E = 1. In this case, the sample value sequence when the input digital acoustic data is in the data format is a random number, and it cannot be said that the data format seems to be acoustic data. Therefore, when the input digital data is compressed, compression cannot be expected even if a compression method specialized for acoustic data is used. On the other hand, if the variation of the square value of the coefficient is large, E becomes a large value. It can be said that it seems to be acoustic data because there is a large variation in the value of the coefficient (spectrum) in the frequency domain. Therefore, when compressing input digital data, the amount of data after compression can be reduced by using a compression method specialized for acoustic data. That is, a high compression effect can be expected. This also means that the correlation between sample values is strong.
Therefore, also by this method, it can be estimated whether the input digital data is acoustic data.

[Modification 2]
Energy / F or energy / E can also be used as an index indicating the likelihood of acoustic data. Energy / F or energy / E corresponds to approximation of the energy of the prediction error, and it can be determined that the smaller the value, the more likely it is acoustic data. Therefore, when this index is used, the data format determination unit 130 compares the index (energy / F or energy / E) value corresponding to each data format, and the data format having the smallest value and the value of the index. Is obtained (S130). The index comparison unit 131 of the data format determination unit 130 compares the acoustic data-like index (energy / F or energy / E) with a threshold value, and determines that the data is likely to be acoustic data if it is smaller than the threshold value. It is determined that the sound data does not appear (S131).

[Modification 3]
In the first embodiment and the first and second modifications, the determination is simply made by comparison with the threshold value in step S131. However, depending on whether the data format of the input signal is 1 byte, 2 bytes or more, or depending on the magnitude of energy when the sample value sequence is the amplitude of the waveform, the determination of whether or not acoustic data is included may change. . Therefore, in this modification, an example in which the processing of the data format determination unit 130 is changed is shown.

FIG. 3 shows a processing flow of the data format discrimination device 100 in this modification. In this processing flow, steps S132 to S136 are executed instead of step S131 of FIG. Steps S132 to S136 will be described below. The data format determination unit 103 confirms whether the data format of the input signal is determined to be in 1-byte units (S132). If step S132 is No, the data format determination unit 103 calculates the energy when the sample value sequence is the waveform amplitude (S133), and checks whether the energy is smaller than the threshold (S134). As the threshold value, for example, there is a method of setting the energy to 1/100 of the maximum amplitude. When step S134 is Yes, it progresses to step S241. If step S134 is No, the index comparison unit 131 compares whether the index indicating the likelihood of acoustic data is more likely to be acoustic data than the threshold (S135). In this step, the threshold value for F and E, which are indices indicating the likelihood of acoustic data, may be about 100. When step S135 is Yes, it progresses to step S241, and when it is No, it progresses to step S242. When step S132 is Yes, the index comparison unit 131 compares whether the index indicating the likelihood of acoustic data is more likely to be acoustic data than the threshold (S136). This step is the same as step S131 of the first embodiment (FIG. 2).
Thus, since it is determined in detail whether or not acoustic data is included, depending on the number of bytes and energy, it is possible to expect more accurate determination.

[Second Embodiment]
FIG. 4 shows an example of the functional configuration of the encoding apparatus according to this embodiment. FIG. 5 shows a processing flow of the encoding device 200. The encoding device 200 includes a data recording unit 110, an index determination unit 120, a data format determination unit 130, and an encoding unit 240. The encoding unit 240 includes an acoustic signal encoding unit 241 that is an encoding unit for acoustic signals and a non-acoustic signal encoding unit that is a general-purpose encoding unit such as ZIP. As can be seen from a comparison between FIG. 4 and FIG. 1, the encoding device 200 of FIG. 4 has a configuration in which an encoding unit 240 is added to the acoustic data format determination device 100 of FIG. 1.

The processing flow of FIG. 5 is also a processing flow in which steps S241 and S242 are added to the processing flow of FIG. In step S241, the acoustic signal encoding unit 241 encodes the digital data recorded in the data recording unit 110 according to the data format of the acoustic data determined by the data format determination unit 130. Therefore, the acoustic signal encoding unit 241 needs to be provided with an encoding unit that can support the data format of a plurality of acoustic data. That is, the data format used when the index calculation unit 124 calculates the index of the acoustic data likelihood (step S124) is limited to the range of the encoding method that can be handled by the acoustic signal encoding unit 241. In step S242, the non-acoustic signal encoding unit 242 encodes the digital data recorded in the data recording unit 110 by a general-purpose encoding unit such as ZIP.
As described above, in the present embodiment, the input digital data is encoded as the data format determined to be the most likely sound data when the sound data is included, and general-purpose when the sound data is not included. Therefore, high compression efficiency can be expected.

[Modification]
In the second embodiment, the determination is simply made in comparison with the threshold value in step S131. However, depending on whether the data format of the input signal is 1 byte or 2 bytes or more, or depending on the magnitude of energy when the sample value sequence is the amplitude of the waveform, it is better to use the encoding means for sound. In some cases, it may be better to use acoustic encoding means. Therefore, in this modification, an example in which the processing of the data format determination unit 130 is changed is shown.

FIG. 6 shows a processing flow of the encoding apparatus 200 according to this modification. The processing flow in FIG. 6 is also a processing flow in which steps S241 and S242 are added to the processing flow in FIG. Therefore, the difference between FIG. 6 and FIG. 3 is the same as the difference between FIG. 5 (second embodiment) and FIG. 2 (first embodiment).
Thus, since it is determined in detail whether or not acoustic data is included, depending on the number of bytes and energy, it is possible to expect more accurate determination. Therefore, more efficient encoding can be expected.

In the first and second embodiments, the case where the input data may be digital acoustic data obtained by sampling an acoustic signal has been described. However, if the input data may be obtained by sampling analog waveform data, the present invention can be used with other time-series data. In this case, the present invention can be applied to general waveform data only by changing “determination of the likelihood of acoustic data” in each embodiment to “determination of the likelihood of waveform data”.
In addition, said embodiment can also read and implement the program which makes a computer perform each step of the said method. Also, as a method for reading into the computer, the program is recorded on a computer-readable recording medium, and the program is read from the recording medium into the computer, or the program recorded in the server or the like is read into the computer through an electric communication line or the like. There are methods.

[Experimental example]
In order to evaluate the effects of the present invention, simulation results in the first embodiment are shown below. The simulation conditions are as follows. The first input signal is 2 bytes / sample, 2048 samples / frame data amount (175 frames), and 2-channel acoustic data. The data format set in advance is the number of channels (1, 2, 3). The second input signal is text data in units of 1 byte. The calculated index is energy, F (p = 1, 2, 3, 20). The data format of the acoustic data assumed when calculating F is the same as the data format of the first input (2 bytes, 2 channels). In the simulation, energy and F were calculated for the first nine frames of the input signal. For example, in the first input signal, since the number of channels is 2, it corresponds to 5 frames of the first channel and 4 frames of the second channel.

The simulation result for the first input is shown in FIG. 7, and the simulation result for the second input is shown in FIG. In the case of the first input, the first two frames have a small acoustic data-like index value, but the other values are large. In particular, in the sixth frame at p = 20, the value is as large as 27608. On the other hand, in the case of the second input, all the values of F are 10 or less, which is a small value.
When the first input was encoded as acoustic data, it was compressed to 30.90%, and when encoded by ZIP, it was compressed to 88.14%. When the second input was encoded as acoustic data, it was compressed to 96.40%, and when encoded using ZIP, it was compressed to 20.36%.

Therefore, it can be seen that by evaluating digital data using the index (F) of the likelihood of sound data, it can be determined whether sound data is included, and efficient coding can be expected.
Note that the value of F increases monotonically as p is increased. Further, when the value of E is calculated as a power spectrum obtained by smoothing Y _j or a power spectrum obtained by adding a small positive value, it is known that the value approximates the value of F when p is increased. . Therefore, although it is understood that the number of channels should be 3 even when p is small, it is desirable to use F or E or energy / F or energy / E with p as a value of about 20 as an index. Moreover, at least p should be 3 or more.

The figure which shows the function structural example of the acoustic data format discrimination | determination part of 1st Embodiment. The figure which shows the processing flow of the acoustic data format discrimination | determination part of 1st Embodiment. The figure which shows the processing flow of the acoustic data format discrimination | determination part of the modification 3 of 1st Embodiment. The figure which shows the function structural example of the encoding apparatus of 2nd Embodiment. The figure which shows the processing flow of the encoding apparatus of 2nd Embodiment. The figure which shows the processing flow of the encoding apparatus of the modification of 2nd Embodiment. The figure which shows the simulation result in the case of making acoustic data into input. The figure which shows the simulation result in the case of inputting text data.

Claims (10)

A data format determination method for determining whether or not input digital data includes acoustic data,
For a sample value string in the input digital data,

Where p is a predetermined positive integer, k _i (i = 1 to p) is a PARCOR coefficient, and an index calculation step for obtaining
When the value of F obtained in the index calculation step is greater than or equal to a preset threshold, it is determined that the input digital data includes acoustic data, and when the value of F is smaller than the threshold, input And a data format determining step for determining that the digital data does not include acoustic data . A data format determination method for determining whether or not input digital data includes acoustic data,
For a sample value string in the input digital data,

Where M is a predetermined positive integer, Y _j (j = 0 to M−1) is an index calculation step for obtaining the square of the j-th frequency domain coefficient ,
When the value of E obtained in the index calculation step is greater than or equal to a preset threshold, it is determined that the input digital data includes acoustic data, and when the value of E is smaller than the threshold, input And a data format determining step for determining that the digital data does not include acoustic data . A data format determination method for determining whether or not input digital data includes acoustic data,
For sample value sequences in the input digital data , energy / F, but

, P is a predetermined positive integer, k _i (i = 1 to p) is a PARCOR coefficient, an index calculation step,
When the energy / F value obtained in the index calculation step is smaller than a preset threshold value , it is determined that the input digital data includes acoustic data, and the energy / F value is equal to or greater than the threshold value. A data format discrimination method comprising: a data format discrimination step for determining that the input digital data does not include acoustic data in some cases . A data format determination method for determining whether or not input digital data includes acoustic data,
For the sample value string in the input digital data , energy / E, but

, M is a predetermined positive integer, Y _j (j = 0 to M−1) is an index calculation step for obtaining the square of the j-th frequency domain coefficient ,
The index calculation digital data input when the value of the energy / E obtained there et beforehand set threshold value smaller than in step is determined to contain sound data, the value of the energy / E is the threshold value A data format determination method, comprising: a data format determination step for determining that the input digital data does not include acoustic data in the case of the above . A data format discrimination device for discriminating whether or not input digital data includes acoustic data,
A data recording unit for recording the input digital data;
For a sample value string in the input digital data,

Where p is a predetermined positive integer, k _i (i = 1 to p) is a PARCOR coefficient ,
When the value of F obtained by the index calculation unit is greater than or equal to a preset threshold, it is determined that the input digital data includes acoustic data, and when the value of F is smaller than the threshold, input A data format discriminating device comprising: a data format discriminating unit that judges that the digital data that has been recorded does not contain acoustic data . A data format discrimination device for discriminating whether or not input digital data includes acoustic data,
A data recording unit for recording the input digital data;
For a sample value string in the input digital data,

Where M is a predetermined positive integer, and Y _j (j = 0 to M−1) is the square of the j-th frequency domain coefficient ;
When the value of E obtained by the index calculation unit is greater than or equal to a preset threshold, it is determined that the input digital data includes acoustic data, and when the value of E is smaller than the threshold, input A data format discriminating device comprising: a data format discriminating unit that judges that the digital data that has been recorded does not contain acoustic data . A data format discrimination device for discriminating whether or not input digital data includes acoustic data,
A data recording unit for recording the input digital data;
For the sample value sequence in the input digital data , energy / F, but

, P is a predetermined positive integer, k _i (i = 1 to p) is a PARCOR coefficient ,
If the energy / F value obtained by the index calculation unit is smaller than a preset threshold value , it is determined that the input digital data includes acoustic data, and the energy / F value is equal to or greater than the threshold value. A data format discriminating device comprising a data format discriminating unit that judges that input digital data does not include acoustic data . A data format discrimination device for discriminating whether or not input digital data includes acoustic data,
A data recording unit for recording the input digital data;
For the sample value sequence in the input digital data , energy / E, where

, M is a predetermined positive integer, Y _j (j = 0 to M−1) is the square of the jth frequency domain coefficient ,
When the energy / E value obtained by the index calculation unit is smaller than a preset threshold value , it is determined that the input digital data includes acoustic data, and the energy / E value is equal to or greater than the threshold value. A data format discriminating apparatus comprising: a data format discriminating unit that judges that input digital data does not include acoustic data . The program which makes a computer perform the method in any one of Claims 1-4 . A computer-readable recording medium on which the program according to claim 9 is recorded.

Images