JP2000003194A - Voice compressing device and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain the reproducing signals having a good S/N ratio even though voice signals having any frequency bands are inputted. SOLUTION: The device is provided with an input signal analyzer 2, which analyzes the frequency characteristic of inputted voices, a differencing circuit 1, which conducts a differential pulse code modulation(DPCM) differencing operation based on any one of the plural compressing characteristics (difference equations), and a difference equation controller 3 which selectively switches any one of the difference equations based on the analysis result of the inputted voices. By analyzing the frequency characteristic of the inputted voices and selectively switching the difference equations, that are used in the circuit 1, based on the analysis result, a most suitable difference equation is selected in accordance with the frequency band of the inputted voices and a voice compression is conducted accordingly.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an audio compression device and a recording medium, and more particularly, to a DPCM as an audio compression system.
(Differential Pulse Code Modulation) or AD
It is suitable for use in an apparatus employing PCM (Adaptive-Differential-PCM).

[0002]

2. Description of the Related Art Conventionally, when transmitting or storing an audio signal, the purpose of the present invention is to reduce the amount of transmitted data and extend the time that can be stored in storage media (memory, tape, disk, etc.). As shown in FIG. 8, voice compression and decompression are performed. In FIG. 8, an audio compression signal generated from an input audio by the audio compression device 81 is transmitted via a transmission path (not shown) or stored in a storage medium (not shown). Then, if necessary, the audio decompression device 8
2 and is output as reproduced sound.

[0003] As a compression method of an audio signal, for example, AD
PCM (Adaptive-Differential-PCM) is generally known. This compression method using ADPCM is a method of quantizing the difference between adjacent sample values when sampling a waveform.
dulation), the quantization step size (scaling coefficient) is adaptively changed according to the waveform amplitude. For example, where the amplitude is large, the step size is increased so that it can follow a change in the waveform.

For example, in the case of a PCM audio signal, 16 bits are required per sample, but when DPCM is applied, it can be compressed to about 7 bits. In some cases, it cannot be done. Therefore, by applying a scale factor that is linearly predicted from previously generated DPCM data to a voice compressed signal by the DPCM, it is possible to sufficiently express voice having a large change even with only 7 bits. ing.

[0005] The ADPCM is adopted as a PHS audio compression system, for example. That is, a recent digital telephone line has a sampling frequency of 8 KHz, the number of bits per sample is 8 bits, and the transmission rate is 64 Kbps, but when voice compression is performed by ADPCM, the number of quantization bits is reduced. Is 4 bits, and the transmission rate is 32 Kbps.

When such a PHS is considered, the voice compression device 81 in FIG. 8 is located in the PHS terminal on the transmission side, and the voice decompression device 82 is located in the PHS terminal on the reception side. The audio compression signal generated from the input audio by the audio compression device 81 on the transmission side is transmitted through a telephone line, decompressed by the audio decompression device 82 on the reception side, and output as reproduced audio.
In this example, the audio compression device 81 and the audio decompression device 82 are provided in different terminals, but may be provided in the same housing.

For example, an audio reproducing apparatus or an audio recording apparatus having a function of compressing and storing an audio signal in a memory such as a DRAM, a magnetic tape, or a recording medium such as a disk, and expanding and reproducing the audio signal as necessary. For playback devices, etc.,
The audio compression device 81 and the audio expansion device 82 are provided in the same housing. Also in this case, as the audio compression method, the above-described compression method using DPCM or ADPCM can be adopted.

FIG. 9 is a diagram showing the configuration of a conventional audio compression apparatus employing DPCM, and FIG.
FIG. 11 is a diagram illustrating a configuration of a conventional audio decompression device employing M. First, the operation of audio compression by DPCM will be described with reference to FIG. PCM audio signal at each sample point
PCM [j] is sequentially input to the difference circuit 91, and the following equation (1) is used.
, The difference between adjacent sample values is calculated. DPCM [j] = PCM [j]-PCM '[j-1] …… (1)

In equation (1), j is a value representing the number of a sample point, DPCM [] is a difference value between adjacent sample values, PCM [] is a sample value of a PCM audio signal, and PCM '[] is a PCM audio signal. This is the value obtained by compressing the signal once, expanding it, and returning it to its original state. In order to reproduce a waveform close to the original waveform with the audio decompression device after compressing the audio signal with the audio compression device,
The difference equation (1) must be used using a value such as PCM '[j-1] instead of PCM [j-1].

The difference value DPCM calculated by the difference circuit 91
[j] is compressed by the normalization / quantization device 92 and output as an audio compression signal code [j]. This audio compression signal code
[j] is also supplied to the denormalization / dequantization device 93, and a difference value DPCM ′ [j] is generated by performing a process reverse to the process performed by the normalization / dequantization device 92. This difference value DPCM '[j] is given to the inverse difference circuit 94, and the inverse processing of the difference circuit 91 is performed according to the following equation (2). PCM '[j] = PCM' [j-1] + DPCM '[j] …… (2)

The sample value PCM '[j] of the PCM audio signal obtained here is a value including a quantization error generated by the normalization / quantization device 92, and is delayed by the delay buffer 95 for a predetermined time. After that, it is supplied to the difference circuit 91 and the inverse difference circuit 94. As described above, by using the sample value PCM '[j] including such a quantization error in the difference equation (1), the voice having a waveform very close to the waveform of the actually input voice can be obtained as shown in FIG. Can be reproduced by the audio decompression device shown in FIG.

Next, the operation of voice expansion by DPCM will be described with reference to FIG. Input audio compression signal
code [j] is given to the denormalization / dequantization device 101, and the difference value DPCM '[j] is obtained by performing a process reverse to the process performed by the normalization / quantization device 92 of the audio compression device. Generated. This difference value DPCM '[j] is calculated by the inverse difference circuit 102
, And the reverse processing of the difference circuit 91 is performed according to the above-described inverse difference equation (2). PC sought here
The sample value PCM ′ [j] of the M audio signal is output as a reproduced audio, and is also supplied to the inverse difference circuit 102 after being delayed by a predetermined time by the delay buffer 103.

[0013]

As described above, the DPC
In the M system (similarly in the ADPCM system), a difference type and an inverse difference type are used in a voice compression device and a voice decompression device. This is because, as a general characteristic of an audio signal, the signal pressure increases as the frequency decreases, so that the compression using the differential equation results in the original PCM.
This is because the absolute value of the audio signal can be expected to be smaller than that of the audio signal (both are expressed in two's complement), and the loss during data compression is small, and the quality of reproduced audio is improved.

However, in the conventional DPCM system,
For these difference formulas and inverse difference formulas, fixed ones are used irrespective of the voice frequency band. In this case, the S / N ratio when a low-frequency audio signal is input is good,
When an audio signal having a frequency higher than a certain level is input, the S / N ratio of the reproduced audio is deteriorated. In more detail, the fixed difference and inverse difference equations yield very good results only for audio signals in one frequency band, but less for audio signals in other frequency bands. However, there is a problem that the S / N ratio is deteriorated.

The present invention has been made to solve such a problem, and has been developed by a DPCM system or an ADPC system.
By changing the difference formula and the inverse difference formula based on the M system according to the frequency band of the input audio signal, a reproduced signal having a good S / N ratio can be obtained even when an audio signal of any frequency band is input. The purpose is to be.

[0016]

An audio compression apparatus according to the present invention is an input signal analysis means for analyzing characteristics of an input audio signal, and means for processing the input audio signal, and has a plurality of filter characteristics. A filter means, and control means for controlling to selectively switch a filter characteristic used by the filter means based on an analysis result of an input audio signal by the input signal analysis means.

Here, the filter means is a difference calculation means by DPCM or ADPCM, and may have a plurality of difference equations for realizing the plurality of filter characteristics.

Further, the input signal analyzing means may be means for analyzing a frequency component of the input audio signal.
In this case, the input signal analysis means may analyze a frequency component of the input audio signal by FFT. Further, the frequency component may be analyzed by counting the number of times the sign is inverted in the fixed time section of the input audio signal. Further, the frequency component may be analyzed based on the ratio between the signal size of the input audio signal within a certain time interval and the signal size after filtering. Furthermore, the frequency analysis according to at least one of claims 5 and 6 may be performed a plurality of times over a plurality of time sections.

Further, the control means may switch to a filter characteristic such that the gain of the filter is minimized in the frequency component of the input voice analyzed by the input signal analysis means. Further, a means for transmitting the selection signal of the filter characteristics together with the compressed signal may be provided. Also, an audio compression apparatus of a system having a compressed signal and a multiplication coefficient as a data format after compression, and may include means for transmitting the selection signal of filter characteristics and the multiplication coefficient together with the compression signal.

The computer-readable recording medium of the present invention comprises: input signal analyzing means for analyzing a frequency component of an input audio signal; and filter means for performing a difference operation on the input audio signal by DPCM or ADPCM. Filter means having a plurality of difference formulas for realizing the filter characteristics of the above, and control to selectively switch the difference formula used by the filter means based on the analysis result of the input audio signal by the input signal analysis means. A program for causing a computer to function as control means for executing the program.

[0021]

DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram illustrating a configuration example of a voice compression device according to the present invention, and FIG. 2 is a diagram illustrating a configuration example of a voice decompression device used as a pair with the voice compression device.

In FIG. 1, reference numeral 1 denotes a difference circuit which calculates a difference between successive sample values when sampling a waveform of an input PCM audio signal PCM [j]. The difference equation used in the difference circuit 1 of the present embodiment is, for example, the following equation (11)
To (18). DPCM [j] = PCM [j]-2 * PCM '[j-1] + PCM' [j-2] …… (11) DPCM [j] = PCM [j]-PCM '[j-1] + PCM '[j-2] / 2 …… (12) DPCM [j] = PCM [j]-PCM' [j-1] …… (13) DPCM [j] = PCM [j]-PCM '[j -1] / 2 + PCM '[j-2] …… (14) DPCM [j] = PCM [j] + PCM' [j-2] …… (15) DPCM [j] = PCM [j] + PCM '[j-1] / 2 + PCM' [j-2] …… (16) DPCM [j] = PCM [j] + PCM '[j-1] + PCM' [j-2] / 2… … (17) DPCM [j] = PCM [j] + 2 * PCM '[j-1] + PCM' [j-2] …… (18)

Reference numeral 2 denotes an input signal analyzer, and the input P
The CM audio signal is subjected to frequency analysis to detect a frequency band of the input audio. Numeral 3 denotes a differential control device, which outputs a differential expression selection signal control in accordance with the result of analysis by the input signal analyzer 2, and outputs any one of the eight differential expressions (11) to (18) in the differential circuit 1. Controls whether audio compression is performed using an expression. That is, the difference value DPCM [j] is calculated as (PCM [j], PCM '[j-1], PCM'
As a function of [j-2]), one of the difference expressions (11) to (18) is selected by the difference expression selection signal control according to the frequency band of the input audio signal.

Numeral 4 denotes a normalizing / quantizing device, and the difference value DPCM calculated by the difference circuit 1 according to any of the difference formulas.
By performing normalization and quantization processing on [j], compression is performed to generate a compressed code signal code [j].
In this way, for example, from a 16-bit input PCM audio signal PCM [j], for example, a 7-bit compressed code signal code
[j] can be generated. This compressed code signal
code [j] is given to the inverse normalization / inverse quantization device 5 and the signal mixer 8. The denormalization / dequantization device 5 generates a difference value DPCM ′ [j] by performing a process reverse to the process performed by the normalization / quantization device 4.

Numeral 6 denotes an inverse difference circuit which performs processing opposite to that of the above-described difference circuit 1 so that the original PCM audio signal PCM
Generate a sample value PCM '[j] close to [j]. The inverse difference equations used in the inverse difference circuit 6 are the above-described difference equations (11) to (1).
The following equations (21) to (28) respectively correspond to (8). PCM '[j] = 2 * PCM' [j-1]-PCM '[j-2] + DPCM' [j] …… (21) PCM '[j] = PCM' [j-1]-PCM ' [j-2] / 2 + DPCM '[j] ...... (22) PCM' [j] = PCM '[j-1] + DPCM' [j] …… (23) PCM '[j] = PCM' [j-1] / 2-PCM '[j-2] + DPCM' [j] …… (24) PCM '[j] =-PCM' [j-2] + DPCM '[j] …… (25 ) PCM '[j] = -PCM' [j-1] / 2-PCM '[j-2] + DPCM' [j] …… (26) PCM '[j] = -PCM' [j-1] -PCM '[j-2] / 2 + DPCM' [j] …… (27) PCM '[j] = 2 * PCM' [j-1]-PCM '[j-2] + DPCM' [j] …… (28) These eight sample values PCM '[j] are (PCM' [j-1], PCM '
[j-2], DPCM '[j]) as a function of the difference expression selection signal contr
ol selects one of them.

The PCM audio signal PCM '[j] obtained by the inverse difference circuit 6 according to any one of the inverse difference equations is delayed by the delay buffer 7 for a fixed time, and then transmitted to the difference circuit 1 and the inverse difference circuit 6. Supplied. By using the sample value PCM '[j] including such a quantization error in the difference expression and the inverse difference expression, the audio having a waveform very close to the waveform of the actually input audio can be expanded as shown in FIG. Playback on the device is possible.

As described above, the differential expression control device 3 selects an expression to be actually used from eight expressions as the differential expression in the differential circuit 1 and the inverse differential expression in the inverse differential circuit 6. As to which formula is selected by the difference type control device 3, basically, the input PCM audio signal is subjected to frequency analysis by the input signal analyzer 2, and the most suitable formula for the detected frequency is set. It is also possible for the user to arbitrarily set it from outside.

When one of the difference formula and the inverse difference formula is selected by using the difference formula control device 3, it is necessary to inform the speech decompression device of which formula is selected by some means. For this purpose, in the present embodiment, for example, the signal mixer 8 is provided at the output stage of the audio compression device, and the differential expression selection signal control including the expression number and the like obtained as a result of the analysis is mixed with the audio compression signal and transmitted. ing. As will be described later, a signal distribution device is provided on the audio decompression device side to separate a formula number added to a header or the like and transmitted.

Next, in FIG. 2, reference numeral 11 denotes a signal distribution device which converts a compressed code signal co
de [j] and the differential expression selection signal control are separated, and the separated compressed code signal code [j] is denormalized and dequantized by the denormalizer / dequantizer 12.
, And a difference expression selection signal control is supplied to the inverse difference circuit 13. The inverse normalization / inverse quantization device 12
The difference value DPCM ′ [j] is generated by performing a process opposite to the process performed by the normalization / quantization device 4 of the audio compression device in FIG. 1 (the same process as the denormalization / dequantization device 5). .

The inverse difference circuit 13 applies a difference value DPCM '[j] generated by the inverse normalization / inverse quantization device 12 to
A process reverse to that of the difference circuit 1 in FIG. 1 according to any one of the formulas (21) to (28) selected by the difference formula selection signal control (the same process as the inverse difference circuit 6). I do. The PCM audio signal PCM '[j] obtained here is output as a reproduced audio, and after being delayed by a predetermined time by the delay buffer 14, is supplied to the inverse difference circuit 13.

Next, the meaning of the eight difference equations (11) to (18) in the difference circuit 1 will be described with reference to FIG. FIG. 3 is a diagram showing gain characteristics (filter characteristics of the difference circuit 1) of the difference equation with respect to the frequency. The horizontal axis of the figure shows the frequency of the input PCM audio signal, and the vertical axis shows the absolute value obtained by each difference equation. Shows the average of The three gain characteristics shown in FIG. 3 are respectively obtained by the above-described difference equations (11) and (1).
This corresponds to (5) and (18). Here, 3
Although eight gain characteristics are shown, eight difference expressions (11) to (1)
In the eight gain characteristics corresponding to 8), the frequency at which the underpeak is located is slightly shifted.

In order to obtain a good S / N ratio in reproduced sound, it is desirable that the gain be as small as possible. Therefore, it is desirable to use the eight difference expressions (11) to (18) for a PCM audio signal having a frequency at which an underpeak is located in each gain characteristic. That is, the difference expression (1
It is desirable to use 1) for a low-frequency PCM audio signal, to use the difference equation (15) for an intermediate-frequency PCM audio signal, and to use the difference equation (18) for a high-frequency PCM audio signal.

Conventionally, a differential expression having a characteristic similar to that of the so-called 1 / f fluctuation characteristic of sound (a characteristic that the signal pressure of a sound decreases in proportion to 1 / frequency) is fixed and used. Was. On the other hand, in the present embodiment, depending on where the frequency band of the input PCM audio signal is, compression is performed by selectively using a difference expression of a gain characteristic having an underpeak near the frequency. ing. As a result, the level of the expected value obtained after the DPCM can be suppressed to a low level for the audio signal of any frequency band, and the S / N ratio of the reproduced audio can be kept good.

Next, a method for determining the switching of the difference formula (input P
The method for determining the frequency band of the CM audio signal will be described. Here, by analyzing the frequency components included in the input PCM audio signal and selecting the most suitable DPCM difference equation, it is possible to obtain a reproduced audio signal having a good S / N ratio.

Generally, FFT (Fast Fourier Transform) and DFT are used as means for digitally analyzing frequency components.
(Discrete Fourier Transform) is known, and can be applied to the present embodiment. For example, as a result of performing an FFT analysis on the input PCM audio signal, the PCM audio signal is determined to be a signal having a frequency component with the highest peak. Then, a difference equation having a characteristic having an underpeak near the frequency is selectively used.

According to the frequency analysis means using the FFT, the signal pressure and the phase of each frequency component in the section where the frequency analysis is performed can be calculated relatively accurately, but the amount of memory and the number of operations are extremely large, and the amount of hardware is large. It has the drawback that it is enormous and consumes a lot of power. Therefore, in the present embodiment, the following other frequency analysis means is provided.

The first example of the frequency analysis means according to the present embodiment is to analyze the frequency by counting the number of times the PCM audio signal is inverted (the number of times the signal value crosses "0") within a fixed time interval. Is what you do. That is, as shown in FIG. 4, when the input PCM audio signal is viewed within a certain time interval, the number of zero crossings decreases as the frequency decreases, and the number of zero crosses increases as the frequency increases. Therefore, it is possible to analyze the frequency component of the PCM audio signal by counting the number of times of the zero cross within a certain time.

A second example of the frequency analysis means according to the present embodiment is a method for summing the absolute value of the signal pressure of the input PCM audio signal and the absolute value of the difference value after DPCM processing within a fixed time interval. This is to analyze by comparing with the sum. That is, as shown in FIG. 5, when the frequencies of the voices having the same signal pressure as shown in FIG. 4 are different, if DPCM processing is performed, the amplitude of the obtained difference value becomes smaller at the lower frequency. On the other hand, the higher the frequency, the greater the amplitude of the obtained difference value.

That is, the ratio of the sum of the absolute values of the input PCM audio signals within a certain time period to the sum of the absolute values of the difference values after the filtering process by the DPCM differs depending on the frequency. Therefore, for example, a simple means for filtering the input PCM audio signal at a plurality of different frequencies is provided in the input signal analyzer 2, and the average value (absolute value average) of the value after the filter processing within a certain period of time is prepared. Or, the root mean square) is calculated. Then, by looking at the ratio between the calculation result and the average loudness of the input voice, the frequency of the currently input PCM voice signal is inferred.

As described above, in the second example, several filters are required for analyzing the frequency components. The filters are more detailed and complicated as the filters corresponding to the eight differential expressions in the differential circuit 1 are used. This is not necessary, and the circuit scale does not become so large. Further, the filters corresponding to the eight difference expressions in the difference circuit 1 may be used for analyzing the frequency components.

According to the first and second examples of the frequency analysis means described above, the required hardware amount is F
There is an advantage that the number of memory, the number of operations, and the power consumption are small as compared with the case of FT.
The input PCM audio signal is not always a clean sine wave as shown in FIG. Therefore,
If the method according to the first example or the second example described above is used alone, it is not always possible to analyze a correct frequency component.

Therefore, as a third example of the frequency analysis means, the method of the first example or the second example described above is performed not only once in a certain time interval but in a plurality of times in different time intervals. By performing this operation twice, the accuracy of the analysis result can be improved. Further, as a fourth example, instead of using the method of the first example or the second example described above alone, by using both methods together, or by using both methods several times in different time intervals, By doing so, the accuracy of the analysis result can be further increased.

Next, a second embodiment of the audio compression apparatus according to the present invention will be described. In the above-described embodiment, the case where DPCM is adopted as the compression method of the audio signal has been described. On the other hand, in the second embodiment, a description will be given of an audio compression apparatus that transmits a scale factor mixed with a compressed code signal code [j].

FIG. 6 is a diagram showing an example of the configuration of an audio compression apparatus according to the second embodiment, in which the same blocks as those shown in FIG. 1 are denoted by the same reference numerals. FIG.
FIG. 3 is a diagram showing a configuration example of a voice decompression device according to a second embodiment, and the same blocks as those shown in FIG. 2 are denoted by the same reference numerals.

In the following description, SubInfo [k] is a compression auxiliary signal, SubInfo [k] .SF is a scale factor signal (for example, 5 bits), and SubInfo [k] .LN is a differential selection signal (for example, 3 bits). ), N is the length of the PCM to which the compression auxiliary signal is applied, and SFtbl [n] is a scale factor table (for example, SFtbl
[n] = 2 ^{(n / 4)} ).

That is, the input PCM signal PCM [j] and the compressed code signal Code [j] corresponding to the index of each compressed auxiliary signal SubInfo [k] are as follows. SubInfo [0]: PCM [0] to PCM [N-1], Code [0] to Code [N-1] SubInfo [1]: PCM [N] to PCM [2 * N-1], Code [N ] To Code [2 * N-1] SubInfo [2]: PCM [2 * N] to PCM [3 * N-1], Code [2 * N] to Code [3 * N-1] SubInfo [3] : PCM [3 * N] to PCM [4 * N-1], Code [3 * N] to Code [4 * N-1] SubInfo [4]: PCM [4 * N] to PCM [5 * N- 1], Code [4 * N]-Code [5 * N-1] ……

Hereinafter, the operation of the present embodiment will be described. First, the operation of the audio compression device shown in FIG. 6 will be described.
In the input signal analyzer 62, the input PCM audio signal PC
M [j] is analyzed, and the analysis result is provided to the control device 63 to select a scale factor and a difference expression. Here, the difference equation (11) is used. The normalization / quantization device 4 compresses the difference value DPCM [j] calculated by the difference circuit 1 according to the difference expression (11) according to the following expressions (31) and (32). Code [j] = DPCM [j] / SFtbl [tmp # 11 [j] .SF] …… (31) Code [j] = DPCM [j] / SFtbl [SubInfo [k] .SF] …… (32)

Here, the relationship between Expressions (31) and (32) will be described. In the normalization / quantization device 4 of FIG.
(32) will be used,
This is the processing for the input signal analyzer 62 corresponding to the preprocessing, and the scale factor signal SubInfo [k] .SF has not been determined yet. In the input signal analyzer 62, the scale factor signal SubInfo [k] .SF and the differential expression selection signal SubInfo
Equation (31) is used for the purpose of determining [k] .LN, and is used temporarily by the normalization / quantization device 4 for that purpose.

In the present embodiment, the delay buffer 61 for delaying the input PCM audio signal by the time required for the processing of the input signal analyzer 62 is provided in the preceding stage of the difference circuit 1. Then, the input signal analyzer 62 and the control circuit 63, when applying the expression (31) to the result calculated in accordance with the expression (11) in the difference circuit 1 by the normalization / quantization device 4, generate the compressed code signal Code [j] is a scale factor coefficient tmp # 11 that falls within the 7-bit representation range shown in the example.
[j]. Find the minimum value of SF.

The minimum value is a mapping for the difference value DPCM [j], and has the following value. DPCM [j] tmp # 11 [j] .SF -64 ≤ DPCM [j] <64: tmp # 11 [j] .SF = 0 -76 ≤ DPCM [j] <76: tmp # 11 [j] .SF = 1 -90 ≤ DPCM [j] <90: tmp # 11 [j] .SF = 2 -107 ≤ DPCM [j] <107: tmp # 11 [j] .SF = 3 -128 ≤ DPCM [j] < 128: tmp # 11 [j] .SF = 4 ……

The minimum value is obtained for the following reason. That is, from the above equation (31), when the value of SFtbl [] increases, the quantization step size increases, and the S / N ratio of the reproduced sound deteriorates. Therefore, since the goal is to improve the quality of the reproduced sound, the scale factor coefficient tm as small as possible
The value of p # 11 [j] .SF is desirable, and the answer is to choose the minimum value.

The processing for associating the scale factor coefficients is performed from j = 0 to N−1, and the largest tmp # 11
[j] Ask for SF. This is because it is necessary to select a value that does not cause overflow or underflow when Expression (31) is used. This solution is called SubInfo # 11 [k] .SF. That is, SubInfo # 11 [k] .SF = max (tmp # 11 [k * N], tmp # 11 [k * N + 1],
... tmp # 11 [(k + 1) * N-1]).

Although only the difference equation (11) is shown here, on the assumption that the difference equation is selected,
In order to find the most appropriate equation number and scale factor, the scale factor corresponding to each difference equation is determined, and the combination of the equation number and scale factor with the smallest scale factor value is determined by the final input signal analyzer. The solution is 62. That is, SubInfo [k] .SF = min (SubInfo # 11 [k] .SF, SubInfo # 12
[k] .SF, SubInfo # 13 [k] .SF,... SubInfo # 18 [k] .SF) SubInfo [k] .LN = number of the above minimum value.

The control device 63 includes the input signal analyzer 62
And outputs a difference expression selection signal SubInfo [k] .LN and a scale factor signal SubInfo [k] .SF in accordance with the analysis result of the above. The difference expression selection signal SubInfo [k] .LN is input to the difference circuit 1, the inverse difference circuit 6, and the signal mixer 8, and the scale factor signal SubInfo [k] .SF is normalized and quantized by the normalization / quantization device 4 and the inverse normalization. -Input to the inverse quantization device 5. The scale factor signal SubInfo used in the normalization / quantization device 4
[k] .SF is also given to the signal mixer 8.

Next, the audio decompression device shown in FIG. 7 will be described. As shown in FIG. 7, in the present embodiment, a control device 71 is provided at an output stage of the signal distribution device 11. The control device 71 supplies the difference type selection signal SubInfo [k] .LN to the inverse difference circuit 13 and the scale factor signal SubI
nfo [k] .SF is controlled to be supplied to the inverse normalization / inverse quantization device 12.

In the above embodiment, eight equations (11) to (18) are used as the differential equations to be selected. However, the present invention is not limited to this number. A large difference formula or a small difference formula may be selectively used. When a smaller difference equation is used, only the binomial equations shown in Equations (13) and (15) may be used, or only the other three terms may be used. Or 2
One term formula and three term formulas may be prepared and one of them may be used selectively.

Also, the present embodiment has a system in which an audio compression device and an audio decompression device are provided in separate casings, such as a PHS, and an audio compression signal is transmitted via a transmission path. The present invention can be applied to a system in which an audio compression device and an audio expansion device are provided in the same housing, such as an audio reproduction device or an audio recording / reproduction device.

(Other Embodiments of the Present Invention) Each block shown in FIG. 1 and FIG. 2 or FIG. 6 and FIG.
In the present embodiment, although hardware is used, CP
U, a ROM, a RAM, and the like.
May be realized according to the work program stored in the.

In this case, the work program itself realizes the functions of the above-described embodiment, and the program itself and means for supplying the program to a computer, for example, a recording medium storing such a program are provided by the present invention. Make up the invention. As a recording medium for storing such a program, in addition to the ROM and RAM, for example, a floppy disk, a hard disk, an optical disk,
Magneto-optical disk, CD-ROM, CD-I, CD-R,
A CD-RW, a DVD, a zip, a magnetic tape, a nonvolatile memory card, or the like can be used.

When the computer executes the supplied program code, not only the functions of the above-described embodiment are realized, but also the OS (operating system) or other operating system running on the computer. Needless to say, even when the functions of the above-described embodiments are realized in cooperation with application software or the like, such program codes are included in the embodiments of the present invention.

Further, after the supplied program code is stored in the memory provided in the function expansion board of the computer or the function expansion unit connected to the computer, the function expansion board or the function expansion unit is specified based on the instruction of the program code. It is needless to say that the present invention also includes a case where the CPU or the like provided in the first embodiment performs part or all of the actual processing, and the processing realizes the functions of the above-described embodiments.

[0062]

As described above, according to the present invention, the characteristics of the input voice are analyzed, and the filter characteristics (differential expression) used by the filter means are selectively switched based on the analysis result. The most suitable difference equation can be selected according to the characteristics (for example, frequency characteristics) of the input voice to perform the voice compression, and the reproduced voice having a good S / N ratio can be obtained even if a voice signal of any frequency band is input. Can be obtained.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a configuration of an audio compression device according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a configuration of a voice decompression device according to the first embodiment of the present invention.

FIG. 3 is a diagram showing a gain characteristic of a difference equation with respect to a frequency.

FIG. 4 is a diagram for explaining a first example of frequency analysis means according to the present invention.

FIG. 5 is a diagram for explaining a second example of the frequency analysis means according to the present invention.

FIG. 6 is a diagram illustrating a configuration of an audio compression device according to a second embodiment of the present invention.

FIG. 7 is a diagram illustrating a configuration of a voice decompression device according to a second embodiment of the present invention.

FIG. 8 is a diagram showing a configuration of a conventional audio compression / decompression device.

FIG. 9 is a diagram showing a configuration of a conventional audio compression device.

FIG. 10 is a diagram showing a configuration of a conventional audio decompression device.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 difference circuit 2 input signal analyzer 3 difference type control device 4 normalization / quantization device 5 denormalization / dequantization device 6 reverse difference circuit 7 delay buffer 8 signal mixer 11 signal distribution device 12 denormalization / dequantization Device 13 inverse difference circuit 14 delay buffer 61 delay buffer 62 input signal analyzer 63 control device 71 control device

Claims (11)

[Claims] 1. An input signal analysis means for analyzing characteristics of an input audio signal, a means for processing the input audio signal, a filter means having a plurality of filter characteristics, and an input sound by the input signal analysis means Control means for controlling the filter characteristics used by the filter means to be selectively switched based on a signal analysis result. 2. The filter means according to claim 1, wherein said filter means is DPCM or A.
2. The audio compression device according to claim 1, wherein the audio compression device is a difference calculation unit based on DPCM, and has a plurality of difference expressions for realizing the plurality of filter characteristics. 3. The audio compression apparatus according to claim 1, wherein said input signal analysis means is means for analyzing a frequency component of said input audio signal. 4. The audio compression apparatus according to claim 3, wherein said input signal analysis means analyzes a frequency component of said input audio signal by FFT. 5. The audio compression device according to claim 3, wherein said input signal analysis means analyzes a frequency component by counting the number of times of sign inversion in a predetermined time interval of said input audio signal. . 6. The apparatus according to claim 1, wherein said input signal analyzing means analyzes a frequency component based on a ratio of a signal size of the input audio signal within a predetermined time interval to a signal size after filtering. Item 4. An audio compression device according to item 3. 7. At least one of claims 5 and 6
An audio compression apparatus characterized in that the frequency analysis described in (1) is performed a plurality of times over a plurality of time sections. 8. The sound according to claim 3, wherein said control means switches to a filter characteristic such that a gain of the filter is minimized in a frequency component of the input sound analyzed by said input signal analyzing means. Compression device. 9. The apparatus according to claim 1, further comprising means for transmitting a selection signal of the filter characteristic together with a compression signal.
9. The audio compression device according to any one of claims 8 to 8. 10. An audio compression apparatus of a type having a compressed signal and a multiplication coefficient as a data format after compression, comprising means for transmitting a selection signal and a multiplication coefficient of the filter characteristic together with the compression signal. Claim 1.
9. The audio compression device according to any one of claims 8 to 8. 11. An input signal analysis means for analyzing a frequency component of an input audio signal, comprising:
Alternatively, a filter means for performing a difference operation by ADPCM, the filter means having a plurality of difference formulas for realizing a plurality of filter characteristics, and the filter based on an analysis result of an input audio signal by the input signal analysis means. A computer-readable recording medium on which a program for causing a computer to function as control means for selectively switching a differential expression used by the means is recorded.