JP2006145782A - Encoding device and method for audio signal - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To reduce a throughput of quantization processing in encoding of an audio signal. <P>SOLUTION: The system comprises a frame division section (1), an auditory psychology operation section (2), a filter bank (3), a scale factor calculation section (4) which weights the spectrum of each frequency band by the operation results of the auditory psychology operation section (2), a quantization step determination section (7) which determines the quantization step over the entire part of the spectrum by subtracting the information amount over the entire part of the spectrum after the quantization from the auditory information amount that the entire part of the spectrum having prior to the weighted quantization and integrating the coefficient obtained from the increment width of the quantization roughness, a spectrum quantization section (8), a bit shaping section (9) which outputs a bit stream shape obtained by shaping the bit stream. The quantization step determination section (7) predicts the information amount over the entire part of the quantized spectrum based on the the amount of the bit allocated to a frame which is an object for encoding. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an audio signal encoding apparatus and method.

  In recent years, high-quality and high-efficiency audio signal encoding technology has been developed for DVD-Video audio tracks, portable audio players using semiconductor memory, HDD, etc., music distribution via the Internet, music to home servers in home LAN It is widely used for accumulation, etc., and it is widely spread and its importance is increasing.

  Many of such audio signal encoding techniques perform time-frequency conversion using a conversion encoding technique. For example, in MPEG-2 AAC, Dolby Digital (AC-3), etc., a filter bank is composed of a single orthogonal transform such as MDCT, and is used for MPEG-1 Audio Layer III (MP3) and ATRAC (MD In the encoding method, a filter bank is configured by connecting subband division filters such as QMF and orthogonal transforms in multiple stages.

  In these transform coding techniques, by performing masking analysis using human auditory characteristics, spectral components that have been determined to be masked are removed, or quantization errors that are masked are allowed to be expressed. Therefore, the amount of information is reduced and the compression efficiency is increased.

  In many of these transform coding techniques, the amount of information held in a spectrum is compressed by nonlinearly quantizing the spectrum component. For example, in MP3 and AAC, the amount of information is compressed by raising each spectral component to the power of 0.75.

  Also, with these transform coding techniques, the input signals converted into frequency components by the filter bank are grouped into divided frequency bands set based on the human auditory frequency resolution, and each divided frequency band is quantized at the time of quantization. The normalization coefficient is determined from the result of auditory analysis, and the amount of information is reduced by expressing the frequency component by a combination of the normalization coefficient and the quantized spectrum. This normalization coefficient is actually a variable for adjusting the quantization roughness for each divided band. When the normalization coefficient changes by 1, the quantization roughness changes by one step. In MPEG-2 AAC, this divided frequency band is called a scale factor band (SFB), and the normalization coefficient is called a scale factor.

  In these transform coding systems, the amount of code is controlled by controlling the quantization roughness of the entire frame, which is a coding unit. In many transform coding schemes, the quantization roughness is controlled in steps with an integer power of a certain radix, and this integer is called a quantization step. In the MPEG audio standard, this quantization step for setting the quantization roughness of the entire frame is called “global gain” or “common scale factor”. The scale factor described above is expressed as a relative value to the quantization step, thereby reducing the amount of information necessary for the sign of these variables.

  For example, in MP3 and AAC, when these variables change by 1, the actual quantization roughness changes by 2 3/16 power.

  In the transform coding quantization process, the scale factor is controlled to reflect the result of the auditory operation, and the quantization distortion is controlled so that the quantization error is masked. At the same time, the quantization step is controlled to control the entire frame. The amount of code of the entire frame must be controlled by appropriately adjusting the quantization roughness of the frame. Since these two kinds of numerical values that determine the quantization roughness have a significant influence on the encoding quality, it is required to perform these two controls simultaneously and efficiently with caution and accuracy.

  In the MPEG-1 Audio Layer III (MP3) standard (ISO / IEC 11172-3) and MPEG-2 AAC standard (ISO / IEC 13818-7), the scale factor and global gain are appropriately controlled during quantization. As a method, a method has been introduced in which a repetitive process is performed by a double loop of a distortion control loop (outer loop) and a code amount control loop (inner loop). Hereinafter, this method will be described with reference to the drawings. For convenience, the case of MPEG-2 AAC will be described as an example.

  FIG. 13 is a simple flowchart of the quantization process described in the ISO / IEC standard.

  First, in step S501, all SFB scale factors and global gains are initialized to 0, and a distortion control loop (outer loop) is entered.

  In the distortion control loop, first, a code amount control loop (inner loop) is executed.

  In the code amount control loop, first, in step S502, one frame, that is, 1024 spectral components are quantized according to the following quantization formula.

In Equation (1), Xq is a quantized spectrum, x _i is a spectrum before quantization (MDCT coefficient), global_gain is a global gain, and scalefac is an SFB scale factor including this spectral component.

  Next, in step S503, the number of bits used for one frame when these quantized spectra are Huffman-coded is calculated and compared with the number of bits allocated to the frame in S504. If the number of used bits is larger than the allocated number of bits, the global gain is increased by 1 in S505, the quantization roughness is increased, and the process returns to the spectral quantization in S502 again. This repetition is performed until the number of bits necessary after quantization becomes smaller than the allocated number of bits, the global gain at this point is determined, and the code amount control loop is terminated.

  In step S506, the quantization error is calculated by dequantizing the spectrum quantized by the code amount control loop and taking the difference from the spectrum before quantization. This quantization error is collected for each SFB.

  In step S507, it is checked whether the scale factor has become larger than 0 in all SFBs, or whether the quantization error is within the allowable error range. If there is an SFB that does not satisfy any of these conditions, the process proceeds to step S508, where the scale factor of the SFB in which the quantization error is not within the allowable error range is increased by 1, and the distortion control loop process is repeated again. The permissible error for each SFB is obtained before the quantization process by auditory calculation.

  As described above, the quantization processing method described in the ISO standard is composed of a double loop, and the global gain and the scale factor can only be controlled by one step. Spectral quantization and bit computation will be repeated endlessly until convergence.

  Here, for example, in the case of MPEG-2 AAC, the spectrum quantization is a process with a large calculation amount because the calculation of the expression (1) is performed 1024 times every time the process is performed once. In addition, since there are 11 types of Huffman code tables that are searched during the bit calculation, if the Huffman code table is fully searched, the calculation amount of the bit calculation inevitably increases.

  Furthermore, in the distortion control loop, the quantization error is calculated after inverse quantization, but this processing also requires a calculation amount. Therefore, it takes a huge amount of processing before the double loop converges.

  In order to solve this problem, various attempts have been made to reduce the processing amount by reducing the number of repetitions of the double loop.

  For example, Japanese Patent Laid-Open No. 2003-271199 (Patent Document 1) discloses that a common scale factor and a scale factor are controlled in a step-by-step manner according to the number of steps determined according to the characteristics of the Huffman code table. A method of reducing the processing amount by reducing the number of loops of each of the heavy loops is disclosed.

  Japanese Patent Laid-Open No. 2001-184091 (Patent Document 2) discloses a method in which an estimation value of a quantization step is first calculated, a scale factor is calculated according to MNR, and then a normal inner loop is executed. Has been.

  In addition, the paper “A robust and efficient implementation of MPEG-2 / 4 AAC Natural Audio Coders” (AES 112th Convention Paper, 2002) (Non-Patent Document 1) by ADDuenes, R. Perez, B. Rivas et al. By using the modified form of (1) and the permissible error energy for each SFB obtained by auditory analysis, the scale factor is calculated appropriately prior to spectral quantization, thereby enabling the distortion control loop outside the double loop. A method for removing the problem and reducing the processing amount is introduced.

  By using these conventional techniques, the convergence of the double loop of the quantization process can be accelerated, and the amount of the quantization process can be reduced to some extent.

JP 2003-271199 A JP 2001-184091 A.D.Duenes, R.Perez, B.Rivas et al., “A robust and efficient implementation of MPEG-2 / 4 AAC Natural Audio Coders”, AES 112th Convention Paper (2002)

  However, with the conventional technology, it is impossible to prevent the double loop described in the ISO standard document from being completely repeated. Therefore, the quantization process is finished unless the spectrum quantization is repeated several to several tens of times. The amount of quantization processing in the entire encoding processing was still large.

  In particular, it is possible to eliminate the outer distortion control loop by calculating the scale factor in advance using the result of auditory calculation in the double loop, but the quantization step should be calculated before quantization. Was impossible with conventional technology.

  Therefore, in the conventional technique, the spectrum quantization and the bit calculation in the code amount control loop are repeatedly performed, and there is a problem that the processing amount is wasted.

  The present invention has been made in view of the above-described problems, and an object thereof is to greatly reduce the amount of quantization processing in audio signal encoding.

  The present invention is basically based on the idea that the overall quantization roughness can be obtained by dividing the amount of information before quantization by the amount of information after quantization. It is what you want to ask before. Here, since the quantization roughness is generally obtained by multiplying the radix by the quantization step, when taking the logarithm with the base as the base to obtain the quantization step, the division of the information amount is the difference of the information amount. To change. An accurate quantization step can be obtained by multiplying this difference by a coefficient determined by the quantization step size. In addition, the actual amount of information after quantization can only be obtained after quantization, but since it can be predicted from the amount of code assigned to the frame, accurate quantization prior to quantization is made using this prediction. It is the one that seeks the conversion step.

  For example, an audio signal encoding apparatus according to one aspect of the present invention includes a frame dividing unit that divides an audio input signal into processing unit frames for each channel, analyzes the audio input signal, determines a transform block length, and calculates auditory masking. And a filter that blocks a frame to be processed according to the conversion block length determined by the psychoacoustic operation unit and converts a time domain signal in the frame into a set of one or more frequency spectra. A bank unit, a scale factor calculation unit that divides the frequency spectrum output from the filter bank unit into a plurality of frequency bands, and weights the spectrum of each frequency band according to a calculation result of the auditory psychological calculation unit; and the scale Whole spectrum before quantization weighted by the factor calculator Quantization that determines the quantization step of the entire frame before spectral quantization by subtracting the information amount of the entire spectrum after quantization from the auditory information amount and integrating the coefficient obtained from the step size of the quantization roughness A step determination unit, a spectrum quantization unit that quantizes the frequency spectrum sequence using the scale factor and the quantization step, and a quantized spectrum output from the spectrum quantization unit according to a prescribed format A quantization unit that predicts the information amount of the entire quantized spectrum based on a bit amount allocated to a frame to be encoded. A spectral information amount prediction unit is included.

  In the present invention, after calculating and determining the scale factor for the first time, the quantization step can be calculated almost accurately by the calculation using the value, so that the quantization is performed by almost one spectral quantization and bit calculation. Can be terminated.

  ADVANTAGE OF THE INVENTION According to this invention, the processing amount of the quantization process in the encoding of an audio signal can be reduced significantly.

  DESCRIPTION OF EMBODIMENTS Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the drawings.

(First embodiment)
FIG. 1 is a diagram illustrating a configuration example of an audio signal encoding device according to the present embodiment. In the figure, a thick line indicates a data signal, and a thin line indicates a control signal.

  In the illustrated configuration, reference numeral 1 denotes a frame divider that divides an audio input signal into frames as processing units. Here, the audio input signal divided into frame units is sent to an auditory psychological calculator 2 and a filter bank 3 described later.

  An auditory psychological calculator 2 analyzes an audio input signal for each frame and performs a masking calculation in a divided frequency band that is more detailed than SFB. As a result of this calculation, the block type is output to the filter bank 3, and the signal-to-mask ratio (SMR) for each SFB is output to the scale factor calculator 4.

  Reference numeral 3 denotes a filter bank, which performs windowing of the block type specified by the auditory psychological calculator 2 on the time signal input from the frame divider 1 and then performs time-frequency conversion with the specified block length. And convert to frequency spectrum.

  4 is a scale factor calculator, which calculates the allowable error energy for each SFB from the SMR (signal-to-mask ratio) for each SFB and the frequency spectrum, and determines the scale factors of all the SFBs based on it.

  A spectrum allocation bit calculator 5 calculates the number of bits allocated to the quantized spectrum code.

  Reference numeral 6 denotes a quantized spectral total amount predictor that predicts the quantized spectral total amount based on the number of spectrum allocation bits.

  Reference numeral 7 denotes a quantization step calculator that calculates the amount of auditory information held by the spectrum before quantization and subtracts the amount of spectrum information after quantization obtained from the total amount of spectrum after quantization.

  A spectrum quantizer 8 quantizes each frequency spectrum.

  Reference numeral 9 denotes a bit shaper that creates a bit stream by appropriately shaping the scale factor and the quantized spectrum into a prescribed format and outputs the bit stream.

  An audio signal processing operation in the audio signal encoding apparatus having the above configuration will be described below.

  In this embodiment, for convenience of explanation, MPEG-2 AAC is described as an example of an encoding method. However, other encoding methods to which a similar quantization method can be applied are realized in exactly the same manner. Is possible.

  First, prior to processing, each unit is initialized. Initialization sets the quantization step and all scale factor values to zero.

  An audio input signal such as an audio PCM signal is divided into frame units by a frame divider 1 and sent to an auditory psychological calculator 2 and a filter bank 3. In the case of the MPEG-2 AAC LC (Low-Complexity) profile, one frame is composed of 1024 sample PCM signals, and this signal is transmitted.

  The psychoacoustic computing unit 2 appropriately analyzes the input signal sent from the frame divider 1, performs auditory masking analysis, scales the block type to filter bank 3, and the signal-to-mask ratio (SMR) for each SFB. Each is output to the factor calculator 4. In addition, since analysis and masking calculation performed by the auditory psychological calculator 2 are known in the art, detailed description thereof will not be given.

  In the filter bank 3, the current input signal for one frame sent from the frame divider 1 and the input signal of the previous frame received at the previous conversion are combined into two frames according to the block type output by the psychoacoustic operator 2. Min, 2048 sample time signals are converted to frequency components. In the present embodiment, the input signal of the previous frame is held in a buffer in the filter bank 3. Here, if the block type uses a long block length, 2048 samples of the input signal are used as one block, windowing according to the block type is performed, MDCT is performed, and 1024 frequency spectra are output. . When a short block length is used, MDCT is performed after windowing 256 samples starting from the 448th sample of the 2048 samples of the input signal, and a conversion that outputs 128 frequency components is input. The signal is shifted eight times while shifting by 128 samples to obtain eight sets of frequency spectra.

  The scale factor calculator 4 calculates the allowable error energy for each SFB from the spectrum component output from the filter bank 3 and the SMR value for each SFB output from the psychoacoustic operator 2, and based on this, the scale factor for each SFB is calculated. Calculate the factor. Since the calculation method of the scale factor based on the allowable error energy is known in the art, the details are not described here. For example, if the method described in Non-Patent Document 1 described above is used, MPEG-2 AAC , The scale factor scalefac [b] in SFB b can be obtained by the following equation.

In equation (2), x _avg is the average level of the spectral components included in SFB b. Xmin [b] is the allowable error energy of SFB b. If the spectral energy of SFB b is energy [b], the signal-to-mask ratio is SMR [b], and the number of included spectra is sfb_width [b], xmin [b] is obtained by the following equation.

  The spectrum allocation bit calculator 5 calculates the number of bits when the scale factor output from the scale factor calculator 4 is Huffman encoded, and subtracts it from the specified number of frame bits, thereby assigning bits to the quantized spectrum. The number is calculated and output to the quantized spectral total amount predictor 6.

  The quantized spectrum total amount predictor 6 performs a prediction calculation of the quantized spectrum total amount based on the number of bits output from the spectrum allocation bit calculator 5. In this embodiment, this calculation is performed using an approximate expression created based on the actual measurement of the relationship between the number of spectrum allocation bits and the total amount of quantized spectrum when quantized by a conventional quantizer. To do. For example, if this approximate expression is F (x) and the spectrum allocation bits are spectrum_bits, the quantized spectrum prediction total amount can be obtained by the following expression.

In the quantization step calculator 7, first, the sum of values obtained by weighting each frequency spectrum output from the filter bank 3 by the scale factor is taken, and based on this, the auditory frequency spectrum before quantization has Calculate the amount of information.
Next, the information amount of the quantized spectrum is calculated based on the quantized spectrum total amount output from the quantized spectrum total amount predictor 6.

  Finally, the quantization step, which is the quantization roughness of the entire frame, is obtained by subtracting the information amount of the quantization spectrum from the auditory information amount of the pre-quantization spectrum and multiplying by the coefficient obtained from the step size of the quantization roughness. calculate.

  Specifically, in the case of MPEG-2 AAC, the predicted value of the quantization step is obtained by calculating the following equation.

が、量子化前のスペクトル全体が持つ聴覚情報量であり、各スペクトルに、スケールファクタによって聴覚上の重み付けがなされた値の総計である。また、右辺の第２項である log₂Σ_iX_q が、量子化後のスペクトルが持つ情報量であり、このうち、Σ_iX_q は量子化スペクトルの総計であり、量子化スペクトル総量予測器６によって予測された値である。この値は前述したように例えば近似式（４）を計算することによって得られる。 In Equation (5), Xq is a quantized spectrum, xi is a spectrum before quantization, global_gain is a global gain (quantization step), and scalefac is an SFB scale factor including this spectrum component. In addition, the range of i that takes the total is one frame, that is, 0 ≦ i ≦ 1023.
Here, in the expression (5), the first term on the right side shown below.

Is the amount of auditory information that the entire spectrum before quantization has, and is the sum of the values of each spectrum weighted auditorily by a scale factor. In addition, log ₂ Σ _i X _q which is the second term on the right side is the information amount of the spectrum after quantization, and among these, Σ _i X _q is the total of the quantized spectrum, and the quantized spectrum total amount prediction The value predicted by the device 6. This value is obtained, for example, by calculating the approximate expression (4) as described above.

  Equation (5) can be obtained by appropriately modifying the spectral quantization equation (1).

  The spectrum quantizer 8 quantizes 1024 frequency spectra according to the scale factor output from the scale factor calculator 4 and the quantization step output from the quantization step calculator 7. Specifically, for example, in the case of MPEG-2 AAC, a quantized spectrum is calculated according to Equation (1), and the number of bits consumed in the entire frame is counted.

  If the number of used bits exceeds the number of spectrum allocated bits, the quantization step is increased until the number of used bits falls within the number of spectrum allocated bits, and spectrum quantization is performed again. However, since the calculation of the quantization step calculator 7 is accurate, in most cases, only one quantization spectrum calculation and bit calculation are performed.

  The scale factor and quantized spectrum of each SFB are shaped into a bit stream according to the format determined by the bit shaper 8 and output.

  As described above, the audio signal encoding apparatus according to the present embodiment predicts the total amount of spectrum after quantization from the amount of bits allocated to the frame, and uses this information amount of information in the entire spectrum before and after quantization. By calculating the difference of, the quantization step is predicted almost accurately before spectral quantization. Thereby, since the repetition for adjusting the quantization step is reduced, the quantization process can be completed quickly.

(Second Embodiment)
The present invention can also be implemented as a software program that operates on a general-purpose computer such as a personal computer (PC). Hereinafter, this case will be described with reference to the drawings.

  FIG. 5 is a diagram illustrating a configuration example of the audio signal encoding device according to the present embodiment.

  In the configuration shown in the figure, reference numeral 100 denotes a CPU, which performs operations for audio signal encoding processing, logic determination, and the like, and controls each component via a bus 102.

  A memory 101 stores a basic I / O program in the configuration example of the present embodiment, a program code being executed, data necessary for program processing, and the like.

  Reference numeral 102 denotes a bus, which transfers an address signal indicating a component to be controlled by the CPU 100, transfers a control signal of each component to be controlled by the CPU 100, and transfers data between the components. Do.

  Reference numeral 103 denotes a terminal for instructing device activation, setting of various conditions and input signals, and encoding start.

  Reference numeral 104 denotes an external storage device that provides an external storage area for storing data, programs, and the like, and is realized by, for example, a hard disk device. Here, programs and data including the OS are stored, and the stored data and programs are called by the CPU 100 when necessary. As will be described later, an audio signal encoding processing program is also installed in the external storage device 104.

  Reference numeral 105 denotes a media drive. Programs, data, digital audio signals, and the like recorded on a recording medium (for example, a CD-ROM) are loaded into the audio signal encoding apparatus by being read by the media drive 105. In addition, various data and execution programs stored in the external storage unit 104 can be written in a recording medium.

  A microphone 106 collects actual sound and converts it into an audio signal. Reference numeral 107 denotes a speaker, which can output arbitrary audio signal data as an actual sound.

  A communication network 108 includes a LAN, a public line, a wireless line, a broadcast wave, and the like. Reference numeral 109 denotes a communication interface, which is connected to the communication network 108. The audio signal encoding apparatus according to this embodiment can communicate with an external device via the communication network 108 via the communication interface 109 to transmit / receive data and programs.

  The audio signal encoding apparatus having such a configuration operates in response to various inputs from the terminal 103. When an input from the terminal 103 is supplied, an interrupt signal is sent to the CPU 100, whereby the CPU 100 reads various control signals stored in the memory 101, and various controls are performed according to the control signals. .

  In the audio signal encoding apparatus according to the present embodiment, the CPU 100 executes the basic I / O program stored in the memory 101, and loads the OS stored in the external storage device 104 to the memory 101. It works by executing Specifically, when the power of this apparatus is turned on, the OS is read from the external storage unit 104 into the memory 101 by the IPL (Initial Program Loading) function in the basic I / O program, and the operation of the OS is started. The

  The audio signal encoding processing program is a program code based on the flowchart of the audio signal encoding processing procedure shown in FIG.

  FIG. 6 is a diagram showing a content configuration example when an audio signal encoding processing program and related data are recorded on a recording medium. In the present embodiment, the audio signal encoding processing program and related data are recorded on a recording medium. As shown in the drawing, directory information of the recording medium is recorded in the head area of the recording medium, and thereafter, an audio signal encoding processing program that is the content of the recording medium and audio signal encoding processing related data are files. It is recorded as.

  FIG. 7 is a schematic diagram showing the introduction of the audio signal encoding processing program into the audio signal encoding device (PC). The audio signal encoding processing program and related data recorded on the recording medium can be loaded into the apparatus through the media drive 105 as shown in FIG. When the recording medium 110 is set in the media drive 105, the audio signal encoding processing program and related data are read from the recording medium 110 and stored in the external storage unit 104 under the control of the OS and the basic I / O program. Is done. After that, these information are loaded into the memory 101 at the time of restart and can be operated.

  FIG. 8 is a diagram showing a memory map in a state where the audio signal encoding processing program according to the present embodiment is loaded into the memory 101 and can be executed. As illustrated, the work area of the memory 101 includes, for example, a reference bit rate, a reference sampling rate, a bit rate, a sampling rate, an assigned bit upper limit value, an average assigned bit, a PE bit, a used bit, a scale factor bit, and a spectrum assigned bit. Pre-quantization spectrum auditory information amount, post-quantization spectrum prediction information amount, allowable error energy, spectrum buffer, quantization spectrum, input signal buffer, scale factor, quantization step, block type, SMR, PE, reserve bit amount Stored.

  FIG. 9 is a diagram illustrating a configuration example of the input signal buffer in the audio signal encoding device according to the present embodiment. In the configuration shown in the figure, the buffer size is 1024 × 3 samples, and for convenience of explanation, every 1024 samples are separated by vertical lines. The input signal is input 1024 samples for one frame from the right and sequentially processed from the left. The illustrated configuration schematically shows an input signal buffer for one channel. In this embodiment, similar buffers are prepared for the channels of the input signal.

  Hereinafter, an audio signal encoding process executed by the CPU 100 in the present embodiment will be described with reference to flowcharts.

  FIG. 2 is a flowchart of audio signal encoding processing in the present embodiment. A program corresponding to this flowchart is included in the audio signal encoding processing program, loaded into the memory 101 as described above, and executed by the CPU 100.

  First, step S <b> 1 is a process in which the user designates an input audio signal to be encoded using the terminal 103. In the present embodiment, the audio signal to be encoded may be an audio PCM file stored in the external storage 104, or a signal obtained by analog / digital conversion of a real-time audio signal captured by the microphone 106. When this process ends, the process proceeds to step S2.

  Step S2 is a process of determining whether or not the input audio signal to be encoded has been completed. If the input signal has ended, the process proceeds to step S11. If not completed, the process proceeds to step S3.

  In step S3, the input signal buffer shown in FIG. 9 shifts the time signal of two frames from the right, that is, 2048 samples to the left by one frame, and newly reads one frame, that is, 1024 samples to the right. This is signal shift processing. This process is performed for all channels included in the input signal. When the process is finished, the process proceeds to step S4.

  Step S4 is a process of analyzing the time signal stored in the input signal buffer and performing the psychoacoustic calculation of the current frame. As a result of this calculation, the block type of the current frame, auditory entropy (PE), and the SMR value for each SFB are calculated and stored in the work area on the memory 101. Here, as the SMR value, when the block length of the current frame is short, eight sets for the short block are calculated, or when the block type is other than that, one set for the long block is calculated. Such auditory computation is well known in the art and will not be described in detail. When the process is finished, the process proceeds to step S5.

  In step S5, according to the block type obtained in step S4, the time signal of the current frame, that is, the signal of 2048 samples (two frames) to the right from the current frame head pointer in FIG. Perform conversion. As a result, in the case of MPEG-2 AAC, when the transform block length is short, eight sets of spectra divided into 128 frequency components are obtained. In the case of a block type having a long block length, one set of spectra divided into 1024 frequency components is obtained. In both cases, a total of 1024 calculated spectra are stored in the spectrum buffer in the work area on the memory 101. When the process is finished, the process proceeds to step S6.

  Step S6 is a process of calculating the allowable error energy from the frequency spectrum obtained in step S5 and the SMR for each SFB obtained in step S4, and using this to calculate the scale factor for each SFB. For example, in the case of MPEG-2 AAC, the scale factor is calculated by the equation (2) in the first embodiment. The allowable error energy and scale factor for each SFB calculated in this process are stored in the work area on the memory 101. When the process is finished, step S7 follows.

  Step S7 is a process of calculating the quantization step from the difference between the auditory information amount of the spectrum before quantization and the information amount of the spectrum after quantization. Details of this processing will be described later with reference to FIG. When the process is finished, step S8 follows.

  In step S8, 1024 frequency spectra are quantized according to the scale factor obtained in step S6 and the quantization step obtained in step S7, and used bits are calculated. The used bits are stored in the work area on the memory 101. This is a process of increasing the quantization step and performing requantization only when the allocated bits are exceeded. Details of this processing will be described later with reference to FIG. When the process is finished, the process proceeds to step S9.

  Step S9 is a process of shaping the quantized spectrum and the scale factor calculated in step S8 according to a format determined by the encoding method, and outputting it as a bit stream. In the present embodiment, the bit stream output by this processing may be stored in the external storage device 104, or may be output to an external device connected to the communication network 108 via the communication interface 109. . When the process is finished, the process proceeds to step S10.

  Step S10 is a process for correcting the number of accumulated bits stored in the memory 101 from the bit amount used in the bit stream output in step S9 and the encoding bit rate. When the process is finished, the process returns to step S2.

  Step S11 is a process of shaping the quantized spectrum that has not yet been output due to the delay caused by the psychoacoustic calculation or orthogonal transformation, etc., into the bit stream and outputting it. When the process is finished, the audio signal encoding process is finished.

  FIG. 3 is a flowchart showing details of the quantization step prediction process in step S7 described above.

  First, step S101 is a process of calculating the number of bits to be used when the scale factor stored in the work area on the memory 101 is encoded according to the format defined by the encoding format. The calculated number of bits is stored in a work area on the memory 101. When the process is finished, step S102 follows.

  Step S102 is a process of calculating the number of bits allocated to the spectrum code by subtracting the number of scale factor bits stored in the memory 101 from the number of bits allocated to the frame. The calculated spectrum allocation bit number is stored in the work area on the memory 101. When the process is finished, step S 103 follows.

  Step S103 is a process of performing prediction calculation of the total amount of quantized spectrum using the number of spectrum allocation bits on the memory 101. This prediction calculation is performed by an approximate expression obtained by conducting an experiment in advance. For example, if this approximate expression is F (x) and the spectrum allocation bits are spectrum_bits, the quantized spectrum prediction total amount can be obtained by the following expression.

  The calculated quantized spectrum prediction total amount is stored in a work area on the memory 101. When the process is finished, step S 104 follows.

  Step S104 is a process of calculating the amount of auditory information that the spectrum before quantization has. The amount of auditory information of the spectrum before quantization is calculated by adding the amount of decrease in quantization roughness due to the scale factor of the SFB that contains the spectrum component to each spectrum component, obtaining the total amount for one frame, and calculating its logarithm It is required by doing. For example, in the case of MPEG-2 AAC, the amount of auditory information that the spectrum before quantization has can be obtained by calculating the following equation.

  The calculated auditory information amount of the pre-quantization spectrum is stored in the work area on the memory 101. When the process is finished, step S105 follows.

  Step S105 is a process of calculating the logarithm of the predicted total amount of the quantized spectrum obtained in step S103 and calculating the predicted information amount of the quantized spectrum. For example, MPEG-2 AAC can be calculated by calculating the following formula.

  That is, by calculating the logarithm of the total quantized spectrum obtained in step S103, the quantized spectrum prediction information amount can be obtained. The quantized spectral information amount calculated by this processing is stored in the work area on the memory 101. When the process is finished, step S 106 follows.

  In step S106, the quantized spectrum prediction information amount obtained in step S105 is subtracted from the auditory information amount of the pre-quantization spectrum obtained in step S104, and the result is determined by the step size of the quantization roughness in step S107. The global gain, that is, the predicted value of the quantization step is calculated. In the case of MPEG-2 AAC, this prediction value is the result of calculating Equation (5) as in the first embodiment.

  The calculated quantization step predicted value is stored in the work area on the memory 101 as a quantization step. When the process is finished, the quantization step prediction process is finished and the process returns.

  FIG. 4 is a flowchart detailing the above-described spectrum quantization processing in step S8.

  Step S201 is a process of quantizing 1024 spectral components stored in the spectral buffer according to the quantization step and scale factor stored on the memory 101. In the case of MPEG-2 AAC, the quantized spectrum is calculated according to the above equation (1). When the process is finished, step S 202 follows.

  Step S202 is a process of calculating the number of bits used when all the quantized spectra calculated in step S201 are encoded. For example, in the case of MPEG-2 AAC, since a plurality of quantized spectra are combined and Huffman encoded, the Huffman code table is searched in this process, and the total number of encoded bits is calculated. . The calculated number of used bits is stored in a work area on the memory 101. When the process is finished, step S203 follows.

  Step S203 is processing for comparing the size of the spectrum allocation bits on the memory 101 and the used bits. As a result of this comparison, if the used bit is larger than the allocated bit, the process proceeds to step S204, the quantization step stored in the memory 101 is increased to reduce the code amount, and the process returns to step S201 again. Although the spectrum is quantized, since an almost accurate quantization step is predicted by the above-described quantization step prediction process, step S204 is rarely actually executed.

  If the used bit is smaller than the allocated bit in the comparison in step S203, the spectrum quantization process is terminated and the process returns.

  As described above, in the audio signal encoding process according to the present embodiment, the information amount of the spectrum after quantization is predicted from the number of bits assigned to the spectrum code, and further, the amount of audio information before quantization is calculated. By taking the difference, it is possible to avoid the adjustment of the quantization step as much as possible by predicting the quantization step almost accurately before performing the actual quantization, greatly increasing the amount of processing required for the quantization process. Can be reduced.

(Third embodiment)
When encoding at a fixed bit rate, the technique of the present invention can be applied even when the accumulated bits accumulated in the bit reservoir are appropriately distributed to each frame according to the characteristics of the input signal. In the present embodiment, this case will be described with reference to the drawings.

  FIG. 10 is a diagram illustrating a configuration example of an audio signal encoding device according to the present embodiment. As in FIG. 1 according to the first embodiment, a thick line in the figure indicates a data flow, and a thin line indicates a control signal flow. In FIG. 10, the same numbers are assigned to components having the same functions as those in FIG.

  In the illustrated configuration, 1 is a frame divider, 2 is an auditory psychological calculator, 3 is a filter bank, 4 is a scale factor calculator, 7 is a quantization step calculator, 8 is a spectral quantizer, and 9 is a bit shaper. It is.

  Reference numeral 11 denotes a PE bit calculator that calculates PE bits, which are the predicted generation code amount of a frame, based on the auditory entropy (PE) of the frame.

  A spectrum allocation bit calculator 12 calculates the number of bits allocated to the spectrum code based on the bit rate, PE bits, accumulated bit amount, scale factor, and the like.

  Reference numeral 13 denotes a bit reservoir that sequentially manages the amount of accumulated bits defined according to the encoding method.

  Reference numeral 14 denotes a quantized spectrum total amount predictor that predicts a quantized spectrum total amount based on frame allocation bits or PE bits depending on conditions.

  Processing operations in the audio signal encoding apparatus having the above configuration will be described below. In this embodiment, MPEG-2 AAC will be described as an example of an encoding method for convenience of explanation, but other encoding methods that perform nonlinear quantization can be realized by the same method.

  First, prior to processing, each unit is initialized. Initialization sets the quantization step and all scale factors to zero.

  The audio input signal is divided into frame units by the frame divider 1 and output to the auditory psychological calculator 2 and the filter bank 3.

  The auditory psychological computing unit 2 appropriately performs auditory masking analysis on the input signal output from the frame divider 1 and outputs SMR and PE for each block type and SFB.

  The filter bank 3 outputs an input signal for two frames, which is a sum of one frame of the input signal output from the frame divider 1 and one frame of the previous frame held in the filter bank 3, from the psychoacoustic analyzer 2. Time frequency conversion is performed according to the block type, and the frequency spectrum is converted.

  The scale factor calculator 4 appropriately calculates the scale factor based on the frequency spectrum output from the filter bank 3 and the SMR value for each SFB output from the auditory psychological calculator 2 as in the first embodiment.

  The PE bit calculator 11 calculates a PE bit from the PE output from the psychoacoustic operator 3. That is, the auditory information amount of the input signal of the frame being processed is converted into the expected code amount when the auditory information is completely encoded. In the case of MPEG-2 AAC, the PE bit calculation formula described in the ISO standard is as follows.

ブロック長がショートのとき：

When the block length is long:

When the block length is short:

  In this embodiment, the PE bit is calculated according to the block length of the block type using this calculation formula as it is.

  The spectrum allocation bit calculator 12 first calculates the number of bits required to encode the scale factor output from the scale factor calculator 4, and then calculates the average bit per frame channel based on the bit rate. The average spectral allocation bit is calculated by obtaining the difference from the quantity.

  Next, this value is compared with the PE bit output from the PE bit calculator 11, and when the PE bit is large, the PE bit is assigned up to the maximum value determined by the accumulated bit amount accumulated in the bit reservoir 13. When the PE bit is small, the average spectrum allocation bit is allocated as it is.

  That is, in this embodiment, the spectrum allocation bits are specifically calculated by the following procedure.

1. Calculate the accumulated bit usage allowance from the accumulated bit amount.
When the block length is long: 10% of the accumulated bit amount,
When the block length is short: 25% of the accumulated bit amount,
Is an allowable use amount of accumulated bits. This is usable_bits.

2. If the average spectrum allocation bit amount is average_bits, the spectrum allocation bit amount and spectrum_bits are determined as follows.
When pe_bits> (average_bits + usable_bits)
spectrum_bits = average_bits + usable_bits;
When pe_bits <average_bits,
spectrum_bits = average_bits;
Otherwise, when average_bits ≦ pe_bits ≦ (average_bits + usable_bits)
spectrum_bits = pe_bits;

  Next, the spectrum allocation bit calculator 12 outputs the PE bit to the quantized spectrum total amount predictor 14 when the PE bit is smaller than the average spectrum allocation bit amount, and when the PE bit is equal to or greater than the average spectrum allocation bit. The spectrum allocation bits calculated in the above procedure are output to the quantized spectrum total amount predictor 14. At this time, bit selection information (hereinafter simply referred to as “selection information”), which is a flag indicating which number of bits has been output to the quantized spectrum total amount predictor 14, is simultaneously output.

  The quantized spectrum total amount predictor 14 predicts the quantized spectrum total amount based on the input selection information and the number of bits. This prediction calculation is performed by an approximate expression obtained by experiment as in the method shown in the first embodiment. In the quantized spectrum total amount predictor 14 in the present embodiment, this approximate expression is switched according to selection information. Perform predictive calculations. For example, assuming that an approximate expression of the total amount of quantized spectrum using spectrum allocation bits is F (x) and an approximate expression of the total amount of quantized spectrum using PE bits is G (x), the total predicted spectrum amount can be obtained by the following expression.

選択情報がPEビットの選択を示している場合：

If the selection information indicates the selection of spectrum allocation bits:

If the selection information indicates PE bit selection:

  In equation (10), bit_rate is the bit rate of the input signal being processed, sampling_rate is the sampling rate of the input signal being processed, base_bit_rate is the reference bit rate, and base_sampling_rate is the reference sampling rate. The reference bit rate and the reference sampling rate are the bit rate and the sampling rate of the input signal when the quantized spectral total amount prediction formula G (x) by PE bits is obtained by experiment, and in the audio signal encoding device in this embodiment This is a predetermined value.

  Here, the reason why the above-described quantized spectrum prediction method is used in the present embodiment will be described below.

  In the present embodiment, the spectrum allocation bit calculator 12 performs bit allocation based on the PE bit. Therefore, the spectrum allocation bits usually reflect the size of the PE bits, that is, the amount of generated code perceptually possessed by the input signal in the frame being processed. However, in the fixed bit rate control, when the size of the PE bit is lower than the average spectrum allocation bit, the average spectrum allocation bit is allocated as it is to the spectrum allocation bit. Therefore, in this case, since the generated code amount perceptually of the input signal is not reflected in the spectrum allocation bits, the prediction error increases when the total quantized spectrum amount is predicted using the spectrum allocation bits. Therefore, in this case, it is possible to predict a more accurate quantized spectrum total amount by predicting the quantized spectrum total amount using the PE bits.

  Further, since the spectrum allocation bits are calculated in consideration of restrictions on the bit rate and sampling rate, they have characteristics that follow changes in the bit rate and sampling rate. On the other hand, in the PE bit, although the value of the original PE itself changes due to a change in the sampling rate, the expressions (8) and (9) themselves do not change even if the bit rate or the sampling rate changes. Therefore, when performing prediction using PE bits, prediction is performed in consideration of the influence of the change rate from the reference bit rate or sampling rate, as shown in Equation (10).

  In this way, one approximate expression G (x) can be applied to any bit rate or sampling rate.

  Returning to the description of FIG. As in the first embodiment, the quantization step calculator 7 calculates the total amount of values obtained by weighting the frequency spectrum output from the filter bank 3 with the scale factor output from the scale factor calculator 4, and the logarithm thereof. To calculate the auditory information amount of the spectrum before quantization. Next, a logarithm of the quantized spectrum total amount predicted by the quantized spectrum total amount predictor 14 is calculated to calculate a spectral information amount after quantization. Further, the quantization step is calculated by taking this difference and multiplying by a coefficient determined by the step size of the quantization roughness. Specifically, the calculation of the above equation (5) is performed.

  As in the first embodiment, the spectrum quantizer 8 uses the scale factor output from the scale factor calculator 4 and the quantization step output from the quantization step calculator 7 to calculate the frequency spectrum output from the filter bank 3. The spectrum is quantized, the necessary number of bits is counted, and compared with the spectrum allocation bits output from the spectrum allocation bit calculator 12. Here, when the necessary number of bits exceeds the spectrum allocation bit, the quantization step is increased as appropriate, and the quantization is performed again. As described above, the predicted value of the quantization step by the quantization step calculator 7 Is almost accurate, so this requantization is rarely performed.

  The quantized spectrum, scale factor, and quantization step that are finally output by the spectrum quantizer 8 are entropy-encoded by the bit shaper 9 and then appropriately shaped into a bit stream format determined by the encoding method and output. .

  At this time, the bit reservoir 13 is notified of the number of bits actually used for the sign, the bit reservoir 13 calculates the difference from the frame bit, and adjusts the accumulated bit amount appropriately by adding or subtracting the increase / decrease to the accumulated bit amount. .

  As described above, even when the accumulated bits accumulated in the bit reservoir according to the input signal are appropriately assigned to the frame at the fixed bit rate as in the present embodiment, the total amount of the quantized spectrum accurately before the quantization. Therefore, it is possible to accurately determine a quantization step before quantization, and to efficiently perform quantization while avoiding repetition of spectrum quantization and bit calculation.

(Fourth embodiment)
The audio encoding device described in the third embodiment can also be implemented as a software program that runs on a general-purpose computer such as a PC. Hereinafter, this case will be described with reference to the drawings.

  The configuration of the audio signal encoding apparatus and the processing content of the audio signal encoding processing program in the present embodiment are generally the same as those in the second embodiment. Therefore, in the present embodiment, the second embodiment will be described. 5, 2, and 6 to 9 will be referred to, and detailed descriptions thereof will be omitted. The difference from the second embodiment is the content of the quantization step prediction process in step S7. Therefore, only the quantization step prediction process in step S7 will be described below.

  FIG. 11 is a flowchart showing details of the quantization step prediction process in step S7 in the present embodiment.

  First, step S301 is a process of calculating PE bits from the PE and block type on the memory 101 obtained by the psychoacoustic calculation process of step S4. Specifically, the PE bit is calculated by selecting the above formula (9) or formula (10) according to the block type as in the third embodiment. The calculated PE bit is stored in the work area on the memory 101. When the process is finished, step S302 follows.

  Step S302 is a process of calculating the number of bits to be used when the scale factor stored in the work area on the memory 101 is encoded in the format defined by the encoding method. The number of scale factor bits calculated by this processing is stored in the work area on the memory 101. When the process is finished, step S 303 follows.

  Step S303 is a process of subtracting the number of scale factor bits stored in the memory 101 from the average number of bits allocated to the frame to calculate the number of bits allocated to the spectrum code, that is, the average number of spectrum allocated bits (average allocated bits). It is. The calculated average number of allocated bits is stored in a work area on the memory 101. When the process is finished, step S304 follows.

  Step S304 is processing for comparing the average number of allocated bits on the memory 101 with the number of PE bits. As a result of the comparison, if the number of PE bits is larger, the process proceeds to step S305. Otherwise, the process proceeds to step S307.

  Step S305 is processing for calculating spectrum allocation bits from the PE bits on the memory 101, the average allocation bits, and the accumulated bit amount. Details of this processing will be described later with reference to FIG. When the process is finished, step S306 follows.

  Step S306 is a process of performing prediction calculation of the total amount of quantized spectrum using the number of spectrum allocation bits on the memory 101. This prediction calculation is performed by an approximate expression obtained by conducting an experiment in advance. For example, if this approximate expression is F (x) and the spectrum allocation bits are spectrum_bits, the quantized spectrum prediction total amount can be obtained by Expression (4) as in the second embodiment.

  The calculated quantized spectrum prediction total amount is stored in a work area on the memory 101. When the process is finished, step S309 follows.

  One step S307 is a process of storing the average allocation bits on the memory 101 in the memory 101 as spectrum allocation bits. That is, the value of the average allocation bit is copied to the spectrum allocation bit. When the process is finished, step S308 follows.

  Step S308 is a process of performing prediction calculation of the total amount of quantized spectrum using the number of PE bits on the memory 101. This prediction calculation is also performed by an approximate expression obtained by conducting an experiment in advance. When this approximate expression is G (x) and the PE bit is pe_bits, the quantized spectral prediction total amount can be obtained by Expression (10) as in the third embodiment.

  The calculated quantized spectrum prediction total amount is stored in a work area on the memory 101. When the process is finished, step S309 follows.

  Step S309 is processing for calculating the amount of auditory information that the spectrum before quantization has. As in the second embodiment, the amount of auditory information of the pre-quantization spectrum is obtained by multiplying each spectral component by a decrease in quantization roughness due to the scale factor of the SFB in which the spectral component is included. It is obtained by calculating the total amount and calculating its logarithm. For example, in the case of MPEG-2 AAC, the amount of auditory information that the spectrum before quantization has can be obtained by calculating the following equation.

  The calculated auditory information amount of the pre-quantization spectrum is stored in the work area on the memory 101. When the process is finished, step S 310 follows.

  Step S310 is a process of calculating the logarithm of the predicted total amount of the quantized spectrum obtained in step S306 or step S308 and calculating the predicted information amount of the quantized spectrum. For example, MPEG-2 AAC can be calculated by calculating the following formula.

  The quantized spectral prediction information amount calculated by this processing is stored in the work area on the memory 101. When the process is finished, step S311 follows.

  In step S311, the quantized spectrum prediction information amount obtained in step S310 is subtracted from the auditory information amount of the pre-quantization spectrum obtained in step S309, and the result is multiplied by a coefficient determined by the quantization roughness step size. The global gain, that is, the predicted value of the quantization step is calculated. In the case of MPEG-2 AAC, this prediction value is the result of calculating Equation (5) as in the first embodiment.

  The calculated quantization step predicted value is stored in the work area on the memory 101 as a quantization step. When the process is finished, the quantization step prediction process is finished and the process returns.

  FIG. 12 is a flowchart showing details of the spectrum allocation bit calculation processing in step S305 in the present embodiment.

In step S401, the number of accumulated bits that can be allocated to this frame is calculated according to the accumulated bit amount and block type in the memory 101, and this value is added to the average allocated bits to calculate the upper limit value of the spectrum allocated bits. It is processing.
In the present embodiment, the number of accumulated bits is determined in the following manner as in the third embodiment.

When the block length is long: 10% of the accumulated bit amount,
When the block length is short: 25% of the accumulated bit amount

  The spectrum allocation bit upper limit value is obtained by adding the value obtained in the above procedure to the average allocation bits on the memory 101.

  The spectrum allocation bit upper limit value obtained by this calculation is stored in the memory 101. When the process is finished, step S 402 follows.

  Step S402 is processing for comparing the PE bit on the memory 101 and the spectrum allocation bit upper limit value. As a result of the comparison, if the number of PE bits is smaller than the spectrum allocation bit upper limit value, the process proceeds to step S403. Otherwise, the process proceeds to step S404.

  Step S403 is processing for storing the PE bits on the memory 101 as spectrum allocation bits. That is, the value of the PE bit is copied to the spectrum allocation bit. When the process is finished, the spectrum allocation bit calculation process is finished and the process returns.

  Step S404 is processing for storing the spectrum allocation bit upper limit value on the memory 101 as a spectrum allocation bit. That is, the spectrum allocation bit upper limit value is copied to the spectrum allocation bit. When the process is finished, the spectrum allocation bit calculation process is finished and the process returns.

  In this process, as described above, setting the upper limit value for the number of bits allocated by the PE bits has an effect of preventing the bit reservoir from being exhausted due to the accumulated bits being depleted.

  As described above, according to the present embodiment, even when the accumulated bits accumulated in the bit reservoir according to the characteristics of the input signal are appropriately assigned to the frame at the fixed bit rate, the quantization is accurately performed before the quantization. By predicting the total amount of quantized spectrum, it is possible to accurately determine the quantization step before quantization, and it is possible to efficiently perform quantization while avoiding repetition of spectrum quantization and bit calculation.

  As described above, in the audio signal encoding processing according to the present invention, the difference in the information amount of the entire spectrum before and after quantization is predicted by predicting the total amount of spectrum after quantization from the amount of bits allocated to the frame. By calculating and predicting the quantization step of the entire frame almost accurately before spectrum quantization, the quantization process can be completed by performing only one spectrum quantization process. Thereby, while maintaining the encoding quality equivalent to that of the conventional technique, it is possible to significantly reduce the amount of processing required for the quantization process compared to the conventional technique.

(Other embodiments)
The present invention is not limited to the embodiment described above.

  In the above-described embodiment, the case of an encoding method in which block switching is performed is handled as an audio encoding device and method. However, the present invention can be similarly applied to an encoding method in which block switching is not performed.

  In addition, the present invention can be variously modified and implemented without departing from the scope of the invention.

  As mentioned above, although embodiment of this invention was explained in full detail, this invention may be applied to the system comprised from several apparatuses, and may be applied to the apparatus which consists of one apparatus.

  As described above, the present invention supplies an audio signal encoding processing program that implements the functions of the embodiments directly or remotely to a system or apparatus, and the computer of the system or apparatus supplies the supplied program code. It is also achieved by reading and executing. In that case, as long as it has the function of a program, the form does not need to be a program.

  Therefore, in order to realize the functional processing of the present invention with a computer, the program code itself installed in the computer and the storage medium storing the program also constitute the present invention. In other words, the claims of the present invention include the computer program itself for realizing the functional processing of the present invention and a storage medium storing the program.

  In this case, the program may be in any form as long as it has a program function, such as an object code, a program executed by an interpreter, or script data supplied to the OS.

  As a storage medium for supplying the program, for example, flexible disk, hard disk, optical disk, magneto-optical disk, MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory card, ROM, DVD (DVD-ROM, DVD-R).

  As another program supply method, a client computer browser is used to connect to an Internet homepage, and the computer program of the present invention itself or a compressed file including an automatic installation function is downloaded from the homepage to a storage medium such as a hard disk. Can also be supplied. It can also be realized by dividing the program code constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. That is, a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer is also included in the scope of the present invention.

  In addition, the program of the present invention is encrypted, stored in a storage medium such as a CD-ROM, distributed to users, and key information for decryption is downloaded from a homepage via the Internet to users who have cleared predetermined conditions. It is also possible to execute the encrypted program by using the key information and install the program on a computer.

  In addition to the functions of the above-described embodiments being realized by the computer executing the read program, the OS running on the computer based on the instruction of the program is a part of the actual processing. Alternatively, the functions of the above-described embodiment can be realized by performing all of them and performing the processing.

  Furthermore, after the program read from the storage medium is written to a memory provided in a function expansion board inserted into the computer or a function expansion unit connected to the computer, the function expansion board or The CPU or the like provided in the function expansion unit performs part or all of the actual processing, and the functions of the above-described embodiments are realized by the processing.

FIG. 1 is a diagram illustrating a configuration example of an audio signal encoding apparatus according to the first embodiment of the present invention. FIG. 2 is a flowchart of audio signal encoding processing according to the second embodiment of the present invention. FIG. 3 is a flowchart of the quantization step prediction process in the second embodiment of the present invention. FIG. 4 is a flowchart of spectrum quantization processing according to the second embodiment of the present invention. FIG. 5 is a diagram illustrating a configuration example of an audio signal encoding device according to the second embodiment of the present invention. FIG. 6 is a diagram showing a content configuration example of a storage medium storing an audio signal encoding processing program according to the second embodiment of the present invention. FIG. 7 is a schematic diagram showing introduction of an audio signal encoding processing program into a PC according to the second embodiment of the present invention. FIG. 8 is a diagram showing an example of a memory map in the second embodiment of the present invention. FIG. 9 is a diagram illustrating a configuration example of an input signal buffer according to the second embodiment of the present invention. FIG. 10 is a diagram illustrating a configuration example of an audio signal encoding device according to the third embodiment of the present invention. FIG. 11 is a flowchart of the quantization step prediction process in the fourth embodiment of the present invention. FIG. 12 is a flowchart of the spectrum allocation bit calculation process in the fourth embodiment of the present invention. FIG. 13 is a flowchart of the quantization process according to the ISO standard.

Claims (10)

A frame dividing unit that divides the audio input signal into processing unit frames for each channel;
An auditory psychological operation unit that analyzes an audio input signal, determines a conversion block length, and calculates auditory masking;
A filter bank unit that blocks a processing target frame according to the conversion block length determined by the auditory psychological calculation unit, and converts a time domain signal in the frame into a set of one or more frequency spectra;
A scale factor calculation unit that divides the frequency spectrum output from the filter bank unit into a plurality of frequency bands, and weights the spectrum of each frequency band according to a calculation result of the auditory psychology calculation unit;
By subtracting the information amount of the whole spectrum after quantization from the auditory information amount of the whole spectrum before quantization weighted by the scale factor calculation unit, by multiplying the coefficient obtained from the step size of the quantization roughness, A quantization step determining unit that determines the quantization step of the entire frame before spectral quantization;
A spectrum quantization unit that quantizes the frequency spectrum sequence using the scale factor and the quantization step;
A bit shaping unit that creates and outputs a bitstream obtained by shaping the quantized spectrum output from the spectrum quantization unit according to a prescribed format;
With
The audio signal characterized in that the quantization step determining unit includes a quantized spectrum information amount predicting unit that predicts an information amount of the entire quantized spectrum based on an amount of bits allocated to a frame to be encoded. Encoding device. A frame dividing unit that divides the audio input signal into processing unit frames for each channel;
An auditory psychological operation unit that analyzes an audio input signal, determines a conversion block length, and calculates auditory masking;
A filter bank unit that blocks a processing target frame according to the conversion block length determined by the psychoacoustic operation unit, and converts a time domain signal in the frame into a set of one or more frequency spectra;
A scale factor calculation unit that divides the frequency spectrum output from the filter bank unit into a plurality of frequency bands, and weights the spectrum of each frequency band according to a calculation result of the auditory psychology calculation unit;
A quantized spectrum information amount prediction unit that predicts before quantizing the information amount of the entire quantized spectrum based on the bit amount allocated to the frame to be encoded;
By subtracting the information amount of the whole spectrum after quantization from the auditory information amount of the whole spectrum before quantization weighted by the scale factor calculation unit, and accumulating the coefficient obtained from the step size of the quantization roughness A quantization step determination unit that determines the quantization step of the entire frame before spectral quantization;
A spectrum quantization unit that quantizes the frequency spectrum sequence using the scale factor and the quantization step;
A bit shaping unit that creates and outputs a bitstream obtained by shaping the quantized spectrum output from the spectrum quantization unit according to a prescribed format;
With
The quantized spectral information amount prediction unit predicts the quantized spectral information amount based on auditory entropy when the prediction code amount of an input signal is less than an average frame allocation bit during fixed bit rate encoding. An audio signal encoding device. A frame dividing unit that divides the audio input signal into processing unit frames for each channel;
An auditory psychological operation unit that analyzes an audio input signal, determines a conversion block length, and calculates auditory masking;
A filter bank unit that blocks a processing target frame according to the conversion block length determined by the psychoacoustic operation unit, and converts a time domain signal in the frame into a set of one or more frequency spectra;
A scale factor calculation unit that divides the frequency spectrum output from the filter bank unit into a plurality of frequency bands, and weights the spectrum of each frequency band according to a calculation result of the auditory psychology calculation unit;
A quantized spectrum information amount prediction unit that predicts before quantizing the information amount of the entire quantized spectrum based on the bit amount allocated to the frame to be encoded;
By subtracting the information amount of the whole spectrum after quantization from the auditory information amount of the whole spectrum before quantization weighted by the scale factor calculation unit, and accumulating the coefficient obtained from the step size of the quantization roughness A quantization step determination unit that determines the quantization step of the entire frame before spectral quantization;
A spectrum quantization unit that quantizes the frequency spectrum sequence using the scale factor and the quantization step;
A bit shaping unit that creates and outputs a bitstream obtained by shaping the quantized spectrum output from the spectrum quantization unit according to a prescribed format;
With
The spectrum quantization unit adjusts the quantization step and re-quantizes the spectrum when the code amount used for the quantized spectrum exceeds the assigned code amount. Encoding device.   The audio signal encoding apparatus according to any one of claims 1 to 3, wherein the encoding format is MPEG-1 Audio Layer III.   The audio signal encoding apparatus according to any one of claims 1 to 3, wherein the encoding format is MPEG-2 / 4 AAC. A frame dividing step for dividing the audio input signal into processing unit frames for each channel;
Auditory psychological calculation step that analyzes audio input signal, determines transform block length and calculates auditory masking,
A filter bank processing step that blocks a processing target frame according to the conversion block length determined in the auditory psychological calculation step, and converts a time domain signal in the frame into a set of one or more frequency spectra;
Dividing the frequency spectrum obtained in the filter bank processing step into a plurality of frequency bands, and weighting the spectrum of each frequency band according to the calculation result in the auditory psychological calculation step; and
By subtracting the information amount of the whole spectrum after quantization from the information amount of the whole spectrum before quantization weighted by the scale factor calculation step, and integrating the coefficients obtained from the step size of the quantization roughness, A quantization step determination step for determining the entire quantization step before spectral quantization;
A spectral quantization step of quantizing the frequency spectrum sequence using the scale factor and the quantization step;
A bit shaping step of creating and outputting a bitstream obtained by shaping the quantized spectrum obtained in the spectrum quantization step according to a prescribed format;
Have
The quantization step determining step includes a quantized spectrum total amount predicting step for predicting an information amount of the entire quantized spectrum based on an information amount assigned to a frame to be encoded. Method. A frame dividing step for dividing the audio input signal into processing unit frames for each channel;
Auditory psychological calculation step that analyzes audio input signal, determines transform block length and calculates auditory masking,
A filter bank processing step that blocks a processing target frame according to the conversion block length determined in the auditory psychological calculation step, and converts a time domain signal in the frame into a set of one or more frequency spectra;
Dividing the frequency spectrum obtained in the filter bank processing step into a plurality of frequency bands, and weighting the spectrum of each frequency band according to the calculation result in the auditory psychological calculation step; and
A quantized spectrum information amount prediction step for predicting before quantizing the information amount of the entire quantized spectrum based on the bit amount assigned to the frame to be encoded;
By subtracting the information amount of the entire spectrum after quantization from the auditory information amount of the entire spectrum before quantization weighted by the scale factor calculation step, and integrating the coefficient obtained from the step size of the quantization roughness A quantization step determining step for determining the quantization step for the entire frame before spectral quantization;
A spectral quantization step of quantizing the frequency spectrum sequence using the scale factor and the quantization step;
A bit shaping step of creating and outputting a bitstream obtained by shaping the quantized spectrum obtained by the spectrum quantization according to a prescribed format;
Have
In the quantized spectral information amount predicting step, when the predicted code amount of the input signal is less than the average frame allocation bit at the time of fixed bit rate encoding, the quantized spectral information amount is predicted based on auditory entropy. An audio signal encoding method. A frame dividing step for dividing the audio input signal into processing unit frames for each channel;
Auditory psychological calculation step that analyzes audio input signal, determines transform block length and performs auditory masking calculation,
A filter bank processing step that blocks a processing target frame according to the conversion block length determined in the auditory psychological calculation step, and converts a time domain signal in the frame into a set of one or more frequency spectra;
Dividing the frequency spectrum obtained in the filter bank processing step into a plurality of frequency bands, and weighting the spectrum of each frequency band according to the calculation result in the auditory psychological calculation step; and
A quantized spectrum information amount prediction step for predicting before quantizing the information amount of the entire quantized spectrum based on the bit amount assigned to the frame to be encoded;
By subtracting the information amount of the entire spectrum after quantization from the auditory information amount of the entire spectrum before quantization weighted by the scale factor calculation step, and integrating the coefficient obtained from the step size of the quantization roughness A quantization step determining step for determining the quantization step for the entire frame before spectral quantization;
A spectral quantization step of quantizing the frequency spectrum sequence using the scale factor and the quantization step;
A bit shaping step of creating and outputting a bitstream obtained by shaping the quantized spectrum obtained in the spectrum quantization step according to a prescribed format;
Have
In the spectrum quantization step, when the code amount used for the quantized spectrum exceeds the assigned code amount, the audio signal is characterized in that the quantization step is adjusted to re-quantize the spectrum. Encoding method.   A program for causing a computer to execute the audio signal encoding method according to any one of claims 6 to 8.   A computer-readable storage medium storing the program according to claim 9.

Images