JP4626261B2 - Speech coding apparatus and speech coding method - Google Patents

Description

  The present invention relates to a speech coding apparatus and a speech coding method.

Conventionally, audio signal compression methods include μ-law, ADPCM (Adaptive Differential Pulse Code Modulation), MP3 (MPEG Audio Layer-3) used in music, VSELP (Vector Sum Excited Linear) used in mobile phones, etc. Prediction), G. A CELP (Code-Excited Linear Prediction) type compression method such as 729 has been put into practical use. Patent Document 1 discloses a technique using vector quantization as an audio compression technique.
Japanese Patent Laid-Open No. 10-63299

  When recording conversation or the like in language learning, a sampling frequency of about 16 kHz is considered to be an appropriate frequency that does not require a large amount of data while maintaining the characteristics of each language. However, there is a problem that the compression noise that appears in the CELP compression method is not appropriate for language learning. In addition, although μ-law and ADPCM have sufficient sound quality, since the encoding rate is high, there is a problem that recording time is shortened when these compression methods are used in a portable device. MP3 is mainly intended for compression of high-quality audio, and there is a problem that compression cannot be performed effectively at a sampling frequency of about 16 kHz.

  An object of the present invention is to realize effective speech coding by improving coding efficiency in speech coding.

The speech coding apparatus according to the present invention includes a framing unit that divides an input speech signal into frames, and a frequency that performs frequency transformation on the speech signal for each block unit constituting one frame and calculates a transform coefficient. A conversion unit, a division unit that divides a conversion coefficient of another block in the same frame by a conversion coefficient of one block specified in advance among the conversion coefficients obtained by the frequency conversion unit, and the division unit A vector quantization unit that vector-quantizes the obtained divided coefficients of the transform coefficients for each block for each frequency band, and encodes index data obtained by the vector quantization unit using a predetermined encoding method. And an encoding unit for converting to an encoding unit.
Another speech encoding apparatus according to the present invention includes a framing unit that divides an input speech signal into frames, and performs a transform on the speech signal for each block unit constituting one frame to obtain a transform coefficient. Based on the frequency conversion unit to be calculated and the conversion coefficient obtained by the frequency conversion unit, the sum of absolute values of the conversion coefficients is calculated for each block unit constituting the frame, and the sum of the absolute values is the largest. Divide the transform coefficient of the other block in the same frame by the transform coefficient of the block, and vector quantization of the divided values of the transform coefficients of each block obtained by the divider for each frequency band A vector quantization unit that encodes the index data obtained by the vector quantization unit using a predetermined encoding method.
Furthermore, another speech encoding apparatus according to the present invention includes a framing unit that divides an input speech signal into frames, and performs a frequency conversion on the speech signal for each block unit constituting one frame to obtain a transform coefficient. A frequency conversion unit to be calculated and a block having a value having the maximum absolute value among the conversion coefficients calculated for each block unit constituting the frame are obtained, and the conversion coefficient of the block is used to determine other blocks in the same frame. A division unit that divides the transform coefficient, a vector quantization unit that collectively vector-quantizes the divided values of the transform coefficient of each block obtained by the division unit for each frequency band, and the vector quantization unit. And an encoding unit that encodes the index data obtained by a predetermined encoding method.

According to the present invention, among the transform coefficients obtained by frequency transform, the transform coefficient of another block in the same frame is divided by the transform coefficient of a predetermined one block in the frame. The range of the absolute value of the subsequent transform coefficient is narrowed, the efficiency of vector quantization, entropy coding, etc. is improved, and the sound quality can be improved.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
First, the configuration in the present embodiment will be described.

  FIG. 1 shows a configuration of speech encoding apparatus 100 according to an embodiment of the present invention. As shown in FIG. 1, the speech encoding apparatus 100 includes an A / D converter 1, a DC (Direct Current) removing unit 2, a framing unit 3, a level adjusting unit 4, a frequency converting unit 5, a dividing unit 6, a frequency The rearrangement unit 7, the vector quantization unit 8, the entropy encoding unit 9, the rate controller 10, and the data deletion unit 11 are configured.

  The A / D conversion unit 1 converts the input audio analog signal into a digital signal and outputs the digital signal to the DC removal unit 2. The sampling frequency is preferably about 16 kHz, but may be 11.025 kHz, 22.05 kHz, or the like.

The DC removal unit 2 removes the direct current component of the audio signal input from the A / D conversion unit 1 and outputs it to the framing unit 3. The reason why the DC component of the audio signal is removed is that the DC component is almost irrelevant to the sound quality. The removal of the direct current component can be realized by, for example, a high-pass filter. An example of the high-pass filter is represented by Expression (1).

  The framing unit 3 divides the signal input from the DC removal unit 2 into frames that are processing units of encoding (compression) and outputs the frames to the level adjustment unit 4. Here, one frame has a length including one or more, preferably four or more blocks. One block is a unit for performing one MDCT (Modified Discrete Cosine Transform), and has a length corresponding to the order of MDCT. Hereinafter, each block constituting one frame is referred to as an MDCT block. FIG. 2 shows the relationship between the input signal and each frame, and FIG. 3 shows the relationship between one frame and each MDCT block. As shown in FIG. 3, each MDCT block has an overlap portion that is half the length of the previous MDCT block and the MDCT block. In addition, as shown in FIG. 2, each frame has an overlapping portion that is half the length of the previous frame and the MDCT block.

音声信号では、１０ビット程度に制圧することが考えられる。なお、音声再生時には、振幅が制圧目標ビット以下に制圧された信号を元に戻す必要があるため、shift_bitを表す信号が、音声符号化信号の一部として出力される。 The level adjustment unit 4 adjusts the level of the input audio signal for each frame, and outputs the level-adjusted signal to the frequency conversion unit 5. Level adjustment is to make the maximum value of the amplitude of a signal included in one frame fall within a specified number of bits (hereinafter referred to as suppression target bits). For example, if the maximum amplitude of a signal in one frame is nbit and the suppression target bit number is N, all the signals in the frame are LSB (Least Significant Bit: least significant) for the number of shift_bits that satisfy Expression (2). This can be realized by shifting to the bit) side.

It can be considered that the audio signal is suppressed to about 10 bits. Note that, during audio reproduction, a signal whose amplitude is suppressed to a suppression target bit or less needs to be restored, so a signal representing shift_bit is output as part of the encoded audio signal.

ここで、ｈ_nは窓関数であり、式（４）で表される。

なお、ブロック長Ｍは、16kHz程度のサンプリング周波数の音声では、２５６程度の値が考えられる。 The frequency conversion unit 5 performs frequency conversion on the signal input from the level adjustment unit 4 and outputs the result to the division unit 6. In the present embodiment, a case where MDCT (Modified Discrete Cosine Transform) is used as frequency conversion is shown. When the length of the MDCT block is M and the input signal is {x _n | n = 0,..., M−1}, the MDCT coefficient {X _k | k = 0,. ).

Here, h _n is a window function, and is represented by Expression (4).

Note that the block length M may be a value of about 256 for audio having a sampling frequency of about 16 kHz.

The division unit 6 performs a process of dividing the MDCT coefficient of another MDCT block by the MDCT coefficient of the MDCT block designated in advance among the MDCT blocks for one frame obtained by the frequency conversion. For example, there are four MDCT blocks M _{0 to} M ₃ in one frame, and each MDCT block has a sixth-order MDCT coefficient. The MDCT coefficient of the _i -th MDCT block is expressed as M _i = {m _in | n = 0,..., 5}, and the MDCT block designated in advance as a division value in the division processing is M ₀ . As a result of dividing the MDCT coefficients of other MDCT blocks M _{1 to} M ₃ in the same frame by the MDCT coefficient of M ₀ , M ′ _i = {m ′ _in | n = 0,..., 5} It becomes like this.
_{m 'in = m in / m} 0n i = 1,2,3, n = 0, ..., 5 (5)
If the absolute value of the division value m _0n is less than 1, the value of the MDCT coefficient becomes larger due to the division, so that the absolute value of the division value m _0n is 1 or more (| m _0n |> 1) Only perform division processing.

For example, when the MDCT coefficients of the MDCT blocks M _{0 to} M ₃ are expressed by the following formula (6), M ′ _i (i = 1, 2, 3) after the division processing is expressed by the following formula (7).

The frequency rearrangement unit 7 rearranges the MDCT coefficients obtained by the division process of the division unit 6 for each frequency, collectively vectorizes the coefficients in the same frequency band, and outputs them to the vector quantization unit 8. As described above, when signals in the same frequency band are collectively vectorized, for example, when many stationary signals are included, the accuracy of subsequent vector quantization is improved. When there are m MDCT blocks in one frame and M / 2 MDCT coefficients are calculated in each MDCT, the j-th frequency band is summarized assuming that the j-th MDCT coefficient of the i-th MDCT block is X _ij . The vector F _j is F _j = {X _ij | i = 0,..., M−1}, j = 0,.

The vector quantization unit 8 has a VQ (Vector Quantization) table that stores representative vectors representing a plurality of speech patterns. The vector F _j created by the frequency rearrangement unit 7 and each representative stored in the VQ table. The vectors are compared, and the index indicated by the most similar representative vector is output as a code to the entropy encoding unit 9.

代表ベクトルの数ｋと符号化対象ベクトルのベクトル長Ｎは、ベクトル量子化に要する処理時間やＶＱテーブルの容量等を勘案して決定される。例えば、ベクトル長を２にして代表ベクトル数を２５６にしたり、ベクトル長を４にして代表ベクトル数を８１９２（＝２¹³）にしたりするなど、自由な組み合わせが考えられる。 For example, {s _j | j = 1,..., N} is an encoding target vector having a vector length N, and k representative vectors stored in the VQ table are {V _i | i = 1,. If V _i = {v _ij | j = 1,..., N}, the error e _i between the encoding target vector and each element v _ij of the i-th representative vector stored in the VQ table is minimized. I (index) is an output code. The equation for calculating the error e _i shown in equation (8).

The number k of representative vectors and the vector length N of the encoding target vector are determined in consideration of the processing time required for vector quantization, the capacity of the VQ table, and the like. For example, a free combination is conceivable, for example, the vector length is 2 and the number of representative vectors is 256, or the vector length is 4 and the number of representative vectors is 8192 (= 2 ¹³ ).

Since audio often has different characteristics in the high frequency part and the low frequency part, in this embodiment, different VQ tables are used in the high frequency and the low frequency. The VQ table storing the high frequency representative vector is referred to as the high frequency VQ table 8a, and the VQ table storing the low frequency representative vector is referred to as the low frequency VQ table 8b. In the vector F _j = {X _ij | i = 0,..., M−1}, j = 0,..., M / 2-1 created by the frequency rearrangement unit 7, the boundary between the high frequency and the low frequency is What is necessary is just to divide j which shows a frequency band into half simply. _{That, F 0, F 1, ...} , low the _{F M / 4-1, F M /} 4, F M / 4 + 1, ..., a F _{M / 2-1} may be set to the high band. Therefore, the low-frequency vectors F ₀ , F ₁ ,..., F _{M / 4-1} are compared with the representative vectors stored in the low-frequency VQ table 8b, and the index indicated by the most similar representative vector is used as a code. Is output. Similarly, the high-frequency vectors F _{M / 4} , F _{M / 4 + 1} ,..., F _{M / 2-1} are compared with the representative vectors stored in the high-frequency VQ table 8a, and the most similar representatives are compared. The index indicated by the vector is output as a code.

  The entropy encoding unit 9 performs entropy encoding on the code input from the vector quantization unit 8 and outputs the result to the rate controller 10. Entropy coding is a coding method that uses the statistical properties of a signal to convert a code into a shorter code. Huffman coding, arithmetic coding, and code by a range coder (Range Coder) There is. Details of the entropy encoding will be described later with reference to FIGS.

  The rate controller 10 determines whether or not the amount of code data obtained by entropy coding is larger than a preset target data amount, and the amount of code data obtained by entropy coding is larger than the target data amount. When it determines with it being large, it requests | requires suppression of the data amount of a code | symbol with respect to the data deletion part 11. FIG. When it is determined that the data amount of the code obtained by entropy coding is less than or equal to the target data amount, the rate controller 10 outputs the code obtained by entropy coding as an encoded (compressed) audio signal. . The encoded speech signal output from the rate controller 10 is recorded on a recording medium or transmitted to an external device via a communication network.

Data deleting unit 11, by the rate controller 10, if the data amount of the code obtained by the entropy coding is determined to be greater than the target amount of data, energy | F _j | ² deletes the minimum bandwidth, after deletion Are output to the entropy encoding unit 9 and the entropy encoding is requested again.

<Entropy coding>
Hereinafter, Huffman coding and coding by a range coder will be described as examples of entropy coding applied in the present embodiment.

(Huffman coding)
Huffman coding is a method of compressing the entire data amount by assigning short codes to symbols with high appearance frequency and assigning long codes to symbols with low appearance frequency. For example, assume that there is 100 characters of data consisting of four symbols {a, b, c, d}. When a binary code (fixed length code) having the same length is assigned to all symbols, 2 bits are required to represent the four symbols, so the data amount of 100 characters is 2 [bit] × 100 = 200 [bit].

  In Huffman coding, a binary code is assigned according to the appearance frequency of each symbol. FIG. 4 shows an example of a binary code assigned to each symbol when the appearance frequency of each symbol a, b, c, d in 100-character data is 10, 70, 1, 19 respectively. Show. As shown in FIG. 4, when symbols 100, 0, 101, and 11 are assigned to the symbols a, b, c, and d, respectively, the data amount of 100 characters is 3 [bit] × 10 + 1 [bit] ×. 70 + 3 [bit] × 1 + 2 [bit] × 19 = 141 [bit], and the data amount is compressed to 70% of the data amount of the fixed-length code.

(Encoding by range coder)
Assume that a set of symbols included in the original signal before encoding is S = {s _i | i = 1,..., N}, and the appearance probability of each symbol s _i is p _i . Furthermore, symbol strings sorted in a predetermined order each symbol s _i in the original signal _{{s 1, s 2, ...} , s n} in, lined before the symbol s _{k (k} ≧ 2) Let G _k be the total appearance probability of each symbol. That is, G _k is expressed as in Expression (9).

In the encoding by the range coder, a symbol string indicated by an input signal is indicated by a numerical value based on a table (hereinafter referred to as an occurrence probability table) in which appearance probabilities p _i and G _i are stored in association with each symbol. Process to set the range (lower limit, width). The range (lower limit, width) set for the input signal is determined based on the range set for the signal input immediately before and the occurrence probability table.

The range set when the signal s _{k to} be encoded is input is range ', the lower limit is low', and the range set when the signal before the signal s _k is input is range, Assuming that the lower limit is low, the width range ′ and the lower limit low ′ are expressed as in Expression (10) and Expression (11), respectively.
range '= range × _pk (10)
low '= low + range × G _k (11)
The range ′ and low ′ calculated by Expression (10) and Expression (11) are the range and low when the next signal is input.

  The calculation processing shown in Expression (10) and Expression (11) is performed until there is no input signal. Between the range low and low + range determined based on the range and low calculated when the last signal is input. Is output as a code value.

FIG. 5 shows an example of range coder encoding. FIG. 5A shows an example of the occurrence probability table when the set of symbols included in the original signal is S = {s ₁ = a, s ₂ = b, s ₃ = c, s ₄ = d}. . FIG. 5B shows an example of encoding for the symbol string {baca}. FIG. 5B shows a case where the code indicating the symbol string is a decimal number, the initial value of low is 0, and the initial value of range is 10 ⁶ . In FIG. 5B, the “input signal” item indicates the input symbol, the “symbol string” item indicates the symbol string input so far, and the “low” item is expressed by the equation (11). “Low ′” is calculated, and the “range” item indicates “range ′” calculated by Expression (10). The “range” item indicates a range of code values determined from low and range. In FIG. 5B, the notation [x, y) means that the code value Z satisfies x ≦ Z <y. According to FIG. 5B, one of the code values Z satisfying 593750 ≦ Z <603125 (for example, 600000) is output as a result of encoding the symbol string {baca}.

  As described above, in encoding by the range coder, each symbol input is encoded using a predetermined appearance probability, and therefore, from an information source in which the appearance probability of each symbol included in the original signal is fixed. This is very effective. However, it is extremely rare that a signal to be encoded is generated from an information source having a constant appearance probability. Therefore, in the encoding by the above range coder, the appearance probability of each symbol is not adapted to the signal to be encoded. Therefore, in the present embodiment, in the range coder encoding, the appearance probability is updated every time a signal is input, so that it can be adapted to an actual signal. Hereinafter, encoding by the range coder of this embodiment will be described.

Similarly to the above, a set of symbols included in the original signal before encoding is S = {s _i | i = 1,..., N}. Assuming that the appearance frequency of symbols s _i included in the original signal is g _i , the sum of the appearance frequencies g _i is cum, and the appearance probability of each symbol s _i is p _i , cum and p _i are respectively expressed by Equation (12), It is expressed as equation (13).

The entropy encoding unit 9 has an occurrence probability table 81 as shown in FIG. 6 as a table for setting the width range and the lower limit low for the input signal. As shown in FIG. 6, the occurrence probability table 81 stores the items of the appearance frequency g _i , the appearance probability p _i , and G _i in association with each symbol. The definition of G _i is as shown in Expression (9).

式（１４）及び式（１５）で算出されたrange'、low'が、次の信号が入力されたときのrange、lowとなる。 The range set when the encoding target signal s _k is input to the entropy encoding unit 9 is set as range ', the lower limit is set as low', and is set when the signal before the signal s _k is input. Assuming that the obtained width is range and the lower limit is low, the width range ′ and the lower limit low ′ are respectively expressed as Expression (14) and Expression (15).

The range ′ and low ′ calculated by Expression (14) and Expression (15) are the range and low when the next signal is input.

When range and low are calculated by the input of the signal s _k , the entropy encoding unit 9 adds 1 to the appearance probability g _k as shown in the equation (16), and uses the calculated appearance probability g _k ′. Let it be a new g _k .
g _k '= g _k +1 (16)
The entropy encoding unit 9 recalculates the cum, the appearance probabilities p _i and G _i with the addition of the appearance probability g _k , and updates the occurrence probability table 81. The entropy encoding unit 9 performs these processes until there is no input signal, and encodes a value between the range low to low + range determined based on the range and low calculated when the last signal is input. Output as a value.

7 and 8 show examples of range coder encoding according to this embodiment. FIG. 7A shows a default occurrence probability table 81 when the set of symbols included in the original signal is S = {s ₁ = a, s ₂ = b, s ₃ = c, s ₄ = d}. An example is shown. It is assumed that p _i and G _i of the default occurrence probability table 81 shown in FIG. 7A are the same as the occurrence probability table shown in FIG. FIG. 7B shows an example of encoding for the same symbol string {baca} as the symbol string shown in FIG. Also in FIG. 7B, the symbol indicating the symbol string is a decimal number, the initial value of low is 0, and the initial value of range is 10 ⁶ . In FIG. 7B, the “input signal” item indicates the input symbol, the “symbol string” item indicates the symbol string input so far, and the “low” item is expressed by the equation (15). “Low ′” is calculated, and the “range” item indicates “range ′” calculated by the equation (14). The “range” item indicates a range of code values determined from low and range. The “occurrence probability table” item indicates an occurrence probability table updated every time a symbol is input. FIG. 8 shows an occurrence probability table updated every time a symbol is input. According to FIG. 7B, by updating the occurrence probability table for each input of the symbol, the “range” indicated by the symbol string {baca} is the case where the occurrence probability table shown in FIG. 5B is fixed. Unlike the above, one of the code values Z satisfying 591992 ≦ Z <599757 is output as a result of encoding the symbol string {baca}.

  FIG. 9 shows a configuration of a speech decoding apparatus 200 that decodes a speech signal encoded (compressed) by the speech encoding apparatus 100. As shown in FIG. 9, the speech decoding apparatus 200 includes an entropy decoding unit 21, an inverse vector quantization unit 22, a time order rearrangement unit 23, a multiplication unit 24, a frequency inverse transformation unit 25, a level reproduction unit 26, and a frame synthesis unit. 27, a D / A converter 28. Note that the structure may be such that the speech encoding apparatus 100 and the speech decoding apparatus 200 are integrally provided in one housing, or each may be provided separately.

  The entropy decoding unit 21 decodes the signal encoded by entropy encoding and outputs the decoded signal to the inverse vector quantization unit 22. The inverse vector quantization unit 22 has a high frequency VQ table 22a and a low frequency VQ table 22b as tables storing representative vectors indicating a plurality of speech patterns, and a signal (index) input from the entropy decoding unit 21. ) Are extracted and output to the time order rearrangement unit 23. The high frequency VQ table 22a and the low frequency VQ table 22b are respectively the same as the high frequency VQ table 8a and the low frequency VQ table 8b shown in FIG.

  The time order rearrangement unit 23 rearranges the vectors input from the inverse vector quantization unit 22 in time order and outputs the vectors to the multiplication unit 24. The multiplying unit 24 multiplies the MDCT coefficient of the MDCT block specified in advance among the MDCT blocks for one frame obtained by the time-order rearranging unit 23 by the MDCT coefficient of the other MDCT block, and inverses the frequency of the multiplication result. The data is output to the conversion unit 25. Here, the MDCT block designated in advance is the same as the MDCT block used as a division value in the division unit 6 of the speech encoding apparatus 100 at the time of encoding.

  The frequency inverse transform unit 25 performs inverse MDCT on the signal (MDCT coefficient) input from the multiplication unit 24 and outputs the result to the level reproduction unit 26. The level reproduction unit 26 adjusts the level of the signal input from the frequency inverse conversion unit 25, returns it to the original level, and outputs it to the frame synthesis unit 27.

  The frame synthesizing unit 27 synthesizes frames that are processing units of encoding and decoding, and outputs the combined signal to the D / A converting unit 28. The D / A converter 28 converts the digital signal input from the frame synthesizer 27 into an analog signal and outputs it as an audio reproduction signal.

Next, the operation in this embodiment will be described.
First, the speech encoding process executed in speech encoding apparatus 100 will be described with reference to the flowchart of FIG.

  First, when an audio analog signal is input, the input audio analog signal is converted into an audio digital signal in the A / D converter 1 (step S1). Hereinafter, the audio digital signal to be encoded is simply referred to as an audio signal. Next, the DC removal unit 2 deletes the DC component of the audio signal (step S2), and the framing unit 3 divides the audio signal after the DC component deletion into frames (step S3).

  Next, the level adjustment unit 4 adjusts the level of the input audio signal for each frame (step S4), and the frequency conversion unit 5 applies MDCT to the audio signal after level adjustment (step S5). ).

  Next, the division unit 6 performs a division process (frequency conversion coefficient division process) on the MDCT coefficient obtained in step S5 for each frame (step S6). In step S6, as shown in FIG. 11, the MDCT coefficient of the MDCT block designated in advance among the MDCT blocks for one frame obtained by the frequency converter 5, and the MDCT coefficients of other MDCT blocks in the same frame are set. Divide (step S20).

  Next, the frequency rearrangement unit 7 rearranges the MDCT coefficients obtained by the frequency conversion coefficient division processing in step S6 for each frequency (step S7), and collectively vectorizes the coefficients in the same frequency band.

  Next, the vector quantization unit 8 compares the high-frequency MDCT coefficient vector with the representative vector stored in the high-frequency VQ table 8a, and stores the low-frequency MDCT coefficient vector in the low-frequency VQ table 8b. The stored representative vectors are compared, and the index indicated by the most similar representative vector is output as a code (step S8).

  Next, entropy encoding is performed for each frame on the speech signal after vector quantization (step S9), and the entropy encoded signal is output to the rate controller 10 as a speech encoded signal. Next, the rate controller 10 determines whether or not the speech encoded signal for one frame input from the entropy encoding unit 9 is equal to or less than a preset target data amount (step S10).

If it is determined in step S10 that the input speech encoded signal is larger than the target data amount (step S10; NO), the data deleting unit 11 uses the energy | F _{j in} the speech encoded signal for one frame. The signal in the band with the minimum | ² is deleted (step S12). When step S12 ends, the process returns to step S9, and entropy coding is performed again on the audio signal of the corresponding frame (step S9).

  When it is determined in step S10 that the input speech encoded signal is equal to or less than the target data amount (step S10; YES), it is determined whether the speech signal of the next frame is input to the entropy encoding unit 9 or not. (Step S11).

  In step S11, when it is determined that the audio signal of the next frame is input to the entropy encoding unit 9 (step S11; YES), the process returns to step S9, and entropy encoding is performed on the frame (step S9). If it is determined in step S11 that entropy encoding has been completed for all frames input to the entropy encoding unit 9 (step S11; NO), the speech encoding process ends.

<Modification of frequency conversion coefficient division>
The frequency conversion coefficient division processing method in step S6 is not limited to the method shown in FIG. Hereinafter, a modification of the frequency conversion coefficient division process of FIG. 11 will be described with reference to the flowchart of FIG.

First, the sum of absolute values of MDCT coefficients is calculated for each MDCT block for one frame (step S30). When the MDCT coefficient of the _i -th MDCT block is M _i = {m _in | n = 0,..., k−1}, the sum SM _i of the absolute values of the MDCT coefficients of the i-th MDCT block is given by Equation (17). It is expressed as follows.

Then, among the MDCT block of the corresponding frame, the sum SM _i of the largest absolute value MDCT block of MDCT coefficients is selected (step S31), in MDCT coefficients of the selected MDCT blocks, other MDCT in the same frame The MDCT coefficient of the block is divided (step S32), and the frequency conversion coefficient division process is completed.

式（１８）と式（７）を比較すると、式（１８）の方が、除算処理後のＭＤＣＴ係数のとり得る範囲がより狭いことがわかる。 For example, in the example shown in Expression (6), when the sum SM _i of the absolute values of the MDCT coefficients is calculated for each MDCT block, SM ₀ = 186, SM ₁ = 201, SM ₂ = 150, SM ₃ = 180. Become. Among these, since SM ₁ is the largest, M ₁ is selected as an MDCT block to be used as a division value. At this time, M ′ _i (i = 0, 2, 3) after the division processing is as shown in Expression (18).

Comparing equation (18) and equation (7), it can be seen that equation (18) has a narrower range of possible MDCT coefficients after division processing.

As another method of frequency conversion coefficient division processing, there is a method of using an MDCT block having an MDCT coefficient having the largest absolute value as a division value. For example, in the example shown in Expression (6), the MCDT coefficient having the largest absolute value in each MDCT block is M ₀ is 112, M ₁ is 120, M ₂ is 97, and M ₃ is 110. Since the largest of these is M ₁ 120, M ₁ is selected as the MDCT block to be used as a division value.

  The multiplication unit 24 of the speech decoding apparatus 200 performs frequency conversion coefficient multiplication processing corresponding to the method of frequency conversion coefficient division processing (hereinafter simply referred to as “division processing”) shown in step S6 of FIG.

  When the division process of FIG. 11 is performed in step S6, the multiplication unit 24 sets the MDCT coefficient of the MDCT block designated in advance among the MDCT blocks for one frame as shown in FIG. The MDCT coefficient of each MDCT block is multiplied (step S40). Here, the MDCT block designated in advance is the same as the MDCT block used as a division value in the division unit 6 of the speech encoding apparatus 100 at the time of encoding.

  When the division process of FIG. 12 is performed in step S6, the multiplication unit 24 determines that the MDCT coefficients of the MDCT block selected in step S31 out of the MDCT blocks for one frame are other values as shown in FIG. The MDCT coefficient of the MDCT block is multiplied (step S50).

  In step S6, when division processing using the MDCT block having the MDCT coefficient having the largest absolute value as the division value is performed, the multiplication unit 24 converts the MDCT coefficient of the MDCT block into the MDCT coefficient of another MDCT block. Is multiplied.

  As described above, according to the speech coding apparatus 100 and the speech decoding apparatus 200 of the present embodiment and the modifications thereof, the MDCT coefficients of other MDCT blocks are obtained using the MDCT coefficients of one MDCT block in one frame. By performing the division, the range of the absolute value of the MDCT coefficient after the division is narrowed, the efficiency of vector quantization, entropy coding and the like is improved, and the sound quality can be improved. Also, by performing the division process only when the absolute value of the division value is 1 or more, the MDCT coefficient is prevented from becoming larger and the vector quantization efficiency is not adversely affected.

Note that the description in the present embodiment can be changed as appropriate without departing from the spirit of the present invention.
For example, although the case where MDCT is used as frequency conversion has been described in the above-described embodiment, other frequency conversion such as FFT (Fast Fourier Transform) may be used.

The block diagram which shows the structure of the audio | voice coding apparatus which concerns on embodiment of this invention. The figure which shows the frame division | segmentation of an input signal. The figure which shows the relationship between 1 frame and each MDCT block. The figure which shows an example of a Huffman code | symbol. The figure which shows an example of the encoding by the conventional range coder. The figure which shows the data structure of the occurrence probability table 81 required for the range coder encoding of this embodiment. The figure which shows an example (the figure (a)) of default occurrence probability table 81, and an example (the figure (b)) of encoding. The figure which shows the example of an update of the occurrence probability table 81. FIG. The block diagram which shows the structure of the audio | voice decoding apparatus which decodes the audio | voice encoded. The flowchart which shows the audio | voice encoding process performed in the audio | voice encoding apparatus of this embodiment. The flowchart which shows the frequency-transform coefficient division process performed in the audio | voice coding apparatus of this embodiment. The flowchart which shows the modification of the frequency conversion coefficient division process of FIG. The flowchart which shows the frequency conversion coefficient multiplication process performed in the audio | voice decoding apparatus of this embodiment. 14 is a flowchart showing a modification of the frequency conversion coefficient multiplication process of FIG.

Explanation of symbols

1 A / D conversion unit 2 DC removal unit 3 Framing unit 4 Level adjustment unit 5 Frequency conversion unit 6 Division unit 7 Frequency rearrangement unit 8 Vector quantization unit 8a High frequency VQ table 8b Low frequency VQ table 9 Entropy code Conversion unit 10 rate controller 11 data deletion unit 81 occurrence probability table 21 entropy decoding unit 22 inverse vector quantization unit 23 time order rearrangement unit 24 multiplication unit 25 frequency inverse conversion unit 26 level reproduction unit 27 frame synthesis unit 28 D / A conversion Unit 100 speech coding apparatus 200 speech decoding apparatus

Claims (6)

A framing unit that divides the input audio signal into frames;
A frequency conversion unit that performs frequency conversion on an audio signal and calculates a conversion coefficient for each block unit constituting one frame;
A division unit that divides a conversion coefficient of another block in the same frame by a conversion coefficient of one block designated in advance among the conversion coefficients obtained by the frequency conversion unit;
A vector quantization unit that collectively vector-quantizes the divided values of the transform coefficients of each block obtained by the division unit for each frequency band;
An encoding unit that encodes the index data obtained by the vector quantization unit using a predetermined encoding method;
A speech encoding apparatus comprising:   A framing unit that divides the input audio signal into frames;
  A frequency conversion unit that performs frequency conversion on an audio signal and calculates a conversion coefficient for each block unit constituting one frame;
  Based on the transform coefficient obtained by the frequency transform unit, the sum of the absolute values of the transform coefficients is calculated for each block unit constituting the frame, and the same transform coefficient of the block having the largest absolute value is the same. A division unit that divides the transform coefficients of other blocks in the frame;
  A vector quantization unit that collectively vector-quantizes the divided values of the transform coefficients of each block obtained by the division unit for each frequency band;
  An encoding unit that encodes the index data obtained by the vector quantization unit using a predetermined encoding method;
A speech encoding apparatus comprising:   A framing unit that divides the input audio signal into frames;
  A frequency conversion unit that performs frequency conversion on an audio signal and calculates a conversion coefficient for each block unit constituting one frame;
  A division unit that obtains a block having a maximum absolute value among the transform coefficients calculated for each block constituting the frame, and divides the transform coefficient of another block in the same frame by the transform coefficient of the block. When,
  A vector quantization unit that collectively vector-quantizes the divided values of the transform coefficients of each block obtained by the division unit for each frequency band;
  An encoding unit that encodes the index data obtained by the vector quantization unit using a predetermined encoding method;
A speech encoding apparatus comprising: A framing step for dividing the input audio signal into frames;
A frequency conversion step of performing a frequency conversion on the audio signal and calculating a conversion coefficient for each block unit constituting one frame;
A division step of dividing a conversion coefficient of another block in the same frame by a conversion coefficient of one block specified in advance among the conversion coefficients obtained in the frequency conversion step;
A vector quantization step for performing vector quantization for each frequency band by dividing the divided values of the transform coefficients of each block obtained in the division step;
An encoding step for encoding the index data obtained by the vector quantization step by a predetermined encoding method;
A speech encoding method comprising:   A framing step for dividing the input audio signal into frames;
  A frequency conversion step of performing a frequency conversion on the audio signal and calculating a conversion coefficient for each block unit constituting one frame;
  Based on the transform coefficient obtained in the frequency transform step, the sum of the absolute values of the transform coefficients is calculated for each block unit constituting the frame, and the same transform coefficient of the block having the largest absolute value is the same. A division step to divide the transform coefficients of other blocks in the frame;
  A vector quantization step for performing vector quantization for each frequency band by dividing the divided values of the transform coefficients of each block obtained in the division step;
  An encoding step for encoding the index data obtained by the vector quantization step by a predetermined encoding method;
A speech encoding method comprising:   A framing step for dividing the input audio signal into frames;
  A frequency conversion step of performing a frequency conversion on the audio signal and calculating a conversion coefficient for each block unit constituting one frame;
  A division step of obtaining a block having a maximum absolute value among the transform coefficients calculated for each block constituting the frame, and dividing the transform coefficient of another block in the same frame by the transform coefficient of the block When,
  A vector quantization step for performing vector quantization for each frequency band by dividing the divided values of the transform coefficients of each block obtained in the division step;
  An encoding step for encoding the index data obtained by the vector quantization step by a predetermined encoding method;
A speech encoding method comprising:

Images