JP4259401B2 - Speech processing apparatus and speech coding method - Google Patents

Description

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

By performing frequency conversion, vector quantization, and entropy coding on the audio signal and deleting a specific frequency component, the audio signal is efficiently compressed. In the process of deleting a specific frequency component, an audio signal is deleted from the frequency component with the smallest energy in order to suppress the data amount after entropy coding to be equal to or less than the target data amount (see, for example, Patent Document 1).
JP 2001-166797 A

  However, in the method of reducing the data amount of the audio signal by deleting the audio signal from the frequency component with low energy, noise may remain if the target data amount is set to be somewhat low.

  An object of the present invention is to reduce noise by deleting a frequency component that exists in isolation from other frequency components in an audio signal encoding (compression) process.

An audio processing device according to the present invention is based on a framing unit that divides an input audio signal into frames, and for each frame obtained by the framing unit, based on the maximum value of the amplitude of the audio signal included in the frame. A level adjusting unit for adjusting a level of the audio signal, a frequency converting unit for performing frequency conversion on the audio signal whose level is adjusted by the level adjusting unit, and a vector quantum for the audio signal obtained by the frequency conversion. A vector quantization unit that performs encoding, an entropy coding unit that performs entropy coding on the speech signal obtained by the vector quantization, and a data amount of the speech signal obtained by the entropy coding unit in advance A determination unit that determines whether the amount of data is larger than a set target data amount, and a sound obtained by the entropy encoding unit by the determination unit; If the data amount of the signal is determined to be greater than the target amount of data, from among said vector quantized audio signal, and a first deletion processing energy to remove the frequency components of the lowest band, on the frequency axis And a data deletion unit that performs a second deletion process that deletes frequency components that have zero adjacent frequency components and an absolute value of an amplitude value greater than a predetermined value, and the audio signal that has been data-deleted by the data deletion unit again. A controller that performs entropy encoding and performs the deletion process and the encoding process until the data amount of the encoded audio signal falls within the target data amount .

  In the second deletion process of the data deletion unit, it is preferable that the predetermined value is a value calculated from a maximum value of amplitude values of all frequency components.

Further, the data deletion unit determines whether or not the value of the high frequency component of the audio signal obtained by the frequency conversion unit is greater than the value of the low frequency component, and the value of the high frequency component of the audio signal is the low frequency component When it is determined that the frequency is equal to or less than the value, it is preferable to delete the high-frequency component that exists more than a preset frequency interval on the frequency axis and execute the second deletion process.

An audio processing device according to the present invention is based on a framing unit that divides an input audio signal into frames, and for each frame obtained by the framing unit, based on the maximum value of the amplitude of the audio signal included in the frame. A level adjusting unit for adjusting a level of the audio signal, a frequency converting unit for performing frequency conversion on the audio signal whose level is adjusted by the level adjusting unit, and a vector quantum for the audio signal obtained by the frequency conversion. A vector quantization unit that performs encoding, an entropy coding unit that performs entropy coding on the speech signal obtained by the vector quantization, and a data amount of the speech signal obtained by the entropy coding unit in advance A determination unit that determines whether the amount of data is larger than a set target data amount, and a sound obtained by the entropy encoding unit by the determination unit; If the data amount of the signal is determined to be greater than the target amount of data, from among said vector quantized audio signal, and a first deletion processing energy to remove the frequency components of the lowest band, on the frequency axis A data deletion unit that performs a second deletion process that deletes frequency components that are separated by a predetermined frequency interval or more, and an audio signal that has been data-deleted by the data deletion unit is re-entropy-encoded and encoded. And a controller that performs the deletion process and the encoding process until the data amount of the audio signal falls within the target data amount .

  Further, the data deletion unit determines whether or not the value of the high frequency component of the audio signal obtained by the frequency conversion unit is greater than the value of the low frequency component, and the value of the high frequency component of the audio signal is the low frequency component When it is determined that the frequency component is equal to or less than the value, it is preferable to delete, from the high-frequency component, a frequency component that exists at a predetermined frequency interval or more on the frequency axis.

  Further, the data deletion unit calculates the logarithm of each frequency component in the audio signal obtained by the frequency conversion unit, and compares the sum of the logarithm of the high frequency component with the sum of the logarithm of the low frequency component, thereby It is preferable to determine whether the value of the high frequency component of the audio signal is greater than the value of the low frequency component.

  The data deletion unit separates the audio signal whose level is adjusted by the level adjustment unit into a high frequency component and a low frequency component by a high pass filter and a low pass filter, It is preferable to determine whether or not the value of the high frequency component obtained by the pass filter is larger than the value of the low frequency component.

The speech coding method according to the present invention divides an input speech signal into frames, and adjusts the level of the speech signal for each frame based on the maximum value of the amplitude of the speech signal included in the frame. The speech signal is subjected to frequency transformation, the speech signal obtained by the frequency transformation is subjected to vector quantization, the speech signal obtained by the vector quantization is subjected to entropy coding, It is determined whether the data amount of the audio signal obtained by entropy encoding is larger than a preset target data amount, and the data amount of the audio signal obtained by entropy encoding is larger than the target data amount. If it is determined that, from among the vector quantized speech signals, energy removes the frequency components of the lowest band, adjacent on the frequency axis In wavenumber component is 0 and the absolute value of the amplitude value removes the larger frequency component than a predetermined value, the re-entropy coding for the data deleted voice signal, the data amount is the target data amount of the encoded speech signal The deletion process and the encoding process are performed until they fall within the range .

The speech coding method according to the present invention divides an input speech signal into frames, and adjusts the level of the speech signal for each frame based on the maximum value of the amplitude of the speech signal included in the frame. The speech signal is subjected to frequency transformation, the speech signal obtained by the frequency transformation is subjected to vector quantization, the speech signal obtained by the vector quantization is subjected to entropy coding, It is determined whether the data amount of the audio signal obtained by entropy encoding is larger than a preset target data amount, and the data amount of the audio signal obtained by entropy encoding is larger than the target data amount. If it is determined that, from among the vector quantized speech signals, energy removes the frequency components of the lowest band, preset on the frequency axis Remove the frequency components present apart over the frequency interval that the re entropy coding for the data deleted voice signal, the deletion process and the code until the data amount of the encoded audio signal falls within the target data quantity It is characterized in that the processing is performed .

According to the present invention, in order to suppress the data amount after entropy encoding to be equal to or less than the target data amount, when performing audio encoding in which a frequency component with low energy is deleted from the audio signal, the adjacent data on the frequency axis is used. Noise can be reduced by deleting frequency components whose matching frequency components are 0 and whose absolute value of the amplitude value is larger than a predetermined value .

In addition, according to the present invention, in order to suppress the amount of data after entropy encoding to be equal to or less than the target data amount, when performing speech encoding in which a frequency component with low energy is deleted from the speech signal, The noise can be reduced by deleting frequency components that are separated by a predetermined frequency interval or more .

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 is a block diagram showing a configuration of a speech processing apparatus 100 according to an embodiment of the present invention. As shown in FIG. 1, the audio processing apparatus 100 includes an A / D conversion unit 1, a DC (Direct Current) removal unit 2, a framing unit 3, a level adjustment unit 4, a frequency conversion unit 5, a frequency rearrangement unit 6, The vector quantization unit 7, entropy encoding unit 8, rate controller 9, and data deletion unit 10 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 compression processing units 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を表す信号を、音声圧縮信号の一部として出力する。 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.

At the time of audio reproduction, since it is necessary to restore the signal whose amplitude is suppressed to the suppression target bit or less, a signal representing shift_bit is output as a part of the audio compression 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 frequency rearrangement 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 | n = 0,. expressed.

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

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

The frequency rearrangement unit 6 rearranges the MDCT coefficients input from the frequency conversion unit 5 for each frequency, collectively vectorizes the coefficients in the same frequency band, and outputs them to the vector quantization unit 7. Thus, 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 by 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 7 has a VQ (Vector Quantization) table in which representative vectors representing a plurality of sound patterns are stored. The vector F _j created by the frequency rearranging unit 6 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 8.

代表ベクトルの数ｋとベクトル長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 (5).

The number of representative vectors k and the vector length N 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 in which the high-frequency representative vector is stored is referred to as a high-frequency VQ table 7a, and the VQ table in which the low-frequency representative vector is stored is referred to as a low-frequency VQ table 7b. In the vector F _j = {X _ij | i = 0,..., M−1}, j = 0,..., M / 2-1 created by the frequency rearrangement unit 6, 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 7b, 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 7a, and the most similar representatives are compared. The index indicated by the vector is output as a code.

  The entropy coding unit 8 performs entropy coding on the code input from the vector quantization unit 7 and outputs the code to the rate controller 9. 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 9 determines whether the data amount of the code obtained by entropy coding is larger than a preset target data amount, and the data amount of the code 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 10. FIG. When it is determined that the data amount of the code obtained by entropy coding is equal to or less than the target data amount, the rate controller 9 outputs the code obtained by entropy coding as a speech compression signal. The audio compression signal output from the rate controller 9 is recorded on a recording medium or transmitted to an external device via a communication network.

Data deleting unit 10, by the rate controller 9, when 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 | with ² to delete the minimum bandwidth, frequency Band that deletes frequency components whose absolute value difference between adjacent frequency components on the axis is larger than a predetermined value, and deletes frequency components that are separated by more than a preset frequency interval on the frequency axis Data deletion processing is executed (see FIGS. 10 to 12). A frequency component in which the absolute value of the difference between the amplitude values of adjacent frequency components on the frequency axis is greater than a predetermined value is a frequency in which the absolute value of the amplitude value of the frequency component is greater than or equal to the predetermined value and adjacent on the frequency axis The frequency component of the band whose component value is 0 is shown. Then, the data deletion unit 10 outputs the audio signal after the band data deletion process to the entropy encoding unit 8 and requests entropy encoding again. The band data deletion process will be described in detail later with reference to FIGS.

<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 (6).

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 (7) and Expression (8), respectively.
range '= range × _pk (7)
low '= low + range × G _k (8)
The range ′ and low ′ calculated by the equations (7) and (8) are the range and low when the next signal is input.

  The calculation processing shown in Expression (7) and Expression (8) is performed until there is no input signal, and the range calculated when the last signal is input, the range determined based on low, between low to low + range 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 (8). “Low ′” is calculated, and the “range” item indicates “range ′” calculated by Expression (7). 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 (9), It is expressed as equation (10).

The entropy encoding unit 8 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 (6).

式（１１）及び式（１２）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 8 is set to range ', the lower limit is set to low', and is set when the signal immediately 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 (11) and Expression (12).

The range ′ and low ′ calculated by Expression (11) and Expression (12) are the range and low when the next signal is input.

When range and low are calculated by inputting the signal s _k , the entropy encoding unit 8 adds 1 to the appearance probability g _k as shown in the equation (12-1), and calculates the calculated appearance probability g _k. Let 'be a new g _k .
g _k '= g _k +1 (12-1)
The entropy encoding unit 8 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 8 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 (12). “Low” is calculated, and the “range” item indicates “range ′” calculated by the equation (11). 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}.

Next, the operation in this embodiment will be described.
First, an audio compression process executed in the audio processing 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, in the frequency rearrangement unit 6, the MDCT coefficients are rearranged for each frequency (step S6), and the coefficients in the same frequency band are collectively vectorized.

  Next, the vector quantization unit 7 compares the high-frequency MDCT coefficient vector with the representative vector stored in the high-frequency VQ table 7a, and stores the low-frequency MDCT coefficient vector in the low-frequency VQ table 7b. The stored representative vectors are compared, and the index indicated by the most similar representative vector is output as a code (step S7).

  Next, entropy encoding is performed for each frame on the speech signal after vector quantization (step S8), and the entropy encoded signal is output to the rate controller 9 as a speech compression signal. Next, the rate controller 9 determines whether or not the audio compression signal for one frame input from the entropy encoding unit 8 is equal to or less than a preset target data amount (step S9).

  If it is determined in step S9 that the input audio compression signal is larger than the target data amount (step S9; NO), the data deletion unit 10 performs a band data deletion process (step S11), and the corresponding processing is performed again. Entropy coding is performed on the frame (step S8). The band data deletion process in step S11 will be described in detail later with reference to FIG.

  If it is determined in step S9 that the input audio compression signal is equal to or less than the target data amount (step S9; YES), it is determined whether or not the audio signal of the next frame has been input to the entropy encoding unit 8. (Step S10). In step S10, when it is determined that the audio signal of the next frame is input to the entropy encoding unit 8 (step S10; YES), the entropy encoding for the frame is performed again (step S8). If it is determined in step S10 that entropy coding has been completed for all frames input to the entropy coding unit 8 (step S10; NO), the speech compression processing is terminated.

  Next, the band data deletion process shown in step S11 of FIG. 9 will be described with reference to the flowchart of FIG.

First, the frequency band (frequency component) with the minimum energy | F _j | ² is deleted from the audio signal obtained by the vector quantization in step S7 (step S20). Next, the value of the high frequency component and the value of the low frequency component of the frequency signal calculated in step S6 are compared, and it is determined whether or not the audio signal to be processed is a signal (waveform) with a lot of high frequency components ( Step S21).

Ｆ_H＞Ｆ_Lである場合に、高周波成分の多い信号（波形）であると判断される。音の大きさは対数的に感じられるため、式（１３）及び式（１４）に示すように、周波数成分の対数をとって合計することにより、処理対象の音声信号が高周波成分の多い信号であるか否かを確実に判断することができる。 The comparison between the low-frequency component and the high-frequency component in step S21 may be performed by dividing the audio signal to be processed into a low-frequency component and a high-frequency component and comparing the totals as follows. If the audio signal to be processed is separated into N frequency components [F _i | i = 0,..., N−1], the total low frequency component F _L and the total high frequency component F _H are respectively Are defined as in the following equations (13) and (14).

When F _H > F _L, it is determined that the signal has a high frequency component (waveform). Since the loudness of the sound is felt logarithmically, as shown in the equations (13) and (14), the logarithm of the frequency components is taken and totaled so that the audio signal to be processed is a signal having a high frequency component. It can be reliably determined whether or not there is.

  In step S21, when it is determined that the audio signal to be processed is a signal (waveform) with many high frequency components (step S21; YES), an isolated frequency component deletion process is performed (step S23), and this band data deletion process is performed. Ends. The isolated frequency component deletion processing in step S23 will be described in detail later with reference to FIG.

  In step S21, when it is determined that the audio signal to be processed is a signal (waveform) with a small number of high frequency components (step S21; NO), the frequency interval of the high frequency component signals that is set in advance on the frequency axis is greater than or equal to A high-frequency noise deletion process is performed to delete frequency components that are present apart (step S22). The high-frequency noise elimination process in step S22 will be described in detail later with reference to FIG. When the high-frequency noise removal processing is completed, in order to prevent a clip sound from entering other than the high-frequency components, the isolated frequency component deletion processing is performed on the audio signal after the high-frequency noise removal processing (step S23), and this band data The deletion process ends.

Ｘ_H＞Ｘ_Lである場合に、高周波成分の多い信号（波形）であると判断される。 The low frequency component and the high frequency component shown in the equations (13) and (14) are separated using an HPF (High Pass Filter) and an LPF (Low Pass Filter). Can also be done. In this case, the audio signal level-adjusted by the level adjustment unit 4 is output to the data deletion unit 10, and the data deletion unit 10 uses the LPF and HPF to convert the audio signal into a low frequency component and a high frequency component as follows. To be separated. Assuming that the audio signal output from the level adjusting unit 4 is {x _n | n = 0,..., M−1}, the low frequency component total X _L and the high frequency component total X _H are respectively expressed by the following formulas ( 15), defined as equation (16).

When X _H > X _L, it is determined that the signal has a high frequency component (waveform).

Next, with reference to the flowchart of FIG. 11, the high-frequency noise deletion process shown in step S22 of FIG. 10 will be described in detail. In the following, it is assumed that the audio signal is separated into N frequency components [F _i | i = 0,..., N−1], and the sampling frequency of the audio signal is f _s (Hz), which is a processing target. The band of the high frequency component is set to S (Hz) or more, and T is a preset frequency interval.

First, the counter value j for specifying a frequency component is set to _{j = S × 2N / f s} ( step S30). The high-frequency component that is the target of the high-frequency noise elimination processing is higher than the j-th frequency component. Next, it is determined whether or not the current counter value j is less than N (step S31). If it is determined in step S31 that the counter value j is greater than or equal to N (step S31; NO), the high frequency noise elimination process ends.

If it is determined in step S31 that the counter value j is less than N (step S31; YES), it is determined whether or not the j-th frequency component F _j is 0 (step S32). If it is determined in step S32 that the j-th frequency component F _j is not 0 (step S32; NO), the counter value j is incremented (step S33), and the process returns to step S31.

If it is determined in step S32 that the j-th frequency component F _j is 0 (step S32; YES), j + 1 is set as a new counter value i (step S34), and this counter value i is less than N. It is determined whether or not there is (step S35).

When it is determined in step S35 that the counter value i is less than N (step S35; YES), it is determined whether or not the i-th frequency component F _i is 0 (step S36). If it is determined in step S36 that the i-th frequency component F _i is 0 (step S36; YES), the counter value i is incremented (step S37), the process returns to step S35, and the current counter value i is N. It is determined whether it is less than (step S35).

If it is determined in step S36 that the i-th frequency component F _i is not 0 (step S36; NO), it is determined whether j + T is less than i (step S38). If it is determined in step S35 that the counter value i is greater than or equal to N (step S35; NO), the process proceeds to step S38, and it is determined whether j + T is less than i (step S38).

If it is determined in step S38 that j + T is equal to or greater than i (step S38; NO), the counter value i is set as the counter value j (step S39), and the counter value j is incremented (step S33). Return to S31. In step S38, the case of j + T is determined to be less than i (step S38; YES), N-1 th frequency component from the j-th are removed _{(F j ~F N-1 =} 0) ( step S40) This high-frequency noise removal process ends. As a result, of the high-frequency component signals, frequency components that are separated by a predetermined frequency interval or more on the frequency axis are deleted.

  Next, the isolated frequency component deletion process shown in step S23 of FIG. 10 will be described in detail with reference to the flowchart of FIG. Also in the following isolated frequency component deletion processing, the number of frequency components of the audio signal is N.

  First, a counter value i for designating a frequency component is set to 1 (step S50). Next, it is determined whether or not the current counter value i is less than N−1 (step S51). In step S51, when it is determined that the counter i is equal to or greater than N−1 (step S51; NO), the isolated frequency component deletion process ends.

If it is determined in step S51 that the counter value i is less than N−1 (step S51; YES), the frequency components F _i−1 and F _{i + 1} before and after the i-th frequency component F _i are 0. Further, it is determined whether or not the absolute value | F _i | of the i-th frequency component F _i is larger than a preset value P (step S52). Here, the preset value P is preferably calculated from the maximum value of the amplitude values of all frequency components. The maximum absolute value max of all frequency components is expressed as shown in Equation (17).
max = MAX (| F _i |) i = 0,…, N-1 (17)
Here, MAX is a function that takes the maximum value. A value obtained by multiplying the maximum value max shown in Expression (17) by a preset magnification r can be set as P (P = r · max). This magnification r takes, for example, a value of 0.1.

In step S52, if F _i-1 = 0, F _{i + 1} = 0, and | F _i |> P are not satisfied (step S52; NO), the counter value i is incremented (step S54), and the process returns to step S51. . In step S52, when F _i-1 = 0, F _{i + 1} = 0, and | F _i |> P are satisfied (step S52; YES), the i-th frequency component F _i is set to 0 (step S53). The counter value i is incremented (step S54), and the process returns to step S51. To delete frequency components satisfying F _i-1 = 0, F _{i + 1} = 0, | F _i |> P, the absolute value of the difference in amplitude value between adjacent frequency components on the frequency axis is set in advance. This corresponds to deleting a frequency component larger than the value P.

  As described above, according to the audio processing device 100 of the present embodiment, by deleting a frequency component whose absolute value of the difference in amplitude value between adjacent frequency components on the frequency axis is larger than a predetermined value, The included noise can be reduced.

  Also, if the value of the high frequency component of the audio signal to be processed is less than the value of the low frequency component, the high frequency component is included in the audio signal by deleting the frequency component that exists more than the preset frequency interval. Noise can be reduced.

  Note that the description in the present embodiment can be changed as appropriate without departing from the spirit of the present invention.

式（１８）のｂ_iを、実施形態における周波数成分Ｆ_iの代わりに用いることができる。例えば、ｂ_iが削除対象となった場合は、[Ｘ_pi|p=0,…,K-1]が削除される。 For example, each calculation in the above-described embodiment can collectively process a plurality of sets of MDCT coefficients as described below. That is, if there are K MDCT blocks and the i-th MDCT coefficient of the p-th MDCT block is X _pi , the i-th frequency component b _i is defined as in Expression (18).

B _i in equation (18) can be used in place of the frequency component F _{i in} the embodiment. For example, if b _i is subject to deletion, | is deleted _{[X pi p = 0, ...} , K-1].

The block diagram which shows the structure of the audio processing 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 flowchart which shows the audio | voice compression process performed in the audio | voice processing apparatus of this embodiment. The flowchart which shows the detail of the zone | band data deletion process shown by FIG. FIG. 11 is a flowchart showing details of high-frequency noise removal processing shown in FIG. 10. FIG. 11 is a flowchart showing details of an isolated frequency component deletion process shown in FIG. 10.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 A / D conversion part 2 DC removal part 3 Framing part 4 Level adjustment part 5 Frequency change part 6 Frequency rearrangement part 7 Vector quantization part 7a VQ table 7b for low regions 8 VQ table 8 for low regions 8 Entropy encoding part 81 Occurrence probability table 9 Rate controller (determination unit)
10 Data Deletion Unit 100 Voice Processing Device

Claims (9)

A framing unit that divides the input audio signal into frames;
For each frame obtained by the framing unit, a level adjusting unit that adjusts the level of the audio signal based on the maximum value of the amplitude of the audio signal included in the frame;
A frequency conversion unit that performs frequency conversion on the audio signal whose level is adjusted by the level adjustment unit;
A vector quantization unit that performs vector quantization on the audio signal obtained by the frequency conversion;
An entropy encoding unit that performs entropy encoding on the speech signal obtained by the vector quantization;
A determination unit that determines whether or not the data amount of the audio signal obtained by the entropy encoding unit is larger than a preset target data amount;
When the determination unit determines that the data amount of the audio signal obtained by the entropy encoding unit is larger than the target data amount, the vector quantization-encoded audio signal has a minimum energy band. A data deleting unit that performs a first deleting process that deletes a frequency component and a second deleting process that deletes a frequency component whose frequency component adjacent to the frequency axis is 0 and whose amplitude value is larger than a predetermined value. When,
A control unit that performs entropy encoding again on the audio signal whose data has been deleted by the data deletion unit, and performs the deletion process and the encoding process until the data amount of the encoded audio signal falls within the target data amount,
An audio processing apparatus comprising:   The speech processing apparatus according to claim 1, wherein, in the second deletion process of the data deletion unit, the predetermined value is a value calculated from a maximum value of amplitude values of all frequency components. The data deletion unit determines whether the value of the high frequency component of the audio signal obtained by the frequency conversion unit is greater than the value of the low frequency component, and the value of the high frequency component of the audio signal is the value of the low frequency component If it is determined to be equal to or less than, it deletes the high frequency components present further than preset frequency intervals on the frequency axis, claim 1, characterized in that performing the second deletion processing or 2. The speech processing apparatus according to 2. A framing unit that divides the input audio signal into frames;
For each frame obtained by the framing unit, a level adjusting unit that adjusts the level of the audio signal based on the maximum value of the amplitude of the audio signal included in the frame;
A frequency conversion unit that performs frequency conversion on the audio signal whose level is adjusted by the level adjustment unit;
A vector quantization unit that performs vector quantization on the audio signal obtained by the frequency conversion;
An entropy encoding unit that performs entropy encoding on the speech signal obtained by the vector quantization;
A determination unit that determines whether or not the data amount of the audio signal obtained by the entropy encoding unit is larger than a preset target data amount;
When the determination unit determines that the data amount of the audio signal obtained by the entropy encoding unit is larger than the target data amount, the vector quantization-encoded audio signal has a minimum energy band. A data deletion unit that performs a first deletion process of deleting a frequency component, and a second deletion process of deleting a frequency component that exists at a predetermined frequency interval or more apart on the frequency axis;
A control unit that performs entropy encoding again on the audio signal from which data has been deleted by the data deletion unit, and performs the deletion process and the encoding process until the data amount of the encoded audio signal falls within the target data amount;
An audio processing apparatus comprising: The data deletion unit determines whether the value of the high frequency component of the audio signal obtained by the frequency conversion unit is greater than the value of the low frequency component, and the value of the high frequency component of the audio signal is the value of the low frequency component 5. The audio processing according to claim 4 , wherein, when it is determined that the frequency component is equal to or less than the frequency component, a frequency component that exists at a predetermined frequency interval or more on the frequency axis is deleted from the high-frequency component. apparatus. The data deletion unit calculates the logarithm of each frequency component in the audio signal obtained by the frequency conversion unit, and compares the sum of the logarithm of the high-frequency component with the sum of the logarithm of the low-frequency component. audio processing apparatus according to claim 3 or 5 the value of the high frequency component and judging a greater or not than the value of the low frequency components of the. The data deleting unit separates the audio signal whose level is adjusted by the level adjusting unit into a high-frequency component and a low-frequency component by a high-pass filter and a low-pass filter, and the high-pass filter and the low-pass filter audio processing apparatus according to claim 3 or 5 obtained value of the high-frequency component and judging a greater or not than the value of the low frequency components by. Divide the input audio signal into frames,
For each frame, adjust the level of the audio signal based on the maximum amplitude of the audio signal included in the frame,
Frequency conversion is applied to the audio signal whose level is adjusted,
Apply vector quantization to the audio signal obtained by the frequency conversion,
Entropy coding is performed on the speech signal obtained by the vector quantization,
It is determined whether the data amount of the audio signal obtained by the entropy encoding is larger than a preset target data amount,
If it is determined that the data amount of the audio signal obtained by the entropy encoding is larger than the target data amount , the frequency component of the band with the minimum energy is deleted from the vector quantized audio signal , Delete frequency components whose frequency components adjacent to each other on the frequency axis are 0 and whose absolute value of the amplitude value is larger than a predetermined value ,
A speech encoding method, wherein the speech signal from which data has been deleted is entropy-encoded again, and the deletion process and the encoding process are performed until the data amount of the encoded speech signal falls within the target data amount . Divide the input audio signal into frames,
For each frame, adjust the level of the audio signal based on the maximum amplitude of the audio signal included in the frame,
Frequency conversion is applied to the audio signal whose level is adjusted,
Apply vector quantization to the audio signal obtained by the frequency conversion,
Entropy coding is performed on the speech signal obtained by the vector quantization,
It is determined whether the data amount of the audio signal obtained by the entropy encoding is larger than a preset target data amount,
If it is determined that the data amount of the audio signal obtained by the entropy encoding is larger than the target data amount , the frequency component of the band with the minimum energy is deleted from the vector quantized audio signal , Deletes frequency components that are separated by more than a preset frequency interval on the frequency axis ,
A speech encoding method, wherein the speech signal from which data has been deleted is entropy-encoded again, and the deletion process and the encoding process are performed until the data amount of the encoded speech signal falls within the target data amount .

Images