WO2004040552A1

WO2004040552A1 - Transcoder and coder conversion method

Info

Publication number: WO2004040552A1
Application number: PCT/JP2003/012859
Authority: WO
Inventors: Kazunori Ozawa
Original assignee: Nec Corporation
Priority date: 2002-10-31
Filing date: 2003-10-08
Publication date: 2004-05-13
Also published as: CN1708786A; EP1564723A1; CA2504174A1; AU2003271119A1; EP1564723A4; KR100715014B1; JP4438280B2; CN100498933C; JP2004151424A; DE60321712D1; KR20050061579A; HK1077913A1; EP1564723B1

Abstract

An inter-conversion transcoder includes: a spectrum parameter calculation circuit (100) for calculating a spectrum parameter for a signal x(n) obtained by decoding a first code; a coefficient calculation circuit (130) for receiving a spectrum parameter and converting it into a coefficient of the signal subjected to band extension; a noise generation circuit (120) for outputting a noise signal whose band is limited; a gain circuit (140) for multiplying the output signal of the noise generation circuit by a gain; a combining filter circuit (170) for receiving the output signal from the noise generation circuit (120) and the coefficient from the coefficient calculation circuit (130) and outputting a high-band signal y(n) for band extension; a sampling frequency conversion circuit (180) for outputting a signal s(n) which is the signal x(n) up-sampled to a predetermined sampling frequency; an adder (190) for adding the high-band signal y(n) and the signal s(n) to form a band-extension signal z(n); and a second encoding circuit (195) for encoding the band-extension signal z(n) by the second encoding method and outputting the result.

Description

Technical Field Transcoder and transcoding method

The present invention relates to a transcoder for mutually converting a code according to a first encoding method and a code according to a second encoding method, and when converting a first code into a second code, a frequency band of a signal. Transcoder that expands and converts

Background art

A method is known in which the frequency band of an audio signal encoded and reproduced at a low bit rate is extended on the receiving side without transmitting auxiliary information for band extension from the transmitting side (for example, see Non-Patent Document 1). ).

Non-Patent Document 1: P. Jax, P. Vary, "Wideband extension of telephone speech using hidden markov model," Pro IEEE Speech Coding Workshop, pp. 133-13, 2000

In the conventional method described in the above document 1, the receiving side searches for filter coefficients after band expansion using an HMM (Hidden Markov Model).

On the other hand, in a transcoder that performs mutual conversion between a code using the first coding method and a code using the second coding method, the band of the signal is extended when converting from the first code to the second code. There is no precedent for transcoders that convert after conversion.

The conventional method described in the above-mentioned document by P. Jax and P. Vary (Non-Patent Document 1) requires modeling of the spectral envelope of wideband speech and the filter coefficients by HMM, as follows. There were serious problems.

In other words, it is necessary to determine the parameters of the HMM model in advance from a large amount of speech database offline, which requires a great deal of calculation time and cost.

Also, when performing band extension processing in real time on the receiving side, a search using the H This required a search, which required a lot of computation. Disclosure of the invention

Therefore, a main object of the present invention is to provide a transcoder for mutually converting between a code according to the first encoding method and a code according to the second encoding method, when converting from the first code to the second code. It is another object of the present invention to provide a transcoder and a code conversion method capable of performing band expansion conversion with good sound quality with a relatively small amount of operation when converting a signal after expanding the signal band.

According to one aspect, a transcoder according to the present invention is a transcoder for mutually converting between a code according to a first encoding method and a code according to a second encoding method, and according to the first aspect. For example, a spectrum parameter calculation unit that receives a code coded by the first coding method and decodes the data by the first coding method to calculate a spectrum parameter representing the spectrum characteristics. A noise generation unit that generates a noise signal; a coefficient calculation unit that calculates a fill coefficient after shifting the frequency of the spectrum parameter; and a gain unit that provides an appropriate gain to the output of the noise generation unit A synthesis filter unit that passes an output of the gain unit through a synthesis filter configured using the coefficient to reproduce a band-extended signal; and converts a sampling frequency of the input signal, and An adder for adding and outputting the output signal of the data section, is configured to output a second code by encoding the output signals of the adders second encoding method. According to a second aspect, the transcoder according to the present invention is a transcoder for mutually converting between a code according to a first coding scheme and a code according to a second coding scheme, A spectrum parameter calculation unit that receives a code coded by the coding scheme, decodes the code by the first coding scheme, and calculates a spectrum parameter representing a spectrum characteristic; and a pitch period from the input signal. And an adaptive codebook section for generating an adaptive codebook component based on the pitch period and a past sound source signal; a noise generating section for generating a noise signal; and shifting the frequency of the spectral parameter over and over. A coefficient calculation unit for obtaining a filter coefficient in step (1), and an appropriate gain given to at least one of the output of the noise generation unit and the output of the adaptive codebook unit, followed by adding A gain section for outputting a No., synthetic fill constructed using the coefficients A synthesis filter for inputting the sound source signal in the evening and reproducing the band extension signal; an adder for converting a sampling frequency of the reproduction signal and adding an output signal of the synthesis filter to output the converted signal; The output signal of the adder is encoded by a second encoding method to output a second code.

According to a third aspect, a transcoder according to the present invention is a transcoder for mutually converting between a code according to a first encoding scheme and a code according to a second encoding scheme, A spectrum parameter calculation unit that receives a code coded according to a coding scheme, decodes the code according to a first coding scheme, and calculates a spectrum parameter representing a spectrum characteristic, and calculates a pitch period from the input signal. An adaptive codebook unit that calculates and generates an adaptive codebook component based on the pitch period and a past sound source signal; a noise generation unit that generates a noise signal; and a frequency shift of the spectral parameter. A coefficient calculation unit for obtaining a filter coefficient; and an appropriate gain given to at least one of the output of the noise generation unit and the output of the adaptive codebook unit, followed by addition. And a gain section that outputs a signal, a pitch pre-filter is applied to the sound source signal using the pitch period, and the pitch pre-filter output signal is input to a synthesis filter configured using the coefficients to generate a band extension signal. A synthesis filter section for reproducing, an adder for converting the sampling frequency of the reproduction signal, and adding and outputting an output signal of the synthesis filter section, and a second encoding method for outputting the output signal of the adder. And outputs the second code.

Further, according to the present invention, the output of the adaptive codebook section may be passed through a low-pass filter having a predetermined power-off frequency.

Further, according to the present invention, a post-filter is configured using a weighting coefficient obtained by weighting the coefficient, and a band extension signal is reproduced by passing an output signal of the synthesis filter unit through the BOST filter. Good.

According to one aspect of the method according to the present invention, in a transcoding method of a transcoder for mutually converting a code according to a first encoding method and a code according to a second encoding method, Decoding the code coded according to the first decoding method according to the first decoding method and outputting a decoded signal;

Calculate and output spectrum parameters representing spectrum characteristics from the decoded signal Steps to

Shifting the frequency of the spectral parameter and calculating and outputting a filter coefficient;

Applying a gain to the output signal from the noise generator;

Passing the output signal with the gain through a synthesis filter configured using the filter coefficient, and outputting a signal in a band necessary for band conversion;

Adding a signal obtained by converting the decoded signal at a predetermined sampling frequency and an output signal of the synthesis filter;

Encoding the result of the addition using the second encoding method to obtain and output a second code.

Alternatively, according to another aspect of the method of the present invention, a step of decoding a code coded by the first coding method by a first decoding method and outputting a decoded signal;

Calculating and outputting a spectrum parameter representing a spectrum characteristic from the decoded signal;

Calculating a pitch period from the decoded signal and generating an adaptive codebook component based on the pitch period and a past excitation signal;

Outputting a sound source signal by adding a gain to at least one of the output of the noise generator and the adaptive codebook component and adding the gain;

Outputting a signal of a band necessary for passband conversion of the sound source signal to a synthesis filter configured using the filter coefficient;

Alternatively, according to another aspect of the method of the present invention, a code encoded by the first encoding method is decoded by a first decoding method, and a decoded signal is output. Steps

Performing a pitch prefill on the sound source signal using the pitch period;

Outputting a signal of a band necessary for band conversion through a signal subjected to the pitch prefilling process to a synthetic filter formed by using the filter coefficient; and Adding the converted signal and the output signal of the composite filter;

According to another aspect of the method according to the present invention, the sound source signal from the gain section is subjected to a prefilling process using the pitch period in a pitch prefilling process, and The method may further include a step of inputting an output signal from the pitch prefill.

According to another aspect of the method according to the present invention, the output signal of the synthesis filter unit is added to a post-filter configured using a weighting coefficient obtained by weighting the filter coefficient from the counting calculation unit. May be input. According to another aspect of the method according to the present invention, an output of a periodic signal generating unit that generates a periodic signal using a pitch period is supplied to a gain unit instead of an output signal from the adaptive codebook unit. You may.

According to the present invention, when a code according to the first encoding method is input, converted to a code according to the second encoding method, and output, the signal is expanded after expanding the band of the signal. Performance A high-frequency signal is generated by processing with a small amount of computation, and a sampling signal of a narrow-band input signal is added to the converted signal to generate a band-extended signal (for example, a 7 kHz band). The effect is that

Also, according to the present invention, an adaptive codebook signal is generated using a delay calculated from a narrowband input signal based on a past sound source signal in a high frequency part, and is multiplied by an appropriate gain to generate a noise signal. By adding the above, there is an effect that a band extension signal with good sound quality can be generated when periodicity is required for a high-frequency portion signal such as a vowel.

Furthermore, according to the present invention, a pitch prefilter is used for a sound source signal using a delay, or a post-filter is provided by weighting a coefficient from a coefficient calculation circuit to provide a post-filter. This has the effect of generating a band-extended signal with excellent sound quality. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a diagram showing a configuration of a first embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of the second exemplary embodiment of the present invention.

FIG. 3 is a diagram showing a configuration of a third exemplary embodiment of the present invention.

FIG. 4 is a diagram showing a configuration of a fourth embodiment of the present invention.

FIG. 5 is a diagram showing a configuration of a fifth embodiment of the present invention.

FIG. 6 is a diagram showing a configuration of a modification of the second embodiment of the present invention. BEST MODE FOR CARRYING OUT THE INVENTION

Embodiments will be described with reference to the drawings in order to explain the present invention in more detail. In the following, the first code is coded for a narrow band input signal in the 4 kHz band, and the transcoder converts this to a signal in the 5 kHz band or 7 kHz band. It is assumed that the band is extended and then the second code is used to obtain the second code.

FIG. 1 is a block diagram showing a configuration of a first embodiment of a transcoder according to the present invention. Referring to FIG. 1, the transcoder includes a first decoding circuit 105 Addition of spectrum parameter calculation circuit 100, noise generation circuit 120, coefficient calculation circuit 130, synthesis filter circuit 170, sampling frequency conversion circuit 180 , A second encoding circuit 195, a voiced / unvoiced discriminating circuit 200, a gain adjusting circuit 310, and a gain circuit 140.

The first decoding circuit 105 receives a code according to the first encoding method, decodes the code according to the first decoding method, and outputs a decoded signal x (n).

The spectral parameter calculation circuit 100 divides the decoded signal x (n) into frames (for example, 10 ms), and calculates a spectrum parameter of a predetermined order P for each frame. Here, the spectral parameter is a parameter that represents the spectral outline of the audio signal for each frame, and a well-known LPC (linear predictive coding) analysis or the like can be used for this calculation. Further, the spectral parameter overnight calculation unit 100 converts the linear prediction coefficients Q; i (i = l,... P) calculated by the LPC analysis into LSPs (Line Spectrum Pairs) suitable for quantization and interpolation. : Line spectrum pair) Converted to parameters and output. Here, for the conversion from the linear prediction coefficient to LSP, for example, the following paper is referred to (Non-Patent Document 2).

Non-Patent Document 2: Sugamura, Itakura "Speech Information Compression Using Line Spectrum Pair (LSP) Speech Analysis and Synthesis", IEICE Transactions, J64_A, pp. 599-606, 1981

The coefficient calculation circuit 130 receives the spectral parameter data output from the spectrum parameter calculation circuit 100 and converts it into a coefficient of a band-extended signal. For this conversion, for example, well-known methods such as a method of simply shifting the frequency of LSP to a higher frequency, a nonlinear conversion method, and a linear conversion method can be used. Here, using all or part of the LSP parameters, the existing frequency band of the LSP is shifted to a higher frequency band, and then converted to a linear prediction coefficient of order P. Output to

The noise generation unit 120 generates a noise signal with the average amplitude normalized to a predetermined level, a band-limited noise signal for a time length equal to the frame length, and outputs the noise signal to the gain circuit 140. In this embodiment, white noise is used as an example of the noise signal, but another noise signal may be used.

The voiced / unvoiced discrimination circuit 200 inputs the narrowband input signal x (n) and outputs the signal for each frame. Determine if the issue is voiced or unvoiced. As a voiced / unvoiced determination, for example, for a narrowband input signal x (n), a normalized autocorrelation function D (T) up to a predetermined delay time m is calculated according to equation (1). The maximum value of (T) is obtained, and if the maximum value of D (T) is larger than a predetermined threshold value, it is determined that the voiced voice is used.

D (T) = x (n) x (n-Γ)] / [£ χ ² (η-Γ)] (1)

Then, the voiced / unvoiced discrimination circuit 200 outputs the voiced / unvoiced discrimination information to the gain adjustment circuit 210. In equation (1), N is the number of samples for calculating the normalized autocorrelation.

The gain adjustment circuit 310 inputs voiced / unvoiced discrimination information from the voiced / unvoiced discrimination circuit 200, adjusts the gain given to the noise signal according to voiced / unvoiced, and outputs the same to the gain circuit 140. .

The gain circuit 140 inputs the gain from the gain adjustment circuit 310, multiplies the output signal of the noise generation circuit 120 by the gain, and outputs the result to the synthesis filter circuit 170. The synthesis filter circuit 170 inputs the output signal from the gain circuit 140, and further inputs a coefficient of a predetermined order from the coefficient calculation section 130 to configure a filter, thereby expanding the band. It outputs the high-frequency signal y (n) necessary for the conversion.

The sampling frequency conversion circuit 180 up-samples the narrowband input signal x (n) to a predetermined sampling frequency, and outputs an up-sampled signal s (n).

The adder 190 adds the output signal y (n) of the synthesis filter circuit 170 to the output signal s (n) of the sampling frequency conversion circuit 180, and finally a signal whose band is extended. Form z (n) and output.

The second encoding circuit 195 receives the output signal z (n) of the adder 190, performs encoding by a second encoding method, obtains a second code, and outputs the second code.

This concludes the description of the first embodiment.

FIG. 2 is a block diagram showing a configuration of the second exemplary embodiment of the present invention. Figure 2 For reference, the transcoder according to the second embodiment of the present invention includes a first decoding circuit 105, a spectrum parameter calculation circuit 100, an adaptive codebook circuit 110, and a noise generation circuit 1 2 0, coefficient calculation circuit 130, gain circuit 340, synthesis filter circuit 170, sampling frequency conversion circuit 180, adder 160, adder 190, A second encoding circuit 195, a voiced / unvoiced discrimination circuit 200, and a gain adjustment circuit 210. In FIG. 2, the same elements as those in FIG. 1 are denoted by the same reference numerals. Hereinafter, differences from the first embodiment will be mainly described, and descriptions of the same elements as those in FIG. 1 will be omitted as appropriate. Referring to FIG. 2, the second embodiment of the present invention includes an adaptive code book circuit 110 and an adder 160 in addition to the configuration of FIG.

The voiced / unvoiced discriminating circuit 200 receives the narrow-band input signal x (n) and discriminates whether the signal for each frame is voiced or unvoiced. As a voiced / unvoiced determination, for example, for a narrowband input signal x (n), a normalized autocorrelation function D (T) up to a predetermined delay time m is calculated according to the above equation (1). The maximum value of D (T) is obtained, and if the maximum value of D (T) is larger than a predetermined threshold, it is determined as voiced, otherwise it is determined to be unvoiced, and the result of the determination is determined by the gain adjustment circuit 210. Output to

The voiced / unvoiced discrimination circuit 200 supplies the value of T that maximizes the normalized self-correlation function D (T) to the adaptive codebook circuit 110 as the pitch period T in the voiced frame. .

The adaptive codebook circuit 110 inputs the adaptive codebook delay T from the voiced / unvoiced discrimination circuit 200 and adapts it based on the past sound source signal v (n) according to the following equation (2). Generates and outputs the code vector p (n). p (n) = v (n-T) (2) The gain adjustment circuit 210 inputs voiced / unvoiced discrimination information from the voiced / unvoiced discrimination circuit 200, and adapts the adaptive code according to whether it is voiced or unvoiced. The gain of the book signal and the gain of the noise signal are adjusted and supplied to the gain circuit 340.

The gain circuit 340 inputs the gain from the gain adjustment circuit 210 and the adaptive code The output signal of at least one of the book circuit 110 and the noise generation circuit 120 is multiplied by a gain and output to the adder 160.

The adder 160 outputs two types of signals output from the gain circuit 340 (two signals obtained by multiplying at least one output signal of the adaptive code book circuit 110 and the noise generation circuit 120 by a gain). And outputs the result to the synthesis filter circuit 170 and the adaptive codebook circuit 110.

The synthesis filter circuit 170 inputs the output signal of the adder 160 and further inputs a predetermined order coefficient (filter coefficient) from the coefficient calculator 130 to form a filter. Outputs the signal y (n) in the high frequency range necessary for band extension.

According to the second embodiment of the present invention, an adaptive codebook signal is generated using a delay calculated from a narrowband input signal based on a past sound source signal in a high frequency part, and is multiplied by an appropriate gain. By adding to the noise signal, a band extension signal with good sound quality can be generated when periodicity is required for a high frequency signal such as a vowel. This concludes the description of the second embodiment.

As a modification of the second embodiment of the present invention, as shown in FIG. 6, a structure including a periodic signal generation circuit 115 instead of the adaptive code book circuit 110 of FIG. It is good also as a result. The periodic signal generating circuit 115 receives the pitch period from the voiced / unvoiced discriminating circuit 200, generates a periodic signal using the pitch period, and outputs the periodic signal to the gain circuit 340. The configuration other than the periodic signal generation circuit 115 is the same as that of the second embodiment.

FIG. 3 is a block diagram showing a configuration of the third exemplary embodiment of the present invention. Referring to FIG. 3, a transcoder according to a third embodiment of the present invention includes a first decoding circuit 105, a spectrum parameter calculation circuit 100, and an adaptive codebook circuit 110. , A noise generation circuit 120, a coefficient calculation circuit 130, a gain circuit 300, a synthesis filter circuit 170, a sampling frequency conversion circuit 180, and an adder 190. , A second encoding circuit 195, a voiced / unvoiced discrimination circuit 200, a gain adjustment circuit 210, and a pitch pre-filter circuit 400. In FIG. 3, the same or equivalent elements as those in FIGS. 1 and 2 are denoted by the same reference numerals. In the following, the differences from the second embodiment will be mainly described, and the same as FIGS. 1 and 2 will be described. Description of the elements is omitted. In the present embodiment, a pitch prefill circuit 400 is provided.

The gain circuit 300 inputs the gain from the gain adjustment circuit 210, multiplies the output signal of the adaptive codebook circuit 110 and the noise generation circuit 120 by the gain, and adds and adds two types of signals. The result is output to the pitch pre-filter circuit 400.

The pitch pre-filter circuit 400 receives the delay T (pitch cycle) from the voiced / unvoiced discrimination circuit 200 and receives the sound source signal v (n) from the gain circuit 300 to obtain the following equation (3). And then output to the composite fill circuit 170 after performing the pitch prefilling.

According to this embodiment, by using the pitch pre-filter circuit 400 for the sound source signal using the delay, it is possible to generate a band extension signal having good sound quality. This concludes the description of the third embodiment.

In this embodiment, as in the modification of the second embodiment, a periodic signal generation circuit may be used instead of the adaptive codebook circuit 110. In this case, the periodic signal generation circuit inputs a signal from the voiced / unvoiced discrimination circuit 200, calculates a pitch period, generates a periodic signal based on the pitch period, and outputs the signal to the gain circuit 300.

FIG. 4 is a block diagram showing a configuration of a fourth embodiment of the present invention. Referring to FIG. 4, a transcoder according to a fourth embodiment of the present invention includes a first decoding circuit 105, a spectral parameter overnight calculation circuit 100, and an adaptive codebook circuit 110. , A noise generation circuit 120, a coefficient calculation circuit 130, a gain circuit 340, an adder 160, a synthesis filter circuit 170, and a sampling frequency conversion circuit 180. , An adder 190, a second encoding circuit 195, a voiced / unvoiced discrimination circuit 200, a gain adjustment circuit 210, and a low-pass filter circuit 500. Have. In FIG. 4, the same or equivalent elements as those in FIG. 2 are denoted by the same reference numerals. Less than In the following, a description will be given mainly of a difference from the second embodiment, and a description of the same elements as those in FIG. 2 will be omitted. Referring to FIG. 4, a low-pass filter circuit 500 having an output of the adaptive codebook circuit 110 as an input is provided.

The low-pass filter (LPF) circuit 500 passes the output signal of the adaptive code book circuit 110 to the gain circuit 340 after passing the low-pass signal according to equation (4) according to equation (4). I do.

p '(n) = (n) ¾ (n) (4)

The cut-off frequency of the low-pass filter circuit 500 is determined in advance, and may be, for example, 6 kHz. In equation (4), h (n) indicates the impulse response of the low-pass filter, and the symbol "" indicates the convolution operation.

This concludes the description of the fourth embodiment of the present invention. In the fourth embodiment of the present invention, as in the modification of the second embodiment, a periodic signal generation circuit can be used instead of the adaptive codebook circuit 110. In this case, a signal is input from the voiced / unvoiced discriminating circuit 200 to calculate a pitch period, a periodic signal based on the pitch period is generated, and output to the gain circuit 340.

FIG. 5 is a block diagram showing a configuration of a fifth exemplary embodiment of the present invention. Referring to FIG. 5, a transcoder according to a fifth embodiment of the present invention includes a first decoding circuit 105, a spectral parameter overnight calculation circuit 100, and an adaptive codebook circuit 110. , A noise generation circuit 120, a coefficient calculation circuit 130, a gain circuit 300, a composite filter circuit 170, a sampling frequency conversion circuit 180, and an adder 190. , A second encoding circuit 195, a voiced / unvoiced discrimination circuit 200, a gain adjustment circuit 210, a pitch pre-filter 400, and a post-filter 600. In FIG. 5, the same or equivalent elements as those in FIG. 3 are denoted by the same reference numerals. In the following, differences from the third embodiment will be mainly described, and description of the same elements as those in FIG. 3 will be omitted. In the present embodiment, a post-fill 600 is further provided in addition to the configuration of the third embodiment.

The post-filter 600 inputs the coefficient (filter coefficient) from the coefficient calculation circuit 130, weights the coefficient, performs post-filtering according to equation (5), and outputs the output. Output to 0. y ′ (n) = y (n) −∑ a; ri ¹ y (ni) + ∑ a) r ₂ ¹ (nii) (5) According to the present embodiment, the post filter 600 is used. As a result, a band extension signal with good sound quality can be generated. This concludes the description of the fifth embodiment.

Note that, also in the fifth embodiment of the present invention, a periodic signal generation circuit can be used instead of the adaptive codebook circuit 110, similarly to the modification of the second embodiment. In this case, a signal is input from the voiced / unvoiced discriminating circuit 200 to calculate the pitch period, generate a periodic signal based on the pitch period, and output it to the gain circuit 300.

The configurations of the respective embodiments may be combined, such as using the postfill described in the fifth embodiment in the first embodiment. Although the present invention has been described with reference to the above embodiments, the present invention is not limited to only the configuration of the above embodiments, but may be embraced within the scope of the claims. Needless to say, it includes various variations and modifications that can be made by those skilled in the art. Industrial applicability

As described above, according to the present invention, when converting from a code of the first encoding method to a code of the second encoding method, it is possible to generate a band extension signal with good sound quality, And the like.

Claims

1. In a transcoder that mutually converts a code according to the first encoding method and a code according to the second encoding method,

A first decoding unit that inputs a code encoded by the first encoding method, performs decoding by the first decoding method, and outputs a decoded signal;

A spectrum parameter calculation unit that receives the decoded signal, calculates and outputs spectrum parameters representing spectrum characteristics, and

A noise generator for generating a noise signal;

A coefficient calculator for calculating and outputting a filter coefficient after shifting the frequency of the spectrum parameter,

A gain unit for giving a gain to the output signal from the noise generation unit and outputting the gain signal; and a synthesis filter configured using the filter coefficient from the coefficient calculation unit, and receiving the output signal from the gain unit. A synthesis filter unit that outputs a signal in a band necessary for band conversion through the synthesis filter,

A sampling frequency conversion circuit that outputs a signal obtained by converting the decoded signal at a predetermined sampling frequency,

An adder that adds and outputs an output signal of the sampling frequency conversion circuit and an output signal of the synthesis filter unit;

A second encoding unit that receives an output signal of the adder, performs encoding by the second encoding method, obtains and outputs a second code,

A transcoder, comprising:

2. In a transcoder for mutually converting a code according to the first encoding method and a code according to the second encoding method,

A spectrum parameter calculation unit that receives the decoded signal and calculates spectrum parameters representing spectrum characteristics;

Calculating a pitch period from the decoded signal, based on the pitch period and a past sound source signal; An adaptive codebook section for generating an adaptive codebook component; a noise generating section for generating a noise signal;

A coefficient calculator for calculating a filter coefficient after shifting the frequency of the spectrum parameter,

An output signal of the noise generation unit, and a gain unit for adding a gain to at least one of the output signals of the adaptive codebook unit and adding the resultant to output a sound source signal;

A synthesis filter configured to use the filter coefficient from the coefficient calculation unit, inputting a sound source signal from the gain unit, passing the signal through the synthesis filter, and outputting a signal in a band required for band conversion; Phil Yube and

An adder for adding and outputting the output signal of the sampling frequency conversion circuit and the output signal of the synthesis filter unit;

A transcoder, comprising:

3. In a transcoder for mutually converting between a code according to the first encoding method and a code according to the second encoding method,

An adaptive codebook unit that calculates a pitch period from the decoded signal and generates an adaptive codebook component based on the pitch period and a past excitation signal;

A noise generator for generating a noise signal;

A gain section for adding a gain to at least one of the output signal of the noise generation section and the output signal of the adaptive codebook section and adding the resultant to output a sound source signal; The pitch period is used to pre-fill the sound source signal from the gain section.

A composite filter configured using the filter coefficient from the coefficient calculator, and outputting a signal of a band necessary for band conversion by passing an output signal of the pitch prefill filter to the composite filter; With synthetic fill evening part,

A transcoder, comprising:

4. In a transcoder for mutually converting a code according to the first encoding method and a code according to the second encoding method,

A spectrum parameter overnight calculation unit that receives the decoded signal and calculates a spectrum parameter representing a spectrum characteristic,

A periodic signal generation unit that calculates a pitch period from the decoded signal and generates a periodic signal using the pitch period;

A noise generator for generating a noise signal;

A coefficient calculator for calculating a filter coefficient after shifting the frequency of the spectrum parameter;

A gain unit for adding a gain to at least one of the output signal of the noise generation unit and the output signal of the periodic signal generation unit and adding the resultant to output a sound source signal;

A synthesis filter configured using the filter coefficient from the coefficient calculator; inputting a sound source signal from the gain unit to output a signal in a band necessary for band conversion through the synthesis filter; A synthesis filter unit;

Outputting a signal obtained by converting the decoded signal at a predetermined sampling frequency A sampling frequency conversion circuit,

A transcoder, comprising:

5. In the transcoder for mutually converting between the code according to the first encoding method and the code according to the second encoding method,

A first decoding unit that inputs a code coded by the first coding method, performs decoding by the first decoding method, and outputs a decoded signal;

A noise generator for generating a noise signal;

A coefficient calculator for calculating a fill coefficient after shifting the frequency of the spectrum parameter,

A pre-fill for the sound source signal from the gain unit using the pitch period; and a synthesis filter configured using the filter coefficient from the coefficient calculation unit. The output of the pitch pre-filter during the synthesis filter A synthesis filter section that passes a signal and outputs a signal in a band necessary for band conversion,

The output signal of the adder is input and encoded by the second encoding method. A second encoding unit that obtains and outputs the code of

A transcoder, comprising:

6. A low-pass filter for receiving an output signal of the adaptive codebook section, passing a signal having a frequency equal to or lower than a predetermined cut-off frequency, and outputting the signal. 4. The transcoder according to paragraph 2 or 3.

7. A post filter configured using a weighting coefficient obtained by weighting the filter coefficient from the counting calculation unit,

The output signal of the synthesis filter unit is passed through the post filter to reproduce a band-converted signal,

The said adder adds the output signal of the said post filter and the output signal of the said sampling frequency conversion circuit instead of the output signal of the said synthesis filter part, and outputs it. 7. The transcoder according to any one of items 1 to 6.

8. The adaptive codebook unit receives the decoded signal from the first decoding unit, and outputs a pitch period from a voiced / unvoiced determination circuit that outputs voiced / unvoiced determination information and pitch period information; 4. The transcoder according to claim 2, further comprising an adaptive codebook circuit that inputs a sound source signal input to the filter section.

9. The periodic signal generation unit receives a pitch period from a voiced / unvoiced determination circuit that inputs the decoded signal from the first decoding unit and outputs voiced / unvoiced determination information and pitch period information. The transcoder according to claim 4 or 5, further comprising a periodic signal generation circuit.

10. The pitch pre-filter receives the decoded signal from the first decoding unit and inputs a pitch period from a voiced / unvoiced discriminating circuit that outputs voiced / unvoiced discrimination information and pitch period information. 6. The transcoder according to claim 3, wherein a pitch pre-filling is performed on a sound source signal from the gain unit, and then the signal is output to the synthesis filter unit.

11. A voiced / unvoiced discriminating circuit that inputs the decoded signal from the first decoding unit and outputs voiced / unvoiced discrimination information, And a gain adjustment circuit that inputs voiced / unvoiced discrimination information from the voiced / unvoiced discrimination unit, and adjusts a gain of an output signal from the noise signal generation unit according to whether the voiced / unvoiced discrimination is performed.

2. The gain control unit according to claim 1, wherein the gain unit includes a gain circuit that receives a gain signal from the gain adjustment circuit and applies a gain to an output signal from the noise generation unit. Transcoder.

12. A voiced / unvoiced discrimination circuit that inputs the decoded signal from the first decoding unit and outputs voiced / unvoiced discrimination information and pitch period information,

A gain adjustment circuit that inputs voiced / unvoiced discrimination information from the voiced / unvoiced discrimination unit, and adjusts the gain of the adaptive codebook signal and the output signal from the noise generation unit according to whether it is voiced or unvoiced;

With

A gain circuit that receives a gain signal from the gain adjustment circuit, multiplies at least one of output signals of the adaptive codebook unit and the noise generation unit by a gain, and outputs the multiplied signal;

A second adder that outputs two signals corresponding to the output signal of the adaptive codebook unit and the output signal of the noise generation unit, respectively, and that is output from the gain circuit.

4. The transcoder according to claim 2, wherein an output signal of the second adder is supplied to the synthesis filter section and the adaptive codebook section.

13. A voiced / unvoiced discrimination circuit that receives the decoded signal from the first decoding unit and outputs voiced / unvoiced discrimination information and pitch period information.

A voice / unvoiced discriminating information from the voiced / unvoiced discriminating unit, and a gain adjusting circuit that adjusts a gain of an output signal from the periodic signal generating unit and the noise generating unit according to whether the voiced / unvoiced is determined,

With

The gain unit receives a gain signal from the gain adjustment circuit, and outputs a signal obtained by multiplying at least one output signal of the periodic signal generation unit and the noise generation unit by a gain. Gain circuit

A second adder for adding and outputting two types of signals respectively corresponding to the output signals of the periodic signal generation unit and the noise generation unit, which are output from the gain circuit,

6. The transcoder according to claim 4, wherein an output signal of the second adder is supplied to the synthesis filter section.

1 4. The coefficient calculation unit shifts the frequency of the spectral parameter to a higher frequency to obtain a filter coefficient,

The synthesis filter unit reproduces a signal whose band has been extended,

2. The sampling frequency conversion circuit according to claim 1, wherein the decoded signal from the first decoding unit is input, up-sampled to a predetermined sampling frequency, and output. A transcoder according to any one of paragraphs 13 to 13.

15. The noise generator, wherein the average amplitude is normalized to a predetermined level, and the noise generator generates a band-limited noise signal for a time length equal to a frame length. Item 15. The transcoder according to any one of Items 13 to 13.

16. The voiced / unvoiced discrimination circuit calculates a normalized autocorrelation function for the decoded signal up to a predetermined delay time, obtains a maximum value of the normalized autocorrelation function, and calculates the maximum value. Is determined as voiced if it is larger than a predetermined threshold value, and is otherwise determined as unvoiced, and outputs a determination result to the gain adjustment circuit as the voiced / unvoiced determination information. The transcoder according to item 1.

17. The voiced / unvoiced discriminating circuit calculates a normalized autocorrelation function of the decoded signal up to a predetermined delay time, obtains a maximum value of the normalized autocorrelation function, and obtains the maximum value. Is larger than a predetermined threshold value, it is determined as voiced, otherwise it is determined as unvoiced, and the determination result is output to the gain adjustment circuit as the voiced / unvoiced determination information. 13. The transcoder according to claim 12, wherein a delay value for maximizing an autocorrelation function is supplied to the adaptive codebook unit as a pitch period.

18. The voiced / unvoiced discrimination circuit calculates a normalized autocorrelation function of the decoded signal up to a predetermined delay time, obtains a maximum value of the normalized autocorrelation function, and obtains the maximum value. Is larger than a predetermined threshold value, it is determined as voiced, otherwise it is determined as unvoiced, and the determination result is output to the gain adjustment circuit as the voiced / unvoiced determination information. The transcoder according to claim 13, wherein a delay value maximizing an autocorrelation function is supplied to the periodic signal generation unit as a pitch period.

19. A transcoder code conversion method for mutually converting a code according to the first coding method and a code according to the second coding method,

Decoding the code encoded by the first encoding method by a first decoding method and outputting a decoded signal;

Calculating and outputting spectral parameters representing spectral characteristics from the decoded signal;

Calculating and outputting a filter coefficient after shifting the frequency of the spectral parameter;

Applying a gain to the output signal from the noise generator;

Encoding the addition result by the second encoding method to obtain and output a second code;

A code conversion method, comprising:

20. In a transcoder code conversion method for mutually converting a code according to the first encoding method and a code according to the second encoding method,

Calculating and outputting a spectrum parameter representing a spectrum characteristic from the decoded signal; Calculating a pitch period from the decoded signal, and generating an adaptive codebook component based on the pitch period and a past excitation signal;

Providing a gain to at least one of a noise output from a noise generator and the adaptive codebook component and adding the resultant to output a sound source signal;

Outputting a signal in a band required for passband conversion of the sound source signal to a synthesis filter configured using the filter coefficient;

A step of encoding the addition result by the second encoding method to obtain and output a second code;

A code conversion method, comprising:

21. In a transcoder code conversion method for mutually converting between a code according to the first coding method and a code according to the second coding method,

Calculating a pitch period from the decoded signal, and generating an adaptive codepock component based on the pitch period and a past sound source signal;

Calculating and outputting a filter coefficient after shifting the frequency of the spectrum parameter;

Performing a pitch prefilling process on the sound source signal using the pitch period;

Outputting a signal of a band necessary for band conversion through a signal subjected to the pitch pre-filling process to a synthetic filter configured using the filter coefficient; Adding a signal obtained by converting the decoded signal at a predetermined sampling frequency and an output signal of the synthesis filter;

A code conversion method, comprising:

22. In a transcoder code conversion method for mutually converting a code according to the first encoding method and a code according to the second encoding method,

Calculating a pitch period from the decoded signal and generating a periodic signal using the pitch period;

Outputting a sound source signal by adding a gain to at least one of the noise output from the noise generator and the periodic signal and adding the gain;

A code conversion method, comprising:

23. In a transcoder code conversion method for mutually converting between a code according to the first coding method and a code according to the second coding method,

Calculate and output spectral parameters representing spectral characteristics from the decoded signal Steps to

Performing pitch pre-filtering on the sound source signal using the pitch period;

Outputting a signal of a band necessary for band conversion through the pitch-prefilled signal to a synthesis filter configured by using the filter coefficient, and converting the decoded signal at a predetermined sampling frequency. Adding the obtained signal and the output signal of the synthesis filter;

A code conversion method, comprising:

24. reproducing a band-converted signal through an output signal of the synthesis filter unit in a post-fill configured using a weighting coefficient obtained by weighting the filter coefficient;

Adding the output signal of the post-fill signal to a signal obtained by converting the decoded signal at a predetermined sampling frequency instead of the output signal of the composite fill signal, and outputting the added signal;

The code conversion method according to any one of claims 19 to 23, comprising:

25. The code conversion method according to claim 20 or 21, further comprising a step of subjecting the adaptive codepock component to low-pass filter processing.

26. The noise generator generates and outputs a noise signal whose band-limited is equal to the frame length, with the average amplitude normalized to a predetermined level. The code conversion method according to any one of claims 19 to 25, comprising a step.

27. A step of discriminating voiced / unvoiced by a voiced / unvoiced discriminating unit for inputting the decoded signal and outputting voiced / unvoiced discrimination information;

Adjusting the gain of the output signal from the noise generating unit according to whether the voiced / unvoiced discrimination information is voiced or unvoiced;

20. The code conversion method according to claim 19, further comprising: applying the adjusted gain to an output signal from the noise generation unit.

28. A step of discriminating voiced / unvoiced by a voiced / unvoiced discriminating unit that inputs the decoded signal and outputting voiced / unvoiced discrimination information and pitch period information;

Adjusting the gain of at least one of the adaptive codebook signal and the output signal from the noise generating unit according to whether the voiced / unvoiced discriminating information from the voiced / unvoiced discriminating unit is voiced or unvoiced.

Multiplying at least one of the adaptive codebook signal and the output signal of the noise generator by the adjusted gain signal, and outputting the multiplied signal;

And outputting a signal obtained by adding the adaptive codebook signal obtained by multiplying at least one of the gains and the output signal of the noise generating unit as a sound source signal. Item or the code conversion method described in Item 21.

2 9. A step of discriminating voiced / unvoiced by a voiced / unvoiced discriminating unit that inputs the decoded signal and outputting voiced / unvoiced discrimination information and pitch cycle information

Adjusting the gain of at least one of the periodic signal and the output signal from the noise generating unit according to whether the voiced / unvoiced discriminating information from the voiced / unvoiced discriminating unit is voiced or unvoiced;

Multiplying at least one of the periodic signal and the output signal of the noise generator by the adjusted gain, and outputting the multiplied signal;

Outputting, as a sound source signal, a signal obtained by adding the periodic signal obtained by multiplying at least one of the gains and an output signal of the noise generation unit; The code conversion method according to claim 22 or claim 23, comprising:

30. The voiced / unvoiced discriminating circuit calculates a normalized autocorrelation function for the decoded signal up to a predetermined delay time, obtains a maximum value of the normalized autocorrelation function, and obtains the maximum value. 28.The method according to claim 27, further comprising the step of: discriminating voiced if the predetermined value is larger than a predetermined threshold, and discriminating voiced otherwise, and outputting a discrimination result as the voiced / unvoiced discrimination information. The transcoding method described in the above.

31. The voiced / unvoiced determination circuit calculates a normalized autocorrelation function for the decoded signal up to a predetermined delay time, obtains a maximum value of the normalized autocorrelation function, and obtains the maximum value. Is larger than a predetermined threshold value, it is determined as voiced, otherwise it is determined as unvoiced, and the result of the determination is output as the voiced / unvoiced determination information. In the voiced frame, the normalized autocorrelation function is 30. The code conversion method according to claim 28, further comprising a step of outputting a value of the delay to be maximized as a pitch period.