AUDIO QUANTIFIER AND AUDIO DEQUANTIFIER AND RELATED METHODS

Abstract

An audio quantizer for quantizing a plurality of audio information elements comprises: a first-stage vector quantizer (141, 143) for quantizing the plurality of audio information elements to determine a first-stage vector quantization result and a plurality of intermediate quantized elements corresponding to the first-stage vector quantization result; a residual element determinator (142) for calculating a plurality of residual elements from the plurality of intermediate quantized elements and the plurality of audio information elements;and a second-stage vector quantizer (145) for quantizing the plurality of residual elements to obtain a second-stage vector quantization result, wherein the first-stage vector quantization result and the second-stage vector quantization result are a quantized representation of the plurality of audio information elements.;

Description

AUDIO QUANTIFIER AND AUDIO DEQUANTIFIER AND RELATED METHODS Field of Invention The present invention relates to audio signal processing and, for example, can be applied in a stereo MDCT processing of, for example, IVAS. Furthermore, the present invention can be applied in the joint encoding of stereo spectral noise shaping parameters Background of the Invention Spectral noise shaping shapes quantization noise in the frequency domain so that the human ear perceives minimal quantization noise and, therefore, the perceptual quality of the decoded output signal can be maximized. Spectral noise shaping is a technique used in most state-of-the-art transform-based audio codecs. Advanced Audio Coding (AAC) In this approach [1][2], the MDCT spectrum is partitioned into a series of non-uniform scaling factor bands. For example, at 48 kHz, the MDCT has 1024 coefficients and is partitioned into 49 scaling factor bands. In each band, a scaling factor is used to modify the scale of the MDCT coefficients in that band. A scalar quantizer with a constant step size is then employed to quantize the scaled MDCT coefficients. On the decoder side, an inverse scaling is performed on each band, shaping the quantization noise introduced by the scalar quantizer. The 49 scaling factors are encoded in the bitstream as additional information. Encoding the scaling factors typically requires a significantly high number of bits due to the relatively large number of factors and the high precision required. This can become problematic with low bit rates and / or low latency. TCX based on MDCT In an MDCT-based TCX, a transform-based audio codec used in the MPEG-D USAC [3] and 3GPP EVS [4] standards, spectral noise shaping is done with the help of LPC-based perception filters, similar perception filter as used in recent ACELP-based speech codecs (e.g., AMR-WB). In this approach, a set of 16 linear prediction coefficients (LPCs) is first estimated on a previously emphasized input signal. The LPCs are then weighted and quantized. The frequency response of the weighted and quantized LPCs is then calculated in 64 evenly spaced bands. The MDCT coefficients are then scaled in each band using the calculated frequency response. The scaled MDCT coefficients are then quantized using a scalar quantizer with a step size controlled by an overall gain. In the decoder, inverse scaling is performed in every 64 bands, shaping the noise of MA / t / ZUZÓ / U Ί OU4y quantization introduced by the scalar quantifier. This approach has a distinct advantage over the AAC approach: it requires encoding only 16 parameters (LPC) + 1 (global gain) as additional information (as opposed to the 49 parameters in AAC). Furthermore, 16 LPCs can be efficiently encoded with a small number of bits by employing an LSF representation and a vector quantizer. Consequently, the MDCT-based TCX approach requires fewer bits of additional information than the AAC approach, which can make a significant difference at low bit rates and / or low latency. Enhanced TCX based on MDCT (psychoacoustic LPC) An improved TCX system based on MDCT is published in [5]. In this new approach, autocorrelation (for estimating LPCs) is no longer performed in the time domain, but is calculated in the MDCT domain using an inverse transform of the MDCT coefficient energies. This allows the use of a non-uniform frequency scale simply by grouping the MDCT coefficients into 64 non-uniform bands and calculating the energy of each band. It also reduces the complexity required to calculate the autocorrelation. New spectral noise (SNS) configuration In an improved technique for spectral noise shaping as described [6] and implemented in the low-complexity communication codec (LC3 / LC3plus), a low bit rate can be achieved without substantial loss of quality through scaling modification on the encoder side. This is accomplished by using a larger number of scaling factors and by subsampling the scaling parameters on the encoder side into a second set of 16 scaling parameters (SNS parameters). Thus, additional information about the low bit rate is obtained on the one hand, and yet high-quality spectral processing of the audio signal spectrum is achieved due to the fine scaling modification on the other. Stereo Linear Prediction (SLP) In the thesis described in [7], a set of linear prediction coefficients is calculated considering not only inter-frame prediction but also one-to-another channel prediction. The calculated set of two-dimensional coefficients is quantized and encoded using techniques similar to those of single-channel linear prediction, but without considering residual quantization in the context of the thesis. However, the described implementation comes with high latency and significant complexity and is therefore quite unsuitable for a real-time application requiring low latency, such as communication systems. In a stereo system such as the MDCT-based system described in [8], the discrete LR channel signals are preprocessed to scale the spectra using frequency-domain noise shaping to the bleached domain. Then, joint stereo processing is performed to optimally quantize and encode the bleached spectra. The scaling parameters for the spectral noise shaping techniques described above are quantized and encoded independently for each channel. This results in twice the bit rate of additional information that must be sent to the decoder via the bitstream. MA / t / ZUZÓ / U Ί OU4y Brief Description of the Invention An object of the present invention is to provide an improved or more efficient encoding / decoding concept. This object is achieved by means of an audio quantizer of claim 1, and an audio dequantizer of claim 16, a quantization method of claim 31, a dequantization method of claim 32 or a computer program according to claim 33. The present invention is based on the finding that bit rate savings can be achieved in cases where the L, R, or, more generally, two or more channels of a multichannel signal are correlated. In this case, the parameters extracted for both channels are quite similar. Therefore, joint quantization encoding of the parameters is applied, resulting in a significant bit rate saving. This bit rate saving can be used in several different ways. One way is to spend the saved bit rate on encoding the center signal so as to improve the overall perceptual quality of the stereo or multichannel signal. Another way is to achieve a lower overall bit rate in a case where the encoding of the center signal, and therefore the overall perceptual quality, is not improved but left at the same quality. In a preferred embodiment, according to a first aspect, an audio encoder comprises a scaling parameter calculator for calculating a first set of jointly encoded scaling parameters and a second set of jointly encoded scaling parameters for a first set of scaling parameters for a first channel of the multichannel audio signal and for a second set of scaling parameters for a second channel of the multichannel audio signal. The audio encoder further comprises a signal processor for applying the first set of scaling parameters to the first channel and for applying the second set of scaling parameters to the second channel of the multichannel audio signal. The signal processor further derives multichannel audio data from the first and second channel data obtained by applying the first and second sets of scaling parameters, respectively.The audio encoder additionally has a coded signal shaper to use the multichannel audio data and information about the first group of jointly coded scale parameters and information about the second group of jointly coded scale parameters to obtain a coded multichannel audio signal. Preferably, the scale parameter calculator is configured to be adaptive so that, for each frame or subframe of the multichannel audio signal, a determination is made as to whether to perform joint encoding of scale parameters or separate encoding of scale parameters. In a further mode, this determination is based on a similarity analysis between the channels of the multichannel audio signal under consideration. In particular, the similarity analysis is performed by calculating an energy for the jointly encoded parameters and, specifically, an energy for a set of scale parameters from the first group and the second group of jointly encoded scale parameters. Specifically, the scale parameter calculator calculates the first group as a sum of the corresponding first and second scale parameters and calculates MA / t / ZUZÓ / U Ί OU4y the second group as a difference between the first and second corresponding scale parameters. In particular, the second group, and preferably the scale parameters representing the difference, are used to determine the similarity measurement in order to decide whether to perform joint coding of the scale parameters or separate coding of the scale parameters. This situation can be signaled by a stereo or multichannel flag. Furthermore, it is preferred to specifically quantize the scale parameters using a two-stage quantization process. A first-stage vector quantizer quantizes the plurality of scale parameters, or more generally, audio information elements, to determine a first-stage vector quantization result and to determine a plurality of intermediate quantization elements corresponding to the first-stage vector quantization result. In addition, the quantizer comprises a residual element determinator to calculate a plurality of residual elements from the plurality of intermediate quantized elements and the plurality of audio information elements.Furthermore, a second-stage vector quantizer is provided to quantize the plurality of residual elements to obtain a second-stage vector quantization result, wherein the first-stage vector quantization result and the second-stage vector quantization result together represent the quantized representation of the plurality of audio information elements that are, in one modality, the scale parameters. In particular, the audio information elements can be jointly encoded scale parameters or separately encoded scale parameters. In addition, other audio information elements can be any audio information element that is useful for vector quantization.In particular, aside from scaling parameters or scaling factors as specific audio information elements, other useful audio information elements for vector quantization include spectral values ​​such as MDCT or FFT lines. Other audio information elements that can be vector quantized include time-domain audio values ​​such as audio sample values, time-domain audio sample groups, spectral-domain frequency line groups, LPC data, or other envelope data, whether a spectral or time-envelope representation. In a preferred implementation, the residual element determiner calculates, for each residual element, a difference between the corresponding audio information elements, such as a scale parameter, and a corresponding intermediate quantized element, such as a quantized scale parameter or scale factor. Furthermore, the residual element determiner is configured to amplify or weight, for each residual element, a difference between a corresponding audio information element and a corresponding intermediate quantized element so that the plurality of residual elements is greater than the corresponding difference, or to amplify or weight the plurality of audio information elements and / or the plurality of intermediate quantized elements before calculating a difference between the amplified elements to obtain the residual elements.This procedure allows for useful control of quantization error. This is particularly relevant when the second group of audio information elements, such as the various scale parameters, are quite small. MA / t / ZUZÓ / U Ί OU4y Typically, in the case where the first and second channels are correlated such that joint quantization has been determined, the residual elements are typically quite small. Therefore, when the residual elements are amplified, the quantization result will include more values ​​that are not quantized to 0 compared to a case where this amplification has not been performed. Thus, amplification on the encoder or quantization side can be useful. This is particularly the case when, as in another preferred mode, the second group of co-encoded scale parameters, such as the difference scale parameters, is quantized. Because these additional scale parameters are small anyway, a situation can arise where, without amplification, most of the different scale parameters are quantized to 0. Therefore, to avoid this situation, which could result in a loss of stereo imaging and thus a loss of psychoacoustic quality, the amplification is performed so that only a small number, or almost none, of the additional scale parameters are quantized to 0. This, of course, reduces the bit rate savings.Because of this, however, the quantized residual data elements are still only small; that is, they result in quantization indices that represent small values, and the bit rate increase is not very high, since quantization indices for small values ​​are encoded more efficiently than quantization indices for larger values. This can be further improved by performing an additional entropic coding operation that further favors small quantization indices over larger quantization indices in terms of bit rate. In another preferred embodiment, the first-stage vector quantizer is a vector quantizer that has a given codebook, and the second-stage vector quantizer is an algebraic vector quantizer that results in a quantization index in a codebook number, a vector index in a base codebook, and a Voronoi index. Preferably, both the vector quantizer and the algebraic vector quantizer are configured to perform split-level vector quantization, where both quantizers have the same split-level procedure.Furthermore, the first-stage and second-stage vector quantizers are configured such that the number of bits, and therefore the precision, of the first-stage vector quantizer result is greater than the number of bits, or the precision, of the second-stage vector quantizer result; or vice versa. In other configurations, the first-stage vector quantizer has a fixed bit rate, while the second-stage vector quantizer has a variable bit rate. Therefore, in general, the characteristics of the first-stage and second-stage vector quantizers differ from each other. In a preferred embodiment of an audio decoder for decoding an audio signal encoded according to the first aspect, the audio decoder comprises a scale parameter decoder for decoding information about the co-encoded scale parameters. Furthermore, the audio decoder MA / t / ZUZÓ / U Ί OU4y has a signal processor, where the scale parameter decoder is configured to combine a jointly encoded scale parameter from the first group and the jointly encoded scale parameter from the second group using different combination rules to obtain the scale parameters for the first set of scale parameters and the scale parameters for the second set of scale parameters that are then used by the signal processor. According to the additional aspect of the present invention, an audio dequantizer is provided comprising a first-stage vector dequantizer, a second-stage vector dequantizer, and a combiner for combining the plurality of intermediate quantizer information elements obtained by the first-stage vector dequantizer and the plurality of residual elements obtained from the second-stage vector dequantizer to obtain a dequantized plurality of audio information elements. The first aspect of co-coding scale parameters can be combined with the second aspect related to two-stage vector quantization. On the other hand, the two-stage vector quantization aspect can be applied to separately encoded scale parameters, such as scale parameters for a left and right channel, or it can be applied to mid-scale parameters as another type of audio information element. Therefore, the second aspect of two-stage vector quantization can be applied independently of the first aspect or in conjunction with it. Subsequently, the preferred embodiments of the present invention are summarized. In a stereo system using transform-based coding (MDCT), the scaling parameters extracted from any of the techniques described in the introductory section for frequency-domain noise shaping on the encoder side must be quantized and encoded to be included as additional information in the bitstream. Then, on the decoder side, the scaling parameters are decoded and used to scale the spectrum of each channel to shape the quantization noise in a minimally perceptible way. Independent coding of the spectral noise shaping parameters of the two channels: left and right can be applied. The parameters for modifying the spectral noise shaping scale are adaptively encoded, either independently or jointly, depending on the degree of correlation between the two channels. In summary: • A mean / lateral representation of the scaling modification parameters is calculated. • The energy of the lateral parameters is calculated. • Depending on the energy – which indicates the degree of correlation between the two signals – the parameters are encoded: • Independently: like the current approach, using for each channel, for example, a two-stage vector quantization (VQ) • Jointly: MA / t / ZUZÓ / U Ί OU4y • The middle vector is encoded using, for example, a two-stage vector quantization. The side vector is encoded using a coarser quantization scheme, for example, assuming that the first-stage output VQ comprises quantized values ​​of zero and applying only the second-stage quantization, for example, an algebraic vector quantizer (AVQ). • An additional bit is used to signal whether the quantized side vector is zero or not. • An additional bit to signal whether the two channels are encoded jointly or independently is sent to the decoder. Figure 24 shows an MDCT-stereo-based encoder implementation as described in detail in [8]. An essential part of the stereo system described in [8] is that stereo processing is performed on the bleached spectra. Therefore, each channel undergoes preprocessing, where for each frame, after windowing, the time-domain block is transformed into the MDCT domain. Then, time noise shaping (TNS) is applied adaptively, either before or after spectral noise shaping (SNS) depending on the signal characteristics. After spectral noise shaping, joint stereo processing is performed; that is, an adaptive MS, L / R decision per band to efficiently quantize and encode the bleached-spectrum coefficients.As a next step, stereo intelligent space-filling (IGF) analysis is performed, and the respective information bits are written to the bitstream. Finally, the processed coefficients are quantized and encoded. Reference numbers similar to those in Figure 1 have been added. The calculation and processing of the scaling factors are carried out in the SNS blocks between the two TNS blocks in Figure 24. The block window illustrates a windowing operation. The MCLT block stands for Modified Complex Lapse Transform. The MDCT block stands for Modified Discrete Cosine Transform. The block power spectrum represents the calculation of a power spectrum. The block switching decision represents an analysis of the input signal to determine the block lengths to be used for the window.The TNS block represents the temporal noise shaping and this feature is performed before or after the spectrum scaling modification in the SNS block. In the MDCT-stereo codec implementation described above, on the encoder side, the discrete LR channels are preprocessed to scale the spectra using frequency-domain noise shaping to the bleached domain. Then, joint stereo processing is performed to optimally quantize and encode the bleached spectra. On the decoder side, as illustrated in Figure 25 and described in [8], the encoded signal is decoded, and inverse quantization and inverse stereo processing are performed. The spectrum of each channel is then whitened by spectral noise shaping parameters recovered from the bitstream. Reference numbers similar to those in Figure 1 have been added. Decoding and scaling factor processing are carried out in blocks 220 in Figure 25. The blocks indicated in the figure relate to the blocks in the encoder in Figure 24 and typically perform the corresponding inverse operations. The window and OLA block performs a ML / t / ZUZÓ / U Ί OU4U synthesis window creation operation and a superposition and subsequent summation operation to obtain the L and R time domain output signals. The frequency domain noise shaping (FDNS) applied to the system in [8] is replaced here with SNS as described in [6]. Block diagrams in Figure 1 and Figure 2 show a block diagram of the SNS processing path for the encoder and decoder, respectively. Preferably, a low bit rate can be achieved without substantial loss of quality through scaling modification on the encoder side by using a larger number of scaling factors and subsampling the scaling parameters on the encoder side into a second set of scaling parameters or scaling factors. The scaling parameters in this second set, which are then encoded and transmitted or stored through an output interface, are smaller than the first set of scaling parameters. This results in both fine scaling and a low bit rate on the encoder side. On the decoder side, the small transmitted number of scale factors is decoded by a scale factor decoder to obtain a first set of scale factors where the number of scale factors or scale parameters in the first set is greater than the number of scale factors or scale parameters in the second set, and then again, fine scaling is performed using the larger number of scale parameters on the decoder side within a spectral processor to obtain a fine scaling spectral representation. Therefore, you get a low bit rate on one hand and yet high-quality spectral processing of the audio signal spectrum on the other. Spectral noise shaping, as used in preferred modes, is implemented using only a very low bit rate. Therefore, this spectral noise shaping can be an essential tool even in a low-bit-rate transform-based audio codec. Spectral noise shaping shapes quantization noise in the frequency domain so that the human ear perceives minimal quantization noise, thus maximizing the perceptual quality of the decoded output signal. The preferred methods are based on spectral parameters calculated from amplitude-related measurements, such as the energies of a spectral representation. In particular, band energies, or more generally, band-related amplitude measurements, are calculated as the basis for scaling parameters. The bandwidths used in calculating these band-related amplitude measurements increase from lower to higher bands to approximate the characteristics of human hearing as closely as possible. Preferably, the division of the spectral representation into bands is performed according to the well-known Bark scale. In additional modalities, linear domain scale parameters are calculated, particularly for the first set of scale parameters with the high number of scale parameters, and this high number of scale parameters becomes a logarithmic domain. A logarithmic domain is generally a domain, ML / t / ZUZÓ / U Ί OU4y in which small values ​​are expanded and large values ​​are compressed. Then, the subsampling or downsampling operation of the scale parameters is performed in the logarithmic domain, which can be a logarithmic domain with base 10, or a logarithmic domain with base 2, the latter being preferred for implementation purposes. The second set of scale factors is then calculated in the logarithmic domain, and preferably, vector quantization of the second set of scale factors is performed, where the scale factors are in the logarithmic domain. Therefore, the result of the vector quantization indicates logarithmic domain scale parameters. The second set of scale factors or scale parameters has, for example, half the number of scale factors of the first set, or even a third, or even more preferably, a quarter.The small, quantized number of scale parameters in the second set of scale parameters is then brought into the bitstream and transmitted from the encoder side to the decoder side, or stored as an encoded audio signal along with a quantized spectrum that has also been processed using these parameters. This processing further involves quantization using an overall gain. Preferably, however, the encoder derives from these quantized logarithmic-domain second scale factors a set of linear-domain scale factors, which is the third set of scale factors. The number of scale factors in the third set of scale factors is greater than, and preferably equal to, the number of scale factors in the first set of scale factors.Then, on the encoder side, these interpolated scaling factors are used to process the spectral representation, where the processed spectral representation is finally quantized and, in any case, encoded by entropy, such as by Huffman coding, arithmetic coding, or vector quantization-based coding, etc. In the decoder that receives an encoded signal with a low number of spectral parameters along with the encoded representation of the spectral signal, the low number of scale parameters is interpolated to a high number of scale parameters. This results in a first set of scale parameters where the number of scale factors in the second set of scale factors or scale parameters is less than the number of scale parameters in the first set, i.e., the set calculated by the parameter / scale factor decoder. Then, a spectral processor located within the device for decoding the encoded audio signal processes the decoded spectral representation using this first set of scale parameters to obtain a scaled-up spectral representation.A converter to convert the scaled modified spectral representation then operates to finally obtain a decoded audio signal that is preferably in the time domain. Other modalities result in additional advantages, which are discussed later. In preferred modalities, spectral noise shaping is performed using 16 scaling parameters similar to the scaling factors used in [6], [8], or [1]. These parameters are obtained in the encoder by first calculating the MDCT spectrum energy in 64 non-uniform bands (similar to the 64 non-uniform bands of the prior technique 3), then applying some processing to the 64 energies (smoothing, pre-emphasis, minimum noise threshold, conversion). MA / t / ZUZÓ / U 1 OU4y logarithmic), then by subsampling the 64 processed energies by a factor of 4 to obtain 16 parameters which are finally normalized and scaled. These 16 parameters are then quantized using vector quantization (using similar vector quantization as used in technique 2 / 3 above). The quantized parameters are then interpolated to obtain 64 interpolated scaled parameters. These 64 scaled parameters are then used to directly shape the MDCT spectrum in the 64 non-uniform bands. Similar to techniques 2 and 3 above, the scaled MDCT coefficients are then quantized using a scalar quantizer with a step size controlled by an overall gain. In a further embodiment, the information about the jointly encoded scale parameters for one of the two groups, such as the second group preferably related to the side scale parameters, does not comprise quantization indices or other quantization bits, but only information such as a flag or a single bit indicating that the scale parameters for the second group are all zero for a portion or frame of the audio signal. This information is determined by the encoder through analysis or other means and is used by the decoder to synthesize the second group of scale parameters based on this information, such as by generating zero scale parameters for the time portion or frame of the audio signal, or it is used by the decoder to calculate the first and second sets of scale parameters using only the first group of jointly encoded scale parameters. In a further embodiment, the second group of co-encoded scale parameters is quantized using only the second quantization stage of the two-stage quantizer, which is preferably a variable-rate quantizer stage. In this case, it is assumed that the first stage results in all quantized values ​​being zero, so only the second stage is effective. In yet another embodiment, only the first quantization stage of the two-stage quantizer, which is preferably a fixed-rate quantizer stage, is applied, and the second stage is not used at all for a portion of time or frame of the audio signal. This case corresponds to a situation where it is assumed that all residual elements are zero or smaller than the smallest or first quantization step size of the second quantization stage. Brief Description of the Figures Preferred embodiments of the present invention are subsequently analyzed with respect to the attached figures, in which: Figure 1 illustrates a decoder according to the first aspect; Figure 2 illustrates an encoder according to the first aspect; Figure 3a illustrates another encoder according to the first aspect; Figure 3b illustrates another implementation of an encoder according to the first aspect; Figure 4a illustrates an additional modality of a decoder according to the first aspect; Figure 4b illustrates another type of decoder; Figure 5 illustrates an additional modality of an encoder; MA / t / ZUZÓ / U Ί OU4y Figure 6 illustrates an additional modality of an encoder; Figure 7a illustrates a preferred implementation of a vector quantifier according to a first or second aspect; Figure 7b illustrates an additional quantifier according to the first or second aspect; Figure 8a illustrates a decoder according to a first aspect of the present invention; Figure 8b illustrates an encoder according to the first aspect of the present invention; Figure 9a illustrates an encoder according to the second aspect of the present invention; Figure 9b illustrates a decoder according to the second aspect of the present invention; Figure 10 illustrates a preferred implementation of a decoder according to the first or second aspect; Figure 11 is a block diagram of a device for encoding an audio signal; Figure 12 is a schematic representation of a preferred implementation of the scale factor calculator in Figure 1; Figure 13 is a schematic representation of a preferred implementation of the subsampler in Figure 1; Figure 14 is a schematic representation of the scale factor encoder of Figure 4; Figure 15 is a schematic illustration of the spectral processor in Figure 1; Figure 16 illustrates a general representation of an encoder on one side and a decoder on the other side that implements spectral noise shaping (SNS); Figure 17 illustrates a more detailed representation of the encoder side on one side and the decoder side on the other side where temporal noise shaping (TNS) is implemented together with spectral noise shaping (SNS); Figure 18 illustrates a block diagram of a device for decoding an encoded audio signal; Figure 19 illustrates a schematic illustration that illustrates details of the scale factor decoder, spectral processor, and spectrum decoder of Figure 8; Figure 20 illustrates a subdivision of the spectrum into 64 bands; Figure 21 illustrates a schematic representation of the subsampling operation on one hand and the interpolation operation on the other; Figure 22a illustrates a time-domain audio signal with overlapping frames; Figure 22b illustrates an implementation of the converter in Figure 1; Figure 22C illustrates a schematic representation of the converter in Figure 8; Figure 23 illustrates a histogram comparing different inventive procedures; Figure 24 illustrates one type of encoder; and Figure 25 illustrates one type of decoder. MA / t / ZUZÓ / U Ί OU4y Detailed Description of the InventionFigure 8a-8b illustrates an audio decoder for decoding an encoded audio signal comprising multichannel audio data comprising data for two or more audio channels and information about jointly encoded scaling parameters. The decoder comprises a scaling parameter decoder 220 and a signal processor 210, 212, 213 illustrated in Figure 8a as a single element. The scaling parameter decoder 220 receives information about the first jointly encoded group and the second group of scaling parameters, where, preferably, the first group of scaling parameters are mid-scale parameters and the second group of scaling parameters are side-scale parameters.Preferably, the signal processor receives the first channel representation of the multichannel audio data and the second channel representation of the multichannel audio data and applies the first set of scaling parameters to a first channel representation derived from the multichannel audio data and applies the second set of scaling parameters to the second channel representation derived from the multichannel audio data to obtain the first and second channels of the decoded audio signal at the output of block 210, 212, 213 in Figure 8a. Preferably, the co-coded scaling parameters comprise information about the first group of co-coded scaling parameters such as mid-scale parameters and information about a second group of co-coded scaling parameters such as side-scale parameters.Furthermore, the 220 scale parameter decoder is configured to combine a jointly encoded scale parameter from the first group and a jointly encoded scale parameter from the second group using a first combination rule to obtain a scale parameter from the first set of scale parameters, and to combine the same jointly encoded scale parameters from the first and second groups using a second combination rule that is different from the first combination rule to obtain a scale parameter from the second set of scale parameters. Therefore, the 220 scale parameter decoder applies two different combination rules. In a preferred embodiment, the two different combination rules are a combination rule of addition on one hand and a combination rule of subtraction on the other. However, in other embodiments, the first combination rule may be a combination rule of multiplication and the second combination rule may be a combination rule of quotient or division. Therefore, all other pairs of combination rules are also useful depending on the representation of the corresponding scale parameters of the first and second groups, or of the first and second sets of scale parameters. Figure 8b illustrates a corresponding audio encoder for encoding a multichannel audio signal comprising two or more channels. The audio encoder comprises a scaling parameter calculator 140, a signal processor 120, and a coded signal former 1480, 1500. The scaling parameter calculator 140 is configured to calculate a first set of jointly encoded scaling parameters and a second set of jointly encoded scaling parameters from a first set of scaling parameters for a first channel of the multichannel audio signal and from a second set of scaling parameters for a second channel. ML / t / ZUZÓ / U Ί OU4y of the multichannel audio signal. In addition, the signal processor is configured to apply the first set of scaling parameters to the first channel of the multichannel audio signal and to apply the second set of scaling parameters to the second channel of the multichannel audio signal to derive encoded multichannel audio data. The multichannel audio data is derived from the scaled first and second channels, and the multichannel audio data is used by the encoded signal former 1480,1500 together with the information on the first and second sets of scaling parameters encoded together to obtain the encoded multichannel audio signal at the output of block 1500 in Figure 8b. Figure 1 illustrates a further implementation of the decoder in Figure 8a. Specifically, the bitstream is fed into signal processor 210, which typically performs entropy decoding and inverse quantization, along with intelligent gap-filling (IGF) procedures and inverse stereo processing of the scaled or bleached channels. The output of block 210 is either scaled or bleached left and right, or, more generally, several decoded channels of a multichannel signal. The bitstream comprises additional information bits for the left and right scale parameters in the case of separate encoding, and side information bits for the jointly encoded and scaled scale parameters, illustrated as scale parameters M and S in Figure 1.This data is entered into the scale parameter or scale factor decoder 220 which, at its output, generates the decoded left scale factors and the decoded right scale factors which are then applied in the spectrum block 212,230 to finally obtain a preferably MDCT spectrum for left and right which can then be converted into a time domain using a certain inverse MDCT operation. The implementation of the corresponding encoder side is provided in Figure 2. Figure 2 begins with an MDCT spectrum having a left and right channel, which are fed into a spectrum shaper 120a. The output of spectrum shaper 120a is fed into a processor 120b, which, for example, performs stereo processing, intelligent gap-filling operations on the encoder side, and corresponding encoding (entropy) and quantization operations. Therefore, blocks 120a and 120b together represent the signal processor 120 in Figure 8b. Furthermore, for the purpose of calculating the scaling factors performed on the SNS (spectral noise shaping) scaling factors calculated from block 120b, an MDST spectrum is also provided. The MDST spectrum, along with the MDCT spectrum, is then fed into a power spectrum calculator 110a.Alternatively, the power spectrum calculator 110a can operate directly on the input signal without an MDCT or MDST spectrum procedure. Another way would be to calculate the power spectrum of a DFT operation instead of an MDCT and an MDST operation, for example. Furthermore, the scaling factors are calculated using the scaling parameter calculator 140, illustrated in Figure 2 as a block quantization encoding of scaling factors. In particular, the outputs of block 140, depending on the similarity between the first and second channels, are either separately coded scaling factors for left and right or jointly coded scaling factors for M and S. This is illustrated in Figure 2 to the right of block 140. Therefore, in this... MA / t / ZUZÓ / U Ί OU4y implementation, block 110b calculates the scale factors for left and right and block 140 then determines, whether separate coding, i.e., coding for the left and right scale factors is better or worse than coding of jointly coded scale factors, i.e., M and S scale factors derived from the scale factors separated by the two different combination rules such as an addition on one side and a subtraction on the other side. The result of block 140 are additional information bits for L, R or M, S which, together with the result of block 120b, are fed into an output bit stream illustrated by Figure 2. Figure 3a illustrates a preferred implementation of the encoder in Figure 2 or Figure 8b. The first channel is input into block 1100a, which determines the separate scaling parameters for the first channel, i.e., for the L channel. The second channel is input into block 1100b, which determines the separate scaling parameters for the second channel, i.e., for the R channel. The scaling parameters for the left channel and the scaling parameters for the right channel are then proportionally reduced by a subsampler 130a for the first channel and a subsampler 130b for the second channel. The results are subsampled parameters (DL) for the left channel and subsampled parameters (DR) for the right channel. These DL and DR data are then fed into a joint scale parameter determinator 1200. The joint scale parameter determinator 1200 generates the first group of jointly encoded scale parameters, such as mean scale parameters or M, and a second group of jointly encoded scale parameters, such as side scale parameters or S. Both groups are fed into the corresponding vector quantizers 140a, 140b to obtain quantized values, which are then fed into a final entropy encoder 140c and encoded to obtain information about the jointly encoded scale parameters. The 140c entropy encoder can be implemented to perform an arithmetic entropy coding algorithm or an entropic coding algorithm with a one-dimensional Huffman code table or with one or more dimensional Huffman code tables. Another implementation of the encoder is illustrated in Figure 3b, where subsampling is not performed with separate scale parameters such as left and right, as illustrated in 130a and 130b in Figure 3a. Instead, the order of operations for joint scale parameter determination and subsequent subsampling is reversed using the corresponding subsamplers 130a and 130b. Whether to use the implementation in Figure 3a or Figure 3b depends on the specific implementation, with the implementation in Figure 3a being preferred. This is because joint scale parameter determination 1200 is already performed on the subsampled scale parameters; that is, the two different combination rules performed by the scale parameter calculator 140 are typically carried out on fewer inputs compared to the case in Figure 3b. Figure 4a illustrates the implementation of a decoder for decoding an encoded audio signal that has multichannel audio data comprising data for two or more audio channels and co-encoded scaling parameter information. However, the decoder in Figure 4a is only one part of the whole. MA / t / ZUZÓ / U Ί OU4y decoder of figure 8a, since only a part of the signal processor and, in particular, the corresponding channel scaling modifiers 212a, 212b are illustrated in figure 4a. With regard to the scaling parameter decoder 220, this element comprises an entropy decoder 2200 that reverses the procedure performed by the corresponding block 140c in figure 3a. In addition, the entropy decoder outputs jointly quantized encoded scaling parameters, such as quantized M scaling parameters and quantized S scaling parameters. The corresponding groups of scale parameters are fed into dequantizers 2202 and 2204 in order to obtain dequantized values ​​for M and S. These dequantized values ​​are then fed into a separate scale parameter determinator 2206 which generates scale parameters for left and right, i.e., separate scale parameters.These corresponding scale parameters are fed into interpolators 222a and 222b to obtain the interpolated scale parameters for the left (IL) and the interpolated scale parameters for the right (IR). Both sets of data are then fed into channel scale modifiers 212a and 212b, respectively. Additionally, the channel scale modifiers receive the first channel representation after the entire procedure performed by block 210 in Figure 1, as an example. Consequently, channel scale modifier 212b also receives its corresponding second channel representation as output from block 210 in Figure 1. A final channel scale modification, or shape spectrum as it is called in Figure 1, is then performed to obtain a shaped left and right spectral channel, illustrated as the MDCT spectrum in Figure 1.Then, a final frequency domain to time domain conversion can be performed for each channel illustrated in 240a, 240b to finally obtain a first decoded channel and a second decoded channel of a multichannel audio signal in a time domain representation. In particular, the 220 scale parameter decoder illustrated in the left portion of Figure 4a can be included within an audio decoder as shown in Figure 1 or as shown collectively in Figure 4a, but it can also be included as a local decoder within an encoder as will be shown with respect to Figure 5 which explicitly shows the 220 local scale parameter decoder at the output of the 140 scale parameter encoder. Figure 4b illustrates a further implementation where, compared to Figure 4a, the order of interpolation and scaling parameter determination is reversed to determine the separate scaling parameters. In particular, interpolation takes place with the jointly encoded scaling parameters M and S using interpolators 222a and 222b in Figure 4b, and the interpolated jointly encoded scaling parameters such as IM and IS are fed into the separate scaling parameter determinator 2206. The output of block 2206 then consists of the oversampled scaling parameters, i.e., the scaling parameters for each of the, for example, 64 bands illustrated in Figure 21. Figure 5 illustrates an additional preferred implementation of the encoder of Figure 8b, Figure 2, or Figure 3a and Figure 3b. The first and second channels are fed into an optional time-domain-to-frequency converter such as 100a and 100b in Figure 5. The spectral representation produced by blocks 100a and 100b is input into a MA / t / ZUZÓ / U Ί OU4y channel scale modifier 120a that modifies the scale individually of the spectral representation for the left and right channels. Therefore, channel scale modifier 120a performs a spectrum operation as illustrated in 120a of Figure 2. The output of the channel scale modifier is fed into a channel processor 120b of Figure 5, and the processed channel output of block 120b is fed into the coded signal former 1480,1500 to obtain the coded audio signal. Furthermore, to determine the separately or jointly encoded scaling parameters, a similarity calculator 1400 is provided, which receives, as input, the first and second channels directly in the time domain. Alternatively, the similarity calculator can receive the first and second channels at the output of the time-to-frequency converters 100A and 100B, i.e., the spectral representation. Although Figure 6 will show that the similarity between the two channels is calculated based on the second set of jointly encoded scale parameters—that is, the side scale parameters—it should be noted that this similarity can also be calculated directly from the time-domain or spectral-domain channels without explicitly calculating the jointly encoded scale parameters. Alternatively, the similarity can also be determined based on the first set of jointly encoded scale parameters—that is, the mid-scale parameters. Specifically, joint encoding is possible when the energy of the side scale parameters is below a certain threshold.Similarly, the energy of the mid-scale parameters in a frame can also be measured, and joint coding can be determined when the energy of the mid-scale parameters exceeds another threshold, for example. Therefore, many different methods can be implemented to determine the similarity between the first and second channels to decide whether to co-code or separately encode scale parameters. However, it should be noted that the determination for co- or separate encoding of scale parameters does not necessarily have to be identical to the determination of joint stereo coding for the channels—that is, whether two channels are co-coded using a mid / side representation or separately encoded in an L, R representation.The determination of the joint coding of the scaling parameters is performed independently of the determination of stereo processing for the actual channels, since the determination of any stereo processing performed in block 120b in Figure 2 is done after and following a scaling modification or spectrum shaping using scaling factors for mid and lateral. Specifically, as illustrated in Figure 2, block 140 can determine joint coding. Therefore, as illustrated by the arrow in Figure 2 pointing to block 140, the scaling factors for M and S can occur within this block.In the case of applying a local scale parameter decoder 220 within the encoder of Figure 5, then the scale parameters actually used to shape the spectrum, although they are scale parameters for left and scale parameters for right, are nevertheless derived from the encoded and decoded scale parameters for middle and side. With respect to Figure 5, a mode 1402 decision module is provided. The mode decision module MA / t / ZUZÓ / U Ί OU4U Block 1402 receives the output from the similarity calculator 1400 and decides on separate scaling parameters when the channels are not sufficiently similar. However, when the channels are determined to be similar, then joint scaling parameters are determined by block 1402, and the information—whether separate joint scaling or a change in scaling parameters is applied—is signaled by a corresponding additional information or flag 1403, illustrated in Figure 5, which is provided from block 1402 to the coded signal former 1480, 1500. Furthermore, the encoder comprises the scale parameter encoder 140, which receives the scale parameters for the first channel and the scale parameters for the second channel and encodes the scale parameters either separately or jointly, as controlled by the mode decision module 1402.The scaling parameter encoder 140 can, in one mode, output the scaling parameters for the first and second channels as indicated by the dashed lines, so that the channel scaling modifier 120a performs scaling with the corresponding first and second channel scaling parameters. However, it is preferable to apply a local scaling parameter decoder 220 within the encoder so that the channel scaling is carried out with the scale parameters encoded and decoded locally, and the dequantized scale parameters are applied for channel scaling within the encoder. This has the advantage that exactly the same situation occurs within the channel scaling modifier in both the encoder and the decoder, at least with respect to the scale parameters used for channel scaling or spectrum shaping. Figure 6 illustrates a further preferred embodiment of the present invention with respect to the audio encoder. An MDCT spectrum calculator 100 is provided, which may, for example, be a time-domain to frequency-domain converter that applies an MDCT algorithm. In addition, a power spectrum calculator 110a is provided, as illustrated in Figure 2. The separate scale parameters are calculated using a corresponding calculator 1100, and for the purpose of calculating the jointly encoded scale parameters, an addition block 1200a and a subtraction block 1200b are used. Then, for the purpose of determining similarity, a per-frame power calculation is performed using the side parameters, i.e., the second group of jointly encoded scale parameters.In block 1406, a comparison is made with a threshold. This block, which is similar to the mode decision module 1402 for the frame in Figure 5, outputs the mode flag or stereo flag for the corresponding frame. Furthermore, information is provided to the controllable encoder, which performs separate or joint encoding on the current frame. To this end, the controllable encoder 140 receives the scale parameters calculated by block 1100 (i.e., the separate scale parameters) and also receives the jointly encoded scale parameters (i.e., those determined by blocks 1200a and 1200b). Block 140 preferentially generates a zero flag for the frame when it determines that all of a frame's side parameters should be quantized to 0. This result occurs when the first and second channels are very close to each other, and the differences between the channels, and therefore the differences between the scale factors, are such that these differences are smaller than the lowest quantization threshold applied by the quantizer. MA / t / ZUZÓ / U Ί OU4y included in block 140. Block 140 generates information about the scale parameters coded together or separately for the corresponding box. Figure 9a illustrates an audio quantizer for quantizing a plurality of audio information elements. The audio quantizer comprises a first-stage vector quantizer 141, 143 for quantizing the plurality of audio information elements, such as scale factors, scale parameters, spectral values, etc., to determine a first-stage vector quantization result 146. Furthermore, blocks 141 and 143 generate a plurality of intermediate quantized elements corresponding to the first-stage vector quantization result. The intermediate quantized elements are, for example, the values ​​associated with the first-stage result.When the first-stage result identifies a particular codebook with, for example, 16 certain (quantized) values, then the intermediate quantized elements are the 16 values ​​associated with the codebook vector index, which is the first-stage result 146. The intermediate quantized elements and the audio information elements in the input to the first-stage vector quantizer 141, 143 are fed into a residual element determinator to calculate a plurality of residual elements from the plurality of intermediate quantized elements and the plurality of audio information elements. This is done, for example, by calculating a difference for each item between the original element and the quantized element.The residual elements are fed into a second-stage vector quantizer 145 to quantize the plurality of residual elements to obtain the second-stage vector quantization result. Then, the first-stage vector quantization result at the output of blocks 141, 143 and the second-stage result at the output of block 145 together represent the quantized representation of the plurality of audio information elements that are encoded by an optional encoded signal former 1480, 1500, which outputs the quantized audio information elements that are not only quantized in the preferred mode but are additionally encoded by entropy. A corresponding audio dequantizer is illustrated in Figure 9b. The audio dequantizer comprises a first-stage vector dequantizer 2220 for dequantizing a first-stage quantization result included in the quantized plurality of audio information elements to obtain a plurality of intermediate quantized audio information elements. In addition, a second-stage vector dequantizer 2260 is provided and configured to dequantize a second-stage vector quantization result included in the quantized plurality of audio information elements to obtain a plurality of residual elements.Both the intermediate elements of block 2220 and the residual elements of block 2260 are combined by a combiner 2240 to combine the plurality of intermediate quantized audio elements and the plurality of residual elements to obtain a dequantized plurality of audio information elements. In particular, the intermediate quantized elements in the output of block 2220 are separately encoded scale parameters, such as for L and R, or the first group of jointly encoded scale parameters, for example, for M, and the residual elements can represent jointly encoded side scale parameters, for example, the second group of jointly encoded scale parameters. MA / t / ZUZÓ / U Ί OU4y Figure 7a illustrates a preferred implementation of the first-stage vector quantizer 141,143 of Figure 9a. In step 701, vector quantization of a first subset of scale parameters is performed to obtain a first quantization index. In step 702, vector quantization of a second subset of scale parameters is performed to obtain a second quantization index. Additionally, depending on the implementation, vector quantization of a third subset of scale parameters is performed, as illustrated in block 703, to obtain a third quantization index, which is optional. The procedure in Figure 7a applies when there is split-level quantization. As an example, the audio input signal is split into 64 bands, illustrated in Figure 21. These 64 bands are subsampled to 16 bands / scale factors, so that the entire bandwidth is covered by 16 scale factors.These 16 scale factors are quantized by the first-stage vector quantizer 141,143 in a split-level mode illustrated in Figure 7a. The first 8 scale factors of the 16 scale factors in Figure 21, obtained by subsampling the original 64 scale factors, are vector-quantized by step 701 and thus represent the first subset of scale parameters. The remaining 8 scale parameters for the upper 8 bands represent the second subset of scale parameters, which are vector-quantized in step 702. Depending on the implementation, the entire set of scale parameters or audio information elements does not necessarily have to be separated into exactly two subsets; it can also be done into three or even more subsets. Regardless of how many splits are performed, the indices for each level together represent the first-stage result. As discussed with respect to Figure 14, these indices can be combined using an index combinator in Figure 14 to obtain a single first-stage index. Alternatively, the first-stage result may consist of the first index, the second index, a potential third index, and probably even more indices that are not combined but are encoded by entropy as is. In addition to the corresponding indices that form the first-stage result, steps 701, 702, and 703 also provide the intermediate scale parameters used in block 704 to calculate the residual scale parameters for the frame. Therefore, step 705, performed, for example, by block 142 in Figure 9a, yields the residual scale parameters, which are then processed by an algebraic vector quantization performed by step 705 to generate the second-stage result. Thus, the first-stage result and the second-stage result are generated for the separate scale parameters L, the separate scale parameters R, and the first group of joint scale parameters M.However, as illustrated in Figure 7b, the (algebraic) vector quantization of the second group of jointly encoded scale parameters or side scale parameters is only performed by step 706, which is found in a preferred implementation identical to step 705, and is again performed by block 142 in Figure 9a. In a further embodiment, the information about the jointly encoded scale parameters for one of the two groups, such as the second group preferably related to the side scale parameters, does not comprise quantization indices or other quantization bits, but only information such as a flag or a single MA / t / ZUZÓ / U Ί OU4y bit indicating that the scale parameters for the second group are all zero for a portion or frame of the audio signal or are all at a certain value, such as a small value. This information is determined by the encoder through analysis or other means and is used by the decoder to synthesize the second group of scale parameters based on this information, such as by generating zero scale parameters for the time portion or frame of the audio signal, or by generating certain value scale parameters, or by generating small random scale parameters, all of which are, for example, smaller than the smallest or first quantization stage. Alternatively, it is used by the decoder to calculate the first and second sets of scale parameters using only the first group of jointly encoded scale parameters.Therefore, instead of performing step 705 in Figure 7a, only the all-zero flag for the second group of jointly encoded scale parameters is written as the result of the second step. The calculation in block 704 can also be omitted in this case and replaced with a decision module to determine whether to set and transmit the all-zero flag. This decision module can be controlled by a user input indicating a jump in the encoding of the S parameters as a whole, or by bit rate information, or it can actually perform an analysis of the residual elements.Therefore, for the frame that has the all-zero bit, the scale parameter decoder does not perform any combination but calculates the second set of scale parameters only using the first group of jointly encoded scale parameters, such as by dividing the encoded scale parameters of the first group by two or by weighting using another default value. In an additional embodiment, the second group of co-coded scale parameters is quantized using only the second quantization stage of the two-stage quantizer, which is preferably a variable-rate quantizer stage. In this case, the first stage is assumed to result in all quantized values ​​being zero, so only the second stage is effective. This case is illustrated in Figure 7b. In an even further mode, only the first quantization stage, such as 701, 702, and 703 of the two-stage quantizer in Figure 7a, is applied. This is preferably a fixed-rate quantization stage, and the second stage, 705, is not used at all for a portion of time or a frame of the audio signal. This case corresponds to a situation where it is assumed that all residual elements are zero or smaller than the smallest or first quantization step size of the second quantization stage. Then, in Figure 7b, element 706 would correspond to elements 701, 702, and 703 of Figure 7a, and element 704 could also be omitted and replaced by a decision module to determine whether or not to use the first-stage quantization.This decision module can be controlled by user input or bit rate information, or an analysis of the residual elements can actually be performed to determine that the residual elements are small enough for the accuracy of the second group of jointly coded scale parameters quantized by the individual stage to be sufficient. In a preferred implementation of the present invention, further illustrated in Figure 14, the algebraic vector quantizer 145 further performs a split-level calculation and, preferably, performs the MA / t / ZUZÓ / U Ί OU4y is the same split-level operation performed by the vector quantizer. Therefore, the subsets of residual values ​​correspond, with respect to the band number, to the subset of scale parameters. In the case of having two split levels, that is, for the first 8 subsampled bands in Figure 21, the algebraic vector quantizer 145 generates the first level result. In addition, the algebraic vector quantizer 145 generates a second level result for the 8 subsampled scale factors or upper scale parameters, or, generally, audio information elements. Preferably, the algebraic vector quantizer 145 is implemented as the algebraic vector quantizer defined in section 5.2.3.1.6.9 of ETSI TS 126 445 V13.2.0 (2016-08) cited in reference (4), where the result of the corresponding multi-rate lattice vector quantization is a codebook number for every 8 elements, a vector index in the base codebook, and an 8-dimensional Voronoi index. However, if only one codebook is available, the codebook number can be omitted, and only the vector index in the base codebook and the corresponding n-dimensional Voronoi index are required. Therefore, these elements—element a, element b, element c, only element c, and element c—for each level of the algebraic vector quantization result represent the second-stage quantization result. Subsequently, reference is made to Figure 10 which illustrates a corresponding decoding operation that coincides with the encoding of Figure 7a, 7b or the encoding of Figure 14 according to the first or second aspect of the present invention or according to both aspects. In step 2221 of Figure 10, the quantized mid-scale factors are retrieved; that is, the second group of jointly encoded scale factors. This is done when the stereo mode flag or element 1403 in Figure 5 indicates a true value. A first-stage decoding 2223 and a second-stage decoding 2261 are then performed to redo the procedures carried out by the encoder in Figure 14 and, in particular, by the algebraic vector quantizer 145 described with respect to Figure 14 or with respect to Figure 7a. In step 2225, it is assumed that the side scale factors are all 0. In step 2261, it is verified by means of the flag value 0 whether there are indeed non-zero quantized scale factors for the frame.If the flag value 0 indicates that there are non-zero side scale factors for the frame, then the quantized side scale factors are retrieved and decoded using second-stage decoding 2261 or by performing block 706 of Figure 7b only. In block 2207, the jointly encoded scale parameters are transformed back into separately encoded scale parameters, and the quantized left and right scale parameters are then output, which can then be used for inverse scaling of the spectrum in the decoder. When the stereo mode flag value indicates a value of zero or when it is determined that a separate encoding has been used within the frame, then only the first-stage decoding 2223 and the second-stage decoding 2261 are performed for the left and right scale factors, and since the left and right scale factors are already in the separately encoded representation, no such transformation is required ML / t / ZUZÓ / U Ί OU4y as block 2207. The efficient encoding and decoding process of the SNS scaling factors needed to modify the spectrum scale before stereo processing on the encoder side and after inverse stereo processing on the decoder side is described below to show a preferred implementation of the present invention as an example pseudocode with comments. Joint quantification and coding of scale factors Calculate lateral of the M scale factors of each channel snsly snsry calculate the total energy of the side enerside. ener_side=O; for (i =0; i < M; i++) { side[i] = sns l[i] - snsr[¡]; energ_side =ener_side + side[i]A2; } If the inside is less than a certain threshold, the two signals are highly correlated and the encoding must be done jointly or independently. if(ener_side <threshold) codificar factores de escala conjuntamente { Signal MS encoding to bitstream Calculate the mean of the M scale factors of each snsly snsR for (i = 0; i < M; i++) { mid[¡] = (snsl[i] + snsrfi]) * 0.5f; } Quantize mean with first-stage vector quantization (VQ), the function returns the stochastic codebook indexl_Ί and the intermediate quantized mean parameters mid_q. indexl_1 = sns_1st_cod( mid, mid_q ); Quantify mean with second-stage algebraic vector quantization (AVQ), the function returns split-dimension indices and the final quantized mean mid_q. indexl_2 = sns_2st_cod( mid, mid_q ); Quantize side - assume coarse quantization and set all quantized parameters to zero. for (i = 0; i < M; i++) { s ide_q[i] = Of} MA / t / ZUZÓ / U Ί OU4y Quantize sideq with second-stage algebraic vector quantization (AVQ), the function returns split-dimensional indices and the final quantized sideq. indexr_2 = sns_2st_cod( side, side_q); Detect if the quantized scale factors are zero; if so, signal it to the bitstream with a bit if (flag_zero) { send signal bit to bitstream} Transform the quantized scale factors back to the Ion LR representation for (i = 0; i < M; i++) { snsl_q[i] = mid_q[i] + side_q[i] * 0.5f; snsr_q[i] = mid_q[¡] - side_q[i] * 0.5f; }} else code scaling factors independently { Signal LR encoding to bitstream Quantify left channel scaling factors with first stage vector quantization (VQ), the function returns the stochastic codebook indexl_7 and the quantized snsl parameters snsjq indexl_1 = sns_1st_cod( snsl, snsl_q ); Quantify left channel scale factors with second-stage algebraic vector quantization (AVQ), the function returns split-dimensional indices and the final quantized snsl snsl q indexl_2 = sns_2st_cod( snsl, snsl_q ); Quantify right channel scaling factors with first stage vector quantization (VQ), the function returns the stochastic codebook indexip 7 and the quantized snsr parameters snsr_q indexr_1 = sns_1 st_cod( snsr, snsr_q); Quantify right channel scale factors with second-stage algebraic vector quantization (AVQ), the function returns split-dimensional indices and the final quantized snsr snsr q indexr_2 = sns_2st_cod( snsr, snsr_q ); } Emit quantized S NS scaling factors snsl_qy snsr_q to enhance spectrum scaling modification. ML7 t / ZUZO / U IOU4U Any type of quantization can be used, for example, uniform or non-uniform scalar quantization and arithmetic or entropy coding, to represent the parameters. In the described implementation, as can be seen in the algorithm description, a 2-stage vector quantization scheme is implemented: • First stage: 2 divisions (8 dimensions each) with 5 bits each, therefore encoded with 10 bits • Second stage: algebraic vector quantization (AVQ), again 2 divisions with remainder scaling modification, where the codebook indices are encoded by entropy and therefore uses a variable bit rate. Since the side signal for highly correlated channels can be considered small, using, for example, a reduced-scale 2-stage AVQ is sufficient to represent the corresponding SNS parameters. By omitting the 1-stage VQ for these signals, significant complexity and bit savings can be achieved in the encoding of the SNS parameters. A pseudocode description of each implemented quantization stage is provided below. First stage with vector quantization of 2 divisions using 5 bits for each division: ΜΛ / t / ZUZÓ / U Ί OU4y codebook index = sns 1 st cc¿( input : sns parameters vector to quantize output : sns_q quantized sns scale to meter ) { divide the vector coefficients ci in half jO = 0; j1 = M / 2; initialize minimum distance dist_min = 1.0e30f; pointer to memory location of stored codebooks p = sns_vq_cdbk1; indexO = 0; Split vector quantization Use 5-bit representation 32=2^5 to find the optimal index with the minimum distance for (i = 0; i < 32; i++ ) { dist = 0.0; for(j = j0;j <j1;j++) { Obtain the difference of the sns parameters with each of the quantized 8-dimensional vectors, which are stored sequentially in memory. temp = sns[j] - *p++; calculate distance dist = dist + temp * temp; } return codebook index with minimum distance if ( dist < dist_min ) { dist_min = dist; indexO = i; }} Having found the optimal vector index, obtain quantified values ​​of the first M / 2 S NS scale factors from the codebook Will point the address in memory to the selected codebook p = &sns_vq_cdbk1 [indexO * ( M / 2 )]; for (j =j0; j <j1;j++) { snsq[j] = *p++; Increment pointer by one} Repeat the procedure for the second division of the vector jO =M / 2; ML / t / ZUZÓ / U IOU4U j1 = ML; dist_min = 1.0e30f; p = sns_vq_cdbk2; index 1 = 0; for (i = 0; i < 32; i++) { dist = 0.0; for(j =j0;j <j1;j++) { temp = sns[j] - *p_dico++; dist += temp * temp; } if ( dist < dist_min ) { dist_min = dist; index 1 = i; }} Get [the quantized values ​​for the remaining factors of the codebook p = &sns_vq_cdbk2[index1 * ( M / 2 )]; for (j =j0; j <j1;j++) { snsqlj] =*p++; } The final index is the sum of the indices of the first division + second division multiplied by the factor 2Λ5=32. Therefore, only one index needs to be multiplexed in the bit stream index = indexO + (index! << 5 ); MA / IZ / ZUZO / U1 OU4y return index; Second-stage algebraic vector quantization: inputsns, normalized vector to quantify input / output snsq, i: first stage o: 1st+2nd stage output index[], ) { scale = 1.0 / 2.5; Calculate first stage quantization residue and residual scale for finer quantization for ( i = 0; i < M; i++ ) { x[i] = (sns[¡] - snsq[¡]) / scale; } Quantify the remainder using AVQ (algebraic code vector) used in EVS for second-stage quantization of LPC coefficients [4], where x is the remainder, xq is the quantized remainder returned from the function, 2 marks the 2-split process, and indx is an array containing the codebook indices for each split AVQ_cod_lpc( x, xq, indx, 2 ); Refine the quantified S NS scale factors by adding the quantized residue that concludes the second quantization stage for (i = 0; i < M; i++) { snsq[¡] = snsq[i] + scale * xq[¡]; }} The indices emitted from the encoding process are finally packaged into the bitstream and sent to the decoder. The AVQ procedure disclosed earlier for the second stage is preferably implemented as described in EVS with reference to high-speed LPC (subclause 5.3.3.2.1.3) in the MDCT-based TCX chapter. Specifically, the algebraic vector quantizer used in the second stage is indicated in 5.3.3.2.1.3.4, and the algebraic VQ used to quantize the refinement is described in subclause 5.2.3.1.6.9. In one modality, for each index, there is a set of codewords for the base codebook index and a set of codewords for the Voronoi index, all of which are entropy-encoded and therefore variable bit-rate.Therefore, the AVQ parameters in each subband j consist of the codebook number, the vector index in the base codebook, and the n- (such as 8-) dimensional Voronoi index. Decoding of scaling factors. ML / t / ZUZÓ / U 1 OU4y At the decoder end, the indices are extracted from the bitstream and used to decode and derive the quantized values ​​of the scale factors. A pseudocode example of the procedure is provided later. The 2-stage decoding procedure is described in detail in the pseudocode below. MA / t / ZUZÓ / U 1 OU4y Read the stereo bit signaling encoding of the hits stream if (stereo_mode is true ) { Read Indices to retrieve the quantified inmedia scale factors. From first stage encoding, enter indexl I and return quantized mean, honey q sns_1st_dec( indexl_1, mid_q); Second stage decoding, introduce indices indexl_2 and return final quantized mean mid_c\ sns_2st_dec( mid_q, indexl_2 ); Assume that the quantified lateral scale factors are zero after the first stage for (i=0; i <M; i++) { s ide_q[i] = O.t} If the bitstream signals that the lateral scale factors are not zero, perform second-stage decoding if (flag_zero is false) { Introduce second-stage indices indexr_2 and return quantized side, side_q sns_2st_dec( side_q, indexr_2 ); } Transform quantified scale factors S NS lateral-media to LR for (i = 0; i < M; i++) { S NS_QI[i] = mid_q[i] + side_q[i] * 0.5f; SNS_Qr[¡] = mid_q[¡] - side_q[i] *0.5f; }} else { Two-stage decoding to retrieve quantified scale factors S NS LR First-stage decoding L sns_1st_dec( indexl_1, SNS_QI ); Second stage decoding L sns_2st_dec( SNS_QI, indexl); First stage decoding R sns_1 st_dec( *indexr++, S NS_Qr); Second stage decoding R [1] sns_2st_dec( SNS_Qr, indexr); } Return quantified scaling factors for each channel to modify the scale of the decoded spectrum The sns 2st dec( input / output snsq, i: lera stage o: lera+2nd stage inputindx, i: index[j (4 bits per word) ) { floatscale = 1.0 / 2.5; Derive from indices índx ios M quantized residues xq, of the AVQ decoding function of 2 divisions. AVQ_dec_lpc( indx, xq, 2 ); Reconstruct the final quantized SNS parameters by adding the modified residuals on a for (i = 0; i < M; i++) { snsq[i] = snsq[¡] + scale * (float) xq[i]; }} Regarding scaling or amplification / weighting of the residual on the encoder side and scaling or attenuation / weighting on the decoder side, the weighting factors are not calculated separately for each value or division. Instead, a single weight or a small number of different weights (as an approximation to avoid complexity) is used to scale all parameters. This scaling is a factor that determines the balance between, for example, coarse quantization (more zero quantizations), bit rate savings, and quantization accuracy (with corresponding spectral distortion). It can be predetermined in the encoder so that this default value does not need to be transmitted to the decoder, but it can be fixed or initialized in the decoder to save transmission bits.Therefore, further scaling of the residue would require more bits but would have minimal spectral distortion, whereas downscaling would save additional bits, and if the spectral distortion is kept within an acceptable range, that could serve as a means of additional bit rate savings. Advantages of Preferred Modalities • Substantial bit savings when two channels are correlated and SNS parameters are co-coded. An example of the bit-per-frame savings achieved in the system described in the previous section is shown below: • Independent: 88.1 bits on average • New independent: 72.0 bits on average • New set: 52.1 bits on average where • “Independent” is the MDCT-stereo implementation described in [8] using SNS [6] for the FDNS encoding ML / t / ZUZÓ / U Ί OU4y the two channels only independently with 2-stage VQ • First stage: 8-bit trained codebook (16 dimensions) • Second stage: AVQ of the residue scaled by a factor of 4 (variable bit rate) • New independent refers to the above-described modality of the invention where the correlation of the two channels is not high enough and they are encoded separately, using a new 2-stage VQ approach as described above and the residue is scaled by a reduced factor of 2.5 • “New set” refers to the jointly encoded case (also described above), where again in the second stage the residue is scaled by a reduced factor of 2.5. • Another advantage of the proposed method is the reduction in computational complexity. As shown in [6], the new SNS is more computationally optimal than the LPC-based FDNS described in [5] due to the autocorrelation calculations required to estimate the LPCs. Therefore, when comparing the computational complexity of the MDCT-based stereo system in [8] using the improved LPC-based FDNS [5] to an implementation where the new SNS [6] replaces the LPC-based approach, there are savings of approximately 6 WMOPS at a sampling rate of 32 kHz. Furthermore, the new two-stage quantization with VQ for the first stage and reduced-scale AVQ for the second stage achieves a further reduction in computational complexity. For the modality described in the previous section, the computational complexity is further reduced by approximately 1 WMOPS at a sampling rate of 32 kHz, with acceptable spectral distortion compensation. Summary of Aspects or Preferred Modalities 1. Joint coding of spectral noise shaping parameters, where the mean / lateral representation of the parameters is calculated and the mean is coded using quantization and entropic coding and the lateral is coded using a coarser quantization scheme. 2. Adaptively determine whether noise shaping parameters should be coded independently or jointly based on channel correlation or coherence. 3. Signaling bit sent to determine if the parameters were encoded independently or jointly. 4. Applications based on the MDCT-stereo implementation: • with bits where the side coefficients are zero • that the SNS is used • that the power spectrum is used to calculate the SNS • that 2 divisions with 5 bits are used in the first stage. • Adjusting the scaling of the second-stage AVQ residue can further reduce the number of bits for second-stage quantization. Figure 23 illustrates a comparison of the number of bits for both channels in line with an implementation of the current previous technique (described as independent above), the new independent implementation of MA / t / ZUZÓ / U 1 OU4y in accordance with the second aspect of the present invention and for the new joint implementation in accordance with the first aspect of the present invention. Figure 23 illustrates a histogram where the vertical axis represents the frequency of occurrence and the horizontal axis illustrates the intervals of the total number of bits to encode the parameters for both channels. Subsequently, additional preferred modalities are illustrated where specific emphasis is given to the calculation of the scale factors for each audio channel and where further specific emphasis is given to the specific application of undersampling or oversampling of the scale parameters, which is applied either before or after the calculation of the jointly encoded scale parameters as illustrated with respect to Figure 3a, Figure 3b. Figure 11 illustrates an apparatus for encoding an audio signal 160. The audio signal 160 is preferably available in the time domain, although other representations of the audio signal, such as a prediction domain or any other domain, would also be useful. The apparatus comprises a converter 100, a scaling factor calculator 110, a spectral processor 120, a subsampler 130, a scaling factor encoder 140, and an output interface 150. The converter 100 is configured to convert the audio signal 160 into a spectral representation. The scaling factor calculator 110 is configured to calculate a first set of scaling parameters or scaling factors from the spectral representation. The other channel is received in block 120, and the scaling parameters of the other channels are received by block 140. Throughout this specification, the term “scale factor” or “scale parameter” is used to refer to the same parameter or value; that is, a value or parameter used, after processing, to weight certain spectral values. This weighting, when performed in the linear domain, is actually a multiplication operation with a scale modification factor. However, when weighting is performed in a logarithmic domain, the weighting operation with a scale factor is performed by an actual addition or subtraction operation. Therefore, for the purposes of this application, scaling not only means multiplication or division, but also, depending on the specific domain, addition or subtraction, or generally any operation by which the spectral value, for example, is weighted or modified using the scale factor or scale parameter. The subsampler 130 is configured to subsample the first set of scale parameters to obtain a second set of scale parameters, where a second number of the scale parameters in the second set of scale parameters is less than a first number of scale parameters in the first set of scale parameters. This is also described in the inset of Figure 11, which indicates that the second number is less than the first number. As illustrated in Figure 11, the scale factor encoder is configured to generate a coded representation of the second set of scale factors, and this coded representation is forwarded to the output interface 150.Due to the fact that the second set of scale factors has a smaller number of scale factors than the first set of scale factors, the bit rate for transmitting or storing the encoded representation of the second set of scale factors is lower compared to a situation in the. ML / t / ZUZÓ / U Ί OU4y that the subsampling of the scale factors performed in subsampler 130 would not have been carried out. Furthermore, the spectral processor 120 is configured to process the spectral rendering output by the converter 100 in Figure 11 using a third set of scaling parameters, the third set of scaling parameters or scaling factors having a third number of scaling factors that is greater than the second number of scaling factors, wherein the spectral processor 120 is configured to use, for the purpose of spectral processing, the first set of scaling factors as already available from block 110 through line 171. Alternatively, the spectral processor 120 is configured to use the second set of scaling factors as output by the subsampler 130 for the calculation of the third set of scaling factors as illustrated by line 172.In a further implementation, the spectral processor 120 uses the encoded representation output of the scale factor / parameter encoder 140 for the purpose of calculating the third set of scale factors, as illustrated by line 173 in Figure 11. Preferably, the spectral processor 120 does not use the first set of scale factors, but instead uses the second set of scale factors calculated by the subsampler, or even more preferably, it uses the encoded representation, or generally, the second quantized set of scale factors, and then performs an interpolation operation to interpolate the second quantized set of spectral parameters to obtain the third set of scale parameters, which has a larger number of scale parameters due to the interpolation operation. Therefore, the coded representation of the second set of scale factors generated by block 140 comprises a codebook index for a preferably used scale parameter codebook or a set of corresponding codebook indices. In other embodiments, the coded representation comprises the quantized scale parameters of quantized scale factors obtained when the codebook index, the set of codebook indices, or, more generally, the coded representation is input into a decoder-side vector decoder or any other decoder. Preferably, the spectral processor 120 uses the same set of scaling factors that is also available on the decoder side, i.e., it uses the second quantized set of scaling parameters together with an interpolation operation to finally obtain the third set of scaling factors. In a preferred embodiment, the third number of scale factors in the third set of scale factors is equal to the first number of scale factors. However, a smaller number of scale factors is also useful. For example, 64 scale factors could be derived in block 110 and then subsampled from these 64 scale factors to 16 scale factors for transmission. Interpolation could then be performed not necessarily to 64 scale factors, but to 32 scale factors in spectral processor 120. Alternatively, interpolation could be performed to an even larger number, such as more than 64 scale factors, as appropriate, provided that the number of scale factors transmitted in the encoded output signal 170 is less than the number of scale factors calculated in block 110 or calculated and used in block 120 of Figure 11. Preferably, the 110 scale factor calculator is configured to perform several illustrated operations MA / t / ZUZÓ / U Ί OU4y in Figure 12. These operations refer to a calculation 111 of a band-related amplitude measurement, where the spectral representation for one channel is entered in block 111. The calculation for the other channel will be performed similarly. A preferred band-related amplitude measurement is band-related energy, but other amplitude-related measurements can also be used, for example, the sum of the magnitudes of the band-related amplitudes or the sum of squared amplitudes, which corresponds to energy. However, apart from the power of 2 used to calculate band-related energy, other powers can also be used, such as a power of 3 that would reflect the loudness of the signal, and even different powers of integers, such as powers of 1.5 or 2.5, can be used to calculate band-related amplitude measurements. Even powers less than 1 can be used.0 provided that it is ensured that the values ​​processed by these powers are positive. An additional operation performed by the scaling factor calculator can be interband smoothing 112. This interband smoothing is preferably used to smooth out any instabilities that may appear in the vector of amplitude-related measurements obtained by step 111. If this smoothing were not performed, these instabilities would be amplified when converted to a logarithmic domain later, as illustrated in 115, especially at spectral values ​​where the energy is close to 0. However, in other modes, interband smoothing is not performed. Another preferred operation performed by the scale factor calculator 110 is the pre-emphasis operation 113. This pre-emphasis operation serves a similar purpose to a pre-emphasis operation used in an LPC-based perceptual filter of MDCT-based TCX processing, as discussed earlier in relation to the previous technique. This procedure increases the amplitude of the conformal spectrum at low frequencies, resulting in reduced quantization noise at low frequencies. However, depending on the implementation, the pre-emphasis operation - like the other specific operations - does not necessarily have to be performed. Another optional processing operation is the minimum noise threshold addition processing 114.This procedure improves the quality of signals containing very high spectral dynamics, such as, for example, glockenspiel, by limiting the amplitude amplification of the shaped spectrum in the valleys. This has the indirect effect of reducing quantization noise at the peaks, at the cost of increased quantization noise in the valleys. In the valleys, quantization noise is not perceptible anyway due to the masking properties of the human ear, such as the absolute hearing threshold, pre-masking, post-masking, or the general masking threshold. This indicates that, typically, a fairly low-volume tone relatively close in frequency to a high-volume tone is not perceptible at all; that is, it is completely masked or only approximately perceived by the human auditory mechanism. Thus, this spectral contribution can be quantized rather roughly. However, the minimum noise threshold addition operation 114 does not necessarily have to be performed. Furthermore, block 115 indicates a logarithmic domain conversion. Preferably, a transformation of MA / t / ZUZÓ / U Ί OU4y, the output of one of the blocks 111, 112, 113, 114 in Figure 12 is implemented in a logarithmic domain. A logarithmic domain is a domain in which values ​​close to 0 are expanded and large values ​​are compressed. Preferably, the logarithmic domain has a base of 2, but other logarithmic domains can also be used. However, a logarithmic domain with a base of 2 is better for implementation in a fixed-point signal processor. The output of the scale factor calculator 110 is a first set of scale factors. As illustrated in Figure 12, each of blocks 112 to 115 can be combined; that is, the output of block 111, for example, could already be the first set of scaling factors. However, all processing operations, and particularly the logarithmic domain conversion, are preferred. Therefore, the scaling factor calculator could even be implemented by performing only steps 111 and 115 without the procedures in steps 112 to 114, as an example. The output of block 115 yields a set of scaling parameters for one channel (such as L), and a similar calculation can yield a set of scaling parameters for the other channel (such as R). Therefore, the scale factor calculator is configured to perform one, two, or more of the procedures illustrated in Figure 12 as indicated by the input / output lines connecting various blocks. Figure 13 illustrates a preferred implementation of the subsampler 130 of Figure 11 again for a single channel. The data for the other channel are calculated similarly. Preferably, low-pass filtering, or more generally, filtering with a certain window w(k), is performed in step 131, and then a downsampling operation is performed on the filtered result. Because the low-pass filtering 131 and, in preferred embodiments, the downsampling operation 132 are both arithmetic operations, the filtering 131 and the downsampling 132 can be performed within a single operation, as will be described later. Preferably, the downsampling operation is performed in such a way as to overlap the individual groups of scale parameters of the first set of scale parameters.Preferably, a scaling factor is superimposed on the filtering operation between two downscaled calculated parameters. Therefore, step 131 performs a low-pass filter on the scaling parameter vector before the downscaling. This low-pass filter has an effect similar to the propagation function used in psychoacoustic models. It reduces quantization noise at the peaks, at the cost of increased quantization noise around the peaks, where it is perceptually masked to a greater degree than the quantization noise at the peaks themselves. Furthermore, the subsampler additionally performs an average removal step 133 and an additional scaling step 134. However, the low-pass filtering operation 131, the average removal step 133, and the scaling step 134 are only optional steps. Therefore, the subsampler illustrated in Figure 13 or illustrated in Figure 11 can be implemented to perform only step 132 or to perform two steps illustrated in Figure 13, such as step 132 and one of steps 131, 133, and 134. Alternatively, the ML / t / ZUZÓ / U 1 OU4y subsampler can perform all four steps or only three of the four steps illustrated in Figure 13 as long as the subsampling / reduction operation 132 is performed. As described in Figure 13, the audio operations in Figure 13 performed by the subsampler are performed in the record-type domain in order to obtain better results. Figure 15 illustrates a preferred implementation of the spectral processor. The spectral processor 120, included within the encoder of Figure 11, comprises an interpolator 121 that receives the second quantized set of scale parameters for each channel or, alternatively, for a group of jointly encoded scale parameters, and generates the third set of scale parameters for a channel or for a group of jointly encoded scale parameters, where the third number is greater than the second number and preferably equal to the first number. Furthermore, the spectral processor comprises a linear domain converter 120. Spectral shaping is then performed in block 123 using the linear scale parameters on one hand and the spectral representation obtained from the converter 100 on the other.Preferably, a subsequent temporal noise shaping operation, i.e., a prediction on frequency, is performed in order to obtain spectral residual values ​​at the output of block 124, while additional TNS information is forwarded to the output interface as indicated by arrow 129. Finally, the spectral processor 125,120b has at least one scalar quantizer / encoder configured to receive a single global gain for the entire spectral representation (i.e., for a complete frame), as well as stereo processing functionality and IGF processing functionality, etc. Preferably, the global gain is derived based on certain bitrate considerations. Therefore, the global gain is set so that the encoded representation of the spectral representation generated by block 125,120b meets certain requirements, such as a bitrate requirement, a quality requirement, or both. The global gain can be calculated iteratively or in a feed-forward measurement, as appropriate.Global gain is generally used in conjunction with a quantizer, and a high global gain typically results in coarser quantization, while a low global gain results in finer quantization. Therefore, in other words, a high global gain results in a larger quantization step size, whereas a low global gain results in a smaller quantization step size when using a fixed quantizer. However, other quantizers can also be used in conjunction with global gain functionality, such as a quantizer that incorporates some form of compression for high values—that is, some form of nonlinear compression so that, for example, higher values ​​are compressed more than lower values.The above dependence between global gain and quantization roughness holds true when global gain is multiplied by pre-quantization values ​​in the linear domain, corresponding to addition in the logarithmic domain. However, if global gain is applied by division in the linear domain or by subtraction in the logarithmic domain, the dependence is reversed. The same is true when global gain represents an inverse value. MA / t / ZUZÓ / U Ί OU4y Subsequently, preferred implementations of the individual procedures described with respect to Figures 11 to 15 are provided. Detailed step-by-step description of preferred modes ENCODER: MA / t / ZUZÓ / U Ί OU4U * Pasol: Energy per band (111) The band energies EB(n) are calculated as follows: for b = 0 ...NB- 1 where are the MDCT coefficients, NB= 64 is the number of bands and Ind(n) are the band indices. The bands are not uniform and follow the perceptually relevant Bark scale (smaller at low frequencies, larger at high frequencies). • Step 2: Smoothing (112) The energy per band EB(b) is smoothed using ^.75-^(0) + 0.25-^(1) Es(b) = 0.25 Eb(62) + 0.75 Es(63) , if b = O , if b = 63 0.25 EB(b — 1) + 0.5 EB(b) + 0.25 EB(b + 1) , otherwise Note: This step is mainly used to smooth out possible instabilities that may appear in the EB(bf) vector. If not smoothed, these instabilities are amplified when converted to a logarithmic domain (see step 5), especially in the valleys where the energy is close to 0. • Step 3: Pre-emphasis (113) The band-smoothed energy Es(b) is pre-emphasized using b-gtut Ep^b) = Es(b~) 1010 63 for b= 0. .63 with gtM controls the pre-emphasis tilt and depends on the sampling frequency. Es, for example, 18 at 16 kHz and 30 at 48 kHz. The pre-emphasis used in this step has the same purpose as the pre-emphasis used in the LPC-based perception filter of the previous technique 2, increasing the amplitude of the shaped spectrum at low frequencies, which results in reduced quantization noise at low frequencies. * Step 4: Minimum noise threshold (114) A minimum noise threshold of -40dB is added for EP(b~) using EP(b) = max(EP(bfnoiseFloor) for b = 0. .63 with the minimum noise threshold calculated by ÍZuncleEpW40noiseFloor = max ----—--10 io, 232) \ 64 / This step improves the quality of signals containing very high spectral dynamics such as, for example, glockenspiel, by limiting the amplitude amplification of the shaped spectrum in the valleys, which has the indirect effect of reducing quantization noise at the peaks, at the cost of an increase in quantization noise in the valleys where it is not noticeable anyway. IVIA / t / ZUZÓ / UI OU4y • Step 5: Logarithm (115) Then a transformation is performed in the logarithmic domain using for b = 0.. 63 • Step 6: Subsampling (131,132) The vector EL(b) is then reduced by a factor of 4 using r 5 w(0)E¿(0) + w( / c) EL(4b + k — 1) k=i E^b) = < w(k) EL(4-b -I- k - 1) + w(5)EL(63) k=O w(k) EL(4-b + k — 1) Vc=0 ,if b = 0 , if b = 15 , otherwise With z. (123321) w(k) = j—, —, —, —, —,—fv J112 12 12 12 12 12J This step applies a low-pass filter (w(k)) to the vector E^b) before the decay. This low-pass filter has an effect similar to the propagation function used in psychoacoustic models: it reduces quantization noise at the peaks, at the cost of increased quantization noise around the peaks where it is perceptually masked anyway. * Step 7: Mean Removal and Scale Modification (133,134) The final scale factors are obtained after mean removal and scaling by a factor of 0.85 scf(n) = 0.85 IE^n) — J for n = 0.. 15 Since the codec has additional global gain, the average can be removed without any loss of information. Removing the average also allows for more efficient vector quantization. The scaling modification of 0.85 slightly compresses the amplitude of the noise shaping curve. This has a perceptual effect similar to the propagation function mentioned in Step 6: reduced quantization noise at the peaks and increased quantization noise at the valleys. * Step 8: Quantification (141,142) Scale factors are quantized using vector quantization, producing indices that are then packed into the bitstream and sent to the decoder, and quantized scfQ(n) scale factors. * Step 9: Interpolation (121,122) The quantified scale factors scfQ(n) are interpolated using sc / Qint(0) = scfQ(0) scfQint(X) = scfQ(0) scfQintC4n + 2) = scfQ(n) + — (scfQ(n + 1) — scfQínY)vz scfQint(4n + 3) = scfQ(n) + ~(scfQ(n + 1) — scfQ(n)) z scfQint(4n + 4) = scfQ(n) + ~(scfQ(n + 1) - scfQ(n)) 8vz scfQint(4n + 5) = scfQ(n) + -(scfQ(n + 1) - scfQ(n))vpara para para para 0..14 0..14 0..14 0..14 MA / t / ZUZÓ / U Ί OU4y scfQint(62) = scfQ(15) + - (sc / Q(15) - scfQ^l^} O z scfQint{63') = scfQQlS) + -(sc / Q(15) — sc / (?(14)) 8vy transformed back into linear domain using 9snsW = 2scf^int^ for b = 0. .63 Interpolation is used to obtain a smooth noise shaping curve and thus avoid any large amplitude jumps between adjacent bands. • Step 10: Spectral conformation (123) The SNS gSNsW scaling factors are applied to the MDCT frequency lines for each band separately in order to generate the XSW conformal spectrum XW XsW=----77R para k = Ind(Jf)..Ind(b + 1) — 1, para b = 0..63 dsNsW Figure 18 illustrates a preferred implementation of an apparatus for decoding an encoded audio signal 250 (a stereo signal encoded as L, R or M, S) comprising information about an encoded spectral representation and information about an encoded representation of a second set of scaling parameters (encoded separately or jointly). The decoder comprises an input interface 200, a spectrum decoder 210 (e.g., performing IGF processing or inverse stereo processing or dequantization processing), a scaling factor / parameter decoder 220, a spectral processor 230 (e.g., for R, L), and a converter 240 (e.g., for R, L).Input interface 200 is configured to receive the encoded audio signal 250 and to extract the encoded spectral representation, which is forwarded to spectrum decoder 210. It also extracts the encoded representation of the second set of scale factors, which is forwarded to scale factor decoder 220. Spectrum decoder 210 is configured to decode the encoded spectral representation to obtain a decoded spectral representation, which is forwarded to spectral processor 230. Scale factor decoder 220 is configured to decode the second encoded set of scale parameters to obtain a first set of scale parameters, which is forwarded to spectral processor 230. The first set of scale factors has a greater number of scale factors or parameters than the second set.The spectral processor 230 is configured to process the decoded spectral representation using the first set of scaling parameters to obtain a scaled-up spectral representation. The converter 240 then converts the scaled-up spectral representation to finally obtain the decoded audio signal 260, which is either a stereo signal or a multichannel signal with more than two channels. Preferably, the scale factor decoder 220 is configured to operate substantially in the same manner as discussed with respect to the spectral processor 120 in Figure 11 in relation to the calculation of the third set of scale factors or scale parameters as discussed in relation to blocks 141 or 142 and, particularly, with respect to blocks 121, 122 in Figure 15. Specifically, the scale factor decoder is configured to perform substantially the same procedure for interpolation and transformation back into the linear domain as discussed above with respect to step 9. Therefore, as illustrated in Figure 19, the scale factor decoder 220 is configured to apply a decoding codebook 221 to one or more indexes per frame representing the encoded scale parameter representation.Next, an interpolation is performed in block 222, which is substantially the same interpolation that has been analyzed with respect to block 121 in Figure 15. Then, a linear domain converter 223 is used, which is substantially the same linear domain converter 122 that has been analyzed with respect to Figure 15. However, in other implementations, blocks 221, 222, and 223 may operate differently than what has been analyzed with respect to the corresponding blocks on the encoder side. Furthermore, the spectrum decoder 210 illustrated in Figure 18 or 19 comprises a dequantizer / decoder block that receives the encoded spectrum as input and outputs a dequantized spectrum, preferably dequantized using the overall gain transmitted from the encoder side to the decoder side within the encoded audio signal. Block 210 may also perform IGF processing or inverse stereo processing such as MS decoding. The dequantizer / decoder 210 may, for example, include an arithmetic or Huffman decoder function that receives some type of code as input and produces quantization indices representing spectral values.These quantization indices are then fed into a dequantizer along with the overall gain, and the output are dequantized spectral values ​​that can then be subjected to TNS processing, such as inverse frequency prediction, in a TNS decoder processing block 211, which is optional. In particular, the TNS decoder processing block also receives the additional TNS information generated by block 124 in Figure 15, as indicated by line 129. The output of the TNS decoder processing step 211 is fed into a spectral shaping block 212, which operates for each channel separately using separate scaling factors. The first set of scaling factors calculated by the scale factor decoder is applied to the decoded spectral representation, which may or may not be further processed. MA / t / ZUZÓ / U Ί OU4y by TNS as appropriate, and the output is the scaled modified spectral representation for each channel that is then entered into the converter 240 of figure 18. Subsequently, additional procedures for preferred decoder modes are analyzed. DECODER: * Step 1: Quantification (221) The vector quantizer indices produced in encoder step 8 are read from the bitstream and used to decode the quantized scaling factors scfQ(n). * Step 2: Interpolation (222,223) Same as step 9 of the encoder. * Step 3: Spectral conformation (212) The SNS scaling factors (gSNsCb~) are applied to the MDCT frequency lines quantized for each band separately in order to generate the decoded spectrum X(k) as indicated in the following code. X^k) = Xs(kj gsNs(b) para k= Ind^by.Ind^b + 1) — 1, para b= 0..63 Figure 16 and Figure 17 illustrate a general encoder / decoder configuration, where Figure 16 represents an implementation without TNS processing, while Figure 17 illustrates an implementation that includes TNS processing. Similar functionalities illustrated in Figure 16 and Figure 17 correspond to similar functionalities in the other figures when identical part numbers are indicated. In particular, as illustrated in Figure 16, the input signal 160, for example, a stereo or multichannel signal, is fed into a transform stage 110, and subsequently, spectral processing 120 is performed. Specifically, the spectral processing is reflected by an SNS encoder indicated by part numbers 123, 110, 130, and 140, which indicates that the block SNS encoder implements the functionalities indicated by these part numbers.Following the SNS encoder block, a quantization encoding operation 120b, 125 is performed and the encoded signal is entered into the bit stream as indicated in 180 in Figure 16. Then, the bit stream 180 is produced on the decoder side and after an inverse quantization and decoding illustrated by reference number 210, the SNS decoder operation illustrated by blocks 210, 220, 230 in Figure 18 is performed so that, in the end, after an inverse transform 240, the decoded output signal 260 is obtained. Figure 17 illustrates a representation similar to that in Figure 16, but indicates that, preferably, TNS processing is performed after SNS processing on the encoder side and, correspondingly, TNS processing 211 is performed before SNS processing 212 with respect to the processing sequence on the decoder side. Preferably, the additional TNS tool is used between spectral noise shaping (SNS) and quantization / encoding (see block diagram below). TNS (temporal noise shaping) also shapes quantization noise, but it also performs time-domain shaping (unlike the frequency-domain shaping of SNS). TNS is useful for signals containing sharp attacks and for signals ΜΛ / t / ZUZÓ / U Ί OU4y of voice. TNS is generally applied (in AAC, for example) between the transform and SNS. Preferably, however, TNS is applied to the shaped spectrum. This avoids some artifacts produced by the TNS decoder when operating the codec at low bit rates. Figure 20 illustrates a preferred subdivision of the spectral coefficients or spectral lines as obtained by block 100 on the band-encoder side. In particular, it indicates that the lower bands have fewer spectral lines than the upper bands. Specifically, the x-axis in Figure 20 corresponds to the band index and illustrates the preferred 64-band mode, while the y-axis corresponds to the spectral line index, illustrating 320 spectral coefficients in a table. Figure 20 specifically illustrates, as an example, the super-wideband (SWB) case with a sampling frequency of 32 kHz. In the case of broadband, the situation with respect to the individual bands is such that one frame results in 160 spectral lines and the sampling frequency is 16 kHz so that, for both cases, one frame has a time length of 10 milliseconds. Figure 21 illustrates further details of the preferred subsampling performed in the subsampler 130 of Figure 11 or the corresponding oversampling or interpolation as performed in the scale factor decoder 220 of Figure 18 or as illustrated in block 222 of Figure 19. Along the x-axis, the index for bands from 0 to 63 is given. In particular, there are 64 bands ranging from 0 to 63. The 16 subsampled points corresponding to scfQ (i) are illustrated as vertical lines 1100. In particular, Figure 21 illustrates how a certain grouping of scale parameters is carried out to finally obtain the subsampled point 1100. As an example, the first block of four bands consists of (0,1,2,3) and the midpoint of this first block is at 1.5 indicated by the element 1100 at index 1.5 along the x-axis. Correspondingly, the second block of four bands is (4,5,6, 7), and the midpoint of the second block is 5.5. The 1110 windows correspond to the w(k) windows analyzed with respect to the subsampling step 6 described above. It can be observed that these windows are centered on the subsampled points and there is an overlap of one block on each side, as analyzed previously. The interpolation step 222 in Figure 19 recovers the 64 bands from the 16 subsampled points. This is shown in Figure 21 when calculating the position of any of the 1120 lines based on the two subsampled points indicated in 1100 around a given 1120 line. The following example illustrates this. The position of the second band is calculated based on the two vertical lines that surround it (1.5 and 5.5): 2=1.5+1 / 8x(5.5-1.5). Correspondingly, the position of the third band based on the two vertical lines 1100 around it (1.5 and 5.5): 3=1.5+3 / 8x(5.5-1.5). MA / t / ZUZÓ / U Ί OU4y A specific procedure is performed for the first two bands and the last two bands. For these bands, interpolation cannot be performed because there would be no vertical lines or values ​​corresponding to vertical lines 1100 outside the range of 0 to 63. Therefore, to address this problem, extrapolation is performed as described in step 9: interpolation as described above for the two bands 0,1 on one hand and 62 and 63 on the other. Subsequently, a preferred implementation of the converter 100 in Figure 11 is analyzed on the one hand, and the converter 240 in Figure 18 on the other hand. In particular, Figure 22a illustrates a program for indicating the framing performed on the encoder side within converter 100. Figure 22b illustrates a preferred implementation of converter 100 of Figure 11 on the encoder side, and Figure 22c illustrates a preferred implementation of converter 240 on the decoder side. The converter 100 on the encoder side is preferably implemented to produce a frame with overlapping frames, such as a 50% overlap, where frame 2 overlaps with frame 1, and frame 3 overlaps with frames 2 and 4. However, other overlaps or non-overlapping processing are also possible, but a 50% overlap is preferred, along with an MDCT algorithm. To this end, the converter 100 comprises an analysis window 101 and a downstream spectral converter 102 to perform FFT processing, MDCT processing, or any other type of time-to-spectrum conversion processing to obtain a frame sequence corresponding to a sequence of spectral representations, as shown in Figure 11, in the blocks downstream of the converter 100. Accordingly, the scaled spectral representations are fed into the converter 240 of Figure 18. Specifically, the converter comprises a time converter 241 that implements either an inverse FFT operation, an inverse MDCT operation, or a corresponding spectrum-to-time conversion operation. The output is fed into a synthesis window 242, and the output of the synthesis window 242 is fed into a superposition-sum processor 243 to perform a superposition-sum operation to ultimately obtain the decoded audio signal.In particular, the overlay-sum processing in block 243, for example, performs a sample-by-sample addition between the corresponding samples of the second half of, for example, frame 3 and the first half of frame 4 so that the audio sample values ​​for the overlay between frame 3 and frame 4 are obtained as indicated by element 1200 in figure 22a. Similar overlay-sum operations are performed in a sample-by-sample manner to obtain the remaining audio sample values ​​of the decoded audio output signal. It will be mentioned here that all the alternatives or aspects as discussed above and all the aspects as defined by independent claims in the following claims may be used individually, that is, without any other alternative or object than the alternative, object, or independent claim contemplated. However, in other embodiments, two or more of the alternatives or aspects or independent claims may be combined with each other, and in still other embodiments, all the aspects or alternatives and all the MA / t / ZUZÓ / U Ί OU4y independent claims. Although further aspects are described above, the appended claims indicate two distinct aspects, namely, an audio decoder, an audio encoder, and related methods using co-encoding of scaling parameters for channels of a multichannel audio signal, or an audio quantizer, an audio dequantizer, and related methods. These two aspects may be combined or used separately, as the case may be, and the inventions according to these aspects are applicable to audio processing applications other than the specific applications described above. In addition, reference is made to additional figures 3a, 3b, 4a, 4b, 5, 6, 8a, 8b that illustrate the first aspect and figures 9a, 9b that illustrate the second aspect and figures 7a, 7b that illustrate the second aspect as it applies within the first aspect. An inventively encoded signal can be stored on a digital storage medium or a non-transient storage medium, or it can be transmitted on a transmission medium such as a wireless transmission medium or a wired transmission medium such as the Internet. Although some aspects have been described in the context of an apparatus, it is evident that these aspects also represent a description of the corresponding method, where a block or apparatus corresponds to a method step or a feature of a method step. Similarly, the aspects described in the context of a method step also represent a description of a corresponding block, element, or feature of a corresponding apparatus. Depending on certain implementation requirements, the embodiments of the invention can be implemented in hardware or software. Implementation can be carried out using a digital storage medium, for example, a floppy disk, DVD, CD, ROM, PROM, EPROM, EEPROM, or FLASH memory, having electronically readable control signals stored therein, which cooperate (or are capable of cooperating) with a programmable computer system so that the respective method is performed. Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperating with a programmable computer system, so that each of the methods described herein is carried out. In general, embodiments of the present invention can be implemented as a computer program product with program code, the program code being operative for performing one of the methods when the computer program product is executed on a computer. The program code can, for example, be stored on a machine-readable medium. Other modalities include the computer program to perform one of the methods described herein, stored on a machine-readable carrier or a non-transient storage medium. In other words, a modality of the inventive method is, therefore, a computer program that has program code to perform one of the methods described herein, when the computer program MA / t / ZUZÓ / U 1 OU4y is run on a computer. An additional form of inventive methods is, therefore, a data carrier (or a digital storage medium, or a computer-readable medium) comprising, recorded thereon, the computer program for carrying out one of the methods described herein. A further embodiment of the inventive method is, therefore, a data stream or a sequence of signals representing the computer program for carrying out one of the methods described herein. The data stream or the sequence of signals can, for example, be configured to be transferred over a data communication connection, such as the Internet. An additional modality comprises a processing means, for example, a computer or a programmable logic device, configured to, or adapted to, perform one of the methods described herein. One modality also includes a computer that has installed on it the computer program to perform one of the methods described herein. In some embodiments, a programmable logic device (for example, a field-programmable gate array) can be used to perform some or all of the functionalities of the methods described herein. In some embodiments, a field-programmable gate array can cooperate with a microprocessor to perform one of the methods described herein. In general, the methods are preferably implemented using any hardware device. The embodiments described above are merely illustrative of the principles of the present invention. It is understood that modifications and variations of the arrangements and details described herein will be obvious to others skilled in the art. Therefore, it is intended that the scope of the imminent patent claims be limited only by the specific details presented herein for the purpose of describing and explaining the embodiments. Subsequently, additional modalities / examples are summarized: 1. Audio decoder for decoding an encoded audio signal comprising multichannel audio data comprising data for two or more audio channels and jointly encoded scaling parameter information comprising: a scale parameter decoder (220) for decoding the information about the jointly encoded scale parameters to obtain a first set of scale parameters for a first channel of a decoded audio signal and a second set of scale parameters for a second channel of the decoded audio signal; and a signal processor (210, 212, 230) for applying the first set of scale parameters to a first channel representation derived from the multichannel audio data and for applying the second set of scale parameters to a second channel representation derived from the multichannel audio data to obtain the first and second channels of the decoded audio signal, MA / t / ZUZÓ / U Ί OU4y wherein the jointly coded scale parameters comprise information about a first group of jointly coded scale parameters and information about a second group of jointly coded scale parameters, and wherein the scale parameter decoder (220) is configured to combine a jointly coded scale parameter from the first group and a jointly coded scale parameter from the second group using a first combination rule to obtain a scale parameter from the first set of scale parameters, and using a second combination rule that is different from the first combination rule to obtain a scale parameter from the second set of scale parameters. 2. Audio decoder of example 1, wherein the first group of jointly encoded scale parameters comprises mid-scale parameters and the second group of jointly encoded scale parameters comprises side-scale parameters, and wherein the scale parameter decoder (220) is configured to use, in the first combination rule, an addition and to use, in the second combination rule, a subtraction. 3. Audio decoder of example 1 or 2, wherein the encoded audio signal is arranged in a sequence of frames, wherein a first frame comprises the multichannel audio data and the jointly encoded scale parameter information, and wherein a second frame comprises separately encoded scale parameter information, and wherein the scale parameter decoder (220) is configured to detect that the second frame comprises separately encoded scale parameter information and to calculate the first set of scale parameters and the second set of scale parameters. 4. Audio decoder of Example 3, wherein the first frame and the second frame each comprise additional state information indicating, in a first state, that the first frame comprises information about the jointly encoded scale parameters and, in a second state, that the second frame comprises information about the separately encoded scale parameters, and wherein the scale parameter decoder (220) is configured to read the additional state information from the second frame, to detect that the second frame comprises information about the separately encoded scale parameters based on the additional state information read, or to read the additional state information from the first frame, and to detect that the first frame comprises information about the jointly encoded scale parameters using the additional state information read. 5. Audio decoder of one of the previous examples, wherein the signal processor (210, 212, 230) is configured to decode the multichannel audio data to derive the first channel representation and the second channel representation, wherein the first channel representation and the second channel representation are spectral domain representations having spectral sampling values, and ML / t / ZUZÓ / U Ί OU4y 14. Audio decoder of one of Examples 11-13, wherein the first dequantization stage (2220) comprises using an index to a first codebook having a first number of entries or using an index representing a first precision, wherein the second dequantization stage (2260) comprises using an index to a second codebook having a second number of entries or using an index representing a second precision, and wherein the second number is less than or greater than the first number or the second precision is less than or greater than the first precision. 15. Audio decoder of one of the preceding examples, wherein the information about the second group of jointly encoded scale parameters indicates that the second group of jointly encoded scale parameters are all zero or at a certain value for a frame of the encoded audio signal, and wherein the scale parameter decoder (220) is configured to use, in the combination using the first rule or the second rule, a jointly encoded scale parameter that is zero or at a certain value or that is a synthesized jointly encoded scale parameter, or wherein, for the frame comprising the information of all zero or a certain value, the scale parameter decoder (220) is configured to determine the second set of scale parameters using only the first group of jointly encoded scale parameters without a combination operation. 16. Audio decoder of one of Examples 9 or 10, wherein the scale parameter decoder (220) is configured to use, as the first quantization mode, the first dequantization stage (2220) and the second dequantization stage (2260), and the combiner (2240) receives, as input, a result from the first dequantization stage (2220) and a result from the second dequantization stage (2260), and to use, as the second dequantization stage, the first dequantization stage (2220) of the first quantization mode. 17. An audio encoder for encoding a multichannel audio signal comprising two or more channels, comprising: a scale parameter calculator (140) calculate a first group of jointly encoded scale parameters and a second group of jointly encoded scale parameters from a first set of scale parameters for a first channel of the multichannel audio signal and from a second set of scale parameters for a second channel of the multichannel audio signal; a signal processor (120) for applying the first set of scaling parameters to the first channel of the multichannel audio signal and for applying the second set of scaling parameters to the second channel of the multichannel audio signal and for deriving multichannel audio data; and a coded signal former (1480,1500) for using the multichannel audio data and information about the first set of jointly coded scaling parameters and information about the second set of jointly coded scaling parameters to obtain a coded multichannel audio signal. 18. Audio encoder of example 17, where the signal processor (120) is configured, in the application, MA / t / ZUZÓ / U Ί OU4y to encode the first group of jointly encoded scale parameters and the second group of jointly encoded scale parameters to obtain information about the first group of jointly encoded scale parameters and information about the second group of jointly encoded scale parameters, to locally decode the information about the first and second groups of jointly encoded scale parameters to obtain a first set of locally decoded scale parameters and a second set of locally decoded scale parameters, and to modify the scale of the first channel using the first set of locally decoded scale parameters and to modify the scale of the second channel using the second set of locally decoded scale parameters, or wherein the signal processor (120) is configured, in the application, to quantize the first group of jointly encoded scale parameters and the second group of jointly encoded scale parameters to obtain a first quantized group of jointly encoded scale parameters and a second quantized group of jointly encoded scale parameters, to locally decode the first and second quantized groups of jointly encoded scale parameters to obtain a first set of locally decoded scale parameters and a second set of locally decoded scale parameters, and to scale the first channel using the first set of locally decoded scale parameters and to scale the second channel using the second set of locally decoded scale parameters. 19. Audio encoder of example 17 or 18, wherein the scale parameter calculator (140) is configured to combine a scale parameter from the first set of scale parameters and a scale parameter from the second set of scale parameters using a first combination rule to obtain a jointly encoded scale parameter from the first group of jointly encoded scale parameters, and using a second combination rule different from the first combination rule to obtain a jointly encoded scale parameter from the second group of jointly encoded scale parameters. 20. The audio encoder of Example 19, wherein the first group of jointly encoded scale parameters comprises mid-scale parameters and the second group of jointly encoded scale parameters comprises side-scale parameters, and wherein the scale parameter calculator (140) is configured to use, in the first combination rule, an addition and to use, in the second combination rule, a subtraction. 21. The audio encoder of one of the examples 17-20, where the scale parameter calculator is MA / t / ZUZÓ / U Ί OU4y is configured to process a sequence of frames from the multichannel audio signal, wherein the scale parameter calculator (140) is configured to calculate the first and second sets of jointly encoded scale parameters for a first frame of the frame sequence, and to analyze a second frame of the frame sequence to determine a separate encoding mode for the second frame, and wherein the encoded signal former (1480,1500) is configured to introduce additional status information into the encoded audio signal indicating a separate encoding mode for the second frame or a joint encoding mode for the first frame, and information about the first set and second set of separately encoded scale parameters for the second frame. 22. The audio encoder of one of Examples 17-21, wherein the scale parameter calculator (140) is configured to calculate the first set of scale parameters for the first channel and the second set of scale parameters for the second channel, to reduce the sampling of the first and second sets of scale parameters to obtain a first subsampled set and a second subsampled set; and to combine a scale parameter from the first subsampled set and the second subsampled set using different combination rules to obtain a jointly encoded scale parameter from the first group and a jointly encoded scale parameter from the second group, or wherein the scale parameter calculator (140) is configured to calculate the first set of scale parameters for the first channel and the second set of scale parameters for the second channel, combine one scale parameter from the first set and one scale parameter from the second set using different combination rules to obtain a jointly encoded scale parameter from the first group and a jointly encoded scale parameter from the second group, and to subsample from the first group of jointly encoded scale parameters to obtain a first subsampled group of jointly encoded scale parameters, and to reduce the sampling of the second group of jointly encoded scale parameters to obtain a second subsampled group of jointly encoded scale parameters,where the first subsampled group and the second subsampled group represent information about the first group of jointly coded scale parameters and information about the second group of jointly coded scale parameters. 23. Audio encoder from example 21 or 22, ML / t / ZUZÓ / U Ί OU4y where the scale parameter calculator (140) is set to calculate a similarity of the first channel and the second channel in the second frame and to determine the separate encoding mode in case a calculated similarity is in a first relationship with a threshold or to determine the joint encoding mode in case the calculated similarity is in a second different relationship with the threshold. 24. The audio encoder of Example 23, wherein the scale parameter calculator (140) is configured to calculate, for the second frame, a difference between the scale parameter of the first set and the scale parameter of the second set for each band, process each difference of the second frame so as to remove negative signs to obtain processed differences of the second frame, combine the processed differences to obtain a similarity measurement, compare the similarity measurement with the threshold, and decide in favor of separate encoding mode, when the similarity measurement is greater than the threshold, or decide in favor of joint encoding mode, when the similarity measurement is less than the threshold. 25. The audio encoder of one of Examples 17-24, wherein the signal processor (120) is configured to quantize the first group of jointly encoded scale parameters using a first-stage quantization function (141,143) to obtain one or more first quantization indices as a first-stage result and to obtain a first intermediate group of jointly encoded scale parameters, to compute (142) a first residual group of jointly encoded scale parameters from the first group of jointly encoded scale parameters and the first intermediate group of jointly encoded scale parameters, and to quantize the first residual group of jointly encoded scale parameters using a second-stage quantization function (145) to obtain one or more quantization indices as a second-stage result. 26. The audio encoder of one of Examples 17-25, wherein the signal processor (120) is configured to quantize the second group of jointly encoded scale parameters using a single-stage quantization function to obtain one or more quantization indices as the single-stage result, or wherein the signal processor (120) is configured to quantize the first group of jointly encoded scale parameters using at least one first-stage quantization function and one second-stage quantization function, and wherein the signal processor (120) is configured to quantize the second group of jointly encoded scale parameters using a single-stage quantization function, wherein the single-stage quantization function is selected from the first-stage quantization function and the second-stage quantization function. 27. The audio encoder of one of the examples 21-26, wherein the scale parameter calculator (140) ML / t / ZUZÓ / U Ί OU4U is configured to quantize the first set of scale parameters using a first-stage quantization function (141,143) to obtain one or more first quantization indices as a first-stage result and to obtain a first intermediate set of scale parameters, to compute (142) a first residual set of scale parameters from the first set of scale parameters and the first intermediate set of scale parameters, and to quantize the first residual set of scale parameters using a second-stage quantization function (145) to obtain one or more quantization indices as a second-stage result, or wherein the scale parameter calculator (140) is configured to quantize the second set of scale parameters using a first-stage quantization function (141,143) to obtain one or more first quantization indices as a first-stage result and to obtain a second intermediate set of scale parameters, to calculate (142) a second residual set of scale parameters from the second set of scale parameters and the second intermediate set of scale parameters, and to quantize the second residual set of scale parameters using a second-stage quantization function (145) to obtain one or more quantization indices as a second-stage result. 28. Audio encoder of Example 25 or 27, wherein the second-stage quantization function (145) uses an amplification or weighting value less than 1 to increment the first residual group of jointly encoded scaling parameters or the first or second residual set of scaling parameters before performing vector quantization, wherein the vector quantization is performed using incremented residual values, and / or wherein, by way of example, the weighting or amplification value is used to divide a scaling parameter by the weighting or amplification value, wherein the weighting value is preferably between 0.1 and 0.9, or more preferably between 0.2 and 0.6, or even more preferably between 0.25 and 0.4, and / or where the same amplification value is used for all scaling parameters of the first residual group of jointly coded scaling parameters or the first or second residual set of scaling parameters. 29. The audio encoder of one of Examples 25-28, wherein the first-stage quantization function (141,143) comprises at least one codebook having a first number of entries corresponding to a first size of one or more quantization indices, wherein the second-stage quantization function (145) or the single-stage quantization function comprises at least one codebook having a second number of entries corresponding to a second size of one or more quantization indices, and ML / t / ZUZÓ / U Ί OU4y where the first number is greater or less than the second number or the first size is greater or less than the second size, or where the first stage quantization function (141,143) is a fixed-rate quantization function and where the second stage quantization function (145) is a variable-rate quantization function. 30. The audio encoder of one of Examples 15-29, wherein the scale parameter calculator (140) is configured to receive a first MDCT representation for the first channel and a second MDCT representation for the second channel, to receive a first MDST representation for the first channel and a second MDST representation for the second channel, to calculate a first power spectrum for the first channel from the first MDCT representation and the first MDST representation and a second power spectrum for the second channel from the second MDCT representation and the second MDST representation, and to calculate the first set of scale parameters for the first channel from the first power spectrum and to calculate the second set of scale parameters for the second channel from the second power spectrum. 31. The audio encoder of Example 30, wherein the signal processor (120) is configured to scale the first MDCT representation using information derived from the first set of scaling parameters, and to scale the second MDCT representation using information derived from the second set of scaling parameters. 32. The audio encoder of one of Examples 17-31, wherein the signal processor (120) is configured to further process a scaled-up first-channel representation and a scaled-up second-channel representation using joint multichannel processing to derive a processed multichannel representation of the multichannel audio signal, to optionally further process using spectral band replication processing or intelligent gap-filling processing or bandwidth enhancement processing, and to quantize and encode a channel representation of the multichannel audio signal to obtain the multichannel audio data. 33. The audio encoder of one of Examples 17-34, which is configured to determine, for a frame of the multichannel audio signal, information about the second group of jointly coded scale parameters as all zero or all of a certain value information indicating the same value or a zero value for all of the jointly coded scale parameters of the frame, and wherein the coded signal former (1480, 1500) is configured to use the all zero or all of a certain value information to obtain the coded multichannel audio signal. 34. The audio encoder of one of the examples 17-34, wherein the scale parameter calculator (140) is configured ML / t / ZUZÓ / U Ί OU4y to calculate the first group of jointly coded scale parameters and the second group of jointly coded scale parameters for a first frame, to calculate the first group of jointly coded scale parameters for a second frame, wherein, in the second frame, the jointly coded scale parameters are neither calculated nor coded, and wherein the coded signal former (1480, 1500) is configured to use a flag as information about the second group of jointly coded scale parameters indicating that, in the second frame, any jointly coded scale parameters from the second group are not included in the coded multichannel audio signal. 35. Method for decoding an encoded audio signal comprising multichannel audio data comprising data for two or more audio channels and jointly encoded scaling parameter information comprising: decoding information about the jointly encoded scale parameters to obtain a first set of scale parameters for a first channel of a decoded audio signal and a second set of scale parameters for a second channel of the decoded audio signal; and applying the first set of scale parameters to a first channel representation derived from the multichannel audio data and applying the second set of scale parameters to a second channel representation derived from the multichannel audio data to obtain the first and second channels of the decoded audio signal, wherein the jointly encoded scale parameters comprise information about a first group of jointly encoded scale parameters and information about a second group of jointly encoded scale parameters,and wherein the decoding comprises combining a jointly coded scale parameter from the first group and a jointly coded scale parameter from the second group using a first combination rule to obtain a scale parameter from the first set of scale parameters, and using a second combination rule that is different from the first combination rule to obtain a scale parameter from the second set of scale parameters. 36. A method for encoding a multichannel audio signal comprising two or more channels, comprising: calculating a first set of jointly encoded scale parameters and a second set of jointly encoded scale parameters from a first set of scale parameters for a first channel of the multichannel audio signal and from a second set of scale parameters for a second channel of the multichannel audio signal; Apply the first set of scaling parameters to the first channel of the multichannel audio signal and apply the second set of scaling parameters to the second channel of the multichannel audio signal to derive audio data ML / t / ZUZÓ / U 1 OU4y multichannel; and use the multichannel audio data and information about the first group of jointly encoded scale parameters and information about the second group of jointly encoded scale parameters to obtain a coded multichannel audio signal. 37. Computer program to perform, when executed on a computer or processor, the method of example 31 or example 32. [1] ISO / IEC 11172-3, Information technology - Coding of moving pictures and associated audio for digital storage media at up to about 1.5 Mbit / s - Part 3: Audio, 1993. [2] ISO / IEC 13818-7, Information technology - Generic coding ofmoving pictures and associated audio Information - Parte 7: Advanced Audio Coding (AAC), 2003. [3] ISO / IEC 23003-3; Information technology - MPEG audio technologies - Parte 3: Unified speech and audio coding. [4] 3GPP TS 26.445, Codee forEnhanced Volee Services (EVS)', Descripción algorítmica detallada. [5] G. Markovic, G. Fuchs, N. Rettelbach, C. Helmrich y B. Schubert, ‘LINEAR PREDICTION BASED CODING SCHEME USING SPECTRAL DOMAIN NOISE SHAPNG“. Patente de los Estados Unidos 9,595,262 B2,14 de marzo de 2017. [6] E. Ravelli, M. Schnell, C. Benndorf, M. Lutzky y M. Dietz, “Apparatus and method for encoding and decoding an audio signa / using downsampling or interpolation ofseale parameters. Publicación internacional WO 2019091904 A1, 511 2018. [7] A. Biswas, Advances in Perceptual Stereo Audio Coding Using LinearPrediction Techniques, Eindhoven: Technical University of Eindhoven, 2017. [8] G. Markovic, E. Ravelli, M. Schnell, S. Dóhla, W. Jaegars, M. Dietz, C. Heimrich, E. Fotopoulou, M. Multrus, S. Bayer, G. Fuchs y J. Herre, “APPARATUS AND METHOD FOR MDCT M / S STEREO WITH GLOBAL ILD WITH IMPROVED MID / SIDE DECISION1. Publicación internacional WO2017EP5117.

Claims

1. An audio quantizer for quantizing a plurality of audio information elements, comprising: a first-stage vector quantizer (141, 143) for quantizing the plurality of audio information elements to determine a first-stage vector quantization result and a plurality of intermediate quantized elements corresponding to the first-stage vector quantization result; a residual element determinator (142) for calculating a plurality of residual elements from the plurality of intermediate quantized elements and the plurality of audio information elements;and a second-stage vector quantizer (145) for quantizing the plurality of residual elements to obtain a second-stage vector quantization result, wherein the first-stage vector quantization result and the second-stage vector quantization result are a quantized representation of the plurality of audio information elements.; 2. Audio quantizer according to claim 1, wherein the residual element determinator (142) is configured to calculate, for each residual element, a difference between a corresponding audio information element and a corresponding intermediate quantized element.

3. Audio quantizer according to claim 1 or 2, wherein the residual element determinator (142) is configured to amplify or weight, for each residual element, a difference between a corresponding audio information element and a corresponding intermediate quantized element such that the plurality of residual elements is greater than the corresponding differences, or to amplify or weight the plurality of audio information elements and / or the plurality of intermediate quantized elements before calculating a difference between amplified elements to obtain the residual elements.

4. Audio quantizer according to any of the preceding claims, wherein the residual element determinator (142) is configured to divide the corresponding differences between the plurality of intermediate quantized elements and the audio information elements by a predetermined factor that is less than 1 or to multiply the corresponding differences between the plurality of intermediate quantized elements and the audio information elements by a predetermined factor that is greater than 1.

5. Audio quantizer according to any of the preceding claims, wherein the first-stage vector quantizer (141,143) is configured to perform quantization with a first quantization precision, wherein the second-stage vector quantizer (145) is configured to perform quantization with a second quantization precision, and wherein the second quantization precision is less or greater than the first quantization precision, or wherein the first-stage vector quantizer (141,143) is configured to perform fixed-rate quantization and wherein the second-stage vector quantizer (145) is configured to perform variable-rate quantization.

6. Audio quantizer according to any of the preceding claims, wherein the first-stage vector MA / t / ZUZÓ / U Ί OU4y quantizer (141,143) is configured to use a first-stage codebook having a first number of entries, wherein the second-stage vector quantizer (145) is configured to use a second-stage codebook having a second number of entries, and wherein the second number of entries is less than or greater than the first number of entries.

7. An audio quantizer according to any one of the preceding claims, wherein the audio information elements are scaling parameters for a frame of an audio signal usable for scaling time-domain audio samples of an audio signal in a time domain or usable for scaling spectral-domain audio samples of an audio signal in a spectral domain, wherein each scaling parameter is usable for scaling at least two time-domain or spectral-domain audio samples, wherein the frame comprises a first number of scaling parameters, wherein the first-stage vector quantizer (141, 143) is configured to perform a division of the first number of scaling parameters into two or more sets of scaling parameters, and wherein the first-stage vector quantizer (141,143) is configured to determine a quantification index for each set of scale parameters to obtain a plurality of quantification indices representing the first quantification result.

8. Audio quantizer according to claim 7, wherein the first-stage vector quantizer (141,143) is configured to combine a first quantization index for the first set and a second quantization index for the second set to obtain a single index as the first quantization result.

9. Audio quantizer according to claim 8, wherein the first-stage vector quantizer (141,143) is configured to multiply one of the first and second indices by a number corresponding to the number of bits of the first and second indices and to add a multiplied index and an unmultiplied index to obtain the individual index.

10. Audio quantizer according to any of the preceding claims, wherein the second-stage vector quantizer (145) is an algebraic vector quantizer, wherein each index comprises a base codebook index and a Voronoi extension index.

11. Audio quantizer according to any of the preceding claims, wherein the first-stage vector quantizer (141,143) is configured to perform a first division of the plurality of audio information elements, wherein the second-stage vector quantizer (145) is configured to perform a second division of the plurality of residual elements, wherein the first division results in a first quantity of subsets of audio information elements and the second division results in a second quantity of subsets of residual elements, wherein the first quantity of subsets is equal to the second quantity of subsets.

12. Audio quantizer according to any of the preceding claims, wherein the first vector quantizer is configured to produce, from a first codebook search, a first index having a first number of bits, wherein the second vector quantizer is configured to produce, for a second codebook search, a second index having a second number of bits, the second number of bits being less than or greater than the first number of bits.

13. Audio quantizer according to claim 12, wherein the first number of bits is a number of bits between 4 and 7, and wherein the second number of bits is a number of bits between 3 and 6.

14. An audio quantizer according to any one of the preceding claims, wherein the audio information elements comprise, for a first frame of a multichannel audio signal, a first plurality of scale parameters for a first channel of the multichannel audio signal, and a second plurality of scale parameters for a second channel of the multichannel audio signal, wherein the audio quantizer is configured to apply the first and second stage vector quantizers to the first and second pluralities of the first frame, wherein the audio information elements comprise, for a second frame of the multichannel audio signal, a third plurality of mid-scale parameters and a fourth plurality of side-scale parameters, and wherein the audio quantizer is configured to apply the first and second stage vector quantizers to the third plurality of mid-scale parameters.and to apply the second stage vector quantizer to the fourth plurality of sidescale parameters and not to apply the first stage vector quantizer (141,143) to the fourth plurality of sidescale parameters., 15. Audio quantizer according to claim 14, wherein the residual element determinator (142) is configured to amplify or weight, for the second frame, the fourth plurality of sidescale parameters, and wherein the second-stage vector quantizer (145) is configured to process amplified or weighted sidescale parameters for the second frame of the multichannel audio signal.

16. An audio dequantizer for dequantizing a quantized plurality of audio information elements, comprising: a first-stage vector dequantizer (2220) for dequantizing a first-stage vector quantization result included in the quantized plurality of audio information elements to obtain a plurality of intermediate quantized audio information elements; a second-stage vector dequantizer (2260) for dequantizing a second-stage vector quantization result included in the quantized plurality of audio information elements to obtain a plurality of residual elements; and a combiner (2240) for combining the plurality of intermediate quantized information elements and the plurality of residual elements to obtain a dequantized plurality of audio information elements.

17. Audio dequantizer according to claim 16, wherein the combiner (2240) is configured to calculate, for each dequantized information element, a sum between a corresponding intermediate quantized audio information element and a corresponding residual element.

18. An audio dequantizer according to any one of claim 16 or 17, wherein the combiner (2240) is configured to attenuate or weight the plurality of residual elements, such that the attenuated residual elements are smaller than the corresponding residual elements before attenuation, and wherein the combiner (2240) is configured to add the attenuated residual elements to the corresponding intermediate quantized audio information elements, or wherein the combiner (2240) is configured to use an attenuation or weighting value less than 1 to attenuate the plurality of jointly encoded residual elements or scaling parameters before combining, wherein combining is done using attenuated residual values, and / or wherein, by way of example,The weighting or attenuation value is used to multiply a scaling parameter by the weighting or amplification value, wherein the weighting value is preferably between 0.1 and 0.9, or more preferably between 0.2 and 0.6, or even more preferably between 0.25 and 0.4, and / or wherein the same attenuation or weighting value is used for all scaling parameters of the plurality of residual elements or any jointly coded scaling parameter.

19. Audio dequantizer according to claim 18, wherein the combiner (2240) is configured to multiply a corresponding residual element by a weighting factor that is less than one or to divide a corresponding residual element by a weighting factor that is greater than one.

20. Audio dequantizer according to any one of claims 16 to 19, wherein the first-stage dequantizer is configured to perform dequantization with a first precision, wherein the second-stage dequantizer is configured to perform dequantization with a second precision, wherein the second precision is less or greater than the first precision.

21. An audio dequantizer according to any one of claims 16 to 20, wherein the first-stage dequantizer is configured to use a first-stage codebook having a first number of entries, wherein the second-stage dequantizer is configured to use a second-stage codebook having a second number of entries, and wherein the second number of entries is less than or greater than the first number of entries, or wherein the first-stage dequantizer is configured to receive, for a first codebook retrieval, a first index having a first number of bits, wherein the second-stage vector dequantizer (2260) is configured to receive, for a second codebook retrieval, a second index having a second number of bits, the second number of bits being less than or greater than the first number of bits, or wherein, by way of example,the first number of bits is a number of bits between 4 and 7, and where, as an example, the second number of bits is a number of bits between 3 and 6.

22. An audio dequantizer according to any one of claims 16 to 21, wherein the dequantized plurality of audio information elements are scaling parameters for a frame of an audio signal usable for scaling time-domain audio samples of an audio signal in a time domain or usable for scaling spectral-domain audio samples of an audio signal in a spectral domain, wherein each scaling parameter is usable for scaling at least two time-domain or spectral-domain audio samples, wherein the frame comprises a first number of scaling parameters, wherein the first-stage dequantizer is configured to determine, from two or more result indices for the first-stage vector quantization result, a first set and a second set of scaling parameters,and wherein the first-stage vector dequantizer (2220) or the combiner (2240) is configured to combine the first set of scale parameters and the second set of scale parameters into one vector to obtain the first number of intermediate quantized scale parameters.

23. Audio dequantizer according to claim 22, wherein the first-stage vector dequantizer (2220) is configured to recover, as a result of first-stage dequantization, an individual combined index and to process the individual combined index to obtain the two or more result indices.

24. Audio dequantizer according to claim 23, wherein the first-stage dequantizer is configured to retrieve the first result index by determining a remainder of a division and to retrieve the second result index by determining an integer result of the division.

25. Audio dequantizer according to any one of claims 16 to 24, wherein the second-stage vector dequantizer (2260) is an algebraic vector dequantizer, wherein each index comprises a base codebook index and a Voronoi extension index.

26. Audio dequantizer according to any one of claims 16 to 25, wherein the first-stage vector dequantizer (2220) or the combiner (2240) is configured to combine a first set of scale parameters and a second set of scale parameters from a quantization division in a frame of an audio signal, wherein the second-stage vector dequantizer (2260) is configured to combine a first set of residual parameters and a second set of residual parameters from a residual parameter division, and wherein a number of divisions addressed by the first vector dequantizer and another number of divisions addressed by the second-stage vector dequantizer (2260) are equal.

27. Audio dequantizer according to any one of claims 16 to 26, wherein the first-stage vector dequantizer (2220) is configured to use a first index having a first number of bits to generate the plurality of intermediate quantized audio information elements, and wherein the second-stage vector dequantizer (2260) is configured to use, as an index, a second index having a second number of bits to obtain the plurality of residual elements, wherein the second number of bits is less than or greater than the first number of bits.

28. Audio dequantizer according to claim 27, wherein the first number of bits is between four and seven, and the second number of bits is between three and six.

29. An audio dequantizer according to any one of claims 16 to 28, wherein the quantized plurality of audio information elements comprises, for a first frame of a multichannel audio signal, a first plurality of scaling parameters for a first channel of the multichannel audio signal and a second plurality of scaling parameters for a second channel of the multichannel audio signal, wherein the audio dequantizer is configured to apply the first-stage vector dequantizer (2220) and the second-stage vector dequantizer (2260) to the first and second pluralities of the first frame, wherein the quantized plurality of audio information elements comprises, for a second frame of the multichannel audio signal, a third plurality of mid-scale parameters and a fourth plurality of side-scale parameters,and wherein the audio dequantizer is configured to apply the first-stage vector dequantizer (2220) and the second-stage vector dequantizer (2260) to the third plurality of mid-scale parameters and to apply the second-stage vector dequantizer (2260) to the fourth plurality of side-scale parameters and not to apply the first-stage vector dequantizer (2220) to the fourth plurality of side-scale parameters.

30. Audio dequantizer according to claim 29, wherein the combiner (2240) is configured to attenuate, for the second frame, the fourth plurality of sidescale parameters before further using or further processing the fourth plurality of sidescale parameters.

31. A method for quantizing a plurality of audio information elements, comprising: a first-stage vector that quantizes the plurality of audio information elements to determine a first-stage vector quantization result and a plurality of intermediate quantized elements MA / t / ZUZÓ / U 1 OU4y corresponding to the first-stage vector quantization result; calculating a plurality of residual elements from the plurality of intermediate quantized elements and the plurality of audio information elements; and a second-stage vector that quantizes the plurality of residual elements to obtain a second-stage vector quantization result, wherein the first-stage vector quantization result and the second-stage vector quantization result are a quantized representation of the plurality of audio information elements.

32. A method for dequantizing a quantized plurality of audio information elements, comprising: dequantizing a first-stage vector quantization result included in the quantized plurality of audio information elements to obtain a plurality of intermediate quantized audio information elements; dequantizing a second-stage vector quantization result included in the quantized plurality of audio information elements to obtain a plurality of residual elements; and combining the plurality of intermediate quantized information elements and the plurality of residual elements to obtain a dequantized plurality of audio information elements.

33. A computer-readable means for quantifying a plurality of audio information elements comprising the method of claim 31.

34. A computer-readable means for dequantizing a quantized plurality of audio information elements comprising the method of claim 32.

Images