Audio signal encoder, audio signal decoder, method for encoding or decoding an audio signal using an aliasing-cancellation

Abstract

An audio signal decoder (200) for providing a decoded representation (212) of an audio content on the basis of an encoded representation (310) of the audio content comprises a transform domain path (230, 240, 242, 250, 260) configured to obtain a time-domain representation (212) of a portion of the audio content encoded in a transform-domain mode on the basis of a first set (220) of spectral coefficients, a representation (224) of an aliasing-cancellation stimulus signal and a plurality of linear-prediction-domain parameters (222). The transform domain path comprises a spectrum processor (230) configured to apply a spectrum shaping to the first set of spectral coefficients in dependence on at least a subset of the linear-prediction-domain parameters, to obtain a spectrally-shaped version (232) of the first set of spectral coefficients. The transform domain path comprises a first frequency-domain-to-time-domain converter (240) configured to obtain a time-domain representation of the audio content on the basis of the spectrally-shaped version of the first set of spectral coefficients. The transform domain path comprises an aliasing-cancellation stimulus filter configured to filter (250) the aliasing-cancellation stimulus signal (324) in dependence on at least a subset of the linear-prediction-domain parameters (222), to derive an aliasing-cancellation synthesis signal (252) from the aliasing-cancellation stimulus signal. The transform domain path also comprises a combiner (260) configured to combine the time-domain representation (242) of the audio content with the aliasing-cancellation synthesis signal (252), or a post-processed version thereof, to obtain an aliasing reduced time-domain signal.

Description

(19) *EP004651130A1* (11) EP 4 651 130 A1 (12) EUROPEAN PATENT APPLICATION (43) Date of publication: 19.11.2025 Bulletin 2025 / 47 (21) Application number: 25171588.4 (22) Date of filing: 19.10.2010 (51) International Patent Classification (IPC): G10L 19 / 18 (2013.01) G10L 19 / 02 (2013.01) G10L 19 / 04 (2013.01) G10L 19 / 03 (2013.01) G10L 19 / 12 (2013.01) G10L 19 / 20 (2013.01) (52) Cooperative Patent Classification (CPC): G10L 19 / 03; G10L 19 / 0212; G10L 19 / 04; G10L 19 / 12; G10L 19 / 18; G10L 19 / 20 (84) Designated Contracting States: AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR (30) Priority: 20.10.2009 US 25346809 P (62) Document number(s) of the earlier application(s) in accordance with Art. 76 EPC: 24160714.2 / 4 358 082 10771705.0 / 2 491 556 (71) Applicants: • Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V. 80686 München (DE) • VoiceAge Corporation Montreal, QC H3R 2H6 (CA) • Koninklijke Philips N.V. 5656 AG Eindhoven (NL) • Dolby International AB Dublin, D02 VK60 (IE) (72) Inventors: • Bessette, Bruno Sherbrooke, Quebec, J1N 4G5 (CA) • Villemoes, Lars 175 56 Järfälla (SE) • Salami, Redwan Canada H4N 4A2, Saint-Laurent, Quebec (CA) • den Brinker, Albertus C. 5644 Eindhoven (NL) • Neuendorf, Max 91058 Erlangen (DE) • Geiger, Ralf 91058 Erlangen (DE) • Gournay, Philippe Sherbrooke, Quebec, J1L 0A2 (CA) • Lefebvre, Roch Magog, J1X 0L6 (CA) • Grill, Bernhard 91058 Erlangen (DE) • Lecomte, Jérémie 91058 Erlangen (DE) • Bayer, Stefan 91058 Erlangen (DE) • Rettelbach, Nikolaus 91058 Erlangen (DE) (74) Representative: Burger, Markus Schoppe, Zimmermann, Stöckeler Zinkler, Schenk & Partner mbB Patentanwälte Radlkoferstraße 2 81373 München (DE) Remarks: This application was filed on 22.04.2025 as a divisional application to the application mentioned under INID code 62. (54) AUDIO SIGNAL ENCODER, AUDIO SIGNAL DECODER, METHOD FOR ENCODING OR DECODING AN AUDIO SIGNAL USING AN ALIASING‑CANCELLATION (57) An audio signal decoder (200) for providing a decoded representation (212) of an audio content on the basis of an encoded representation (310) of the audio content comprises a transform domain path (230, 240, 242, 250, 260) configured to obtain a time-domain re- presentation (212) of a portion of the audio content en- coded in a transform-domain mode on the basis of a first set (220) of spectral coefficients, a representation (224) of an aliasing-cancellation stimulus signal and a plurality of linear-prediction-domain parameters (222). The trans- formdomain path comprises a spectrumprocessor (230) configured to apply a spectrum shaping to the first set of spectral coefficients in dependence on at least a subset of the linear-prediction-domain parameters, to obtain a spectrally-shaped version (232) of the first set of spectral coefficients. The transformdomain path comprises a first frequency-domain-to-time-domain converter (240) con- figured to obtain a time-domain representation of the audio content on the basis of the spectrally-shaped ver- sion of the first set of spectral coefficients. The transform domain path comprises an aliasing-cancellation stimulus filter configured to filter (250) the aliasing-cancellation stimulus signal (324) in dependence on at least a subset of the linear-prediction-domain parameters (222), to de-EP 4 65 1 13 0 A 1 Processed by Luminess, 75001 PARIS (FR) (Cont. next page) rive an aliasing-cancellation synthesis signal (252) from the aliasing-cancellation stimulus signal. The transform domain path also comprises a combiner (260) configured to combine the time-domain representation (242) of the audio content with the aliasing-cancellation synthesis sig- nal (252), or apost-processedversion thereof, to obtainan aliasing reduced time-domain signal. 2 EP 4 651 130 A1 Description Technical Field

[0001] Embodiments according to the invention create an audio signal decoder for providing a decoded representation of an audio content on the basis of an encoded representation of the audio content.

[0002] Embodiments according to the invention create anaudio signal encoder for providinganencoded representation of an audio content comprising a first set of spectral coefficients, a representation of an aliasing-cancellation stimulus signal and a plurality of linear-prediction-domain parameters on the basis of an input representation of the audio content.

[0003] Embodiments according to the invention create a method for providing a decoded representation of an audio content on the basis of an encoded representation of the audio content.

[0004] Embodiments according to the invention create a method for providing an encoded representation of an audio content on the basis of an input representation of the audio content.

[0005] Embodiments according to the invention create a computer program for performing one of said methods.

[0006] Embodiments according to the invention create a concept for a unification of unified-speech-and-audio-coding (also designated briefly as USAC) windowing and frame transitions. Background of the Invention

[0007] In the following some background of the invention will be explained in order to facilitate the understanding of the invention and advantages thereof.

[0008] During the past decade, big effort has been input on creating the possibility to digitally store and distribute audio content. One important achievement on this way is the definition of the International Standard ISO / IEC 14496‑3. Part 3 of this Standard is related to a coding and decoding of audio contents, and sub-part 4 of part 3 is related to general audio coding. ISO / IEC 14496, part 3, sub-part 4 defines a concept for encoding and decoding of general audio content. In addition, further improvements have been proposed in order to improve the quality and / or reduce the required bitrate. Moreover, it has been found that the performance of frequency-domain based audio coders is not optimal for audio contents comprising speech. Recently, a unified speech-and-audio codec has been proposed which efficiently combines techniques from both words, namely speech coding and audio coding. For some details, reference is made to the publication "A Novel Scheme for Low Bitrate Unified Speech and Audio Coding - MPEG-RM0 " of M. Neuendorf et al. (presented at the 126th Convention of the Audio Engineering Society, May 7‑10, 2009, Munich, Germany).

[0009] In such an audio coder, some audio frames are encoded in the frequency-domain and some audio frames are encoded in the linear-prediction-domain.

[0010] However, it has been found that it is difficult to transition between frames encoded in different domains without sacrificing a significant amount of bitrate.

[0011] In view of this situation, there is a desire to create a concept for encoding and decoding an audio content comprising both speech and general audio, which allows for efficient realization of transitions between portions encoded using different modes. Summary of the Invention

[0012] Embodiments according to the invention create an audio signal decoder for providing a decoded representation of an audio content on the basis of an encoded representation of an audio content. The audio signal decoder comprises a transform domain path (for example, a transform-coded excitation linear-prediction-domain-path) configured to obtain a time domain representation of the audio content encoded in a transform domainmode on the basis of a first set of spectral coefficients, a representation of an aliasing-cancellation stimulus signal, and a plurality of linear-prediction-domain parameters (for example, linear-prediction-coding filter coefficients). The transform domain path comprises a spectrum processor configured toapply a spectral shaping to the (first) set of spectral coefficients in dependenceonat least a subset of linear-prediction-domain parameters to obtain a spectrally-shaped version of the first set of spectral coefficients. The transform domain path also comprises a (first) frequency-domain-to-time-domain-converter configured to obtain a time- domain representation of the audio content on the basis of the spectrally-shaped version of the first set of spectral coefficients. The transform domain path also comprises an aliasing-cancellation-stimulus filter configured to filter the aliasing-cancellation stimulus signal in dependence on at least a subset of the linear-prediction-domain parameters, to derive an aliasing-cancellation synthesis signal from the aliasing-cancellation stimulus signal. The transformdomain path also comprises a combiner configured to combine the time-domain representation of the audio content with the aliasing- cancellation synthesis signal, or a post-processed version thereof, to obtain an aliasing-reduced time-domain signal.

[0013] Thisembodimentof the invention isbasedon thefinding thatanaudiodecoderwhichperformsaspectral shaping of the spectral coefficients of the first set of spectral coefficients in the frequency-domain, andwhich computes an aliasing- 3 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 cancellation synthesis signal by time-domain filtering an aliasing-cancellation stimulus signal, wherein both the spectral shapingof the spectral coefficientsand the time-domain filteringof thealiasing-cancellation-stimulus signal areperformed in dependence on linear-prediction-domain parameters, is well-suited for transitions from and to portions (for example, frames) of the audio signal encoded with different noise shaping and also for transitions from or to frames which are encoded in different domains. Accordingly, transitions (for example, between overlapping or non-overlapping frames) of the audio signal, which are encoded in different modes of amulti-mode audio signal coding, can be rendered by the audio signal decoder with good auditory quality and at a moderate level of overhead.

[0014] Forexample, performing thespectral shapingof thefirst set of coefficients in the frequency-domainallowshaving the transitions between portions (for example, frames) of the audio content encoded using different noise shaping concepts in the transform domain, wherein an aliasing-cancellation can be obtained with good efficiency between the different portions of the audio content encoded using different noise shaping methods (for example, scale-factor-based noise shaping and linear-prediction-domain-parameter-based noise-shaping). Moreover, the above-described concepts also allows for an efficient reduction of aliasing artifacts between portions (for example, frames) of the audio content encoded in different domains (for example, one in the transform domain and one in the algebraic-code-excited-linear- prediction-domain). The usage of a time-domain filtering of the aliasing-cancellation stimulus signal allows for an aliasing- cancellation at the transition from and to a portion of the audio content encoded in the algebraic-code-excited-linear- predictionmode even if the noise shaping of the current portion of the audio content (whichmay be encoded, for example, in a transform-coded-excitation linear prediction-domain mode) is performed in the frequency-domain, rather than by a time-domain filtering.

[0015] To summarize the above, embodiments according to the present invention allow for a good tradeoff between a required side information and a perceptual quality of transitions between portions of the audio content encoded in three different modes (for example, frequency-domain mode, transform-coded-excitation linear-prediction-domain mode, and algebraic-code-excited-linear-prediction mode).

[0016] In a preferred embodiment, the audio signal decoder is a multi-mode audio signal decoder configured to switch between a plurality of coding modes. In this case, the transform domain branch is configured to selectively obtain the aliasing cancellation synthesis signal for a portion of the audio content following a previous portion of the audio content whichdoesnot allow for analiasing-cancelling overlap-and-addoperationor followedbyasubsequent portionof theaudio contentwhichdoesnot allow for analiasing-cancellingoverlap-and-addoperation. It hasbeen found that theapplicationof a noise shaping, which is performed by the spectral shaping of the spectral coefficients of the first set of spectral coefficients, allows for a transition between portions of the audio content encoded in the transform domain and using different noise shaping concepts (for example, a scale-factor-based noise shaping concept and a linear-prediction- domain-parameter-based noise shaping concept) without using the aliasing-cancellation signals, because the usage of the first frequency-domain-to-time-domain converter after the spectral shapingallows for anefficient aliasing-cancellation between subsequent frames encoded in the transformdomain, even if different noise-shaping approaches are used in the subsequent audio frames. Thus, bitrate efficiency can be obtained by selectively obtaining the aliasing-cancellation synthesis signal only for transitions from or to a portion of the audio content encoded in a non-transform domain (for example, in an algebraic code-excited-linear-prediction-mode).

[0017] In a preferred embodiment, the audio signal decoder is configured to switch between a transform-coded- excitation-linear-prediction-domain mode, which uses a transform-coded-excitation information and a linear-prediction- domain parameter information, and a frequency-domain mode, which uses a spectral coefficient information and a scale factor information. In this case, the transform-domain-path is configured toobtain the first set of spectral coefficients on the basis of the transform-coded-excitation information and to obtain the linear-prediction-domain parameters on the basis of the linear-prediction-domain-parameter information. The audio signal decoder comprises a frequency domain path configured to obtain a time-domain representation of the audio content encoded in the frequency-domain mode on the basis of a frequency-domain mode set of spectral coefficients described by the spectral coefficient information and in dependence on a set of scale factors described by the scale factor information. The frequency-domain path comprises a spectrumprocessor configured toapply a spectral shaping to the frequency-domainmodeset of spectral coefficients, or to a pre-processed version thereof, in dependence on the scale factors to obtain a spectrally-shaped frequency-domain mode set of spectral coefficients. The frequency-domain path also comprises a frequency-domain-to-time-domain converter configured to obtain a time-domain representation of the audio content on the basis of the spectrally-shaped frequency-domain-mode set of spectral coefficients. The audio signal decoder is configured such that time-domain representations of two subsequent portions of the audio content, one of which two subsequent portions of the audio content is encoded in the transform-coded-excitation linear-prediction-domain mode, and one of which two subsequent portions of the audio content is encoded in the frequency-domain mode, comprise a temporal overlap to cancel a time- domain aliasing caused by the frequency-domain-to-time-domain conversion.

[0018] As already discussed, the concept according to the embodiments of the invention is well-suited for transitions between portions of the audio content encoded in the transform-coded-excitation-linear-predication-domainmode and in the frequency-domainmode. A very good quality aliasing-cancellation is obtained due to the fact that the spectral shaping 4 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 is performed in the frequency-domain in the transform-coded-excitation-linear-prediction-domain mode.

[0019] In a preferred embodiment, the audio signal decoder is configured to switch between a transform-coded- excitation-linear-prediction-domain-mode which uses a transform-coded-excitation information and a linear-prediction- domain parameter information, and an algebraic-code-excited-linear-prediction mode, which uses an algebraic-code- excitation-information and a linear-prediction-domain-parameter information. In this case, the transform-domain path is configured to obtain the first set of spectral coefficients on the basis of the transform-coded-excitation information and to obtain the linear-prediction-domain parameters on the basis of the linear-prediction-domain-parameter information. The audio signal decoder comprises an algebraic-code-excited-linear-prediction path configured to obtain a time-domain representation of the audio content encoded in the algebraic-code-excited-linear-prediction (also designated briefly with ACELP in the following) mode, on the basis of the algebraic-code-excitation information and the linear-prediction-domain parameter information. In this case, the ACELP path comprises an ACELP excitation processor configured to provide a time-domain excitation signal on the basis of the algebraic-code-excitation information and a synthesis filter configured to perform a time-domain filtering, to provide a reconstructed signal on the basis of the time-domain excitation signal and in dependence on linear-prediction-domain filter coefficients obtained on the basis of the linear-prediction-domain para- meter information. The transform domain path is configured to selectively provide the aliasing-cancellation synthesis signal for a portion of the audio content encoded in the transform-coded-excitation linear-prediction-domain mode following a portion of the audio content encoded in the ACELP mode and for a portion of the content encoded in the transfer-coded-excitation-linear-prediction-domainmode preceding a portion of the audio content encoded in the ACELP mode. It has been found that the aliasing-cancellation synthesis signal is very well-suited for transitions between portions (for example, frames) encoded in the transform-coded-excitation-linear-prediction-domain (in the following also briefly designated as TCX-LPD) mode and the ACELP mode.

[0020] In a preferred embodiment, the aliasing-cancellation stimulus filter is configured to filter the aliasing-cancellation stimulus signals in dependence on linear-prediction-domain filter parameters which correspond to a left-sided aliasing folding point of the first frequency-domain-to-time-domain converter for a portion of the audio content encoded in theTCX- LPDmode following a portion of the audio content encoded in theACELPmode. Thealiasing-cancellation stimulus filter is configured to filter the aliasing-cancellation stimulus signal in dependence on linear-prediction-domain filter parameters which correspond to a right-sided aliasing folding point of the second frequency-domain-to-time-domain converter for a portion of the audio content encoded in the transform-coded-excitation-linear-prediction-mode preceding a portion of the audio content encoded in the ACELPmode. By applying linear-prediction-domain filter parameters, which correspond to the aliasing folding points, an extremely efficient aliasing-cancellation can be obtained. Also, the linear-prediction-domain filter parameters, which correspond to the aliasing folding points, are typically easily obtainable as the aliasing folding pointsareoftenat the transition fromone frame to thenext, such that the transmissionof said linear-prediction-domainfilter parameters is required anyway. Accordingly, overheads are kept to a minimum.

[0021] In a further embodiment, the audio signal decoder is configured to initialize memory values of the aliasing- cancellation stimulus filter to zero for providing the aliasing-cancellation synthesis signal, and to feed M samples of the aliasing-cancellation stimulus signal into the aliasing-cancellation stimulus filter to obtain corresponding non-zero input response samples of the aliasing-cancellation synthesis signal, and to further obtain a plurality of zero-input response samples of the aliasing-cancellation synthesis signal. The combiner is preferably configured to combine the time-domain representation of the audio content with the non-zero input response samples and the subsequent zero-input response samples, toobtainanaliasing-reduced time-domain signal at a transition fromaportionof theaudio content encoded in the ACELP mode to a portion of the audio content encoded in the TCX-LPD mode following the portion of the audio content encoded in the ACELP mode. By exploiting both, the non-zero input response samples and the zero-input response samples, a very good usage can be made of the aliasing-cancellation stimulus filter. Also, a very smooth aliasing- cancellation synthesis signal can be obtained while keeping a number of required samples of the aliasing-cancellation stimulus signal as small as possible.Moreover, it has been found that a shape of the aliasing-cancellation synthesis signal is verywell-adapted to typical aliasingartifactsbyusing theabove-mentionedconcept. Thus, averygood tradeoffbetween coding efficiency and aliasing-cancellation can be obtained.

[0022] In a preferred embodiment, the audio signal decoder is configured to combine awindowed and folded version of at least a portion of a time-domain representation obtained using theACELPmodewith a time-domain representation of a subsequent portion of the audio content obtained using the TCX-LPDmode, to at least partially cancel an aliasing. It has been found that the usage of such aliasing-cancellation mechanisms, in addition to the generation of the aliasing cancellation synthesis signal, provides the possibility of obtaining an aliasing-cancellation in a very bitrate efficient manner. In particular, the required aliasing-cancellation stimulus signal can be encodedwith high efficiency if the aliasing- cancellation synthesis signal is supported, in the aliasing-cancellation, by the windowed and folded version of at least a portion of a time-domain representation obtained using the ACELP mode.

[0023] In a preferred embodiment, the audio signal decoder is configured to combine a windowed version of a zero impulse responseof the synthesis filter of theACELPbranchwith a time-domain representation of a subsequent portion of the audio content obtained using the TCX-LPD mode, to at least partially cancel an aliasing. It has been found that the 5 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 usageof suchazero impulse responsemayalsohelp to improve the codingefficiency of the aliasing-cancellation stimulus signal, because thezero impulse responseof the synthesis filter of theACELPbranch typically cancels at least a part of the aliasing in the TCX-LPD-encoded portion of the audio content. Accordingly, the energy of the aliasing-cancellation synthesis signal is reduced, which, in turn, results in a reduction of the energy of the aliasing-cancellation stimulus signal. However, encoding signals with a smaller energy is typically possible with reduced bitrate requirements.

[0024] In a preferred embodiment, the audio signal decoder is configured to switch between aTCX-LPDmode, inwhich a capped frequency-domain-to-time-domain transform is used, a frequency-domain mode, in which a tapped frequency- domain-to time-domain transform is used, as well as an algebraic-code-excited-linear-prediction mode. In this case, the audio signal decoder is configured to at least partially cancel an aliasing at a transition between a portion of the audio content encoded in the TCX-LPD mode and a portion of the audio content encoded in the frequency-domain mode by performing an overlap-and-add operation between time domain samples of subsequent overlapping portions of the audio content. Also, the audio signal decoder is configured to at least partially cancel an aliasing at a transition betweena portion of the audio content encoded in the TCX-LPDmode and a portion of the audio content encoded in the ACELPmode using the aliasing-cancellation synthesis signal. It has been found that the audio signal decoder also is well-suited for switching between different modes of operation, wherein the aliasing cancels very efficiently.

[0025] Inapreferredembodiment, theaudiosignal decoder is configured toapplyacommongainvalue foragainscaling ofa time-domain representationprovidedby thefirst frequency-domain-to-time-domainconverter of the transformdomain path (for example, TCX-LPD path) and for a gain scaling of the aliasing-cancellation stimulus signal or the aliasing- cancellation synthesis signal. It has been found that a reuse of this common gain value both for the scaling of the time- domain representation provided by the first frequency-domain-to-time-domain converter and for the scaling of the aliasing-cancellation stimulus signal or aliasing-cancellation synthesis signal allows for the reduction of bitrate required at a transition between portions of the audio content encoded in different modes. This is very important, as a bitrate requirement is increased by the encoding of the aliasing-cancellation stimulus signal in the environment of a transition between portions of the audio content encoded in the different modes.

[0026] In a preferred embodiment, the audio signal decoder is configured to apply, in addition to the spectral shaping performed in dependence on at least the subset of linear-prediction-domain parameters, a spectrumdeshaping to at least a subset of thefirst set of spectral coefficients. In this case, theaudio signal decoder is configured toapply thespectrumde- shaping to at least a subset of a set of aliasing-cancellation spectral coefficients from which the aliasing-cancellation stimulus signal is derived. Applying a spectral deshaping both, to the first set of spectral coefficients, and to the aliasing- cancellation spectral coefficients from which the aliasing cancellation stimulus signal is derived, ensures that the aliasing cancellation synthesis signal iswell-adapted to the "main " audio content signal provided by the first frequency-domain-to- time-domain converter. Again, the coding efficiency for encoding the aliasing cancellation stimulus signal is improved.

[0027] In a preferred environment, the audio signal decoder comprises a second frequency-domain-to-time-domain converter configured to obtain a time-domain representation of the aliasing-cancellation stimulus signal in dependenceon asetof spectral coefficients representing thealiasing-cancellationstimulussignal. In this case, thefirst frequency-domain- to-time-domain converter is configured to perform a lapped transform, which comprises a time-domain aliasing. The second frequency-domain-to-time-domain converter is configured to performanon-lapped transform.Accordingly, a high coding efficiency can be maintained by using the lapped transform for the "main " signal synthesis. Nevertheless, the aliasing-cancellation achieved using an additional frequency-domain-to-time-domain conversion, which is non-lapped. However, it has been found that the combination of the lapped frequency-domain-to-time-domain conversionand thenon- lapped frequency-domain-to-time-domain conversion allows for amore efficient encoding of transitions that a single non- lapped frequency-domain-to-time-domain transition.

[0028] An embodiment according to the invention creates an audio signal encoder for providing an encoded repre- sentation of an audio content comprising a first set of spectral coefficients, a representation of an aliasing-cancellation stimulus signal and a plurality of linear-prediction-domain parameters on the basis of an input representation of the audio content. The audio signal encoder comprises a time-domain-to-frequency-domain converter configured to process the input representation of the audio content, to obtain a frequency-domain representation of the audio content. The audio signal encoder also comprises a spectral processor configured to apply a spectral shaping to a set of spectral coefficients, or to a pre-processed version thereof, in dependence on a set of linear-prediction-domain parameters for a portion of the audio content to be encoded in the linear-prediction-domain, to obtain a spectrally-shaped frequency-domain representa- tion of the audio content. The audio signal encoder also comprises an aliasing-cancellation information provider configured to provide a representation of an aliasing-cancellation stimulus signal, such that a filtering of the aliasing- cancellation stimulus signal in dependence on at least a subset of the linear prediction domain parameters results in an aliasing-cancellation synthesis signal for cancelling aliasing artifacts in an audio signal decoder.

[0029] The audio signal encoder discussed here is well-suited for cooperation with the audio signal encoder described before. In particular, the audio signal encoder is configured to provide a representation of the audio content in which a bitrate overhead required for cancelling aliasingat transitions betweenportions (for example, framesor sub-frames) of the audio content encoded in different modes is kept reasonably small. 6 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55

[0030] Further embodiments according to the invention create a method for providing a decoded representation of the audio content and amethod for providing an encoded representation of an audio content. Saidmethods are based on the same ideas as the apparatus discussed above.

[0031] Embodiments according to the invention create computer programs for performing one of said methods. The computer programs are also based on the same considerations. Brief Description of the Figures

[0032] Embodimentsaccording to thepresent inventionwill subsequently bedescribed taking reference to theenclosed figures, in which: Fig. 1 shows a block schematic diagram of an audio signal encoder, according to an embodiment of the invention; Fig. 2 shows a block schematic diagram of an audio signal decoder, according to an embodiment of the invention; Fig. 3a shows a block schematic diagram of a reference audio signal decoder according to working draft 4 of the Unified Speech and Audio Coding (USAC) draft standard; Fig. 3b shows a block schematic diagram of an audio signal decoder, according to another embodiment of the in- vention; Fig. 4 shows a graphical representation of a reference window transition according to working draft 4 of the USAC draft standard; Fig. 5 shows a schematic representation of window transitions which can be used in an audio signal coding, ac- cording to an embodiment of the invention; Fig. 6 shows a schematic representation providing an overview over all window types used in an audio signal en- coder according to an embodiment of the invention or an audio signal decoder according to an embodiment of the invention; Fig. 7 shows a table representation of allowed window sequences, which may be used in an audio signal encoder according to an embodiment of the invention, or and audio signal decoder according to an embodiment of the invention; Fig. 8 shows a detailed block schematic diagram of an audio signal encoder, according to an embodiment of the invention; Fig. 9 shows a detailed block schematic diagram of an audio signal decoder according to an embodiment of the invention; Fig. 10 shows a schematic representation of forward-aliasing-cancellation (FAC) decoding operations for transi- tions from and to ACELP; Fig. 11 shows a schematic representation of a computation of an FAC target at an encoder; Fig. 12 shows a schematic representation of a quantization of an FAC target in the context of a frequency-domain- noise-shaping (FDNS); Table 1 shows conditions for the presence of a given LPC filter in a bitstream; Fig. 13 shows a schematic representation of a principle of a weighted algebraic LPC inverse quantizer; Table 2 shows a representation of possible absolute and relative quantization modes and corresponding bitstream signaling of "mode _lpc "; Table 3 shows a table representation of coding modes for codebook numbers nk; 7 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 Table 4 shows a table representation of a normalization vector W for AVQ quantization; Table 5 shows a table representation of mapping for a mean excitation energy E ; Table 6 shows a table representation of a number of spectral coefficients as a function of "mod[]; " Fig. 14 shows a representation of a syntax of a frequency-domain channel stream "fd_channel_stream() "; Fig. 15 shows a representation of a syntax of a linear-prediction-domain channel stream "lpd_channel_stream() "; and Fig. 16 shows a representation of a syntax of the forward aliasing-cancellation data "fac_data() ". Detailed Description of the Embodiments 1. Audio Signal Decoder according to Fig. 1

[0033] Fig. 1 shows a block schematic diagram of an audio signal encoder 100, according to an embodiment of the invention. The audio signal encoder 100 is configured to receive an input representation 110 of an audio content and to provide, on the basis thereof, an encoded representation 112 of the audio content. The encoded representation 112 of the audio content comprises a first set 112a of spectral coefficients, a plurality of linear-prediction-domain parameters 112b and a representation 112c of an aliasing-cancellation stimulus signal.

[0034] Theaudio signal encoder 100 comprises a time-domain-to-frequency-domain converter 120which is configured to process the input representation 110 of the audio content (or, equivalently, a pre-processed version 110’ thereof), to obtain a frequency-domain representation 122 of the audio content (which may take the form of a set of spectral coefficients).

[0035] The audio signal encoder 100 also comprises a spectral processor 130 which is configured to apply a spectral shaping to the frequency-domain representation 122 of the audio content, or to a pre-processed version 122’ thereof, in dependence on a set 140 of linear-prediction-domain parameters for a portion of the audio content to be encoded in the linear-prediction-domain, to obtain a spectrally-shaped frequency-domain representation 132 of the audio content. The first set 112a of spectral coefficients may be equal to the spectrally-shaped frequency-domain representation 132 of the audio content, or may be derived from the spectrally-shaped frequency-domain representation 132 of the audio content.

[0036] The audio signal encoder 100 also comprises an aliasing-cancellation information provider 150, which is configured to provide a representation 112c of an aliasing-cancellation stimulus signal, such that a filtering of the aliasing-cancellation stimulus signal in dependence on at least a subset of the linear-prediction-domain parameters 140 results in an aliasing-cancellation synthesis signal for cancelling aliasing artifacts in an audio signal decoder.

[0037] It should also be noted that the linear-prediction-domain parameters 112b may, for example, be equal to the linear-prediction-domain parameters 140.

[0038] The audio signal encoder 110 provides information which is well-suited for a reconstruction of the audio content, even if different portions (for example, frames or sub-frames) of the audio content are encoded in different modes. For a portion of the audio content encoded in the linear-prediction-domain, for example, in a transform-coded-excitation linear- prediction-domainmode, the spectral shaping, which brings along a noise shaping and therefore allows a quantization of the audio contentwith a comparatively small bitrate, is performedafter the time-domain-to-frequency-domain conversion. This allows for an aliasing cancelling overlap-and-add of a portion of the audio content encoded in the linear-prediction- domain with a preceding or subsequent portion of the audio content encoded in a frequency-domain mode. By using the linear-prediction-domain parameters 140 for the spectral shaping, the spectral shaping is well-adapted to speech-like audio contents, such that a particularly good coding efficiency can be obtained for speech-like audio contents. Moreover, the representation of the aliasing-cancellation stimulus signal allows for an efficient aliasing-cancellation at transitions from or towards a portion (for example, frame or sub-frame) of the audio content encoded in the algebraic-code-excited- linear-prediction mode. By providing the representation of the aliasing-cancellation stimulus signal in dependence on the linear prediction domain parameters, a particularly efficient representation of the aliasing-cancellation stimulus signal is obtained, which can be decoded at the side of the decoder taking into consideration the linear-prediction-domain parameters, which are known at the decoder anyway.

[0039] To summarize, the audio signal encoder 100 is well-suited for enabling transitions between portions of the audio content encoded in different codingmodesand is capable of providinganaliasing-cancellation information in aparticularly compact form. 8 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 2. Audio Signal Decoder according to Fig. 2

[0040] Fig. 2 shows a block schematic diagram of an audio signal decoder 200 according to an embodiment of the invention.Theaudiosignal decoder200 isconfigured to receiveanencoded representation210of theaudiocontentand to provide, on the basis thereof, the decoded representation 212 of the audio content, for example, in the form of an aliasing- reduced-time-domain signal.

[0041] The audio signal decoder 200 comprises a transform domain path (for example, a transform-coded-excitation linear-prediction-domain path) configured to obtain a time-domain representation 212 of the audio content encoded in a transform domain mode on the basis of a (first) set 220 of spectral coefficients, a representation 224 of an aliasing- cancellation stimulus signal and a plurality of linear-prediction-domain parameters 222. The transform domain path comprises a spectrum processor 230 configured to apply a spectral shaping to the (first) set 220 of spectral coefficients in dependence on at least a subset of the linear-prediction-domain parameters 222, to obtain a spectrally-shaped version 232 of the first set 220 of spectral coefficients. The transform domain path also comprises a (first) frequency-domain-to- time-domain converter 240 configured to obtain a time-domain representation 242 of the audio content on the basis of the spectrally-shaped version 232 of the (first) set 220 of spectral coefficients. The transform domain path also comprises an aliasing-cancellation stimulus filter 250, which is configured to filter the aliasing-cancellation stimulus signal (which is represented by the representation 224) in dependence on at least a subset of the linear-prediction-domain parameters 222, to derive an aliasing-cancellation synthesis signal 252 from the aliasing-cancellation stimulus signal. The transform domain path also comprises a combiner 260 configured to combine the time-domain representation 242 of the audio content (or, equivalently, a post-processed version 242’ thereof) with the aliasing-cancellation synthesis signal 252 (or, equivalently, a post-processed version 252’ thereof), to obtain the aliasing-reduced time-domain signal 212.

[0042] The audio signal decoder 200 may comprise an optional processing 270 for deriving the setting of the spectrum processor 230, which performs, for example, a scaling and / or frequency-domain noise shaping, from at least a subset of the linear-prediction-domain parameters.

[0043] The audio signal decoder 200 also comprises an optional processing 280, which is configured to derive the settingof thealiasing-cancellation stimulusfilter 250,whichmay, for example, performasynthesis filtering for synthesizing the aliasing-cancellation synthesis signal 252, from at least a subset of the linear-prediction-domain parameters 222.

[0044] Theaudio signal decoder 200 is configured to provide an aliasing-reduced time domain signal 212, which iswell- suited for a combination both, with a time-domain signal representing an audio content and obtained in a frequency- domainmode of operation, and to / in combinationwith a time-domain signal representing an audio content and encoded in an ACELP mode of operation. Particularly good overlap-and-add characteristics exist between portions (for example, frames) of the audio content decoded using a frequency-domain mode of operation (using a frequency-domain path not shown inFig. 2) andportions (for example, a frameor sub-frame) of theaudio content decodedusing the transformdomain path of Fig. 2, as the noise shaping is performed by the spectrum processor 230 in the frequency-domain, i.e. before the frequency-domain-to-time-domain conversion 240. Moreover, particularly good aliasing-cancellations can also be obtained between a portion (for example, a frame or sub-frame) of the audio content decoded using the transform domain path of Fig. 2 and a portion (for example, a frame or sub-frame) of the audio content decoded using an ACELP decoding path due to the fact that the aliasing-cancellation synthesis signal 252 is provided on the basis of a filtering of an aliasing-cancellation stimulus signal in dependence on linear-prediction-domain parameters. An aliasing-cancellation synthesis signal 252, which is obtained in this manner, is typically well-adapted to the aliasing artifacts which occur at the transition between a portion of the audio content encoded in the TCX-LPD mode and a portion of the audio content encoded in the ACELP mode. Further optional details regarding the operation of the audio signal decoding will be described in the following. 3. Switched Audio Decoders according to Figs. 3a and 3b

[0045] In the following, the concept of a multi-mode audio signal decoder will briefly be discussed taking reference to Figs. 3a and 3b. 3.1 Audio Signal Decoder 300 according to Fig. 3a

[0046] Fig. 3a shows a block schematic diagram of a reference multi-mode audio signal decoder, and Fig. 3b shows a block schematic diagram of a multi-mode audio signal decoder, according to an embodiment of the invention. In other words, Fig. 3a shows a basic decoder signal flow of a reference system (for example, according to working draft 4 of the USACdraft standard), andFig. 3b showsabasic decoder signal flowof aproposed systemaccording to anembodiment of the invention.

[0047] The audio signal decoder 300 will be described first taking reference to Fig. 3a. The audio signal decoder 300 comprisesabitmultiplexer 310,which is configured to receive an input bitstreamand toprovide the information included in 9 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 the bitstream to the appropriate processing units of the processing branches.

[0048] The audio signal decoder 300 comprises a frequency-domain mode path 320, which is configured to receive a scale factor information 322 and an encoded spectral coefficient information 324, and to provide, on the basis thereof, a time-domain representation 326of anaudio frameencoded in the frequency-domainmode. The audio signal decoder 300 also comprises a transform-coded-excitation-linear-prediction-domain path 330, which is configured to receive an encoded transform-coded-excitation information 332 and a linear-prediction coefficient information 334, (also designated asa linear-prediction coding information, or asa linear-prediction-domain informationor asa linear-prediction-coding filter information) and to provide, on the basis thereof, a time-domain representation of an audio frame or audio sub-frame encoded in the transform-coded-excitation-linear-prediction-domain (TCX-LPD) mode. The audio signal decoder 300 also comprises an algebraic-code-excited-linear-prediction (ACELP) path 340, which is configured to receive an encoded excitation information342anda linear-prediction-coding information344 (alsodesignatedasa linearpredictioncoefficient information or as a linear prediction domain information or as a linear-prediction-coding filter information) and to provide, on the basis thereof, a time-domain linear-prediction-coding information, to as representation of an audio frame or audio sub-frame encoded in the ACELP mode. The audio signal decoder 300 also comprises a transition windowing, which is configured to receive the time-domain representations 326, 336, 346 of frames or sub-frames of the audio content encoded in the different modes and to combine the time domain representation using a transition windowing.

[0049] The frequency-domain path 320 comprises an arithmetic decoder 320a configured to decode the encoded spectral representation 324, to obtain a decoded spectral representation 320b, an inverse quantizer 320d configured to provide an inversely quantized spectral representation 320e on the basis of the decoded spectral representation 320b, a scaling 320e configured to scale the inversely quantized spectral representation 320d in dependence on scale factors, to obtain a scaled spectral representation 320f and a (inverse) modified discrete cosine transform 320g for providing a time- domain representation 326 on the basis of the scaled spectral representation 320f.

[0050] The TCX-LPD branch 330 comprises an arithmetic decoder 330a configured to provide a decoded spectral representation 330b on the basis of the encoded spectral representation 332, an inverse quantizer 330c configured to provide an inversely quantized spectral representation 330d on the basis of the decoded spectral representation 330b, a (inverse) modified discrete cosine transform 330e for providing an excitation signal 330f on the basis of the inversely quantized spectral representation 330d, and a linear-prediction-coding synthesis filter 330g for providing the time-domain representation 336 on the basis of the excitation signal 330f and the linear-prediction-coding filter coefficients 334 (also sometimes designated as linear-prediction-domain filter coefficients).

[0051] The ACELP branch 340 comprises an ACELP excitation processor 340a configured to provide an ACELP excitation signal 340bon thebasis of theencodedexcitation signal 342anda linear-prediction-coding synthesis filter 340c for providing the time-domain representation 346 on the basis of the ACELP excitation signal 340b and the linear- prediction-coding filter coefficients 344. 3.2 Transition Windowing according to Fig. 4

[0052] Taking reference now to Fig. 4, the transition windowing 350 will be described in more detail. First of all, the general framing structure of an audio signal decoder 300 will be described. However, it should be noted that a very similar framing structurewith onlyminor differences, or even an identical general framing structure, will be used in the other audio signal encoders or decoders described herein. It should also be noted that audio frames typically comprise a length of N samples, wherein Nmay be equal to 2048. Subsequent frames of the audio contentmay be overlapping by approximately 50%, for example, by N / 2 audio samples. An audio framemay be encoded in the frequency-domain, such that the N time- domain samples of an audio frame are represented by a set of, for example, N / 2 spectral coefficients. Alternatively, the N time-domain samples of an audio framemay also be represented by a plurality of, for example, eight sets of, for example, 128 spectral coefficients. Accordingly, a higher temporal resolution can be obtained.

[0053] If theN time-domain samples of an audio frame are encoded in the frequency-domainmode using a single set of spectral coefficients, a single window such as, for example, a so-called "STOP_START "window, a so-called "AACLong " window, a so-called "AAC Start " window, or a so-called "AAC Stop " window may be applied to window the time domain samples 326 provided by the inverse modified discrete cosine transform 320g. In contrast, a plurality of shorter windows, for example of the type "AACShort ", may be applied to window the time-domain representations obtained using different setsof spectral coefficients, if theN time-domain samplesof anaudio frameareencodedusingaplurality of setsof spectral coefficients.Forexample, separateshortwindowsmaybeapplied to time-domain representationsobtainedon thebasisof individual sets of spectral coefficients associated with a single audio frame.

[0054] An audio frame encoded in the linear-prediction-domainmodemay be sub-divided into a plurality of sub-frames, which are sometimes designated as "frames ". Each of the sub-framesmay be encoded either in the TCX-LPDmode or in the ACELP mode. Accordingly, however, in the TCX-LPD mode, two or even four of the sub-frames may be encoded together using a single set of spectral coefficients describing the transform encoded excitation.

[0055] Asub-frame (or agroupof twoor four sub-frames) encoded in theTCX-LPDmodemaybe representedbyaset of 10 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 spectral coefficients and one or more sets of linear-prediction-coding filter coefficients. A sub-frame of the audio content encoded in theACELPdomainmaybe representedbyanencodedACELPexcitation signal andoneormore setsof linear- prediction-coding filter coefficients.

[0056] Taking referencenow toFig. 4, the implementationof transitionsbetween framesor sub-frameswill bedescribed. In the schematic representation of Fig. 4, abscissas 402a to 402i describe a time in terms of audio samples, and ordinates 404a to 404i describe windows and / or temporal regions for which time domain samples are provided.

[0057] At reference numeral 410, a transition between two overlapping frames encoded in the frequency-domain is represented. At reference numeral 420, a transition from a sub-frame encoded in the ACELPmode to a frame encoded in the frequency-domainmode is shown.At reference numeral 430, a transition froma frame (or a sub-frame) encoded in the TCX-LPDmode (also designated as "wLPT " mode) to a frame encoded in the frequency-domain mode as illustrated. At reference numeral 440, a transition betweena frameencoded in the frequency-domainmodeanda sub-frameencoded in the ACELP mode is shown. At reference numeral 450, a transition between sub-frames encoded in the ACELP mode is shown.At reference numeral 460, a transition fromasub-frameencoded in theTCX-LPDmode to a sub-frameencoded in theACELPmode is shown.At referencenumeral 470, a transition froma frameencoded in the frequency-domainmode to a sub-frame encoded in the TCX-LPD mode is shown. At reference numeral 480, a transition between a sub-frame encoded in the ACELP mode and a sub-frame encoded in the TCX-LPD mode is shown. At reference numeral 490, a transition between sub-frames encoded in the mode is shown.

[0058] Interestingly, the transition from theTCX-LPDmode to the frequency-domainmode,which is shownat reference numeral 430, is somewhat inefficient or even TCX-LPD very inefficient due to the fact that a part of the information transmitted to the decoder is discarded. Similarly, transitions between the ACELP mode and the TCX-LPD mode, which are shown at reference numerals 460 and 480, are implemented inefficiently due to the fact that a part of the information transmitted to the decoder is discarded. 3.3 Audio Signal Decoder 360 according to Fig. 3b

[0059] In the following, the audio signal decoder 360, according to an embodiment of the invention will be described.

[0060] The audio signal 360 comprises a bit multiplexer or bitstream parser 362, which is configured to receive a bitstream representation 361 of an audio content and to provide, on the basis thereof, information elements to a different branches of the audio signal decoder 360.

[0061] The audio signal decoder 360 comprises a frequency-domain branch 370 which receives an encoded scale factor information 372 and an encoded spectral information 374 from the bitstreammultiplexer 362 and to provide, on the basis thereof, a time-domain representation 376 of a frame encoded in the frequency-domain mode. The audio signal decoder 360 also comprises a TCX-LPD path 380 which is configured to receive an encoded spectral representation 382 and encoded linear-prediction-coding filter coefficients 384 and to provide, on the basis thereof, a time-domain repre- sentation 386 of an audio frame or audio sub-frame encoded in the TCX-LPD mode.

[0062] The audio signal decoder 360 comprises an ACELPpath 390which is configured to receive an encodedACELP excitation 392 and encoded linear-prediction-coding filter coefficients 394 and to provide, on the basis thereof, a time- domain representation 396 of an audio sub-frame encoded in the ACELP mode.

[0063] The audio signal decoder 360 also comprises a transition windowing 398, which is configured to apply an appropriate transitionwindowing to the time-domain representations376, 386, 396of the framesandsub-framesencoded in the different modes, to derive a contiguous audio signal.

[0064] It should be noted here that the frequency-domain branch 370 may be identical in its general structure and functionality to the frequency-domain branch 320, even though there may be different or additional aliasing-cancellation mechanisms in the frequency-domain branch 370. Moreover, the ACELP branch 390 may be identical to the ACELP branch 340 in its general structure and functionality, such that the above description also applies.

[0065] However, the TCX-LPDbranch 380 differs from the TCX-LPD branch 330 in that the noise-shaping is performed before the inverse-modified-discrete-cosine-transform in the TCX-LPD branch 380. Also, the TCX-LPD branch 380 comprises additional aliasing cancellation functionalities.

[0066] The TCX-LPD branch 380 comprises an arithmetic decoder 380a which is configured to receive an encoded spectral representation 382 and to provide, on the basis thereof, a decoded spectral representation 380b. The TCX-LPD branch380also comprisesan inversequantizer 380c configured to receive thedecodedspectral representation380band to provide, on the basis thereof, an inversely quantized spectral representation 380d. The TCX-LPD branch 380 also comprises a scaling and / or frequency-domain noise-shaping 380e which is configured to receive the inversely quantized spectral representation 380d and a spectral shaping information 380f and to provide, on the basis thereof, a spectrally shaped spectral representation 380g to an inverse modified-discrete-cosine-transform 380h, which provides the time- domain representation 386 on the basis of the spectrally shaped spectral representation 380g. The TCX-LPD branch 380 also comprises a linear-prediction-coefficient-to-frequency-domain transformer 380i which is configured to provide the spectral scaling information 380f on the basis of the linear-prediction-coding filter coefficients 384. 11 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55

[0067] Regarding the functionality of the audio signal decoder 360 it can be said that the frequency-domain branch 370 and the TCX-LPD branch 380 are very similar in that each of them comprises a processing chain having an arithmetic decoding, an inverse quantization, a spectrum scaling and an inverse modified-discrete-cosine-transform in the same processing order. Accordingly, the output signals 376, 386 of the frequency-domain branch 370 and of the TCX-LPD branch 380 are very similar in that theymay both be unfiltered (with the exception of a transitionwindowing) output signals of the inversemodified-discrete-cosine-transforms. Accordingly, the time-domain signals 376, 386are verywell-suited for an overlap-and-add operation, wherein a time-domain aliasing-cancellation is achieved by the overlap-and-add opera- tion. Thus, transitions between an audio frame encoded in the frequency-domain mode and an audio frame or audio sub- frame encoded in the TCX-LPD mode can be efficiently performed by a simple overlap-and-add operation without requiring any additional aliasing-cancellation information and without discarding any information. Thus, a minimum amount of side information is sufficient.

[0068] Moreover, it should be noted that the scaling of the inversely quantized spectral representation, which is performed in the frequency-domain path 370 in dependence on a scale factor information, effectively brings along a noise-shaping of the quantization noise introduced by the encoder-sided quantization and the decoder-sided inverse quantization 320c, which noise-shaping is well-adapted to general audio signals such as, for example, music signals. In contrast, the scaling and / or frequency-domain noise-shaping 380e, which is performed in dependence on the linear- prediction-coding filter coefficients, effectively bringsalonganoise-shapingof a quantization noise causedbyanencoder- sided quantization and the decoder-sided inverse quantization 380c, which is well-adapted to speech-like audio signals. Accordingly, the functionality of the frequency-domain branch 370 and of the TCX-LPD branch 380 merely differs in that different noise-shaping is applied in the frequency-domain, such that a coding efficiency (or audio quality) is particularly good for general audio signals when using the frequency-domain branch 370, and such that a coding efficiency or audio quality is particularly high for speech-like audio signals when using the TCX-LPD branch 380.

[0069] It should be noted that the TCX-LPD branch 380 preferably comprises additional aliasing-cancellation mechan- isms for transitions between audio frames or audio sub-frames encoded in the TCX-LPDmode and in the ACELPmode. Details will be described below. 3.4 Transition Windowing according to Fig. 5

[0070] Fig. 5 shows a graphic representation of an example of an envisionedwindowing scheme, whichmay be applied in theaudio signal decoder 360or in anyother audio signal encoders anddecoders according to thepresent invention. Fig. 5 represents a windowing at possible transitions between frames or sub-frames encoded in different of the nodes. Abscissas 502a to 502i describe a time in terms of audio samples and ordinates 504a to 504i describe windows or sub- frames for providing a time-domain representation of an audio content.

[0071] Agraphical representation at reference numeral 510 shows a transition between subsequent frames encoded in the frequency-domainmode. As can be seen, a time-domain samples provided for a first right half of a frame (for example, by an inversemodified discrete cosine transform (MDCT) 320g) are windowed by a right half 512 of a window, whichmay, for example, be of window type "AAC Long " or of window type "AAC Stop ".

[0072] Similarly, the time-domain samples provided for a left half of a subsequent second frame (for example, by the MDCT320g)maybewindowedusinga left half 514of awindow,whichmay, for example, beofwindow type "AACLong "or "AACStart ". The right half 512may, for example, compriseacomparatively long right sided transitionslopeand the left half 514 of the subsequent window may comprise a comparatively long left sided transition slope. A windowed version of the time-domain representation of the first audio frame (windowed using the rightwindowhalf 512) and awindowed version of the time-domain representation of the subsequent second audio frame (windowed using the left window half 514)may be overlapped and added. Accordingly, aliasing, which arises from the MDCT, may be efficiently cancelled.

[0073] Agraphical representation at reference numeral 520 shows a transition froma sub-frameencoded in theACELP mode to a frame encoded in the frequency-domain mode. A forward-aliasing-cancellation may be applied to reduce aliasing artifacts at such a transition.

[0074] A graphical representation at reference numeral 530 shows a transition from a sub-frame encoded in the TCX- LPDmode toa frameencoded in the frequency-domainmode.Ascanbeseen ,awindow532 isapplied to the time-domain samplesprovidedby the inverseMDCT380hof theTCX-LPDpath,whichwindow532may, for example, beofwindow type "TCX256 ", "TCX512 ", or "TCX1024 ". Thewindow532maycomprisea right-sided transition slope533of length128 time- domain samples. A window 534 is applied to time-domain samples provided by the MDCTof the frequency-domain path 370 for the subsequent audio frame encoded in the frequency-domain mode. The window 534 may, for example, be of window type "Stop Start " or "AAC Stop ", and may comprise a left-sided transition slope 535 having a length of, for example, 128 time-domain samples. The time-domain samples of the TCX-LPDmode sub-framewhich are windowed by the right-sided transition slope 533 are overlapped and added with the time-domain samples of the subsequent audio frame encoded in the frequency-domain mode which are windowed by the left-sided transition slope 535. The transition slopes 533 and 535 arematched, such that an aliasing-cancellation is obtained at the transition from the TCX-LPD-mode- 12 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 encoded sub-frameand the subsequent frequency-domain-mode-encoded sub-frame. The aliasing-cancellation ismade possible by the execution of the scaling / frequency-domain noise-shaping 380e before the execution of the inverseMDCT 380h. In other words, the aliasing-cancellation is caused by the fact that both, the inverse MDCT 320g of the frequency- domainpath370and the inverseMDCT380hof theTCX-LPDpath380are fedwith spectral coefficients towhich thenoise- shaping has already been applied (for example, in the form of the scaling factor-dependent scaling and the LPC filter coefficient dependent scaling).

[0075] A graphical representation at reference numeral 540 shows a transition from an audio frame encoded in the frequency-domain mode to a sub-frame encoded in the ACELP mode. As can be seen, a forward aliasing-cancellation (FAC) is applied in order to reduce, or even eliminate, aliasing artifacts at said transition.

[0076] A graphical representation at reference numeral 550 shows a transition from an audio sub-frame encoded in the ACELP mode to another audio sub-frame encoded in the ACELP mode. No specific aliasing-cancellation processing is required here in some embodiments.

[0077] A graphical representation at reference numeral 560 shows a transition from a sub-frame encoded in the TCX- LPD mode (also designated as wLPT mode) to an audio sub-frame encoded in the ACELP mode. As can be seen, time- domain samples provided by theMDCT 380h of the TCX-LPD branch 380 are windowed using awindow 562, whichmay, for example, be of window type "TCX256 ", "TCX512 " or "TCX1024 ".Window 562 comprises a comparatively short right- sided transition slope 563. Time-domain samples provided for the subsequent audio sub-frame encoded in the ACELP mode comprise a partial temporal overlapwith audio samples provided for the preceding TCX-LPD-mode-encoded audio sub-frame which are windowed by the right-sided transition slope 563 of the window 562. Time-domain audio samples provided for the audio sub-frame encoded in the ACELP mode are illustrated by a block at reference numeral 564.

[0078] As can be seen, a forward aliasing-cancellation signal 566 is added at the transition from the audio frame encoded in the TCX-LPD mode to the audio frame encoded in the ACELP mode in order to reduce or even eliminate aliasing artifacts. Details regarding the provision of the aliasing-cancellation signal 566 will be described below.

[0079] A graphical representation at reference numeral 570 shows a transition from a frame encoded in the frequency- domain mode to a subsequent frame encoded in the TCX-LPD mode. Time-domain samples provided by the inverse MDCT 320g of the frequency-domain branch 370may be windowed by a window 572 having a comparatively short right- sided transition slope 573, for example, by a window of type "Stop Start " or a window of type "AACStart ". A time-domain representation provided by the inverse MDCT 380h of the TCX-LPD branch 380 for the subsequent audio sub-frame encoded in the TCX-LPDmodemay bewindowed by a window 574 comprising a comparatively short left-sided transition slope 575, which window 574 may, for example, be of window type "TCX256 ", TCX512 ", or "TCX1024 ". Time-domain sampleswindowed by the right-sided transition slope 573 and time-domain samples windowed by the left-sided transition slope 575 are overlapped and added by the transition windowing 398, such that aliasing artifacts are reduced, or even eliminated. Accordingly, no additional side information is required for performing a transition froman audio frameencoded in the frequency-domain mode to an audio sub-frame encoded in the TCX-LPD mode.

[0080] A graphical representation at reference numeral 580 shows a transition from an audio frame encoded in the ACELPmode to an audio frame encoded in the TCX-LPDmode (also designated as wLPT mode). A temporal region for which time-domain samples are provided by the ACELP branch is designated with 582. A window 584 is applied to time- domain samples provided by the inverse MDCT 380h of the TCX-LPD branch 380. Window 584, which may be of type "TCX256 ", TCX512 ", or "TCX1024 ", may comprise a comparatively short left-sided transition slope 585. The left-sided transition slope 585 of the window 584 partially overlaps with the time-domain samples provided by the ACELP branch, which are represented by the block 582. In addition, an aliasing-cancellation signal 586 is provided to reduce, or even eliminate, aliasingartifactswhichoccurat the transition from theaudiosub-frameencoded in theACELPmode to theaudio sub-frame encoded in the TCX-LPD mode. Details regarding the provision of the aliasing-cancellation signal 586 will be discussed below.

[0081] Aschematic representation at referencenumeral 590 showsa transition fromanaudio sub-frameencoded in the TCX-LPD mode to another audio sub-frame encoded in the TCX-LPD mode. Time-domain samples of a first audio sub- frame encoded in the TCX-LPDmode are windowed using a window 592, which may, for example, be of type "TCX256 ", TCX512 ", or "TCX1024 ", and which may comprise a comparatively short right-sided transition slope 593. Time-domain audio samples of a second audio sub-frame encoded in the TCX-LPD mode, which are provided by the inverse MDCT 380h of the TCX-LPD branch 380 are windowed, for example, using a window 594 which may be of the window type "TCX256 ", TCX512 ", or "TCX1024 " andwhichmay comprise a comparatively short left-sided transition slope 595. Time- domain sampleswindowedusing the right-sided transitional slope593and time-domain sampleswindowedusing the left- sided transition slope 595 are overlapped and added by the transitional windowing 398. Accordingly, aliasing, which is caused by the (inverse) MDCT 380h is reduced, or even eliminated. 4. Overview over all Window Types

[0082] In the following, an overview of all window types will be provided. For this purpose, reference is made to Fig. 6, 13 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 which shows a graphical representation of the different window types and their characteristics. In the table of Fig. 6, a column 610 describes a left-sided overlap length, which may be equal to a length of a left-sided transition slope. The column 612 describes a transform length, i.e. a number of spectral coefficients used to generate the time-domain representationwhich iswindowedby the respectivewindow.Thecolumn614describesa right-sidedoverlap length,which may be equal to a length of a right-sided transition slope. A column 616 describes a name of thewindow type. The column 618 shows a graphical representation of the respective window.

[0083] A first row 630 shows the characteristics of a window of type "AAC Short ". A second row 632 shows the characteristics of a window of type "TCX256 ". A third row 634 shows the characteristics of a window of type "TCX512 ". A fourth row 636 shows the characteristics of windows of types "TCX1024 " and "Stop Start ". A fifth row 638 shows the characteristics of awindowof type "AACLong ".A sixth row640shows the characteristics of awindowof type "AACStart ", and a seventh row 642 shows the characteristics of a window of type "AAC Stop ".

[0084] Notably, the transition slopes of the windows of types "TCX256 ", TCX512 ", and "TCX1024 " are adapted to the right-sided transition slope of the window of type "AAC Start " and to the left-sided transition slope of the window of type "AACStop ", in order to allow for a time-domain aliasing-cancellation by overlapping and adding time-domain representa- tions windowed using different types of windows. In a preferred embodiment, the left-sided window slopes (transition slopes) of all of thewindow types having identical left-sided overlap lengthsmay be identical, and the right-sided transition slopes of all window types having identical right-sided overlap lengths may be identical. Also, left-sided transition slopes and right-sided transition slopes having an identical overlap lengths may be adapted to allow for an aliasing-cancellation, fulfilling the conditions for the MDCTaliasing-cancellation. 5. Allowed Window Sequences

[0085] In the following, allowed window sequences will be described, taking reference to Fig. 7, which shows a table representation of such allowed windowed sequences. As can be seen from the table of Fig. 7, an audio frame encoded in the frequency-domainmode, the time-domain samplesofwhicharewindowedusingawindowof type "AACStop ",maybe followed by an audio frame encoded in the frequency-domain mode, the time-domain samples of which are windowed using a window of type "AAC Long " or a window of type "AAC Start ".

[0086] An audio frame encoded in the frequency-domainmode, the time-domain samples of which arewindowed using awindowof type "AACLong "maybe followedbyanaudio frameencoded in the frequency-domainmode, the time-domain samples of which are windowed using a window of type "AAC Long " or "AAC Start ".

[0087] Audio frames encoded in the linear prediction mode, the time-domain samples of which are windowed using a window of type "AACStart ", using eight windows of type "AACShort " or using awindow of type "AACStopStart ", may be followed by an audio frame encoded in the frequency-domain mode, the time-domain samples of which are windowed using eight windows of type "AACShort ", using awindowof type "AACShort " or using awindowof type "AACStopStart ". Alternatively, audio frames encoded in the frequency-domain mode, the time-domain samples of which are windowed using a window of type "AACStart ", using eight windows of type "AACShort " or using a window of type "AACStopStart " may be followed by an audio frame or sub-frame encoded in the TCX-LPDmode (also designated as LPD-TCX) or by an audio frame or audio sub-frame encoded in the ACELP mode (also designated as LPD ACELP).

[0088] An audio frame or audio sub-frame encoded in the TCX-LPDmodemay be followed by audio frames encoded in the frequency-domain mode, the time-domain samples of which are windowed using eight "AAC Short " windows, and using "AAC Stop " window or using an "AAC StopStart " window, or by an audio frame or audio sub-frame encoded in the TCX-LPD mode or by an audio frame or audio sub-frame encoded in the ACELP mode.

[0089] Anaudio frameencoded in theACELPmodemaybe followedbyaudio framesencoded in the frequency-domain mode, the time-domain samples of which are windowed using eight "AACShort " windows, using an "AACStop " window, using an "AACStopStart " window, by an audio frame encoded in the TCX-LPDmode or by an audio frame encoded in the ACELP mode.

[0090] For transitions from an audio frame encoded in the ACELP mode towards an audio frame encoded in the frequency-domain mode or towards an audio frame encoded in the TCX-LPD mode, a so-called forward-aliasing- cancellation (FAC) is performed. Accordingly, an aliasing-cancellation synthesis signal is added to the time-domain representation at such a frame transition, whereby aliasing artifacts are reduced, or even eliminated. Similarly, a FAC is alsoperformedwhenswitching froma frameor sub-frameencoded in the frequency-domainmode, or froma frameor sub- frame encoded in the TCX-LPD mode, to a frame or sub-frame encoded in the ACELP mode.

[0091] Details regarding the FAC will be discussed below. 6. Audio Signal Encoder according to Fig. 8

[0092] In the following, a multi-mode audio signal encoder 800 will be described taking reference to Fig. 8.

[0093] The audio signal encoder 800 is configured to receive an input representation 810 of an audio content and to 14 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 provide, on the basis thereof, a bitstream 812 representing the audio content. The audio signal encoder 800 is configured to operate in different modes of operation, namely a frequency-domain mode, a transform-coded-excitation-linear- prediction-domain mode and an algebraic-code-excited-linear-prediction-domain mode. The audio signal encoder 800 comprises and encoding controller 814 which is configured to select one of the modes for encoding a portion of the audio content in dependence on characteristics of the input representation 810 of the audio content and / or in dependence on an achievable encoding efficiency or quality.

[0094] Theaudio signal encoder800comprisesa frequency-domainbranch820which is configured toprovideencoded spectral coefficients822, encodedscale factors824, andoptionally, encodedaliasing-cancellation coefficients826, on the basis of the input representation 810 of the audio content. The audio signal encoder 800 also comprises a TCX-LPD branch 850 configured to provide encoded spectral coefficients 852, encoded linear-prediction-domain parameters 854 and encoded aliasing-cancellation coefficients 856, in dependence on the input representation 810 of the audio content. The audio signal decoder 800 also comprises an ACELP branch 880 which is configured to provide an encoded ACELP excitation 882 and encoded linear-prediction-domain parameters 884 in dependence on the input representation 810 of the audio content.

[0095] The frequency-domain branch 820 comprises a time-domain-to-frequency-domain conversion 830 which is configured to receive the input representation 810of the audio content, or a pre-processed version thereof, and to provide, on the basis thereof, a frequency-domain representation 832of the audio content. The frequency-domain branch820also comprises a psychoacoustic analysis 834, which is configured to evaluate frequency masking effects and / or temporal masking effects of the audio content, and to provide, on the basis thereof, a scale factor information 836 describing scale factors. The frequency-domain branch 820 also comprises a spectral processor 838 configured to receive the frequency- domain representation 832 of the audio content and the scale factor information 836 and to apply a frequency-dependent and time-dependent scaling to the spectral coefficients of the frequency-domain representation 832 in dependence on the scale factor information836, to obtain a scaled frequency-domain representation 840of theaudio content. The frequency- domain branch also comprises a quantization / encoding 842 configured to receive the scaled frequency-domain repre- sentation840and toperformaquantizationandanencoding inorder toobtain theencodedspectral coefficients822on the basis of the scaled frequency-domain representation 840. The frequency-domain branch also comprises a quantizatio- n / encoding 844 configured to receive the scale factor information 836 and to provide, on the basis thereof, an encoded scale factor information 824. Optionally, the frequency-domain branch 820 also comprises an aliasing-cancellation coefficient calculation 846 which may be configured to provide the aliasing-cancellation coefficients 826.

[0096] The TCX-LPD branch 850 comprises a time-domain-to-frequency-domain conversion 860, which may be configured to receive the input representation 810 of the audio content, and to provide on the basis thereof, a frequency-domain representation 861 of the audio content. The TCX-LPD branch 850 also comprises a linear-predic- tion-domain-parameter calculation 862which is configured to receive the input representation 810 of the audio content, or a pre-processed version thereof, and to derive one or more linear-prediction-domain parameters (for example, linear- prediction-coding-filter-coefficients) 863 from the input representation 810of the audio content. TheTCX-LPDbranch850 also comprises a linear-prediction-domain-to-spectral domain conversion 864, which is configured to receive the linear- prediction-domain parameters (for example, the linear-prediction-coding filter coefficients) and to provide a spectral- domain representationor frequency-domain representation 865on thebasis thereof. The spectral-domain representation or frequency-domain representation of the linear-prediction-domain parameters may, for example, represent a filter response of a filter defined by the linear-prediction-domain parameters in a frequency-domain or spectral-domain. The TCX-LPD branch 850 also comprises a spectral processor 866, which is configured to receive the frequency-domain representation861, or apre-processed version861’ thereof, and the frequency-domain representationor spectral domain representation of the linear-prediction-domain parameters 863. The spectral processor 866 is configured to perform a spectral shaping of the frequency-domain representation 861, or of the pre-processed version 861’ thereof, wherein the frequency-domain representation or spectral domain representation 865 of the linear-prediction-domain parameters 863 serves to adjust the scaling of the different spectral coefficients of the frequency-domain representation 861 or of the pre- processed version 861’ thereof. Accordingly, the spectral processor 866 provides a spectrally shaped version 867 of the frequency-domain representation 861 or of the pre-processed version 861’ thereof, in dependence on the linear- prediction-domain parameters 863. The TCX-LPD branch 850 also comprises a quantization / encoding 868 which is configured to receive the spectrally shaped frequency-domain representation 867 and to provide, on the basis thereof, encoded spectral coefficients 852. TheTCX-LPDbranch850also comprises another quantization / encoding 869,which is configured to receive the linear-prediction-domain parameters 863 and to provide, on the basis thereof, the encoded linear-prediction-domain parameters 854.

[0097] The TCX-LPD branch 850 further comprises an aliasing-cancellation coefficient provision which is configured to provide the encoded aliasing-cancellation coefficients 856. The aliasing cancellation coefficient provision comprises an error computation 870 which is configured to compute an aliasing error information 871 in dependence on the encoded spectral coefficients, aswell as in dependenceon the input representation 810of the audio content. Theerror computation 870 may optionally take into consideration an information 872 regarding additional aliasing-cancellation components, 15 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 which can be provided by other mechanisms. The aliasing-cancellation coefficient provision also comprises an analysis filter computation 873which is configured toprovide an information 873adescribinganerror filtering in dependenceon the linear-prediction-domain parameters 863. Thealiasing-cancellation coefficient provision also comprises anerror analysis filtering 874, which is configured to receive the aliasing error information 871 and the analysis filter configuration information 873a, and to apply an error analysis filtering, which is adjusted in dependence on the analysis filtering information 873a, to the aliasing error information 871, to obtain a filtered aliasing error information 874a. The aliasing- cancellation coefficientprovisionalsocomprisesa time-domain-to-frequency-domainconversion875,whichmay take the functionality of a discrete cosine transform of type IV, and which is configured to receive the filtered aliasing error information874aand toprovide, on thebasis thereof, a frequency-domain representation875aof the filteredaliasingerror information 874a. The aliasing-cancellation coefficient provision also comprises a quantization / encoding 876 which is configured to receive the frequency-domain representation 875a and, to provide on the basis thereof, encoded aliasing- cancellation coefficients 856, such that the encoded aliasing-cancellation coefficients 856 encode the frequency-domain representation 875a.

[0098] The aliasing-cancellation coefficient provision also comprises an optional computation 877 of an ACELP contribution to an aliasing-cancellation. The computation 877 may be configured to compute or estimate a contribution to analiasing-cancellationwhich canbederived fromanaudio sub-frameencoded in theACELPmodewhichprecedesan audio frame encoded in the TCX-LPDmode. The computation of the ACELP contribution to the aliasing-cancellationmay comprise a computation of a post-ACELP synthesis, a windowing of the post-ACELP synthesis and a folding of the windowed post-ACELP synthesis, to obtain the information 872 regarding the additional aliasing-cancellation compo- nents, whichmay be derived from a preceding audio sub-frame encoded in the ACELPmode. In addition, or alternatively, the computation 877 may comprise a computation of a zero-input response of a filter initialized by a decoding of a preceding audio sub-frame encoded in the ACELP mode and a windowing of said zero-input response, to obtain the information 872 about the additional aliasing-cancellation components.

[0099] In the following, the ACELP branch 880 will briefly be discussed. The ACELP branch 880 comprises a linear- prediction-domain parameter calculation 890 which is configured to compute linear-prediction-domain parameters 890a on the basis of the input representation 810 of the audio content. The ACELP branch 880 also comprises an ACELP excitation computation 892 configured to compute an ACELP excitation information 892 in dependence on the input representation 810 of the audio content and the linear-prediction-domain parameters 890a. The ACELP branch 880 also comprises an encoding 894 configured to encode the ACELP excitation information 892, to obtain the encoded ACELP excitation 882. In addition, the ACELP branch 880 also comprises a quantization / encoding 896 configured to receive the linear-prediction-domain parameters 890a and to provide, on the basis thereof, the encoded linear-prediction-domain parameters 884.

[0100] The audio signal decoder 800 also comprises a bitstream formatter 898 which is configured to provide the bitstream812on thebasis of theencodedspectral coefficients 822, theencodedscale factor information824, the aliasing- cancellation coefficients 826, the encoded spectral coefficients 852, the encoded linear-prediction-domain parameters 852, the encoded aliasing-cancellation coefficients 856, the encoded ACELP excitation 882, and the encoded linear- prediction-domain parameters 884.

[0101] Details regarding the provision of the encoded aliasing-cancellation coefficients 852 will be described below. 7. Audio Signal Decoder according to Fig. 9

[0102] In the following, an audio signal decoder 900 according to Fig. 9 will be described.

[0103] Theaudio signal decoder 900according toFig. 9 is similar to theaudio signal decoder 200according toFig. 2 and also to the audio signal decoder 360 according to Fig. 3b, such that the above explanations also hold.

[0104] The audio signal decoder 900 comprises a bit multiplexer 902 which is configured to receive a bitstream and to provide information extracted from the bitstream to the corresponding processing paths.

[0105] The audio signal decoder 900 comprises a frequency-domain branch 910, which is configured to receive encoded spectral coefficients 912 and an encoded scale factor information 914. The frequency-domain branch 910 is optionally configured to also receive encoded aliasing-cancellation coefficients, which allow for a so-called forward- aliasing-cancellation, for example, at a transition between an audio frame encoded in the frequency-domainmode and an audio frame encoded in the ACELPmode. The frequency-domain path 910 provides a time-domain representation 918 of the audio content of the audio frame encoded in the frequency-domain mode.

[0106] The audio signal decoder 900 comprises a TCX-LPD branch 930, which is configured to receive encoded spectral coefficients 932, encoded linear-prediction-domain parameters 934 and encoded aliasing-cancellation coeffi- cients 936, and toprovide, on thebasis thereof, a time-domain representation of anaudio frameor a sub-frameencoded in the TCX-LPDmode. The audio signal decoder 900 also comprises an ACELP branch 980, which is configured to receive an encoded ACELP excitation 982 and encoded linear-prediction-domain parameters 984, and to provide, on the basis thereof, a time-domain representation 986 of an audio frame or audio sub-frame encoded in the ACELP mode. 16 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 7.1 Frequency Domain Path

[0107] In the following, details regarding the frequency domain path 910 will be described. It should be noted that the frequency-domain path is similar to the frequency-domain path 320 of the audio decoder 300, such that reference ismade to the above description. The frequency-domain branch 910 comprises an arithmetic decoding 920, which receives the encoded spectral coefficients 912 and provides, on the basis thereof, the coded spectral coefficients 920a, and an inverse quantization 921 which receives the decoded spectral coefficients 920a, and provides, on the basis thereof, inversely quantized spectral coefficients 921a. The frequency-domain branch 910 also comprises a scale factor decoding 922, which receives theencodedscale factor informationandprovides, on thebasis thereof, a decodedscale factor information 922a. The frequency-domain branch comprises a scaling 923which receives the inversely quantized spectral coefficients 921a and scales the inversely quantized spectral coefficients in accordance with the scale factors 922a, to obtain scaled spectral coefficients 923a. For example, scale factors 922amay be provided for a plurality of frequency bands, wherein a plurality of frequencybins of the spectral coefficients 921aareassociated to each frequency-band.Accordingly, frequency band-wise scaling of the spectral coefficients 921amay be performed. Thus, a number of scale factors associatedwith an audio frame is typically smaller than a number of spectral coefficients 921a associated with the audio frame. The frequency-domain branch 910 also comprises an inverse MDCT 924, which is configured to receive the scaled spectral coefficients 923a and to provide, on the basis thereof, a time-domain representation 924a of the audio content of the current audio frame.The frequencydomainbranch910also, optionally, comprisesacombining925,which is configured to combine the time-domain representation 924a with an aliasing-cancellation synthesis signal 929a, to obtain the time- domain representation 918. However, in some other embodiments the combining 925may be omitted, such that the time- domain representation 924a is provided as the time-domain representation 918 of the audio content.

[0108] In order to provide the aliasing-cancellation synthesis signal 929a, the frequency-domain path comprises a decoding 926a, which provides decoded aliasing-cancellation coefficients 926b, on the basis of the encoded aliasing- cancellation coefficients 916, and a scaling 926c of aliasing-cancellation coefficients, which provides scaled aliasing- cancellation coefficients 926d on the basis of the decoded aliasing-cancellation coefficients 926b. The frequency-domain pathalsocomprisesan inversediscrete-cosine-transformof type IV927,which is configured to receive thescaledaliasing- cancellation coefficients 926d, and to provide, on the basis thereof, an aliasing-cancellation stimulus signal 927a,which is input into a synthesis filtering 927b. The synthesis filtering 927b is configured to perform a synthesis filtering operation on the basis of the aliasing-cancellation stimulus signal 927a and in dependence on synthesis filtering coefficients 927c, which are provided by a synthesis filter computation 927d, to obtain, as a result of the synthesis filtering, the aliasing- cancellation signal 929a. The synthesis filter computation 927d provides the synthesis filter coefficients 927c in dependence on the linear-prediction-domain parameters, which may be derived, for example, from linear-prediction- domain parameters provided in the bitstream for a frame encoded in the TCX-LPD mode, or for a frame provided in the ACELP mode (or may be equal to such linear-prediction-domain parameters).

[0109] Accordingly, the synthesis filtering 927b is capable of providing the aliasing-cancellation synthesis signal 929a, which may be equivalent to the aliasing-cancellation synthesis signal 522 shown in Fig. 5, or to the aliasing-cancellation synthesis signal 542 shown in Fig. 5. 7.2 TCX-LPD Path

[0110] In the following, theTCX-LPDpath of theaudio signal decoder 900will brieflybediscussed. Further detailswill be provided below.

[0111] The TCX-LPD path 930 comprises a main signal synthesis 940 which is configured to provide a time-domain representation 940a of the audio content of an audio frame or audio sub-frame on the basis of the encoded spectral coefficients 932 and the encoded linear-prediction-domain parameters 934. The TCX-LPDbranch 930 also comprises an aliasing-cancellation processing which will be described below.

[0112] The main signal synthesis 940 comprises an arithmetic decoding 941 of spectral coefficients, wherein the decoded spectral coefficients 941a are obtained on the basis of the encoded spectral coefficients 932. The main signal synthesis 940 also comprises an inverse quantization 942, which is configured to provide inversely quantized spectral coefficients 942a on the basis of the decoded spectral coefficients 941a. An optional noise filling 943may beapplied to the inversely quantized spectral coefficients 942a to obtain noise-filled spectral coefficients. The inversely quantized and noise-filled spectral coefficient 943a may also be designated with r[i]. The inversely quantized and noise-filled spectral coefficients943a, r[i]maybeprocessedbyaspectrumde-shaping944, toobtainspectrumde-shapedspectral coefficients 944a, which are also sometimes designated with r[i]. A scaling 945 may be configured as a frequency-domain noise shaping 945. In the frequency-domain noise-shaping 945, a spectrally shaped set of spectral coefficients 945a are obtained, which are also designated with rr[i]. In the frequency-domain noise-shaping 945, contributions of the spectrally de-shaped spectral coefficients 944a onto the spectrally shaped spectral coefficients 945a are determined by frequency- domain noise-shaping parameters 945b, which are provided by a frequency-domain noise-shaping parameter provision 17 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 whichwill be discussed in the following. Bymeans of the frequency-domain noise-shaping 945, spectral coefficients of the spectrally de-shaped set of spectral coefficients 944a are given a comparatively large weight, if a frequency-domain response of a linear-prediction filter described by the linear-prediction-domain parameters 934 takes a comparatively small value for the frequencyassociatedwith the respective spectral coefficient (out of theset 944aof spectral coefficients) under consideration. In contrast, a spectral coefficient out of the set 944a of spectral coefficient is given a comparatively larger weight when obtaining the corresponding spectral coefficients of the set 945a of spectrally shaped spectral coefficients, if the frequency-domain response of a linear-prediction filter described by the linear-prediction-domain parameters 934 takes a comparatively small value for the frequency associatedwith the spectral coefficient (out of the set 944a) under consideration. Accordingly, a spectral shaping, which is defined by the linear-prediction-domain parameters 934, is applied in the frequency-domain when deriving the spectrally-shaped spectral coefficient 945a from the spectrally de-shaped spectral coefficient 944a.

[0113] The main signal synthesis 940 also comprises an inverse MDCT 946, which is configured to receive the spectrally-shaped spectral coefficients 945a, and to provide, on the basis thereof, a time-domain representation 946a. A gain scaling 947 is applied to the time-domain representation 946a, to derive the time-domain representation 940a of the audio content from the time-domain signal 946a. A gain factor g is applied in the gain scaling 947, which is preferably a frequency-independent (non-frequency selective) operation.

[0114] The main signal synthesis also comprises a processing of the frequency-domain noise-shaping parameters 945b, which will be described in the following. For the purpose of providing the frequency-domain noise-shaping parameters 945b, the main signal synthesis 940 comprises a decoding 950, which provides decoded linear-predic- tion-domain parameters 950a on the basis of the encoded linear-prediction-domain parameters 934. The decoded linear- prediction-domain parameters may, for example, take the form of a first set LPC1 of decoded linear-prediction-domain parameters and a second set LPC2 of linear-prediction-domain parameters. The first set LPC1 of the linear-prediction- domain parameters may, for example, be associated with a left-sided transition of a frame or sub-frame encoded in the TCX-LPD mode, and the second set LPC2 of linear-prediction-domain parameters may be associated with a right-sided transition of the TCX-LPD encoded audio frame or audio sub-frame. The decoded linear-prediction-domain parameters are fed into a spectrum computation 951, which provides a frequency-domain representation of an impulse response defined by the linear-prediction-domain parameters 950a. For example, separate sets of frequency-domain coefficients X0[k]may be provided for the first set LPC1 and for the second set LPC2 of decoded linear-prediction-domain parameters 950.

[0115] Again computation 952maps the spectral valuesX0[k] onto gain values,wherein a first set of -gain values g1[k] is associated with the first set LPC1 of spectral coefficients and wherein a second set of gain values g2[k] is associated with the second set LPC2of spectral coefficients. For example, the gain valuesmaybe inversely proportional to amagnitudeof the corresponding spectral coefficients. A filter parameter computation 953may receive thegain values952aandprovide, on the basis thereof, filter parameters 945b for the frequency-domain shaping 945. For example, filter parameters a[i] and b[i] may be provided. The filter parameters 945d determine the contribution of spectrally de-shaped spectral coefficients 944a onto the spectrally-scaled spectral coefficients 945a. Details regarding a possible computation of the filter parameters will be provided below.

[0116] The TCX-LPD branch 930 comprises a forward-aliasing-cancellation synthesis signal computation, which comprises two branches. A first branch of the (forward) aliasing-cancellation synthesis signal generation comprises a decoding 960, which is configured to receive encoded aliasing-cancellation coefficients 936, and to provide on the basis thereof, decodedaliasing-cancellation coefficients 960a,which are scaledbya scaling 961 in dependence onagain value g to obtain a scaled aliasing-cancellation coefficients 961a. The same gain value gmay be used for the scaling 961 of the aliasing-cancellation coefficients 960aand for the gain scaling 947of the time-domain signal 946a provided by the inverse MDCT 946 in some embodiments. The aliasing-cancellation synthesis signal generation also comprises a spectrum de- shaping 962, which may be configured to apply a spectrum de-shaping to the scaled aliasing-cancellation coefficients 961a, to obtain gain scaled and spectrum de-shaped aliasing-cancellation coefficients 962a. The spectrum de-shaping 962may be performed in a similarmanner to the spectrumde-shaping 944, which shall be described inmore detail below. The gain-scaled and spectrum de-shaped aliasing-cancellation coefficients 962a are input into an inverse discrete- cosine-transform of type IV, which is designatedwith reference numeral 963, and which provides an aliasing-cancellation stimulus signal 963a as a result of the inverse-discrete-cosine-transform which is performed on the basis of the gain- scaled spectrally de-shaped aliasing-cancellation coefficients 962a. A synthesis filtering 964 receives the aliasing- cancellation stimulus signal 963a and provides a first forward aliasing-cancellation synthesis signal 964a by synthesis filtering the aliasing-cancellation stimulus signal 963a using a synthesis filter configured in dependence on synthesis filter coefficients 965a, which are provided by the synthesis filter computation 965 in dependence on the linear-prediction- domain parameters LPC1, LPC2. Details regarding the synthesis filtering 964 and the computation of the synthesis filter coefficients 965a will be described below.

[0117] The first aliasing-cancellation synthesis signal 964a is consequently based on the aliasing-cancellation coeffi- cients 936 as well as on the linear-prediction-domain-parameters. A good consistency between the aliasing-cancellation 18 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 synthesis signal 964a and the time-domain representation 940a of the audio content is reached by applying the same scaling factor g both in theprovision of the time-domain representation 940aof the audio content and in theprovision of the aliasing-cancellation synthesis signal 964, and by applying similar, or even identical, spectrumde-shaping 944, 962 in the provision of the time-domain representation 940a of the audio content and in the provision of the aliasing-cancellation synthesis signal 964.

[0118] The TCX-LPD branch 930 further comprises a provision of additional aliasing-cancellation synthesis signals 973a, 976a in dependence on a precedingACELP frameor sub-frame. This computation 970 of an ACELP contribution to the aliasing-cancellation is configured to receive ACELP information such as, for example a time-domain representation 986 provided by theACELPbranch 980 and / or a content of an ACELP synthesis filter. The computation 970 of theACELP contribution to aliasing-cancellation comprises a computation 971 of a post-ACELP synthesis 971a, a windowing 972 of the post-ACELPsynthesis 971aanda folding 973of the post-ACELPsynthesis 972a.Accordingly, awindowedand folded post-ACELP synthesis 973a is obtained by the folding of the windowed post-ACELP synthesis 972a. In addition, the computation 970 of an ACELP contribution to the aliasing cancellation also comprises a computation 975 of a zero-input response, whichmay be computed for a synthesis filter used for synthesizing a time-domain representation of a previous ACELP sub-frame,wherein the initial state of said synthesis filtermay be equal to the state of theACELPsynthesis filter at the end of the previousACELP sub-frame. Accordingly, a zero-input response975a is obtained, towhich awindowing 976 is applied in order to obtain awindowedzero-input response976a. Further details regarding theprovision of thewindowed zero-input response 976a will be described below.

[0119] Finally, a combining 978 is performed to combine the time-domain representation 940a of the audio content, the first forward-aliasing-cancellation synthesis signal 964a, the second forward-aliasing-cancellation synthesis signal 973a and the third forward-aliasing-cancellation synthesis signal 976a. Accordingly, the time-domain representation 938 of the audio frame or audio sub-frame encoded in the TCX-LPD mode is provided as a result of the combining 978, as will be described in more detail below. 7.3 ACELP Path

[0120] In the following, the ACELP branch 980 of the audio signal decoder 900 will briefly be described. The ACELP branch980 comprises adecoding 988of the encodedACELPexcitation 982, to obtain a decodedACELPexcitation 988a. Subsequently, an excitation signal computation and post-processing 989 of the excitation are performed to obtain a post- processed excitation signal 989a. The ACELP branch 980 comprises a decoding 990 of linear-prediction-domain parameters 984, to obtain decoded linear-prediction-domain parameters 990a. The post-processed excitation signal 989a is filtered, and the synthesis filtering 991 performed, in dependence on the linear-prediction-domain parameters 990a to obtain a synthesized ACELP signal 991a. The synthesized ACELP signal 991a is then processed using a post- processing 992 to obtain the time-domain representation 986 of an audio sub-frame encoded in the ACELP load. 7.4 Combining

[0121] Finally, a combining 996 is performed in order to obtain the time-domain representation 918 of an audio frame encoded in the frequency-domainmode, the time-domain representation 938 of an audio frame encoded in the TCX-LPD mode, and the time-domain representation 986 of an audio frame encoded in the ACELPmode, to obtain a time-domain representation 998 of the audio content.

[0122] Further details will be described in the following. 8. Encoder and Decoder Details 8.1 LPC Filter 8.1.1 Tool Description

[0123] In the following, details regarding the encoding and decoding using linear-prediction coding filter coefficients will be described.

[0124] In the ACELPmode, transmitted parameters include LPC filters 984, adaptive and fixed-codebook indices 982, adaptive and fixed-codebook gains 982.

[0125] In the TCXmode, transmitted parameters include LPC filters 934, energy parameters, and quantization indices 932ofMDCTcoefficients.This sectiondescribes thedecodingof theLPCfilters, for exampleof theLPCfilter coefficientsa1 to a16, 950a, 990a. 19 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 8.1.2 Definitions

[0126] In the following, some definitions will be given.

[0127] The parameter "nb_lpc " describes an overall number of LPC parameters sets which are decoded in the bit stream.

[0128] The bitstream parameter "mode_lpc " describes a coding mode of the subsequent LPC parameters set.

[0129] The bitstream parameter "lpc[k][x] " describes an LPC parameter number x of set k.

[0130] The bitstream parameter "qn k " describes a binary code associated with the corresponding codebook numbers nk. 8.1.3 Number of LPC Filters

[0131] The actual number of LPC filters "nb_lpc " which are encoded within the bitstream depends on the ACELP / TCX mode combination of the superframe, wherein a super frame may be identical to a frame comprising a plurality of sub- frames. The ACELP / TCX mode combination is extracted from the field "lpd_mode " which in turn determines the coding modes, "mod[k] " for k=0 to 3, for each of the 4 frames (also designated as sub-frames) composing the superframe. The mode value is 0 for ACELP, 1 for short TCX (256 samples), 2 for medium size TCX (512 samples), 3 for long TCX (1024 samples). It should benotedhere that the bitstreamparameter "lpd_mode "whichmaybeconsideredasabit-field "mode " defines the coding modes for each of the four frames within the one superframe of the linear-prediction-domain channel stream (which corresponds to one frequency-domainmodeaudio frame such as, for example, an advanced-audio-coding frame or an AAC frame). The codingmodes are stored in an array "mod[] " and take values from 0 to 3. Themapping from the bitstream parameter "LPD_mode " to the array "mod[] " can be determined from table 7.

[0132] Regarding the array "mod[0... 3] " it can be said that the array "mod[] " indicates the respective coding modes in each frame. For details reference is made to table 8, which describes the coding modes indicated by the array "mod[].

[0133] In addition to the 1 to 4 LPC filters of the superframe, an optional LPC filter LPC0 is transmitted for the first super- frame of each segment encoded using the LPD core codec. This is indicated to the LPC decoding procedure by a flag "first_lpd_flag " set to 1.

[0134] The order in which the LPC filters are normally found in the bitstream is: LPC4, the optional LPC0, LPC2, LPC1, and LPC3. The condition for the presence of a given LPC filter within the bitstream is summarized in Table 1.

[0135] Thebitstream isparsed toextract thequantization indicescorresponding toeachof theLPCfilters requiredby the ACELP / TCX mode combination. The following describes the operations needed to decode one of the LPC filters. 8.1.4 General Principle of the Inverse Quantizer

[0136] Inverse quantization of an LPC filter, which may be performed in the decoding 950 or in the decoding 990, is performedas described in Fig. 13. TheLPCfilters are quantized using the line-spectral-frequency (LSF) representation. A first-stage approximation is first computed as described in section 8.1.6. An optional algebraic vector quantized (AVQ) refinement 1330 is then calculated as described in section 8.1.7. The quantized LSF vector is reconstructed by adding 1350 the first-stage approximation and the inverse-weighted AVQ contribution 1342. The presence of an AVQ refinement depends on the actual quantizationmodeof the LPCfilter, as explained in section8.1.5. The inverse-quantized LSF vector is later on converted into a vector of LSP (line spectral pair) parameters, then interpolated and converted again into LPC parameters. 8.1.5 Decoding of the LPC quantization mode

[0137] In the following, the decoding of the LPCquantizationmodewill be described, whichmay be part of the decoding 950 of or the decoding 990.

[0138] LPC4 is alwaysquantizedusinganabsolutequantization approach.Theother LPCfilters canbequantizedusing eitheranabsolutequantizationapproach, oroneof several relativequantizationapproaches.For theseLPCfilters, thefirst informationextracted from thebitstream is thequantizationmode.This information isdenoted "mode_lpc " and is signaled in the bitstream using a variable-length binary code as indicated in the last column of Table 2. 8.1.6 First-stage approximation

[0139] For each LPC filter, the quantizationmode determines how the first-stage approximation of Fig. 13 is computed.

[0140] For the absolute quantization mode (mode_lpc=0), an 8-bit index corresponding to a stochastic VQ-quantized first stage approximation is extracted from the bitstream. The first-stage approximation 1320 is then computed by a simple table look-up. 20 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55

[0141] For relative quantizationmodes, the first-stageapproximation is computed usingalready inverse-quantizedLPC filters, as indicated in the second columnof Table 2. For example, for LPC0 there is only one relative quantizationmode for which the inverse-quantizedLPC4filter constitutes thefirst-stageapproximation. ForLPC1, thereare twopossible relative quantizationmodes, onewhere the inverse-quantized LPC2 constitutes the first-stage approximation, the other for which the average between the inverse-quantized LPC0 and LPC2 filters constitutes the first-stage approximation. As all other operations related to LPC quantization, computation of the first-stage approximation is done in the line spectal frequency (LSF) domain. 8.1.7 AVQ refinement 8.1.7.1 General

[0142] The next information extracted from the bitstream is related to the AVQ refinement needed to build the inverse- quantizedLSFvector. Theonly exception is for LPC1: thebitstreamcontainsnoAVQrefinementwhen this filter is encoded relatively to (LPC0+LPC2) / 2.

[0143] TheAVQ isbasedon the8-dimensionalRE8 lattice vector quantizer used toquantize thespectrum inTCXmodes in AMR-WB+. Decoding the LPC filters involves decoding the two 8-dimensional sub-vectors B̂k , k=1 and 2, of the weighted residual LSF vector.

[0144] The AVQ information for these two subvectors is extracted from the bitstream. It comprises two encoded codebook numbers "qn1 " and "qn2 ", and the corresponding AVQ indices. These parameters are decoded as follows. 8.1.7.2 Decoding of codebook numbers

[0145] The first parameters extracted from the bitstream in order to decode the AVQ refinement are the two codebook numbers nk, k=1 and 2, for each of the two subvectors mentioned above. The way the codebook numbers are encoded depends on the LPC filter (LPCO to LPC4) and on its quantization mode (absolute or relative). As shown in Table 3, there are four different ways to encode nk. The details on the codes used for nk are given below. nk modes 0 and 3: The codebook number nk is encoded as a variable length code qnk, as follows: Q2 → the code for nk is 00 Q3 → the code for nk is 01 Q4 → the code for nk is 10 Others: the code for nk is 11 followed by: Q5 → 0 Q6 → 10 Q0 → 110 Q7 → 1110 Q8 → 11110 etc. nk mode 1: The codebook number nk is encoded as a unary code qnk, as follows: Q0 → unary code for nk is 0 Q2 → unary code for nk is 10 Q3 → unary code for nk is 110 Q4 → unary code for nk is 1110 etc. nk mode 2: The codebook number nk is encoded as a variable length code qnk, as follows: Q2 → the code for nk is 00 Q3 → the code for nk is 01 Q4 → the code for nk is 10 21 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 Others: the code for nk is 11 followed by: Q0 → 0 Q5 → 10 Q6 → 110 etc. 8.1.7.3 Decoding of AVQ indices

[0146] Decoding the LPC filters involves decoding the algebraic VQ parameters describing each quantized sub-vector B̂k of theweighted residual LSF vectors. Recall that each blockBk has dimension 8. For each block B̂k, three sets of binary indices are received by the decoder: a) the codebook number nk, transmitted using an entropy code "qnk " as described above; b) the rank Ik of a selected lattice point z in a so-called base codebook, which indicates what permutation has to be applied to a specific leader to obtain a lattice point z; c) and, if the quantized block B̂k (a lattice point) was not in the base codebook, the 8 indices of the Voronoi extension index vectork; from theVoronoi extension indices, anextension vectorv canbecomputed. Thenumber of bits in each component of index vector k is given by the extension order r, which can be obtained from the code value of index nk. The scaling factor M of the Voronoi extension is given by M = 2r.

[0147] Then, from the scaling factorM, the Voronoi extension vector v (a lattice point inRE8) and the lattice point z in the base codebook (also a lattice point in RE8), each quantized scaled block B̂k can be computed as:

[0148] When there isnoVoronoi extension (i.e.nk<5,M=1andz=0), thebasecodebook iseither codebookQ0,Q2,Q3or Q4 from M. Xie and J.‑P. Adoul, "Embedded algebraic vector quantization (EAVQ) with application to wideband audio coding, "IEEE InternationalConferenceonAcoustics, Speech, andSignalProcessing (ICASSP),Atlanta,GA,USA, vol. 1, pp. 240‑243, 1996.Nobitsare then required to transmit vectork.Otherwise,whenVoronoi extension isusedbecause B̂k is largeenough, then onlyQ3 orQ4 from theabove reference is usedasabase codebook. The selection ofQ3 orQ4 is implicit in the codebook number value nk. 8.1.7.4Computation of the LSF weights

[0149] At the encoder, the weights applied to the components of the residual LSF vector before AVQ quantization are: with: whereLSF1sf is the 1st stageLSFapproximation andW is a scaling factorwhich depends on thequantizationmode (Table 4).

[0150] Thecorresponding inverseweighting1340 isappliedat thedecoder to retrieve thequantized residualLSFvector. 22 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 8.1.7.5 Reconstruction of the inverse-quantized LSF vector

[0151] The inverse-quantized LSF vector is obtained by, first, concatenating the twoAVQ refinement subvectors B̂1 and B̂2 decoded as explained in sections 8.1.7.2 and 8.1.7.3 to formone singleweighted residual LSF vector, then, applying to this weighted residual LSF vector the inverse of the weights computed as explained in section 8.1.7.4 to form the residual LSF vector, and then again, adding this residual LSF vector to the first-stage approximation computed as in section 8.1.6. 8.1.8 Reordering of Quantized LSFs

[0152] Inverse-quantized LSFs are reordered and a minimum distance between adjacent LSFs of 50 Hz is introduced before they are used. 8.1.9 Conversion into LSP parameters

[0153] The inverse quantization procedure described so far results in the set of LPCparameters in the LSFdomain. The LSFs are then converted to the cosine domain (LSPs) using the relation qi = cos(ωi), i=1, . . . ,16 with ωi being the line spectral frequencies (LSF). 8.1.10 Interpolation of LSP parameters

[0154] For each ACELP frame (or sub-frame), although only one LPC filter corresponding to the end of the frame is transmitted, linear interpolation is used to obtain a different filter in each sub-frame (or part of a sub-frame) (4 filters per ACELP frame or sub-frame). The interpolation is performed between the LPC filter corresponding to the end of the previous frame (or sub-frame)and the LPC filter corresponding to the end of the (current) ACELP frame. Let LSP(new) be thenewavailable LSPvector andLSP(old) thepreviously available LSPvector. The interpolatedLSPvectors for theNsfr=4 sub-frames are given by

[0155] The interpolated LSP vectors are used to compute a different LP filter at each sub-frame using the LSP to LP conversion method described in below. 8.1.11 LSP to LP Conversion

[0156] For each sub-frame, the interpolated LSP coefficients are converted into LP filter coefficients ak, 950a, 990a, which are used for synthesizing the reconstructed signal in the sub-frame. By definition, the LSPs of a 16th order LP filter are the roots of the two polynomials and which can be expressed as and with 23 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 and where qi, I = 1,...,16 are the LSFs in the cosine domain also called LSPs. The conversion to the LP domain is done as follows. The coefficients of F1(z) and F2(z) are found by expanding the equations above knowing the quantized and interpolated LSPs. The following recursive relation is used to compute F1(z): for i = 1 to 8 for j = i - 1 down to 1 end end with initial values f1(0) = 1 and f1(‑1) = 0. The coefficients of F2(z) are computed similarly by replacing q2i‑1 by q2i.

[0157] Once the coefficients ofF1(z) andF2(z) are found,F1(z) andF2(z) ismultiplied by 1+z‑1 and 1‑z‑1, respectively, to obtain F’1(z) and F’2(z); that is

[0158] Finally, the LP coefficients are computed from f’1(i) and f’2(i) by

[0159] This is directly derived from the equation , and considering the fact that and are symmetric and asymmetric polynomials, respectively. 8.2.ACELP

[0160] In the following, some details regarding the processing performed by the ACELP branch 980 of the audio signal decoder 900 will be explained to facilitate the understanding of the aliasing-cancellation mechanisms, which will subsequently be described. 8.2.1 Definitions

[0161] In the following, some definitions will be provided.

[0162] The bitstream element "mean_energy " describes the quantized mean excitation energy per frame. The bitstream element "acb_index[sfr] " indicates the adaptive codebook index for each sub-frame.

[0163] The bitstream element "ltp_filtering_flag[sfr] " is an adaptive codebook excitation filtering flag. The bitstream element "lcb_index[sfr] " indicates the innovation codebook index for each sub-frame. The bitstream element "gains[sfr] " describes quantized gains of the adaptive codebook and innovation codebook contribution to the excitation.

[0164] Moreover, for details regarding theencodingof thebitstreamelement "mean_energy ", reference ismade to table 5. 24 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 8.2.2 Setting of the ACELP excitation buffer using the past FD synthesis and LPC0

[0165] In the following, an optional initialization of the ACELP excitation buffer will be described, which may be performed by a block 990b. In case of a transition from FD to ACELP, the past excitation buffer u(n) and the buffer containing thepast pre-emphasizedsynthesisŝ(n) areupdatedusing thepastFDsynthesis (includingFAC)andLPC0(i.e. the LPC filter coefficients of the filter coefficient set LPCO) prior to the decoding of the ACELP excitation. For this the FD synthesis is pre-emphasized by applying the pre-emphasis filter (1‑0.68z‑1), and the result is copied to ŝ(n). The resulting pre-emphasized synthesis is then filtered by the analysis filter (z) using LPC0 to obtain the excitation signal u(n). 8.2.3 Decoding of CELP excitation

[0166] If the mode in a frame is a CELP mode, the excitation consists of the addition of scaled adaptive codebook and fixed codebook vectors. In each sub-frame, the excitation is constructed by repeating the following steps: The information required to decode the CELP information may be considered as the encoded ACELP excitation 982. It should also be noted that the decoding of the CELP excitation may be performed by the blocks 988, 989 of the ACELP branch 980. 8.2.3.1 Decoding of adaptive codebook excitation, in dependence on the bitstream element "acb index[] "

[0167] The received pitch index (adaptive codebook index) is used to find the integer and fractional parts of the pitch lag.

[0168] The initial adaptive codebook excitation vector v’(n) is found by interpolating the past excitation u(n) at the pitch delay and phase (fraction) using an FIR interpolation filter.

[0169] The adaptive codebook excitation is computed for the sub-frame size of 64 samples. The received adaptive filter index (ltp_filtering_flag[]) is then used to decidewhether the filtered adaptive codebook is v(n)= v’(n) or v(n) = 0.18v’(n) + 0.64v’(n - 1) + 0.18v’(n - 2). 8.2.3.2 Decoding of innovation codebook excitation using the bitstream element "icb_index[] "

[0170] The received algebraic codebook index is used to extract the positions and amplitudes (signs) of the excitation pulses and to find the algebraic codevector c(n). That is where mi and si are the pulse positions and signs and M is the number of pulses.

[0171] Once the algebraic codevector c(n) is decoded, a pitch sharpening procedure is performed. First the c(n) is filtered by a pre-emphasis filter defined as follows:

[0172] The pre-emphasis filter has the role to reduce the excitation energy at low frequencies. Next, a periodicity enhancement is performed by means of an adaptive pre-filter with a transfer function defined as: wheren is the sub-frame index (n=0,..,63), andwhereT is a rounded version of the integer partT0and fractional partT0,frac of the pitch lag and is given by: 25 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55

[0173] The adaptive pre-filter Fp(z) colors the spectrum by damping inter-harmonic frequencies, which are annoying to the human ear in case of voiced signals. 8.2.3.3 Decoding of adaptive and innovative codebook gains, described by the bitstream element "gains[] "

[0174] The received 7-bit index per sub-frame directly provides the adaptive codebook gain ĝp and the fixed-codebook gain correction factor γ̂. Thefixedcodebookgain is thencomputedbymultiplying thegaincorrection factor byanestimated fixed codebook gain. The estimated fixed-codebook gain g’c is found as follows. First, the average innovation energy is found by

[0175] Then the estimated gain G’c in dB is found by whereE is thedecodedmeanexcitationenergyper frame.Themean innovativeexcitationenergy ina frame,E, is encoded with 2 bits per frame (18, 30, 42 or 54 dB) as "mean_energy ".

[0176] The prediction gain in the linear domain is given by

[0177] The quantized fixed-codebook gain is given by 8.2.3.4 Computing the reconstructed excitation

[0178] The following steps are for n = 0, ..., 63. The total excitation is constructed by: where c(n) is the codevector from the fixed-codebook after filtering it through the adaptive pre-filter F(z). The excitation signal u’(n) is used to update the content of the adaptive codebook. The excitation signal u’(n) is then post-processed as described in the next section to obtain the post-processed excitation signal u(n) used at the input of the synthesis filter 1 / Â(z). 8.3 Excitation Post-processing 8.3.1General

[0179] In the following, the excitation signal post-processingwill be described, whichmay be performed at block 989. In other words, for signal synthesis a post-processing of excitation elements may be performed as follows. 8.3.2 Gain Smoothing for Noise Enhancement

[0180] A nonlinear gain smoothing technique is applied to the fixed-codebook gain ĝc in order to enhance excitation in noise.Basedon the stability andvoicingof the speechsegment, thegain of the fixed-codebook vector is smoothed inorder to reduce fluctuation in the energy of the excitation in case of stationary signals. This improves the performance in case of stationary background noise. The voicing factor is given by 26 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 with whereEvandEcare theenergies of the scaledpitch codevector and scaled innovation codevector, respectively (rvgives a measure of signal periodicity) . Note that since the value of rv is between ‑1 and 1, the value of λ is between 0 and 1. Note that the factor λ is related to the amount of unvoicingwith a value of 0 for purely voiced segments and a value of 1 for purely unvoiced segments.

[0181] Astability factor θ is computed based on a distancemeasure between the adjacent LP filters. Here, the factor θ is related to the ISF distance measure. The ISF distance is given by where fi are the ISFs in the present frame, and fi (p) are the ISFs in the past frame. The stability factor θ is given by

[0182] The ISF distance measure is smaller in case of stable signals. As the value of θ is inversely related to the ISF distance measure, then larger values of θ correspond to more stable signals. The gain-smoothing factor Sm is given by

[0183] The value ofSm approaches 1 for unvoiced and stable signals, which is the case of stationary background noise signals. For purely voiced signals, or for unstable signals, the value of Sm approaches 0. An initial modified gain g0 is computed by comparing the fixed-codebook gain ĝc to a threshold given by the initial modified gain from the previous sub- frame, g‑1. If ĝc is larger or equal to g‑1, then g0 is computed by decrementing ĝc by 1.5 dB bounded by g0 ≥ g‑1. If ĝc is smaller than g‑1, then g0 is computed by incrementing ĝc by 1.5 dB constrained by g0 ≤ g‑1.

[0184] Finally, the gain is updated with the value of the smoothed gain as follows 8.3.3 Pitch Enhancer

[0185] A pitch enhancer schememodifies the total excitation u’(n) by filtering the fixed-codebook excitation through an innovation filter whose frequency response emphasizes the higher frequencies and reduces the energy of the low frequency portion of the innovative codevector, andwhose coefficients are related to the periodicity in the signal. A filter of the form is used where cpe = 0.125(1 + rv), with rv being a periodicity factor given by rv = (Ev - Ec) / (Ev + Ec) as described above. The filtered fixed-codebook codevector is given by and the updated post-processed excitation is given by

[0186] The above procedure can be done in one step by updating the excitation 989a, u(n) as follows 27 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 8.4 Synthesis and Post-processing

[0187] In the following, the synthesis filtering 991 and the post-processing 992 will be described. 8.4.1 General

[0188] The LP synthesis is performed by filtering the post-processed excitation signal 989a u(n) through the LP synthesis filter 1 / Â(z).The interpolated LP filter per sub-frame is used in the LP synthesis filtering the reconstructed signal in a sub-frame is given by

[0189] The synthesized signal is then de-emphasized by filtering through the filter 1 / (1‑0.68z‑1) (inverse of the pre- emphasis filter applied at the encoder input). 8.4.2 Post-processing of the synthesis signal

[0190] After LP synthesis, the reconstructed signal is post-processed using low-frequency pitch enhancement. Two- band decomposition is used and adaptive filtering is applied only to the lower band. This results in a total post-processing, that is mostly targeted at frequencies near the first harmonics of the synthesized speech signal.

[0191] The signal is processed in twobranches. In the higher branch the decoded signal is filtered by ahigh-pass filter to produce the higher band signal sH. In the lower branch, the decoded signal is first processed through an adaptive pitch enhancer, and then filtered through a low-pass filter to obtain the lower band post-processed signal sLEF. The post- processed decoded signal is obtained by adding the lower band post-processed signal and the higher band signal. The object of the pitch enhancer is to reduce the inter-harmonic noise in the decoded signal, which is achieved here by a time- varying linear filter with a transfer function and described by the following equation: where α is a coefficient that controls the inter-harmonic attenuation, T is the pitch period of the input signal ŝ(n), and sLE(n) is the output signal of the pitch enhancer. Parameters T and α vary with time and are given by the pitch tracking module. Withavalueofα =0.5, thegainof thefilter is exactly 0at frequencies1 / (2T), 3 / (2T), 5 / (2T), etc.; i.e. at themidpoint between theharmonic frequencies1 / T,3 / T,5 / T,etc.Whenαapproaches0, theattenuationbetween theharmonics producedby the filter decreases.

[0192] To confine the post-processing to the low frequency region, the enhanced signal sLE is low pass filtered to produce the signal sLEFwhich is added to the high-pass filtered signal sH to obtain the post-processed synthesis signal sE.

[0193] An alternative procedure equivalent to that described above is used which eliminates the need to high-pass filtering. This is achieved by representing the post-processed signal sE(n) in the z-domain as where PLT(z) is the transfer function of the long-term predictor filter given by 28 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 and HLP(z) is the transfer function of the low-pass filter.

[0194] Thus, the post-processing is equivalent to subtracting the scaled low-pass filtered long-termerror signal from the synthesis signal ŝ(n).

[0195] The value T is given by the received closed-loop pitch lag in each sub-frame (the fractional pitch lag rounded to the nearest integer). A simple tracking for checking pitch doubling is performed. If the normalized pitch correlation at delay T / 2 is larger than 0.95 then the value T / 2 is used as the new pitch lag for post-processing.

[0196] The factor α is given by where ĝp is the decoded pitch gain.

[0197] Note that in TCXmode and during frequency domain coding the value of α is set to zero. A linear phase FIR low- pass filter with 25 coefficients is used, with a cut-off frequency at 5Fs / 256 kHz (the filter delay is 12 samples). 8.5 MDCT based TCX

[0198] In the following, theMDCTbasedTCXwill be described in detail, which is performedby themain signal synthesis 940 of the TXC-LPD branch 930. 8.5.1 Tool description

[0199] When the bitstream variable "core_mode " is equal to 1, which indicates that the encoding is made using linear- prediction-domain parameters, and when one or more of the three TCX modes is selected as the "linear prediction- domain " coding, i.e. one of the 4 array entries of mod[] is greater than 0, the MDCT based TCX tool is used. The MDCT based TCX receives the quantized spectral coefficients 941a from the arithmetic decoder 941. The quantized coefficients 941a (or an inversely quantized version 942a thereof) are first completed by a comfort noise (noise filling 943). LPCbased frequency-domain noise shaping 945 is then applied to the resulting spectral coefficients 943a (or a spectrally de-shaped version944a thereof)andan inverseMDCT transformation946 isperformed toget the time-domain synthesis signal 946a. 8.5.2 Definitions

[0200] In the following, some definitions will be provided. The variable "lg " describes a number of quantized spectral coefficients output by the arithmetic decoder. The bitstream element "noise_factor " describes a noise level quantization index. The variable "noise level " describes a level of noise injected in a reconstructed spectrum. The variable "noise[] " describes a vector of generated noise. The bitstream element "global_gain " describes a re-scaling gain quantization index. The variable "g " describes a re-scaling gain. The variable "rms " describes a root mean square of the synthesized time-domain signal, x[]. The variable "x[] " describes a synthesized time-domain signal. 8.5.3 Decoding Process

[0201] TheMDCT-based TCX requests from the arithmetic decoder 941 a number of quantized spectral coefficients, lg, which is determined by the mod[] value. This value (lg) also defines the window length and shape which will be applied in the inverseMDCT. Thewindow,whichmay beapplied during or after the inverseMDCT946, is composed of three parts, a left side overlap of L samples, amiddle part of ones ofMsamplesanda right overlap part ofR samples. Toobtain anMDCT window of length 2*lg, ZL zeros are added on the left and ZR zeros on the right side. In case of a transition from or to a SHORT_WINDOW, the correspondingoverlap region LorRmayneed to be reduced to 128 in order to adapt to the shorter windowslopeof theSHORT_WINDOW.Consequently the regionMand thecorrespondingzero regionZLorZRmayneed to be expanded by 64 samples each.

[0202] The MDCTwindow, which may be applied during the inverse MDCT 946 or following the inverse MDCT 946, is given by 29 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55

[0203] Table 6 shows a number of spectral coefficients as a function of mod[].

[0204] The quantized spectral coefficients, quant[] 941a, delivered by the arithmetic decoder 941, or the inversely quantized spectral coefficients 942a, are optionally completed by a comfort noise (noise filling 943). The level of the injected noise is determined by the decoded variable noise _factor as follows:

[0205] Anoise vector, noise[], is then computed using a random function, random_sign(), delivering randomly the value ‑1 or +1.

[0206] The quant[] and noise[] vectors are combined to form the reconstructed spectral coefficients vector, r[] 942a, in a way that the runs of 8 consecutive zeros in quant[] are replaced by the components of noise[]. A run of 8 non-zeros are detected according to the formula:

[0207] One obtains the reconstructed spectrum 943a as follows:

[0208] A spectrum de-shaping 944 is optionally applied to the reconstructed spectrum 943a according to the following steps: 1. calculate the energyEm of the 8-dimensional block at indexm for each 8-dimensional block of the first quarter of the spectrum 2. compute the ratio Rm=sqrt(Em / EI), where I is the block index with the maximum value of all Em 3. if Rm<0.1, then set Rm=0.1 4. if Rm<Rm‑1, then set Rm=Rm‑1

[0209] Each 8-dimensional block belonging to the first quarter of spectrum are then multiplied by the factor Rm. Accordingly, the spectrally de-shaped spectral coefficients 944a are obtained.

[0210] Prior to applying the inverse MDCT 946, the two quantized LPC filters LPC1, LPC2 (each of which may be described by filter coefficients a1 to a10) corresponding to both extremity of the MDCT block (i.e. the left and right folding points) are retrieved (block 950), their weighted versions are computed, and the corresponding decimated (64 points, whatever the transform length) spectrums 951a are computed (block 951). These weighted LPC spectrums 951a are computed by applying anODFT (odd discrete Fourier transform) to the LPCfilter coefficients 950a. A complexmodulation is applied to the LPC coefficients before computing the ODFT so that the ODFT frequency bins (used in the spectrum computation 951) are perfectly aligned with the MDCT frequency bins (of the inverse MDCT 946). For example, the weightedLPCsynthesis spectrum951aofagivenLPCfilterÂ(z) (defined, for example,by time-domainfilter coefficientsa1 to a16) is computed as follows: 30 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 with where ŵ[n], n = 0...lpc_order + 1, are the (time-domain) coefficients of the weighted LPC filter given by:

[0211] The gains g[k] 952a can be calculated from the spectral representation X0[k], 951a of the LPC coefficients according to: where M=64 is the number of bands in which the calculated gains are applied.

[0212] Let gl[k] and g2[k], k=0. . . 63, be the decimated LPC spectrums corresponding respectively to the left and right folding points computed as explained above. The inverse FDNS operation 945 consists in filtering the reconstructed spectrum r[i], 944a using the recursive filter: where a[i] and b[i], 945b are derived from the left and right gains g1[k], g2[k], 952a using the formulas:

[0213] In the above, the variable k is equal to i / (lg / 64) to take into consideration the fact that the LPC spectrums are decimated.

[0214] The reconstructed spectrum rr[], 945a is fed in an inverseMDCT946.Thenon-windowedoutput signal, x[], 946a, is re-scaled by the gain, g, obtained by an inverse quantization of the decoded "global_gain " index: where rms is calculated as:

[0215] The rescaled synthesized time-domain signal 940a is then equal to: 31 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55

[0216] After rescaling, the windowing and overlap add is applied, for example, in the block 978.

[0217] The reconstructed TCX synthesis x(n) 938 is then optionally filtered through the pre-emphasis filter (1‑ 0.68z‑1). The resulting pre-emphasized synthesis is then filtered by the analysis filter (z) in order to obtain the excitation signal. The calculated excitation updates the ACELP adaptive codebook and allows switching from TCX to ACELP in a subsequent frame. The signal is finally reconstructed by de-emphasizing the pre-emphasized synthesis by applying the filter 1 / (1 - 0.68z‑1), Note that the analysis filter coefficients are interpolated in a sub-frame basis.

[0218] Note also that the length of the TCX synthesis is given by the TCX frame length (without the overlap): 256, 512 or 1024 samples for the mod[] of 1,2 or 3 respectively. 8.6 Forward Aliasing-Cancellation (FAC) Tool 8.6.1 Forward Aliasing-Cancellation Tool Description

[0219] The following describes forward-aliasing cancellation (FAC) operations which are performed during transitions betweenACELPand transformcoding (TC) (for example, in the frequency-domainmodeor in theTCX-LPDmode) inorder to get the final synthesis signal. The goal of FAC is to cancel the time-domain aliasing introduced by TC and which cannot be cancelled by the preceding or followingACELP frame.Here the notion of TC includesMDCTover long and short blocks (frequency-domain mode) as well as MDCT-based TCX (TCX-LPD mode).

[0220] Fig. 10 represents the different intermediate signals which are computed in order to obtain the final synthesis signal for theTC frame. In theexample shown, theTC frame (for example, a frame1020encoded in the frequency-domain mode or in the TCX-LPDmode) is both preceded and followed by an ACELP frame (frames 1010 and 1030). In the other cases (an ACELP frame followed by more than one TC frame, or more than one TC frame followed by an ACELP frame) only the required signals are computed.

[0221] Taking reference to Fig. 10 now, an overview over the forward-aliasing-cancellation will be provided, wherein it should be noted that the forward-aliasing-cancellation will be performed by the blocks 960, 961, 962, 963, 964, 965 and 970.

[0222] In the graphical representation of the forward-aliasing-cancellation decodingoperations,which are shown inFig. 10, abscissas 1040a, 1040b, 1040c, 1040d describe a time in terms of audio samples. An ordinate 1042a describes a forward-aliasing-cancellation synthesis signal, for example, in termsof anamplitude.Anordinate1042bdescribes signals representing an encoded audio content, for example, an ACELP synthesis signal and a transform coding frame output signal. An ordinate 1042c describes ACELP contributions to an aliasing-cancellation such as, for example, a windowed ACELP zero-impulse response and a windowed and folded ACELP synthesis. An ordinate 1042d describes a synthesis signal in an original domain.

[0223] As can be seen, a forward-aliasing-cancellation synthesis signal 1050 is provided at a transition from the audio frame1010encoded in theACELPmode to theaudio frame1020encoded in theTCX-LPDmode.The forward-aliasing-to- cancellation synthesis signal 1050 is provided byapplying the synthesis filtering 964andanaliasing-cancellation stimulus signal 963a,which is provided by the inverseDCTof type IV 963. The synthesis filtering 964 is based on the synthesis filter coefficients 965a, which are derived from a set LPC1 of linear-prediction-domain parameters or LPC filter coefficients. As can be seen in Fig. 10, a first portion 1050a of the (first) forward-aliasing-cancellation synthesis signal 1050may be a non- zero-input response provided by the synthesis filtering 964 for a non-zero aliasing-cancellation stimulus signal 963a. However, the forward-aliasing-cancellation synthesis signal 1050 also comprises a zero-input response portion 1050b, which may be provided by the synthesis filtering 964 for a zero-portion of the aliasing-cancellation stimulus signal 963a. Accordingly, the forward-aliasing-cancellation synthesis signal 1050 may comprise a non-zero-input response portion 1050a and a zero-input response portion 1050b. It should be noted that the forward-aliasing-cancellation synthesis signal 1050,may preferably be provided on the basis of the set LPC1 of linear-prediction-domain parameters, which is related to the transitionbetween the frameor sub-frame1010, and the frameor sub-frame1020.Moreover, another forwardaliasing- cancellation synthesis signal 1054 is provided at a transition from the frame or sub-frame 1020 to the frame or sub-frame 1030. The forward-aliasing-cancellation synthesis signal 1054 may be provided by synthesis filtering 964 of an aliasing- cancellation stimulus signal 963a, which is provided by an inverse DCT IV, 963 on the basis of the aliasing-cancellation coefficients. It should be noted that the provision of the forward aliasing-cancellation synthesis signal 1054may be based on a set of linear-prediction-domain parameters LPC2, which are associated to the transition between the frame or sub- frame 1020 and the subsequent frame or sub-frame 1030.

[0224] In addition, additional aliasing-cancellation synthesis signals 1060, 1062 will be provided at a transition from an ACELP frame or sub-frame 1010 to a TXC-LPD frame or sub-frame 1020. For example, a windowed and folded version 973a, 1060 of anACELP synthesis signal 986, 1056may beprovided, for example, by the blocks 971, 972, 973. Further, a windowedACELPzero-input-response976a, 1062will be provided, for example, by theblocks 975, 976.For example, the windowed and folded ACELP synthesis signal 973a, 1060 may be obtained by windowing the ACELP synthesis signal 32 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 986, 1056 and by applying a temporal folding 973 of the result of the windowing, as will be described in more detail below. ThewindowedACELPzero-input-response976a, 1062maybeobtainedbyprovidingazero-input toasynthesis filter 975, which is equal to the synthesis filter 991, which is used to provide the ACELP synthesis signal 986, 1056, wherein an initial state of the synthesis filter 975 is equal to a state of the synthesis filter 981 at the end of the provision of the ACELP synthesis signal 986, 1056 of the frame or sub-frame 1010. Thus, the windowed and folded ACELP synthesis signal 1060 may be equivalent to the forward aliasing-cancellation synthesis signal 973a, and the windowed ACELP zero-input- response 1062 may be equivalent to the forward aliasing-cancellation synthesis signal 976a.

[0225] Finally, the transform coding frame output the signal 1050a, whichmay equal to a windowed version of the time- domain representation 940a, as combined with the forward aliasing-cancellation synthesis signals 1052, 1054, and the additional ACELP contributions 1060, 1062 to the aliasing-cancellation. 8.6.2 Definitions

[0226] In the following, somedefinitionswill be provided. Thebitstreamelement "fac_gain " describes a7-bit gain index. The bitstream element "nq[i] " describes a codebook number. the syntax element "FAC[i] " describes forward aliasing- cancellation data. The variable "fac_length " describes a length of a forward aliasing-cancellation transform,whichmaybe equal to 64 for transitions from and to a window of type "EIGHT_SHORT_SEQUENCES " and which may be 128 otherwise. The variable "use_gain " indicates the use of explicit gain information. 8.6.3 Decoding Process

[0227] In the following, the decoding process will be described. For this purpose, the different steps will briefly be summarized. 1. Decode AVQ parameters (block 960) - The FAC information is encoded using the same algebraic vector quantization (AVQ) tool as for the encoding of LPC filters (see section 8.1). - For i=0 . . . FAC transform length: ∘ A codebook number nq[i] is encoded using a modified unary code ∘ The corresponding FAC data FAC[i] is encoded with 4*nq[i] bits - A vector FAC[i] for i=0,...,fac_length is therefore extracted from the bitstream 2. Apply a gain factor g to the FAC data (block 961) - For transitions with MDCT-based TCX (wLPT), the gain of the corresponding "tex_coding " element is used - For other transitions, a gain information "fac_gain " has been retrieved from the bitstream (encoded using a 7-bits scalar quantizer). The gain g is calculated as g=10fac_gain / 28 using that gain information. 3. In the case of transitions betweenMDCT based TCX and ACELP, a spectrum de-shaping 962 is applied to the first quarter of the FACspectral data 961a. The de-shaping gains are those computed for the correspondingMDCTbased TCX (for usage by the spectrum de-shaping 944) as explained in section 8.5.3 so that the quantization noise of FAC and MDCT-based TCX have the same shape. 4. Compute the inverse DCT-IV of the gain-scaled FAC data (block 963). - The FAC transform length, fac_length, is by default equal to 128 - For transitions with short blocks, this length is reduced to 64. 5. Apply (block 964) the weighted synthesis filter 1 / Ŵ(z) (described, for example, by the synthesis filter coefficients 965a) to get the FAC synthesis signal 964a. The resulting signal is represented on line (a) in Fig. 10. - The weighted synthesis filter is based on the LPC filter which corresponds to the folding point (in Fig. 10 it is identifiedasLPC1 for transitions fromACELP toTCX-LPDandasLPC2 for transitions fromwLPDTC(TCX-LPD) to ACELP or LPC0 for transitions from FD TC (frequency code transform coding) to ACELP) - The same LPC weighting factor is used as for ACELP operations: 33 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 Ŵ(z)=A(z / γ1), where γ1=0.92 - To compute the FAC synthesis signal 964a, the initial memory of the weighted synthesis filter 964 is set to 0 - For transitions from ACELP, the FAC synthesis signal 1050 is further extended by appending the zero-input response (ZIR) 1050b of the weighted synthesis filter (128 samples) 6. In the case of transitions from ACELP, compute the windowed past ACELP synthesis 972a, fold it (for example, to obtain the signal 973a or to the signal 1060) and add to it thewindowedZIR signal (for example, the signal 976a or the signal 1062).TheZIR response is computedusingLPC1.Thewindowapplied to the fac_lengthpastACELPsynthesis samples is: The resulting signal is represented on line (c) in Fig. 10 and denoted as the ACELP contribution (signal contributions 1060, 1062). 7. Add the FAC synthesis 964a, 1050 (and the ACELP contribution 973a, 976a, 1060, 1062 in the case of transitions fromACELP) to theTC frame (which is represented as line (b) in Fig. 10) (or to awindowed version of the time-domain representation 940a) in order to obtain the synthesis signal 998 (which is represented as line (d) in Fig. 10). 8.7 Forward Aliasing-Cancellation (FAC) encoding process

[0228] In the following, some details regarding the encoding of the information required for the forward aliasing- cancellationwill be described. In particular, the computation and encoding of the aliasing-cancellation coefficients 936will be described.

[0229] Fig. 11 shows the processing steps at the encoder when a frame 1120 encoded with Transform Coding (TC) is preceded and followed by a frame 1110, 1130 encoded with ACELP. Here the notion of TC includes MDCTover long and short blocks as in AAC, as well as MDCT-based TCX (TCX-LPD). Figure 11 shows time-domain markers 1140 and frame boundaries 1142, 1144. The vertical dotted lines show the beginning 1142 and end 1144 of the frame 1120 encoded with TC. LPC1 and LPC2 indicate the centre of the analysis window to calculate two LPC filters: LPC1 calculated at the beginning 1142 of the frame 1120 encoded with TC, and LPC2 calculated at the end 1144 of the same frame 1120. The frame 1110 at the left of the "LPC1 " marker is assumed to have been encoded with ACELP. The frame 1130 at the right of the marker "LPC2 " is also assumed to have been encoded with ACELP.

[0230] There are four lines 1150, 1160, 1170, 1180 in Fig. 11. Each line represents a step in the calculation of the FAC target at the encoder. It is to be understood that each line is time aligned with the line above.

[0231] Line 1 (1150) of Fig. 11 represents the original audio signal, segmented in frames 1110, 1120, 1130 as stated above. The middle frame 1120 is assumed to be encoded in the MDCT domain, using FDNS, and will be called the TC frame. The signal in the previous frame1110 is assumed to have been encoded in ACELPmode. This sequence of coding modes (ACELP, thenTC, thenACELP) is chosen so as to illustrate all processing in FACsince FAC is concernedwith both transitions (ACELP to TC and TC to ACELP).

[0232] Line 2 (1160) of Fig. 11 corresponds to the decoded (synthesis) signals in each frame (whichmay be determined by the encoder by using knowledge of the decoding algorithm). The upper curve 1162, which extends from beginning to end of the TC frame, shows the windowing effect (flat in the middle but not at the beginning and end). The folding effect is shown by the lower curves 1164, 1166 at the beginning and end of the segment (with "‑ " sign at the beginning of the segment and "+ " sign at the end of the segment). FAC can then be used to correct these effects.

[0233] Line 3 (1170) of Fig. 11 represents the ACELP contribution, used at the beginning of the TC frame to reduce the coding burden of FAC. This ACELP contribution is formed of two parts: 1) the windowed, folded ACELP synthesis 877f, 1170 from the end of the previous frame, and 2) the windowed zero-input response 877j, 1172 of the LPC1 filter.

[0234] It should be noted here that thewindowed and folded ACELP synthesis 1110may be equivalent to thewindowed 34 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 and folded ACELP synthesis 1060, and that the windowed zero-input-response 1172may be equivalent to the windowed ACELP zero-input-response 1062. In other words, the audio signal encoder may estimate (or calculate) the synthesis result 1162, 1164, 1166, 1170, 1172, which will be obtained at the side of an audio signal decoder (blocks 869a and 877).

[0235] The ACELP error which is shown in line 4 (1180) is then obtained by simply subtracting Line 2 (1160) and Line 3 (1170) from Line 1 (1150) (block 870). An approximate view of the expected envelope of the error signal 871, 1182 in the time domain is shown on Line 4 (1180) in Fig. 11. The error in the ACELP frame (1120) is expected to be approximately flat in amplitude in the time domain. Then the error in the TC frame (betweenmarkers LPC1 and LPC2) is expected to exhibit the general shape (time domain envelope) as shown in this segment 1182 of Line 4 (1180) in Fig. 11.

[0236] To efficiently compensate the windowing and time-domain aliasing effects at the beginning and end of the TC frame on Line 4 of Fig. 10, and assuming that the TC frame uses FDNS, FAC is applied according to Fig. 11. It should be noted that Fig. 11 describes this processing for both the left part (transition fromACELP to TC) and the right part (transition from TC to ACELP) of the TC frame.

[0237] To summarize, the transform coding frame error 871, 1182, which is represented by the encoded aliasing- cancellation coefficients 856, 936 is obtained by subtracting both, the transform coding frame output 1162, 1164, 1166 (described, for example, by signal 869b), and the ACELP contribution 1170, 1172 (described, for example, by signal 872) from the signal 1152 in the original domain (i.e. in the time-domain). Accordingly, the transform coding frame error signal 1182 is obtained.

[0238] In the following, the encoding of the transform coding frame error 871, 1182 will be described.

[0239] First, a weighting filter 874, 1210, W1(z) is computed from the LPC1 filter. The error signal 871, 1182 at the beginning of the TC frame 1120 on Line 4 (1180) of Fig. 11 (which is also called the FAC target in Figs. 11 and 12) is then filtered throughW1(z),which has as initial state, or filter memory, the ACELP error 871, 1182 in the ACELP frame 1120 on Line 4 of Fig. 11. The output of filter 874, 1210W1(z) at the top of Fig. 12 then forms the input of a DCT-IV transform 875, 1220. The transform coefficients 875a, 1222 from the DCT-IV 875, 1220 are then quantized and encoded using the AVQ tool 876 (represented by Q, 1230). This AVQ tool is the same that is used for quantizing the LPC coefficients. These encoded coefficients are transmitted to the decoder. The output of AVQ 1230 is then the input of an inverse DCT-IV 963, 1240 to forma time-domain signal 963a, 1242. This time-domain signal is then filtered through the inverse filter 964, 1250, 1 / W1(z) which has zero-memory (zero initial state). Filtering through 1 / W1(z) is extended past the length of the FAC target using zero-input for the samples that extend after the FAC target. The output 964a, 1252 of filter 1250, 1 / W1(z) is the FAC synthesis, which is the correction signal (for example, signal 964a) that may now be applied at the beginning of the TC frame to compensate for the windowing and Time-Domain Aliasing effects.

[0240] Now, turning to the processing for thewindowing and time-domain aliasing correction at the end of the TC frame, weconsider the bottompart of Fig. 12. Theerror signal 871, 1182bat theendof theTC frame1120onLine4of Fig. 11 (FAC target) is filtered throughfilter 874, 1210;W2(z),whichhasas initial state, or filtermemory, theerror in theTC frame1120on Line 4 of Fig. 11. Then all further processing steps are the same as for the upper part of Fig. 12 which dealt with the processingof theFAC target at thebeginningof theTC frame,with theexceptionof theZIRextension in theFACsynthesis.

[0241] Note that the processing in Fig. 12 is performed completely (from left to right) when applied at the encoder (to obtain the local FAC synthesis), whereas at the decoder side the processing in Fig. 12 is only applied starting from the received decoded DCT-IV coefficients. 9. Bitstream

[0242] In the following, some details regarding the bitstream will be described in order to facilitate the understanding of the present invention. It should be noted here that a significant amount of configuration informationmay be included in the bitstream.

[0243] However, an audio content of a frame encoded on the frequency-domain mode is mainly represented by a bitstreamelement named "fd_channel_stream() ". This bitstreamelement "fd_channel_stream() " comprisesaglobal gain information "global_gain ", encoded scale factor data "scale_factor_data() ", and arithmetically encoded spectral data "ac_spectral_data ". In addition, the bitstream element "fd_channel _stream() " selectively comprises forward aliasing- cancellation data including a gain information (also designated as "fac_data(1) "), if (and only if) a previous frame (also designated as "superframe " in some embodiments) has been encoded in the linear-prediction-domainmode and the last sub-frame of the previous frame was encoded in the ACELP mode. In other words, a forward-aliasing-cancellation data including again information is selectively provided for a frequency-domainmodeaudio frame, if the previous frameor sub- frame was encoded in the ACELP mode. This is advantageous, as an aliasing-cancellation can be effected by a mere overlap-and-add functionality betweenaprevious audio frameor audio sub-frameencoded in theTCX-LPDmodeand the current audio frame encoded in the frequency-domain mode, as has been explained above.

[0244] For details, reference is made to Fig. 14, which shows a syntax representation of the bitstream element "fd_channel_stream() " which comprises the global gain information "global_gain ", the scale factor data "scale_fac- tor_data() ", the arithmetically coded spectral data "ac_spectral_data() ". The variable "core_mode_last " describes a last 35 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 core mode and takes the value of zero for a scale factor based frequency-domain coding and takes the value of one for a coding based on linear-prediction-domain parameters (TCX-LPD or ACELP). The variable "last lpd_mode " describes an LPDmode of a last frame or sub-frame and takes the value of zero for a frame or sub-frame encoded in the ACELPmode.

[0245] Taking reference now to Fig. 15, the syntax will be described for a bitstream element "lpd_channel_stream() ", which encodes the information of an audio frame (also designated as "superframe ") encoded in the linear-prediction- domainmode. The audio frame ("superframe ") encoded in the linear-prediction-domainmodemay comprise a plurality of sub-frames (sometimes also designated as "frames ", for example, in combination with the terminology "superframe "). Thesub-frames (or "frames ")maybeof different types, such that someof thesub-framesmaybeencoded in theTCX-LPD mode, while other of the sub-frames may be encoded in the ACELP mode.

[0246] The bitstream variable "acelp _core_mode " describes the bit allocation scheme in case an ACELP is used. The bitstream element "lpd_mode " has been explained above. The variable "first_tcx_flag " is set to true at the beginning of each frame encoded in the LPDmode. The variable "first_lpd_flag " is a flag which indicates whether the current frame or superframe is the first of a sequence of frames or superframes which are encoded in the linear-prediction coding domain. Thevariable "last_lpd " is updated todescribe themode (ACELP;TCX256;TCX512;TCX1024) inwhich the last sub-frame (or frame) was encoded. As can be seen at reference numeral 1510, forward-aliasing-cancellation data without a gain information ("fac_data_(0) ") are included for a sub-frame which is encoded in the TCX-LPD mode (mod[k]>0] if the last sub-framewas encoded in the ACELPmode (last_lpd_mode==0) and for a sub-frame encoded in the ACELPmode (mod [k]==0) if the previous sub-frame was encoded in the TCX-LPD mode (last_lpd_mode>0).

[0247] If, in contrast, the previous framewas encoded in the frequency-domain mode (core_mode_last=0) and the first sub-frameof thecurrent frame isencoded in theACELPmode (mod[0]==0), forward-aliasing-cancellationdata includinga gain information ("fac_data(1) ") are contained in the bitstream element "lpd_channel_stream ".

[0248] Tosummarize, forward-aliasing-cancellationdata includingadedicated forward-aliasing-cancellationgain value are included in thebitstream, if there isadirect transitionbetweena frameencoded in the frequency-domainanda frameor sub-frame encoded in the ACELPmode. In contrast, if there is a transition between a frame or sub-frame encoded in the TCX-LPD mode and a frame or sub-frame encoded in the ACELP mode, a forward-aliasing-cancellation information without a dedicated forward-aliasing-cancellation gain value is included in the bitstream.

[0249] Taking reference now to Fig. 16, the syntax of the forward-aliasing-cancellation data, which is described by the bitstream element "fac_data() " will be described. The parameter "useGain " indicates whether there is a dedicated forward-aliasing-cancellation gain value bitstream element "fac_gain ", as can be seen at reference numeral 1610. In addition, the bitstream element "fac_data " comprises a plurality of codebook number bitstream elements "nq[i] " and a number of "fac_data " bitstream elements "fac[i] ".

[0250] The decoding of said codebook number and said forward-aliasing-cancellation data has been described above. 10. Implementation Alternatives

[0251] Although some aspects have been described in the context of an apparatus, it is clear that these aspects also represent a description of the correspondingmethod,where ablock or device corresponds to amethod step or a feature of a method step. Analogously, aspects described in the context of a method step also represent a description of a corresponding block or item or feature of a corresponding apparatus. Some or all of the method steps may be executed by (or using) a hardware apparatus, like for example, amicroprocessor, a programmable computer or an electronic circuit. In some embodiments, some one or more of the most important method steps may be executed by such an apparatus.

[0252] The inventive encoded audio signal can be stored on a digital storage medium or can be transmitted on a transmission medium such as a wireless transmission medium or a wired transmission medium such as the Internet.

[0253] Depending on certain implementation requirements, embodiments of the invention can be implemented in hardwareor in software.The implementation canbeperformedusingadigital storagemedium, for exampleafloppydisk, a DVD, a Blue-Ray, a CD, a ROM, a PROM, an EPROM, an EEPROM or a FLASHmemory, having electronically readable control signals stored thereon, which cooperate (or are capable of cooperating) with a programmable computer system such that the respective method is performed. Therefore, the digital storage medium may be computer readable.

[0254] Some embodiments according to the invention comprise a data carrier having electronically readable control signals, which are capable of cooperatingwith a programmable computer system, such that one of themethods described herein is performed.

[0255] Generally, embodiments of the present invention can be implemented as a computer program product with a program code, the program code being operative for performing one of themethods when the computer program product runs on a computer. The program code may for example be stored on a machine readable carrier.

[0256] Other embodiments comprise the computer program for performing oneof themethods described herein, stored on a machine readable carrier.

[0257] In otherwords, anembodiment of the inventivemethod is, therefore, a computer programhavingaprogramcode for performing one of the methods described herein, when the computer program runs on a computer. 36 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55

[0258] A further embodiment of the inventive methods is, therefore, a data carrier (or a digital storage medium, or a computer-readable medium) comprising, recorded thereon, the computer program for performing one of the methods described herein. The data carrier, the digital storage medium or the recorded medium are typically tangible and / or non- transitionary.

[0259] A further embodiment of the inventivemethod is, therefore, a data stream or a sequence of signals representing the computer program for performing one of the methods described herein. The data stream or the sequence of signals may for example be configured to be transferred via a data communication connection, for example via the Internet.

[0260] A further embodiment comprises a processingmeans, for example a computer, or a programmable logic device, configured to or adapted to perform one of the methods described herein.

[0261] A further embodiment comprises a computer having installed thereon the computer program for performing one of the methods described herein.

[0262] A further embodiment according to the invention comprises an apparatus or a system configured to transfer (for example, electronically or optically) a computer program for performing one of themethods described herein to a receiver. The receivermay, for example, beacomputer, amobile device, amemorydevice or the like. Theapparatusor systemmay, for example, comprise a file server for transferring the computer program to the receiver.

[0263] In someembodiments, a programmable logic device (for example a field programmable gate array)maybe used to perform some or all of the functionalities of themethods described herein. In some embodiments, a field programmable gate array may cooperate with a microprocessor in order to perform one of the methods described herein. Generally, the methods are preferably performed by any hardware apparatus.

[0264] The above described embodiments are merely illustrative for the principles of the present invention. It is understood that modifications and variations of the arrangements and the details described herein will be apparent to others skilled in the art. It is the intent, therefore, to be limited only by the scope of the impending patent claims and not by the specific details presented by way of description and explanation of the embodiments herein. 11. Conclusion

[0265] In the following, the present proposal for the unification of unified-speech-and-audio-coding (USAC) windowing and frame transitions will be summarized.

[0266] Firstly, an introduction will be given and some background information described. A current design (also designated as a reference design) of the USAC reference model consists of (or comprises) three different coding modules. For each given audio signal section (for example, a frame or sub-frame) one codingmodule (or codingmode) is chosen to encode / decode that section resulting in different coding modes. As these modules alternate in activity, special attention needs to be paid to the transitions from onemode to the other. In the past, various contributions have proposed modifications addressing these transitions between coding modes.

[0267] Embodiments according to the present invention create an envisioned overall windowing and transition scheme. The progress that has been achieved on the way towards completion of this scheme will be described, displaying very promising evidence for quality and systematic structural improvements.

[0268] Thepresent document summarizes theproposedchanges to the referencedesign (which is alsodesignatedasa working draft 4 design) in order to create amore flexible coding structure for USAC, to reduce overcoding and reduce the complexity of the transform coded sections of the codec.

[0269] In order toarriveat awindowingschemewhichavoids costly non-critical sampling (overcoding), twocomponents are introduced, which may be considered as being essential in some embodiments: 1) the forward-aliasing-cancellation (FAC) window; and 2) frequency-domain noise-shaping (FDNS) for the transform coding branch in the LPDcore codec (TCX, also known as TCX-LPD or wLPT).

[0270] The combination of both technologies makes it possible to employ a windowing scheme which allows highly flexible switching of transform length at a minimum bit demand.

[0271] In the following the challenges of reference systems will be described to facilitate the understanding of the advantagesprovided by theembodiments according to the invention. A reference concept according to theworking draft 4 of the USAC draft standard consists of a switched core codec working in conjunction with a pre‑ / post-processing stage consisting of (or comprising) MPEG surround and an enhanced SBR module. The switched core features a frequency- domain (FD) codec and a linear-predictive-domain (LPD) codec. The latter employs an ACELP module and a transform coder working in the weighted domain ("weighted Linear Prediction Transform" (wLPT), also known as transform-coded- excitation, (TCX)). It has been found that due to the fundamentally different coding principles, the transitions between the modes are especially challenging to handle. It has been found that care has to be taken that the modes intermingle 37 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 efficiently.

[0272] In the following, the challenges which arise at the transitions from time-domain to frequency-domain (ACELP↔wLPT, ACELP↔FD) will be described. It has been found that transitions from time-domain coding to trans- form-domain coding are tricky, in particular, as the transformcoder is based on the transformdomain aliasing-cancellation (TDAC) property of neighboring blocks in the MDCT. It has been found that a frequency domain coded block cannot be decoded in its entirety without additional information from its adjacent overlapping blocks.

[0273] In the following, thechallengeswhichappearat transitions from thesignal domain to the linear-predictive-domain (FD↔ACELP,FD↔wLPT)will bedescribed. It hasbeen found that the transitions toand from the linear-predictive-domain imply a transition of different quantization noise-shaping paradigms. It has been found that the paradigmsutilize a different way of conveying and applying psychoacoustically motivated noise-shaping information, which can cause discontinuities in the perceived quality at places where the coding mode changes.

[0274] In the following, details regarding a frame transitionmatrix of a reference concept according to theworking draft 4 of theUSACdraft standardwill be described.Due to the hybrid nature of the referenceUSAC referencemodel, there are a multitudeof conceivablewindow transitions. The3-by‑3 table in Fig. 4 displaysanoverviewof these transitions as theyare currently implemented according to the concept of the working draft 4 of the USAC draft standard.

[0275] Thecontributions listedaboveeachaddressoneormoreof the transitiondisplayed in the tableofFig. 4. It isworth noting that the non-homogenous transitions (the ones not on the main diagonal) each apply different specific processing steps,whichare the result of a compromisebetween trying toachievecritical sampling, avoidingblockingartefacts, finding a common windowing scheme, and allowing for an encoder closed-loopmode decision. In some cases, this compromise comes at the cost of discarding coded and transmitted samples.

[0276] In following, some proposed system changes will be described. In other words, improvements of the reference concept according to the USAC working draft 4 will be described. In order to tackle the listed difficulties at the window transitions, embodiments according to the invention introduce twomodifications to theexisting system,whencompared to the concepts according to the reference system according to the working draft 4 of the USAC draft standard. The first modification aims at universally improving the transition from time-domain to frequency-domain by adopting a supple- mental forward-aliasing-cancellation window. The second modification assimilates the processing of signal- andlinear- predictiondomainsby introducinga transmutationstep for theLPCcoefficients,which thencanbeapplied in the frequency domain.

[0277] In the following, the concept of frequency-domain noise shaping (FDNS) will be described, which allows for the application of the LPC in the frequency-domain. The goal of this tool (FDNS) is to allow TDAC processing of the MDCT coders which work in different domains. While the MDCT of the frequency-domain part of the USAC acts in the signal domain, the wLPT (or TCX) of the reference concept operates in the weighted filtered domain. By replacing the weighted LPCsynthesis filter,which is used in the referenceconcept, by anequivalent processing step in the frequency-domain, the MDCT of both transform coders operate in the same domain and TDAC can be accomplished without introducing discontinuities in quantization noise-shaping.

[0278] In other words, the weighted LPC synthesis filter 330g is replaced by the scaling / frequency-domain noise- shaping 380e in combination with the LPC to frequency-domain conversion 380i. Accordingly, the MDCT 320g of the frequency-domain path and the MDCT 380h of the TCX-LPD branch operate in the same domain, such that transform domain aliasing-cancellation (TDAC) is achieved.

[0279] In the following, some details regarding the forward-aliasing-cancellation window (FAC window) will be de- scribed. The forward-aliasing-cancellation (FAC)window has already been introduced and described. This supplemental window compensates the missing TDAC information which - in a continuously running transform code - is usually contributed by the following or preceding window. Since the ACELP time-domain coder exhibits no overlap to adjacent frames, the FAC can compensate for the lack of this missing overlap.

[0280] It has been found that by applying theLPCfilter in the frequency-domain, theLPDcodingpath looses someof the smoothing impact of the interpolated LPC filtering between ACELP and wLPT (TCX-LPD) coded segments. However, it has been found that, since the FAC was designed to enable a favorable transition at exactly this place, it can also compensate for this effect.

[0281] As a consequence of introducing the FAC window and FDNS, all conceivable transitions can be accomplished without any inherent overcoding.

[0282] In the following, some details regarding the windowing scheme will be described.

[0283] HowtheFACwindowcan fuse the transitionsbetweenACELPandwLPThasalreadybeendescribed.For further details, reference is made to the following document: ISO / IEC JTC1 / SC29 / WG11, MPEG2009 / M16688, June-July 2009, London, United Kingdom, "Alternatives for windowing in USAC ".

[0284] Since the FDNS shifts the wLPT into the signal domain, the FAC window can now be applied to both, the transitions from / to the ACELP to / fromwLPTand also from / to ACELP to / from FDmode in exactly the samemanner (or, at least, in a similar manner).

[0285] Similarly, theTDACbased transformcoder transitionswhichwerepreviouslypossibleexclusively in-betweenFD 38 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 windowsor in-betweenwLPTwindows (i.e. from / toFD to / fromFD; or from / towLPT to / fromwLPT) cannowalsobeapplied when transgressing from the frequency-domain to wLPT, or vice-versa. Thus, both technologies combined allow for the shifting of the ACELP framing grid 64 samples to the right (towards "later " in the time axis). By doing so, the 64 sample overlap-addononeendand theextra-long frequency-domain transformwindowat the other endare no longer required. In both cases, a 64 samples overcoding can be avoided in embodiments according to the invention when compared to the reference concepts. Most importantly, all other transitions stay as they are and no further modifications are necessary.

[0286] In the following the new frame transitionmatrix will briefly be discussed. An example for a new transitionmatrix is provided in Fig. 5. The transitions on themain diagonal stay as theywere in working draft 4 of theUSACdraft standard. All other transitions canbedealtwithby theFACwindowor straightforwardTDAC in the signal domain. In someembodiments only two overlap lengths between adjacent transform domain windows are needed for the above scheme, namely 1024 samples and 128 samples, though other overlap lengths are also conceivable. 12. Subjective Evaluation

[0287] It shouldbenoted that two listening testshavebeenconducted to show that at thecurrent stateof implementation the proposed new technology does not compromise the quality. Eventually, embodiments according to the invention are expected toprovidean increase inquality due to thebit savingsat theplaceswheresampleswerepreviouslydiscarded.As another side effect, the classifier control at the encoder can bemuchmore flexible since themode transitions are no longer afflicted with non-critical sampling. 13. Further Remarks

[0288] To summarize the above, the present description describes an envisioned windowing and transition scheme for theUSACwhichhas several virtues, compared to theexisting scheme, used inworking draft 4 of theUSACdraft standard. Theproposedwindowingand transitionschememaintainscritical sampling inall transform-coded frames,avoids theneed for non-power-of-two transforms and properly aligns all transform-coded frames. The proposal is based on two new tools. The first tool, forward-aliasing-cancellation (FAC), is described in the reference [M16688]. The second tool, frequency- domain noise-shaping (FDNS), allows processing frequency-domain frames and wLPT frames in the same domain without introducing discontinuities in the quantization noise shaping. Thus, all mode transitions in USAC can be handled with these two basic tools, allowing harmonized windowing for all transform-coded modes. Subjective tests results were also provided in the present description, showing that the proposed tools provide equivalent or better quality compared to the reference concept according to the working draft 4 of the USAC draft standard. References

[0289] [M16688] ISO / IEC JTC1 / SC29 / WG11, MPEG2009 / M16688, June-July 2009, London, United Kingdom, "Alter- natives for windowing in USAC "

[0290] In the following, additional embodiments and aspects of the invention will be described which can be used individually or in combination with any of the features and functionalities and details described herein.

[0291] According to a first aspect, an audio signal decoder 200; 360; 900 for providing a decoded representation 212; 399; 998 of an audio content on the basis of an encoded representation 210; 361; 901 of the audio content comprises: a transformdomain path 230, 240, 242, 250, 260; 270, 280; 380; 930 configured toobtain a timedomain representation 212; 386; 938 of a portion of the audio content encoded in a transform domainmode on the basis of a first set 220; 382; 944a of spectral coefficients, a representation 224; 936 of an aliasing-cancellation stimulus signal and a plurality of linear- prediction-domain parameters 222; 384;950a, wherein the transform domain path comprises a spectrum processor 230; 380e; 945 configured to apply a spectral shaping to the first set 944a of spectral coefficients in dependence on at least a subset of the linear-prediction-domain parameters, to obtain a spectrally-shaped version 232; 380g; 945a of the first set of spectral coefficients, wherein the transform domain path comprises a first frequency-domain-to-time-domain converter 240; 380h; 946 configured to obtain a time-domain representation of the audio content on the basis of the spectrally- shaped version of the first set of spectral coefficients; wherein the transform domain path comprises an aliasing- cancellation stimulus filter 250; 964 configured to filter an aliasing-cancellation stimulus signal 224; 963a in dependence on at least a subset of the linear-prediction-domain parameters 222; 384; 934, to derive an aliasing-cancellation synthesis signal 252; 964a from the aliasing-cancellation stimulus signal; and wherein the transform domain path also comprises a combiner 260; 978configured to combine the time-domain representation 242; 940aof theaudio contentwith the aliasing- cancellation synthesis signal 252; 964, or a post-processed version thereof, to obtain an aliasing-reduced time-domain signal.

[0292] According to a second aspect when referring back to the first aspect, the audio signal decoder is a multi-mode audio signal decoder configured to switch between a plurality of codingmodes, andwherein the transformdomain branch 39 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 230; 240, 250, 260, 270, 280; 380; 930 is configured to selectively obtain the aliasing-cancellation synthesis signal 252; 964a for a portion 1020 of the audio content following a previous portion 1010 of the audio content which does not allow for an aliasing-cancelling overlap-and-add operation or for a portion of the audio content followed by a subsequent portion 1030 of the audio content which does not allow for an aliasing-cancelling overlap-and-add operation.

[0293] According to a third aspectwhen referring back to anyoneof the first or secondaspects, the audio signal decoder is configured to switch between a transform-coded-excitation-linear-prediction-domain mode, which uses a transform- coded-excitation information 932 and a linear-prediction-domain parameter information 934, and a frequency-domain mode,which uses a spectral coefficient information 912 and a scale factor information 914; wherein the transform-domain path 930 is configured to obtain the first set 944a of spectral coefficients on the basis of the transform-coded-excitation information 932, and to obtain the linear-prediction-domain-parameters 950a on the basis of the linear-prediction-domain parameter information 934; wherein the audio signal decoder comprises a frequency-domain path 910 configured to obtain a time-domain representation 918 of the audio content encoded on the frequency-domain mode on the basis of a frequency-domain mode set of spectral coefficients 921a described by the spectral coefficient information 912 and in dependence on a set 922a of scale factors 922 described by the scale factor information 914, wherein the frequency- domain path 910 comprises a spectrum processor 923 configured to apply a spectral shaping to the frequency-domain mode set of spectral coefficients 921a, or to a pre-processed version thereof, in dependence on the set 922a of scale factors, to obtain a spectrally-shaped frequency-domainmode set 923a of spectral coefficients, and when the frequency- domain path 910 comprises a frequency-domain-to-time-domain converter 924a configured to obtain a time domain representation 924 of the audio content on the basis of the spectrally shaped frequency-domain mode set of spectral coefficients 923a; wherein the audio signal decoder is configured such that time-domain representations of two sub- sequent portions of the audio content, one of which two subsequent portions of the audio content is encoded in the transform-coded-excitation-linear-prediction-domain mode and one of which two subsequent portions of the audio content is encoded in the frequency-domainmode, comprise a temporal overlap to cancel a time-domain-aliasing caused by the frequency-domain-to-time-domain conversion.

[0294] According toa fourth aspectwhen referring back toanyoneof the first to third aspects, theaudio signal decoder is configured to switch between a transform-coded-excitation-linear-prediction-domain mode, which uses a transform- coded-excitation information 932 and a linear-prediction-domain parameter information 934, and an algebraic code- excited-linear-prediction ACELPmode, which uses an algebraic-code excitation information 982 and a linear-prediction- domain parameter information 984; wherein the transform-domain path 930 is configured to obtain the first set 944a of spectral coefficients on the basis of the transform-coded-excitation information 932, and to obtain the linear-prediction- domainparameters950aon thebasisof the linear-prediction-domainparameter information934;wherein theaudio signal decoder comprises an algebraic-code-excitation-linear-prediction path 980 configured to obtain a time domain repre- sentation 986 of the audio content encoded in the ACELPmode on the basis of the algebraic-code-excitation information 982 and the linear-prediction-domain parameter information 984; wherein the ACELP path 980 comprises an ACELP excitation processor 988, 989 configured to provide a time-domain excitation signal 989a on the basis of the algebraic- code excitation information 982 and using a synthesis filter 991 configured to perform a time-domain filtering of the time- domain excitation signal to provide a reconstructed signal 991a on the basis of the time-domain excitation signal 989aand in dependence on linear-prediction-domain filter coefficients 990a obtained on the basis of the linear-prediction-domain parameter information 984; wherein the transform domain path 930 is configured to selectively provide the aliasing- cancellation synthesis signal 964 for a portion of the audio content encoded in the transform-coded-excitation-linear- prediction-domainmode followingaportionof theaudio contentencoded in theACELPmode,and foraportionof theaudio content encoded in the transform-coded-excitation-linear-prediction-domain mode preceding a portion of the audio content encoded in the ACELP mode.

[0295] According to a fifth aspect when referring back to the fourth aspect, the aliasing-cancellation stimulus filter 964 is configured to filter the aliasing-cancellation stimulus signal 963a in dependence on the linear-prediction-domain filter parameters 950a; LPC1 which correspond to a left-sided aliasing folding point of the first frequency-domain-to-time- domain converter 946 for a portion of the audio content encoded in the transform-coded-excitation-linear-prediction- domainmode following aportion of the audio content encodedon theACELPmode, andwherein thealiasing-cancellation stimulus filter 964 is configured to filter the aliasing-cancellation stimulus signals 963a in dependence on the linear- prediction-domain filter parameters 950a; LPC2 which correspond to a right-sided aliasing folding point of the first frequency-domain-to-time-domain converter 946 for a portion of the audio content encoded in the transform-coded- excitation-linear-prediction-domain mode preceding a portion of the audio content encoded on the ACELP mode.

[0296] According toasixthaspectwhen referringback toanyoneof the fourth tofifthaspects, theaudio signal decoder is configured to initialize memory values of the aliasing-cancellation stimulus filter 964 to zero for providing the aliasing- cancellation synthesis signal, to feedM samples of the aliasing-cancellation stimulus signal into the aliasing-cancellation stimulus filter 964, to obtain corresponding non-zero-input response samples of the aliasing-cancellation synthesis signal 964a, and to further obtain a plurality of zero-input response samples of the aliasing-cancellation synthesis signal; and wherein the combiner is configured to combine the time-domain representation 940a of the audio content with the non- 40 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 zero-input response samples and the subsequent zero-input response samples to obtain an aliasing-reduced time- domain signal at a transition fromaportionof theaudio content encoded in theACELPmode toa subsequent portionof the audio content encoded in the transform-coded-excitation-linear-prediction-domain mode.

[0297] According to a seventh aspect when referring back to any one of the fourth to sixth aspects, the audio signal decoder is configured to combine a windowed and folded version 973a; 1060 of at least a portion of the time-domain representation obtained using theACELPmodewith a time-domain representation 940; 1050a of a subsequent portion of the audio content obtained using the transform-coded-excitation-linear-prediction-domain mode, to at least partially cancel an aliasing.

[0298] According to an eighth aspect when referring back to any one of the fourth to seventh aspects, the audio signal decoder is configured to combine a windowed version 976a; 1062 of a zero-input response of the synthesis filter of the ACELPbranchwith a time-domain representation 940a; 1058 of a subsequent portion of the audio content obtained using the transform-coded-excitation-linear-prediction-domain mode, to at least partially cancel an aliasing.

[0299] According to a ninth aspect when referring back to any one of the fourth to eighth aspects, the audio signal decoder is configured to switch between a transform-coded-excitation-linear-prediction-domain mode, in which a lapped frequency-domain-to-time-domain transform is used, a frequency-domainmode, inwhich a lapped frequency-domain-to- time-domain transform is used, and an algebraic-code-excitation-linear-prediction mode, wherein the audio signal decoder is configured to at least partially cancel an aliasing at a transition between a portion of the audio content encoded in the transform-coded-excitation-linear-prediction-domain mode and a portion of the audio content encoded in the frequency-domain mode by performing an overlap-and-add operation between time-domain samples of subsequent overlapping portions of the audio content; andwherein the audio signal decoder is configured to at least partially cancel an aliasingat a transition betweenaportionof the audio content encoded in the transform-coded-excitation-linear-prediction- domain mode and a portion of the audio content encoded in the algebraic-code-excited-linear-prediction-domain mode using the aliasing-cancellation synthesis signal 964a.

[0300] According to a tenth aspectwhen referring back to anyoneof the first to ninth aspects, theaudio signal decoder is configured toapplyacommongain valueg foragainscaling947of a time-domain representation946aprovidedby thefirst frequency-domain-to-time-domain converter 946 of the transform domain path 930 and for a gain scaling 961 of the aliasing-cancellation stimulus signal 963a or the aliasing-cancellation synthesis signal 964a.

[0301] According to an eleventh aspect when referring back to any one of the first to tenth aspects, the audio signal decoder is configured to apply, in addition to the spectral shapingperformed independence onat least the subset of linear- prediction-domain parameters, a spectrum deshaping 944 to at least a subset of the first set of spectral coefficients, and wherein theaudio signal decoder is configured toapply the spectrumdeshaping962 toat least a subset of a set of aliasing- cancellation spectral coefficients from which the aliasing-cancellation stimulus signal 963a is derived.

[0302] According to a twelfth aspect when referring back to any one of the first to eleventh aspects, the audio signal decoder comprises a second frequency-domain-to-time-domain converter 963 configured to obtain a time-domain representation of the aliasing-cancellation stimulus signal 963a in dependence on a set of spectral coefficients 960a representing the aliasing-cancellation stimulus signal, wherein the first frequency-domain-to-time-domain converter is configured to perform a lapped transform, which comprises a time-domain aliasing, and wherein the second frequency- domain-to-time-domain converter is configured to perform a non-lapped transform.

[0303] According to a thirteenth aspect when referring back to any one of the first to twelfth aspects, the audio signal decoder is configured toapply thespectral shaping to thefirst setof spectral coefficients independenceon thesame linear- prediction-domain parameters, which are used for adjusting the filtering of the aliasing-cancellation stimulus signal.

[0304] According to a fourteenth aspect an audio signal encoder 100; 800 for providing an encoded representation 112; 812of anaudio content comprisinga first set 112a; 852of spectral coefficients, a representationof analiasing-cancellation stimulus signal 112c; 856 and a plurality of linear-prediction-domain parameters 112b; 854 on the basis of an input representation 110; 810 of the audio content, comprises: a time-domain-to-frequency-domain converter 120; 860 configured to process the input representation of the audio content, to obtain a frequency-domain representation 112; 861 of the audio content; a spectral processor 130; 866 configured to apply a spectral shaping to the frequency-domain representation of the audio content, or to a pre-processed version thereof, in dependence on a set of linear-prediction- domain parameters 140; 863 for a portion of the audio content to be encoded in the linear-prediction-domain, to obtain a spectrally-shaped frequency-domain representation 132; 867 of the audio content; and an aliasing-cancellation informa- tion provider 150, 870, 874, 875, 876 configured to provide a representation 112c; 856 of an aliasing-cancellation stimulus signal, such that a filtering of the aliasing-cancellation stimulus signal in dependence on at least a subset of the linear- prediction-domain parameters results in an aliasing-cancellation synthesis signal for cancelling aliasing artifacts in an audio signal decoder.

[0305] According to a fifteenth aspect, amethod for providing adecoded representation of anaudio content on thebasis of an encoded representation of the audio content, comprises the steps of: obtaining a time-domain representation of a portion of the audio content encoded in a transform domain mode on the basis of a first set of spectral coefficients, a representation of an aliasing-cancellation stimulus signal and the plurality of linear-prediction-domain parameters, 41 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 whereinaspectral shaping is supplied to thefirst set of spectral coefficients independenceonat least asubset of the linear- prediction-domain parameters, to obtain a spectrally shaped version of the first set of spectral coefficients, and wherein a frequency-domain-to-time-domain conversion is applied to obtain a time-domain representation of the audio content on the basis of the spectrally-shaped version of the first set of spectral coefficients, and wherein the aliasing-cancellation stimulus signal is filtered in dependence of at least a subset of the linear-prediction-domain parameters, to derive an aliasing-cancellation synthesis signal from the aliasing-cancellation stimulus signal, and wherein the time-domain representation of the audio content is combined with the aliasing-cancellation synthesis signal, or a post-processed version thereof, to obtain an aliasing-reduced-time-domain signal.

[0306] According to a sixteenth aspect, a method for providing an encoded representation of an audio content comprising a first set of spectral coefficients, a representation of an aliasing-cancellation stimulus signal, and a plurality of linear-prediction-domain parameters on the basis of an input representation of the audio content, comprises the steps of: performing a time-domain-to-frequency-domain conversion to process the input representation of the audio content, to obtain a frequency-domain representation of the audio content; applying a spectral shaping to the frequency-domain representation of the audio content, or to a pre-processed version thereof, in dependence of a set of linear-prediction- domain parameters for a portion of the audio content to be encoded in the linear-prediction-domain, to obtain a spectrally- shaped frequency-domain representation of the audio content; and providing a representation of an aliasing-cancellation stimulus signal, such that a filtering of the aliasing-cancellation stimulus signal in dependence on at least a subset of the linear-prediction-domain parameters results in an aliasing-cancellation synthesis signal for cancelling aliasing artifacts in an audio signal decoder.

[0307] A seventeenth aspect relates to a computer program for performing the method according to aspects 15 or 16, when the computer program runs on a computer. Claims 1. Amulti-modeaudio signal decoder (200; 360; 900) for providing adecoded representation (212; 399; 998) of an audio content on the basis of an encoded representation (210; 361; 901) of the audio content, wherein the multi-mode audio signal decoder is configured to switch between three modes, a frequency-domain mode, which uses a spectral coefficient information and a scale factor information , a transform-coded-excitation linear-prediction-domain mode, which uses a transform-coded-excitation information and a linear-prediction- domain parameter information, and an algebraic-code-excited-linear-predictionmode, which uses an algebraic- code-excitation-information and a linear-prediction-domain-parameter information, the audio signal decoder comprising: a transform domain path (230, 240, 242, 250, 260; 270, 280; 380; 930) configured to obtain a time domain representation (212;386; 938), in the formofanaliasing-reduced time-domainsignal, of aportionof theaudio content encoded in a transform domain mode on the basis of a first set (220; 382; 944a) of spectral coefficients, on the basis of a representation (224; 936) of an aliasing-cancellation stimulus signal and on the basis of a plurality of linear-prediction-domain parameters (222; 384;950a), wherein the transform domain path comprises a spectrum processor (230; 380e; 945) configured to apply a spectral shaping to the first set (944a) of spectral coefficients in dependence onat least a subset of the linear- prediction-domain parameters, to obtain a spectrally-shaped version (232; 380g; 945a) of the first set of spectral coefficients, wherein the transform domain path comprises a first frequency-domain-to-time-domain converter (240; 380h; 946) configured toobtain a time-domain representation (242; 940a) of the audio content on thebasis of the spectrally-shaped version of the first set of spectral coefficients; wherein the transformdomainpathcomprisesanaliasing-cancellation stimulus filter (250; 964) configured to filter the aliasing-cancellation stimulus signal (224; 963a) in dependence on at least a subset of the linear- prediction-domain parameters (222; 384; 934), to derive an aliasing-cancellation synthesis signal (252; 964a) for cancelling aliasing artifacts from the aliasing-cancellation stimulus signal; and wherein the transform domain path also comprises a combiner (260; 978) configured to combine the time- domain representation (242; 940a) of the audio content with the aliasing-cancellation synthesis signal (252; 964), or a post-processed version thereof, to obtain the aliasing-reduced time-domain signal as the decoded representation (212) of the audio content; wherein the transform domain path is a transform-coded-excitation linear-prediction-domain path, wherein themulti-mode audio signal decoder comprises a postprocessing stage comprising a spectral band replication ,SBR, module; and 42 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 wherein the multi-mode audio signal decoder comprises a postprocessing stage comprising an MPEG surround module. 2. Amethod for providing a decoded representation of an audio content on the basis of an encoded representation of the audio content, the method comprising: obtaining a time-domain representation of a portion of the audio content encoded in a transform-coded-excitation linear-prediction-domainmodeon thebasis of a first set of spectral coefficients, on thebasis of a representationof an aliasing-cancellation stimulus signal and on the basis of the plurality of linear-prediction-domain parameters, wherein a spectral shaping is applied to the first set of spectral coefficients in dependence on at least a subset of the linear-prediction-domain parameters, to obtain a spectrally shaped version of the first set of spectral coefficients, and wherein a frequency-domain-to-time-domain conversion is applied to obtain a time-domain representation of the audio content on the basis of the spectrally-shaped version of the first set of spectral coefficients, and wherein the aliasing-cancellation stimulus signal is filtered in dependence of at least a subset of the linear- prediction-domain parameters, to derive an aliasing-cancellation synthesis signal from the aliasing-cancellation stimulus signal, and wherein the time-domain representation of theaudio content is combinedwith thealiasing-cancellation synthesis signal, or a post-processed version thereof, to obtain an aliasing-reduced-time-domain signal, wherein the method is a multi-mode decoding method, wherein the method comprises switching between three modes, a frequency-domain mode, which uses a spectral coefficient information and a scale factor information , the transform-coded-excitation linear-prediction- domain mode, which uses a transform-coded-excitation information and a linear-prediction-domain parameter information, and an algebraic-code-excited-linear-prediction mode, which uses an algebraic-code-excitation- information and a linear-prediction-domain-parameter information; and wherein the method comprises performing a spectral band replication ,SBR, postprocessing; and wherein the method comprises performing an MPEG surround postprocessing. 3. Acomputer program for performing themethodaccording to claim2,when the computer program runsonacomputer. 43 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 44 EP 4 651 130 A1 45 EP 4 651 130 A1 46 EP 4 651 130 A1 47 EP 4 651 130 A1 48 EP 4 651 130 A1 49 EP 4 651 130 A1 50 EP 4 651 130 A1 51 EP 4 651 130 A1 52 EP 4 651 130 A1 53 EP 4 651 130 A1 54 EP 4 651 130 A1 55 EP 4 651 130 A1 56 EP 4 651 130 A1 57 EP 4 651 130 A1 58 EP 4 651 130 A1 59 EP 4 651 130 A1 60 EP 4 651 130 A1 61 EP 4 651 130 A1 62 EP 4 651 130 A1 63 EP 4 651 130 A1 64 EP 4 651 130 A1 65 EP 4 651 130 A1 66 EP 4 651 130 A1 67 EP 4 651 130 A1 68 EP 4 651 130 A1 69 EP 4 651 130 A1 70 EP 4 651 130 A1 71 EP 4 651 130 A1 72 EP 4 651 130 A1 73 EP 4 651 130 A1 74 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 75 EP 4 651 130 A1 5 10 15 20 25 30 35 40 45 50 55 76 EP 4 651 130 A1 REFERENCES CITED IN THE DESCRIPTION This list of references cited by the applicant is for the reader’s convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard. Non-patent literature cited in the description • M. NEUENDORF et al. A Novel Scheme for Low Bitrate Unified Speech and Audio Coding - MPEG- RM0. 126th Convention of the Audio Engineering Society, May 7‑10, 2009, Munich, Germany, 07 May 2009

[0008] • M. XIE ; J.‑P. ADOUL. Embedded algebraic vector quantization (EAVQ) with application to wideband audio coding. IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), 1996, vol. 1, 240-243

[0148] 发明名称：音频信号编码器、音频信号解码器、使用混叠消除对音频 信号进行编码或解码的方法 摘要：一种音频信号解码器（200），用于基于音频内容的编码表示（310） 提供音频内容的解码表示（212），包括变换域路径（230、240、242、 250、260），该变换域路径被配置为基于第一组频谱系数（220）、混 叠消除激励信号的表示（224）以及多个线性预测域参数（222），获 取以变换域模式编码的音频内容中的部分的时域表示（212）。该变换 域路径包括频谱处理器（230），该频谱处理器配置为根据线性预测域 参数的至少子集对第一组频谱系数应用频谱整形，以获取第一组频谱 系数的频谱整形版本（232）。该变换域路径包括第一频域到时域转换 器（240），该第一频域到时域转换器被配置为基于第一组频谱系数的 频谱整形版本来获取音频内容的时域表示。该变换域路径包括混叠消 除激励滤波器，该混叠消除激励滤波器被配置为根据线性预测域参数 （222）的至少子集对混叠消除激励信号（324）进行滤波（250），以 从混叠消除激励信号导出混叠消除合成信号（252）。该变换域路径还 包括组合器（260），该组合器被配置为将音频内容的时域表示（242） 与混叠消除合成信号（252）或其后处理版本进行组合，以获取混叠 减少的时域信号。 摘要

Claims

1. A multi-mode audio signal decoder (200; 360; 900) for providing a decoded representation (212; 399; 998) of an audio content on the basis of an encoded representation (210; 361; 901) of the audio content, wherein the multi-mode audio signal decoder is configured to switch between three modes, a frequency-domain mode, which uses a spectral coefficient information and a scale factor information , a transform-coded-excitation linear-prediction-domain mode, which uses a transform-coded-excitation information and a linear-prediction-domain parameter information, and an algebraic-code-excited-linear-prediction mode, which uses an algebraic-code-excitation-information and a linear-prediction-domain-parameter information, the audio signal decoder comprising: a transform domain path (230, 240, 242, 250, 260; 270, 280; 380; 930) configured to obtain a time domain representation (212; 386; 938), in the form of an aliasing-reduced time-domain signal, of a portion of the audio content encoded in a transform domain mode on the basis of a first set (220; 382; 944a) of spectral coefficients, on the basis of a representation (224; 936) of an aliasing-cancellation stimulus signal and on the basis of a plurality of linear-prediction-domain parameters (222; 384;950a), wherein the transform domain path comprises a spectrum processor (230; 380e; 945) configured to apply a spectral shaping to the first set (944a) of spectral coefficients in dependence on at least a subset of the linear-prediction-domain parameters, to obtain a spectrally-shaped version (232; 380g; 945a) of the first set of spectral coefficients, wherein the transform domain path comprises a first frequency-domain-to-time-domain converter (240; 380h; 946) configured to obtain a time-domain representation (242; 940a) of the audio content on the basis of the spectrally-shaped version of the first set of spectral coefficients; wherein the transform domain path comprises an aliasing-cancellation stimulus filter (250; 964) configured to filter the aliasing-cancellation stimulus signal (224; 963a) in dependence on at least a subset of the linear-prediction-domain parameters (222; 384; 934), to derive an aliasing-cancellation synthesis signal (252; 964a) for cancelling aliasing artifacts from the aliasing-cancellation stimulus signal; and wherein the transform domain path also comprises a combiner (260; 978) configured to combine the time-domain representation (242; 940a) of the audio content with the aliasing-cancellation synthesis signal (252; 964), or a post-processed version thereof, to obtain the aliasing-reduced time-domain signal as the decoded representation (212) of the audio content; wherein the transform domain path is a transform-coded-excitation linear-prediction-domain path, wherein the multi-mode audio signal decoder comprises a postprocessing stage comprising a spectral band replication ,SBR, module; and wherein the multi-mode audio signal decoder comprises a postprocessing stage comprising an MPEG surround module.

2. A method for providing a decoded representation of an audio content on the basis of an encoded representation of the audio content, the method comprising: obtaining a time-domain representation of a portion of the audio content encoded in a transform-coded-excitation linear-prediction-domain mode on the basis of a first set of spectral coefficients, on the basis of a representation of an aliasing-cancellation stimulus signal and on the basis of the plurality of linear-prediction-domain parameters, wherein a spectral shaping is applied to the first set of spectral coefficients in dependence on at least a subset of the linear-prediction-domain parameters, to obtain a spectrally shaped version of the first set of spectral coefficients, and wherein a frequency-domain-to-time-domain conversion is applied to obtain a time-domain representation of the audio content on the basis of the spectrally-shaped version of the first set of spectral coefficients, and wherein the aliasing-cancellation stimulus signal is filtered in dependence of at least a subset of the linear-prediction-domain parameters, to derive an aliasing-cancellation synthesis signal from the aliasing-cancellation stimulus signal, and wherein the time-domain representation of the audio content is combined with the aliasing-cancellation synthesis signal, or a post-processed version thereof, to obtain an aliasing-reduced-time-domain signal, wherein the method is a multi-mode decoding method, wherein the method comprises switching between three modes, a frequency-domain mode, which uses a spectral coefficient information and a scale factor information , the transform-coded-excitation linear-prediction-domain mode, which uses a transform-coded-excitation information and a linear-prediction-domain parameter information, and an algebraic-code-excited-linear-prediction mode, which uses an algebraic-code-excitation-information and a linear-prediction-domain-parameter information; and wherein the method comprises performing a spectral band replication ,SBR, postprocessing; and wherein the method comprises performing an MPEG surround postprocessing.

3. A computer program for performing the method according to claim 2, when the computer program runs on a computer.

Images