ENCODER, DECODER AND METHODS FOR TRANSFORMATION OF SIGNAL DEPENDENT ZOOM ON SPACE AUDIO OBJECT CODIN

专利摘要:
a decoder for generating an audio output signal, comprising one or more audio output channel (s) from a downmix signal, The downmix signal encodes one or more signals (s) from the audio object.The decoder comprises a control unit (181) for defining an activation indication in an activation state, depending on a signal property of, at least one of one or more audio object signal (s), furthermore, the decoder comprises a first analysis module (182) for transforming the downmix signal to obtain a first transformed downmix comprising a plurality of audio channels. primary subband Figure 1c
公开号:BR112015007650A2
申请号:R112015007650-5
申请日:2013-10-02
公开日:2019-11-12
发明作者:Sascha Disch；Bernd Edler；Oliver Hellmuth；Jurgen Herre；Thorsten Kastner；Jouni Paulus
申请人:Fraunhofer Ges Forschung；
IPC主号:

专利说明:

ENCODER. DECODER AND METHODS FOR
TRANSFORMATION OF SIGNAL DEPENDENT ZOOM IN THE CODING OF THE
SPACE AUDIO OBJECT.

[0005 already recently, in the dimpc of. audio coding _f techniques psxamét.rxc 's for transmission / storage: f XO':. Ό The t. í X US: t SZi: HQ ΚίΡΧί : <ÍOS / pinai: S ^:: dd Hdig <
mtMticanal / mplx fob jato were introduced, for example, by the drupe of Special iotas in images in motion IMMPEG Γ Tor; eg Picture Kx | d.rtd Groopj and others. an example ® with MPEG Surround [MPS) as a channel-oriented approach [MPS, BCClv on MPEG audio-spatial object coding from MPEG with an object-oriented approach [JSC _f 3AGC, tÓAOÇ. / f SAOCf] »Gotta oxrxentaOa approach pç.iu object s dsnom _: ina.da as''separation of informed source (ISS1, ISS2, 1.55.1, ISS4, ISS5, TSSfej. These techniques have the objective.' · xx and to reconstruct an audio scene of desired output or an object of the desired audio source based on a · nomnmxx of channels / objectoc and additional information that will describe the forward audio scene / weapon is the objects of the audio source in the audio scene.
Γ100 £ 1 èstimatiya es application of the channel / project related to the additional information in these systems and made in a selective form of time-free. The slip systems of these systems employ the trainings of time “frequency. with the Discrete Fourier Transform (DFT | bfecrete fbarier Tramsfoxm). <the Short Duration Fourier Transform (STFT i Inert Time .Fourier Transform) or filter bank, such as bench and Quadratuna spread felt (QMF 1 Quadrature; s .xàXx:;... · 'aaxte.rj <χ bite-ss & ncapabí IPX :: dK ^ BAAC-Sa ^ oestes systems .Fig'Ura described in Example 3 _f utiliuando c $ & MFEG of GC.
In the case of ST ET, the temporal dimension is
represented by the block number oox cap and the spectral oiour-slo is captured by the oPeero (position) of the spectral ocotient. In the case of QMF, the time dimension is represented by the number of the time period and the spectral dimension is captured by the n a subband band, the first aspect of QMF is employed by the subsequent application of a second filter stage :, all _: the filter bank is called hybrid QMF and precise resolution sub-bands are called hybrid sub-bands , [Ó008] Conformo already meriuionadó above, in S.ACC the general processing is performed in a selective form of frequency of top u can be described as follows, among each frequency fairs, as described in λ Ί • '' í ' 1 *
Signals 11 d c- audio object from input c>
S í'vp 'Π C- XC * xi X.- S-: ·: C. · 1: . <U *: -SS t .0; _:: ·:, >> · p C- X T: UC 'pr ecéssamé-nfto of the codif leader, us e l ma nde w ma tr i. cd® domam.í x, closed in the elements dp ^ · In addition, the encoder extracts additional information that describes the òáraàtexistloas of the audio input objects (additional information estimator module (Sic i side-lnformatlotes ümatorlf. For Mr EG SAOC, the ratios of the object's powers are the most basic form of this additional information,
- Demnmia signals and additional information are transmitted / stored <For this purpose, the signal (sj of audio and demamíx can be compressed (sj, for example, using the well-known perceptual audio encoders in the MFEG -1/2 of Layer II co III (also: conhaclao coco .mp3} _f MPlf-2 / ^; 1 - Audio Coding
Advanced (AAC | Advanced Audio Codinn} etc.
~~ At sztxamtcase> 3 the reception, oacouiftcaocr cò.uçex your icon tries to recover the sonars in obr etc. xx., X: j (separation of object ⁷ '} from demnmix signals (decoded}, using additional information transmitted »These approximate object signals â _; · ... óç are then mixed into a target scene represented by M channels of the audio output y <.... p using a presentation matrix described by the efficient bp - - ίήν in Figure 3, The desired alvp scene could, in the extreme case, be the representation of only one sign of erection out of the mixture (scenario of separation of origin), but also any other arbitrary acoustic scene that includes the transmitted objects »êor For example, the output can be a single-channel, 3-channel stereo or 5.1 multichannel target scene.
: Ç $: üÚÁ: Jí: PS: í SXS themes: COb: ΐΡΡΑ-Ο-: Π®:
they can use a time-free transformation t / f} with temporal resolution and aesthetic resolution »Bs colder an e.n'iTs. :. n & òd · 'QT.sXii *: óB Χ'ζ ~! 5 Q 1.UÇ-âO <· '£ f Í.X <^ L .í OTTCíí βΠVQ.). a compromise between frequency resolution and ^ time »f0010J The effect of a fixed t / f resolution can be demonstrated <the example of typical object signals in a mismatch or audio signal. For example, the spectra of tonal sounds exhibit a harmeninamest structure related to a fundamental frequency, and varies over tones »Frequency energy. For final sates, a high frequency response. The representation of t / f used is useful to separate the tonal spectral regions of band streats from a signal mix. On the contrary, transient signals, such as drum sounds, usually have a temporal ostrutera. distinct: substantial energy is only present for short periods of time and is dispersed over one. wide range of frequencies, iterates these signals, a high t temporal solution of the useful tZl repre- sentation is advantageous to separate the transient signal part from the signal mix, fOOll] Current audio object coding schemes will offer only limited variability in the selection of time · frequency of processing 3 ACQ. tor and x sr; rg 1 o, ob P u GS AOC [3 àCu j (3AG C1] [SAOC 2 J is limited to the top-frequency resolution that can be obtained by using the so-called Hybrid Quadrotor Mirror Bunco Hyb ri d- QPr j Syprid Quadrature dirsor Filter .Sank: · the subsequent aggravation of a parametric band, Thus, the re-recovery of the object in the standard 3AQC ícPisG SAOC> as standardized in [SACC1} gcralbente presents the gross frequency resolution of Hybrid-QEF leading to the sound modulated crosstalk of other audio objects (for example, double speech disturbances in the diacurac or auditory xugosity disturbances in music).
[0012] Audio object encoding schemes, such as the Cod if.ination of B.iual Binaural [BCG | Binaural Cue Coding] and Cedi.fi cation of Par and rich Audio Sources [JSC ’Ca reme trio Joint-Coding of Audio Sources], are also.
limited to use of a fixed resolution filter bank. The current choice of a fixed resolution or transformation filter bank always involves a predetermined compromise in terms of the distinction between the temporal properties is and sounds like that of the co dl f 1 o t ection <
[1) 013] In the field of informed source separation (iIS informed source separation), it has been suggested to dynamically adapt the length of time-frequency transformation to signal properties [1317] as well as cdnbêdidb of perceptual audio coding schemes, for example example, Advanced Audio Coding (AAC) [AAC].
[GG14] The object of the present invention is to provide further enhancements to the audio object encoding »The object of the present invention is solved by a decodifleader, in accordance with claim 1, by an encoder _f according to claim '7, by one ^: decudification method, according to. claim 13, by a method for encoding, in accordance with claim 14, is by a computer program, according to claim Im.
Γ QQl, 5] In contrast to eagc of the state of the art, applications are forueoida to dynamically adapt the resolution of time-frequency to the signal in a compatible regressive way, in a way that
- continuous bit streams of the SAGE parameter originating from a standard SA.GC encoder (MPEG SAOC, as standardized in [SAQC]) can also be decoded by an improved decoder with a perceptual quality that is comparable to that obtained color · a standard double fleader, ~ improved ShOCi parameter bit fluxes can be optimized with Optimal quality with the improved deccdificoder, and continuous and standard SAQC parameter bit streams can be mixed, for example In a scenario of a multi-point control unit (MCd | multi-pofnt control unit), it is a common bit stream. can be decoded with an improved or standard decodlfleader. [0016) For the aforementioned properties, it is useful to provide a representation of the common ram / filter bank that can be dynamically adapted to the resolution of time-frequency, both to support the decoding of the new improved SbOC data as well as , at the same time, for the regressive compatible mapping of traditional standard SAOC data. The merging of the improved SAGC ® data with the standard ãAOC data is possible given as a common representation.
(0: 0.1.7..1 An improved SAOC perceptual quality can be obtained by adapting the resolution of the filter bank rupture frequency, which is used to estimate or used to synthesize the signals of the audio object in the specific properties of the object For example, if the audio object is almost stationary for a certain period of time, the parameter estimation and synthesis is favorably performed in a raw time resolution and an accurate frequency resolution. audio or not containing transient ^ostacionâ:. neriodo streams during a given time, the parameter estimation and synthesis will va s · n 'AICS aa' sn v realmddaa jutliieandsi u.ma ρ r c..... i s of time: and a gross resolution of threescence, thus, the dynamic adaptation of the filter bank or transformation allows a high frequency separation in the spectral separation of signals almost stationary. onáriçs to avoid hi form between objects, and high temporal precision for object initiates for transient events in order to reduse pre- and post echoes>
ί O01 s As for time, the quality of the traditional SA.OC can be obtained by mapping the data of the standard SâOC in the time-frequency grid provided by the adaptive transformation of the inventive regressive compatible signal that depends on the additional information describing the ca ra otaristi ca s, οοί smnail ic dp5 | _: éodíií fuqadãv .mdecs séísrfrficár ^:: iãs®í íO- £ t uados of mhdt <
improved as standard using a common transformation allows for direct backward compatibility for applications covering, the mixing of the new and improved SAOC data.
(0G2Gí An oseedif leader for generating an audio output signal, comprising one more audio channel (s) ..... of a dcwnmlr signal, comprising a plurality of d <wnmi..r samples of o time, and provided Q code encodes them or more signals from the audio object FGQ2 3.] Q decoder comprises a window sequence generator or detersion of the plurality of analysis windows, characterized for each of the analysis windows comprises a plurality of samples of the time domain of the dounu.ix signal, each analysis window of the plurality of analysis windows has a window length indicating the number of domain domnmfx samples time of said analysis window, the sequence generator per window is configured to determine the plurality of analysis windows, so that the length of the window of each analysis window depends on a property of the signal, at least , one of two or more signals from the audio object, it is available, the oucedifleader comprises a t / f analysis module to transform the plurality of time domain domnmfz samples from each analysis window into the plurality of analysis windows in one time domain in one. time frequency domain, depending on the window length of said analysis window, to enter a transformed domamir, [0023] In addition said ·, the decoder comprises an unmixed unit for mixing the dcwnmfx trapsformadQ based on the additional information parametric over two or. more; signals from the audio object to obtain the audio output signal, [0024] According to an application, c · window sequencing generator can be configured to determine the plurality of analysis windows, from which a transient indicating such change dc signal, minus one, two or more signals of the audio object being encoded by the domem.ix signal is comprised of a primary analysis window of the plurality of analysis windows and a secondary analysis window of the plurality of windows of analysis, in which a center: r., - of the primary analysis window is defined by a location t of the transient, according to c <~ t -!. ·>, and a center c ^ of the primary analysis window is defined by the location f of the transient®, according to Odm ο ^, ι · ^::: te .p, where l _á and 1> are numbers _x [Q025] In an application, the window sequence generator can be centigred to determine the plurality of windows of analysis, of so that a transient® indicating a signal change of at least one of two or more signals from the audio object being encoded: by the domamix signal it is understood by a primary analysis window. the plurality of analysis windows, in which a center o; - the primary analysis window s defined by a loci t of the transients, according to ~ t, in which a center c>..; an analysis window, secondary to the plurality of analysis windows, is defined by a location t of the transient®, according to «t ~ e in which a center c> · ,; of a third window of the plurality analysis: 4 ^ windows of analysis o defined by An ibcálisáçãc f of the transient®, according to ccc ^::: 0 la> · where 1,: _i and 1 ^ are numbers>
[0026] According to. In one application, the window sequence generator can be configured to determine the maximum number of analysis windows, in such a way that one of the plurality of analysis windows comprises both in the first number of signal areas in the time domain. ; secondumumórái es amps ^: tras ^: < ^: Λ : ®ter Pd gcminio ^ gives <time, in which the second number of time domain signal samples is greater than the first number of time domain signal samples, and one of the plurality analysis windows is used. number of analysis windows comprises the first number of signal samples from the time domain as -the referenced analysis window ax> comprises a transient ©, indicating a change in the rt signal, at least., one of two or more signals from the audio object being encoded by ^: f0 signal G027] an application, the t / f analysis module can be configured to disturb the time domain domnmix samples of each of the time domain analysis windows; a burst frequency domain employing a QdF filter bank and a. Nyquist filter bank, in which the t / f analysis unit (135) is non-configured to transform the plurality of time domain signal samples from each of the analysis windows depending on the window length of said transition window .
(0028) In addition, an encoder for cooifying two or more signals from the audio object of the input is provided. Each of two or more signals from the incoming audio object comprises a plurality of time domain signal samples. The encoder comprises a sequence unit per window for determining a plurality of analysis cis windows. Window cover oananse comprises a plurality of samples of the signal from the top domain of one of the signals of the incoming audio object, in which each of the windows has a window length indicating the number of signal samples the time domain of that analysis window. , The sequence unit per window is configured to determine the plurality of analysis windows, so that the length · of the yield window of one of the analysis windows depends on a signal property of at least one d® two or more abject signals from the incoming audio.
2900291 In addition, the encoder comprises a t / f analysis unit to transform the time domain signal samples from none of the time domain analysis windows into a time frequency domain to obtain samples of the transformed signal . The t / f analysis unit can be configured to transform the plurality of time domain signal samples from each of the analysis windows depending on the window length of said analysis window .:
[00301 In addition, the encoder comprises the unit of estimation by S1 to determine the additional pathametric information, depending on the sampling of the transformed signal.
[0031j In an application, the encoder asks for eamptepend, it also loves a transient detection unit being configured to determine a plurality of input differences and being configured to determine if a difference between the first among the differences in level, object and the second among the object's nival difference · is greater than a limit value, to determine for each of the analysis windows, if the said analysis window comprises a transient and> indicating a sign change: from at least one of uo.is or more s.ma.is to the input and water entry.
[0032] Depending on an application, the transient detection unit can be configured to employ a detection function: d to determine whether the difference between the first of the differences in level of the object and the
m JJhgfOXA -d)) - MOm [ÕS33] where s indicates an index, where i indicates a first object, where f indicates a second object, in cm i indicates a parametric band. CL2 can, for example, indicate a difference in the level, of the object.
[0034] In one. application, the sequence sequence per window can be configured to determine the plurality of analysis windows; of fear that a transient indicating a signal change of at least one of two nu plus signals from the object u and input audio is understood by a primary analysis window of the plurality of analysis windows is per a secondary analysis window the plurality of
denialo of time, and ® that it is one of the windows of analysis of the plurality of windows of analysis comprises the first number of signal samples of the time domain when the said window of analysis comprises a transients, indicating a signal change of at least one of two or more signals from the incoming audio object, [0137] According to an application, the t / f analysis unit can be configured to transform the signal samples from the time domain of each of the analysis windows: a time domain with a time frequency domain employing a fjW filter bank and a dyquiet filter bank, in which the t / f analysis unit can be configured to transform the plurality of signal samples from the uptime domain ^: one of the analysis windows depending on the window length of the said window [0033] in addition, with a decoder .pdra generating an audio output signal, comprising one or more can the audio output of a dotmmis signal, comprising a plurality of time slot Cornelc samples, is provided. The bmrux signal codes. two cu plus signals from the audio object, the deçodifleader comprises a first analysis sub-module to transform the plurality of time domain demnmix samples to obtain a plurality of sub-bands comprising a plurality of sub-band samples, in addition, the decoder comprises a generation of sequence per window to determine a plurality of analysis windows, where each of the analysis windows comprises a plurality of subband samples of one of the plurality of suo-baudus, one that each window is analyzed. ; of the plurality of 'jars, ®ias of analysis has a window length. indicating the number of sub-band samples of said analysis window, in which the window sequence generator is configured to determine the plurality of analysis windows, so that the length of the window to a window of analysis depends on a property of the signal, by the walls, one of two or more signals of the audio object, In addition to the .- .so, the deçodi.f loader it comprises a second analysis module for transforming the plurality of sample samples from each window of analysis of the plurality of windows and analysis, depending on the window length of said analysis window, to obtain an transforma transform. furthermore, the decoder comprises an unmixed unit for not mixing the trans dcmnm.fr formed with. based on additional information about parameters or more and signal the audio object to get the signal out of the audio, [0039] In addition, an encoder to encode sois or, more signals from the input audio object is provided, each of two or more signals from the incoming audio object cc.mpréende um.a plurality of signal samples from the time domain, C- nud.rf loader comprises a first analysis sub-module to transform the plurality of samples u srnai co _: domxnrc · · The time to obtain plurality; sub-band.s comprising one. plurality of sub-Panda samples. In addition, the coder will use one sequence unit per window to determine a plurality of analysis windows, where each of the analysis windows comprises a plurality of subband bands in a plurality of subband bands. , & t that each of the analysis windows has a length of the window indicating the number of sub-bank tests of said analysis window, and; n that the unit of the sequence per window is configured to determine the plurality of analysis windows, so that you buy mens or da. Each analysis window's window depends on a signal property of at least one of two or. more signals from the incoming audio object. In addition, the cudiiicadar comprises a second analysis module to transform the plurality of sto-bauda samples from each analysis window to the plurality of analysis windows, depending on the length of the window of that window <of the analysis, to obtain samples of the transformed signal, Mém díssa, ocdiiicadar comprises an estimation test by PSI to determine the additional paramátrica information depending on the samples of the transformed signal.
[G94CH also dials, to be used to generate an audio output signal, comprising one or more channels (audio output isj of a domnmfx signal, ® provided »0 femnnífx signal encodes one or more signal isj of the object The deoodifiaador comprises a control unit for defining an indication of activation in a state: of activation, depending on a signal property of at least one of one or more signal (s; of the audio object, In addition, © co-trainer includes a first analysis module for transforming the dornmlx signal to obtain a p rime of the mum 1 xtransformat, with a plurality of primary high-band channels. dif. ·. ca do r compr ends a second analysis module to default, when the activation indication is defined in the activation state, a second domain transformed by three to the reaction of at least one of the sub channels -basic primary for obtaining a plurality of secondary subband channels in which the transformed second domnmfx comprises the primary subband channels that have not been transposed by the second analysis module and the secondary subband rings. In addition, the deoodifioador comprises an unmixed unit in which the mixed unit is configured to not mix the second transformed doonmfr, when the activation indication is defined, in the activation state, by tapping the parametric additional information with one or more signals (is , of the audio object to obtain the audio output signal and not to mix the transformed docnmix printei.ro, when the activation indication is not defined in the activation state, based on the additional parametric information on one or more useful signals) from the audio object to the outside the audio output signal.
(OOllj
In addition, an encoder for encoding a signal from the audio object ^ to the input is provided.
The encoder comprises a centrole unit for defining an indication of. activation in one.
activation stay., depending on -a. signal property of the input audio object signal. In addition, the cudlfreader comprises a first analysis module for transitioning the signal from the incoming audio object to obtain a first signal from the object that transitions audio, wherein the first signal from the transformed audio object comprises a plurality of channels primary subband. In addition, the eedifidadár comprises a second analysis module to generate, when the activation indication is defined in the activation state, a second signal of the audio object transformed by the transformation of, at the ménps, one of the plurality of dp sub channels - primary band to obtain a plurality of secondary subband channels, where the second signal of the transformed audio object comprises the primary subband channels that have not been transformed by the second analysis module and the secondary subband channels Furthermore, the encoder comprises a PSI estimation unit, in which the PSI estimation unit is configured to determine additional parametric information based on the second signal of the transformed audio object, with the activation indication defined in activation status, and to determine additional parametric information based on the first signal of the transformed audio object, when The activation indication is not defined in the activation state.
[QQ <2] In addition, a decoding method: to generate an audio output signal, comprising one cu plus audio output channel (s) from a domnmix signal, and comprising a plurality of sampling methods. domnnlx of time domain, it is closed. The domumix signal encodes more audio signals from the object: The method comprises;
~ determine a plural, isa windows age characterized by the a of the urn of the windows of an a use rounder a plurality of domaximum domain samples. time of the dovamíq signal, in which each analysis window xlurality of analysis windows has a length of the loop indicating the number of dounmix samples time domain of said analysis window ..... and in which the plurality of analysis windows is conducted so that the length of the window of each of the analysis windows depends on a signal property of at least one two or more signals from the audio object.
- transform the plurality of samples of
t. m of time domain of each window of analysis of the alidade of analysis windows of a time domain in time frequency domain, depending on the window length of the said window of the analysis window, to obtain a .Z '1 -4s xS): : 1, - X »Us À À CÍ 'úl' i 'f-::
no xnísrurax or transformed with additional parametric information on two or more audio objects to obtain the output signal of
O.
receive a plurality of signal samples from the
O. The method comprises:
- Determine a number of analysis windows, characterized in that each of the analysis windows comprises a plurality of samples. signal from the time domain of one donates signals from the incoming audio object, where each of the analysis windows has one. cox length of the window indicating the number of signal samples from the time domain of said analysis window and in which the determination of the plurality of analysis windows is conducted in such a way that the length of the window of each of the analysis windows fCepélíG ^ ': ::: GO ί.ϊ.ίΙΟ p O OX 'JÚ UO.Cí vU.- ^:: ' GO ^:: ..; V ·, · p U- Ú.V Χ'ΐΟΧϊΟΌ <Uit. Two or more signals from the incoming audio object.
- Transform the time domain signal samples from each of the time domain analysis windows into a frequency domain. of time to obtain samples of the transformed signal, in which the transformation of the plural age of a.most.ra: s of the signal of the time domain of all of the analysis windows depends on the window length. G-S exi® 'qmúeusã :: by a.na.l. ο so. w determine the additional parametrise information depending on the samples of the processed signal.
[0ο44] In addition, a decoding method generating an audio output signal, comprising one or more channels) of audio output from an atamanxz steal, comprising a plurality of time domain atamnmfx samples, · in which the dormita signal audifíoa two or more signals from the audio object, is provided :. The minutes all comprise;
- Transiornate the plurality of doawílx samples from the delirium of _: time to obtain a plurality of ~ Determine a plurality of: analysis windows, characterized by each of the analysis windows comprising a plurality of samples of aub-bsnda from one of the plurality of sub-bands, in which each analysis window of the plurality of analysis windows has a length of the window indicating the number of sub-hand samples of that window gives analysis, in which the determination of the plurality of analysis windows is necessary so that one of the analysis windows depends on a signal property of at least one of two or more signals from the audio object.
- Transfer to: plurality of subband samples from nothing analysis window of the plurality of analysis windows, depending on the window length of said analysis window, to obtain a transformed fcwm, E:
- Do not mix the transformed dome based on additional parametric information about donated or more signals or object of the auutric to quote a d® iOOiS freckle system] In addition, a method for encoding two or.
more signals from the audio object: dé input, em. that each of two or more signals from the incoming audio object comprises a plurality of time domain signal samples, and provided, the method comprises:
Transform the plurality of time domain signal samples to obtain a plurality of the sub ~ Determine a plurality of analysis windows, where each of the windows gives more plurality of subband samples of one of the plurality of sub-baudas, in which each of the analysis windows has a length of the window indicating the number of subband samples of said analysis window, in which the determination of the plurality of the analysis windows is conducted in a that the length of the window of each of the analysis windows depends on _: a property of the signal of, at least
- I transfer the plurality of sub-band smo-stras of each analysis window in the analysis window glualidaòé, depending on the window length of said analysis window, to obtain samples of the formed rans signal.
Determine the additional parametric information depending on the samples of the transformed signal. [00161 In addition said, a decoding method using an audio output signal comprising one or more channels (rs; audio output of one: I will donate a signal, in which the dwnwnmfx signal encodes two of the signals of the object of audio is provided, the method comprises:
- Define an activation indication in an activation data, depending on the one. Oronrlednde dc · signal
- Transforming the damned n signal to obtain the first transformed domwfx, comprising a plurality of primary subband channels.
Generate, when the activation indication is defined in the activation state, a second tame a t r a n s for p rma: foot la t r ane f a rmation of #. at least one of the primary subband channels to obtain a plurality of senúndárlos subband channels, where the second transformed domnmix comprises the primary subband channels that have not been transformed by the second analysis module and by secondary subband channels, S;
give the second transformed domnmim mix, when the activation indication is defined in the activation state, based on additional parametric information about two or more signals from the audio object to obtain the audio output signal, and not mix the first audio signal. ::,;: ^ quanCÇ :::: <: a>'..O.ittçiv OS ísfíád ^: O defined in the activation state, based on additional information paramdtriáa overdis or more signals of the audio object to obtain the output signal of audio.
Define an activation indication in an activation state, depending on a signal property of at least two or more signals from the incoming audio object.
- Transform each of the si.nais: of the input audio object to obtain u.m. first signal of the formed audio object trains are damaged by the signal of the audio object of s.utrada, in which said first signal of the transitional audio object comprises a plurality of primary subband channels.
- Generate for each of the signals of the incoming audio object, using the indication e and activation and derived in the activation state, a second signal of the audio object transformed by the transformation of. at least one of the primary subband nasales of the first signal of the audio object trains supplied from said signal of the incoming audio object to obtain a plurality of secondary subband channels :, st that o; said second formed domnmlú urans comprise said primary sub-nauda channels that have not been transitioned by the second analysis module and said secondary sub-bauda channels, E:
- Determine the additional parametric cam based on the second transformed audio object signal of each of the input audio object's signals. »When the activation indication is defined in the activation state, and determination of additional information to meet rich based on the first signal of the transformed audio object of _: each one of the signals of the input audio object, when the activation indication is not defined in the activation state.
[004B] In addition, a computer program to implement one of the methods described above, when executed in. a computer or signal processor, is provided.

mono simply be denoted fu, in order to allow the d & c o d i f 1 chain of 7 AO C r and coupe r and o individual objects:
the additional information slayer 17 provides SAOC decadifleader 12 with additional information including parameters from 7.AG0,
For example, in the case of a stereo doendx, SAOC parameters comprise differences in object level (OLD and rei dif fers cea) correlations between objects (IOC | Imtsr-objeot correlations) (cross-correlation parameters between objects), give values docnmix gain (PMG I domamlx gain vai lacei ® differences: co ti vc 1 do c ar- a 1 de do cis (DGLG doxcrnlx obannc.i led differences}, Additional information including the parameters of SAOC, with the mix-dos signal Is the SAOC output data stream received by the SAuC decoder
Γ0076j
The SAOC decoder comprises a
Upmíxer that receives sinau:
domnmlx 18, as well as additional information in order to retrieve and display the audio signals ήe Ô;
in any set of channels se1 oned by 1 the user y _; u yp with the representation being prescribed by the presentation of information inserted in the decoder of 7777 12.
Os: audio signals -a iÇ;
can be inserted in code 10 in. any coding domain, such comes in the spectral domain I of time.
h <:>
θη s audio signals if they are coded 10 in the time domain.
as encoded by PCM the encoder 10 can use a filter stand and a hybrid QFF bank, in order to damage the signals in a spectral domain _: in wed 3 the audio signals are represented and there are several associated sub-basements in different spectral parts, in a specific resolution of the filter bank, If the audio signals are the ,, are already in the representation, the gas expected by the coder 10, some do not have to realize the spectral composition <
[00: 781 Figure 4 shows a signal of _: audio in the spectral domain already mentioned. Ccirno asks to be seen, the audio signal is represented as a plurality of sub-baud signals, each sub-band signal 3C _; . to 30, consists of a temporal sequence of sub-bartda values indicated by the small boxes: 32. As can be seen, the sub-panda values _: 32 of the signals: from sub-band 30- to pCç. they are synchronized with each other on the top, so that, for each consecutive filter bank time period 34, each subband 30 _; to 30:.; comprises an exact subband value 32. As illustrated by the frequency axis 3ê, the subband signals 3¾ to 3 0; <are associated with different frequency regions, it is according to II us trade by the time axis: 38 , the time periods of the filter bank 34 are nonsecutively arranged in time.
[00'39] As described above, the additional information extractor 17 in Figure 3 computes the SêOC parameters of the audio signals given inputs. _; a sv., according to c · pad.rãò SAQC currently implemented, g encodes this computational calculation using a οοηψό / frequency resolution that can be reduced in relation to the

u: r: the objects. 0 computational calculation and oomc follows
the values of the sub-band partoncéntés at a certain portion of the time / frequency 42. f and g denoting a certain pair of audio objects a a s ^, and Re] f denoting the operation of t · ::: ·: · : ·. ..m. .... ...
discard the imaginary part of the complex argument,
Γ00Β4]: Q domnmixer 16 in Figure 3 reduces the objects only by the use of 1-pinned gain factors in each cojection a, as .:. That is, a factor of gain cj is applied to object 1 and then all objects weighted a a-s are added together to obtain a mono dowmix signal, which is exemplified in Figure 3 if Fl, In another case: example of one; two-channel domnmax signal, described in Figure 3 if F ^::: 2, a gain factor dj _Z í is applied to object 1 and then these gain amplified objects are added together to obtain the left-hand annum Lu, b gain factors d / _z <sac> agglomerated under object 1 e, ®ntâo _{: <} the gain amplified products are added together to obtain the right domumim channel 31. dm prooesáámentQ which is analogous to the above must be applied in the case of a . dcmnmlx multi nana. ·. íHz /}, (00351 This domnmlx prescription is signaled next to the decoder by means of dovnmax gains .DMb and, in the case of a stereo domnmix signal, differences in ^: num channel level in DCLLL.
[Ü08Ê] dcmcmdx's earnings are the same amount of *
a-cc-rdo com;
IH / C ', -20: feg _ít) H. + r) (daunmix mono), DÂdr,; = lÔk> g _ÍS (ífj edjeaj, [stereo doratix,! _f [0067; Where ^ε is tm small number hornΐΟί'α [0038} To DCLDa the following formula is applied:
'' Ϊ: </: 4 g
(00191 In m normal mode, the cmemmxex 16 generates the signal
ie dounmlx, from acoxdb cam:
(10) - ........ and’(« £): i [009’3] For a dotntpix mono, ass ('£ 01 , x í t 'l1 & <. P ; i UoJ ' IM ..... b 1' M [0091] _for xii i dcw ^: iO: .ix entered, respectívameúté. [00 9 2]. At ii n, in the formulas mentioned çç.ima, cs GXD and 10 parameters C 3610 ’υ.; Τϊ · ά1 ΪUHÇttO GO3 t 3 .0ΠXS (frog mCj i.O 3
parameters ZWG s DCID S-So a function of d. yes it is observed
that d can aer several · cover and frequency » í 0093] Atsir n., in normal mode, α dewniner 16 mix all objects s _; a .¾ without preferences, that is, with handles all <Diu j GV, Q3 3d f3 ^: 3> τ .6Q 33 .1 ΓΓ;. £ ΗΊ td3 [00941 Nc 1 « decoder, the ucmixer realizes the
intervention of the domnadn procedure and the implementation of presentation 26 information represented by a matrix R (in the veritable literature also called Al in a step of the cal d.ü 1 o Compdt ac 1 dn à 1, as ã b ar, n õ c as o of domnm 1 x of do 1 s O <'i I -x 3 ^: .1.3 i:
implementation in the example of the Additional Information Esirmador [SIBj go-mç part of an encoder of óAOC lü, The encoder of SAOC 10 comprises mixer 16 and the estimator of additional information (SIS) 11. G 3IE conceptually consists of two modules: a module 45 to compute a short-lived t / f representation (for example, STF or ÇíMF · d: and each signal. The computed short-lived t / f representation is inserted into a second module 46, c module of estimation of additional selective information of t / f (t / f-SIt!). The module of t / f ^ SIE 46 computes the additional information for each t / f portion. current basics of SAOC, the time / freewheeling transformation It is fixed and identical for all audio objects a, m .- <Also, the SAOC parameters are determined by scoring the SACC structures which are the same for all audio objects and have the same time / frequency resolution for all audio objects s> to sv, ass im · d & advising the specific needs of the object for precise temporal resolution in some cases or precise spectral resolution in others.
[0100] In the following, the applications of the present invention are described.
[0101] Figure 1 illustrates a deccdifloader for generating an audio output signal, comprising one or more channels (audio output of a downmix signal, comprising a plurality of both time domain, time dcwcix eras. according to an application. The domnmix signal encodes two on _: more signals from the audio object.
[0102]: The conditioner comprises a sequence generator in the window 134 that determines : a plurality of
Ά L. · A * analysis windows (for example, based on additional porametric information, for example, differences in object level), ®m that each of the analysis windows comprises a plurality of samples of: domain dornamis term of sinsí da ttwsu ^ i Each gain d mm da, o. ·. uralz ca e of analysis windows has a length: of the window indicating the number of «most d ama s s of dumí ηi o i empo rt of the retained analysis window. The window sequence generator 134 is configured to determine the plurality of analysis windows, so that the window length of each analysis window depends on a signal property giving at least one. two ts plus signals from the audio object. For example, the length of the window may depend on whether said analysis window comprises a transient, indicating a signal change of at least one of two or more signals from the audio object being encoded by the doaxwix signal.,
(.03.031 Fame determining the plurality of Analysis windows, Ρ sequence generator per window 134 can, for example, additional information for «analysis metrics, for example, differences transmitted at the object level with respect to two more boxes signals of the audio object, to determine the length of the window of the analysis windows, from mo: od .que ·: o : cqmgmme: n: t. o : da _.: j: ra .qe ann : :: caca: one: EA: s : gandras: analysis depends on a signal property of at least one of two or more signals to the audio object, or, for example, to terminate the gluing of o® in the analysis nets, the generator d® sequence by window 134 pods can analyze the forms
39/3 9 jandlá own windows or Cs analysis, ethoxy that there rormas AET-iqmm '; IAW: the puim -dm -apa; would _<up::: go-nor, -pto-- -éae: mg <i. <jÇ: - be rated in the continuous flow of hits from the encoder to the dacodifloader, and the length of the window of each of the analysis windows depends on a signal property of at least one of the two or more signals from the audio object.
[0104 In addition, r: dscoder comprises a pure t / f 135 analysis module transforming the plurality of time domain donnmix samples gives each analysis window the plurality of analysis windows from one time domain to a frequency domain of time, depending on the window length of that analysis window, to obtain the transformed dcmúatix.
In addition, the code includes an unmixed unit 136 for mixing the transformed dosvnm.íx based on additional parametric information about two or more signals from the audio object to obtain the sanai nc output from [OliMj the following applications used a mechanism for constructing the following with the special window »A function of the prototype window f ia,, py a defined for _: Index 0 á n '> tf ~ 1 for a length of the window Λ1. When drawing a single window w.í }, three control points are necessary, namely, the uentrua da jane.:.a previous, acti and pic Xxma, c., J, q, eq,.: »
Γ0137] 0111100000-00, the window function is f 1. n i da c orno
'·* .1. / Ο 17 [Qlll] As explained later with respect to Figure 1Ô, the window sequence generator 124 can, for example, be centigrade to characterize the age of analysis windows, so that a stranger be understood by a primary analysis window of the plurality of analysis windows, in which one within cy of the primary analysis window is defined 'by a translational t lalition, according to one hundred m ^::: t, in which a center c ^ -χ of a window to analyze secunoarxa in the piurá.i.x ue windows and analysis ® defined by a location t of the lens, according to e, ..; ~ t - 1 where an οχη center. of a third analysis window of the plurality of analysis windows is defined by a location t of the transieute, of the heart with c ^ .t ·· ^:::: te ,,, in which l. and.! & are numbers.
[0111 ’] As explained later with respect to Figure 11, the window sequence generator 134 can, for example, be configured to determine the plurality of gn ctnql ..,. c ©, u © g ©. us-.c * uers -s g.,. ..-, -¾ .. .-. O.u'.íc gives windows for analysis both comprising a first number of time domain signal samples or a second number of time domain signal samples, in which u the second number of signal samples in the time domain is greater than the first number of signal samples in the buffer domain, and each of the analysis windows of the plurality of analysis windows comprises the first number of signal samples of the time domain when ra reraaaj an e 1 aq © a na 11 if with m ar re and a tran st & π te.
(íjcid-] one application, with the curf of anal rua rif 13 oe Gonfigurado paxa- transform the samples of the dômamlz of the time domain of each one of the windows of analysis of a time doinin in a domain of f encla using a QMF filter bank and a Nyquist filter bank, where the t / f analysis unit (135) is configured to transform the plurality of signal samples from the dé: time domain of each window of analysis depending on the window failure of said window by ana J. Ise.
[0114] Figure 2 illustrates an encoder for encoding them or more signals from the incoming audio object. 'Ixcj déj hurt '_S:::; or xmas-iSi ί® ; ϊηοθ: θκ <όθκ nbgU'tie · :: in the year à.g c.nt raoa comprises a plurality of time hate signal samples.
[0115] The user comprises one sequence unit per window 102 to determine a plurality of analysis windows. Each of the analysis windows comprises a plurality of signal samples from the time domain of one of the signals of the input audio object, where each of the analysis windows has a window length indicating the number of signal masters in the domain of the analysis window. Ά sequence unit per window 102 is configured to daterolosr the plurality of the analysis windows, so that the curapriisanto d: the oaclâ window one of the analysis windows depends on: a property of the signal, by rracs, one of two or. more signals from the incoming audio object. For example, the greeting of the window may depend çé ç that analysis window comprises a
[ülltj In addition, the encoder comprises an analysis unit of t / f ltd to transfer the signal samples from the buffer domain of each one into the analysis windows of a ..... pain ··ioio of time in one; time frequency domain for obtaining more of the transformed signal. The t / f analysis unit 103 can be configured to transform the plurality of time domain signal samples from each of the analysis windows depending on the window length of said analysis window.
[0117] In addition, the _: encoder comprises the PS1 estimation unit: for determining the additional parametric information depending on the samples of the signal.
[0113] In an application, the cooifloader can, for example, comprise, still, a detection unit transients 101 being configured to determine a plurality of differences in the object level of two or more signals from the incoming apio object and sends confluclated to determine, if a difference snaps a first one among the differences in level of the object and a second one of different levels of: object is greater than a limit value, for cc se rr.ina r: pamá ::: here: ua _K mma .... seas jsSneraS: me ·: ianáuisé:, ::: if the next analysis window comprises a _: ttansient, indicating a change in signal from at least two or more signals from the erAxada audio object.
[0110] According to an application, the transients detection unit 101 is configured to employ a tuning and detection (n.) Guarantee to determine if the difference between a urine and one of the differences in level of the object and the second among the differences level of the object is greater than the limit value, where the detection function dfaj is • 'i © £ l.rÚ gives COu: Ox <
a (n) s jtog (Ôl £ l..0.η I)) log (Hi. £>, (d,
it can, for example, indicate a difference in the level of the object.
of primary analysis of the plurality of analysis windows and by a window of xéoundary analysis of the plurality of analysis windows, in which a center c * of the primary analysis window is defined: by a localization t of the transients, on board with ο _λ - t M, and a center o ,, .. · of the primary analysis window is defined by the inaction location t of the transients ,: of ac o roo with <q.>; ™ te l. _s , e: v que ./. ». and 1. ·> sáç nuamros,
10122J As explained in relation to Figure 10, the sequence unit per window 102 can, for example, be set to determine the plurality of analysis windows, so that a transient® indicating a change, at least, of a signal one of them or real audio object signals: input is understood by a window, of primary analysis of the plurality of analysis windows, in which one. center m of the primary analysis window is defined by XUmétí :: xidoalap.apd: © ::: t ::: :: d ^: d :: :: dt: ΟΟΟ. '. ΟΟ Ppm,: ΟΡ: V:, 1 ::, :: CO, · SUU a center of a secondary analysis window of: pd, Ut: á .1.10.000 sUO :: Us': £ · Á.aS, ::: 0: 0 :::: : 000 ::: 0.1: 0: 0 :: ^::: is: <::::: 00: 10: 0: Xpbt 000 :::::: <UC location cesium transience t, chord with ο _;; . · Ι ~ t ~ ls>, and in yoke: 00 ::, fH ··. · SS '.UTiô. ^:: ΡΟ: 0: Ζ: ®χ: Ο: ό :::: 'jpUem of - . t oO ..; .. e 03 S0 plurality of analysis windows set pair and a T trans location. rente, according to cam <;<<: ~ t + 1 _{<: /} where l. _s and 1 ^ are numbers>
[CH. ] Conforms explained later with respect to Figure 11, the sequence unit per ICt window can, for example, be used to determine the plurality of analysis windows, so that each of the analysis windows plòtàlídáde it comprises both a first number of time domain signal samples and a second number of time domain signal samples, where the second number of time domain signal samples is greater than the first number of samples. ura signal of the dpmánnio de tempo, and in which each of the analysis windows of the plurality of analysis windows comprises the first number of signal samples of the time domain when said analysis window comprises one, transients, bypassing a change sign of, skins less, one of two or more signs of the Indian object of entra.da.
6/89 [01241 D and according to an application, a. t / f analysis unit 103 is configured to transform the signal samples from the time domain of each of the analysis windows of a time domain into a time frequency domain using a QMF filter bank and a filter bank òe Nyquist, in which the analysis unit of the t / f 103 is configured to transform the plurality of signal samples of the time domain of each of the analysis windows depending on the length of the window of the rafar ida analysis window »[012: 5] b next, an improved SAOC using regressive compatible adaptive filter banks, from
gives you the ability to decode the standard bit stream of standard SAOC encoders with good quality. Decryption is limited to. parametric reconstruction only and possible residual flows are ignored.
[0128] Figure 6 depicts a blood diagram of an improved SAQC decoder, according to one application, illustrating the standard decoding SAOC bit stream <The black nsfgrated functional blocks (132, 133, 134, 135) indicate inventive processing. Additional information paramétríea (.PSI) cens yacht in sets of object level differences _(QLD):, correlations between objeto® (lóCb, and a amm.x matrix D used to create the dowrrix signal [Audio UMD) of ® individual objects in the decodifloader. Each defined parameter is associated with one. limit, of the parameter that defines the temporal region in which the parameters are associated®, ia 1AOC standard, the frequency positions of the representation of the underlying teatpsZf representation are grouped in parametric®® band. The band spacing resembles the critical bands in the human auditory system. In addition, several t / f representation structures can be grouped into a parameter structure. Both operations provide a reduction in the amount of additional information needed at the cost of modeling inaccuracies.
rêlrfj As described in the S.AGC standard, COs and 1OC® are used to reduce the unmixed matrix G-ElVj, swimming the elements of E are E (/, /) ~ / OC. _t ^ _OLD! OLO _if approximates the cross-correlation matrix of ohjéto,, fej are indexes of the object, J ® | > and Ü is for the transposition of D. An unmixed MARI® calculator 131 can be configured to calculate the unmixed matrix correctly.
[0133 .; The unmixed matrix is then linearly interpellated by a tempered interpolator. 132 of the matrix no
mlstursda gives structure previous about the structure of parameter up until the limit c parameter at the which S values e® Limados are reached®, heart by SAOC standard, this results
in the unmixed matrices for each time / frequency analysis window, and parametric band.
[0131] The resolution frequency of the resolution band gives representation of time / treqnència in this window of analysis by a unit of adaptation of frequency resolution of window 133. (.making the unmixed mat laughing asked for the parametric tail ή in a time structure is defined with G (f}, the unmixed coefficients are used for all frequency positions within this band for the axle.
[0132] A window sequence generator 134 is configured to use information from the defined range of the PSI parameter to determine an appropriate winding sequence to analyze the incoming audio signal. The main sxigeuc.ia is that when there is a definite limit of the parameter in the FSI, the cross point between the executive analysis windows must not correspond to it. d 'jane.l.amentc also; determines the frequency resolution of the data within each window (used in the expansion of data without my t ure, as previously described).
[01331 The windowed data are then transformed by the t / f 135 analysis module into a frequency domain representation using an appropriate temperature-frequency transformation, for example. Discrete Fourier Transform (OFT), Transforms formed by Coe without Discrete Ho di ficada: Complex (CMDCT | Complex Audifled Dfeorete Cosine Transform ^ or Transformed Fourier Transform with differential empiIhamentc (CDFT ^; S Oddly steered DIscrete Four, er T · a ex · '·>-fi'-'ry'. [Cl 34] Finally, an unmixed unit 136 to pl., causes the nan mixed matrices of position by frequency and structure in the spectral representation of the ddx signal X to obtain ¥ parametric reconstructions.
freckle, y and¥. -1Μ.Λ - a combination 1 d. ear channels dcwumix [31331 The quality what can be obtained < like this it is for the majority of: finitIly features
indistinguishable. d result obtained with a stoc standard deeodifloader.
[0136 ·] It should be noted that the text above describes the reconstruction of individual objects, but in the standard SAQC the presentation is included in the unmixed matrix, that is, it is included in the paramedic intarpolation. As an operation
1 inea r _f u order of operations matters.d i fe rance. is ci x g n a of note.[0137]THE next _: go, flow deoodfication © continup bit in 3AQC me.lhara.da by a decod.rf leader from S3QC
improvement is disc r1 to.
[01331 The main bottom of the decadifier from SAQC improved already described n.a
decoding the bitstream of the standard SãOC. This section will detail how improvements in improved SAÇC 1 strode ddá in PSI can be used to achieve a better perceptual quai i fade1.
[3133] Figure 7 describes the functional blocks main decodifleader, according to an application illustrating the denodification of the knitting of the resolution of
frequency. The bold black background blocks (133, 133, each parametric band. In the frequency resolution used for improvements, for example, in IC14 positions, this is done by replicating the value over the frequency positions that correspond to the parametric band. results in. new üLDs OW *** (/) “K (/. à) <. ZtZ) (ó) and IQCs / OÇTX /) ~ K (/ ₅ f} / tXl fan; * K (/ _S f) ^f a core matrix that defines the assignment of frequency / parametric band positions to the
ÍL iffeè otherwise
correction to obtain the delta function Cf (f) of the remaining size that the OLD and IQC expanded.
[01421 union, ã: delta application unit. 143 applies the delta to expanded OLD values, and> the OLD values of precise resolution obtained are obtained by «= C, í fX (f).
[0143] In a particular application, the calculation of unmixed matrices, can, for example, be done by the unmixed matrix calculator 131 as with continuous bit flow. of the standard decoding 3LOC: G (/) «W /) D '(/ W), with E,. (7) - AX 7 (./}, / (ΪΖZ> H /) OZZ) f'^; (/), e »j (/) Ó (/) Eí /) Ó (./ ') | «If: uesçjadc, a. matrix c and presentation could be multiplied in the matrix r-ac · mixed <»: (/) ♦ The temporal interpellation by the temporal ihterpolator 132 follows
u. us „. pa αχ a α [01441 Eat the frequency resolution in each window can be different [generally lower) from the high frequency nominal resolution, the frequency resolution adaptation unit of window 133 needs to adapt the unmixed matrices to match the resolution of the spectral data of the audio to allow it to be applied. This can be done, for example, by checking the coefficients: leaving the frequency axis for the correct rsscJ.uçào. Or if the resolutions are multiple integers, simply average the high-resolution data from the indexes; that respond. a frequency position in the lower resolution G ^ (è) -l / | âÍ ^ G (/), fu 1451 The sequence information of the continuous bit flow window asks to be used to obtain a couplet time-frequency analysis , ameuts complementary to a mtaxtíxano ^: ípp · καάΡχχαχααρχ ', :: sSttsa ssqueuoia í®: ραne · .1 amstrcc..c- çms be built based on the limits of the parameter, as it is theft in the decoding of the continuous bit flow standard SÃOC. For this, a window sequence generator 134 can be employed.
(146) The time analysis of the audio frequency of domnmfx is then the gox output and a t / f 135 analysis tool using the given windows.
[0147j Finally, the temporalmonte unmixed and spatially spaced (poesiwlmente) unmixed matrices are applied by an unmixed unit 136 in the time / frequency representation of the input audio, and the output channel / can be obtained as a linear combination of the channels input ¥, (/} * [0143} Next, the 3 ACC meIhorasa compatible ccdifiCá-çãe recceasvava and described.
[0149] Now, an improved SACC encoder that produces one. continuous bit flow containing a part of the additional compatible information regressive and further improvements is described. Existing SAOC standard decooifleaders can decode the regressive compatible part of FS1 and produce reconstructions: of the objects »A. Added information used by the improved SAOC decoder improves the perceptual quality of the reconstructions in most cases. In addition, if the improved SACC decodifleader is operating in imt all resources, the improvements can be ignored and a reconstruction of the basic Act is still obtained. It should be noted that the reconstructions of the standard SACC and improved SAOC decoders using only the standard SAOC Policy, differ, but are judged to be very similarly perceived (the difference is similar in nature: as in the standard SAOC bit stream of decoding with an improved SACC decodifloader).
[0150j Figure 3 illustrates a diagram. in blocks of an encoder,: according to a particular application that implements the parametric passage of the encoder described above. Functional blocks in bold letters (1'02, 103; indicate inventive processing. In particular, Figure 8 illustrates a block diagram of the two-stage eodification produced _. for decodifleaders plus coats.
[Olülj First, the signal is subdivided into analysis structures, which are then transformed into a frequency domain. Several analysis structures are grouped into a fixed length parameter structure using, for example, MPdQ lengths 8àOC of the 16 and 32 analysis structures are common, it is assumed that the signal properties remain almost stationary during the párãmeiw structure and can then be characterized with: only a set of parameters, if the signal characteristics change, within the structure of the parameter, the modeling error is presented, and it would be useful to subdivide the structure of the longest parameter into parts in which the almost stationary assumption is unrealized. a, (0132) Transitions can be detected by the transient detection unit lul of all input objects separately, and how much is un. transient event is just one of the objects that will. location is declared as a global transient inoculation. The information of the locations is used to manage an appropriate winding sequence. The structure asks to be based, for example, on the logical deck:
::: S ·: :: | pf d ^ pir; Çp: çqppr: lbu: tltb: (dp jΙαΡΡίΡ: Ipádtâp ;, ^: pp:
that is, a length of a standard signal transformation hlocc, for example, 2048 samples,
- Define the length of the parameter structure, for example, 4Q36 samples, color resonant to 4 standard windows with overlapping SS SSt. The: ® strut ura of the parameter group several: windows together and a single set of signal duseritorns is used for the entire block instead of having disorientations for each window separately <This allows to reduce the amount of PSI.,
If no transients have been detected, use the standard windows and full compliance with the parameter structure,
example, MPEG},. This is felts ago © spectral positions in the parametric bands of ·. MPEG SAQC and estimating XOCs, OLDs and energies of absolute objects {NPG in the bands. Just following the MEEG SAOC notation, the normal irate product of two spectra of the object S _; (/, «) The SX / jaj in W parameter portion is defined as
Σ Σ: W, fl »M (/, njSÇ / jn) /« O
...................... djnd ........................... ....................
ΣΣ ^Κίά '/ ^) b-XÍ ^ íj f
[GlSd] wave the matrix defines d mapping of the representation positions of t / f / ç in the P structure (of the M structures in this parameter structure) in parametric bands $ by '1 iff and 5 w, 4 ^. 1, ik oLiienvíse '' _f e [015'7] 8 'are played with the S. complex. The resolution and spectra.), Can vary between times within a single parametric block., Thus the matrix of mapping converts the data into. a base of common resolution ,, The energy of the maximum object in this parameterization portion is defined as the energy of the maximum object “maxlí ^ jg] ΧΙ $) * Having this value, OLDs are then defined as the normalized object energies s
A W)
Γ0158]
E fFi> TOC can also be obtained from raw materials such as ^:
This concludes the estimate of the negative parts of the standard bit stream continuous SAOC.
[0160]
A .105 gross spectrophotometer reconstruction unit is configured to use the 01 Da and NRGs to reconstruct a gross spectral envelope estimate in the parameter analysis block. The envelope is constructed at the highest frequency resolution used in that block.
Also the original spectrum of each analysis window is used by an active unit of the IQê power spectrum to calculate the power spectrum in this window.
[0162] The ροtoneia spectra obtained are nxansiu beloved ·> was <um :: reprssêntagãaí / ^ da: xiMSsrsçs®: ícçrápmsí ^:: daríí jai-ta frequency by a frequency resolution adaptation unit .10.71 This asks to be done, for example, by interpellation of the spectral power values. Then, the average power spectral profile is calculated by calculating the average of the spectra within the parameter block. This roughly corresponds to the GLD estimate that emits the aggregation of the parametric band. The spectral profile obtained is considered as the OLD of precise resolution.
[01631. The 1.0S delta estimation unit is configured to estimate a correction factor, ^{> s} de.lta, · for example, dividing the OLD of precise resolution by the reconstruction: of the gross power spectrum. As a result, this provides for each frequency position a correction factor (multiply that I could use p 3GM .1.03 V 3 L- ^: <'. LG SQ À.' D.Ç <3 Q pX £ · CS O .Q3 OS 3 SpOC-C 30S
described for use in conjunction with SAQC. Seen in a podant way, it does not have the adjective of deteatax the transients, but instead of the changes in the parameterization of the signal that can also be triggered, for example, by a sound compensation.
[QÍÕij The input signal is divided into overlapping turret structures, and the structures are transformed in the frequency domain, for example, with the Discrete Fourier Transform '[D FT) ,. The complex spectrum is transformed to the: power spectrum by multiplying the values with their complex conjugates (or if already, quadrature their absolute values; then a grouping of the parmostricc band, similar to that used in the standard 3AOT, is nutiladg, to energy. each parametric band in each time period in each object is calculated. The operations are short
P, íô, í ) = Vs, ( „ <! $ '(F, I) f« XX f
ΓϋίηΟ] where 3 - (/ ,. a) is the complex spectrum of object i in time structure h. A. sum passes over the frequency / band positions i. To remove some noise effect from the data,. values are filtered by low-pass with a first order hr filter:
Ρ * ’(Μ« JOl »''!) + 0 <1v) 1n (0dj
[0171] where is the filter return efficiency, for example, u _{;; í} ~ 0.9.
[Cl 77] The main parameterization in SAOC is the differences in level of the object (ÓBbsA The proposed detection method tries to detect when the CUDs would change, so all pairs of the object are inspected with tone 0L £ 1. (&,> ) “P / '^; (à.sã) / Pj (ca') · ds changes in all pairs of abject union are added in a detection function pdr dia) ® Χύ, η ·· 1)] '~ feuíXiW:
, ·: ^Μ '.....' · 'V <': ''. · T ^z '::>
[017 31 Gs worth obtained are compared with a
Limits T filter for filtering small level deviations, and; sxsiçs9: © ®í mrnxma: enter the ':, DSBs: ^nççes:: <çonseçuuxvas and imposts Ass .. IM detection function is
11. if> Π & (ú (to - 0. Wx: η - Z, <a <») <3b ) To '| 0 [0174] In the following, the resolution of the improved SAOC frequency resolution is described.
(0175] The shallow frequency lotion obtained from the analysis of the standard 3Ato is limited to the number of parametric bands, with the maximum value of 28 in the standard SAOC, bias are obtained from a hybrid filter bank consisting of a Q.MF analysis. 64 bands followed by a hybrid filtering stage Γ> 3S S, Ih-dl-lS 03XX3B B, ίQXVΣO, LΠΟΚ '- ^ 3S GÍG ú ^ i ·· SLlGft03ΠG3S The frequency ranges obtained are grouped in parametric bands imitating the resolution of the critical band of the human auditory system. The grouping allows to reduce the bit rate of the necessary additional information.
[017í] The euistentn system produces a reasonable separation quality given the reasonably low bit rate. G main problem is insufficient frequency resolution for a seçáraçao xxçpa u sounds tonars »it is cxio-idc · MPC halo ^'7 other objects around the tonal eemoonentes ma perturnação co rip © vocoder. Emits prejudus.orai Oeste ale can be reduced by increasing the parametric resolution resolution. It was observed that a resolution even higher than 512 bands at the sampling rate 4 4.1 Ht) produces good percaptualmaute separation in the signals: from the east> this resolution would request to be optimized by extending the stagnation and hybrid filtering of the existing system, but ibfidus filters need to be of a high order for a sepafãçáb uriciente taking ® have an aitc cost of computational calculation.
0177] dma ferra simple to obtain the resolution of neocssarian reguenela. is to use a temperature transformation with beats in DfTb. · These can be implemented f i cially via a. Fast transfer and Fourier -EFT Fast Fourier Transform). Instead of an Ornate DFT, CMDCT or GDW are considered® as an alternative®. The difference is that the last two are odd and the btldo spectrum contains pure positive and negative frequencies, tied at a D ET, the frequency positions are alternated with a width of 0.5 positions. At 13Ff w of the positions is friendship at G Hz and another at the Nyquist frequency. The difference between mi 'and CMu) is that they contain an operation and additional modulation that affects the phase spectrum. The emphasis of this is that the complex spectrum (resulting consists of the Transform of Modified Discrete Modem ÇMDCT) and the Transform of Modified Discrete Modem (MDST (éfodified F sore te Sine Transform;
Qi 78] A hundred b-ase transform in length DFT Λ produs an aspect re complex with. ; V values. When the sequence is upset and in vaiur msi _z only A ·· '.' I of these vs flowers are necessary for a perfect reconstruction; the others value A / 2 can be obtained from the data with simple manipulations. Does the analysis normally operate by submitting the structure of A samples signal time domain, applying a window function over the ^; velérass, the <en to t, CS ... MMA: tranrt.ormapa; O _{'::: teas: sndés::} SOS sn janula, the blocks overlap oonsaoutivos têmporaImente 50% and windowing functions are designed so that the consecutive window squares will add up to one unit. This ensures that when the window function is applied twice to the data it smokes once anallsandó ...... the time domain sign, and one: second view after the synthesis transformation before _: overlapping addition 1, the analysis-plus-synthesis chain without signal modifications is without. losses, [0179] Given 504 overlap between the consecutive structures and a compliance of the 2048 sample structure, the effective temporal resolution is 1024 samples (corresponding to 23.2 ms at the sampling rate of 4 4.1 kHz], that does not is it small enough for two reasons first, it would be desirable to be able to encode a continuous bit stream produced by one: <çOaef icádçr :; 0 n eÁQtu jpadrãttiú ^{:: K} e; seegenddp o ®; of analysis in an improved SAOC index with a resolution
signal remain similar enough on the parameter structure to be expensive with a single defined parameter. The structure length of the standard parameters I found in the standard 3AQC is in 32 QMF structures (lengths up to 72 are allowed by the standard). Similar grouping can be accomplished by using a filter bank with a high frequency resolution. When the signal properties do not change during a parameter structure, the grouping provides the coding efficiency without degradation of quality. However, when the signal properties change within the parameter structure, © grouping induces errors. Standard SAOC allows to define a standard grouping length, which is used with. quasi-stationary signals, but also define the sub-blocks of the parameter ·. Sub-blocks define groupings shorter than the standard length, and parameterization is done on each sub-block separately. As a result of the underlying temporal resolution of the © QMF © Bank, the resulting temporal resolution is 64 time domain samples, which is much finer than the resolution obtained using a high frequency resolution fix © filter. This requirement affects the mediated SÁOC dscodifleader.
[0131] Using a filter bank with one. The large transform provides a good frequency resolution, but the temporal resolution is degraded over time (the so-called uncertainty principle). If the properties of the signal change within a single structure of analysis, the low temporal resolution can tire wear and tear at the output of the synthesis. At 0.10, it would be useful to obtain a temporal resolution of the sub-break in the locations of considerable signal changes. The temporal resolution of the substructure takes naturulmentu at a lower frequency resolution but it is assumed that during a signal change the most important temporal resolution ® aspect® to be captured proclaamenta.
This 8zrgence of the temporal resolution of the sub®strutsra mainly affects up to the
It is only improved by following the decodifleader).
used in
The same principle of the solution can be both to use
1 s of analysis s ta rts when the quasi stationary ((none
t. rans í ® n t e de t e q t a do) and when there are no parameter limits.
When either of the two hoists, use the block switching scheme lengths. An exception to this condition can be made at the parameter limits where the undivided structure groups reside and coincide with the cross bridge between two long windows (while decoding the fluXG COritlnUO 00 bit of the standard 3AOC). u assumed that in this case the signal properties remain sufficiently stationary for the high-resolution filter bank. When
A parameter limit is signaled (from the continuous bit stream or transient® detector}, the structure is adjusted to use a shorter structure length, thus improving the response time® a1 1ocaIly <
[0163]
The first two applications use the same sequence construction mocacism as the underlying window. The window function: the prototype is set to index 0 <ύ <Λ --1 for a length of the window hando '* Drawing the single window # three corner ends are:
necessary, namely, the centers of the previous, current window and the next window, q ,. . . _; , q. _f a q ..
I / '( T 3 (q ~~ q „j)), for 0 <a <q ~ q..j μ (a-2q> q„ _; · q. ^ Liq .., - q)), fbrq -q .., <u <q _,; -¾. ^
The 1st season.11 real race of the window u, antae, | q., <a <iq _{i > with »m ~ [q .., I. The function of the prototype window used in these illustrations is the mono sinusoidal window
[QlBTj Figure 9 d illustrates the principle of the switching scheme of the '' crossing block in the translent. In particular, Figure 1 illustrates the adaptation of the asosene La de j ana lamer is normal to use a poxito from a window in the translent. The ill line represents the time domain signal samples, the vertical, ill line and the location i of the detected transients (or have limit of the citi continue flow parameter, and the lines. 113 illustrate the windowing functions and their time intervals Bst® scheme requires deciding the quartz of the overlapping between the two windows nr and 5¾ around the transients, changing the inclination of the window.When the overlapping is done in a small amount, the windows have their points maximum planes to the transmute and the sections where ctttx & .rail deterioration or the transient The overlapping lengths can also be different before and after the transients In this approach, the two windows or frames around the transient® will be adjusted in length. location of the transient © defines the centers of the windows surrounding eomtr 4 and .., in which i. and / are the overlap length before and after the transient, respectively · nont u With these defined, the above equation can be used.
[ülSSJ A seuux.r, the isolation of the trans lent and, according to an application is described.
fOlSdj Figure 10 illustrates the principle of the switching scheme of the transients isolation block, stunned with an application. Orna short window 1¾ is centered on the transient, and the two flat windows and vç _(i are adjusted to complement the short window. Effectively, the viaiabac windows are limited to the transient location, so the anterior section contains only the sign before the transient ©, and the following window contains only p after the transient ©. In this approach, the transients define the centers
,. . _γ >'1> * pfà £ <Λ V. .t. & S ¹ 7 Q-ΠκλΘ íaj defines O: desired window interval before and after the transient. With these defined, the above equation can be used.
[0110j Below, structure of type AAC, according to c > m. urve <<<: tab: i u ca ca e s cr nra ,, [019'11 The: degrees of freedom of two previous schemes of already merit cannot always be unnecessary. The processing of the differentiation transients is also employed in the field of perceptual audio coding. The objective is to reduce the temporal dispersion of the transient that would tire the so-called pru-eeot. In MPEG-2/4 AAC [AAC], two basic lengths of the window are used: LONGO [one hundred and twenty 2048 samples), and CORTO (with 256 samples), in addition to these two transition windows they are also defined to allow the transition from LONG to SHORT and vice versa. As an additional constraint, CüRTÁS windows are used for serious use in groups of 8 windows. Lassa forma, the step between windows and window groups remains at a constant® value gives 102 4 samples.
[0192] If the SAGC system employs an ARC-based codec for object signals, dovnmix, or residual objects, it would be useful to have a structure diagram that can be easily integrated with the codec. For this; reason, a block switching scheme based on the AAC windows is described.
[0193] Figure 11 describes an example of switching the AAC-type block. In particular, Figure 11 illustrates the same: signal with a transient® is the resulting ACA type sequence. It can be seen that the 'location of the temporal transient® and covered with § CilRTÁS windows, which are surrounded by the transition windows from n to the LONG windows, rod® can be seen from the illustration that the transient itself is not centered in a single window. nor in a cross stitch by ganelar miSAl ^r ·. · - · .104 $.
> wtr between two windows. This is because the locations of the window are fixed to a grid, but the grid guarantees the constant pass at the same time. The resulting rounded temporal error is assumed to be small and sufficient to be perceptibly irrelevant compared to the LONG tired frogs only.
The windows are defined as:
LONG janele: ^::: / (. _{Taal; Vi;),} blushing
SHORT window: w _v .- _{: < ,;} . _y DA - / ¼¾ ^ cora
- The transition window from LONG to ctiita / ') Jbf 6 <fí <^ gtt.
V ny.:. 7.0: í p, _r tlAAl, _} cri <.: ...: .. 0. :: <a '' 2 ~ 4 ⁷ ->: v .4. A, · vy / OvV '. fcr ' ^. ..... 44
Ύ xr, <> λ if _fof td »lffl» <_{; i} , <.4 ~ The transition window from SHORT to LDüGA 'Wzçp (η) ™ - ~ ^{; f} ' r ~ U · [01951 In the following, the implementation variants, according to the applications are described.
[0196; Regardless of the block switching scheme, another design choice is: the length of the actual t / f transform. If the main target is to maintain the following operations: simple frequency domain by the analysis structures, a constant transformation length can be used. The length is set to a large appropriate value, for example, corresponding to the length of the most common structure. since the time domain is shortened than this value, otherwise it is preuuc hu.de · with zeros in the cot'primer, so total - It must be observed even though after the prsenóhlméntc with zero the esoect.ro has. In a larger number of positions, the amount of actual information is not high compared to a short malformation. In this case, the K matrix matrices (b, /, á} have the same dlmensees for all values of á.
(C 197] Another high standard is tra n. S, for max a s t u t ura in windows without p th e not Imentc cs beings> This has a less complex computational complexity than with a constant transformation length <. So, then, the differentiated resolution of the nr-tru the consecutive structures will need to be considered with the matxi ncciac K (5 :, /, X
10192] Then, the hybrid hybrid filtering, according to the application is described, [0190] Another possibility to obtain a higher frequency resolution would be to modify the hybrid filter bank used in the DAGC psdrio to a finer resolution, In SAOC By default, only the three most basic of the S4 QMF bands are passed through the Nyquist filter bank which subdivides the band contents as well.
[0200] Figure 12 illustrates extended QMF hybrid filtering. Nyquiçt filters are repeated for each QNE band separately, and the outputs are combined by a single high resolution spectrum. In particular, Figure 12 illustrates how obtaining a frequency resolution comparable to the bare approach in OFT would require the subdivision of each QMF band to be, for example, 18 sub-bands (requiring complex filtering in 32 sub-bands). The disadvantage of this approach is that the necessary filter prototypes are long due to the narrowing of the bands. This gets tired: processing delay and increases the complexity of the calculation qçmputaelanai>
2010201] An alternative way is to implement extended hybrid filtering by replacing the sets of hyquist filters with efficient filter banks / transformations (for example, DFT zoom, Discrete Cosine Transform, etc.). In addition, the distortion contained in the resulting spectral coefficients of alia resolution, which is caused: by the leakage effects of the first filter stage (here: QMF; <can be substantially reduced by a distortion canceling post-pronament: of the spectral coefficients high resolution images similar to the well-known Layer 3 MFFé-lZ2 hybrid filter stand [FBI [MFFG-IJ.
[Q2Q2] Figure 1b illustrates a decudifloader for generating an audio output signal, comprising one or more channels (audio output isi of a dodnmix signal, comprising a plurality of time daminium.x, dvnmi.x samples I woke up with a brilliant application. The doxamix signal encodes two or mods of the audio object., (3203 (The deocdifleader comprises a first analysis sub-module · 161 to transition the plurality of time-domain samples to obtain a time domain) plurality of sub-band comprising a plurality of love strasde su b-ban da.
[0204] In addition, the decodifloader comprises a sequence generator per window 162 for determining a plurality of analysis windows, wherein each of the analysis windows comprises a plurality of atpsbras of sub · —Οχλδ'.ί ÍCC: ma da ρ.:. u.ra.:. The. dace zdés S: Upr: baudas ^ <ss that Ca Ca gmueaa: ue suai z s- e ísssí _: yaPrarmame oe tanét a> s ote íánqaáíSps has the length of the window indicating the number of sub-band samples in the said analysis window. □ window sequence generator 162 is configured to determine the plurality of analysis windows, for example, based on additional parametric information, so that the window length _: of each of the analysis windows depends on a property of the signal of the at least one of two or more audio object signals.
[0205] In addition, the decoder will buy one.
second analysis module 166 to transform the plurality of subband samples from each analysis window from the plurality of analysis windows, depending on the window fulfillment of said analysis window, to obtain a Ç <0 * · ' 4 · v »[6206] In addition, the decodifloader comprises an unmixed unit 164 for not mixing the transformed dovnmix based on additional parametric information: over two or more signals from the audio object to obtain the audio output signal.
[0207] In other words: the transformation and led cm two phrases:. First: phase in t r s n s f o r m a tion, a pinnality of sub-banners each. an buy and addendum a plurality of masters sub-band iras are
created. Then, on; a second phase, another transformation is carried out. Inter alia., The analysis windows used for the second phase determine the time resolution the resolution gives: resulting transformed frequency.
10208] Figure 13 illustrates an example where windows 1 or 2 are used for the chip _; the transformation. Ü ti 1 ir the short windows leads to a frequency resolution range, but a high resolution time. Using short windows to prune, for example, be appropriate ·, when a transient is present in the signals of the encoded audio object (Q <χ, ·, indicates love behind subband, and v < _: .., · indicates samples of dovamí x transierrado in a time frequency domain .;
[020: 9] Figure 14 illustrates an example where the longer windows are used for the transformation than in the example in Figure 13. Using the long windows leads to a high frequency resolution, but a low time resolution. Employing long windows may, for example, be appropriate when a transient is not present in the encoded audio object's signals. [Again, the: ¾¾ indicates the subband samples, and the vS _xS indicates the transformed dopemix samples: in the time frequency domain.) [02:10] Figure 2 b illustrates a corresponding encoder to encode two or more signals from the incoming audio object, according to an application. Each of two or more signals from the input audio object comprises a plurality of time domain signal samples.
[G211] The Q undid leader comprises a first analysis sub-module 171 for transforming the plurality of time domain signal samples to obtain a plurality of sub-bands comprising a plurality of sub-band samples.
[0212] in addition, the encoder comprises a sequence unit per window 172 to determine a plurality of analysis windows, wherein: each of the analysis windows comprises a plurality of sub-band samples from one of the plurality of samples. baudas, in which each analysis window has a window length indicating the number of subband samples of said analysis window, in which the sequence unit per window 172 is configured to determine the plurality of analysis windows, so that the window length of each of the analysis windows depends on a signal property of at least one of two m plus signals from the incoming audio object. For example, a transient detection unit (optional) 175 can provide information if a transient is present in one of the signals of the input audio object is a window sequence unit 172.
[Q213] Furthermore, the codifleader comprises a second analysis module 173 to transform the plurality of subband samples from each analysis window from the plurality of analysis windows, depending on the window length of said analysis window, to obtain samples from s 1 n a. 1 t r ans decorated.
[0214] In addition, the codifloader coitpreeode an estimate unit takes by fSI 174 to determine the parametric additional information depending on the samples
Pufiauí :: tr u u f · rira o o ·
For example, if high frequency resolution is required and a low time resolution is acceptable, then the second analysis module is turned on.
[7217 i In short, if a high time resolution is required and a resolution box - frequency is acceptable, I will switch off the second analysis module *
ÍQ2181 Figure 1c illustrates a deocdifloader for generating an audio output signal comprising an or. me is audio output channel (s) of a sisal from duunm.ix, de. according to such application * Q deenmix signal encodes one or more s 1 in 1 (i. s} of ofo j s t o; d and á d 1 o <
[0219] The decoder consists of a pure 181 conirole unit defining an Activation Indication in one. activation state, depending on a property of the signal, paio rtonós, one of one or more terminal (s) of the audio object * [0218] In addition, the decoded I comprise a first analysis module 182 to transform the dcomx signal to produce a first transformed dwnnfx, comprising a plurality of primary sub-band channels, [0221] Furthermore, © decodifloader comprises a second modulo cie arrali. if 18.1 stops. generate, when ã .1. activation activation is defined in the activation state, a second disturbed duutmi-ή will freeze costumes from, at least, one of the primary subband bands to obtain a plurality of secondary subband channels, where -the second domnmi Transformed .x comprises the primary subband channels that have not been transformed by the second analysis module and the secondary subband channels.
[0222] In addition, the decodifloader comprises an unmixed unit 184, in which the unmixed unit 184 is configured to not mix the second tbmmmfx ttransformed, when the activation indication is defined in the activation state, based on the information additional garamétrioa on one or more signal [is] of the audio object gare to obtain the audio output signal, and not to mix © úrimairp:;. cost ·, 0 ....- 1 ^ urunsibimagp ,; c-uendo z ndlc—- ça © ç.e <.. 11. vaç .. ·· -. © ^; ution is defined in the activation state, based on additional perametric information about one. or more signal (s) of the cHlCLiQ object pbi.i.VÂ obi € r SlHti.l Q © Q £. ηΓΟΡ.Ι.Ό>
[9221] â Figure 1.5 illustrates an example, where a high and three-th resolution is required for an nss.im time resolution is acceptable. Consequently, the control unit 181 turns on the second analysis module by defining the activation indication to the activation state (for example, transformed by the first analysis module 182 (not shown in Fig. 15} to obtain a first transmaterial domamia, uncle example. pi.o, of the r. 1 u, the transformed uoazlaix has three sub-bands. In the most real application scenarios 1stína, the transformed dbwamfx can, for example, have, for example, 32 or 64 sub-bands. the first transformed downmlx is transformed by the second analysis module 13 3 (not the most race in Figure 15) to obtain a second downs, transitioned fx., for example, in Figure 15, the transitioned dowaatix has nine sub-bands. realistloa application scenarios, the transformed dourmia can, for example, have, for example, 512, 1024 or _: 234 8 and ub - ba ndas. The unmixed one way 184 then will not mix the second transistor dpwnmix to get the signal audio output.
[0224] For example, the unmixed unit 184 may receive the indication of activation of the control unit 151. Or, for example, whenever the unmixed unit 184 receives' .ms according to the transferamm of the second analysis module 183, the unmixed unit 184 concludes that the second transformed dbvnmix must not be mixed; whenever unmixed unit 184 does not receive a second transformed downmix from the second analysis module 183, unmixed unit 184 concludes that the first transformed dounmix must not be mixed.
[02251 Figure 15 illustrates an example, where a high resolution of the table top is required and a low resolution of the milling table is acceptable. Conseuusntementé, the
band that is defined in the sub-band transformation state, at xo ot a and x · a zpiçuraiiXíaédb: ae ídés s <; b ~ oan α a> s and ct 'iaa xti o s, o not to transform each of the sub-band sub-channels, the indication of sub-band transformation that is not: defined in the sub-band transformation state.
[0228] Figure 1'7 illustrates an example, where the control unit 181 is shown in Figure 17} has not defined the indication of sub-wave transformation second ssb-banoa in the sub-band transformation state (for example , defining a boolean variable '' sutbsnd transform indication 2 ^W in subband trãnsfarm_indiuatÍQn_2 - true). Thus, the second analysis module 183 (not shown in Figure 17): transforms the second snb -ba n da »á r> a obtertres η ova ss sb · ba n da sdsf 1 na ~
Λ resolution. In the example in Figure 17, the control unit 131 did not define the subband transformation indication of the first and third subband in the aubband transformation state (for example, this can be indicated by the cantrale unit 181 defining ar baniaanas variable's ebb andtra fox m_ 1 ns na ti on IOA ^w 1 and ^VI indication subband transform subband 3 * t f r n & m__ind.i or cat ~ false indication transform and subband 3 -. false). Thus, the second analysis module '183: does not transform the first to the third sub-bzinda. Instead, the first and third subbands are used as subbands of the second transformed abanezx.
[02233 Figure 13 illustrates an example, in which the indication of the transformation of si & -band of the first second sub-bundle in the state of sub-band transfiguration by axeapÂO, in view of the variable boorsana çubband. transform indication 1 in subband rans form indi ca ti an_l ^:::: troe and, for example, defining the arlóvel. boo leans subband transform indication .2 into subband transf orm_indioation _, 2 ~ true) <Thus, the second analysis module 183 (not rnos trade in Figure 18) transforms the first and the second subband to obtain six new flow sub-bands. resolution. In the example in Figure 18, the control city 181 did not define the indication of ransrovmaçao cia sçb — handa su su texaecra Panda in the subband ransformation station (for example, this can be indicated by the d® ucntrcle 181 unit defining the variable oo 1 eanasubban d_t r an sf or m_ indi ca ti © n_ 3 ^{, f} in s ubb> a nd ransform indication 3 ~ false ;. Thus, the second analysis module 183 does not transform: the third subband. therefore, the third sub-band itself is used as a sub-band of the second of the trans for.nado <anu rx <.
02301 At the end of an application, the first analysis hate 181 is configured to transform elnai and doun.uix to obtain the first dovnmf; <trans formed undertaking the plurality of primary sub-band carias mpregandu wn Filtra em quadrature mirror (ÍQMF).
0231 j In one application, the first analysis module 182 is configured to transform the do nmnix signal depending on a first length of the analysis window, m: that the first length of the analysis fan depends on the retained flow rate. de do a. nax, and / or the second analysis and analysis 183 ê conf iguradG ^: to generate, when the activation indication: it is defined in the activation state, the second dovnmix transformed by the transformation of at least one of the primary subband channels depending on a second length of the analysis window, in which the second length of the analysis window depends on the aforementioned signal property. This application turns on and off with the second analysis module: 183, and to define the length of an analysis window
10232] In an application, the decoder is configured to generate the audio output signal comprising a self: more audio output (s) from the doxnmfx signal, was that the donnrafx signal encodes dots or more signals from the object of audio, Control unit 181 is configured to set the activation indication to the activation state, depending on the signal property of at least one of two or more signals from the audio object. In addition, the unmixed unit 134 and configured not to mix c according to the transformed dobnmfx, however: the activation indication is defined in the activation state, based on the additional perramatric information on one or more signal (s) of the audio object for obtain the audio output signal, ® so as not to mix the first transformed dovnmi.x, when the activation indication is not defined in the activation state, based on additional information for two or more signals from the audio object to get the audio output signal.
[0233] A. figure 2 c: illustrate an encoder to encode mm signal dc input audio object, in color

one of at least first additional transformed audio object signals to obtain a plurality of additional secondary subband channels. In addition, the PSI 114 estimate unit is configured to determine the additional parameterization information based on the plurality of additional secondary aub-band channels, when the atxvsection indication is defined in an activation state, [02431 0 method · and apparatus inventive relieves sv anxa gets pravi lover me rrc a on a da p ro ca ss ame ntooe · S AOG of the state of the art using a fixed filter bank or time-frequency transformation. Better subjective audio quality can; be obtained by dynamically adapting the time / frequency resolution of the transformations or filter banks used to analyze and synthesize the audio objects within SAGC. At the same time, disturbances such as pre- and low-feet caused by lack of temporal precision and disturbances such as auditory feedback and double speech caused by speci c practice 1 insufficiency in time: serr éduiz i da s ide ntr o same ISGC system, more importantly, the improved SAOC system equipped with adaptive transformation maintains, backward compatibility with the standard SAOd still providing a good perceptual quality comparable to the standard SM1C.
[024 «] The applications provide an audio encoder or method to encode the audio or related computer program as described in the above. In addition, the applications provide an audio adapter: or audio deodification method or a built-in buffer program as described above. In addition, the lectures provide an encoded audio signal that can be stored by having stored the conferred encoded audio signal described above.
[02 4Sj Although some aspects have been described in the context of a device, it is evident that these aspects also represent a description of the corresponding method, where a binou or diapositive: corresponds to a stage of the method or a character. of a method step. Similarly, the aspects described in the context of a method step also represent a description of a corresponding block or item or feature of a corresponding device.
[024 6) The decomposed inventive signal can be stored in a digital storage medium or it can be transmitted via a transmission medium, such as a
t ran smi s- are without wire or a middle of t ran confession, such like the Internet. [ü247j Depending on in c e r t a s e x i. g e n c 1 a s in imp .1 love h t a ç, as anlie actions of the invention can to be
implemented in hardware or in software, the implementation can be accomplished using a digital storage honeys, for example, a floppy disk, a: DVD, a CD, a ROM memory, a FROM ,. an EfeOM, an EER ^ QM or a FLASH memory, having electronically readable centrole signals stored
in him they are capable to cooperate) common system of computer program â vé1, give way that the respective · method be realigned.
[0240]
Some applications, according to. the invention, comprise a non-transitory data carrier having electronically readable control signals that are capable of cooperating with a suitable computer system, out of fear that one of the methods: described in this document will be carried out. [0249] In general, the applications of the present invention can be implemented as a product of the computer program with a program code, the program code being operative to perform one. of the guandu methods the product of the · computer program is run on a computer. The program code can, for example, be stored on a machine-readable conveyor.
[Q250J Cutras applications ccmoraendsrn õ òrrrama -computer to perform one of the dnsurites methods in this dosing, stored in a _: machine-readable conveyor.
[õzilj In other words, an application of the inventive method is thus one. computer program having a program code to perform one of the procedures described in this document, when the computer program is run on a computer.
[& 2S.2j Another application of the inventive methods is, therefore, a data carrier (or a digital storage medium, or a computer-readable medium) or the computer program for carrying out nr ·; all described in this document;
[v <c.od,: C'U '-. ra ap.s. íca.ção u an inventive metOoo and therefore a data stream or a. sequence of signals representing the computer orthography to perform one of the methods described in this document. The data stream or signal sequence can, for example, be configured to be transferred over a data communication connection, for example, via the Internet.
[0254] Another application comprises a processing medium, for example, a computer or a programmable logic device, configured for or adapted to perform one of the methods described in this document.
[015: 5] The application comprises a computer having the computer program installed on it to perform one of the methods described in this document, [0254] In some applications, a logic device will program (for example, an arrangement of programmable doors or fields) } can be used to perform some or all of the functionality of the methods described in this document. In some applications, an array of programmable field gates can cooperate with a microprocessor in order to perform one of the methods described in this document »In general, the methods are preferably carried out for any hardware device »[0257] The applications described above are merely: illustrative for the principles of the present invention. It is understood that: changes and variations of the provisions and the details described in the presumed document will be evident to others experts in the art »2 intention of the invention, therefore, to be limited only by the es cup of the attached patent claims and not for the specific details presented in the form of a description and explanation of the upgrades in the present department.
[0258j [BCC] C. Faller end E. Bawçgarte, Binaural
Cine Ceding - Bart XI: Schemes and applications, IEEE Trans, on Speech and Audio Proc., Vol. 11, no. 6, Kov. 2013.
[02 5 9] [J S C J C. Fa 1.1 e r, 2 a cam t r i c So i nt - Cedi ng of
Audio Sources, 120th AES Convention, Paris, 2006.
[0260] [SAOCll u. Herre, S. Disch, 01 Hilpert, Q.
Hellmutht From SAC So SAGS - Hecent bevelopmanto In Parametric Coding of Spatial Audio ;, 22 nd Fog local UK AES Conference, Cambridge, OK, April 2607.
[0281] [SAQC2] J, Engdegàrd, B. Rusch, C. Caleb, O.
Belleuth, u, Hilpert, A · Holzer, 2. Terentiev, Oh Rreebaart, 0. Poppens, E. Schuijers and 2. Goren: Spatial Audio Object Coding {SAGO.} - The Upcoming hPKG Standard, on Parametric Object Based Audi © Ceding , 124th AES Convention, Amsterdam, 2003.
(02 62] rSAQCj I SC / TEC, '' KFEG audio technologies Fart 2: Spatial Audio Object Coding (SAGO), I80 / IEC G1C1 / SC2 9 / KG11 (MSEC) International Standard 2 3033-2: 2010.
[0.2631 [AACj Sosi, Hariha; Erandenburg, Karlheinz;
Quachenbush, Schuyler; Fielder, Louis; Akaglri, Kenzo; Fuchs, Hendrik; Diets, Martin, ISG / IEC HFEG-2 Advanced Audio Coding, u '. Audi © Eng. See, vol in, no .... 1.0.- ,. pp. 783-213 ,. ISST.
[0264] [ISSIJ M. Farvain and L. Girin: Informed
Source Separation of underdetermined instantaneous Stereo Mixtures using Source Index Embedding, IEEE iCASSF, .2010, [02631 [1SS2] M. Farvaix, L. Girin, G.-M. Erosstar:
"and. wat erma r king -based method for .informed source separation of audio signals with a single sensor, IEEE Transactions on Audio, Speech and Language: Processing, 2010.
[0266] [ISS3] A. Liutkus and J. rinel and R. Badeau and L. Girin and G, Richard: '' Informed source separation through spectrogram ceding and data embedding, Signal Processing journal _f 2011.
[0267] [1254] A. Ozerov, A. Ld.utkus, R. Badeau, G.
Richard :: * Informed source separation: sonra.e coding meets source separation, IEEE Workshop on Applications of Signal Proaassing to Audio and Acoustics, 2011.
[.02681 [ISS5] Shuhna Ehang and Laurent Girin: An
X n f o rme d So u roe Separate t i on Sy s t em. f o r B p e e ch Si g n a i s, IETERBPEEÇL, 2011
[u260] [ISSS1 L. Girin and J. Pine 1 :: Informed Audio
Source Separation, from Compressed Linear Stereo Mixtures, AES 12nd international Conference: Semantic Audio, 2011.
[0270I [ISS7] Andrew Eeabit, Emmanuel Pincent, and
Mark. D. Plumbley: Benchmarking flexible adaptive timefrequency transforms for underdotermrned audio source ssparat1on, IEEE International Conference on Acoustics, Speech and Signal Prc french, pp. 27-40, 2008.
[0271] [EE] 8, Edict, Aliasing reduction in subbands of cascaded filterbanks with decimation, Electronic Letters, vol. 26, do, 12 :, pp, 1104-1106, dune 1002.
[027 2] [MPEG-1] ISC / IEC JTG1 /S.Ç20./WG11 MPEG,
International Standard I SC, ⁷ : LEG 111.72, Coding of moving pictures and associated audio for digital, storage media at up

权利要求:
Claims (2)
[1]
1. dm deobdif in «dor to generate an output signal, audio, comprising one or more channels (.is) of the audio output of a -mm signal, characterized by the downmfx signal encoding a single signal (s) of the audio object, in which the compres ride decoder:
a control unit (131 to define an activation indication in an activation state, depending on a signal property, ice menus, one or more signal (s) of the audio object, a first module of analysis (Ill) to transform the dounmí.á signal to obtain a first transformed dooncíx, comprising a plurality of primary sub-band channels, a second oe analysis module (183) to generate: when the <activation indication is defined in the state of atiVigac ,,; a < _: <asgun: pC ' ^: do mmz « _: three a sfo roa of <paia t <ransf 0 <rma.çãoíz of at least unt of sub-ban channels of primary to obtain a plurality of secondary subband channels, wherein the second transformed damnmix comprises the primary subband channels that have not been transformed by the second module: analysis and the secondary subband channels, and one. unmixed unit (184}, where the unmixed unit (184) is configured to do not mix the second trainsfermadé dóònmiz, when the indication of activation is defined in the activation state, based on the additional parametric information in one or male signal (s) of the audio object to obtain the signal. audio output, and no activation no and dertnmia in the o and at.i.vaça ©, with tease in the parametric additional information in a © plus signal (s) of the audio object to obtain the audio output signal.
[2]
2, A decoder. the pain according to claim 1, caractsriiad © gela control unit (181) be configured to define the indication of activation in the BtlHGO QS 3 C i VHÇâ O _f QGpGnOBriGG SB pBJ .. G ΓΓιΒΠΟΒ W; dB V <^ OU s inal (ί s) α g Gb jetg of 'âud'1 o com comprise a t ranei ent.e Indicating a sign change of at least one of one or more sign (s) of the object of audio.

subband formation rans be assigned to each of the channels

configured to set the subband transformation indication for each of the primary subband channels into a subband transformation depending on the

watery analysis module (183) is configured to transform each of the primary subband channels, a: subband transformation indication from which it is set to the subband transformation state, to obtain the plurality of © annals sub-band channels, and to transform the sub-band channels


a second analysis module (193) to generate, when the activation indication is defined in the activation state, w second signal of the audio object transformed by the transformation of at least one of the plurality of primary subband channels to obtain a plurality of secondary sub-channel channels, in which the second signal of the transformed audio object comprises the primary sub-band channels that were not transformed by the second analysis module and the secondary sub-band channels, and a unit of estimation by PSI {194 ;, was that the oil estimation unit {1943 is configured to determine additional parametric information based on the second signal of the transformed audio object, when the activation indication is defined in the activation state, and for determine additional parametric information based on the first signal of the transformed audio object, when the activation indication is not defined in the state activation.
An encoder according to claim 3, characterized in that the control unit (191) is configured to define the activation indication in the activation data, depending on whether the signal of the input audio object comprises a transients indicating a signal change of the input audio object signal.
An encoder in accordance with claim 7 or 8, characterized by the _: indication of sub-band transformation being assigned to each of the primary ssb-band channels, wherein the control unit (191) is configured to define the subband transformation indication of each of the primary subband channels in a subband transformation state depending on the signal properties of the input audio object signal, S:
wherein the second analysis module [193 is configured to transform each of the primary subband channels, the subband transformation indication of which is defined in the subband traffic state, to obtain the plurality of secondary subband channels, and in order not to transform each of the secondary subband channels, the indication of subband transformation that is not defined in the subband transformation state.
An encoder according to any one of claims 7 to 9, characterized by the first analysis module (192) being configured to transform each of the signals of the input audio object employing a fi 11; o in sp p and 1.0 o of w a r d.
11. dim encoder according to any of claims 7 to 10, characterized in that the first analysis module® (192) is configured to transform the signal of the incoming audio object, depending on a first length of the analysis window, where the first length of the analysis window depends on the said signal property, or where the second analysis module (193) is configured to generate, with the indication of activation and

8/10
Urfi GO pX · LíHGx. GO3 S ΧΠ · 3 1 S GG OG Ί OÒG 0'6> 6ΚΚ1Χ Ο 0X 3 Ώ3 £ OX'ífi3QOS additional to drive a plurality of other sub-channels under the scope of 1 us, and in which the estimation unit: per PSI Q.34) is configured to determine additional panametric information with. based on the plurality of additional secured sab-band sab-band channels, when the indication of activation is defined in the activation stay.
13 <dm method for decoding through the generation of an audio output signal, it will comprise one or more audio output channel (s) of a m-rr mrame-r ^ imoa- pemds nmãi signal <de -ddistmixt- .ça.gífméaiíí .gomst or more signals from the Audio object, in which the method comprises:
define an activation indication in an activation state, depending on a signal property of at least one of two or more signals from the audio object, transform the signal from dounmis to obtain a first transformed domnmix, comprising a plurality of primary sub-band channels, stop:,: çmo spí si.nd.inéçád: ca defined in the activation state, a second domnmim transformed by the transformation of, through the menus, one of the primary sub-baud channels to obtain a plurality of sub-band to secondary channels, in that the second transformed dmamlx complies with the primary sub-channel channels that

when the fc® activation indication defined in the activation state, based on additional parametric information about two or more signals from the audio object to obtain the audio output signal and not mix the first transformed downmx, when the indication: de activation is not defined in an activation state, based on additional garametric information about two or more signals from the audio object to obtain the audio output signal.
14. A method for encoding two or more signals from the incoming audio object, characterized by half comprising:
define an activation indication in an activation state, depending on a signal property, through menus, one: gives: two more signals: from the input audio object, transform each of the signals from the input audio object to : obtaining a first signal from the transitional audio object of said signal from the incoming audio object, wherein said first signal from the transformed audio object comprises a plurality of subbeam channels, generate for each of the signals of the input audio object, when the activation indication is defined in the activation state, the second signal of the audio object transformed by the transfer of, by monos, a primary snb-band nu eánáís of the first signal of the transformed audio ubjetó said signal from the intermediary audio object to obtain a plurality of secondary subband channels, one that said second transformed ο.ηη ~ ηχ comprises said reed primary scb-band ions that have not been transformed into a second analysis module and referred to secondary sun-baaoa channels, and of a rainf frame 1 _: ona 1 ram with a case on the second signal object transformed audio of one of the signals of the input audio object, when the activation indication is defined in an activation state, and determine the additional information, parameters based on the diagram: O ::}: <sianaa: οο; íOógetçxipêr iàMsáí t .ransi cama oo c <c mda <: an <ms signals from the incoming audio object, · when a. activation indication is not defined in the activation state.
IS. A computer program to implement a method, according to: claim 13 or 14, aractexirado for being executed in a computer

类似技术:

---

公开号 | 公开日 | 专利标题

---

BR112015007650A2|2019-11-12|ENCODER, DECODER AND METHODS FOR TRANSFORMATION OF SIGNAL DEPENDENT ZOOM ON SPACE AUDIO OBJECT CODING

---

ES2687044T3|2018-10-23|Dynamic range control controlled by metadata

---

ES2398573T3|2013-03-20|Reduced number of channel decoding

---

ES2709755T3|2019-04-17|Stereo decoding of complex prediction based on TCMD

---

JP2018511827A|2018-04-26|Audio encoder for encoding multi-channel signals and audio decoder for decoding encoded audio signals

---

ES2881076T3|2021-11-26|Apparatus and method for efficient encoding of object metadata

---

JP5027799B2|2012-09-19|Adaptive grouping of parameters to improve coding efficiency

---

CN102592600B|2016-08-24|Embedded silence and background noise compression

---

EP1711937B1|2009-10-28|Method and apparatus for time scaling of a signal

---

ES2338918T3|2010-05-13|ADAPTIVE RESIDUAL AUDIO CODING.

---

DE602005002833T2|2008-03-13|COMPENSATION OF MULTICANAL AUDIO ENERGY LOSSES

---

DE602005002942T2|2008-07-24|METHOD FOR DISPLAYING MULTI CHANNEL AUDIO SIGNALS

---

ES2529221T3|2015-02-18|Low delay sound coding that alternates predictive coding and transform coding

---

US8515768B2|2013-08-20|Enhanced audio decoder

---

EP1768451A1|2007-03-28|Acoustic signal encoding device and acoustic signal decoding device

---

KR101456639B1|2014-11-04|Coder using forward aliasing cancellation

---

KR20160033777A|2016-03-28|Audio encoder, audio decoder, methods and computer program using jointly encoded residual signals

---

BR112020011035A2|2020-11-17|apparatus and method for encoding or decoding directional audio encoding parameters using different time-frequency resolutions

---

TW530296B|2003-05-01|Method and apparatus for using coding scheme selection patterns in a predictive speech coder to reduce sensitivity to frame error conditions

---

CN105706165B|2020-07-14|Audio encoder, decoder, encoding and decoding methods using noise padding

---

ES2778076T3|2020-08-07|Method and arrangement for smoothing stationary background noise

---

TW201246187A|2012-11-16|Apparatus and methods to perform huffman coding

---

WO2010130225A1|2010-11-18|Audio decoding method and audio decoder

---

EP2573767B1|2017-08-02|Method and device for updating status of synthesis filters

---

BRPI0921082B1|2020-04-07|audio signal encoding apparatus, encoding method and communication device

同族专利:

---

公开号 | 公开日

---

RU2015116645A|2016-11-27|

---

CA2887028A1|2014-04-10|

---

TW201423729A|2014-06-16|

---

CN105190747B|2019-01-04|

---

CN105190747A|2015-12-23|

---

AR092928A1|2015-05-06|

---

US20150221314A1|2015-08-06|

---

EP2904610B1|2021-05-05|

---

MX351359B|2017-10-11|

---

TWI541795B|2016-07-11|

---

EP2717265A1|2014-04-09|

---

MY178697A|2020-10-20|

---

KR101689489B1|2016-12-23|

---

SG11201502611TA|2015-05-28|

---

AR092929A1|2015-05-06|

---

TWI539444B|2016-06-21|

---

CA2887028C|2018-08-28|

---

WO2014053548A1|2014-04-10|

---

CA2886999A1|2014-04-10|

---

ES2880883T3|2021-11-25|

---

MX2015004018A|2015-07-06|

---

MX2015004019A|2015-07-06|

---

AU2013326526B2|2017-03-02|

---

HK1213361A1|2016-06-30|

---

MX350691B|2017-09-13|

---

CN104798131B|2018-09-25|

---

ES2873977T3|2021-11-04|

---

US10152978B2|2018-12-11|

---

EP2717262A1|2014-04-09|

---

JP6185592B2|2017-08-23|

---

JP2015535960A|2015-12-17|

---

CN104798131A|2015-07-22|

---

RU2639658C2|2017-12-21|

---

WO2014053547A1|2014-04-10|

---

EP2904611A1|2015-08-12|

---

EP2904611B1|2021-06-23|

---

RU2625939C2|2017-07-19|

---

KR101685860B1|2016-12-12|

---

KR20150065852A|2015-06-15|

---

RU2015116287A|2016-11-27|

---

EP2904610A1|2015-08-12|

---

JP2015535959A|2015-12-17|

---

KR20150056875A|2015-05-27|

---

US20150279377A1|2015-10-01|

---

US9734833B2|2017-08-15|

---

CA2886999C|2018-10-23|

---

JP6268180B2|2018-01-24|

---

AU2013326526A1|2015-05-28|

---

TW201419266A|2014-05-16|

引用文献:

---

公开号 | 申请日 | 公开日 | 申请人 | 专利标题

---

---

JP3175446B2|1993-11-29|2001-06-11|ソニー株式会社|Information compression method and device, compressed information decompression method and device, compressed information recording / transmission device, compressed information reproducing device, compressed information receiving device, and recording medium|

---

BR0304540A|2002-04-22|2004-07-20|Koninkl Philips Electronics Nv|Methods for encoding an audio signal, and for decoding an encoded audio signal, encoder for encoding an audio signal, apparatus for providing an audio signal, encoded audio signal, storage medium, and decoder for decoding an audio signal. encoded audio|

---

US7392195B2|2004-03-25|2008-06-24|Dts, Inc.|Lossless multi-channel audio codec|

---

KR100608062B1|2004-08-04|2006-08-02|삼성전자주식회사|Method and apparatus for decoding high frequency of audio data|

---

US7630902B2|2004-09-17|2009-12-08|Digital Rise Technology Co., Ltd.|Apparatus and methods for digital audio coding using codebook application ranges|

---

CN101312041B|2004-09-17|2011-05-11|广州广晟数码技术有限公司|Apparatus and methods for multichannel digital audio coding|

---

EP1906706B1|2005-07-15|2009-11-25|Panasonic Corporation|Audio decoder|

---

US7917358B2|2005-09-30|2011-03-29|Apple Inc.|Transient detection by power weighted average|

---

EP1979897B1|2006-01-19|2013-08-21|LG Electronics Inc.|Method and apparatus for processing a media signal|

---

WO2007110823A1|2006-03-29|2007-10-04|Koninklijke Philips Electronics N.V.|Audio decoding|

---

EP2054875B1|2006-10-16|2011-03-23|Dolby Sweden AB|Enhanced coding and parameter representation of multichannel downmixed object coding|

---

RU2411645C2|2006-10-25|2011-02-10|Фраунхофер-Гезелльшафт Цур Фердерунг Дер Ангевандтен Форшунг Е.Ф.|Device and method to generate values of subbands of sound signal and device and method to generate audio counts of time area|

---

CN101636917B|2007-03-16|2013-07-24|Lg电子株式会社|A method and an apparatus for processing an audio signal|

---

EP3712888A3|2007-03-30|2020-10-28|Electronics and Telecommunications Research Institute|Apparatus and method for coding and decoding multi object audio signal with multi channel|

---

EP2158587A4|2007-06-08|2010-06-02|Lg Electronics Inc|A method and an apparatus for processing an audio signal|

---

EP2144229A1|2008-07-11|2010-01-13|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Efficient use of phase information in audio encoding and decoding|

---

WO2010105695A1|2009-03-20|2010-09-23|Nokia Corporation|Multi channel audio coding|

---

KR101387808B1|2009-04-15|2014-04-21|한국전자통신연구원|Apparatus for high quality multiple audio object coding and decoding using residual coding with variable bitrate|

---

EP2249334A1|2009-05-08|2010-11-10|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Audio format transcoder|

---

MX2011013829A|2009-06-24|2012-03-07|Fraunhofer Ges Forschung|Audio signal decoder, method for decoding an audio signal and computer program using cascaded audio object processing stages.|

---

ES2793958T3|2009-08-14|2020-11-17|Dts Llc|System to adaptively transmit audio objects|

---

KR20110018107A|2009-08-17|2011-02-23|삼성전자주식회사|Residual signal encoding and decoding method and apparatus|

---

TWI431611B|2009-10-20|2014-03-21|Dolby Int Ab|Apparatus for providing an upmix signal representation on the basis of a downmix signal representation, apparatus for providing a bitstream representing a multi-channel audio signal, methods, computer program and bitstream using a distortion control sign|

---

JP5645951B2|2009-11-20|2014-12-24|フラウンホッファー−ゲゼルシャフト ツァ フェルダールング デァ アンゲヴァンテン フォアシュンク エー．ファオ|An apparatus for providing an upmix signal based on a downmix signal representation, an apparatus for providing a bitstream representing a multichannel audio signal, a method, a computer program, and a multi-channel audio signal using linear combination parameters Bitstream|

---

WO2011101708A1|2010-02-17|2011-08-25|Nokia Corporation|Processing of multi-device audio capture|

---

CN102222505B|2010-04-13|2012-12-19|中兴通讯股份有限公司|Hierarchical audio coding and decoding methods and systems and transient signal hierarchical coding and decoding methods|

---

EP2717265A1|2012-10-05|2014-04-09|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Encoder, decoder and methods for backward compatible dynamic adaption of time/frequency resolution in spatial-audio-object-coding|AU3073800A|1999-02-01|2000-08-25|Hitachi Limited|Semiconductor integrated circuit and nonvolatile memory element|

---

JP4530464B2|2000-03-09|2010-08-25|ルネサスエレクトロニクス株式会社|Semiconductor integrated circuit|

---

EP2717265A1|2012-10-05|2014-04-09|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Encoder, decoder and methods for backward compatible dynamic adaption of time/frequency resolution in spatial-audio-object-coding|

---

EP2804176A1|2013-05-13|2014-11-19|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Audio object separation from mixture signal using object-specific time/frequency resolutions|

---

CN110085240A|2013-05-24|2019-08-02|杜比国际公司|The high efficient coding of audio scene including audio object|

---

KR102243395B1|2013-09-05|2021-04-22|한국전자통신연구원|Apparatus for encoding audio signal, apparatus for decoding audio signal, and apparatus for replaying audio signal|

---

US20150100324A1|2013-10-04|2015-04-09|Nvidia Corporation|Audio encoder performance for miracast|

---

CN106409303B|2014-04-29|2019-09-20|华为技术有限公司|Handle the method and apparatus of signal|

---

CN105336335B|2014-07-25|2020-12-08|杜比实验室特许公司|Audio object extraction with sub-band object probability estimation|

---

CN107533845B|2015-02-02|2020-12-22|弗劳恩霍夫应用研究促进协会|Apparatus and method for processing an encoded audio signal|

---

EP3067885A1|2015-03-09|2016-09-14|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Apparatus and method for encoding or decoding a multi-channel signal|

---

WO2017064264A1|2015-10-15|2017-04-20|Huawei Technologies Co., Ltd.|Method and appratus for sinusoidal encoding and decoding|

---

GB2544083B|2015-11-05|2020-05-20|Advanced Risc Mach Ltd|Data stream assembly control|

---

US9640157B1|2015-12-28|2017-05-02|Berggram Development Oy|Latency enhanced note recognition method|

---

US9711121B1|2015-12-28|2017-07-18|Berggram Development Oy|Latency enhanced note recognition method in gaming|

---

WO2017134214A1|2016-02-03|2017-08-10|Dolby International Ab|Efficient format conversion in audio coding|

---

US10210874B2|2017-02-03|2019-02-19|Qualcomm Incorporated|Multi channel coding|

---

CN110447243B|2017-03-06|2021-06-01|杜比国际公司|Method, decoder system, and medium for rendering audio output based on audio data stream|

---

CN108694955B|2017-04-12|2020-11-17|华为技术有限公司|Coding and decoding method and coder and decoder of multi-channel signal|

---

CN110870006A|2017-04-28|2020-03-06|Dts公司|Audio codec window size and time-frequency transform|

---

CN109427337B|2017-08-23|2021-03-30|华为技术有限公司|Method and device for reconstructing a signal during coding of a stereo signal|

---

US10856755B2|2018-03-06|2020-12-08|Ricoh Company, Ltd.|Intelligent parameterization of time-frequency analysis of encephalography signals|

---

TWI658458B|2018-05-17|2019-05-01|張智星|Method for improving the performance of singing voice separation, non-transitory computer readable medium and computer program product thereof|

---

WO2020249815A2|2019-06-14|2020-12-17|Fraunhofer-Gesellschaft zur Förderung der angewandten Forschung e.V.|Parameter encoding and decoding|

法律状态:
2018-11-21| B06F| Objections, documents and/or translations needed after an examination request according [chapter 6.6 patent gazette]|

---

2020-08-25| B06U| Preliminary requirement: requests with searches performed by other patent offices: procedure suspended [chapter 6.21 patent gazette]|

---

2021-09-08| B06A| Patent application procedure suspended [chapter 6.1 patent gazette]|

---

2021-10-13| B350| Update of information on the portal [chapter 15.35 patent gazette]|

---

2022-03-03| B09A| Decision: intention to grant [chapter 9.1 patent gazette]|

优先权:

---

申请号 | 申请日 | 专利标题

---

US201261710133P| true| 2012-10-05|2012-10-05|

---

EP13167487.1A|EP2717262A1|2012-10-05|2013-05-13|Encoder, decoder and methods for signal-dependent zoom-transform in spatial audio object coding|

---

PCT/EP2013/070550|WO2014053547A1|2012-10-05|2013-10-02|Encoder, decoder and methods for signal-dependent zoom-transform in spatial audio object coding|

---