EP2357854B1 - Procédé et dispositif de production de signaux audio binauraux individuellement adaptables - Google Patents

Procédé et dispositif de production de signaux audio binauraux individuellement adaptables Download PDF

Info

Publication number
EP2357854B1
EP2357854B1 EP10197378.2A EP10197378A EP2357854B1 EP 2357854 B1 EP2357854 B1 EP 2357854B1 EP 10197378 A EP10197378 A EP 10197378A EP 2357854 B1 EP2357854 B1 EP 2357854B1
Authority
EP
European Patent Office
Prior art keywords
head
transit time
itd
freed
head position
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
EP10197378.2A
Other languages
German (de)
English (en)
Other versions
EP2357854A1 (fr
Inventor
Alexander Lindau
Rolando Jorgos Estrella Cazuriaga
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Deutsche Telekom AG
Original Assignee
Deutsche Telekom AG
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Deutsche Telekom AG filed Critical Deutsche Telekom AG
Publication of EP2357854A1 publication Critical patent/EP2357854A1/fr
Application granted granted Critical
Publication of EP2357854B1 publication Critical patent/EP2357854B1/fr
Priority to HRP20160279TT priority Critical patent/HRP20160279T1/hr
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04S—STEREOPHONIC SYSTEMS 
    • H04S7/00—Indicating arrangements; Control arrangements, e.g. balance control
    • H04S7/30—Control circuits for electronic adaptation of the sound field
    • H04S7/302—Electronic adaptation of stereophonic sound system to listener position or orientation
    • H04S7/303—Tracking of listener position or orientation
    • H04S7/304—For headphones
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04S—STEREOPHONIC SYSTEMS 
    • H04S2420/00—Techniques used stereophonic systems covered by H04S but not provided for in its groups
    • H04S2420/01—Enhancing the perception of the sound image or of the spatial distribution using head related transfer functions [HRTF's] or equivalents thereof, e.g. interaural time difference [ITD] or interaural level difference [ILD]

Definitions

  • the invention relates to a method and a device for generating individually adjustable binaural audio signals.
  • interaural transit time difference also referred to as ITD or interaural time delay for short
  • ILD interaural level difference
  • the ITD can only be meaningfully evaluated for the localization within a frequency range up to approx. 1500 Hz, above this ambiguities prevent this evaluation and the level difference ILD increasingly determines the perceived sound incidence direction. Both aspects are partly interchangeable with each other (so-called "runtime intensity trading"), from which, for example, benefits the stereophony, which can be implemented as either maturity, level or mixed stereophony.
  • a complete, frequency-dependent description of both the temporal and the spectral information of the sound field is the head-related or outer ear transfer function. It is defined as the ratio of the following transfer functions: On the one hand the transfer function measured on the eardrum of a listener (optionally also on the blocked ear canal input of a Artificial head) and the second of the transfer function of a ball microphone in the center of the head in his absence.
  • This so-called head-related transfer function also called HRIR or HRTF for short, where HRIR stands for head-related impulse response and HRTF for head-related transfer function
  • HRIR head-related impulse response
  • HRTF head-related transfer function
  • a method and apparatus for processing interaural time delay (“Method and Apparatus for Interaural Time Delay Processing in 3D Digital Audio") is disclosed in US Pat US 7,174,229 B1 described. In US 6,795,556 Will Modify Head-Related Transfer Function (HRTFs) described. Methods for measuring HRTFs are used in the EP 0912077 B1 or the DE 69523643 T2 described.
  • BRIRs or BRTFs binaural room impulse responses or transfer functions
  • head-related transfer functions which also include the reverberant room.
  • head-related transfer function By means of such a head-related transfer function, one can produce the acoustic impression of so-called virtual sound sources with high plausibility. Namely, if one filters any reverberant-free audio signal with the HRTF of the left and right ear, during playback, e.g. via headphones a sound event with more or less correct direction and distance impression.
  • the theory of binaural technique is somewhat simplistic in that the perfect reconstruction of the sound pressure time course on the eardrum, which is possible in this way, leads to the actual hearing of an identical auditory event.
  • Each sound path corresponds to a room reflection; by weighting the sound paths with the HRTF of the left and right ear corresponding to the direction of incidence of the sound path and after superposition of all such weighted reflections of the room impulse response results in a binaural spatial impulse response of the virtual space.
  • This room can now be made audible by filtering reverberant audio with this BRIR; then one speaks of "auralization".
  • the BRIRs can be measured directly in-situ.
  • the BRIRs are measured using electroacoustic excitation sources (loudspeakers) and a so-called artificial head torso simulator or robot (also referred to as HATS for short) and head and torso simulator.
  • HATS head torso simulator
  • Such a robot (HATS) allows an automated and spatially fine measurement of the BRIRs for head movements of a listener due to a replica of the multiple degrees of freedom of the head movement (eg 4 rotational, 3 translational) and can achieve a much higher sound quality and proximity to the original.
  • HATS head torso simulator
  • Such a robot (HATS) allows an automated and spatially fine measurement of the BRIRs for head movements of a listener due to a replica of the multiple degrees of freedom of the head movement (eg 4 rotational, 3 translational) and can achieve a much higher sound quality and proximity to the original.
  • the HATS FABIAN is realized by a software-controlled automatable robot consisting of an outer ear-shaped artificial head and a generically-modeled human body. The robot is used to achieve natural sound field influence (diffraction, shading, reflection) as a result of the actual measurement of the sound field, the two Microphone diaphragms on the blocked ear canal.
  • a HAT an improved auralization can be achieved.
  • a further improvement by increasing the plausibility of binaural room acoustic simulation arises when the interactivity of the listener is taken into account, i. when the reaction of the simulation to head movements of the listener is taken into account. It would be desirable if any intentional or unconscious head movement could be compensated for, thus contributing to a plausible and error-free spatial hearing. For this, however, the head-related transfer functions must be present as HRTFs or BRIRs for each head position of the listener to be taken into account in a later auralization (possibly with regard to different translatory and rotational degrees of freedom).
  • the object is achieved by a method having the features of claim 1 and by a device having the features of the independent claim.
  • head-related transfer functions e.g. BRIRs extract the interaural transit times to obtain transit time-free transfer functions and calculate from the extracted transit times travel time differences to be used in a later step along with an individual scaling factor to impose audio signals dependent on binaural synthesis the current head position were generated by means of the runtime-freed transfer functions.
  • the head-related transfer functions are freed from the interaural transit times related to a particular anthropometry (eg, an artificial head) and run-time-freed audio signals are generated by binaural synthesis followed by a time delay that is individually weighted ( Scaling factor), with the appropriate for the respective person or user runtimes acted upon, so that a significantly improved spatial hearing can be achieved.
  • the time delay corresponds to a value calculated from the weighting of the time difference calculated for the current head position and the individual scaling factor.
  • the method is dynamically designed by changing from a previous header position to a change occurring the current head position is adaptively changed the time delay between a first value and a second value by means of a sample rate conversion (SRC).
  • SRC sample rate conversion
  • the sampling rate conversion uses a conversion factor to accelerate or decelerate the time-lapsed audio signal by the conversion factor, and the conversion factor used for the sample rate conversion is determined according to the change from the time delay associated with the previous and current head positions.
  • the method can be used to calculate the runtime-free head-related transfer functions for a plurality of head positions and / or for a plurality of audio signal sources.
  • the plurality of head positions it may be e.g. the resolution of the viewing direction in small angle changes or steps, e.g. 1 degree, act.
  • the number of sources can be very large, generating an audio signal per source and for each signal path (left and right ear canal). These can then be superpositioned for each signal path after the individual weighting (delay delay).
  • the extraction of the interaural transit times from the head-related transfer functions carried out at the beginning of the procedure can be carried out, for example, by means of one of the following methods: onset method, interaural cross-correlation method, frequency-dependent group delay time difference formation, subtraction of the frequency-dependent linearly approximated phase gradient or determination of the excess phase component from division of the Ü functions before and after Hilbert transformation.
  • onset method interaural cross-correlation method
  • frequency-dependent group delay time difference formation subtraction of the frequency-dependent linearly approximated phase gradient or determination of the excess phase component from division of the Ü functions before and after Hilbert transformation.
  • the onset method leads to very good results, which will be described in detail later.
  • the invention advantageously solves the problem that, in auralization procedures, normally the head-related transfer functions HRTFs or BRIRs are always valid only for the anthropometry of a particular individual or for a particular artificial head, thereby eliminating individual differences, e.g. those of the head diameter are not exactly represented by the propagation time information contained in the transfer functions, which means that other listeners, ie "foreign" persons, a more or less strongly distorted perception of localization and - in head movements - the sensation of a naturally non-existent, spatial movement of the audio signals (localization instability) experience.
  • the invention avoids localization errors due to a wrong head diameter (deviation from the artificial head).
  • a side aspect of the chosen approach (runtime exemption and quasi-minimal-phase cross-fading) also result in significant improvements in terms of latency aspects and the audible errors in the cross-fading in head movements.
  • the invention also reduces the normally occurring fading errors ("stuttering"), which arise because during a real-time exchange of the HRTFs or BRIRs filters with runtime offsets are blended into one another.
  • This temporal "missalignment” leads to typical comb filter-like fading artefacts, which appear clearly and disturbingly especially in the case of quasi-stationary contents (in the case of speech applications, eg in the case of vowels, in music, for example, in "string carpets”).
  • the invention reduces these fading errors by the transition of the transfer functions and the insertion of runtime differences in the Binauralsynthese temporally successive and not - as usual - take place at the same time.
  • a second step 120 binaural synthesis is carried out by means of the propagation time-freed transfer functions in order to generate runtime-freed audio signals L 'and R', respectively. This will be explained in more detail on the basis of Fig. 5a / b described. The per se known Binauralsynthese is still based on the Fig. 4a / b described.
  • a real-time synthesis is performed to individually apply a scalable time delay to the first audio signals. This will be even closer to the Fig. 3 such as Fig. 5a / b described.
  • the circuit A comprises a plurality of functional blocks 111 to 117, each having a sub-step of the step sequence 110 (s. Fig. 1 ).
  • the Fig. 2 thus illustrates pre-processing of the impulse response data sets, wherein an almost inaudibly accurate extraction of the ITD from empirical HRTF / BRIR data sets can be achieved by onset detection.
  • normal transfer functions BRIR (alternatively also HRIR) are read from a database. Then an oversampling follows by an amount that allows a more than accurate extraction of the runtimes from the impulse responses (eg 10 times with respect to a common audio sample rate of 44.1 or 48kHz) in block 112. Thereafter, in block 113, the onsets (start the audio signals or data) are found. Subsequently, in block 114, the length of the onset-freed (quasi-minimal-phase) impulse response is determined and applied in a block 115 as a vector.
  • BRIR alternatively also HRIR
  • sub-sampling in block 116 results in block 117 in transfer-term-free transfer functions and the extracted transit times, which are calculated and stored as transit time differences ITD.
  • descriptive metadata records can be added to DSI. That on the basis of Fig. 2 illustrated method uses the onset method for determining the runtime-released impulse responses. This will be discussed later.
  • the interaural transit time difference results as the difference between the transit times of the HRTF and BRIR of the left and right ear. These transit times are again given as the sum of the linear-phase (pure delay component) and the allpass-containing (frequency-dependent phase shift distortion without spectral distortions) the so-called excess phase component.
  • x excess n x linear n + x allpass n
  • the interaural transit time difference ITD is frequency-dependent. However, the proportion that is essential for correct localization ( ⁇ 1500 Hz) is relatively constant and can be extracted more or less artifact-free. As a method is particularly suitable onset detection, as it is based on the Fig. 2 is illustrated. Alternatively, the determination of the excess phase component by using the Hilbert transformation, frequency-domain-specific phase gradient matching, maximum of the interaural cross-correlation or the frequency-domain-specific determination of the interaural group delay difference is also suitable.
  • the onset method is accurate enough (compare cross-correlation methods), robust enough (compare phase gradient methods), applicable (compare group delay time difference method) and true to tone color (compare Hilbert method). Intrinsically conditioned, the onset method also conserves possible allpass components of the BRIRs in an advantageous manner; they are not lost, but remain in the runtime-free spectra, which are therefore referred to here as quasi-minimal phase.
  • quasi-minimal phase In formal and criteria-free listening experiments it was confirmed that in the resynthesis of the extracted transit times and the quasi-minimal-phase spectra, as expected, no localization errors occur even with contralateral sound incidence. Further formal listening tests showed that the changes in the reverberation structure due to the Hilbert transformation are audible in every case.
  • the head-related transfer functions are freed from the terms.
  • both can then be fed separately to the resynthesis, with a scalable and tailored to the individual resynthesis can be performed (s. Fig. 3 and Fig. 5a / b ).
  • Advantages of this are latency reduction and a shortening of the HRTFs to be kept (if these, as previously implicitly implemented as FIR filters) are just the extracted runtime.
  • the HRTFs can also be generated as infinite impulse response (IIR) filters, either by modeling / estimating measured HRTFs, which are always compulsory as FIR (finite impulse response) filters, or by parametric modeling of essential features.
  • IIR infinite impulse response
  • the method described here is applicable in principle to any head-related transfer functions generated. However, it is described here using the example of empirical HRTF / BRIR data sets present as FIR filters, as described, for example, in US Pat. with the robot mentioned at the outset (HATS FABIAN, see Lindau et al., 2007).
  • the algorithms of the invention described below relate by way of example to the use of BRIR data sets that can be obtained with such a robot.
  • the method is not limited to these data sets, but applicable to any auralization that realize spatial sound localization by filtering head related impulse responses with audio signals.
  • the method 100 includes in a preprocessing step 110 (see FIG. Fig. 1 as well as subblocks in Fig. 2 ) and a real-time resynthesis step 130 (see FIG. Fig. 1 and also Fig. 3 ), which corresponds to a dynamic binaural synthesis algorithm or step 120 (see FIG. Fig. 1 and also Fig. 4a / b ) (see Fig. 5a / b ).
  • the onset method is the most suitable method among extraction methods.
  • the other methods showed less robustness in empirical data sets of binaural room impulse responses.
  • the Hilbert method also seems to be unsuitable, since it changes due to the inherent energy compaction in the direction of the beginning of the impulse response, the contained reflection structure of the room sound field in any audible extent.
  • the inventors were able to make this plausible on the basis of auditor model numerical preliminary tests and a formal listening test.
  • the onset method is set so that the natural measurement background noise (typically about -50 to -90 dB relative to the magnitude maximum value of the impulse response) is determined and then a threshold is chosen well above it (eg 15 dB higher, ie -35dB rel ).
  • the impulse response data set is then searched by machine and in each case from the beginning of the impulse responses and calculated on the basis of the times of crossing the threshold criterion in the left and right channels of the HRTF / BRIR by subtraction of the ITD (see formula 3).
  • the runtimes are removed and the now runtime-free impulse responses are saved again (block 117).
  • the onset method is applied to the 10-times oversampled time signal, thus obtaining a discretization of the ITD in 2.3 ⁇ s steps. This resolution is about one fifth of the ITD threshold that is just noticeable.
  • the thus extracted runtime or time difference ITD can now be in a text-based List format (eg * .txt, * .csv, * .xml) machine-readable to the run-time freed record.
  • This processing is performed in a first circuit A, which is part of the device (see FIG. Fig. 5a / b ).
  • jack audio server In order to modify the convolution process or to customize the runtime, it is preferable to use a so-called "jack audio server” architecture and thus implement an independent plug-in.
  • Functionally identical solutions can also be realized, for example, from arrangements of special DSP hardware or by means of methods in the context of the VST plug-in architecture (ie based on the VST interface).
  • the computer-controlled device eg PC
  • it After starting the computer-controlled device (eg PC), it reads in a configuration file, a record description file and then the text-based list of the ITDs of the BRIR data record that is currently auralized by the classical convolution process.
  • the missing transit time difference can now be inserted as head position-specific delay time VDL and without audible artifacts in one of the two audio channels.
  • the individualization process reads the head movement data as a data stream of the head tracker HTDAT (eg via an IP-based transport protocol), which it also sends to the folding process as before.
  • the latter During the initialization of the individualization process, the latter must realize the first effective interaural transit time difference, determined by the initial head position, by a time-delayed or anticipatory playback by a fixed amount.
  • the simplicity wg. the first time difference can be assumed to be 0, and the first conversion factor can be assumed to be 1, for example.
  • the fractional ratio formation of the audio block length with the audio block length corrected by the amount of change in the time differences results in a ratio that can be used as a conversion factor in a real-time sample rate conversion algorithm of the highest audio quality. This achieves the adaptation of the changed delay values by a conversion factor that is accelerated or delayed by the conversion factor, in which it interpolates from the present signal new samples at other times corresponding to a higher or lower sampling rate and outputs them instead.
  • the table size of the interpolator low pass can be estimated by suitable formulas. It uses a high-quality, band-limited floating-point interpolator based on an analytically described sinc function with a worst-case signal-to-noise ratio of 97 dB and a bandwidth of 97%.
  • the actual conversion factor per audio block to be processed can be determined in each case using the difference between the ITD belonging to the previous and the current head position.
  • An always available software library allows an inaudible and continuous change of the conversion factor, so that the respective head position corresponding delay difference can be resynthesized correctly.
  • sampling rate conversion ratio fs new / fs old ( ⁇ 1 or> 1)
  • sample rate conversion requires fewer or more samples than the underlying current block size.
  • Another approach is based on a prediction of the individual correction value based on an anthropometric measure. Preliminary examinations were performed with several subjects. In this case, a listening test was performed according to the above-mentioned acoustic scenery; In addition to the virtual source, however, the subjects were able to hear the real sound source. The goal was to change the conversion factor To set the simulation so that when switching between simulation and reality found the best possible match. This experiment was repeated 10 times per subject. In addition, four measures of the head considered appropriate were taken by each person. Next, the prediction of the individual scaling factor averages from the head dimensions was checked by multiple linear regression.
  • the variability of the scaling factor within the sample was ⁇ 4%, ie in the non-individualized case a worst-case error of up to 8% of the ITD could have occurred.
  • the individual forecast halves or quarters (at best) this error.
  • An error of 1.25% of the ITD corresponds to a localization error of just over 1 ° and is thus (again: in the most favorable case) already almost in the range of the currently perceivable change in the local salience.
  • the invention makes it possible to achieve numerous improvements, such as the advantage of error-free cross-fading, the latency minimization, the Doppler effect. Also worth mentioning is the choice of band-limited interpolation for sample rate conversion during generation the variable ITD, the real-time capability of the ITD manipulation, the inaudible extraction method and the plug-in architecture. Likewise, important aspects, such as the effective and separate reduction and interpolation of ITD and ILD (ie HRTF / BRIR spectra), should be emphasized.
  • the individual adaptation of the runtime-freed ITDs to the respective individual can be done by scaling the respective current conversion factor in the context of a sample rate conversion, which enables dynamic adaptation when the head position is changed. If a static condition occurs, i. If the head position does not change, a static adaptation of the ITDs (automatic) is also possible. Or customization is done by scaling the ITDs associated with the previous and current head positions before calculating the (unscaled) conversion factor.
  • the individualization and adaptation of the transit time difference ITD * resulting between the output signals L * and R * can be achieved by scaling the respective current conversion factor (application of the factor ISF to the block VDL / SRC; Fig. 5a / b ).
  • the individualization and adaptation of the transit time difference ITD * resulting between the output signals L * and R * can be achieved by scaling the transit time differences (ITD) corresponding to the travel-time-free head-related transfer functions (xBIR *).
  • the invention can be used in many applications.
  • the proposed method can be a substantial improvement of all existing real-time applications for binaural (room) acoustic simulation (3D Virtual Auditory Displays in General, Spatial Acoustics in Computer Games, Virtual Chat Rooms, Binaural Guidance & Alerting Systems, Binaural Walkthroughs Through virtual architecture or through multimodal media shows).
  • Possible applications are the subsequent addition of commercial 3D audio APIs.
  • the receiver side a multi-media PC presuppose such.
  • binaural teleconferencing via VoIP or binaural streaming of live concerts can gain significantly in perceptible quality.

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Acoustics & Sound (AREA)
  • Signal Processing (AREA)
  • Stereophonic System (AREA)

Claims (11)

  1. Une méthode (100) pour générer des signaux audio binauraux individuellement adaptables (L*, R*), comprenant les étapes suivantes :
    le calcul, à partir de fonctions de transfert de tête (xRIR) associées à diverses positions de la tête (1...k), de fonctions de transfert de tête (xRIR*) dépourvus de temps de transit grâce à l'extraction de temps de transit interaural (TL, TR) et le calcul des différences de temps de fonctionnement (ITD) de transit correspondants (étape 110);
    la génération, au moyen d'une synthèse binaurale, de signaux audio (L', R') dépourvus de temps de transit, chacun relatif à un chemin de signal binaural, dans laquelle, en fonction des données de position de tête (HTDAT), les fonctions de transfert de tête (xRIR*) dépourvues de temps de transit sont utilisées pour une position actuelle de la tête (1... k) (étape 120);
    caractérisée en ce que
    les signaux audio binauraux individuellement adaptables (L*, R*) sont générés à partir des signaux audio (L', R') dépourvus de temps de transit en chargeant pour au moins un chemin de signal le signal audio généré (L') dépourvu de temps de transit un délai (VDL) fonction de la différence de temps de transit (ITD) calculée pour la position de tête réelle (1...k) et en fonction d'un facteur d'échelle individuel (ISF) (étape 130);
    lorsque intervient un changement depuis une position précédente de tête (k=1) vers la position réelle de tête (k=2), le délai (VDL) entre une première valeur et une seconde valeur est changée de manière adaptative au moyen d'une conversion de taux d'échantillonnage (SRC), dans laquelle la conversion de taux d'échantillonnage applique un facteur de conversion pour reproduire le signal audio dépourvu de temps de transit en avance ou en retard (étape 130) ; et
    le facteur de conversion utilisé pour la conversion du taux d'échantillonnage (SRC) est déterminé en fonction du changement de délai correspondant au changement de la position de tête précédente (k=1) vers la position de tête actuelle (k=2) (étape 130).
  2. La méthode (100) de la revendication 1, caractérisée en ce que le délai (VDL) correspond à une valeur qui est calculée à partir d'une pondération de la différence de temps de transit (ITD) calculé pour la position de tête actuelle (1...k) et pour le facteur d'échelle individuel (ISF) (étape 130).
  3. La méthode (100) de la revendication 1, caractérisée en ce que une différence de temps de transit (ITD*) s'accroissant entre des signaux audio binauraux individuellement adaptables (L*, R*) est adaptée individuellement par la mise à l'échelle du facteur de conversion actuel (étape 130).
  4. La méthode (100) de la revendication 3, caractérisée en ce que l'accroissement de la différence de temps de transit (ITD*) est adapté individuellement par la mise à l'échelle des différences de temps de transit (ITD) correspondant aux fonctions de transfert de tête (xRIR*) dépourvue de temps de transit.
  5. La méthode (100) de l'une des revendications précédentes, caractérisée en ce que les fonctions de transfert de tête (xRIR*) dépourvues de temps de transit sont calculées pour une pluralité de positions de tête (1...k) et pour une pluralité de sources de signal audio (1...n) (étape 110).
  6. La méthode (100) de l'une des revendications précédentes, caractérisée en ce que les temps de transit interauraux (TL, TR) sont extraits des fonctions de transfert de tête (xRIR) au moyens d'une des méthodes suivantes : méthode de détermination ; méthode de corrélation traverse interaurale ; par construction de différence (temps de transit) de délai de groupe spécifique de fréquences ; par correspondance de gradient de phase spécifique de gamme de fréquences ou par détermination de phase en excès au moyen d'une transformation Hilbert (étape 110).
  7. La méthode (100) de la revendication 6, caractérisée en ce que la méthode de détermination est appliquée de telle façon qu'un plancher de bruit de mesure naturelle est déterminée et puis est déterminé un seuil situé au-dessus qui reste inférieur au maximum absolu de la réponse impulsionnelle résultant de la fonction de transfert de tête correspondante (étape 110).
  8. La méthode (100) de la revendication 7, caractérisée en ce que le seuil est déterminé pour être au moins 10dB au dessus du plancher de bruit de mesure naturelle et/ou le seuil est déterminé pour être au moins 10dB plus bas que le maximum absolu de la réponse impulsionnelle résultant de la fonction de transfert de tête correspondante (étape 110).
  9. La méthode (100) de la revendication 8, caractérisée en ce que le seuil est dans une gamme moyenne de la dynamique de mesure (étape 110).
  10. La méthode (100) de la revendication 6, caractérisée en ce que la méthode de détermination est appliquée de manière à ce qu'une valeur en pourcentage du maximum absolu de la réponse impulsionnelle résultant d'une fonction de transfert de tête correspondante, est détectée pour être fixée, en particulier comme valeur entre 10% et 90% (étape 110).
  11. Dispositif pour mettre en oeuvre une méthode suivant l'une des revendications précédentes, dans lequel le dispositif pour générer des signaux audio binauraux (L*, R*) pour une perception spatiale, comporte :
    un premier circuit (A) pour extraire à partir des fonctions de transfert de tête (xRIR) qui sont associées à diverses positions de tête (1...k) des fonctions de transfert de tête (xRiR*) dépourvues de temps de transit, et pour en calculer des différences de temps de transit (ITD) ;
    un second circuit (BB) pour générer, à partir de signaux audio de synthèse binauraux (L', R') dépourvus de temps de transit, chacun étant relatif à un chemin de signal binaural, dans lequel ce circuit utilise les fonctions de transfert de tête (xRIR*), dépourvues de temps de transit, en fonction de données de position de tête (HTDAT) et ce pour une position de tête actuelle (1...k) ;
    caractérisé par
    au moins un troisième circuit (BA*, C*) pour générer, à partir des signaux audio (L', R') dépourvus de temps de transit, les signaux audio binauraux individuellement adaptables (L*,R*) en chargeant pour au moins un chemin de signal le signal audio généré (L') dépourvu de temps de transit, avec un délai temporel (VDL) en fonction de la différence de temps de transit (ITD) calculé pour la position de tête courante (1...k) et en fonction d'un facteur d'échelle individuel (ISF) ;
    dans lequel le troisième circuit au moins (BA*; C*) change de manière adaptative le délai temporel (VDL) entre une première valeur et une seconde valeur au moyen d'une conversion du taux d'échantillonnage lorsque intervient un changement de la position de tête précédente (k=1) vers la position de tête actuelle (k=2), dans lequel la conversion de taux d'échantillonnage applique un facteur de conversion pour reproduire en avance ou en retard le signal audio dépourvu de temps de transit ; et
    détermine le facteur de conversion utilisé pour la conversion du taux d'échantillonnage (SRC) en fonction du changement de délai temporel correspondant au changement de la position de tête précédente vers la position de tête actuelle.
EP10197378.2A 2010-01-07 2010-12-30 Procédé et dispositif de production de signaux audio binauraux individuellement adaptables Active EP2357854B1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
HRP20160279TT HRP20160279T1 (hr) 2010-01-07 2016-03-18 Postupak i uređaj za generiranje individualno prilagodljivog binauralnog audio signala

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
DE102010004171 2010-01-07

Publications (2)

Publication Number Publication Date
EP2357854A1 EP2357854A1 (fr) 2011-08-17
EP2357854B1 true EP2357854B1 (fr) 2016-03-09

Family

ID=43759806

Family Applications (1)

Application Number Title Priority Date Filing Date
EP10197378.2A Active EP2357854B1 (fr) 2010-01-07 2010-12-30 Procédé et dispositif de production de signaux audio binauraux individuellement adaptables

Country Status (5)

Country Link
EP (1) EP2357854B1 (fr)
ES (1) ES2571044T3 (fr)
HR (1) HRP20160279T1 (fr)
HU (1) HUE028661T2 (fr)
PL (1) PL2357854T3 (fr)

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105900457B (zh) 2014-01-03 2017-08-15 杜比实验室特许公司 用于设计和应用数值优化的双耳房间脉冲响应的方法和***
EP3473022B1 (fr) 2016-06-21 2021-03-17 Dolby Laboratories Licensing Corporation Suivi de tête pour système audio binaural pré-rendu
US9848273B1 (en) 2016-10-21 2017-12-19 Starkey Laboratories, Inc. Head related transfer function individualization for hearing device
GB2601805A (en) * 2020-12-11 2022-06-15 Nokia Technologies Oy Apparatus, Methods and Computer Programs for Providing Spatial Audio
CN113821190B (zh) * 2021-11-25 2022-03-15 广州酷狗计算机科技有限公司 音频播放方法、装置、设备及存储介质

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
ATE208120T1 (de) 1994-02-25 2001-11-15 Henrik Moller Binaurale synthese, kopfbezogene übertragungsfunktion, und ihre verwendung
US7174229B1 (en) 1998-11-13 2007-02-06 Agere Systems Inc. Method and apparatus for processing interaural time delay in 3D digital audio
GB2351213B (en) 1999-05-29 2003-08-27 Central Research Lab Ltd A method of modifying one or more original head related transfer functions
GB2369976A (en) 2000-12-06 2002-06-12 Central Research Lab Ltd A method of synthesising an averaged diffuse-field head-related transfer function
GB0419346D0 (en) * 2004-09-01 2004-09-29 Smyth Stephen M F Method and apparatus for improved headphone virtualisation

Also Published As

Publication number Publication date
PL2357854T3 (pl) 2016-09-30
HRP20160279T1 (hr) 2016-04-22
EP2357854A1 (fr) 2011-08-17
HUE028661T2 (en) 2016-12-28
ES2571044T3 (es) 2016-05-23

Similar Documents

Publication Publication Date Title
DE69522971T2 (de) Binaurale Synthese, kopfbezogene Übertragungsfunktion, und ihre Verwendung
EP3069530B1 (fr) Procédé et dispositif pour compresser et décompresser des données de champ sonore d'un domaine
DE4328620C1 (de) Verfahren zur Simulation eines Raum- und/oder Klangeindrucks
DE69319456T2 (de) Schallfeldsteuerungssystem
DE102006050068B4 (de) Vorrichtung und Verfahren zum Erzeugen eines Umgebungssignals aus einem Audiosignal, Vorrichtung und Verfahren zum Ableiten eines Mehrkanal-Audiosignals aus einem Audiosignal und Computerprogramm
DE60304358T2 (de) Verfahren zur verarbeitung von audiodateien und erfassungsvorrichtung zur anwendung davon
DE69517192T2 (de) Verfahren und vorrichtung zur effizienten darstellung von dreidimensionalen tonsignalen hoher qualitat
EP1671516B1 (fr) Procede et dispositif de production d'un canal a frequences basses
EP1576847B1 (fr) Systeme de restitution audio et procede de restitution d'un signal audio
EP2357854B1 (fr) Procédé et dispositif de production de signaux audio binauraux individuellement adaptables
Lindau et al. Individualization of dynamic binaural synthesis by real time manipulation of ITD
WO2014187877A2 (fr) Pupitre de mixage, générateur de signal audio, procédé et programme informatique de fourniture d'un signal audio
EP1637012A1 (fr) Dispositif de synthese de champ electromagnetique et procede d'actionnement d'un reseau de haut-parleurs
AT394650B (de) Elektroakustische anordnung zur wiedergabe stereophoner binauraler audiosignale ueber kopfhoerer
EP2891334B1 (fr) Génération de son multivoies à partir de signaux audio stéréo
EP3044972B1 (fr) Dispositif et procédé de décorrélation de signaux de haut-parleurs
DE10321980B4 (de) Vorrichtung und Verfahren zum Berechnen eines diskreten Werts einer Komponente in einem Lautsprechersignal
EP1972181A1 (fr) Dispositif et procédé de simulation de systèmes wfs et de compensation de propriétés wfs influençant le son
EP1471770B1 (fr) Procédé de géneration d'une fonction de transfert partielle approximée
DE112006002548T5 (de) Vorrichtung und Verfahren zur Wiedergabe von virtuellem Zweikanal-Ton
EP1123638B1 (fr) Procede et dispositif pour evaluer la qualite de signaux audio a canaux multiples
DE102011003450A1 (de) Erzeugung von benutzerangepassten Signalverarbeitungsparametern
DE102019217189A1 (de) Wellenfeldverarbeitungsverfahren
EP2503799B1 (fr) Procédé et système de calcul de fonctions HRTF par synthèse locale virtuelle de champ sonore
AT413010B (de) Vorrichtung zur herleitung von räumlichen schallsignalen

Legal Events

Date Code Title Description
PUAI Public reference made under article 153(3) epc to a published international application that has entered the european phase

Free format text: ORIGINAL CODE: 0009012

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

AX Request for extension of the european patent

Extension state: BA ME

17P Request for examination filed

Effective date: 20120216

17Q First examination report despatched

Effective date: 20140103

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

INTG Intention to grant announced

Effective date: 20150922

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

RAP1 Party data changed (applicant data changed or rights of an application transferred)

Owner name: DEUTSCHE TELEKOM AG

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

Free format text: NOT ENGLISH

REG Reference to a national code

Ref country code: AT

Ref legal event code: REF

Ref document number: 780258

Country of ref document: AT

Kind code of ref document: T

Effective date: 20160315

Ref country code: CH

Ref legal event code: EP

REG Reference to a national code

Ref country code: HR

Ref legal event code: TUEP

Ref document number: P20160279

Country of ref document: HR

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

Free format text: LANGUAGE OF EP DOCUMENT: GERMAN

REG Reference to a national code

Ref country code: DE

Ref legal event code: R096

Ref document number: 502010011179

Country of ref document: DE

REG Reference to a national code

Ref country code: HR

Ref legal event code: T1PR

Ref document number: P20160279

Country of ref document: HR

REG Reference to a national code

Ref country code: SE

Ref legal event code: TRGR

REG Reference to a national code

Ref country code: NO

Ref legal event code: T2

Effective date: 20160309

REG Reference to a national code

Ref country code: ES

Ref legal event code: FG2A

Ref document number: 2571044

Country of ref document: ES

Kind code of ref document: T3

Effective date: 20160523

REG Reference to a national code

Ref country code: NL

Ref legal event code: FP

REG Reference to a national code

Ref country code: LT

Ref legal event code: MG4D

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: FI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

Ref country code: GR

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160610

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: RS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

Ref country code: LT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

Ref country code: LV

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

REG Reference to a national code

Ref country code: SK

Ref legal event code: T3

Ref document number: E 21058

Country of ref document: SK

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: EE

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

Ref country code: IS

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160709

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: PT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160711

Ref country code: SM

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

Ref country code: RO

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

REG Reference to a national code

Ref country code: DE

Ref legal event code: R097

Ref document number: 502010011179

Country of ref document: DE

REG Reference to a national code

Ref country code: FR

Ref legal event code: PLFP

Year of fee payment: 7

REG Reference to a national code

Ref country code: HU

Ref legal event code: AG4A

Ref document number: E028661

Country of ref document: HU

PLBE No opposition filed within time limit

Free format text: ORIGINAL CODE: 0009261

STAA Information on the status of an ep patent application or granted ep patent

Free format text: STATUS: NO OPPOSITION FILED WITHIN TIME LIMIT

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: DK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

26N No opposition filed

Effective date: 20161212

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: SI

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MC

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

REG Reference to a national code

Ref country code: IE

Ref legal event code: MM4A

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: LU

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20161230

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: IE

Free format text: LAPSE BECAUSE OF NON-PAYMENT OF DUE FEES

Effective date: 20161230

REG Reference to a national code

Ref country code: FR

Ref legal event code: PLFP

Year of fee payment: 8

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: CY

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MK

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: MT

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

PG25 Lapsed in a contracting state [announced via postgrant information from national office to epo]

Ref country code: AL

Free format text: LAPSE BECAUSE OF FAILURE TO SUBMIT A TRANSLATION OF THE DESCRIPTION OR TO PAY THE FEE WITHIN THE PRESCRIBED TIME-LIMIT

Effective date: 20160309

REG Reference to a national code

Ref country code: HR

Ref legal event code: ODRP

Ref document number: P20160279

Country of ref document: HR

Payment date: 20181220

Year of fee payment: 9

REG Reference to a national code

Ref country code: HR

Ref legal event code: ODRP

Ref document number: P20160279

Country of ref document: HR

Payment date: 20191220

Year of fee payment: 10

REG Reference to a national code

Ref country code: HR

Ref legal event code: ODRP

Ref document number: P20160279

Country of ref document: HR

Payment date: 20201221

Year of fee payment: 11

REG Reference to a national code

Ref country code: HR

Ref legal event code: ODRP

Ref document number: P20160279

Country of ref document: HR

Payment date: 20211223

Year of fee payment: 12

REG Reference to a national code

Ref country code: HR

Ref legal event code: ODRP

Ref document number: P20160279

Country of ref document: HR

Payment date: 20221227

Year of fee payment: 13

REG Reference to a national code

Ref country code: HR

Ref legal event code: ODRP

Ref document number: P20160279

Country of ref document: HR

Payment date: 20231227

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: SK

Payment date: 20231213

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: GB

Payment date: 20231212

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: SE

Payment date: 20231212

Year of fee payment: 14

Ref country code: NO

Payment date: 20231218

Year of fee payment: 14

Ref country code: NL

Payment date: 20231219

Year of fee payment: 14

Ref country code: HU

Payment date: 20231231

Year of fee payment: 14

Ref country code: HR

Payment date: 20231227

Year of fee payment: 14

Ref country code: FR

Payment date: 20231212

Year of fee payment: 14

Ref country code: DE

Payment date: 20231212

Year of fee payment: 14

Ref country code: CZ

Payment date: 20231213

Year of fee payment: 14

Ref country code: BG

Payment date: 20231212

Year of fee payment: 14

Ref country code: AT

Payment date: 20231213

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: PL

Payment date: 20231213

Year of fee payment: 14

Ref country code: BE

Payment date: 20231218

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: ES

Payment date: 20240118

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: CH

Payment date: 20240110

Year of fee payment: 14

PGFP Annual fee paid to national office [announced via postgrant information from national office to epo]

Ref country code: TR

Payment date: 20231221

Year of fee payment: 14

Ref country code: IT

Payment date: 20231229

Year of fee payment: 14