EP3381204B1 - Method and device for estimating sound reverberation - Google Patents

Description

FIELD OF THE INVENTION

The present invention relates to methods and devices for estimating acoustic reverberation.

BACKGROUND OF THE INVENTION

Estimating the acoustic reverberation of a medium is essential for capturing acoustic signals such as speech in a reverberant medium such as, for example, a room in a building.

When a sound is emitted and picked up by a microphone in a reverberant environment, the microphone picks up not only the received signal directly, but also reverberant signals in the medium.

This reverberation is reflected by the impulse response of the medium, from which various known parameters, including the reverberation time, flow. The impulse response can be measured directly by emitting an acoustic pulse in the medium, but this method is cumbersome and difficult to envisage for performing repeated measurements while one or more speakers are speaking in the room.

The reverberation time can be estimated blindly, for example while one or more speakers are speaking. The most common parameter used to represent the reverberation time is the 60 dB RT ₆₀ reverberation time.

For example, the document US2014169575 describes a method for blind estimation of the reverberation time in a room.

However, the reverberation time is not representative of the distance between the transmitter and the microphone, which however has a significant impact on the reverb level. The aforementioned known methods of the type do not therefore make it possible to process the acoustic signals sensed satisfactorily.

• JIMI Y C WEN ET AL: "Blind estimation of reverberation time based on the distribution of signal decay rates", IEEE ICASSP 2008, p. 329-332 .
• FALK T H ET AL: "Temporal Dynamics for Blind Measurement of Room Acoustical Parameters",IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, vol. 59, no. 4, 2010, p. 978-989 .
• EATON JAMES ET AL: "Noise-robust reverberation time estimation using spectral decay distributions with reduced computational cost",2013 IEEE ICASSP, 2013, p. 161-165 .

The following documents are part of the state of the art related to the present invention:

• JIMI YC WEN AND AL: "Blind estimation of reverberation time based on the distribution of signal decay rates", IEEE ICASSP 2008, p. 329-332 .
• FALK TH ET AL: "Temporal Dynamics for Measurement of Room Acoustical Parameters", IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT, vol. 59, no. 4, 2010, p. 978-989 .
• EATON JAMES ET AL: "Noise-robust reverberation time estimation using spectral decay distributions with reduced computational cost," 2013 IEEE ICASSP, 2013, p. 161-165 .

OBJECTS AND SUMMARY OF THE INVENTION

The present invention therefore aims to provide a method for estimating acoustic reverberation, which avoids this disadvantage.

1. (a) une étape de mesure dans laquelle on capte au moins un signal acoustique dans le milieu,
2. (b) une étape d'observation au cours de laquelle on détermine une distribution de taux de décroissance d'énergie acoustique à partir du signal acoustique capté à l'étape (a) et on détermine la fonction caractéristique de la distribution de taux de décroissance d'énergie acoustique,
3. (c) une étape d'estimation au cours de laquelle on estime un temps de réverbération caractéristique et un niveau de réverbération caractéristique du son dans le milieu à partir de données représentatives de la distribution de taux de décroissance d'énergie acoustique déterminée à l'étape (b), la régression étant faite en référence à :
   • des fonctions caractéristiques de référence représentatives respectivement de plusieurs distributions de taux de décroissance d'énergie acoustique,
   • des temps de réverbération caractéristiques de référence correspondant auxdites fonctions caractéristiques de référence,
   • et des niveaux de réverbération caractéristiques de référence correspondant auxdites fonctions caractéristiques de référence.

For this purpose, the invention proposes a method for estimating acoustic reverberation in a medium, comprising the following steps:

1. (a) a measurement step in which at least one acoustic signal is sensed in the medium,
2. (b) an observation step during which an acoustic energy decay rate distribution is determined from the acoustic signal picked up in step (a) and the characteristic function of the decay rate distribution is determined acoustic energy,
3. (c) an estimation step in which a characteristic reverberation time and a characteristic reverberation level of the sound in the medium are estimated from data representative of the acoustic energy decay rate distribution determined at step (b), the regression being made with reference to:
   • reference characteristic functions representative respectively of several distributions of decay rate of acoustic energy,
   • reference characteristic reverberation times corresponding to said characteristic functions reference,
   • and reference characteristic reverberation levels corresponding to said reference characteristic functions.

Thanks to these provisions, and notably because the method of estimation is applied to the distribution of decay rates of acoustic energy, it is possible to determine both a characteristic reverberation time and a reverberation level characteristic of the sound in the middle, reliably. These two parameters make it possible to satisfactorily process the sound signals picked up.

• au cours de l'étape d'estimation (c), on utilise un estimateur à fonction noyau et on détermine simultanément le temps de réverbération caractéristique et le niveau de réverbération caractéristique ;
• au cours de l'étape d'estimation (c), on utilise un estimateur de Nadaraya-Watson ;
• au cours de l'étape d'estimation (c), le niveau de réverbération caractéristique du son dans le milieu (7) est choisi parmi l'indice de clarté C_τ et l'indice de définition D_τ ;
• au cours de l'étape d'observation (b), on détermine les taux de décroissance d'énergie en calculant l'énergie Em du signal acoustique sur des trames successives m de signal, puis en calculant un rapport logarithmique entre l'énergie de deux trames successives : $$\rho \left(m\right)=\log \left(\frac{E_m}{E_{m-1}}\right)$$
  
  • le procédé comporte en outre une phase préliminaire de calibrage comprenant les étapes suivantes :
    • (a') au moins une étape initiale de détermination de signaux de référence dans laquelle on détermine une pluralité de signaux acoustiques de référence correspondant auxdits temps de réverbération caractéristiques de référence et auxdits niveaux de réverbération caractéristiques de référence,
    • (b') au moins une étape initiale d'observation au cours de laquelle, pour chaque signal acoustique de référence, on détermine une distribution de taux de décroissance d'énergie acoustique et la fonction caractéristique de référence ;
  • au cours de ladite étape de détermination de signaux de référence, on détermine par calcul au moins une partie des signaux acoustiques de référence et les temps de réverbération caractéristiques et niveaux de réverbération caractéristiques de référence correspondant auxdits signaux acoustiques de référence, à partir d'un ensemble prédéterminé de réponses impulsionnelles ;
  • au cours de ladite étape de détermination de signaux de référence, on détermine par mesure au moins une partie des signaux acoustiques de référence, les temps de réverbération caractéristiques et les niveaux de réverbération caractéristiques de référence correspondant auxdits signaux acoustiques de référence.

In various embodiments of the method according to the invention, one or more of the following provisions may also be used:

• during the estimation step (c), a kernel function estimator is used and the characteristic reverberation time and the characteristic reverberation level are simultaneously determined;
• during the estimation step (c), a Nadaraya-Watson estimator is used;
• during the estimation step (c), the characteristic reverberation level of the sound in the medium (7) is chosen from the clarity index C _τ and the definition index D _τ ;
• during the observation step (b), the energy decay rates are determined by calculating the energy Em of the acoustic signal on successive signal frames m, and then calculating a logarithmic ratio between the energy of the signal two successive frames: $$\rho \left(m\right)=\log \left(\frac{E_m}{E_{m-1}}\right)$$
  
  • the method further comprises a phase preliminary calibration procedure comprising the following steps:
    • (a ') at least one initial step of determining reference signals in which a plurality of reference acoustic signals corresponding to said reference characteristic reverberation times and said reference characteristic reverb levels are determined,
    • (b ') at least one initial observation step during which, for each reference acoustic signal, an acoustic energy decay rate distribution and the reference characteristic function are determined;
  • during said step of determining reference signals, at least a portion of the reference acoustic signals and the characteristic reverberation times and reference characteristic reverberation levels corresponding to said reference acoustic signals are determined by calculation from a predetermined set of impulse responses;
  • during said step of determining reference signals, at least a portion of the reference acoustic signals, the characteristic reverberation times and the reference characteristic reverberation levels corresponding to said reference acoustic signals are determined by measurement.

1. (a) des moyens de mesure pour capter au moins un signal acoustique émis dans le milieu,
2. (b) des moyens pour déterminer une distribution de taux de décroissance d'énergie acoustique à partir du signal acoustique capté par les moyens de mesure, et pour déterminer la fonction caractéristique de la distribution de taux de décroissance d'énergie acoustique,
3. (c) des moyens pour estimer un temps de réverbération caractéristique et un niveau de réverbération caractéristique du son dans le milieu à partir de données représentatives de la distribution de taux de décroissance d'énergie acoustique, la régression étant faite en référence à :
   • des fonctions caractéristiques de référence représentatives respectivement de plusieurs distributions de taux de décroissance d'énergie acoustique,
   • des temps de réverbération caractéristiques de référence correspondant auxdites fonctions caractéristiques de référence,
   • et des niveaux de réverbération caractéristiques de référence correspondant auxdites fonctions caractéristiques de référence.

Moreover, the invention also relates to a device for estimating acoustic reverberation in a medium, comprising:

1. (a) measuring means for sensing at least one acoustic signal emitted in the medium,
2. (b) means for determining an acoustic energy decay rate distribution from the signal acoustics captured by the measuring means, and to determine the characteristic function of the acoustic energy decay rate distribution,
3. (c) means for estimating a characteristic reverberation time and a reverberation level characteristic of sound in the medium from data representative of the acoustic energy decay rate distribution, the regression being made with reference to:
   • reference characteristic functions representative respectively of several distributions of decay rates of acoustic energy,
   • reference characteristic reverberation times corresponding to said reference characteristic functions,
   • and reference characteristic reverberation levels corresponding to said reference characteristic functions.

BRIEF DESCRIPTION OF THE DRAWINGS

Other features and advantages of the invention will become apparent from the following description of one of its embodiments, given by way of non-limiting example, with reference to the accompanying drawings.

• la figure 1 est une vue schématique illustrant la réverbération du son dans une pièce lorsqu'un sujet parle de façon que ses paroles soient captées par un dispositif selon une forme de réalisation de l'invention,
• la figure 2 est un schéma de principe du dispositif de la figure 1.

On the drawings:

• the figure 1 is a schematic view illustrating the reverberation of sound in a room when a subject speaks so that his words are picked up by a device according to one embodiment of the invention,
• the figure 2 is a schematic diagram of the device of the figure 1 .

DESCRIPTION DETAILED

In the different figures, the same references designate identical or similar elements.

The aim of the invention is to estimate the acoustic reverberation of a medium 7, for example a room in a building as schematized on the figure 1 , so as to process the acoustic signals picked up by an electronic device 1 provided with a microphone 2. The electronic device 1 can be for example a telephone in the example shown, or a computer or other.

When a sound is emitted in the medium 7, for example by a person 3, this sound propagates to the microphone 2 along various paths 4, either direct or after reflection on one or more walls 5, 6 of the medium 7.

As shown on the figure 2 , the electronic device 1 may comprise for example an electronic central unit 8 such as a processor or the like, connected to the microphone 2 and various other elements, including for example a speaker 9, a keyboard 10, a screen 11. L electronic central unit 8 can communicate with an external network 12, for example a telephone network.

• un temps de réverbération caractéristique, par exemple le temps de réverbération à 60 dB RT₆₀,
• et un niveau de réverbération caractéristique (par exemple indice de clarté ou de définition, ou indice de signal direct sur signal réverbéré).

The invention enables the electronic device 1 to measure blindly two characteristic parameters of the reverberation of the medium 7:

• a characteristic reverberation time, for example the reverberation time at 60 dB RT ₆₀ ,
• and a characteristic reverberation level (eg clarity or definition index, or direct signal index on reverberated signal).

These parameters can be used to suppress the effects of echoes or more generally to optimize the sound signals picked up by the microphone 2. The parameters in question are estimated in a repetitive way, so that the device 1 is adapted for example to changes in speakers 3, to movements of speakers 3, to movements of the device 1 or other objects in the medium 7.

Où :

• h est la réponse impulsionnelle du milieu de longueur N_h ,
• n est un indice temporel, par exemple un nombre d'échantillons obtenus par un échantillonnage de pas temporel constant, n étant compris entre 1 et N_h.

The reverberation time at 60dB RT ₆₀ can be defined by the inverse integration method of Manfred R. Schroeder (New Method of Measuring Reverberation Time, The Journal of the Acoustical Society of America, 37 (3): 409, 1965 ) by the EDC Energy Decay Curve curve: $$\mathrm{EDC}\left(\mathrm{not}\right)={\displaystyle {\Sigma }_{k=\mathrm{not}}^{{\mathrm{NOT}}_h}h{\left(k\right)}^2}$$

Or :

• h is the impulse response of the medium of length N _h ,
• n is a time index, for example a number of samples obtained by sampling a constant time step, n being between 1 and N _h .

RT ₆₀ is the time at the time index n required for EDC (n) to decrease by 60 dB.

Although the RT ₆₀ reverberation time is the most commonly used, we could estimate another reverberation time characteristic of the medium 7.

ou par l'indice de définition : $$D_{\tau }=10{\log }_{10}\left(\frac{{\displaystyle {\sum }_{n=0}^{N_{\tau }}{h}^2\left(n\right)}}{{\displaystyle {\sum }_{n=0}^{\infty }{h}^2\left(n\right)}}\right)\mathrm{dB},$$

où :

• N _τ est le nombre d'échantillons à pas temporels constant correspondant au temps τ, généralement compris entre 0.1 ms et 1 s,
• n est un indice temporel compris entre 1 et N _τ , représentatif d'un nombre d'échantillons de pas temporel constant,
• h(n) est la réponse impulsionnelle du milieu 7.

The reverb level is most commonly represented by the clarity index: $${\mathrm{VS}}_{\tau }=10{\mathrm{<figure-callout\; id="10"\; label="log"\; filenames="imgf0001.png"\; state="\{\{state\}\}">log</figure-callout>}}_{10}\left(\frac{{\displaystyle {\Sigma }_{\mathrm{not}=0}^{{\mathrm{NOT}}_{\tau }}{h}^2\left(\mathrm{not}\right)}}{{\displaystyle {\Sigma }_{\mathrm{not}={\mathrm{NOT}}_{\tau }+1}^{\infty }{h}^2\left(\mathrm{not}\right)}}\right)\mathrm{dB},$$

or by the definition index: $$D_{\tau }=10{\mathrm{<figure-callout\; id="10"\; label="log"\; filenames="imgf0001.png"\; state="\{\{state\}\}">log</figure-callout>}}_{10}\left(\frac{{\displaystyle {\Sigma }_{\mathrm{not}=0}^{{\mathrm{NOT}}_{\tau }}{h}^2\left(\mathrm{not}\right)}}{{\displaystyle {\Sigma }_{\mathrm{not}=0}^{\infty }{h}^2\left(\mathrm{not}\right)}}\right)\mathrm{dB},$$

or :

• N _τ is the number of samples with a constant time step corresponding to the time τ , generally between 0.1 ms and 1 s,
• n is a time index between 1 and N _τ , representative of a number of samples of constant time step,
• h ( n ) is the impulse response of medium 7.

These indices have been described in particular by PA Naylor and ND Gaubitch (Speech Dereverberation, Springer, eds edition, 2010 ).

The two most commonly used values of τ are 50 ms and 80 ms, in particular 50 ms (indices C ₅₀ and D ₅₀ ), but other durations are possible and more generally other indices reflecting the ratio of the direct sound to the sound. reverberation could be estimated in the method according to the invention, implemented for example by the aforementioned electronic central unit 8.

1. (a) une étape de mesure de signal acoustique,
2. (b) une étape d'observation au cours de laquelle on détermine une distribution de taux de décroissance d'énergie acoustique à partir des signaux acoustiques mesurés à l'étape (a),
3. (c) une étape d'estimation au cours de laquelle on estime, par régression à partir de la distribution de taux de décroissance d'énergie acoustique déterminée à l'étape (b), un temps de réverbération caractéristique et un niveau de réverbération caractéristique du son dans le milieu 7.

This process comprises the following steps:

1. (a) an acoustic signal measurement step,
2. (b) an observation step during which an acoustic energy decay rate distribution is determined from the acoustic signals measured in step (a),
3. (c) an estimation step in which it is estimated, by regression from the sound energy decay rate distribution determined in step (b), a characteristic reverberation time and a characteristic reverberation level sound in the middle 7.

(a) Measurement step :

During this step, the microphone 2 captures "blind" (that is to say, without prior knowledge of the signals transmitted), an acoustic signal broadcast in the medium 7, for example when the speaker 3 speaks. This signal is sampled and stored in processor 8 or an ancillary memory (not shown).

(b) Observation stage:

During this step, an acoustic energy decay rate distribution is determined from the acoustic signal measured in step (a).

For this purpose, the energy envelope of the reverberated signal d _x (n) as first described by Wen et al. ( JYC Wen, EAP Habets, and PA Naylor, Blind Estimation of Reverberation Time Based on the Distribution of Signal Decay Rates, Acoustics, Speech. and Signal Processing, 2008, ICASSP 2008, IEEE International Conference pages 329-332, March 2008 ).

et ensuite estimer le taux de décroissance d'énergie en calculant le rapport logarithmique entre deux trames successives : $${\lambda }_x\approx \rho \left(m\right)=\log \left(\frac{E_m}{E_{m-1}}\right).$$

By doing a calculation on N _tr frames, signal samples separated by jumps of R signal samples, we can calculate a total energy of the frame m by the formula: $$E_m={\displaystyle {\Sigma }_{i=0}^{{\mathrm{NOT}}_{\omega }-1}{d}_x\left(\mathit{mR}+i\right)}$$

and then estimate the rate of energy decay by calculating the logarithmic ratio between two successive frames: $${\lambda }_x\approx \rho \left(m\right)=\log \left(\frac{E_m}{E_{m-1}}\right).$$

où λ_s and λ_h sont respectivement le taux de décroissance d'énergie du signal anéchoïque émis et du milieu 7 (le signal capté est une convolution du signal anéchoïque émis (parole) avec la réponse impulsionnelle du milieu entre le locuteur 3 et le microphone 2, n étant l'indice temporel défini précédemment).Indeed, the energy envelope d _x (n) can be expressed by the formula: $$d_x\left(\mathrm{not}\right)=\{\begin{array}{cc}\left(e^{{\lambda }_h\mathrm{not}}-e^{{\lambda }_s\mathrm{not}}\right)/\left({\lambda }_h-{\lambda }_s\right)& \mathrm{if}{\lambda }_h\ne {\lambda }_s\\ {\mathit{born}}^{{\lambda }_h{\mathrm{not}}_S}& \mathrm{if}{\lambda }_h={\lambda }_s\end{array}$$

where λ _s and λ _h are respectively the rate of energy decay of the emitted anechoic signal and medium 7 (the signal picked up is a convolution of the emitted anechoic signal (speech) with the impulse response of the medium between the speaker 3 and the microphone 2, n being the temporal index defined above).

ce qui justifie la formule (5) ci-dessus.Since the sum is dominated by the exponential term corresponding to the largest value of λ , the energy decay rate of the reverberated signal λ _x can be approximated by: $${\lambda }_x\approx \max \left[{\lambda }_h,{\lambda }_s\right]$$

which justifies formula (5) above.

The calculation of ρ (m) can typically be done on a number M of frames of at least 2000, corresponding to at least 1 min of signal according to the selected analysis parameters. The frames can have an individual duration of 10 to 100 ms, in particular of the order of 32 ms. The frames may overlap each other, for example with a recovery rate of the order of 50% between successive frames.

Different values of the energy decay rate ρ (m) are obtained, which have a certain statistical distribution (number of realizations or probability of realization as a function of the rate of energy decay ρ (m), as explained, for example, in the article by Wen et al., above).

où X représente ici le taux de décroissance d'énergie ρ(m) susmentionné estimé pour diverses valeurs de m (formule (5)), F_x la distribution cumulée de X et f est une variable sans dimension généralement appelée fréquence angulaire.The characteristic function of the energy decay rate distribution is then determined by the following formula (see Audrey Feuerverger and Roman A. Mureika [The empirical characteristic function and its applications, Ann. Statist., 5 (1): 88-97, 01 1977 ]): $${\phi }_x\left(f\right)={\displaystyle \int e^{\mathit{ifx}}{\mathit{dF}}_x\left(x\right)=\mathrm{E}\left[e^{\mathit{IFX}}\right]}$$

where X represents here the estimated energy decay rate ρ (m) estimated for various values of m (formula (5)), F _x the cumulative distribution of X and f is a dimensionless variable generally called angular frequency.

The characteristic function can be calculated for angular frequencies f ranging for example from 0 to 0.4, in increments of 0.001.

(c) Estimation step :

We start from the characteristic function Φ _{ρ (m)} (f), calculated for p / 2 frequencies f (p being an even natural integer), the frequency range f and their sampling being provided so that | Φ _{ρ (m)} ( f) | is preferably comprised between 0.1 and 1.

Typically, p can be for example between 256 and 2048.

Since the characteristic function is a complex number, it can be represented by a vector of X belonging to R ^p constituting the random input vector x of the estimator used. The random output vector y of the estimator, belonging to R ² , has as component the two estimated parameters, for example (RT ₆₀ , C ₅₀ ) or (RT ₆₀ , D ₅₀ ).

The estimator used may advantageously be a kernel function estimator, for example a Nadaraya-Watson estimator. Such an estimator has the advantage of simultaneously determining the characteristic reverberation time and the characteristic reverberation level.

The estimator in question can be determined in advance in an initial calibration phase, where at least one initial step of determination of reference signals (a ') and at least one initial observation step (b') is implemented.

During the initial step of determining reference signals, a plurality of reference acoustic signals and reference characteristic reverberation times and corresponding reference characteristic reverb levels are determined.

During the initial observation step, the acoustic energy decay rate distribution and the reference characteristic function for each reference acoustic signal are determined identically or similarly to the observation step (b). ) above.

The acoustic reference signals are generally vocal signals of the number of N and correspond to N cases of different figures (different speakers, different positions, different media 7). N can be several hundred or even thousands.

• avec des mesures réelles nouvelles effectuées par exemple avec un dispositif électronique 1 d'un modèle déterminé (dans ce cas, le temps de réverbération caractéristique et le niveau de réverbération caractéristique peuvent être également mesurés) ;
• et / ou avec des signaux acoustiques synthétiques.

The initial step of determining reference signals can be performed:

• with real new measurements made for example with an electronic device 1 of a given model (in this case, the characteristic reverberation time and the characteristic reverberation level can also be measured);
• and / or with synthetic acoustic signals.

In the case of real measurements, these will generally not be done in the particular medium where the electronic device 1 will be used, although this case may be considered.

The aforementioned synthetic acoustic signals can be calculated by convolving pre-recorded impulse responses with pre-recorded speech anechoic signals from different speakers. The prerecorded impulse responses may for example come from impulse response databases, for example from open access databases such as Aachen Impulse Response databases. (http://www.openairlib.net/auralizationdb), MARDY ( Wen et al., Evaluation of speech dereverberation algorithms using the Mardy database, Sept. IWAENC 2006, By is), QueenMary ( R. Stewart and M. Sandler, Database of omnidirectional and b-format room impulse responses, In Acoustics Speech and Signal Processing (ICASSP). 2010 IEEE International Conference on, pages 165-168, March 2010 ), for example with RT ₆₀ reverberation times ranging from 0.3 s to 8 s and C ₅₀ clarity indices from -10 dB to 25 dB. Anechoic speech signals may be signals recorded from various speakers, for example of different ages and sexes, with for example recording times for example of a few minutes, for example of the order of 5 minutes.

The energy decay rate distributions can for example be calculated over frames of 10 to 100 ms, in particular of the order of 32 ms. The frames may overlap each other, for example with a recovery rate of the order of 50% between successive frames. The characteristic functions can be calculated for angular frequencies ranging for example from 0 to 0.4, in increments of 0.001.

où :

• x_i, y_i, i = 1 à N, sont les N réalisations des vecteurs x, y utilisées pour l'étape de calibration,
• K_λ(x, x_i) est une fonction noyau (kernel) de fenêtre λ (λ est une constante également appelée paramètre de lissage),
• x est le vecteur d'entrée inconnu (mesure effectuée à l'étape de mesure (a) en vue d'estimer le vecteur y par la formule y = f̂(x)).

N yields of the aforementioned x and y vectors are thus obtained, and the Nadaraya-Watson estimator can then be determined by the formula: $$\widehat{f}\left(x\right)=\frac{{\displaystyle {\Sigma }_{i=1}^{\mathrm{NOT}}{\mathrm{there}}_i{K}_{\lambda }\left(x,x_i\right)}}{{\displaystyle {\Sigma }_{i=1}^{\mathrm{NOT}}{K}_{\lambda }\left(x,x_i\right)}}.$$

or :

• x _i , y _i , i = 1 to N, are the N realizations of the x, y vectors used for the calibration step,
• K _λ (x, x _i ) is a kernel function of window λ (λ is a constant also called smoothing parameter),
• x is the unknown input vector (measured at the measurement step (a) to estimate the vector y by the formula y = f ( x )).

The kernel function K _λ (x, x _i ) is a function of x and x _i as defined in particular by Scholkopf et al. ( B. Scholkopf and AJ Smola, Learning with Kernels, MIT Press, Cambridge, MA, 2001 ).

In particular, it is possible to use a Gaussian kernel, for example with a window of λ = 5.10 ^-4 (non-limiting example): $$K_{\lambda }\left(x,x_i\right)=\frac{1}{\lambda }{e}^{\frac{-{\Vert x-x_i\Vert }^2}{2\lambda }}.$$

The tests carried out show that the method of the invention is more accurate than the methods of the prior art for determining the reverberation time, and it also makes it possible to determine the reverberation level at the same time as the reverberation time, which is a big gain.

Claims (9)

1. A method for estimating the acoustic reverberations in an environment (7) comprising the following steps:

   (a) a measurement step in which at least one acoustic signal emitted in the environment (7) is captured;

   (b) an observation step during which an acoustic energy decay rate distribution is determined from the acoustic signal measured in step (a) and the characteristic function of the acoustic energy decay rate distribution is determined;

   (c) an estimation step during which a characteristic reverberation time and a characteristic reverberation level of the sound in the environment (7) thereof are estimated by regression from said characteristic function determined in step (b), where the regression is done with reference to:

   - reference characteristic functions representative respectively of several acoustic energy decay rate distributions;

   - reference characteristic reverberation times corresponding to said reference characteristic functions; and

   - reference characteristic reverberation levels corresponding to said reference characteristic functions.
2. The method according to claim 1 wherein during the estimation step (c), a kernel function estimator is used and the characteristic reverberation time and the characteristic reverberation level are determined simultaneously.
3. The method according to claim 2 wherein during the estimation step (c), a Nadaraya-Watson estimator is used.
4. The method according to any one of the preceding claims wherein during the estimation step (c), the characteristic reverberation level of the sound in the environment (7) is chosen among the clarity index C_τ and the definition index D_τ.
5. The method according to any one of the preceding claims wherein during the observation step (b), the energy decay rates are determined by calculating the energy E_m of the acoustic signal on successive signal frames m, and then calculating a logarithmic ratio between the energy of two successive frames: $$\rho \left(m\right)=\log \left(\frac{E_m}{E_{m-1}}\right)$$

   
6. The method according to any one of the preceding claims further comprises a preliminary calibration phase comprising the following steps:

   (a') at least one initial reference signal determination step in which a plurality of reference acoustic signals corresponding to said reference characteristic reverberation times and said reference characteristic reverberation levels are determined;

   (b') at least one initial observation step during which, an acoustic energy decay rate distribution and the reference characteristic function are determined for each reference acoustic signal.
7. The method according to claim 6 wherein during said reference signal determination step, at least one part of the reference acoustic signals and the reference characteristic reverberation times and characteristic reverberation levels corresponding to said reference acoustic signals are determined by calculation from a predetermined set of impulse responses.
8. The method according to claim 6 wherein during said reference signal determination step, at least one part of the reference acoustic signals, the characteristic reverberation times and the reference characteristic reverberation levels corresponding to said reference acoustic signals are determined by measurement.
9. A device for estimating the acoustic reverberations in an environment (7) comprising:

   (a) means of measurement (2) for capturing at least one acoustic signal emitted in the environment (7);

   (b) means of determination (8) of an acoustic energy decay rate distribution from the acoustic signal captured by the means of measurement, and for determining the characteristic function of the acoustic energy decay rate distribution;

   (c) means of estimation (8) of a characteristic reverberation time and a characteristic reverberation level of the sound in the environment from data representative of the acoustic energy decay rate distribution, where the regression is done with reference to:

   - reference characteristic functions representative respectively of several acoustic energy decay rate distributions;

   - reference characteristic reverberation times corresponding to said reference characteristic functions; and

   - reference characteristic reverberation levels corresponding to said reference characteristic functions.

Images