EP0918317B1 - Verfahren zur Frequenzfilterung mittels eines Wiener Filters für die Geräuschunterdrückung von Audiosignalen - Google Patents

Verfahren zur Frequenzfilterung mittels eines Wiener Filters für die Geräuschunterdrückung von Audiosignalen Download PDF

Info

Publication number
EP0918317B1
EP0918317B1 EP98402894A EP98402894A EP0918317B1 EP 0918317 B1 EP0918317 B1 EP 0918317B1 EP 98402894 A EP98402894 A EP 98402894A EP 98402894 A EP98402894 A EP 98402894A EP 0918317 B1 EP0918317 B1 EP 0918317B1
Authority
EP
European Patent Office
Prior art keywords
noise
model
signals
frame
coefficient
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.)
Expired - Lifetime
Application number
EP98402894A
Other languages
English (en)
French (fr)
Other versions
EP0918317A1 (de
Inventor
Dominique Thomson-CSF Prop. Intel. Pastor
Gérard Thomson-CSF Prop. Intel. Reynaud
Pierre-Albert Thomson-CSF Prop. Intel. Breton
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.)
Thales Avionics SAS
Original Assignee
Thales Avionics SAS
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 Thales Avionics SAS filed Critical Thales Avionics SAS
Publication of EP0918317A1 publication Critical patent/EP0918317A1/de
Application granted granted Critical
Publication of EP0918317B1 publication Critical patent/EP0918317B1/de
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering

Definitions

  • the present invention relates to a method of frequency filtering using a Wiener filter.
  • the main areas concern telephone or radiotelephone communications, speech recognition, sound recording on board aircraft civil or military, and more generally of all noisy vehicles, on-board intercoms, etc.
  • noise is caused by engines, air conditioning, ventilation of on-board equipment or aerodynamic noises. All these noises are picked up, at less partially, by the microphone in which the pilot or another member of the crew.
  • one of the characteristics noises is to be very variable over time. In indeed, they are very dependent on the operating regime engines (take-off phase, stabilized speed, etc.).
  • Useful signals i.e. signals representing the conversations, also present particularities: they are most often short-lived.
  • voicing relates to elementary characteristics of pieces of speech, and more specifically concerns vowels, as well only part of the consonants: "b”, “d”, “g”, “j”, etc. These letters are characterized by an audio signal from pseudo-periodic structure.
  • the denoising process take into account this important characteristic of audio signals including speech.
  • These processes generally include the steps following main: a division into frames of the signal to denois, the processing of these frames by a Fourier transform operation (or a similar transform) to move into the field frequency, the proper denoising treatment by digital filtering, and processing, dual of the first, by an inverse Fourier transform, to return to the time domain.
  • the last step is a signal reconstruction. This reconstruction can be obtained by multiplying each of the frames by a window of weighting.
  • Wiener filter in particular a so-called optimal Wiener filter. This one presents the advantage of differentiating the frames successive.
  • Wiener's optimal filtering is at the center of the methods signal processing, based on second order statistical characteristics and therefore of the notion of correlation.
  • Wiener filtering allows the separation of decorrelation signals. Its importance is linked to the simplicity of theoretical calculations. In addition, it can apply to a multitude of specific processes, and in particular, with regard to the preferred application targeted by the invention, the extraction of noise polluting a signal of speech.
  • the invention therefore sets itself the aim of overcoming the disadvantages of filtering methods of the known art, especially the main drawback which has just been recalled: the presence of a parasitic residual noise in the denoised signal, called "musical noise".
  • the invention aims to more generally, to increase the intelligibility of the speech, in its main application.
  • the Wiener filter used for digital filtering is modified in an optimized way by introducing a term of energy compensation aimed at overestimating the level of noise.
  • this compensation term is adaptive.
  • Each block referenced 0 to 5, represents a phase of the process, which itself can be subdivided into elementary steps.
  • the method of the invention includes a step of splitting the signal into frames audiophonic to denois (block 0).
  • the frame signals are not “continuously evolving" signals, but discrete signals obtained by sampling. It is assumed that the signals are sampled at period T e , before digital processing. It is common to then consider 2 p samples for a signal frame, choosing p so that the value 2 p T e is of the order of magnitude of the duration D of a frame.
  • p the value of the duration of a frame.
  • D LGframe ⁇ T e is therefore satisfied.
  • the step of cutting into frames, as indicated in FIG. 1, is therefore preceded by a step of digitization by sampling.
  • the stages of digitization and cutting into frames (block 0) are common to known art.
  • the samples numeric thus created are stored in a buffer memory circulating of type "FIFO" (that is to say of type "first in - first out ”) to be read as frames successive.
  • FIFO type "first in - first out ”
  • the operations carried out in block 1 consists in identify segments of the signal to be denoised that do not contain only noise.
  • the output of this block consists of a series of digital samples representative of noise alone.
  • a noise model is developed at from noisy signals, or more precisely from frames successively read (block 0).
  • block 3 has a step of estimating the spectral density of the frame signal current and its energy calculation.
  • the coefficients of the filter frequency denoising the signal are determined in the manner which will be detailed below.
  • the method of the invention is based on a energy compensation and noise overestimation.
  • the denoised time signal is rebuilt, ensuring the best continuity possible between frames.
  • the signals can be used as is by various methods such as automatic speech recognition. In itself, this phase of the process is common to known art, and there is no no need to detail the reconstruction method or for processing the signals at the output of block 4.
  • the process makes it possible to modify and optimize the coefficients of the Wiener filter used for the phase of denoising proper (block 4), so as to eliminate or, to say the least, strongly attenuate the so-called parasitic noises "Musical".
  • the dispersion is quantified by a coefficient from the analysis carried out in block 2, from the model of noise developed in block 1.
  • the method according to the invention overestimates this spectral density, in y introducing a degree of adaptability in order to optimize the perception of the denoised signal.
  • FIG. 2 very schematically illustrates a Wiener filter used to denoise a noisy signal U (n) .
  • the coefficients of the filter Wiener are modified using parameters determined in blocks 2 and 3, in the way that will now be Detailed.
  • the process according to the invention optimally modifies the coefficients of the Wiener filter and introduces a compensation term energy, artificially overestimating the level of noise, with different levels of adaptivity of this compensation.
  • the exponential attenuation coefficient ⁇ is a term commonly used in the literature on digital filtering and, more specifically, noising. A typical value for this parameter is 0.5.
  • the curve in FIG. 4a shows that the energy of the signal in the frequency band ⁇ , represented by the spectral density ⁇ x , is not negligible.
  • the energy weighting report described below reduces this distortion in the signal noised.
  • noise denoising alone is correct, but it can be too brutal within parts of the useful signal.
  • ⁇ remains close to a typical value equal to 10, when the noisy signal contains only noise, and varies between 0 and 10, when a useful signal is present in the noisy signal. A degree is therefore advantageously introduced adaptivity.
  • FIG. 5 This third modification is illustrated by FIG. 5.
  • This type of filter therefore has good efficiency. in terms of eliminating degraded signal segments in which speech is absent and decrease in distortions inflicted on the wanted speech signal.
  • the probability of generating "musical noise” is also related, as noted, to the variance of of the spectral density of the noise over all of the frames.
  • the value of the overestimation coefficient is made dependent on the statistical properties of the noise.
  • a coefficient, called maximum below is introduced, proportional to the dispersion of the values of spectral densities of the noise.
  • the maximum coefficient is equal to the maximum ratio, for all the frames of the noise model, between the maximum of the spectral density of the frame of the noise model considered, and the maximum of the estimated spectral density of the noise model.
  • this coefficient characterizes the maximum noise disparity for frequency channels carrying significant energy. Multiplied by the coefficient ⁇ , it provides additional proportional attenuation to this disparity.
  • the method is based on continuous research and noise model automatically. This research is done on digitized and stored u (t) signal samples into an input buffer. This memory is capable to memorize all the samples of several frames of the input signal (at least 2 frames and, in the general case, N frames).
  • the noise model sought consists of a succession of several frames including energy stability and the relative energy level suggest that it is ambient noise and not a speech signal or other disturbing noise. We will see later how this automatic search.
  • ambient noise is a signal with a minimum energy stable in the short term. In the short term, it must hear a few frames, and we will see in the example practice given below that the number of frames intended for assess the noise stability is from 5 to 20. The energy must be stable on several frames, otherwise we must assume that the signal contains speech or a noise other than ambient noise. It must be minimal, otherwise, the signal is considered to contain breathing or phonetic speech resembling noise but superimposed on ambient noise.
  • Figure 6 shows a typical configuration time evolution of the energy of a signal microphone at the start of a broadcast, speech, with a breathing noise phase, which goes out during a few tens to hundreds of milliseconds to make room for ambient noise alone, after which an energy level high indicates the presence of speech, to finally return to the ambient noise.
  • N1 5
  • the digital values of all the samples of these N frames are stored.
  • This set of NxP samples constitutes the current noise model. It is used in denoising. Analysis of the following frames continues.
  • the ambient noise changes slowly, the change will be taken into account that the comparison threshold with the stored model is greater than 1. If it evolves more rapidly in the increasing direction, evolution risks not not be taken into account, so it is better to schedule a reset from time to time looking for a noise pattern. For example, on an airplane on the ground at a standstill, the ambient noise will be relatively low, and it should not be that during the takeoff phase the noise model remains frozen on what it was at standstill that a noise model is only replaced by a less energetic or not much more energetic. The reset methods will be explained later. considered.
  • Figure 7 shows a flowchart of automatic noise pattern search operations ambient.
  • n The number of the current frame in a noise model search operation is designated by n and is counted by a counter as the search is carried out.
  • n is set to 1. This number n will be incremented as a model of several successive frames is developed.
  • the model already includes by hypothesis n -1 successive frames meeting the conditions imposed to be part of a model.
  • the signal energy of the frame is calculated by summation of the squares of the numerical values of the samples of the frame. It is kept in memory.
  • the ratio between the energies of the two frames is calculated. If this ratio is between two thresholds S and S 'one of which is greater than 1 and the other of which is less than 1, it is considered that the energies of the two frames are close and that the two frames can be part of a noise model.
  • the frames are declared incompatible and the search is reset by resetting n to 1.
  • the rank n of the current frame is incremented, and in an iterative procedure loop, the energy of the next frame is calculated and a comparison with the energy of the previous frame or previous frames, using thresholds S and S ' .
  • the first type of comparison consists in comparing only the energy of the frame n to l energy of the frame n -1.
  • the second type consists in comparing the energy of the frame n to each of the frames 1 to n -1.
  • the second way leads to a greater homogeneity of the model but it has the disadvantage of not taking into account sufficiently well the cases where the noise level increases or decreases quickly.
  • the energy of the frame of rank n is compared with the energy of the frame of rank n -1 and possibly of other previous frames (not necessarily all for that matter).
  • the number N2 is chosen so as to limit the computation time in the subsequent operations for estimating the spectral noise density.
  • n is less than N2 , the homogeneous frame is added to the previous ones to help build the noise model, n is incremented and the next frame is analyzed.
  • n is equal to N2
  • the frame is also added to the previous n -1 homogeneous frames and the model of n homogeneous frames is stored for use in eliminating noise.
  • the search for a model is also reset by setting n to 1.
  • the previous steps relate to the first model search. But once a model has been stored, it can be replaced at any time by a model more recent.
  • the replacement condition is still a energy condition but this time it relates to the average energy of the model and no longer on the energy of each frame.
  • the new model is considered to be better and it is stored in place of the previous one. Otherwise, the new model is rejected and the old one remains in force.
  • the threshold SR is preferably slightly greater than 1.
  • the SR threshold was less than or equal to 1, we would store the least energetic homogeneous frames each time, which corresponds well to the fact that we consider that ambient noise is the energy level below which we never descend . But, we would eliminate any possibility of evolution of the model if the ambient noise started to increase.
  • the threshold SR is about 1.5. Above this threshold we will keep the old model; below this threshold we will replace the old model with the new one. In both cases, the search will be reinitialized by recommencing the reading of a first frame of the input signal u (t) , and setting n to 1.
  • This inhibition is to prevent certain sounds are taken for noise, when they are useful phonemes, that a noise model based on these sounds be stored and that noise suppression after the development of the model then tends to remove all similar sounds.
  • Ambient noise can indeed increase significantly and quickly, for example during the acceleration phase of the engines of an airplane or other vehicle, air, land or sea.
  • the threshold SR requires that the previous noise model be kept when the average noise energy increases too quickly.
  • Periodicity can be based on the average duration of speech in the application considered; for example the speaking times are in average of a few seconds for the crew of an airplane, and resetting can take place with a frequency of a few seconds.
  • Figure 1 block 1
  • the implementation of the method of developing a noise model ( Figure 1: block 1) and, more general of the process according to the invention, can be done at from non-specialized computers, provided with necessary calculation programs and receiving the digitized signal samples as supplied by an analog-digital converter, via a port adapted.
  • This implementation can also be done from a specialized computer based on signal processors digital, which enables faster processing large number of digital signals.
  • the computers are associated, as is well known, with different types of memories, static and dynamic, for recording the programs and the intermediate data, as well as with circulating memories of the "FIFO" type.
  • the system includes an analog-to-digital converter, for digitizing u (t) signals, and a digital-to-analog converter, as needed, if the denoised signals are to be used in analog form.
  • Figure 8 is a summary diagram all the stages of the filtering process according to the invention, in a preferred embodiment.
  • stages are divided into a first subset steps to determine the parameters depending on the noise model, and a second subset steps to determine the dependent parameters only from the current frame of the signal to be denoised.
  • the first step of the first subset includes an initial step of selecting a suitable noise model to the specific application, advantageously a model of noise determined by the method described above, in reference to Figures 6 and 7.
  • This first subset of steps includes two branches.
  • the energy of the frame is calculated for each frame of the noise model (in the time domain), then the average energy of the frames of the model is calculated, which makes it possible to estimate the average energy of the model, i.e. the parameter E x .
  • the parameter ⁇ x ( ⁇ ) is also used for the calculation of one of the other coefficients of the Wiener filter.
  • the second subset of steps also includes two branches.
  • the energy of the current frame, E u is determined, and in the second branch, the spectral density of the current frame ⁇ u is estimated.
  • the coefficients ⁇ and ⁇ are fixed coefficients predetermined, typically equal to 10 and 0.5, respectively.
  • the invention is not not reduced to the only domain of filtering of signals containing noisy speech, even if this domain constitutes one of the favorite apps.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Quality & Reliability (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Filters That Use Time-Delay Elements (AREA)
  • Soundproofing, Sound Blocking, And Sound Damping (AREA)
  • Noise Elimination (AREA)

Claims (9)

  1. Verfahren zur Frequenzfilterung für die Rauschminderung von verrauschten akustischen Signalen (u(t)), die aus sogenannten Nutzsignalen gemischt mit Rauschsignalen bestehen, wobei das Verfahren mindestens einen Schritt (0) der Zerlegung der akustischen Signale in eine Reihe von identischen Rahmen einer vorbestimmten Länge und einen Schritt der Frequenzfilterung (4) mithilfe eines WIENER-Filters enthält, dadurch gekennzeichnet, daß es außerdem die folgenden Schritte aufweist:
    Erarbeitung eines Rauschmodells (1) ausgehend von den verrauschten Signalen (u(t)) über eine bestimmte Anzahl N von Rahmen, wobei N zwischen einem vorbestimmten unteren und einem vorbestimmten oberen Grenzwert liegt,
    Anwendung einer Fourier-Transformierten auf die N Rahmen,
    Schätzung (2) der spektralen Dichte für jeden Rahmen des Modells,
    Schätzung (2) der mittleren spektralen Dichte des Rauschmodells,
    Berechnung (2) eines statistischen Über-Schätzkoeffizienten ausgehend von diesen beiden Schätzwerten, wobei der statistische Koeffizient dem maximalen Verhältnis zwischen dem Maximum der geschätzten spektralen Dichte eines betrachteten Rahmens des Rauschmodells und dem Maximum der geschätzten spektralen Dichte des Rauschmodells für die N Rahmen des Rauschmodells gleicht,
    Schätzung (3) der spektralen Dichte für jeden Rahmen der verrauschten Signale (u(t)),
    und Veränderung (4) der Koeffizienten des WIENER-Filters für jeden Rahmen der verrauschten Signale (u(t)), damit die folgende Beziehung erfüllt ist: W(ν)=(1-α•maxi•γx(ν)γu(ν) )β
    wobei in dieser Beziehung α und β vorbestimmte Koeffizienten, nämlich ein statischer Koeffizient der Energiekompensation beziehungsweise ein Koeffizient der exponentiellen Dämpfung sind, v die Gesamtheit der Frequenzkanäle der Fourier-Transformierten beschreibt, γu(v) den Schätzwert der spektralen Dichte des verrauschten Rahmens beschreibt, γx(v) die spektrale Dichte des Rauschmodells ist und maxi den statistischen Über-Schätzungskoeffizienten darstellt, der den statischen Koeffizienten α der Energiekompensation verändert.
  2. Verfahren nach Anspruch 1, dadurch gekennzeichnet, daß der statistische Koeffizient maxi der folgenden Beziehung genügt: maxi = Max_Max(vγi(ν))Max(vγx(ν))_ fÅri=1....N
  3. Verfahren nach einem der Ansprüche 1 und 2, dadurch gekennzeichnet, daß es die folgenden zusätzlichen Schritte aufweist:
    Berechnung der mittleren Energie Ex des Rauschmodells,
    Berechnung der Energie Eu des aktuellen Rahmens für jeden Rahmen der verrauschten Signale (u(t)),
    und Multiplikation des statischen Koeffizienten α der Energiekompensation mit einem energetischen Gewichtungskoeffizient gleich dem Verhältnis Ex/Eu, um diese Koeffizienten selektiv für jeden Rahmen der verrauschten Signale (u(t)) durch Anwendung eines permanent variablen Koeffizienten zwischen einem Höchstwert und einem Mindestwert zu verändern, wobei der Höchstwert im wesentlichen den Wert Eins hat, wenn Nutzsignale im verrauschten Signal (u(t)) nicht vorliegen, und im wesentlichen den Wert Null, wenn die Energie der Nutzsignale deutlich größer als die Energie der Rauschsignale ist, wobei die Koeffizienten des WIENER-Filters die folgende Beziehung erfüllen: W(ν)=(1-α•maxi•Ex Eu •γx(ν)γu(ν) )β
  4. Verfahren nach einem beliebigen der vorstehenden Ansprüche, dadurch gekennzeichnet, daß der statische Koeffizient α der Energiekompensation den Wert 10 hat.
  5. Verfahren nach einem beliebigen der vorstehenden Ansprüche, dadurch gekennzeichnet, daß der Koeffizient β der exponentiellen Dämpfung den Wert 0,5 hat.
  6. Verfahren nach einem beliebigen der vorstehenden Ansprüche, dadurch gekennzeichnet, daß es einen einleitenden Verfahrensschritt (0) enthält, der darin besteht, die verrauschten Signale (u(t)) durch Tastung zu digitalisieren, wobei jeder Rahmen p Tastproben enthält.
  7. Verfahren nach Anspruch 6, dadurch gekennzeichnet, daß das Rauschmodell (1) sich durch eine mehrfach wiederholte Suche ergibt, die permanent in den verrauschten Signalen (u(t)) durchgeführt wird, indem N aufeinanderfolgende Rahmen mit je p Tastproben gesucht werden, die Merkmale besitzen, welche für ein Rauschen erwartet werden, indem die NP entsprechenden Tastproben gespeichert werden, um das Rauschmodell zu bilden, und indem die Suche wiederholt wird, um ein neues Rauschmodell zu finden, und, je nach den Merkmalen der beiden Modelle, dieses neue Rauschmodell anstelle des vorausgegangenen zu speichern oder das vorausgegangene Modell beizubehalten.
  8. Verfahren nach einem beliebigen der vorstehenden Ansprüche, bei dem die verrauschten akustischen Signale verrauschte Sprachsignale (u(t)) sind.
  9. Verfahren nach Anspruch 8, dadurch gekennzeichnet, daß die Dauer der Rahmen in einem Bereich zwischen 10 und 20 ms liegt.
EP98402894A 1997-11-21 1998-11-20 Verfahren zur Frequenzfilterung mittels eines Wiener Filters für die Geräuschunterdrückung von Audiosignalen Expired - Lifetime EP0918317B1 (de)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
FR9714641A FR2771542B1 (fr) 1997-11-21 1997-11-21 Procede de filtrage frequentiel applique au debruitage de signaux sonores mettant en oeuvre un filtre de wiener
FR9714641 1997-11-21

Publications (2)

Publication Number Publication Date
EP0918317A1 EP0918317A1 (de) 1999-05-26
EP0918317B1 true EP0918317B1 (de) 2003-08-27

Family

ID=9513645

Family Applications (1)

Application Number Title Priority Date Filing Date
EP98402894A Expired - Lifetime EP0918317B1 (de) 1997-11-21 1998-11-20 Verfahren zur Frequenzfilterung mittels eines Wiener Filters für die Geräuschunterdrückung von Audiosignalen

Country Status (5)

Country Link
US (1) US6445801B1 (de)
EP (1) EP0918317B1 (de)
JP (1) JPH11265198A (de)
DE (1) DE69817507D1 (de)
FR (1) FR2771542B1 (de)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7313518B2 (en) 2001-01-30 2007-12-25 France Telecom Noise reduction method and device using two pass filtering

Families Citing this family (33)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6633842B1 (en) * 1999-10-22 2003-10-14 Texas Instruments Incorporated Speech recognition front-end feature extraction for noisy speech
FR2786308B1 (fr) * 1998-11-20 2001-02-09 Sextant Avionique Procede de reconnaissance vocale dans un signal acoustique bruite et systeme mettant en oeuvre ce procede
FI19992453A (fi) * 1999-11-15 2001-05-16 Nokia Mobile Phones Ltd Kohinanvaimennus
AU2607601A (en) * 1999-12-30 2001-07-16 Comlink 3000 Electromagnetic matched filter based multiple access communications systems
EP1132896A1 (de) * 2000-03-08 2001-09-12 Motorola, Inc. Verfahren zur Frequenzfilterung mittels eines Wiener Filters für die Geräuschunterdrückung von akustischen Signalen
US6766292B1 (en) * 2000-03-28 2004-07-20 Tellabs Operations, Inc. Relative noise ratio weighting techniques for adaptive noise cancellation
FR2808917B1 (fr) * 2000-05-09 2003-12-12 Thomson Csf Procede et dispositif de reconnaissance vocale dans des environnements a niveau de bruit fluctuant
JP4560899B2 (ja) * 2000-06-13 2010-10-13 カシオ計算機株式会社 音声認識装置、及び音声認識方法
EP1170728A1 (de) * 2000-07-05 2002-01-09 Alcatel System zur angepassten Reduktion von Geräuschen bei Sprachsignalen
US6990446B1 (en) * 2000-10-10 2006-01-24 Microsoft Corporation Method and apparatus using spectral addition for speaker recognition
EP1202602B1 (de) * 2000-10-25 2013-05-15 Panasonic Corporation Mikrophonvorrichtung mit Zoom-Einstellung
US7289626B2 (en) * 2001-05-07 2007-10-30 Siemens Communications, Inc. Enhancement of sound quality for computer telephony systems
EP1278185A3 (de) * 2001-07-13 2005-02-09 Alcatel Verfahren zur Verbesserung von Geräuschunterdrückung bei der Sprachübertragung
DE10137348A1 (de) * 2001-07-31 2003-02-20 Alcatel Sa Verfahren und Schaltungsanordnung zur Geräuschreduktion bei der Sprachübertragung in Kommunikationssystemen
US6959276B2 (en) * 2001-09-27 2005-10-25 Microsoft Corporation Including the category of environmental noise when processing speech signals
FR2842064B1 (fr) * 2002-07-02 2004-12-03 Thales Sa Systeme de spatialisation de sources sonores a performances ameliorees
AU2002950530A0 (en) * 2002-08-01 2002-09-12 Lake Technology Limited Approximation sequence processing
US7295250B2 (en) * 2003-07-31 2007-11-13 Broadcom Corporation Apparatus and method for restoring DC spectrum for analog television reception using direct conversion turners
WO2005050623A1 (en) * 2003-11-12 2005-06-02 Telecom Italia S.P.A. Method and circuit for noise estimation, related filter, terminal and communication network using same, and computer program product therefor
US7725314B2 (en) * 2004-02-16 2010-05-25 Microsoft Corporation Method and apparatus for constructing a speech filter using estimates of clean speech and noise
GB2414646B (en) * 2004-03-31 2007-05-02 Meridian Lossless Packing Ltd Optimal quantiser for an audio signal
US7516069B2 (en) * 2004-04-13 2009-04-07 Texas Instruments Incorporated Middle-end solution to robust speech recognition
EP1789956B1 (de) * 2004-09-16 2010-08-04 France Telecom Verfahren zum verarbeiten eines rauschbehafteten tonsignals und einrichtung zur implementierung des verfahrens
US7760887B2 (en) * 2004-10-15 2010-07-20 Lifesize Communications, Inc. Updating modeling information based on online data gathering
WO2007130765A2 (en) * 2006-05-04 2007-11-15 Sony Computer Entertainment Inc. Echo and noise cancellation
US7912567B2 (en) * 2007-03-07 2011-03-22 Audiocodes Ltd. Noise suppressor
JP5152799B2 (ja) * 2008-07-09 2013-02-27 国立大学法人 奈良先端科学技術大学院大学 雑音抑圧装置およびプログラム
JP5152800B2 (ja) * 2008-07-09 2013-02-27 国立大学法人 奈良先端科学技術大学院大学 雑音抑圧評価装置およびプログラム
EP2151821B1 (de) * 2008-08-07 2011-12-14 Nuance Communications, Inc. Rauschunterdrückende Verarbeitung von Sprachsignalen
JP5376635B2 (ja) * 2009-01-07 2013-12-25 国立大学法人 奈良先端科学技術大学院大学 雑音抑圧処理選択装置，雑音抑圧装置およびプログラム
US9886968B2 (en) * 2013-03-04 2018-02-06 Synaptics Incorporated Robust speech boundary detection system and method
US11335312B2 (en) 2016-11-08 2022-05-17 Andersen Corporation Active noise cancellation systems and methods
CN112384973A (zh) 2018-05-04 2021-02-19 安德森公司 针对噪声衰减的多频带频率

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0525408A3 (en) * 1991-07-01 1993-12-22 Eastman Kodak Co Method for multiframe wiener restoration of noisy and blurred image sequences
FR2681715B1 (fr) * 1991-09-25 1994-02-11 Matra Communication Procede de traitement de la parole en presence de bruits acoustiques: procede de soustraction spectrale non lineaire .
SG49334A1 (en) * 1993-12-06 1998-05-18 Koninkl Philips Electronics Nv A noise reduction system and device and a mobile radio station
SE505156C2 (sv) * 1995-01-30 1997-07-07 Ericsson Telefon Ab L M Förfarande för bullerundertryckning genom spektral subtraktion
DE69631955T2 (de) * 1995-12-15 2005-01-05 Koninklijke Philips Electronics N.V. Verfahren und schaltung zur adaptiven rauschunterdrückung und sendeempfänger
FR2744277B1 (fr) 1996-01-26 1998-03-06 Sextant Avionique Procede de reconnaissance vocale en ambiance bruitee, et dispositif de mise en oeuvre

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7313518B2 (en) 2001-01-30 2007-12-25 France Telecom Noise reduction method and device using two pass filtering

Also Published As

Publication number Publication date
FR2771542A1 (fr) 1999-05-28
EP0918317A1 (de) 1999-05-26
JPH11265198A (ja) 1999-09-28
US6445801B1 (en) 2002-09-03
FR2771542B1 (fr) 2000-02-11
DE69817507D1 (de) 2003-10-02

Similar Documents

Publication Publication Date Title
EP0918317B1 (de) Verfahren zur Frequenzfilterung mittels eines Wiener Filters für die Geräuschunterdrückung von Audiosignalen
EP0993671B1 (de) Verfahren zur bestimmung eines rauschmodells in einem gestörten audiosignal
EP1789956B1 (de) Verfahren zum verarbeiten eines rauschbehafteten tonsignals und einrichtung zur implementierung des verfahrens
EP1154405B1 (de) Verfahren und Vorrichtung zur Spracherkennung in einer Umgebung mit variablerem Rauschpegel
EP1356461B1 (de) Rauschverminderungsverfahren und -einrichtung
EP1016072B1 (de) Verfahren und vorrichtung zur rauschunterdrückung eines digitalen sprachsignals
EP0594480B1 (de) Verfahren zur Erkennung von Sprachsignalen
EP2772916B1 (de) Verfahren zur Geräuschdämpfung eines Audiosignals mit Hilfe eines Algorithmus mit variabler Spektralverstärkung mit dynamisch modulierbarer Härte
FR3012928A1 (fr) Modificateurs reposant sur un snr estime exterieurement pour des calculs internes de mmse
EP1131813B1 (de) Verfahren und vorrichtung zur spracherkennung eines mit störungen behafteten akustischen signals
EP1016071B1 (de) Verfahren und vorrichtung zur sprachdetektion
FR3012927A1 (fr) Estimation precise du rapport signal a bruit par progression reposant sur une probabilite de la presence de la parole mmse
EP1016073B1 (de) Verfahren und vorrichtung zur rauschunterdrückung eines digitalen sprachsignals
EP3627510A1 (de) Filterung eines tonsignals, das durch ein stimmerkennungssystem erfasst wurde
EP2515300B1 (de) Verfahren und System für die Geräuschunterdrückung
EP1021805B1 (de) Verfahren und vorrichtung zur verbesserung eines digitalen sprachsignals
WO1999027523A1 (fr) Procede de reconstruction, apres debruitage, de signaux sonores
EP4315328A1 (de) Schätzung einer optimierten maske zur verarbeitung von erfassten tondaten
EP4163866A1 (de) Verfahren, computerprogramm und vorrichtung zur bildverarbeitung durch extraktion der hauptkomponente(n) einer bilddarstellung
CA2389785A1 (fr) Procede et systeme de pre et de post-traitement d'un signal audio pour la transmission sur un canal fortement perturbe
WO2002093553A1 (fr) Estimation de periodes fondamentales de sources concurrentes multiples notamment de son

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): DE FR GB NL

AX Request for extension of the european patent

Free format text: AL;LT;LV;MK;RO;SI

17P Request for examination filed

Effective date: 19990723

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

Owner name: THOMSON-CSF SEXTANT

AKX Designation fees paid

Free format text: DE FR GB NL

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

Owner name: THALES AVIONICS S.A.

GRAH Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOS IGRA

RIC1 Information provided on ipc code assigned before grant

Ipc: 7G 10L 21/02 A

GRAH Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOS IGRA

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

AK Designated contracting states

Designated state(s): DE FR GB NL

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

Ref country code: NL

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: 20030827

Ref country code: GB

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: 20030827

REG Reference to a national code

Ref country code: GB

Ref legal event code: FG4D

Free format text: NOT ENGLISH

REF Corresponds to:

Ref document number: 69817507

Country of ref document: DE

Date of ref document: 20031002

Kind code of ref document: P

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

Ref country code: DE

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: 20031128

NLV1 Nl: lapsed or annulled due to failure to fulfill the requirements of art. 29p and 29m of the patents act
GBV Gb: ep patent (uk) treated as always having been void in accordance with gb section 77(7)/1977 [no translation filed]

Effective date: 20030827

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

26N No opposition filed

Effective date: 20040528

REG Reference to a national code

Ref country code: FR

Ref legal event code: PLFP

Year of fee payment: 18

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

Ref country code: FR

Payment date: 20151023

Year of fee payment: 18

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST

Effective date: 20170731

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

Ref country code: FR

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

Effective date: 20161130