[go: up one dir, main page]

EP1386307B1 - Procede et dispositif pour determiner un niveau de qualite d'un signal audio - Google Patents

Procede et dispositif pour determiner un niveau de qualite d'un signal audio Download PDF

Info

Publication number
EP1386307B1
EP1386307B1 EP02703438A EP02703438A EP1386307B1 EP 1386307 B1 EP1386307 B1 EP 1386307B1 EP 02703438 A EP02703438 A EP 02703438A EP 02703438 A EP02703438 A EP 02703438A EP 1386307 B1 EP1386307 B1 EP 1386307B1
Authority
EP
European Patent Office
Prior art keywords
signal
audio signal
quality
determining
interruptions
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
EP02703438A
Other languages
German (de)
English (en)
Other versions
EP1386307B2 (fr
EP1386307A1 (fr
Inventor
Pero Juric
Bendicht Thomet
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.)
Swissqual License AG
Original Assignee
SwissQual 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
Family has litigation
First worldwide family litigation filed litigation Critical https://patents.darts-ip.com/?family=8183803&utm_source=google_patent&utm_medium=platform_link&utm_campaign=public_patent_search&patent=EP1386307(B1) "Global patent litigation dataset” by Darts-ip is licensed under a Creative Commons Attribution 4.0 International License.
Application filed by SwissQual AG filed Critical SwissQual AG
Priority to EP02703438.8A priority Critical patent/EP1386307B2/fr
Publication of EP1386307A1 publication Critical patent/EP1386307A1/fr
Publication of EP1386307B1 publication Critical patent/EP1386307B1/fr
Application granted granted Critical
Publication of EP1386307B2 publication Critical patent/EP1386307B2/fr
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
    • G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
    • G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
    • G10L25/69—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for evaluating synthetic or decoded voice signals

Definitions

  • the invention relates to a method for determining a quality measure of an audio signal. Furthermore, the invention relates to a device for carrying out this method and a noise suppression module and an interrupt detection and interpolation module for use in such a device.
  • the To judge service quality of a telecommunication network is quality to determine a signal transmitted via the telecommunication network.
  • quality to determine a signal transmitted via the telecommunication network For audio signals, In particular with speech signals, various intrusive methods are known for this purpose. In such methods, as the name suggests, in the system under test intervened by occupying a transmission channel and transmitting therein a reference signal becomes. The quality assessment is then carried out by comparing the known Reference signal with the received signal, for example, subjectively by a or a plurality of test persons. However, this is expensive and therefore expensive.
  • EP 0 980 064 is another intrusive method for machine-aided quality assessment an audio signal, wherein for assessing the transmission quality a spectral similarity value of the known source signal and the received signal is determined. This similarity value is based on a calculation of the covariance the spectra of the source signal and the received signal and a division of the covariance by the standard deviations of the two spectra.
  • intrusive methods generally have the disadvantage that, as already mentioned in the zu testing system must be intervened. To determine the signal quality must namely occupies at least one transmission channel and transmits therein a reference signal become. This transmission channel can not during this time for a data transmission be used. In addition, it is in a broadcasting system such as a broadcasting service in principle possible, the signal source for transmission occupied by test signals, but since this occupies all channels and the test signal to This procedure is extremely impractical. Intrusive Methods are also inappropriate for simultaneously maintaining the quality of a variety of transmission channels to monitor.
  • EP-A-644 526 discloses a non-intrusive process for Noise reduction, which is used to calculate the desired Signal information uses an estimate of the noise energy.
  • the object of the invention is to provide a method of the type mentioned above, which avoids the disadvantages of the prior art and in particular offers a possibility to assess the signal quality of a transmitted over a telecommunications network Signals without knowledge of the originally transmitted signal.
  • a reference signal is first determined from the audio signal. through Comparing the determined reference signal with the audio signal becomes a quality value determined, which is used to determine the quality measure.
  • the inventive method thus allows an assessment of the quality of an audio signal at any terminal of the telecommunication network. Ie. it allows so that the quality assessment of many transmission channels simultaneously, even a simultaneous assessment of all channels would be possible.
  • the quality assessment takes place solely on the basis of the characteristics of the received signal, d. H. without knowledge of the source signal or the signal source.
  • the invention thus not only enables monitoring of the transmission quality of the Telecommunications network, but also, for example, a quality-based cost allocation, a quality-based routing in the network, a test of the coverage ratio
  • a QOS Quality of Service
  • a transmitted over a telecommunications network audio signal has next to the desired Signal information also typically unwanted components such as different noise components which are not in the original source signal were present.
  • the reference signal is determined by the in the received signal received existing Störsignalanmaschine and then from the received signal are removed. By removing the noise from the audio signal are first determined a noisy audio signal, which is preferred as Reference signal is used to assess the transmission quality.
  • the audio signal could, for example, be passed through appropriate filters.
  • a neural network is used for this purpose.
  • the audio signal is not used directly as an input signal.
  • DWT discrete wavelet transform
  • This transformation provides a plurality of DWT coefficients of the audio signal corresponding to the neural Network are supplied as input signal.
  • the neural network delivers at Output a plurality of corrected DWT coefficients, from which with the inverse DWT the reference signal is obtained. This corresponds to the noisy version of the Audio signal.
  • the coefficients of the neural network must be set in this way be that this to the DWT coefficients of a noisy input signal provides the DWT coefficients of the corresponding noisy input signal.
  • the neural network In order to the neural network provides the desired coefficients, it must first with a Set trained by corresponding noisy or noisy signal pairs become.
  • any other information in addition to the quality value provided by the Comparison of the received audio signal determined with the reference signal determined therefrom will be considered, any other information. This can both Information contained in the audio signal, as well as information about the transmission channel or the telecommunications network itself.
  • the quality of the received audio signal for example, by the at Transmission codecs (coder - decoder) influenced. It is difficult to do such Detect signal degradation, for example, at too small codec bit rates a part of the original signal information is lost. However, they are too small Codec bit rates result in a change in the fundamental frequency (pitch) of the audio signal, why examined with advantage the course and the dynamics of the fundamental frequency in the audio signal becomes. Since such changes are easiest based on audio signal sections With vocals, it is first preferable to use signal components in the audio signal detected with vowels and then examined for pitch variations.
  • the received audio signal can namely not only have unwanted signal components, it can also partially on the way desired information has been lost. So can the received audio signal for example, have more or less long signal interruptions.
  • the received audio signal may include various types of audio signals. So For example, it can contain voice, music, noise or silence signals.
  • the quality assessment can be based on all or part of it Signal components take place. In a preferred variant of the invention, the assessment the signal quality, however, limited to the speech signal components.
  • the speech signal components are first extracted from the audio signal and only these speech signal components for determining the quality measure, i. H. to Determination of the reference signal used. To determine the quality value is in In this case, the determined reference signal, of course, not with the received audio signal, but compared only with the voice signal component extracted therefrom.
  • the inventive device for machine-aided determination of a quality measure an audio signal comprises first means for determining a reference signal the audio signal, second means for determining a quality value by means of comparisons the determined reference signal with the audio signal and third means for determining the quality measure taking into account the quality value.
  • the first means for determining a reference signal from the audio signal can be several Include modules. So is preferably a noise suppression module and / or a Interrupt detection and interpolation module provided.
  • noise signal components can be received in the Suppress audio signal. It contains the means to carry out the already described Wavelet transforms and the neural network to determine the new DWT coefficients.
  • the interrupt detection and interpolation module has those Means, on the one hand for detecting signal interruptions in the audio signal and on the other hand, for the polynomial interpolation of short and model-based interpolation be required by medium-length signal interruptions. The determined so Reference signal thus corresponds to a noisy version of the received audio signal and typically has only larger signal interruptions.
  • the information about the signal interruptions of the audio signal is not only used to determine a better reference signal, they can also be used to determine of a better quality.
  • the third means of determination of the quality measure are therefore preferably designed such that information can be taken into account via signal interruptions in the audio signal.
  • the device therefore advantageously has fourth means for determining information on codec-related Signal distortions on.
  • codec-related Signal distortions include, for example, a vocal detection module, with which signal components with vowels can be detected in the audio signal. These vowel signal components will be passed on to an evaluation module, which is based on this Signal components
  • Information about codec-related signal distortions determines which also be used to assess the signal quality.
  • the third funds are corresponding designed such that this information about the codec-related signal distortions can be taken into account when determining the quality measure.
  • the device has, in particular, fifth means for extracting the device Speech signal components from the audio signal. Accordingly, to determine the Reference signal not the audio signal itself, but only the voice signal component noisy and checked for interruptions. Likewise, of course, not the audio signal, but only the voice signal component compared with this reference signal. In order to the determination of the quality measure is based only on the information in Voice signal component, wherein the information from the remaining signal components is not taken into account become.
  • FIG. 1 shows a block diagram of the method according to the invention.
  • a Audio signal 1 determines a quality measure 2, which, for example, also for evaluation the used (not shown) telecommunications network can be used.
  • the audio signal 1 is here understood to mean the signal which is a receiver after transmission via the telecommunication network.
  • This audio signal 1 Namely, typically does not match the one sent by the transmitter (not shown) Signal match, because on the way from the transmitter to the receiver, the transmission signal varied way changed. For example, it goes through different modules such as speech coders and decoders, multiplexers and demultiplexers or even speech enhancers and echo cancellers. But even the transmission channel itself can be a big Have an influence on the signal, which occurs, for example, in the form of interference, fading, Transmission off or interruptions, echo generation, etc. express.
  • the audio signal 1 thus contains not only desired signal components, d. H. the original one Transmission signal, but also unwanted interference signal components. It can also be that Signal portions of the transmission signal are missing, d. H. lost during the transmission are.
  • the evaluation of signal quality is not based on the entire audio signal 1, but only on the basis of the contained therein Speech portion.
  • the audio signal 1 is first recorded with an audio discriminator 3 Voice signal parts 4 examined out. Found speech signal components 4 become further Processing, whereas other signal components such as music 5.1, breaks 5.2 or severe signal interference 5.3 sorted out and otherwise processed or can be discarded.
  • the audio signal 1 piecewise, d. H. to pieces a each about 100 ms to 500 ms, passed to the audio discriminator 3. This decomposes these pieces further in single buffer of about 20 ms in length, processes these buffers and then assigns them each one of the signal groups to be distinguished speech signal, music, pause or strong interference to.
  • the audio discriminator 3 uses, for example, to judge the signal chips an LPC (linear predictive coding) transformation, which uses the coefficients of a the adaptive filter corresponding to the human language tract.
  • LPC linear predictive coding
  • the Assignment of the signal pieces to the different signal groups is based on the Shape of the transmission characteristics of this filter.
  • this voice signal component becomes 4 now a reference signal 6, d. H. the best possible estimate of the sender originally transmitted transmission signal determined.
  • This reference signal estimation takes place in several stages.
  • a noise suppression module 7 are initially undesirable Signal components such as stationary noise or impulse noise from the speech signal component 4 removed or suppressed. This is done with the help of a neural network, which previously by means of a variety of noisy signals as input and each train the corresponding noise-free version of the input signal as a target signal has been. The thus obtained, noisy speech signal 11 is sent to the second stage forwarded.
  • the interruption detection and interpolation module 8 interruptions detected in the audio signal 1 or in the voice signal portion 4 and if possible interpolated, d. H. the missing samples are replaced by appropriately estimated values.
  • the detection of signal interruptions by means of an investigation discontinuities of the signal fundamental frequency (pitch-tracing).
  • the interpolation is performed depending on the length of the detected interruption.
  • model-based interpolations such as a maximum a posteriori, an autoregressive or a frequency-time interpolation applied.
  • For longer Signal interruptions is an interpolation or other signal reconstruction in usually no longer possible in a meaningful way.
  • the comparison module 9 After determining the reference signal 6 with the noise suppression module 7 and the interruption detection and interpolation module 8 it is using the comparison module 9 compared with the voice signal component 4.
  • This comparison can be an algorithm used, for example, in intrusive procedures for comparison the known source signal is used with the received signal. Suitable are, for example, psychoacoustic models, the signals perceptive, d. H. perceptible to compare.
  • the result of this comparison is an intrusive quality value 10.
  • This intrusive quality value 10 the input signals, so the Voice signal component 4 and the reference signal 6, in signal pieces of about 20 to 30 ms Length decomposes and calculates a partial quality value for each signal piece. After about 20 to 30 signal pieces, which corresponds approximately to a signal duration of 0.5 seconds, becomes the intrusive Quality value 10 is determined as the arithmetic mean of these partial quality values. Of the intrusive quality value 10 forms the output signal of the comparison module 9.
  • the transmitted signal on its way from the transmitter to the receiver has an influence on the audio signal 1.
  • These influences exist, for example in that both the fundamental frequency and the higher harmonic frequencies vary the signal. The smaller the bit rate of the speech codecs used, the greater the frequency shifts and thus the signal distortions.
  • the evaluation module 14 divides the vocal signal 13 into signal pieces of about 30 ms and calculates a respective DFT (discrete Fourier transformation) with a frequency resolution of about 2 Hz at a sampling frequency of about 8 kHz. Leave it then determine the fundamental frequency and the higher harmonic frequencies and look for variations. Another feature for evaluating the codec-related Distortion forms the dynamics of the signal spectrum, with a smaller dynamics a poorer signal quality means.
  • the reference values for the dynamic assessment are obtained for the individual vowels from example signals. From the information on the influence of codecs on frequency shifts and spectrum dynamics of the audio signal 1 and the denoised voice signal 11 becomes a codec quality value 15 derived.
  • intrusive quality value 10 and codec quality value 15 also have an interruption quality value 17 taken into account.
  • This value includes information about the length and the number of interruptions detected by the interruption detection and interpolation module 8, in a preferred embodiment of the invention, only the information be taken into account over the long breaks.
  • quality information 18 on the received audio signal 1 or the denoised speech signal 11, which is determined with other modules or examinations will be included in the calculations of quality standard 2.
  • the individual quality values are now scaled such that they are in the range of numbers between 0 and 1, where a quality value of 1 is undiminished quality and Values below 1 indicate a correspondingly reduced quality.
  • the quality measure 2 is finally calculated as a linear combination of the individual quality values, whereby the individual weighting coefficients are determined experimentally and determined that their sum is 1.
  • figure 2 shows the noise suppression module 7.
  • the speech signal component 4 of the audio signal 1 is first subjected to a known DWT 19 (discrete wavelet transformation).
  • DWT's are similar to DFT's used for signal analysis.
  • An essential Difference, however, unlike the ones used in a DFT, is indefinite and thus temporally unlocated sine or cosine waveforms, the use of so-called wavelets, d. H. temporally limited and thus temporally localized Waveforms with mean 0.
  • the speech signal component 4 is divided into signal pieces of about 20 ms to 30 ms, which each of the DWT 19 are subjected.
  • the result of DWT 19 is a set of DWT coefficients 20.1, which is fed as input vector to a neural network 20 become. Its coefficients have previously been trained to become a given Set of DWT coefficients 20.1 of a noisy signal a new set of DWT coefficients 20.2 provide the noisy version of this signal.
  • This new set of DWT coefficient 20.2 will now be sent to IDWT 21, i. H. subjected to the DWT 19 inverse DWT.
  • This IDWT 21 delivers in this way a mostly unencumbered version of the Speech signal portions 4, just the desired, denoised speech signal 11th
  • the training configuration of the neural network 20 is shown in FIG. It is with Training pairs of noisy and noisy versions of sample signals.
  • One unencumbered example signal 22.1 is subjected to the DWT 19 and it becomes a first Theorem 20.3 obtained from DWT coefficients.
  • Even the noisy sample signal 22.2 is subjected to the same DWT 19 and generates a second set 20.4 of DWT coefficients, which is fed into the neural network 20.
  • the output vector of the neural Network 20, the new DWT coefficients 20.5 is placed in a comparator 23 with the first one Theorem 20.3 compared with DWT coefficients. Because of the differences between These two sets of DWT coefficients are corrected 24 of the coefficients of the neural network 20.
  • example signals 22.1, 22.2 uses which human sounds from different Represent languages. It is also beneficial for women as well as women To use male and female voices.
  • the mentioned size of the individually to be processed Signal pieces from 20 ms to 30 ms duration are selected so that the processing of the Voice signal portion 4 are performed regardless of the language and the speaker can. Also pauses and very quiet signal sections are trained, so even these are recognized correctly.
  • a multi-layer perceptron was used with an input layer 25, a hidden layer 26 and a Starting layer 27 used.
  • the perceptron was trained with a backpropagation algorithm.
  • the input layer 25 has a plurality of input neurons 25.1, the hidden layer 26 a plurality of hidden neurons 26.1 and the Output layer 27 on a plurality of output neurons 27.1. Every input neuron 25.1 becomes one of the DWT coefficients 20.1 of the preceding DWT 19 fed.
  • the respective values are determined by the set coefficients of the respective neurons and the value combinations are calculated in each neuron supplies each output neuron 27.1 one of the new DWT coefficients 20.2.
  • the audio discriminator 3 decomposes the signal pieces into individual buffers of Length 20 ms. At a sampling rate of 8 kHz, this corresponds to 160 samples.
  • this Case may be, for example, a neural network 20 with 160 input and output neurons each 25.1, 27.1 and about 50 to 60 hidden neurons 26.1 are used.
  • a time-frequency interpolation is used for the signal reconstruction.
  • Length 8 ms
  • the goal of interpolation is to address this gap.
  • Figure 5 shows such a signal 28 of about 200 samples in length.
  • Figure 5 shows the signal 28 in the temporal domain easier to recognize.
  • On the abscissa axis 32 are the number of samples and on the ordinate axis 33 the magnitudes applied.
  • the interpolation is done in the frequency-time domain.
  • the interruption 29 is easy to recognize as a gap of almost 10 samples.
  • the pitch period 30 of the signal 28 is determined.
  • the interpolation will be information from the samples before and after the gap within this pitch period 30 is taken into account.
  • the signal areas 31.1, 31.2 show those Ranges of the signal 28 each have a pitch period before or after the interruption 29.
  • This Signal ranges 31.1, 31.2 are not identical to the original signal piece at break 29, but still show a high degree of similarity. For small Gaps up to about 10 samples are believed to still provide enough signal information is present in order to be able to carry out a correct interpolation. For longer gaps Additional information from samples of the environment can be used.
  • the invention allows the signal quality of a Judge received audio signal without knowing the original transmission signal. From the signal quality can of course on the quality of the transmission channels used and thus closed on the service quality of the entire telecommunications network become.
  • the fast response times of the inventive method which are on the order of about 100 ms to 500 ms, thus allowing different Applications such as general comparisons of service quality of various Networks or subnets, quality-based cost allocation or quality-based Routing in a network or across multiple networks by means of appropriate Control of network nodes (gateways, routers etc.).

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Computational Linguistics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Transmission Systems Not Characterized By The Medium Used For Transmission (AREA)
  • Detection And Prevention Of Errors In Transmission (AREA)
  • Noise Elimination (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
  • Testing Electric Properties And Detecting Electric Faults (AREA)

Claims (13)

  1. Procédé pour la détermination, assistée par ordinateur, d'une mesure de qualité d'un signal audio, caractérisé en ce que l'on détermine à partir du signal audio un signal de référence qui représente une estimation d'un signal audio initialement émis, et en ce que l'on détermine, au moyen d'une comparaison du signal de référence au signal audio, une valeur de qualité qui est utilisée pour la détermination de la mesure de qualité.
  2. Procédé selon la revendication 1, caractérisé en ce que l'on détermine, en éliminant des composantes bruitées du signal audio, un signal audio non bruité et on utilise celui-ci comme signal de référence.
  3. Procédé selon la revendication 2, caractérisé en ce que l'on détermine le signal audio non bruité en soumettant le signal audio à une transformation d'ondelettes discrète dont les coefficients sont introduits dans un réseau neuronal ayant subi auparavant un apprentissage et dont les signaux de sortie sont soumis à la transformation d'ondelettes discrète inverse.
  4. Procédé selon la revendication 2 ou 3, caractérisé en ce que l'on détecte dans le signal audio non bruité des composantes de signal avec éléments vocaux, en ce que l'on en détermine des informations sur des distorsions de signal dues au codeur-décodeur et en ce que l'on prend en compte celles-ci lors de la détermination de la mesure de qualité.
  5. Procédé selon l'une des revendications 1 à 4, caractérisé en ce que l'on détecte des interruptions de signal dans le signal audio et en ce que l'on détermine le signal de référence en le reconstruisant au moins partiellement au niveau des interruptions de signal, le signal de référence étant reconstruit de préférence avec une interpolation polynomiale en cas d'interruptions de signal courtes et de préférence avec une interpolation basée sur un modèle en cas d'interruptions de signal moyennement longues.
  6. Procédé selon la revendication 5, caractérisé en ce que, lors de la détermination de la mesure de qualité, on prend en compte des informations sur les interruptions de signal.
  7. Procédé selon l'une des revendications 1 à 6, caractérisé en ce que, avant la détermination du signal de référence, on extrait une composante de signal vocale du signal audio, et en ce qu'on limite la détermination de la mesure de qualité à la composante de signal vocale.
  8. Dispositif pour la détermination, assistée par ordinateur, d'une mesure de qualité d'un signal audio, caractérisé en ce qu'il comporte des premiers moyens pour déterminer un signal de référence à partir du signal audio, des deuxièmes moyens pour déterminer une valeur de qualité au moyen d'une comparaison du signal de référence au signal audio et des troisièmes moyens pour déterminer la mesure de qualité en tenant compte de la valeur de qualité, le signal de référence représentant une estimation d'un signal audio initialement émis.
  9. Dispositif selon la revendication 8, caractérisé en ce que les premiers moyens comportent un module de suppression de bruit pour supprimer des composantes bruitées et/ou un module de détection d'interruptions et d'interpolation pour détecter et interpoler des interruptions de signal dans le signal audio, et en ce que les troisièmes moyens sont conçus de telle sorte que des interruptions de signal peuvent être prises en compte lors de la détermination de la mesure de qualité.
  10. Dispositif selon la revendication 8 ou 9, caractérisé en ce qu'il comporte des moyens pour déterminer des distorsions de signal dues au codeur-décodeur, ces moyens comprenant un module de détection d'éléments vocaux pour la détection de composantes de signal vocales dans le signal audio ainsi qu'un module d'évaluation pour la détermination des distorsions de signal dues au codeur-décodeur, les troisièmes moyens étant conçus de telle sorte que les distorsions de signal dues au codeur-décodeur peuvent être prises en compte lors de la détermination de la mesure de qualité.
  11. Dispositif selon l'une des revendications 8 à 10, caractérisé en ce qu'il comporte des moyens pour extraire du signal audio une composante de signal vocale, et en ce qu'il est conçu pour la détermination de la mesure de qualité de la composante de signal vocale.
  12. Dispositif selon la revendication 9, les premiers moyens comportant le module de suppression de bruit, caractérisé en ce que le module de suppression de bruit comporte des moyens pour la mise en oeuvre d'une transformation d'ondelettes discrète en vue du calcul de coefficients de signal d'un signal audio, un réseau neuronal en vue du calcul de coefficients de signal corrigés, ainsi que des moyens pour la mise en oeuvre d'une transformation d'ondelettes inverse des coefficients de signal corrigés en vue de la détermination du signal audio sans composante bruitée.
  13. Dispositif selon la revendication 9, les premiers moyens comportant le module de détection d'interruptions et d'interpolation, caractérisé en ce que le module de détection d'interruptions et d'interpolation comporte des moyens pour détecter des interruptions de signal dans un signal audio ainsi que des moyens pour interpoler des interruptions du signal audio, ces derniers étant conçus de préférence pour une interpolation polynomiale d'interruptions de signal courtes et pour une interpolation, basée sur un modèle, d'interruptions de signal moyennement longues.
EP02703438.8A 2001-03-20 2002-03-19 Procede et dispositif pour determiner un niveau de qualite d'un signal audio Expired - Lifetime EP1386307B2 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
EP02703438.8A EP1386307B2 (fr) 2001-03-20 2002-03-19 Procede et dispositif pour determiner un niveau de qualite d'un signal audio

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
EP01810285A EP1244094A1 (fr) 2001-03-20 2001-03-20 Procédé et dispositif de détermination de la qualité d'un signal audio
EP01810285 2001-03-20
PCT/CH2002/000164 WO2002075725A1 (fr) 2001-03-20 2002-03-19 Procede et dispositif pour determiner un niveau de qualite d'un signal audio
EP02703438.8A EP1386307B2 (fr) 2001-03-20 2002-03-19 Procede et dispositif pour determiner un niveau de qualite d'un signal audio

Publications (3)

Publication Number Publication Date
EP1386307A1 EP1386307A1 (fr) 2004-02-04
EP1386307B1 true EP1386307B1 (fr) 2005-02-09
EP1386307B2 EP1386307B2 (fr) 2013-04-17

Family

ID=8183803

Family Applications (2)

Application Number Title Priority Date Filing Date
EP01810285A Withdrawn EP1244094A1 (fr) 2001-03-20 2001-03-20 Procédé et dispositif de détermination de la qualité d'un signal audio
EP02703438.8A Expired - Lifetime EP1386307B2 (fr) 2001-03-20 2002-03-19 Procede et dispositif pour determiner un niveau de qualite d'un signal audio

Family Applications Before (1)

Application Number Title Priority Date Filing Date
EP01810285A Withdrawn EP1244094A1 (fr) 2001-03-20 2001-03-20 Procédé et dispositif de détermination de la qualité d'un signal audio

Country Status (5)

Country Link
US (1) US6804651B2 (fr)
EP (2) EP1244094A1 (fr)
AT (1) ATE289109T1 (fr)
DE (1) DE50202226D1 (fr)
WO (1) WO2002075725A1 (fr)

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7177430B2 (en) * 2001-10-31 2007-02-13 Portalplayer, Inc. Digital entroping for digital audio reproductions
US7746797B2 (en) * 2002-10-09 2010-06-29 Nortel Networks Limited Non-intrusive monitoring of quality levels for voice communications over a packet-based network
US20040167774A1 (en) * 2002-11-27 2004-08-26 University Of Florida Audio-based method, system, and apparatus for measurement of voice quality
GB2407952B (en) * 2003-11-07 2006-11-29 Psytechnics Ltd Quality assessment tool
US20050228655A1 (en) * 2004-04-05 2005-10-13 Lucent Technologies, Inc. Real-time objective voice analyzer
DE102004029421A1 (de) * 2004-06-18 2006-01-05 Rohde & Schwarz Gmbh & Co. Kg Verfahren und Vorrichtung zur Bewertung der Güte eines Signals
US7856355B2 (en) * 2005-07-05 2010-12-21 Alcatel-Lucent Usa Inc. Speech quality assessment method and system
WO2007098258A1 (fr) * 2006-02-24 2007-08-30 Neural Audio Corporation Système et procédé de conditionnement pour un codec audio
US8949120B1 (en) 2006-05-25 2015-02-03 Audience, Inc. Adaptive noise cancelation
US20080244081A1 (en) * 2007-03-30 2008-10-02 Microsoft Corporation Automated testing of audio and multimedia over remote desktop protocol
AU2009220198B2 (en) * 2008-03-04 2012-11-29 Cardiac Pacemakers, Inc. Implantable multi-length RF antenna
JP4327888B1 (ja) * 2008-05-30 2009-09-09 株式会社東芝 音声音楽判定装置、音声音楽判定方法及び音声音楽判定用プログラム
JP4327886B1 (ja) * 2008-05-30 2009-09-09 株式会社東芝 音質補正装置、音質補正方法及び音質補正用プログラム
WO2011010962A1 (fr) * 2009-07-24 2011-01-27 Telefonaktiebolaget L M Ericsson (Publ) Procédé, ordinateur, programme d’ordinateur et produit progiciel pour estimation de la qualité vocale
US20110178800A1 (en) 2010-01-19 2011-07-21 Lloyd Watts Distortion Measurement for Noise Suppression System
US9558755B1 (en) 2010-05-20 2017-01-31 Knowles Electronics, Llc Noise suppression assisted automatic speech recognition
US8239196B1 (en) * 2011-07-28 2012-08-07 Google Inc. System and method for multi-channel multi-feature speech/noise classification for noise suppression
US9640194B1 (en) 2012-10-04 2017-05-02 Knowles Electronics, Llc Noise suppression for speech processing based on machine-learning mask estimation
US9396738B2 (en) 2013-05-31 2016-07-19 Sonus Networks, Inc. Methods and apparatus for signal quality analysis
US9536540B2 (en) 2013-07-19 2017-01-03 Knowles Electronics, Llc Speech signal separation and synthesis based on auditory scene analysis and speech modeling
DE112015003945T5 (de) 2014-08-28 2017-05-11 Knowles Electronics, Llc Mehrquellen-Rauschunterdrückung
CN106816158B (zh) * 2015-11-30 2020-08-07 华为技术有限公司 一种语音质量评估方法、装置及设备
US10490206B2 (en) * 2016-01-19 2019-11-26 Dolby Laboratories Licensing Corporation Testing device capture performance for multiple speakers
US10283140B1 (en) * 2018-01-12 2019-05-07 Alibaba Group Holding Limited Enhancing audio signals using sub-band deep neural networks
TWI708243B (zh) * 2018-03-19 2020-10-21 中央研究院 於分散式語音辨識中基於小波轉換之語音特徵壓縮及重建系統與方法
CN115798506A (zh) * 2022-11-10 2023-03-14 维沃移动通信有限公司 语音处理方法、装置、电子设备及存储介质

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4897878A (en) * 1985-08-26 1990-01-30 Itt Corporation Noise compensation in speech recognition apparatus
DE3639753A1 (de) * 1986-11-21 1988-06-01 Inst Rundfunktechnik Gmbh Verfahren zum uebertragen digitalisierter tonsignale
US5446492A (en) * 1993-01-19 1995-08-29 Wolf; Stephen Perception-based video quality measurement system
DE4309985A1 (de) * 1993-03-29 1994-10-06 Sel Alcatel Ag Geräuschreduktion zur Spracherkennung
IT1272653B (it) * 1993-09-20 1997-06-26 Alcatel Italia Metodo di riduzione del rumore, in particolare per riconoscimento automatico del parlato, e filtro atto ad implementare lo stesso
US6122610A (en) * 1998-09-23 2000-09-19 Verance Corporation Noise suppression for low bitrate speech coder
JP4462766B2 (ja) * 1999-05-25 2010-05-12 アルゴレックス インコーポレイテッド マルチメディアおよび他の信号の万能型品質測定システム
US20020054685A1 (en) * 2000-11-09 2002-05-09 Carlos Avendano System for suppressing acoustic echoes and interferences in multi-channel audio systems
US6937978B2 (en) * 2001-10-30 2005-08-30 Chungwa Telecom Co., Ltd. Suppression system of background noise of speech signals and the method thereof

Also Published As

Publication number Publication date
ATE289109T1 (de) 2005-02-15
EP1244094A1 (fr) 2002-09-25
EP1386307B2 (fr) 2013-04-17
EP1386307A1 (fr) 2004-02-04
US20020191798A1 (en) 2002-12-19
WO2002075725A1 (fr) 2002-09-26
DE50202226D1 (de) 2005-03-17
US6804651B2 (en) 2004-10-12

Similar Documents

Publication Publication Date Title
EP1386307B1 (fr) Procede et dispositif pour determiner un niveau de qualite d'un signal audio
EP1088300B1 (fr) Procede d'execution d'une evaluation automatisee de la qualite de transmission de signaux audio
DE69614989T2 (de) Verfahren und Vorrichtung zur Feststellung der Sprachaktivität in einem Sprachsignal und eine Kommunikationsvorrichtung
DE69131883T2 (de) Vorrichtung zur Rauschreduzierung
DE69432943T2 (de) Verfahren und Vorrichtung zur Sprachdetektion
DE69131739T2 (de) Einrichtung zur Sprachsignalverarbeitung für die Bestimmung eines Sprachsignals in einem verrauschten Sprachsignal
DE69520067T2 (de) Verfahren und Einrichtung zur Kennzeichnung eines Eingangssignales
DE69420027T2 (de) Rauschverminderung
DE60034026T2 (de) Sprachverbesserung mit durch sprachaktivität gesteuerte begrenzungen des gewinnfaktors
DE2626793B2 (de) Elektrische Schaltungsanordnung zum Bestimmen des stimmhaften oder stimmlosen Zustandes eines Sprachsignals
EP0938831B1 (fr) Evaluation de la qualite, a adaptation auditive, de signaux audio
DE10017646A1 (de) Geräuschunterdrückung im Zeitbereich
EP1091349A2 (fr) Procédé et dispositif pour la réduction de bruit durant la transmission de parole
DE60311619T2 (de) Datenreduktion in Audiokodierern unter Ausnutzung nichtharmonischer Effekte
EP3197181A1 (fr) Procédé de réduction du temps de latence d'un banc de filtrage destiné au filtrage d'un signal audio et procédé de fonctionnement sans latence d'un système auditif
EP3065417B1 (fr) Procede de suppression d'un bruit parasite dans un systeme acoustique
EP1869671B1 (fr) Procede et dispositif pour attenuer le bruit
EP1382034B1 (fr) Procede de determination de valeurs caracteristiques d'intensite de bruits de fond dans des pauses de voix de signaux vocaux
DE60110541T2 (de) Verfahren zur Spracherkennung mit geräuschabhängiger Normalisierung der Varianz
DE3230391C2 (fr)
DE10150519B4 (de) Verfahren und Anordnung zur Sprachverarbeitung
DE4445983C2 (de) Verfahren zur Rauschunterdrückung und Vorrichtungen zur Durchführung der Verfahren
EP1130577B1 (fr) Procédé de reconstruction des basses fréquences du signal de parole à partir de fréquences moyennes
EP3962115B1 (fr) Procédé d'évaluation de la qualité de parole d'un signal vocal au moyen d'un dispositif auditif
DE102013005844B3 (de) Verfahren und Vorrichtung zum Messen der Qualität eines Sprachsignals

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

17P Request for examination filed

Effective date: 20030821

AK Designated contracting states

Kind code of ref document: A1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE TR

AX Request for extension of the european patent

Extension state: AL LT LV MK RO SI

GRAP Despatch of communication of intention to grant a patent

Free format text: ORIGINAL CODE: EPIDOSNIGR1

GRAS Grant fee paid

Free format text: ORIGINAL CODE: EPIDOSNIGR3

GRAA (expected) grant

Free format text: ORIGINAL CODE: 0009210

RIN1 Information on inventor provided before grant (corrected)

Inventor name: THOMET, BENDICHT

Inventor name: JURIC, PERO

AK Designated contracting states

Kind code of ref document: B1

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE TR

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

Ref country code: TR

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: 20050209

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: 20050209

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: 20050209

Ref country code: IE

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: 20050209

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: CH

Ref legal event code: EP

Ref country code: CH

Ref legal event code: NV

Representative=s name: KELLER & PARTNER PATENTANWAELTE AG

RAP2 Party data changed (patent owner data changed or rights of a patent transferred)

Owner name: SWISSQUAL LICENSE AG

REG Reference to a national code

Ref country code: IE

Ref legal event code: FG4D

Free format text: GERMAN

REF Corresponds to:

Ref document number: 50202226

Country of ref document: DE

Date of ref document: 20050317

Kind code of ref document: P

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

Ref country code: AT

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

Effective date: 20050319

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: 20050319

Ref country code: LU

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

Effective date: 20050319

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 NON-PAYMENT OF DUE FEES

Effective date: 20050331

NLT2 Nl: modifications (of names), taken from the european patent patent bulletin

Owner name: SWISSQUAL LICENSE AG

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: 20050509

Ref country code: SE

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: 20050509

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: 20050509

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

Ref country code: ES

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: 20050520

GBT Gb: translation of ep patent filed (gb section 77(6)(a)/1977)

Effective date: 20050516

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

Ref country code: IE

Payment date: 20050628

Year of fee payment: 4

NLV1 Nl: lapsed or annulled due to failure to fulfill the requirements of art. 29p and 29m of the patents act
REG Reference to a national code

Ref country code: IE

Ref legal event code: FD4D

PLBI Opposition filed

Free format text: ORIGINAL CODE: 0009260

PLAX Notice of opposition and request to file observation + time limit sent

Free format text: ORIGINAL CODE: EPIDOSNOBS2

ET Fr: translation filed
26 Opposition filed

Opponent name: ASCOM (SCHWEIZ) AG

Effective date: 20051102

REG Reference to a national code

Ref country code: CH

Ref legal event code: NV

Representative=s name: E. BLUM & CO. PATENTANWAELTE

PLAF Information modified related to communication of a notice of opposition and request to file observations + time limit

Free format text: ORIGINAL CODE: EPIDOSCOBS2

PLBB Reply of patent proprietor to notice(s) of opposition received

Free format text: ORIGINAL CODE: EPIDOSNOBS3

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

Ref country code: BE

Payment date: 20060929

Year of fee payment: 5

REG Reference to a national code

Ref country code: CH

Ref legal event code: PFA

Owner name: SPIRENT COMMUNICATIONS LICENSE AG

Free format text: SWISSQUAL LICENSE AG#METALLSTRASSE 9B#6300 ZUG (CH) -TRANSFER TO- SPIRENT COMMUNICATIONS LICENSE AG#METALLSTRASSE 9B#6300 ZUG (CH)

REG Reference to a national code

Ref country code: FR

Ref legal event code: CD

REG Reference to a national code

Ref country code: CH

Ref legal event code: PFA

Owner name: SPIRENT COMMUNICATIONS LICENSE AG

Free format text: SPIRENT COMMUNICATIONS LICENSE AG#METALLSTRASSE 9B#6300 ZUG (CH) -TRANSFER TO- SPIRENT COMMUNICATIONS LICENSE AG#METALLSTRASSE 9B#6300 ZUG (CH)

BERE Be: lapsed

Owner name: SWISSQUAL A.G.

Effective date: 20070331

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

Ref country code: BE

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

Effective date: 20070331

Ref country code: PT

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

Effective date: 20050709

APBP Date of receipt of notice of appeal recorded

Free format text: ORIGINAL CODE: EPIDOSNNOA2O

APAH Appeal reference modified

Free format text: ORIGINAL CODE: EPIDOSCREFNO

RAP2 Party data changed (patent owner data changed or rights of a patent transferred)

Owner name: SWISSQUAL LICENSE AG

APBQ Date of receipt of statement of grounds of appeal recorded

Free format text: ORIGINAL CODE: EPIDOSNNOA3O

RAP2 Party data changed (patent owner data changed or rights of a patent transferred)

Owner name: SWISSQUAL LICENSE AG

APBU Appeal procedure closed

Free format text: ORIGINAL CODE: EPIDOSNNOA9O

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

Ref country code: FR

Payment date: 20120403

Year of fee payment: 11

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

Ref country code: IT

Payment date: 20120328

Year of fee payment: 11

PUAH Patent maintained in amended form

Free format text: ORIGINAL CODE: 0009272

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

Free format text: STATUS: PATENT MAINTAINED AS AMENDED

27A Patent maintained in amended form

Effective date: 20130417

AK Designated contracting states

Kind code of ref document: B2

Designated state(s): AT BE CH CY DE DK ES FI FR GB GR IE IT LI LU MC NL PT SE TR

REG Reference to a national code

Ref country code: DE

Ref legal event code: R102

Ref document number: 50202226

Country of ref document: DE

REG Reference to a national code

Ref country code: DE

Ref legal event code: R082

Ref document number: 50202226

Country of ref document: DE

Representative=s name: GESTHUYSEN PATENT- UND RECHTSANWAELTE, DE

REG Reference to a national code

Ref country code: CH

Ref legal event code: PCOW

Free format text: NEW ADDRESS: ALLMENDWEG 8, 4528 ZUCHWIL (CH)

Ref country code: CH

Ref legal event code: AELC

Ref country code: CH

Ref legal event code: PFA

Owner name: SWISSQUAL LICENSE AG, CH

Free format text: FORMER OWNER: SPIRENT COMMUNICATIONS LICENSE AG, CH

Ref country code: CH

Ref legal event code: PCOW

Free format text: NEW ADDRESS: BAARERSTRASSE 78, 6300 ZUG (CH)

REG Reference to a national code

Ref country code: DE

Ref legal event code: R102

Ref document number: 50202226

Country of ref document: DE

Effective date: 20130417

REG Reference to a national code

Ref country code: DE

Ref legal event code: R082

Ref document number: 50202226

Country of ref document: DE

Representative=s name: GESTHUYSEN PATENT- UND RECHTSANWAELTE, DE

Effective date: 20130423

Ref country code: DE

Ref legal event code: R081

Ref document number: 50202226

Country of ref document: DE

Owner name: SWISSQUAL LICENSE AG, CH

Free format text: FORMER OWNER: SPIRENT COMMUNICATIONS LICENSE AG, ZUG, CH

Effective date: 20130423

REG Reference to a national code

Ref country code: FR

Ref legal event code: ST

Effective date: 20131129

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: 20130402

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

Ref country code: IT

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

Effective date: 20130319

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

Ref country code: CH

Payment date: 20210218

Year of fee payment: 20

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

Ref country code: GB

Payment date: 20210324

Year of fee payment: 20

Ref country code: DE

Payment date: 20210319

Year of fee payment: 20

REG Reference to a national code

Ref country code: DE

Ref legal event code: R071

Ref document number: 50202226

Country of ref document: DE

REG Reference to a national code

Ref country code: CH

Ref legal event code: PL

REG Reference to a national code

Ref country code: GB

Ref legal event code: PE20

Expiry date: 20220318

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

Ref country code: GB

Free format text: LAPSE BECAUSE OF EXPIRATION OF PROTECTION

Effective date: 20220318