EP4312215A1 - Verfahren zur echtzeitverarbeitung eines tonsignals und vorrichtung zur erfassung eines tonsignals - Google Patents

Verfahren zur echtzeitverarbeitung eines tonsignals und vorrichtung zur erfassung eines tonsignals Download PDF

Info

Publication number
EP4312215A1
EP4312215A1 EP23188331.5A EP23188331A EP4312215A1 EP 4312215 A1 EP4312215 A1 EP 4312215A1 EP 23188331 A EP23188331 A EP 23188331A EP 4312215 A1 EP4312215 A1 EP 4312215A1
Authority
EP
European Patent Office
Prior art keywords
sound signal
signal
sound
transformation
processing
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.)
Pending
Application number
EP23188331.5A
Other languages
English (en)
French (fr)
Inventor
Jean DEMARTINI
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.)
Uss Sensivic
Original Assignee
Uss Sensivic
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 Uss Sensivic filed Critical Uss Sensivic
Publication of EP4312215A1 publication Critical patent/EP4312215A1/de
Pending 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
    • G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
    • 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
    • G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
    • G10L25/21—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being power information

Definitions

  • the present invention relates to the capture and processing of a sound signal, with a view to its analysis.
  • the invention relates to the processing methods which are applied, in real time, to the signal captured by a sensor such as a microphone, before the steps of analyzing this signal.
  • the invention also relates to devices for capturing a sound signal, comprising at least one sensor and equipment for processing the sound signal measured by these sensors.
  • Some of these sound signal capture devices are integrated into devices intended to record, amplify or remotely transmit a sound signal audible to the human ear, representative of the sound signal picked up by a microphone integrated into the sound signal capture device.
  • the processing carried out on the signal picked up by the microphones generally aims to make the sound signal more easily audible.
  • processing may for example include frequency filters favoring the sound frequencies corresponding to the sounds that one wishes to transmit and/or record, in order to improve the signal/noise ratio of the sound signal.
  • the present invention does not apply to such sound signal capture devices intended to transmit and/or record an audible sound signal.
  • the present invention also does not apply to such devices for capturing sound signals, often called voice recognition devices.
  • the present invention does not apply to such devices for capturing sound signals, which relate to sound signals which do not include human words.
  • sound level meters which are devices for capturing a sound signal comprising a sensor such as a microphone associated with a processing device aimed at determining a standardized sound level, expressed in decibels, of the sound signal captured.
  • the real-time processing devices for the sound signal may include, in this case, frequency filters aimed at favoring the frequencies audible to the human ear, so that the measured sound level is representative of the perception of the sound level captured. by the human ear.
  • Such processing of the sound signal does not, however, make it possible to carry out, on the processed signal, an analysis making it possible to extract from this signal information other than that concerning the sound level.
  • the present invention does not apply to such devices for capturing sound signals which only allow superficial analyzes of the sound signal.
  • Devices for capturing a sound signal aimed at identifying a particular event or an anomaly, based on the sound signal captured are also known.
  • sound signal capture devices are known making it possible to capture ambient sound in a public space, and to identify, from the captured sound signal, a particular event or an anomaly.
  • Such equipment can for example be associated with video surveillance cameras, to particularly report to a video surveillance operator the images captured by the video surveillance cameras covering the areas in which a sound signal capture device has detected a sound corresponding to an event particular or an anomaly.
  • the present invention applies to such sound signal capture devices, intended to be placed so as to capture ambient sounds in a public space, and to allow analysis of the sound signal based on the physical data of this sound signal which do not contain language elements. It applies more particularly to such sensors aimed at identifying, from an ambient sound signal, particular events or anomalies.
  • such devices can be placed in locations such as public places, where their sound sensors, such as their microphones, can pick up information such as human voices.
  • the presence of such devices capable of capturing human voices may legitimately raise concerns about the confidentiality of conversations that would be picked up by the microphones.
  • Sound information such as the contents of conversations, or any information allowing the identification of a person based on a voice, is generally considered by legislation to be personal information, which is prohibited from recording. or transmit without the authorization of the persons concerned. It is therefore generally prohibits the use, in public places, of sound signal capture devices capable of recording or transmitting information containing voices, recognizable elements of language or similar signals, even if the device is not intended to carry out recognition of the words spoken or identification of the author of the voice.
  • the present invention aims to overcome these drawbacks of the prior art.
  • the present invention aims in particular to allow the use of devices for capturing a sound signal aimed at identifying a particular event or an anomaly, from the sound signal captured, under conditions such that it can be guaranteed that this device does not transmit or record any personal information contained in the sound signal it picks up.
  • Another objective of the invention is to provide a processing method, which can be implemented in a sound signal capture device, which makes it possible to process a sound signal picked up by a sensor such as a microphone in order to remove it. , in a reliable and irreversible manner, the information being able to be considered as personal information.
  • the invention also aims to provide such a method for processing a sound signal which is capable of carrying out in real time the processing of a sound signal picked up by the sensors.
  • the invention also aims to provide such a method of processing a sound signal which provides a processed signal, coming from the sound signal, containing sufficient information coming from the sound signal initially captured and not not necessarily relating to personal information, to enable the detection and categorization of particular sound events or anomalies.
  • Another objective of the invention is to provide a sound signal capture device, which is capable of capturing a sound signal and processing this sound signal in real time, such that the processed signal provided by this capture device , does not contain information that could be considered personal information.
  • Such a method applies processing for the anonymization of the sound signal.
  • This process makes it possible to obtain a processed signal presenting a large number of physical characteristics allowing its analysis, and in particular the conservation of all its frequency components, in which it is however not possible to find any information which could be considered such as personal information, such as elements of language, or characteristics allowing us to recognize a voice.
  • the processed signal is thus “anonymized”, that is to say it no longer contains the personal information that could be contained in the sound signal.
  • this transformation consists of associating its absolute value with each value of the sound signal.
  • this transformation consists of associating the square of its value with each value of the sound signal.
  • the method comprises digitizing the sound signal, and the transformation is applied to the digitized sound signal.
  • the analysis of the processed signal comprises at least one step of frequency analysis of the processed signal.
  • the analysis of the processed signal is carried out after recording this processed signal.
  • This anonymized processed signal can advantageously be analyzed subsequently to detect and/or categorize particular events or anomalies in the sound signal.
  • This anonymization method can advantageously present all the characteristics described in the present description for the method of processing a sound signal.
  • the invention also relates to a device for capturing a sound signal, comprising a pressure sensor and processing equipment, this processing equipment applying to the sound signal coming from the sensor a transformation isomorphic to the “absolute value” transformation.
  • Such a sound signal capture device makes it possible to provide a processed signal which presents a large part of the information of the sound signal, while not including information which could be considered as personal information.
  • such a sound signal capture device is constituted by a single electronic component, comprising a pressure sensor and an electronic chip ensuring the digitization of the signal coming from the sensor and the application to this signal of the isomorphic transformation to the transformation " absolute value ".
  • Such an electronic component bringing together a pressure sensor and a chip in the same package that can be implemented in an electronic circuit, is similar to components known as “analog microphones”. They are, however, modified, compared to these analog microphones, so that the chip, in addition to digitizing the sound signal coming from the sensor, applies a transformation isomorphic to the “absolute value” transformation to this signal.
  • FIG. 1 represents, schematically, a device for capturing, processing and analyzing a sound signal intended to detect and categorize, in this sound signal, a particular event or an anomaly.
  • This device comprises a sensor 11, for example constituted by a microphone, capable of capturing a sound and generating a signal 110, subsequently called "sound signal", representative of this sound.
  • a sensor 11 for example constituted by a microphone, capable of capturing a sound and generating a signal 110, subsequently called "sound signal”, representative of this sound.
  • the sounds are preferably audible sounds likely to contain a human voice, which are generally between 200 Hz and 3 kHz. Indeed, only sounds containing human voices are generally considered to contain personal information.
  • FIG 2 is a graphic representation of an extract of a sound signal that can be picked up by a sensor 11 such as a microphone.
  • the sound signal 110 supplied by the sensor is a continuous signal represented by the curve 21 oscillating around a neutral value 20.
  • This curve 21 is representative of the pressure variations, around a neutral value 20, which characterizes the sound.
  • the sound signal 110 may consist of an analog signal directly from the sensor 11. It may also, in a conventional manner, have undergone a first preprocessing, such as pre-amplification and filtering by a band-pass filter, intended to eliminate the DC component of the signal and to correct the frequencies which are poorly processed by sensor 11. It is also possible that the sound signal has been digitized.
  • a first preprocessing such as pre-amplification and filtering by a band-pass filter
  • the sound signal 110 generated by the sensor 11 is sent to processing equipment 12 which applies in real time to the sound signal 110 a processing aimed at removing from this signal any information that could be considered as personal information.
  • This processing is anonymization processing within the meaning of the laws and regulations protecting personal data. Processing to anonymize a sound signal whose source is a person consists of eliminating all parts of the signal allowing it to be traced back to the person from whom it came. For a sound signal containing a human voice, anonymization processing consists of eliminating the formants of the voice. Formants are the harmonics of the fundamental frequency of the voice. It is these harmonics that make it possible to identify the speaker. An anonymization processing erasing the formants of the voice prevents the identification of the speaker from the processed signal. On the other hand, if this anonymization processing makes it possible to preserve the fundamental frequency of the sound signal, the processed signal can be analyzed to identify information such as the emotion contained in its voice (anger, irritation, fear, panic, joy, ).
  • the present invention relates in particular to anonymization treatments making it possible to produce a processed signal making it possible to find the information of the fundamental frequencies of the sound signal, but not making it possible to find the information of the harmonics of this sound signal.
  • the processing equipment 12 subjects the sound signal 110 to a transformation isomorphic to the “absolute value” transformation.
  • the processed signal 120 resulting from this processing of the sound signal 110 by the processing equipment 12, makes it possible to find the absolute value of the sound signal 110 coming from the sensor 11.
  • Such a transformation is an unaveraged even function.
  • This function is therefore non-invertible.
  • An even function applying to a variable signal oscillating around an equilibrium value considered as the origin, is a function which associates with each value of the signal, positive in relation to the origin, the same result as for the same value of the signal, negative compared to the origin.
  • the even function therefore generates a variable image depending on the absolute value of the original signal.
  • this function is not averaged, that is to say that at a given moment, the generated image is only a function of the absolute value of the original signal at that precise moment. This function therefore makes it possible to take into account all of the oscillations of the original signal, without smoothing these oscillations.
  • the transformation applied to the sound signal is a non-averaged function.
  • a non-averaged transformation allows the processed signal to retain a large part of the frequency information of the original sound signal, particularly in the audible frequencies between 200 Hz and 3 kHz.
  • the processed signal can thus be subjected to frequency analyzes making it possible to provide numerous information on the nature of the sounds captured in the sound signal.
  • absolute value transformation or a transformation isomorphic to the absolute value, transformed a sound signal irreversibly.
  • absolute value transformation we understand a transformation which associates the absolute value of this number with a number. It is therefore impossible to find the initial sound signal from the processed signal. Furthermore, the processed signal irreversibly loses its information allowing a sound or tone of voice to be recognized. The processed signal is therefore irreversibly anonymized, and no longer contains information likely to be considered as personal information.
  • the processed signal 120 resulting from this transformation carried out by the processing equipment 12, is then sent to analysis equipment 13.
  • This analysis equipment 13 can be intended to analyze the processed signal 120, preferably in real time , in order to identify in this processed signal 120 anomalies or particular events of the sound signal captured by the sensor 11.
  • This processed signal 120 has the great advantage of retaining all the oscillations of the initial sound signal 110, its frequency being doubled compared to at the frequency of the initial signal. It is therefore possible to carry out a frequency analysis, for example to categorize the different sound components of the initial sound signal.
  • the processed signal is sent to recording equipment, in order to record the processed signal 120, for example in order to allow its subsequent analysis. It is also possible that the processed signal is sent to transmission equipment, in order to transmit this processed signal 120 to another device, which can for example carry out its analysis in order to detect and categorize particular events or anomalies of the signal. sound captured by sensor 11.
  • a transmission or recording of the sound signal 120 should not be limited by legislation limiting the transmission or recording of data containing information of a personal nature, this sound signal 120 not containing information that could be considered personal information.
  • FIG. 3 is an example of representation of a processed signal, resulting from the processing of the sound signal represented by the figure 2 , by processing equipment according to one embodiment of the invention.
  • the processing equipment applies an “absolute value” transformation, consisting of associating its absolute value with each value of the sound signal, relative to the balance value. All the values of the processed signal, represented by curve 22, are therefore positive with respect to the neutral value 20.
  • Such an “absolute value” transformation can for example be obtained, on an analog sound signal, by submitting this signal emitted by a sensor such as a microphone to an electronic assembly such as a thresholdless rectifier, well known to man. of career.
  • An example of such an electronic assembly comprising an operational amplifier 43, a diode 44 and a resistor 45, is represented by the Figure 4 .
  • Such an arrangement makes it possible to associate with an input voltage 41 an output voltage 42 whose value corresponds to the absolute value of the input voltage 41.
  • Such an “absolute value” transformation can also be obtained very easily on a digitized sound signal, by applying a program adapted to this signal.
  • FIG. 5 is an example of representation of a processed signal, resulting from the processing of the sound signal represented by the figure 2 , by other processing equipment according to an embodiment of the invention.
  • the processing equipment applies a “squaring” transformation, consisting of associating the square of its value with each value of the sound signal. All the values of the processed signal, represented by curve 32, are therefore positive with respect to the neutral value 30.
  • the “squared” transformation is isomorphic to the “absolute value” transformation. It is in fact possible, by calculating the square root of the processed signal, to obtain the absolute value of the initial sound signal.
  • Such a “squaring” transformation can also be obtained very easily on a digitized sound signal, by applying a program adapted to this signal.
  • a sound signal capture device can bring together a pressure sensor such as a microphone and processing equipment applying to the signal coming from this sensor a transformation isomorphic to the “absolute value” transformation.
  • a pressure sensor such as a microphone
  • processing equipment applying to the signal coming from this sensor a transformation isomorphic to the “absolute value” transformation.
  • Such a sound signal capture device is represented schematically, on the figure 1 , by rectangle 100.
  • this sound signal capture device can be constituted by a single electronic component, comprising a pressure sensor and an electronic chip ensuring the digitization of the signal coming from the sensor and the application to this signal of a transformation isomorphic to the “absolute value” transformation.
  • This electronic component can also include the means making it possible to ensure pre-processing of the signal, such as pre-amplification and filtering by a band-pass filter, for example before digitization.
  • Such an electronic component can advantageously be similar to a component known as a “digital microphone”, which includes a pressure sensor and an electronic chip ensuring the digitization of the signal coming from the sensor.
  • This sensor and this chip can be placed in the same housing of the electronic component, which can itself be implemented in an electronic circuit.
  • the chip is in this case programmed to also perform a transformation isomorphic to the “absolute value” transformation.
  • Such an electronic component can thus make it possible to capture a signal, representative of an ambient sound, and capable of undergoing frequency analysis, while guaranteeing that this signal does not include information that could be considered as information of a personal nature.
  • the combination in the same housing of the sensor and the chip ensuring the transformation of the signal advantageously makes it possible to guarantee that the untransformed sound signal is not used.
  • the implementation of such a microphone in an electronic circuit thus makes it possible to guarantee that this electronic circuit does not use any sound data which could be considered as personal information.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computational Linguistics (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)
  • Circuit For Audible Band Transducer (AREA)
  • Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
EP23188331.5A 2022-07-28 2023-07-28 Verfahren zur echtzeitverarbeitung eines tonsignals und vorrichtung zur erfassung eines tonsignals Pending EP4312215A1 (de)

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
FR2207819A FR3138592A1 (fr) 2022-07-28 2022-07-28 Procédé de traitement en temps réel d'un signal sonore et dispositif de captation de signal sonore

Publications (1)

Publication Number Publication Date
EP4312215A1 true EP4312215A1 (de) 2024-01-31

Family

ID=84359982

Family Applications (1)

Application Number Title Priority Date Filing Date
EP23188331.5A Pending EP4312215A1 (de) 2022-07-28 2023-07-28 Verfahren zur echtzeitverarbeitung eines tonsignals und vorrichtung zur erfassung eines tonsignals

Country Status (2)

Country Link
EP (1) EP4312215A1 (de)
FR (1) FR3138592A1 (de)

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190035381A1 (en) * 2017-12-27 2019-01-31 Intel Corporation Context-based cancellation and amplification of acoustical signals in acoustical environments

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190035381A1 (en) * 2017-12-27 2019-01-31 Intel Corporation Context-based cancellation and amplification of acoustical signals in acoustical environments

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
IAIN COLLINS: "Signal Power and Energy", 14 August 2019 (2019-08-14), XP093021470, Retrieved from the Internet <URL:https://www.youtube.com/watch?v=7I9XEhAup4c> [retrieved on 20230207] *

Also Published As

Publication number Publication date
FR3138592A1 (fr) 2024-02-02

Similar Documents

Publication Publication Date Title
EP0904535B1 (de) Verfahren und vorrichtung zur erfassung und lokalisierung einer reflektierenden schallquelle
CA2923888C (en) Dynamic threshold methods, systems, computer readable media, and program code for filtering noise and restoring attenuated high-frequency components of acoustic signals
Brown et al. Automatic and efficient denoising of bioacoustics recordings using mmse stsa
FR2888946A1 (fr) Procede et appareil pour attenuer le bruit du vent dans des donnees sismiques
CA2657450A1 (fr) Procede et dispositif de diagnostic de l&#39;etat de fonctionnement d&#39;un systeme de sonorisation
CA2891205A1 (fr) Methode et dispositif de detection acoustique d&#39;un dysfonctionnement d&#39;un moteur equipe d&#39;un controle actif du bruit
EP0612943B1 (de) Induktive Überwachungseinrichtung für Ventilklappe
EP4312215A1 (de) Verfahren zur echtzeitverarbeitung eines tonsignals und vorrichtung zur erfassung eines tonsignals
Nassi et al. Lamphone: Passive sound recovery from a desk lamp's light bulb vibrations
FR3088766A1 (fr) Procede et systeme de traitement de signaux audio pour microphone de masque a oxygene d’un aeronef
Cook et al. Automatic classification and reduction of wind noise in spectral data
CA3179399A1 (fr) Systeme de reconnaissance et d&#39;identification de sources sonores en temps reel
FR2724029A1 (fr) Procede de detection de signaux acoustiques provenant de torpilles
Kurniawan et al. Statistical-based audio forensic on identical microphones
Chen et al. Sparse Representation Based on Tunable Q‐Factor Wavelet Transform for Whale Click and Whistle Extraction
Rappaport Establishing a standard for digital audio authenticity: A critical analysis of tools, methodologies, and challenges
Richmond et al. Speaker diarisation of vibroacoustic intelligence from drone mounted laser doppler vibrometers
WO2015177359A1 (fr) Systeme et procede pour mesurer l&#39;impedance acoustique d&#39;une cavite
CN112581975A (zh) 基于信号混叠和双声道相关性的超声波语音指令防御方法
FR3065079B1 (fr) Procede et dispositif de sondage ultrasonore pour l&#39;obtention de courbes de dispersion d&#39;un milieu sonde
FR2846089A1 (fr) Dispositif et procede de test acoustique et vibratoire de de pieces mecaniques
Müller et al. A New Approach to Voice Authenticity
FR2903189A1 (fr) Procede de detection de presence d&#39;insectes xylophages actifs dans un ouvrage en bois
KR102644641B1 (ko) 레이저 용접에 대한 품질 모니터링 장치 및 방법
FR2996675A1 (fr) Procede de restitution d&#39;un signal d&#39;ecoute a partir d&#39;un appareil de diagnostic vibroacoustique et appareil mettant en œuvre un tel procede

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

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

Free format text: STATUS: THE APPLICATION HAS BEEN PUBLISHED

AK Designated contracting states

Kind code of ref document: A1

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

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

Free format text: STATUS: REQUEST FOR EXAMINATION WAS MADE

17P Request for examination filed

Effective date: 20240725

RBV Designated contracting states (corrected)

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