A method for auditory evaluation of a predefined space

Abstract

The invention relates to a method for auditory evaluation of a predefined space (2) comprising the following steps: - taking an acoustic sample over a predefined duration with an acoustic sensor (1), in the predefined space (2), - converting the sample into a data stream corresponding to a stream of signals arriving at the brain with frequency channels, - evaluating for each channel the signal stream according to a biological value scale by determining an acoustic nuisance index for each channel, - selecting a characterization profile of said person including physiological, temporal and psychic characteristics, - selecting a characterization profile of the space including characteristics of an activity present in the space and of a place, - providing the acoustic nuisance indices of the channels as well as the characterization profile of said person and the characterization profile of the space as input to an artificial neural network.Figure 2.

Description

Title of the invention: Method for auditory evaluation of a predefined space

[0001] The invention relates to the field of hearing and more specifically to a method of auditory evaluation of a predefined space by simulating the acoustic perception of a person in that space in order to determine the existence of an acoustic disturbance for that person in that space.

[0002] An increasingly large portion of the world's population is exposed to noise levels harmful to hearing due to the rise in the number and intensity of environmental noise emissions. Faced with the increasing number of noise-related illnesses, more and more sectors of activity and leisure are gradually becoming aware of the impact of noise pollution on health. Whether in the professional or private sphere, addressing this issue can also greatly improve everyone's quality of life, particularly in the workplace.

[0003] Noise has been recognized in France as a cause of occupational disease since 1963, and occupational deafness ranks fourth among occupational diseases (source: CIDB). Furthermore, it must be borne in mind that the social cost of noise is far from negligible. In France, it has been estimated at 57 billion euros, from all sources combined (transport, neighborhood, workplace, school). According to an IFOP / JNA survey, noise and sound pollution are responsible for a loss of productivity of approximately 23 billion euros per year. These few health and economic data demonstrate that noise pollution is a crucial issue for the development of our society.

[0004] The human auditory system is subject to constant stimuli; it is the only sense that does not rest. The human brain thus continuously receives information related to all the acoustic sources present in our sound "sphere," and reproduces sensations that are both quantitative and subjective, and in this respect, may fall outside the framework of a purely rational approach. Auditory sensation is therefore automatically accompanied by a psychoacoustic evaluation that primarily incorporates two types of factors:

[0005] • Auditory factors that are related to the human auditory system and depend on the the way in which an individual's hearing and brain 'physically' perceive sound stimuli.

[0006] • Emotional factors felt in parallel and which take into account the individual's psychological, social and cultural experience, their emotional state and momentary activity.

[0007] From a neuroscientific perspective, auditory sensation can be broadly defined as the transformation of an external sound event into internal neuronal activity in the brain. It is the first step in a chain of physical, biochemical, and neurological events, ranging from the energetic stimulation of a sensory organ, in this case the ear, to its unconscious reception by the brain, whereas sound perception corresponds to a set of neuronal processes by which the human becomes aware of it. Auditory sensation thus corresponds to the raw information decoded by the brain and, in this respect, can be likened to an emotion, while the perception resulting from an additional complex cognitive process can be likened to a feeling of emotion, an "awareness."

[0008] Acoustic disturbance or noise nuisance is a parasitic auditory information providing a neuronal interference that is detrimental to the performance of a main brain activity, and is perceived emotionally as unpleasant.

[0009] However, the difficulty in assessing acoustic discomfort lies, for example, in the fact that listening to loud music at a concert is not perceived negatively by audience members, nor is the sound of a gunshot by a hunter participating in, for example, a clay pigeon shooting event, although in these cases it is well known that the ear is subjected to significant damage. Furthermore, it must be noted that discomfort is not always associated with a high noise level. For example, the sound of a dripping tap, the persistent hum of an air conditioner, or a neighbor's conversation are considered very unpleasant and bothersome.

[0010] In some cases, the pronounced and continuous tonal emergence (noise concentrated on certain frequencies) can trigger a phenomenon of pregnance in brain activity by saturating certain neuronal pathways and lead to a situation of obsessive annoyance (sound of a vacuum cleaner, of neighbors...) without being of a very high level.

[0011] The acoustic regulations currently in force do not allow for addressing this acoustic nuisance.

[0012] The present invention aims to evaluate the phenomenon of acoustic annoyance, in particular by a more reliable and objective approach than that permitted by methods known in the prior art and which are based solely on physical characteristics of the sound signal such as sound pressure and / or frequency.

[0013] To this end, the invention relates to a method for auditory evaluation of a predefined space by simulating the acoustic perception of a person in that space in order to determine the existence of acoustic discomfort for that person, comprising the following steps:

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020]

[0021] - to take, using at least one acoustic sensor, in particular a binaural acoustic sensor, an acoustic sample in the predefined space over a predefined duration of at least one minute, - convert the acoustic sample into a data stream corresponding to a neuronal-type signal stream arriving at the brain with frequency channels between 16Hz and 16000Hz, the neuronal-type signal stream for each frequency channel being formed by pulses whose repetition rate per second is a function of the intensity of the acoustic signal of the frequency channel considered - evaluate for each channel the neuronal-type signal stream according to a biological value scale by determining an acoustic nuisance index for each channel, - select a characterization profile of said person including physiological, temporal and psychological characteristics, - select a space characterization profile including characteristics of an activity present in the space and of a place, - provision of the acoustic nuisance indices of the frequency channels as well as the characterization profile of said person and the characterization profile of the input space of an artificial neural network to assess audibly the predefined space by a classification according to at least a first category corresponding to the occurrence of an acoustic annoyance perceived by the person and a second category corresponding to the perception of an absence of an acoustic annoyance perceived by the person. The process may exhibit one or more of the following aspects, taken alone or in combination: For example, 620 frequency channels are used, ranging from 16 Hz to 16,000 Hz. The repetition rate is typically between 0 and 330 per second. Physiological characteristics include at least one of the following parameters: age, specifically an age range, and sex. Temporal characteristics include, for example, at least one of the following parameters: month of the year, time slot of the day such as morning, afternoon, evening or night. Psychic characteristics may include at least one of the following parameters: an attentional state such as awake, drowsy or asleep and an emotional state such as calm / serene, stressed / anxious, joyful / cheerful, sad / melancholic or irritated / angry. The biological value scale can be represented by a sigmoid-type function: f ( x ) “7^7777'

[0022] where x is the intensity of a stimulus, f(x) the physiological response, a a sensitivity factor controlling the evolution law and xo the tipping point.

[0023] The sensitivity factor a is determined empirically in relation to a test protocol.

[0024] The characteristics of the activity include, for example, at least one of the following parameters: manual, sporting, recreational, musical, computer, reading, writing, discussion, meditation / reflection activity.

[0025] The location characteristics include in particular at least one of the following parameters: office, school, meeting room, workshop, means of transport, performance hall, restaurant, sports hall, open air.

[0026] The invention also relates to a computer program product comprising one or more sequences of stored instructions accessible to a processor and which, when executed by the processor, enables the latter to carry out the steps of the process as described above.

[0027] The invention also relates to a method for reducing acoustic disturbance perceived by a person in a predefined space, in which - As a first step, an auditory evaluation of the predefined space is carried out using a procedure as described above, - if the result of the first step is classified according to a category corresponding to the occurrence of acoustic disturbance perceived by the person, the said space is modified, in particular by the installation / arrangement of acoustic absorption elements in this space, - we repeat the first and second steps until we obtain, as a result of the first step, a classification according to a category corresponding to the absence of acoustic disturbance.

[0028] The invention further relates to a device for implementing a method as described above, comprising: - at least one acoustic sensor, in particular a binaural acoustic sensor, configured to be able to take an acoustic sample in a predefined space over a predefined period of at least one minute, - a converter of the acoustic sample into a data stream corresponding to a neuronal-type signal stream arriving at the brain with frequency channels between 16Hz and 16000Hz, the neuronal-type signal stream for each frequency channel being formed by pulses whose repetition rate per second is a function of the intensity of the acoustic signal of the frequency channel considered - an evaluation unit to assess the neuronal-type signal stream for each channel according to a biological value scale by determining an acoustic nuisance index for each channel, - a first selection unit configured to select a characterization profile of said person including physiological, temporal and psychological characteristics, - a second selection unit configured to select a space characterization profile including characteristics of an activity present in the space and of a location, - an artificial neural network processing unit configured to receive as input acoustic nuisance indices from frequency channels as well as characterization profiles of said person and characterization of the input space, the artificial neural network processing unit being configured to auditorily evaluate the predefined space by a classification according to at least a first category corresponding to the occurrence of an acoustic annoyance perceived by the person and a second category corresponding to the perception of an absence of an acoustic annoyance perceived by the person.

[0029] The device may have one or more of the following aspects taken alone or in combination:

[0030] Frequency channels between 16Hz and 16000 Hz include, for example, 620 channels.

[0031] The repetition frequency of the acoustic sample converter is in particular between 0 and 330 per second for each channel.

[0032] Physiological characteristics include in particular at least one of the following parameters: age, in particular an age range, sex.

[0033] The temporal characteristics include, for example, at least one of the following parameters: month of the year, time slot of the day such as morning, afternoon, evening or night.

[0034] The psychic characteristics include in particular at least one of the following parameters: an attentional state such as awake, drowsy or asleep and an emotional state such as calm / serene, stressed / anxious, joyful / cheerful, sad / melancholic or irritated / angry.

[0035] The biological value scale of the biological assessment unit can be represented by a sigmoid-type function:

[0036] .........1........,

[0037] where x is the intensity of a stimulus, f(x) the physiological response, a a sensitivity factor controlling the evolution law and xo the tipping point.

[0038] Other features and advantages of the invention will become clear from the following description, which is by way of example and not limitation, with reference to the accompanying drawings, in which:

[0039] [Fig. 1] Fig. 1 is a flowchart showing different stages of the process according to the present invention in a first embodiment, and

[0040] [Fig.2] The [Fig.2] is a simplified diagram of an embodiment of a device for implementing a process according to the present invention according to a first embodiment.

[0041] [Fig.3] The [Fig.3] is a flowchart showing different stages of a process for reducing acoustic discomfort perceived by a person in a predefined space.

[0042] In the various figures, identical elements bear the same reference numbers. The following embodiments are examples.

[0043] In this description, certain elements or parameters can be indexed, such as first element or second element, first parameter and second parameter, first criterion and second criterion, etc. In this case, it is simply a matter of indexing to differentiate and name similar but not identical elements, parameters, or criteria. This indexing does not imply any priority of one element, parameter, or criterion over another, and such designations can easily be interchanged without departing from the scope of this description.

[0044] The method of the invention can be implemented at least in part on a computer. In this context, unless otherwise indicated, it is understood that, throughout this description, discussions using terms such as "computing", "computing" and "generation" or similar, refer to the action and / or processes of a computer or computer system, or similar electronic computing device, which manipulates and / or transforms data represented as physical quantities, such as electronic quantities, in the registers and / or memories of the computer system into other data similarly represented as physical quantities in the memories, registers or other devices for storing, transmitting or displaying information of the computer system.

[0045] A computer program product comprising one or more sequences of stored instructions accessible to a processor and which, when executed by the processor, enables the latter to perform the steps of the process is also proposed.

[0046] Such a computer program may be stored on a computer-readable storage medium, such as, but not limited to, any type of disk, including floppy disks, optical discs, CD-ROMs, magnetic-optical discs, read-only memories (ROMs), random-access memories (RAMs), electrically programmable read-only memories (EPROMs), electrically erasable and programmable read-only memories (EEPROMs), magnetic or optical cards, or any other type of medium suitable for storing electronic instructions, and capable of being coupled to a computer system bus.

[0047] A computer-readable medium containing one or more sequences of instructions for the computer program product is therefore proposed. This allows the method to be implemented anywhere.

[0048] The methods and displays presented here are not intrinsically linked to any particular computer or other device. Various general-purpose systems can be used with programs conforming to the teachings of this document, or it may be convenient to construct a more specialized device to carry out the desired method. The desired structure for a variety of such systems will appear in the description below. Furthermore, the embodiments of the present invention are not described with reference to any particular programming language. It will be appreciated if a variety of programming languages ​​can be used to implement the teachings of the inventions as described herein.

[0049] We will now describe the auditory evaluation process of a predefined space 2 by simulating the acoustic perception of a person in this space in order to determine the existence of acoustic discomfort for this person and the associated device,

[0050] A possible embodiment of the present invention will be described with reference to Figures 1 and 2, of which [Fig.1] shows a flowchart showing different steps of the process according to the present invention according to a first embodiment and [Fig.2] presents a simplified diagram of an embodiment of a device for the implementation of a process.

[0051] As can be seen in [Fig.2], the device for implementing the method according to the invention includes at least one acoustic sensor 1. This sensor 1 is configured to be able to take, according to a first step 100 (see [Fig.1]), in a predefined space 2, an acoustic sample over a predefined period of at least one minute.

[0052] An acoustic sensor 1 may include a single microphone. According to the example described, it is for example a binaural acoustic sensor comprising for example two microphones spaced apart by a distance which corresponds to the average distance between two human ears and whose sensitive parts are directed in opposite directions.

[0053] The binaural sensor 1 enables binaural recording which is based on a replica of the auditory system implementing a set of microphones arranged so as to be able to accurately recreate the difference in time and intensity perceived between the two ears of a human.

[0054] The predefined space 2 is, for example, a space whose acoustic quality must be evaluated. It could be, for example, a room in a building, an office, a hall, an enclosed workplace in a factory production area, but also an open-air location.

[0055] The acoustic sample is one or more recordings over a predefined duration of at least one minute. It may also consist of several samples taken at different times and aggregated, for example in the morning, at midday and in the evening, or samples taken from different locations in the room.

[0056] In the case of a binaural acoustic sensor 1, we have for example two acoustic samples which can be processed in parallel independently.

[0057] The binaural acoustic sensor 1 is for example connected by wire or wireless means to a specific computer 3 which includes for example an input to receive and process the signals from the binaural sensor 1.

[0058] Thus, the computer 3 includes, for example, a converter 5 configured to convert, in a step 102, the acoustic sample(s) into a data stream corresponding to a stream of neuronal-type signals arriving at the brain with frequency channels ranging from 16 Hz to 16,000 Hz. The neuronal-type signal stream for each frequency channel is formed by pulses whose repetition rate per second is a function of the intensity of the acoustic signal of the frequency channel considered.

[0059] It has been found that the human being is capable of discerning about 620 frequency channels (140 below 500 Hz and 420 above 500 Hz).

[0060] Each channel can be considered to correspond to a neuronal channel in the human brain. Consequently, the acoustic sample is processed so as to divide the frequency range between 16Hz and 16000Hz into 620 frequency channels.

[0061] Of course, the number of frequency channels for the present method can be decreased or increased without departing from the scope of the present invention.

[0062] According to the present method, the processing, and in particular the conversion of the various channels into a neuronal-type signal stream, is carried out individually for each frequency channel. The frequency channels are notably processed in parallel.

[0063] To do this, the acoustic sample for each frequency channel is converted into a value between 0 and 330, which corresponds to a biological response between 0 and 330 action potentials (or "AP").

[0064] Indeed, a fundamental element of neuronal communication is the Action Potential (AP). A neuron that has just been stimulated (lit up) transmits a message to another neuron by sending electrochemical impulses called "action potentials" (AP) through its axon.

[0065] The amplitude of the impulses emitted by a neuron is approximately constant, and therefore the information conveyed is found in their frequency of occurrence or repetition. Given their duration, nerve impulses cannot follow one another unless 3 milliseconds apart. In this respect, this duration can be considered the natural unit of processing time for nerve impulses in the brain. It can be deduced that the maximum firing frequency of a neuron, and therefore of neuronal communication, is approximately 300-400 Hz per neuronal pathway, resulting in a biological value scale between 0 and 330, with 330 corresponding to the maximum number of impulses that can be emitted by a neuron. The repetition rate per second, or action potential, is thus a function of the intensity of the acoustic signal in the frequency channel under consideration.

[0066] In view of the elements set forth above, the neuronal information of hearing can therefore be expressed by a 620 x 330 matrix, corresponding to 620 parallel channels sampling the audible frequency range (between 16Hz and 16000 Hz), and 330 levels sampling the biological value of the sound force (between 0 and 330 Action Potentials per second).

[0067] For each neuronal channel identified by a position on the basilar membrane and associated with a sound frequency, the number of action potentials will thus oscillate between 0 and a maximum value which has been set at 330 PA / s taking into account the duration of a typical PA, and give rise to a sound perception whose evolution responds to a sensation as described above, within the physical limits delimited by the hearing threshold and the pain threshold, which are extreme situations which limit the sensory response of hearing.

[0068] Then, according to a step 104, the neuronal type signal flow is evaluated for each channel using an evaluation unit 7 (see [Fig.2]) according to a biological value scale by determining an acoustic nuisance index for each channel.

[0069] The biological value scale is, for example, represented by a sigmoid-type function:

[0070]

[0071] where x is the intensity of a stimulus, f(x) the physiological response, a a sensitivity factor controlling the evolution law and xo the tipping point.

[0072] Generally for a channel, the sensitivity factor a is between 0.08 and 0.15 and xo the switching point is expressed in a non-dimensional way is therefore 0.5. The sensitivity factor a can also be determined empirically with respect to a test protocol.

[0073] The factor a of the sigmoid function can be likened to the overall neuronal sensitivity of an individual. The tipping point xo of the sigmoid function is likened to a median value of the spectral dynamics of the ear.

[0074] The intensity x of the stimulus is therefore a value between 0 and 330 for each channel.

[0075] It has been shown that biological values ​​can be represented by a universal "sigmoid" type evolution law as a function of the amplitude of the stimulation. Such a law reflects that, for example, when one is exposed to a low-level noise, one is not very aware of the discomfort caused, and that beyond a certain threshold the discomfort becomes evident and then rapidly worsens to reach an unbearable level, roughly following this evolution.

[0076] This also allows us to understand that the initial sensation arising from an acoustic stimulus may appear indistinct because it is drowned in "neuronal background noise," but it can become increasingly distinct and then culminate in a saturation zone where the sensation varies only slightly, without us knowing precisely where the limit lies as the stimulus increases in intensity. This is, for example, what we observe in the field of acoustics for the perception of sound intensity, or in a completely different field for the perception of pain. The tipping point of this function (where the derivative reaches its maximum) can be seen as a statistical threshold beyond which the sensation is confirmed and where full awareness of it is achieved.

[0077] Then the computer 3 can include a first selection unit 9 configured to select according to a step 106 a characterization profile of said person including physiological, temporal and psychic characteristics and a second selection unit 11 configured to select according to a step 108 a characterization profile of the space including characteristics of an activity present in the space and of a place.

[0078] Physiological characteristics include at least one of the following parameters: age, in particular an age range, sex.

[0079] The temporal characteristics include at least one of the following parameters: month of the year, time slot of the day such as morning, afternoon, evening or night.

[0080] The psychic characteristics include at least one of the following parameters: an attentional state such as awake, drowsy or asleep and an emotional state such as calm / serene, stressed / anxious, joyful / cheerful, sad / melancholic or irritated / angry.

[0081] The characteristics of the activity include at least one of the following parameters: manual, sporting, recreational, musical, computer, reading, writing, discussion, meditation / reflection activity.

[0082] The location characteristics include at least one of the following parameters: office, school, meeting room, workshop, means of transport, performance hall, restaurant, sports hall, outdoors.

[0083] All these characteristics can, for example, be entered via a computer interface with, for example, a screen and a mouse allowing these characteristics to be selected from drop-down menus.

[0084] The computer further includes an artificial neural network processing unit 13 configured to receive as input acoustic nuisance indices for each channel as well as characterization profiles of said person and characterization of the input space.

[0085] Of course, units 5, 7, 9, 11, and 13 can be part of one or more computers. They may or may not be separate physical units. More specifically, they may be software routines for processing or performing one or more steps as described above.

[0086] The processing unit 13 is configured to evaluate during a step 110 the predefined space 2 by auditory means by a classification according to at least a first category corresponding to the occurrence of an acoustic annoyance perceived by the person and a second category corresponding to the perception of an absence of an acoustic annoyance perceived by the person.

[0087] The final assessment of the annoyance associated with a sound stimulus was carried out by the processing unit 13 by determining an acoustic nuisance index between 0 and 100%. To do this, the artificial neural network processing unit 13 takes into account in particular the frequency channels having the highest number of pulses per second or PA / s.

[0088] Five ranges of values ​​of the noise nuisance index can be distinguished corresponding to intervals: [0.00-0.25[ ; [0.25-0.75[ ; [0.75-0.95[ ; [0.95-1.00] on the following verbal noise nuisance scale: [very low; low; medium; high; very high].

[0089] The ranges very low; low; medium are for example assimilated to the second category corresponding to the perception of an absence of acoustic disturbance perceived by the person and the ranges high; very high to the first category corresponding to the occurrence of acoustic disturbance perceived by the person.

[0090] Thus, acoustic disturbance can be assessed more reliably and with greater precision.

[0091] The method and device described above can be used to reduce acoustic disturbance perceived by a person in a predefined space 2.

[0092] With reference to [Fig.3], to achieve a reduction of an acoustic annoyance perceived by a person in a predefined space 2, according to a first step 200, the auditory evaluation of the predefined space 2 is carried out as described above with steps 100 to 110 (Steps 100 to 110 are identical to those of steps 100 to 110 of [Fig.1]).

[0093] Then, if the result of the first step 200 classifies the evaluation result according to a category corresponding to the occurrence of an acoustic disturbance perceived by the person, we proceed according to a step 202 during which we proceed to modify said space.

[0094] Such a modification consists, for example, of installing / or arranging sound-absorbing elements in this space, such as sound-absorbing panels. To do this, particular consideration is given, for example, to the position of the binaural sensor 1 in the predefined space 2. Thus, sound-absorbing panels can be placed around the position of the binaural sensor 1 in the predefined space 2 to create a space protected by sound insulation / attenuation around this space, or by placing sound insulation or attenuation panels on walls or at ceiling level. Then, the first and second steps 200 and 202 are repeated until the result of the first step 200 is a classification according to a category corresponding to the absence of acoustic disturbance. Such a process is advantageous because it is possible to optimize the number of sound insulation panels to be installed.

[0095] The acoustic quality of a predefined space 2 can thus be improved more objectively by taking into account more precisely the neuronal process of sound formation in the human brain and the potential reactions of the brain to acoustic stimulation.

Claims

Demands

1. A method for the auditory evaluation of a predefined space (2) by simulating the acoustic perception of a person in that space in order to determine the existence of acoustic discomfort for that person, comprising the following steps: - taking, with at least one acoustic sensor (1), in particular a binaural acoustic sensor, in the predefined space (2), an acoustic sample over a predefined period of at least one minute, - converting the acoustic sample into a data stream corresponding to a stream of neuronal-type signals arriving at the brain with frequency channels between 16Hz and 16000Hz,The neuronal signal flow for each frequency channel is formed by impulses whose repetition rate per second is a function of the acoustic signal intensity of the frequency channel considered; evaluate the neuronal signal flow for each channel according to a biological value scale by determining an acoustic nuisance index for each channel; select a characterization profile for said person including physiological, temporal, and psychological characteristics; select a characterization profile for the space including characteristics of an activity present in the space and of a place.- provision of the acoustic nuisance indices of the frequency channels as well as the characterization profile of said person and the characterization profile of the input space of an artificial neural network to auditorily evaluate the predefined space (2) by a classification according to at least a first category corresponding to the occurrence of an acoustic annoyance perceived by the person and a second category corresponding to the perception of an absence of an acoustic annoyance perceived by the person.

2. A method according to claim 1, wherein 620 frequency channels are used, ranging from 16Hz to 16000 Hz.

3. A method according to claim 1 or 2, wherein the repetition frequency is between 0 and 330 per second.

4. A method according to any one of claims 1 to 3, wherein the physiological characteristics include at least one of the following parameters: age, in particular an age range, sex.

5. A method according to any one of claims 1 to 4, wherein the time characteristics include at least one of the following parameters: month of the year, time slot of the day such as morning, afternoon, evening or night.

6. A method according to any one of claims 1 to 5, wherein the psychic characteristics include at least one of the following parameters: an attentional state such as awake, drowsy or asleep and an emotional state such as calm / serene, stressed / anxious, joyful / cheerful, sad / melancholic or irritated / angry.

7. A method according to any one of claims 1 to 6, wherein the biological value scale is represented by a sigmoid-type function: - where x is the intensity of a stimulus, f(x) a physiological response, a a sensitivity factor controlling the evolution law and xo a tipping point.

8. A method according to claim 7, wherein the sensitivity factor a is determined empirically with respect to a test protocol.

9. A method according to any one of claims 1 to 8, wherein the characteristics of the activity include at least one of the following parameters: manual, sporting, recreational, musical, computer, reading, writing, discussion, meditation / reflection activity.

10. A method according to any one of claims 1 to 9, wherein the location features include at least one of the following parameters: office, school, meeting room, workshop, means of transport, performance hall, restaurant, sports hall, outdoors.

11. A computer program product comprising one or more sequences of stored instructions accessible to a processor and which, when executed by the processor, enables the latter to perform the steps of the process according to any one of claims 1 to 10

12. A method for reducing acoustic disturbance perceived by a person in a predefined space (2), in which - according to a first step, the auditory evaluation of the predefined space (2) is carried out according to any one of claims 1 to 10, - if the result of the first step is classified according to a category corresponding to the occurrence of an acoustic disturbance perceived by the person, the said space is modified, in particular by the installation / arrangement of acoustic absorption elements in this space, - the first and second steps are repeated until the result of the first step is a classification according to a category corresponding to the absence of an acoustic disturbance.

13. A device for implementing a method according to any one of claims 1 to 10, comprising: - at least one acoustic sensor (1), in particular a binaural acoustic sensor, configured to be able to take an acoustic sample in a predefined space (2) over a predefined period of at least one minute, - a converter (5) of the acoustic sample into a data stream corresponding to a stream of neuronal-type signals arriving at the brain with frequency channels between 16Hz and 16000Hz, the neuronal-type signal stream for each frequency channel being formed by pulses whose repetition rate per second is a function of the intensity of the acoustic signal of the frequency channel considered - an evaluation unit (7) for evaluating the neuronal-type signal stream for each channel according to a biological value scale by determining an acoustic nuisance index for each channel,- a first selection unit (9) configured to select a characterization profile of said person comprising physiological, temporal and psychological characteristics, - a second selection unit (11) configured to select a characterization profile of the space comprising characteristics of an activity present in the space and of a place, - an artificial neural network processing unit (13) configured to receive as input acoustic nuisance indices from the frequency channels as well as characterization profiles of said person and characterization profiles of the space as input, the artificial neural network processing unit (13) being configured, to assess audibly the predefined space (2) by a classification according to at least a first category corresponding to the occurrence of an acoustic disturbance perceived by the person and a second category corresponding to the perception of an absence of an acoustic disturbance perceived by the person.

14. Device according to claim 13, wherein the frequency channels between 16Hz and 16000 Hz comprise 620 channels.

15. Device according to claim 13 or 14, wherein the repetition frequency of the acoustic sample converter is between 0 and 330 per second for each channel.

16. Device according to any one of claims 13 to 15, wherein the physiological characteristics include at least one of the following parameters: age, in particular an age range, sex.

17. Device according to any one of claims 13 to 16, wherein the time characteristics include at least one of the following parameters: month of the year, time slot of the day such as morning, afternoon, evening or night.

18. Device according to any one of claims 13 to 17, wherein the psychic characteristics include at least one of the following parameters: an attentional state such as awake, drowsy or asleep and an emotional state such as calm / serene, stressed / anxious, joyful / cheerful, sad / melancholic or irritated / angry.

19. Device according to any one of claims 13 to 18, wherein the biological value scale of the biological assessment unit is represented by a sigmoid-type function: f(x) - , , - where x is the intensity of a stimulus, f(x) the physiological response, a a sensitivity factor controlling the evolution law and xo the tipping point.

Images