Hearing system comprising at least one hearing instrument worn in or on the user's ear, and method for operating such a hearing system

Description

[0001] The invention relates to a method according to the preamble of claim 1 for operating a hearing system for supporting a user's hearing, comprising at least one hearing instrument worn in or on the user's ear. The invention further relates to such a hearing system.

[0002] A hearing instrument is generally defined as an electronic device that supports the hearing of a person wearing the hearing instrument (hereinafter referred to as the "wearer" or "user"). In particular, the invention relates to hearing instruments designed to fully or partially compensate for the hearing loss of a hearing-impaired user. Such a hearing instrument is also referred to as a "hearing aid." There are also hearing instruments designed to protect or improve the hearing of users with normal hearing, for example, to enable improved speech comprehension in complex listening situations.

[0003] Hearing instruments in general, and hearing aids in particular, are usually designed to be worn in or on the user's ear, particularly as behind-the-ear devices (also known as BTE devices) or in-the-ear devices (also known as ITE devices). In terms of their internal structure, hearing instruments usually have at least one (acousto-electrical) input transducer, a signal processing unit (signal processor), and an output transducer. During operation of the hearing instrument, the input transducer receives airborne sound from the hearing instrument's environment and converts this airborne sound into an input audio signal (i.e., an electrical signal that conveys information about the ambient sound). This input audio signal is also referred to below as the "recorded sound signal." The input audio signal is processed (i.e., converted) in the signal processing unit.The signal processing unit outputs a correspondingly processed audio signal (also referred to as the "output audio signal" or "modified sound signal") to support the user's hearing, particularly to compensate for a user's hearing loss. The signal processing unit outputs a correspondingly processed audio signal (also referred to as the "output audio signal" or "modified sound signal") to the output transducer. In most cases, the output transducer is designed as an electro-acoustic transducer, which converts the (electrical) output audio signal back into airborne sound, whereby this airborne sound—modified compared to the ambient sound—is emitted into the user's ear canal. In a hearing instrument worn behind the ear, the output transducer, also referred to as the "receiver," is usually integrated outside the ear in a housing of the hearing instrument. In this case, the sound emitted by the output transducer is guided into the user's ear canal via a sound tube.Alternatively, the output transducer can be located in the ear canal, outside the housing worn behind the ear. Such hearing instruments are also referred to as receiver-in-canal (RIC) devices. Hearing instruments worn in the ear, which are so small that they do not protrude beyond the ear canal, are also referred to as completely in-canal (CIC) devices.

[0004] In other designs, the output transducer can also be an electromechanical transducer that converts the output audio signal into structure-borne sound (vibrations), which is then transmitted, for example, to the user's skull. There are also implantable hearing instruments, particularly cochlear implants, and hearing instruments whose output transducers directly stimulate the user's auditory nerve.

[0005] The term "hearing system" refers to a single device or a group of devices and, where appropriate, non-physical functional units that together provide the functions required for the operation of a hearing aid. In the simplest case, the hearing system may consist of a single hearing instrument. Alternatively, the hearing system may comprise two cooperating hearing instruments to serve both ears of the user. In this case, it is referred to as a "binaural hearing system." Additionally or alternatively, the hearing system may include at least one other electronic device, such as a remote control, a charger, or a programming device for the or each hearing aid.Modern hearing systems often feature a control program, particularly in the form of an app, instead of a remote control or dedicated programming device. This control program is designed for implementation on an external computer, particularly a smartphone or tablet. The external computer is usually not part of the hearing system itself and, in particular, is not usually provided by the hearing system manufacturer.

[0006] A common problem for hearing-impaired users, and to a lesser extent for users with normal hearing, is that conversation partners are difficult to understand in listening situations where several people are speaking (multi-speaker environments). This problem can be partially remedied by direction-dependent attenuation (beamforming) of the input audio signal. Corresponding algorithms are usually set to selectively emphasize a portion of the ambient sound coming from the front over other noise sources, so that the user can better understand a conversation partner as long as they are looking at them. However, this type of signal processing disadvantageously limits the user's ability to interact with their environment. For example, the user cannot turn their head away from the conversation partner during a conversation without running the risk of losing track.Furthermore, conventional directional attenuation also increases the risk that the user will not understand or even notice other people who want to participate in the conversation but are outside the directional lobe.

[0007] The application is based on the task of enabling users of a hearing system to understand speech better, especially in a multi-speaker environment.

[0008] A method according to the preamble of claim 1 and a corresponding hearing system are known from EP 1 303 166 A2. Similar methods and hearing systems are known from US 2012 128186 A1 and DE 10 2015 210 652 A1.

[0009] With regard to a method, this object is achieved according to the invention by the features of claim 1. With regard to a hearing aid system, the object is achieved according to the invention by the features of claim 14. Advantageous and partly inventive embodiments or further developments of the invention are set out in the subclaims and the following description.

[0010] The invention generally relates to a hearing system for supporting the hearing of a user, wherein the hearing system has at least one hearing instrument worn in or on one ear of the user. As described above, in simple embodiments of the invention the hearing system can consist exclusively of a single hearing instrument. Preferably, however, in addition to the hearing instrument the hearing system comprises at least one further component, e.g. another (in particular of the same type) hearing instrument for supplying the other ear of the user, a control program (in particular in the form of an app) for execution on an external computer (in particular a smartphone) of the user and / or at least one further electronic device, e.g. a remote control or a charger.The hearing instrument and the at least one other component exchange data with each other, with data storage and / or data processing functions of the hearing system being divided between the hearing instrument and the at least one other component.

[0011] The hearing instrument has at least one input transducer for receiving a sound signal (particularly in the form of airborne sound) from the environment of the hearing instrument, a signal processing unit for processing (modifying) the received sound signal to assist the user's hearing, and an output transducer for outputting the modified sound signal. If the hearing system has another hearing instrument to supply the user's other ear, this additional hearing instrument preferably also has at least one input transducer, a signal processing unit, and an output transducer. Instead of a second hearing instrument with an input transducer, signal processing unit, and output transducer, a hearing instrument can also be provided for the second ear that does not have an output transducer itself, but only receives sound and transmits it - with or without signal processing - to the hearing instrument of the first ear.Such so-called CROS or BiCROS instruments are used particularly for users with unilateral deafness.

[0012] The or each hearing instrument of the hearing system is, in particular, available in one of the designs described above (BTE device with internal or external output transducer, ITE device, e.g., CIC device, hearing implant, especially cochlear implant, etc.). In the case of a binaural hearing system, both hearing instruments are preferably designed in the same way.

[0013] The or each input transducer is, in particular, an acousto-electrical transducer that converts ambient airborne sound into an electrical input audio signal. To enable direction-dependent analysis and processing of the recorded sound signal, the hearing system preferably comprises at least two input transducers, which can be arranged in the same hearing instrument or—if present—divided between the two hearing instruments of the hearing system. The output transducer is preferably designed as an electro-acoustic transducer (receiver), which in turn converts the audio signal modified by the signal processing unit into airborne sound. Alternatively, the output transducer is designed to emit structure-borne sound or to directly stimulate the user's auditory nerve.

[0014] The signal processing unit preferably comprises a plurality of signal processing functions that are applied to the recorded sound signal, i.e., the input audio signal, in order to process it to support the user's hearing. The signal processing functions include, in particular, any selection of the following functions: frequency-selective amplification, dynamic compression, spectral compression, direction-dependent attenuation (beamforming), noise suppression, e.g., classic noise suppression using a Wiener filter or active noise cancellation (ANC), active feedback cancellation (AFC), wind noise suppression, voice activity detection, own voice detection and processing (Own Voice Detection, Own Voice Processing), tinnitus masking, etc.Each of these functions, or at least a large portion of them, can be parameterized using one or more signal processing parameters. A signal processing parameter is a variable that can be assigned different values ​​to influence the operation of the associated signal processing function. In the simplest case, a signal processing parameter can be a binary variable that switches the respective function on and off. In more complex cases, hearing aid parameters are formed by scalar floating-point numbers, binary or continuously variable vectors, or multidimensional arrays, etc. An example of such signal processing parameters is a set of gain factors for a number of frequency bands of the signal processing unit, which define the frequency-dependent amplification of the hearing instrument.

[0015] During the process performed by the hearing system, a sound signal from the hearing instrument's surroundings is recorded by at least one input transducer of the hearing instrument. The recorded sound signal (input audio signal) is modified in a signal processing step to support the user's hearing ability. The modified sound signal is output via the hearing instrument's output transducer.

[0016] According to the method, in an analysis step, speech intervals are detected by analyzing the recorded sound signal in which the recorded sound signal contains (spoken) speech of a speaker different from the user.

[0017] Here and in the following, a speech interval generally refers to a temporally limited section of the recorded sound signal that contains spoken speech. Speech intervals that contain the user's own speech are referred to as "own speech intervals." In In contrast, speech intervals that contain the speech of at least one speaker other than the user, regardless of the language type - i.e. regardless of whether the speaker speaks English, German, French, etc. - are referred to as "foreign speech intervals".

[0018] To avoid linguistic ambiguities, only persons other than the user will be referred to as "talkers" in the following. The user himself is not considered a "speaker" here or in the following, even if he or she speaks.

[0019] According to the invention, in the analysis step, different speakers are identified by analyzing the recorded sound signal in recognized foreign speech intervals. The word "identify" is used in the sense that each of the identified speakers is recognizably distinguished from other speakers. In the analysis step, each recognized foreign speech interval is assigned to the speaker speaking in this foreign speech interval. Signal components from people speaking simultaneously (e.g., signal components of the user and at least one speaker, or signal components of several speakers) are preferably separated from one another using signal processing and processed separately. A self-speech interval and a foreign speech interval, or foreign speech intervals of different speakers, can overlap in time.In alternative embodiments of the invention, temporal sections of the recorded sound signal are always assigned to only one of the speakers involved, even if they contain speech components of several people.

[0020] According to the invention, for each recognized foreign language interval, the associated speaker is classified during an interaction classification according to whether or not this speaker is in a direct communication relationship with the user. Speakers who are in a direct communication relationship with the user are referred to below as "main talkers." Speakers who are not in a direct communication relationship with the user are referred to below as "secondary talkers." "Communication" is defined here and below as at least an attempted (conscious or unconscious) transmission of information between a speaker and the user through spoken language. A direct communication relationship exists when information is transmitted directly (without mediation by other persons or means) between the speaker and the user.For the present proceedings, four cases of a direct communication relationship are particularly relevant, namely . Firstly, the case in which the user and the speaker speak to each other, secondly, the case in which the speaker addresses the user directly and the user consciously listens to the speaker, thirdly, the case in which the speaker addresses the user directly, but the user does not consciously listen to the speaker (this includes in particular the case in which the user does not notice the speaker and the speaker's communication with the user at all), and fourthly, the case in which the speaker does not address the user directly, but in which the user consciously listens to the speaker.

[0021] Conversely, a direct communication relationship does not exist if the speaker does not address the user directly and the user does not consciously listen to the speaker.

[0022] In a multi-speaker environment, each of the multiple speakers can be classified as a primary speaker or a secondary speaker. Thus, there can be multiple primary speakers and / or multiple secondary speakers simultaneously. Interaction classification is still performed in a time-resolved manner. An identified speaker can therefore change their status as primary speaker or secondary speaker depending on their current communication relationship with the user. A speaker previously classified as a secondary speaker becomes the primary speaker when a direct communication relationship between them and the user arises. Likewise, a speaker previously classified as the primary speaker becomes a secondary speaker when the direct communication relationship between them and the user ends (e.g., when the speaker and the user each permanently turn to other conversation partners).

[0023] Depending on this interaction classification (i.e., depending on whether the speaker assigned to a recognized foreign speech interval was classified as the main speaker or as a secondary speaker), the detected foreign speech intervals are modified in different ways in the signal processing step, in particular by applying different settings of the signal processing parameters. For example, direction-dependent attenuation (beamforming) is applied to a greater extent to foreign speech intervals assigned to a speaker identified as the main speaker than to foreign speech intervals assigned to a secondary speaker. In other words, the directional lobe of the beamformer is preferentially and particularly strongly directed towards a speaker identified as the main speaker, while signal components from secondary speakers are preferentially attenuated or processed with little or no directionality.

[0024] By classifying the identified speakers into main and secondary speakers and by processing foreign speech intervals differently depending on this (interaction) classification, it is possible to particularly highlight parts of the recorded sound signal that are attributable to the main speaker and thus make them better or easier to perceive for the user.

[0025] In an advantageous embodiment of the method, the interaction classification is based on an analysis of the spatial orientation of the or each identified speaker relative to the user, and in particular to their head orientation. In the analysis step, the spatial orientation and, optionally, a distance of this speaker relative to the user's head are recorded for at least one identified speaker (preferably for each identified speaker) and taken into account in the interaction classification. For example, the finding that a user turns to an identified speaker particularly frequently and / or for a long time, so that this speaker is predominantly positioned in front of the user's head, is considered an indication that this speaker should be classified as the main speaker.A speaker who is always or at least predominantly positioned to the side or rear of the user's head, on the other hand, tends to be classified as a secondary speaker.

[0026] If the distance between identified speakers is also taken into account in interaction classification, a speaker's distance within a certain distance range is considered an indication that this speaker is a primary speaker. A speaker located at a comparatively large distance from the user's head, on the other hand, is more likely to be classified as a secondary speaker.

[0027] In addition or alternatively to spatial orientation, the sequence in which foreign speech intervals and own speech intervals alternate (turn-taking) is preferably also taken into account for interaction classification. In this analysis step, own speech intervals in which the user is speaking are also detected. For at least one (preferably for each) identified speaker, a temporal sequence of the assigned foreign speech intervals and the recognized own speech intervals is recorded and taken into account in the interaction classification. A speaker whose assigned foreign speech portions alternate with own speech portions without overlap or with only a comparatively small overlap and comparatively short intervening speech pauses tends to be classified as the main speaker, since such turn-taking is considered an indication that the speaker is in a two-way conversation with the user.Foreign speech intervals that are temporally uncorrelated with own speech intervals (for example, those that on average have a large overlap with own speech intervals) are counted as foreign speech intervals of cross-speakers.

[0028] Optionally, turn-taking between two speakers (different from the user) is also analyzed and considered for interaction classification. Foreign speech intervals of different speakers that alternate with each other without overlap or with only a slight overlap are considered an indication that the speakers assigned to these foreign speech intervals are in conversation with each other and thus – in the absence of other indications of passive or active user participation – should be classified as crosstalkers.

[0029] Additionally or alternatively, in preferred embodiments of the invention, interaction classification is based on the volume and / or the signal-to-noise ratio of the recorded sound signal. In the analysis step, an average volume (level) and / or a signal-to-noise ratio is determined for each detected foreign speech interval and taken into account in the interaction classification. Foreign speech intervals with a volume within a predetermined range or a comparatively good signal-to-noise ratio tend to be assigned to a primary speaker, while a comparatively low volume or a low signal-to-noise ratio during a foreign speech interval is considered an indication that the assigned speaker is a secondary speaker.In an advantageous embodiment, temporal changes in the spatial distribution of speakers (varying speaker positions or varying number of speakers) are analyzed during interaction classification.

[0030] Additionally or alternatively, the interaction classification preferably takes into account a physiological reaction of the user during a foreign speech interval. The physiological reaction, i.e. a temporal change in a recorded physiological parameter (e.g. heart rate (pulse rate), skin resistance, body temperature, condition of the ear muscle, eye position and / or brain activity), is determined in particular by means of at least one biosensor integrated into the hearing system or externally, e.g. a heart rate monitor, a skin resistance measuring device, a skin thermometer or an EEG sensor. A significant physiological reaction of the user during a foreign speech interval, i.e. a comparatively large change in the recorded physiological parameter, is interpreted as an indication that the speaker assigned to this foreign speech interval is a main speaker.A speaker whose foreign speech intervals do not elicit a significant physiological response from the user tends to be classified as a crosstalker.

[0031] Furthermore, behavioral patterns are advantageously also analyzed and used for interaction classification, e.g. changes in the pattern of head and torso movements, movement in space (approaching or distancing with respect to unfamiliar speakers), changes in the manner of speaking (in particular intensity, tonality, speaking rate / number of words per unit of time) or speaking time, change in sitting position, comprehension questions, selection of dialogue partners, etc. For example, the user's behavior can be analyzed with regard to his or her selection of certain unfamiliar speakers within the potential main speakers determined on the basis of the distance or direction analysis.For example, disproportionately more frequent interaction with a nearby speaker or fixation on a specific speaker (recognized by a low intensity or absence of head movements) while the user fixates less strongly on other speakers (recognized by a greater intensity of head movements) are considered as an indication that different main speakers are perceived differently.

[0032] Preferably, a combination of several of the criteria described above (spatial distribution of speakers, turn-taking, volume and / or signal-to-noise ratio of speech contributions, and the user's physiological response) and optionally one or more additional criteria are considered in the interaction classification. In this case, the interaction classification is preferably based on an examination of several criteria for coincidence (where, for example, a speaker is classified as the main speaker if several indicators are met simultaneously) or on a weighted consideration of the fulfillment or non-fulfillment of the several individual criteria.

[0033] The different modification of the recognized foreign speech intervals of main speakers and secondary speakers in the signal processing step is expressed according to the invention in that that foreign speech intervals of the or each speaker classified as the main speaker are dynamically compressed to a lesser extent than foreign speech intervals of the or each speaker classified as a secondary speaker (in particular are processed without compression), and / or that foreign speech intervals of the or each speaker classified as the main speaker are subjected to lesser noise reduction (active noise cancelling) than foreign speech intervals of the or each speaker classified as a secondary speaker.

[0034] Optionally, additionally, foreign speech intervals of the speaker or speakers classified as the main speaker are amplified to a greater extent and / or subjected to direction-dependent attenuation (beamforming) to a greater extent than foreign speech intervals of the speaker or speakers classified as the secondary speaker; the directional lobe of the beamforming algorithm is particularly directed towards the main speaker.

[0035] The difference in the processing of the voice components of the main and secondary speakers is fixed (unchangeable) in practical embodiments of the invention.

[0036] In a preferred variant of the invention, however, this difference is varied depending on the communication quality. For the speaker or speakers classified as the primary speaker, a measure (i.e., a parameter) of the communication quality is determined, which is characteristic of the success of the information transmission between this primary speaker and the user and / or of the user's listening effort associated with this communication.The aforementioned measure of communication quality (also referred to as "quality measure" for short) has a comparatively high value, for example, if the user perceives the information conveyed by a speaker classified as the main speaker without noticeably increased listening effort; it has a comparatively low value, however, if the user shows increased listening effort during foreign speech intervals of this main speaker, noticeably does not understand the information conveyed by this main speaker, or does not notice this main speaker at all.

[0037] The quality measure is preferably a continuously variable value, for example, a floating-point number that can assume a variable value between two predetermined limits. Alternatively, in simple embodiments of the invention, the quality measure can also be a binary value. In each of the aforementioned cases, the modification of the foreign speech intervals assigned to this primary speaker in the signal processing step is carried out depending on the specified quality measure.

[0038] When determining the quality measure, the same or similar criteria as for the interaction classification are preferably used. Thus, the quality measure is preferably based on the spatial orientation and / or the distance of the main speaker relative to the user's head, based on the temporal sequence (turn-taking) of the foreign speech intervals assigned to the main speaker of the recognized own speech intervals, based on the physiological reaction of the user during a foreign speech interval assigned to the main speaker, based on the volume of the main speaker's voice share and / or based on an evaluation of behavioral patterns as described above determined. Fixation of the main speaker by the user (i.e., an exceptionally strong focus of the user on the main speaker), an exceptionally close distance between the user and the main speaker, a physiological reaction of the user characteristic of exceptionally high listening effort or frustration, and an increased volume of the main speaker's voice are considered indicators of poor communication quality.

[0039] Additionally or alternatively, in an advantageous embodiment of the invention, a spectral property, in particular a fundamental frequency (pitch), of the user's voice is determined for at least one own speech interval and / or a spectral property of the main speaker's voice is determined for at least one foreign speech interval assigned to the main speaker. In these cases, the quality measure is determined (exclusively or at least also) based on the spectral property of the user's voice or on the spectral property of the main speaker's voice. For example, a fundamental frequency of the user's or main speaker's voice that is elevated compared to a normal value is considered an indication of poor communication quality. This variant of the invention is based on the finding that the user or other speakers typically tend to raise their voice in situations with poor communication quality.

[0040] Additionally or alternatively, a volume of the recorded sound signal (in particular, the volume of the user's own voice or the main speaker) is determined, preferably for at least one self- or foreign-speech interval, and taken into account when determining the quality measure. This variant of the invention is based on the experience that people (and thus in particular the user of the hearing system and the main speakers communicating with it) tend to speak louder in situations with poor communication quality.

[0041] Additionally or alternatively, a speech rhythm or the speech rate (speech speed) of the user is preferably determined for at least one self-speech interval, and / or a speech rhythm of the main speaker is determined for at least one foreign speech interval assigned to a main speaker. The speech rhythm of the user or the main speaker is taken into account when determining the quality measure. This variant of the invention is based on the finding that people (and thus also the user of the hearing system and the main speakers communicating with it) tend to speak with a different speech rhythm in situations with poor communication quality compared to normal situations.For example, poor communication quality often manifests itself in a slowed speech rhythm, as the user or other speaker attempts to achieve better understanding by speaking slowly. On the other hand, situations with poor communication quality can also be associated with an exceptionally accelerated speech rhythm as a result of dissatisfaction on the part of the user or other speaker. An exceptionally increased or decreased speech rhythm of the user or main speaker is therefore considered an indication of poor communication quality.

[0042] The speech analysis unit preferably calculates the quality measure based on a weighted analysis of the indicators described above. Alternatively, the speech analysis unit sets the quality measure to a value indicating poor communication quality if several of the above-mentioned indicators are met simultaneously.

[0043] The hearing system according to the invention is generally configured to automatically carry out the above-described method according to the invention. The hearing system is thus configured to record a sound signal from the environment of the hearing instrument using the at least one input transducer of the at least one hearing instrument, to modify the recorded sound signal in the signal processing step to support the hearing ability of a user, and to output the modified sound signal using the output transducer of the hearing instrument. The hearing system is further configured to detect foreign speech intervals in the analysis step, to identify different speakers in detected foreign speech intervals, and to assign each foreign speech interval to the speaker speaking in this foreign speech interval.Finally, the hearing system is also designed to classify the assigned speaker as the main speaker or as a secondary speaker for each recognized foreign speech interval during the interaction classification, and to carry out the modification of the recognized foreign speech intervals in different ways in the signal processing step depending on the interaction classification (i.e. depending on whether the assigned speaker was classified as the main speaker or as a secondary speaker).

[0044] The hearing system's mechanism for automatically implementing the method according to the invention is of a programming and / or circuitry nature. The hearing system according to the invention thus comprises programming means (software) and / or circuitry (hardware, e.g., in the form of an ASIC) that automatically implement the method according to the invention during operation of the hearing system. The programming or circuitry for implementing the method can be located exclusively in the hearing instrument (or hearing instruments) of the hearing system. Alternatively, the programming or circuitry for implementing the method is distributed across the hearing instrument(s) and at least one other device or software component of the hearing system.For example, programming means for carrying out the method are distributed between at least one hearing instrument of the hearing system and a control program installed on an external electronic device (in particular a smartphone).

[0045] The above-described embodiments of the method according to the invention correspond to corresponding embodiments of the hearing system according to the invention. The above statements regarding the method according to the invention can be applied accordingly to the hearing system according to the invention, and vice versa.

[0046] In In preferred embodiments, the hearing system is particularly designed to, in the analysis step to record a spatial orientation (and optionally a distance) of this speaker relative to the user's head for at least one (preferably for each) identified speaker and to take this into account in the interaction classification, also to recognize own speech intervals, to record a temporal sequence (turn-taking) of the assigned foreign speech intervals and the recognized own speech intervals for at least one (preferably for each) identified speaker and to take this into account in the interaction classification, to determine an average volume and / or a signal-to-noise ratio for each recognized foreign speech interval and to take this into account in the interaction classification, and / or to record a physiological reaction of the user for each recognized foreign speech interval and to take this into account in the interaction classification.

[0047] In further preferred embodiments, the hearing system is particularly designed to, in the signal processing step To amplify foreign speech intervals of the speaker or speakers classified as the main speaker to a greater extent than foreign speech intervals of the speaker or speakers classified as secondary speakers, To dynamically compress foreign speech intervals of the speaker or speakers classified as the main speaker to a lesser extent than foreign speech intervals of the speaker or speakers classified as secondary speakers (in particular not to compress them at all), To subject foreign speech intervals of the speaker or speakers classified as the main speaker to a lesser degree of noise reduction (active noise cancelling) and / or feedback suppression (active feedback cancelling) than foreign speech intervals of the speaker or speakers classified as secondary speakers,and / or subject foreign speech intervals of the speaker or speakers classified as the main speaker to a greater degree of direction-dependent attenuation (beamforming) than foreign speech intervals of the speaker or speakers classified as secondary speakers.

[0048] In further preferred embodiments, the hearing system is particularly configured to record a measure of the communication quality (quality measure) for the speaker or each speaker classified as the main speaker and to modify the foreign speech intervals assigned to this main speaker as a function of this quality measure.

[0049] The hearing system is particularly designed to measure the quality in the manner described above based on the spatial orientation (and / or distance) of the main speaker relative to the user's head, based on the temporal sequence (turn-taking) of the foreign speech intervals assigned to the main speaker and the recognized own speech intervals, based on the user's physiological reaction during a foreign speech interval assigned to the main speaker, based on a spectral property, in particular the fundamental frequency, of the user's and / or a main speaker's voice, based on the volume of an own speech interval and / or a foreign speech interval (in particular based on the volume of the user's own voice or the voice of the main speaker), based on the speech rhythm (speech speed) of the user and / or a main speaker, and / or based on an evaluation of behavioral patterns as described above to determine.

[0050] In the following, exemplary embodiments of the invention are explained in more detail with reference to a drawing. In the drawings: Fig. 1 shows a schematic representation of a hearing system consisting of a single hearing instrument in the form of a hearing aid that can be worn behind the ear of a user, Fig. 2 shows a flowchart of a method for operating the hearing system from Fig. 1 , Fig. 3 in a flow diagram an alternative embodiment of the method, and Fig. 4 in representation according to Fig. 1 an alternative embodiment of the hearing system, in which it comprises a hearing instrument in the form of a hearing aid that can be worn behind the ear and a control program implemented in a smartphone.

[0051] Identical parts and sizes are always marked with the same reference symbols in all figures.

[0052] Fig. 1 shows a hearing system 2 with a single hearing aid 4, i.e., a hearing instrument configured to support the hearing of a hearing-impaired user. In the example shown here, the hearing aid 4 is a BTE hearing aid worn behind the user's ear.

[0053] Optionally, in a further embodiment of the invention, the hearing system 2 comprises a second hearing aid (not expressly shown) for supplying the user's second ear.

[0054] Within a housing 5, the hearing aid 4 comprises at least one microphone 6 (two microphones in the example shown) as an input transducer and a receiver 8 as an output transducer. The hearing aid 4 further comprises a battery 10 and a signal processing unit in the form of a signal processor 12. The signal processor 12 preferably comprises both a programmable subunit (for example, a microprocessor) and a non-programmable subunit (for example, an ASIC). The signal processor 12 comprises a (voice recognition) unit 14 and a (speech analysis) unit 16. In addition, the signal processor 12 optionally has a (physiology analysis) unit 18, which evaluates signals from one or more—also optional—biosensors 19, e.g., signals from a heart rate monitor, a skin resistance sensor, a body temperature sensor, and / or an EEG sensor.

[0055] Preferably, the units 14 to 18 are designed as software components that are implemented in the signal processor 12. The or each biosensor 19 can - as exemplified in Fig. 1 shown - be integrated in the hearing aid 2. The physiological analysis unit 18 can additionally or alternatively also receive signals from one or more external (ie arranged outside the housing 5) biosensors.

[0056] The signal processor 12 is supplied with an electrical supply voltage U from the battery 10.

[0057] During normal operation of the hearing aid 4, the microphones 6 pick up airborne sound from the surroundings of the hearing aid 4. The microphones 6 convert the sound into an (input) audio signal I containing information about the picked up sound. The input audio signal I is fed to the signal processor 12 within the hearing aid 4.

[0058] The signal processor 12 processes the input audio signal I using a plurality of signal processing algorithms, for example directional attenuation (beamforming), noise and / or feedback suppression, dynamic compression and frequency-dependent amplification based on audiogram data, to compensate for the user's hearing loss. The respective operation of the signal processing algorithms, and thus of the signal processor 12, is determined by a variety of signal processing parameters. The signal processor 12 outputs an output audio signal O, which contains information about the processed and thus modified sound, to the earpiece 8.

[0059] The receiver 8 converts the output sound signal O into a modified airborne sound. This modified airborne sound is transmitted into the user's ear canal via a sound channel 20 connecting the receiver 8 to a tip 22 of the housing 5, as well as via a flexible sound tube (not explicitly shown) connecting the tip 22 to an earpiece inserted into the user's ear canal.

[0060] The voice recognition unit 14 generally detects the presence of voice (ie spoken language, regardless of the person speaking) in the input audio signal I. The voice recognition unit 14 therefore does not distinguish between the user's voice and the voice of another speaker and thus generally identifies speech intervals, ie time periods in which the input audio signal I contains spoken speech.

[0061] The speech analysis unit 16 evaluates the recognized speech intervals and determinesthe orientation of the speaking person with respect to the user's head, in particular by comparing the voice component in several differently directed variants of the input audio signal I (beamformer signals) or by adjusting the direction of greatest amplification or attenuation of a directional algorithm (beamformer) such that the voice component in the input audio signal I is maximized, (an estimated value for) the distance of the speaking person with respect to the user's head; in order to estimate this distance, the speech analysis unit 16 evaluates in particular the volume in combination with averaged angular velocities and / or amplitudes of the temporal change in the orientation of the speaking person, wherein orientation changes based on a head rotation of the user are preferably subtracted or otherwise disregarded;During this evaluation, the speech analysis unit 16 takes into account that, from the user's perspective, the orientation of another speaker typically changes more quickly and over a larger area the smaller the distance between that speaker and the user; additionally or alternatively, to estimate the distance, the speech analysis unit 16 evaluates the fundamental frequency and the relationship between the formants of vowels in speech intervals of the user and / or the other speakers; in doing so, the speech analysis unit 16 uses the knowledge that the aforementioned sizes of speakers are typically varied in a characteristic manner depending on the distance to the respective listener (depending on the distance to the listener, the speaking person typically varies their manner of speaking between whispering, normal speech and shouting in order to make themselves understood);Furthermore, the hearing system 2 is preferably configured to detect wirelessly communicating electronic mobile devices (e.g. smartphones, other hearing aids, etc.) in the user's environment, to determine the respective distance of the detected devices and to use the detected distance values ​​to determine or verify the plausibility of the distance of the speaking person to the user's head, the fundamental frequency (pitch) of the input audio signal I or the voice component in the input audio signal I, the speech rhythm (speech speed) of the speaking person, the level (volume) of the input audio signal I or the voice component in the input audio signal I and / or a signal-to-noise ratio of the input audio signal I. ;

[0062] Based on one or more of the above-mentioned variables, the speech analysis unit 16 distinguishes the recognized speech intervals into own speech intervals in which the user speaks, and foreign speech intervals in which a speaker (different from the user) speaks.

[0063] The speech analysis unit 16 recognizes eigenspeech intervals in particular by the fact that an unchanged orientation from the front is determined for the voice portion of the input audio signal I in these intervals with respect to the user's head.

[0064] In addition, the speech analysis unit 16 optionally evaluates the fundamental frequency and / or the speech rhythm of the voice component for the purpose of self-voice recognition and compares this, for example, with stored reference values ​​of the fundamental frequency or the speech rhythm of the user.

[0065] Again, or alternatively, the speech analysis unit 16 applies other methods known per se for self-voice recognition, such as those known, for example, from US 2013 / 0148829 A1 or WO 2016 / 078786 A1. Again, additionally or alternatively, the speech analysis unit 16 uses structure-borne sound information or a recorded inner-ear sound signal for self-voice recognition; this is based on the knowledge that the user's own voice is measured with a significantly stronger proportion in the structure-borne sound transmitted via the user's body or in the inner-ear sound signal measured in the user's ear canal than in the input audio signal I corresponding to the ambient sound.

[0066] On the other hand, the speech analysis unit evaluates 16 recognized foreign speech intervals in order to distinguish between different speakers and thus to assign each foreign speech interval to a specific individual speaker.

[0067] The speech analysis unit 16 evaluates the analyzed foreign speech intervals, for example, with regard to the fundamental frequency and / or speech rhythm. In addition, the speech analysis unit 16 preferably also evaluates the orientation and, if applicable, the distance of the detected speech signals in order to distinguish between different speakers. This evaluation particularly utilizes the fact that, on the one hand, the position of a speaker relative to their head and to other speakers cannot change abruptly, and, on the other hand, that two speakers cannot be in the same location at the same time. The speech analysis unit 16 therefore evaluates constant or continuously varying orientation and distance values ​​as an indication that the corresponding speech signals originate from the same speaker.Conversely, the speech analysis unit 16 evaluates uncorrelated changes in orientation and distance values ​​of the respective voice portion of two foreign speech intervals as an indication that the corresponding speech signals originate from different speakers.

[0068] The speech analysis unit 16 preferably creates profiles of recognized speakers with respective reference values ​​for several of the aforementioned variables (fundamental frequency, speech rhythm, orientation, distance) and, when analyzing each foreign speech interval, determines whether the corresponding variables are compatible with the reference values ​​from one of the profiles. If this is the case, the speech analysis unit 16 assigns the foreign speech interval to the respective profile (and thus to the respective recognized speaker). Otherwise, the speech analysis unit 16 assumes that the analyzed foreign speech interval can be assigned to a new (as yet unknown) speaker and creates a new profile for this speaker.

[0069] If several people (e.g., the user and at least one other speaker, or several speakers other than the user) are speaking simultaneously at a given time, the speech analysis unit 16 evaluates the respective voice components, preferably separately, using spatial source separation (using beamforming algorithms). In this case, this results in several self- and / or foreign speech intervals that overlap in time.

[0070] The speech analysis unit 16 further records information about the temporal length and sequence of the own speech intervals and the foreign speech intervals, as well as the speakers assigned to the foreign speech intervals. Using this information, the speech analysis unit 16 determines parameters relevant to turn-taking between the own speech intervals and the foreign speech intervals of a specific speaker. "Turn-taking" refers to the organization of the speech contributions of two speakers in a conversation, particularly the sequence of these speakers' speech contributions. Relevant parameters of "turn-taking" include, in particular, uninterrupted speech contributions (TURNS) of the speakers, overlaps (OVERLAPS), gaps (LAPSES), pauses (PAUSES), and switches (SWITCHES), as defined, for example, in S.A. Chowdhury et al., "Predicting User Satisfaction from Turn-Taking in Spoken Conversations," Interspeech 2016.

[0071] The language analysis unit 16 specifically determines the temporal length or temporal frequency of TURNS of the user and / or TURNS of the other speaker, where a TURN is an interval of one's own or another's speech without a PAUSE, during which the respective conversation partner is silent; the temporal length or temporal frequency of PAUSES, where a PAUSE is an interval of the input audio signal I without a speech component that separates two consecutive TURNS of the user or two consecutive TURNS of the other speaker, provided that the temporal length of this interval exceeds a predetermined threshold; optionally, PAUSES between TURNS of the user and PAUSES between TURNS of the or each other speaker are each recorded and evaluated separately; alternatively, all PAUSES are recorded and evaluated together;the temporal length or temporal frequency of LAPSES, where a LAPSE is an interval of the input audio signal I without speech component between a TURN of the user and a subsequent TURN of the other speaker or between a TURN of the other speaker and a subsequent TURN of the user if the temporal length of this interval exceeds a predetermined threshold; optionally, LAPSES between a TURN of the user and a TURN of the other speaker and LAPSES between a TURN of the other speaker and a TURN of the user are each recorded and evaluated separately; alternatively, all LAPSES are recorded and evaluated together; ; the temporal length or temporal frequency of OVERLAPS, where an OVERLAP is an interval of the input audio signal I in which both the user and the other speaker are speaking; preferably, such an interval is only counted as an OVERLAP if the temporal length of this interval exceeds a predetermined threshold; optionally, OVERLAPS between a TURN of the user and a subsequent TURN of the other speaker and OVERLAPS between a TURN of the other speaker and a subsequent TURN of the user are each recorded and evaluated separately; alternatively, all OVERLAPS are recorded and evaluated together;and / or the temporal frequency of SWITCHES, where a SWITCH is a transition from a TURN of the user to a TURN of the other speaker or a transition from a TURN of the other speaker to a subsequent TURN of the user without an OVERLAP or intermediate PAUSE, in particular a transition within a certain temporal threshold; optionally, SWITCHES between a TURN of the user and a TURN of the other speaker and SWITCHES between a TURN of the other speaker and a subsequent TURN of the user are recorded and evaluated separately; alternatively, all SWITCHES are recorded and evaluated together.

[0072] The parameters relevant for turn-taking are determined separately for each recognized speaker. The speech analysis unit 16 thus determines separately for each recognized speaker the temporal sequence of the foreign speech intervals of this individual speaker with the own speech intervals (and thus how this speaker interacts with the user).

[0073] Based on the analysis described above, the speech analysis unit 16 performs an interaction classification, during which each speaker is classified as a "main speaker" or a "secondary speaker" as described above.

[0074] The speech analysis unit 16 preferably evaluates several of the above-described parameters in a comparative analysis, in particular one or more parameters relevant for turn-taking, the orientation and distance of the respective speaker to the user's head, the level of the respective speaker's voice share and optionally also the level of the user's voice share and / or the fundamental frequency of the speaker's voice share and optionally also the fundamental frequency of the user's voice share.

[0075] The language analysis unit evaluates 16 as indications that a particular speaker is a main speaker. the finding that the user predominantly turns towards this speaker during the foreign speech intervals assigned to this speaker, so that the voice share of this speaker comes predominantly from the front; the user's attention to the speaker is interpreted as an indication that the user is consciously listening to the speaker, regardless of whether the speaker addresses the user directly or not; the finding that the user and this speaker predominantly speak in a temporally synchronized manner (in particular, alternately), so that the own speech intervals and foreign speech intervals of this speaker are in a temporally correlated sequence; the speech analysis unit 16 recognizes this in particular by a comparatively high frequency of SWITCHES and / or a comparatively low frequency of OVERLAPS and / or a comparatively low frequency of LAPSES in the communication between the user and this speaker;for example, the speech analysis unit 16 compares the frequencies of SWITCHES, OVERLAPS and LAPSES with corresponding threshold values; a temporally correlated sequence of the speech contributions of the user and a speaker is, on the one hand, an indicator of a reciprocal conversation between the speaker and the user; a correlated sequence of the speech contributions, however, also allows a communication situation to be identified in which a speaker wants to make contact with the user, even if the user may not even notice this, or a communication situation in which the user interrupts his or her own speech in order to listen to the speaker; the finding that this speaker (absolutely or in comparison to other speakers) is within a certain distance from the user's head; this is based on the knowledge that conversation partners often have a comparatively narrow distance of, for example, between 80 centimeters and 2 meters;depending on the cultural background of the user and the speaker, the ambient noise level, the location of the conversation and the group size) to each other, while closer or further distances between conversation partners rarely occur. For this purpose, the speech analysis unit 16 compares the distance of the speaker to the user's head with stored threshold values; these threshold values ​​can be fixed or varied depending on the user; a reduction in the distance between the user and the speaker, especially if it is temporally correlated with speech contributions from the user and / or this speaker, is optionally also considered as an indication of a direct communication relationship; and the finding that the voice share of this speaker lies within a predetermined level range;This is based on the knowledge that conversation partners usually adjust the volume of their own voice so that it can be clearly heard by the other conversation partner, depending on the ambient conditions, and is neither too quiet nor too loud; the predetermined level range is optionally varied depending on the background noise level; an increase in the speaker's volume is optionally also interpreted as an indication that the speaker is attempting to contact the user.

[0076] Optionally, the speech analysis unit 16 also evaluates signals from the physiological analysis unit 18 during interaction classification. As an indication that a specific recognized speaker is a main memory, the speech analysis unit 16 evaluates the finding that the signals from the biosensor 19 evaluated by the physiological analysis unit 18 (or, if the hearing aid 2 accesses signals from multiple biosensors, at least one of these signals) show a change that correlates with the foreign speech intervals of this speaker over time. The detection of a physiological reaction that correlates with the foreign speech intervals of a specific speaker suggests increased attention on the part of the user and thus conscious listening on the part of the user.

[0077] During the interaction classification, the speech analysis unit 16 classifies a specific recognized speaker as the primary speaker, particularly if several (e.g., at least two or at least three) of the above-described indicators for this speaker are met. Otherwise, this speaker is classified as a secondary speaker. This classification is variable over time. A speaker classified as the primary speaker can be reclassified as a secondary speaker, and vice versa.

[0078] Depending on whether a particular recognized speaker has been classified as a main speaker or a secondary speaker, the voice components of this speaker in the input audio signal I are processed differently by the signal processor 12. In this case, a foreign speech interval, if the speaker assigned to this speech interval has been classified as the main speaker, amplified with a larger gain factor, dynamically compressed to a lesser extent, subjected to less noise or feedback suppression, and / or subjected to a greater degree of directional attenuation as a foreign language interval whose associated speaker was classified as a secondary speaker.

[0079] If the input audio signal I contains voice components of several speakers at a certain point in time, the voice components of these several speakers separated by source separation are processed differently by the signal processor 12 for the speaker or speakers classified as main speakers and the speaker or speakers classified as secondary speakers.

[0080] In Fig. 2 A concrete sequence of the method performed by the hearing system 2 is illustrated by way of example. During normal operation of the hearing aid 4, the voice recognition unit 14 then checks in a step 30 whether the input audio signal I contains spoken speech. If this is the case (Y), the voice recognition unit 14 instructs the signal processor 12 to perform a subsequent step 32. Otherwise (N), the voice recognition unit 14 repeats step 30. The voice recognition unit 14 thereby separates speech intervals of the input audio signal I from intervals without a voice component. Step 32 and the subsequent further steps of the method are only performed in recognized speech intervals.

[0081] In step 32, the signal processor 12 performs source separation. The signal processor 12 uses beamforming algorithms to detect spatially different noise sources (in particular, people speaking) in the input audio signal I and separates the signal components corresponding to these noise sources from each other to enable different processing of these signal components. Typically, the input audio signal I contains only the voice component of a single person speaking (namely, the user or a different speaker).

[0082] In a subsequent step 34, the speech analysis unit 16 checks whether the voice component detected in the input audio signal I (or, if applicable, one of the voice components detected in the input audio signal I) contains the user's own voice. If this is the case (Y), the speech analysis unit 16 applies a subsequent step 36 to this voice component.

[0083] Otherwise (N), ie, to voice portions of the input audio signal I that contain the voice of a speaker different from the user, a step 38 is applied by the speech analysis unit 16. In this way, the speech analysis unit 16 analyzes the own speech intervals and the foreign speech intervals in the input audio signal I separately from one another.

[0084] To determine the parameters relevant for turn-taking, the speech analysis unit 16 determines the start and end times for the or each recognized eigenspeech interval in step 36.

[0085] In a subsequent step 40, the speech analysis unit 16 initiates the setting of signal processing parameters of the signal processor 12 for the detected self-speech interval, which are optimized for processing the user's own voice. The signal processor 12 then returns to step 30 in the process execution.

[0086] For each detected foreign speech interval, the speech analysis unit 16 identifies the respective speaker in step 38 by determining characteristics (orientation, distance, fundamental frequency, speech rhythm) of the voice component in the input audio signal I in the manner described above and comparing them with corresponding reference values ​​of stored speaker profiles. The speech analysis unit 16 assigns the respective foreign speech interval - if possible - to a compatible speaker profile or, if not, creates a new speaker profile. The speech analysis unit 16 also checks which speakers are active in the user's current listening situation. Speaker profiles that cannot be assigned to a foreign speech interval over a certain period of time (e.g., depending on the group size of the speakers and the listening environment) are deleted by the speech analysis unit 16.

[0087] For each of the identified speakers, the speech analysis unit 16 also records the start and end times of the respective assigned foreign speech intervals in a step 41 in order to determine the parameters relevant for turn-taking.

[0088] In a step 42, the speech analysis unit 16 checks whether more than one speaker is active in the user's current listening situation, ie whether more than one of the stored speaker profiles is present or active.

[0089] In this case (Y), the speech analysis unit 16 performs a subsequent step 44. Otherwise (N), the signal processor 12 returns to step 30. The further procedure is thus only performed in multi-speaker environments.

[0090] In step 44, the speech analysis unit 16 uses the start and end times of the own and foreign speech intervals recorded in steps 36 and 41 to determine the parameters (TURNS, PAUSES, LAPSES, OVERLAPS, SWITCHES) relevant for turn-taking between the user and each identified speaker and their length or temporal frequency.

[0091] In a subsequent step 46, the speech analysis unit 16 performs the interaction classification described above. It assesses whether several of the aforementioned indicators for a direct communication relationship between the user and the speaker to whom the currently tested foreign language interval is assigned are met.

[0092] If this is the case (Y), the speech analysis unit 16 classifies this speaker as the main speaker and notes this classification in the associated speaker profile. It then initiates, in a step 48, the

[0093] Speech interval the setting of hearing aid parameters that are optimized for processing voice components of a main speaker, in particular a comparatively high amplification of the input audio signal I, a comparatively low dynamic compression, and a comparatively strong direction-dependent attenuation (beamforming) of the input audio signal I, whereby the directional lobe of the beamforming algorithm is aligned in particular according to the orientation of this main speaker to the user's head.

[0094] Otherwise (N), the speech analysis unit 16 classifies this speaker as a crosstalker and also notes this classification in the associated speaker profile. It then initiates, in a step 50, the setting of hearing aid parameters for the relevant foreign speech interval that are optimized for processing voice components of a crosstalker, in particular a comparatively low amplification of the input audio signal I, a comparatively strong dynamic compression, and a comparatively low direction-dependent attenuation.

[0095] In both cases, i.e. after step 48 as well as after step 50, the signal processor 12 then returns to step 30.

[0096] In Fig. 3 is a variant of the procedure from Fig. 2 The procedure according to Fig. 3 corresponds in large part to the above based on Fig. 2 It comprises in particular the steps 30 to 50 described above. In addition, the method according to Fig. 3 but two additional steps 52 and 54.

[0097] Step 52 is performed after the interaction classification (step 46) if the speaker to whom the current foreign language interval is assigned has been classified as the main speaker.

[0098] In this case, the speech analysis unit 16 checks whether a difficult listening situation (i.e. associated with increased listening effort and / or frustration of the user) exists based on the parameters relevant for turn-taking, based on the volume and the fundamental frequency of the user's own voice in a previous own speech interval, the volume and the fundamental frequency of the speaker's voice in the current foreign speech interval and optionally based on the signals of the or each biosensor 19 evaluated by the physiology analysis unit 18.

[0099] If this is the case (Y), the speech analysis unit 16 initiates, in step 54, an adjustment of the signal processing parameters to be set in step 48. For example, the speech analysis unit 16 increases the gain factor to be applied to the processing of vocal components of a main speaker or decreases the dynamic compression to be applied in this case.

[0100] Otherwise (N), i.e. if the test carried out in step 52 does not reveal any indications of a difficult listening situation, step 54 is skipped in the process execution.

[0101] Through steps 52 and 54, the signal processing parameters are adjusted as needed in a kind of control loop to make it easier for the user to understand the direct communication partner(s) (main speaker) in difficult listening situations. The adjustments to the signal processing parameters made in step 54 are gradually reversed once the difficult listening situation has ended and the test performed in step 52 produces a negative result over a certain period of time.

[0102] Fig. 4 shows a further embodiment of the hearing system 2, in which it comprises, in addition to the hearing aid 4 (or two hearing aids of this type for supplying both ears of the user), a control software. This control software is referred to below as hearing app 60. The hearing app 60 is in the Fig. 4 In the example shown, it is installed on a smartphone 62. The smartphone 62 itself is not part of the hearing system 2. Rather, the smartphone 62 is used by the hearing app 60 merely as a resource for storage space and computing power.

[0103] During operation of the hearing system 2, the hearing aid 4 and the hearing app 62 exchange data via a wireless data transmission connection 64. The data transmission connection 64 is based, for example, on the Bluetooth standard. The hearing app 62 accesses a Bluetooth transceiver of the smartphone 62 to receive data from the hearing aid 4 and to send data to it. The hearing aid 4, in turn, includes a Bluetooth transceiver to send data to the hearing app 60 and to receive data from this app.

[0104] In the execution according to Fig. 4 are parts of the documents required for the implementation of the procedure in accordance with Fig. 2 or Fig. 3 required software components are not implemented in the signal processor 12, but rather in the hearing app 60. For example, in the embodiment according to Fig. 4 the speech analysis unit 16 or parts thereof are implemented in the hearing app 60.

[0105] The invention is particularly clear from the exemplary embodiments described above, but is not limited to these exemplary embodiments. Rather, further embodiments of the invention can be derived by those skilled in the art from the claims and the above description. List of reference symbols

[0106] 2Hearing system 4Hearing aid 5Case 6Microphone 8Receiver 10Battery 12Signal processor 14(Voice recognition) unit 16(Speech analysis) unit 18Physiology analysis unit 19Biosensor 20Sound channel 22Tip 30Step 32Step 34Step 36Step 38Step 40Step 41Step 42Step 44Step 46Step 48Step 50Step 52Step 54Step 60Hearing app 62Smartphone 64Data transmission connection USupply voltage I(Input) audio signal O(Output) audio signal

Claims

1. A method for operating a hearing system (2) for assisting the hearing of a user, having at least one hearing instrument (4) worn in or on the ear of the user, - wherein a sound signal from an environment of the hearing instrument (4) is recorded by means of an input transducer of the hearing instrument (4), - wherein the recorded sound signal is modified in a signal processing step to assist the hearing of the user, - wherein the modified sound signal is output by means of an output transducer (8) of the hearing instrument (4), - wherein in an analysis step - foreign speech intervals are recognized, in which the recorded speech signal contains speech of a speaker different from the user, - various speakers are identified and recognized foreign speech intervals, and wherein each foreign speech interval is assigned to the speaker who speaks in this foreign speech interval, characterized in that - for each recognized foreign speech interval, the assigned speaker is classified in the course of an interaction classification as to whether this speaker is in a direct communication relationship with the user as a main speaker or whether this speaker is not in a direct communication relationship with the user as a secondary speaker, - in the signal processing step, the modification of the recognized foreign speech intervals takes place depending on the interaction classification, and - in the signal processing step - foreign speech intervals of the or each speaker classified as a main speaker are dynamically compressed to a greater extent than foreign speech intervals of the or each speaker classified as a secondary speaker, and / or - foreign speech intervals of the or each speaker classified as a main speaker are subjected to last noise reduction and / or feedback suppression than foreign speech intervals of the or each speaker classified as a secondary speaker.

2. The method as claimed in claim 1, wherein in the analysis step, for at least one identified speaker, a spatial orientation of this speaker relative to the head of the user is detected and taken into consideration in the interaction classification.

3. The method as claimed in claim 1 or 2, - wherein in the analysis step, own speech intervals are also recognized, in which the recorded sound signal contains speech of the user, - wherein for the at least one identified speaker, a chronological sequence of the assigned foreign speech intervals and the recognized own speech intervals is detected and taken into consideration in the interaction classification.

4. The method as claimed in any one of claims 1 to 3, wherein in the analysis step, for each recognized foreign speech interval, an average volume and / or a signal-to-noise ratio is ascertained and taken into consideration in the interaction classification.

5. The method as claimed in any one of claims 1 to 4, wherein in the analysis step, for each recognized foreign speech interval, a physiological reaction of the user is detected and taken into consideration in the interaction classification.

6. The method as claimed in any one of claims 1 to 5, wherein in the signal processing step - foreign speech intervals of the or each speaker classified as a main speaker are amplified to a greater extent than foreign speech intervals of the or each speaker classified as a secondary speaker and / or - foreign speech intervals of the or each speaker classified as a main speaker are subjected to direction-dependent damping to a greater extent than foreign speech intervals of the or each speaker classified as a secondary speaker.

7. The method as claimed in any one of claims 1 to 6, wherein for the or each speaker classified as a main speaker, a measure of the communication quality is detected, which is characteristic for the success of the information transmission between this main speaker and the user and the hearing exertion of the user connected thereto, and wherein in the signal processing step, the modification of the foreign speech intervals assigned to this main speaker takes place depending on the measure of the communication quality.

8. The method as claimed in claim 7, wherein a spatial orientation of the or each main speaker relative to the head of the user is detected and wherein the measure of the communication quality is ascertained on the basis of this spatial orientation.

9. The method as claimed in claim 7 or 8, wherein own speech intervals are recognized, in which the recorded sound signal contains speech of the user, wherein for the or each main speaker, a chronological sequence of the assigned foreign speech intervals and the recognized own speech intervals is detected, and wherein the measure of the communication quality is ascertained on the basis of this chronological sequence.

10. The method as claimed in any one of claims 7 to 9, wherein for each foreign speech interval of the or each main speaker, a physiological reaction of the user is detected and wherein the measure of the communication quality is ascertained on the basis of this physiological reaction.

11. The method as claimed in any one of claims 7 to 10, wherein for at least one own speech interval, a spectral property of the voice of the user is ascertained and / or wherein for at least one foreign speech interval assigned to the main speaker, a spectral property of the voice of the main speaker is ascertained, and wherein the measure of the communication quality is ascertained on the basis of the spectral property of the voice of the user or on the basis of the spectral property of the voice of the main speaker.

12. The method as claimed in any one of claims 7 to 11, wherein for at least one own speech interval, a volume is ascertained and / or wherein for at least one foreign speech interval assigned to the main speaker, a volume is ascertained, and wherein the measure of the communication quality is ascertained on the basis of this volume of the own speech interval or foreign speech interval.

13. The method as claimed in any one of claims 7 to 12, wherein for at least one own speech interval, a speech rhythm is ascertained and / or wherein for at least one foreign speech interval assigned to the main speaker, speech or the is ascertained, and wherein the measure of the communication quality is ascertained on the basis of this speech rhythm of the user or the main speaker.

14. A hearing system (2) for assisting the hearing of a user having at least one hearing instrument (4) worn in or on the ear of the user, wherein the hearing instrument (4) comprises: - an input transducer (6) for recording a sound signal from an environment of the hearing instrument (4), - a signal processing unit (12) for modifying the recorded sound signal to assist the hearing of the user, and - an output transducer (8) for outputting the modified sound signal, wherein the hearing system (2) is configured to automatically perform the method as claimed in any one of claims 1 to 13.