CN104717593B - Position learns audiphone - Google Patents
Position learns audiphone Download PDFInfo
- Publication number
- CN104717593B CN104717593B CN201410777867.8A CN201410777867A CN104717593B CN 104717593 B CN104717593 B CN 104717593B CN 201410777867 A CN201410777867 A CN 201410777867A CN 104717593 B CN104717593 B CN 104717593B
- Authority
- CN
- China
- Prior art keywords
- hearing aid
- device system
- aid device
- signal
- detector
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000012545 processing Methods 0.000 claims abstract description 98
- 238000004422 calculation algorithm Methods 0.000 claims description 80
- 230000033228 biological regulation Effects 0.000 claims description 28
- 230000008859 change Effects 0.000 claims description 20
- 230000005236 sound signal Effects 0.000 claims description 19
- 230000007613 environmental effect Effects 0.000 claims description 16
- 230000004044 response Effects 0.000 claims description 12
- 238000012986 modification Methods 0.000 claims description 3
- 230000004048 modification Effects 0.000 claims description 3
- 230000006698 induction Effects 0.000 claims description 2
- 241000894007 species Species 0.000 description 60
- 210000005069 ears Anatomy 0.000 description 18
- 238000000034 method Methods 0.000 description 16
- 238000013507 mapping Methods 0.000 description 13
- 230000000875 corresponding effect Effects 0.000 description 12
- 206010011878 Deafness Diseases 0.000 description 6
- 230000008901 benefit Effects 0.000 description 6
- 230000010370 hearing loss Effects 0.000 description 6
- 231100000888 hearing loss Toxicity 0.000 description 6
- 208000016354 hearing loss disease Diseases 0.000 description 6
- 238000003860 storage Methods 0.000 description 6
- 230000006978 adaptation Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 5
- 230000006835 compression Effects 0.000 description 5
- 238000007906 compression Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 240000006409 Acacia auriculiformis Species 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000000284 resting effect Effects 0.000 description 4
- 241000208340 Araliaceae Species 0.000 description 3
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 3
- 235000003140 Panax quinquefolius Nutrition 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 230000005611 electricity Effects 0.000 description 3
- 235000008434 ginseng Nutrition 0.000 description 3
- 230000001965 increasing effect Effects 0.000 description 3
- 239000013589 supplement Substances 0.000 description 3
- 230000001629 suppression Effects 0.000 description 3
- 238000009825 accumulation Methods 0.000 description 2
- 235000021170 buffet Nutrition 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000007717 exclusion Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000007935 neutral effect Effects 0.000 description 2
- 230000012385 regulation of binding Effects 0.000 description 2
- 238000005728 strengthening Methods 0.000 description 2
- 230000032258 transport Effects 0.000 description 2
- 241000209140 Triticum Species 0.000 description 1
- 235000021307 Triticum Nutrition 0.000 description 1
- 230000003213 activating effect Effects 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- DMBHHRLKUKUOEG-UHFFFAOYSA-N diphenylamine Chemical compound C=1C=CC=CC=1NC1=CC=CC=C1 DMBHHRLKUKUOEG-UHFFFAOYSA-N 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012432 intermediate storage Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- INFDPOAKFNIJBF-UHFFFAOYSA-N paraquat Chemical compound C1=C[N+](C)=CC=C1C1=CC=[N+](C)C=C1 INFDPOAKFNIJBF-UHFFFAOYSA-N 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000008929 regeneration Effects 0.000 description 1
- 238000011069 regeneration method Methods 0.000 description 1
- 230000003252 repetitive effect Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000002463 transducing effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Landscapes
- Telephone Function (AREA)
- Circuit For Audible Band Transducer (AREA)
Abstract
The present invention relates to position to learn audiphone.A kind of new hearing aid device system is provided, its it is determined that acoustic environment species when include geographical position and user feedback, for automatically adjusting for signal processing parameter.
Description
Technical field
New hearing aid device system is provided with response at acoustic environment, geographical position and the hearing aid signal of user feedback
Reason the improved of parameter automatically selects and adjusted.Particularly, new hearing aid device system is characterised by being based on geographical position and pattra leaves
The optimization of the hearing aid signal processing parameter of this increment preference induction.
Background technology
Traditional hearing aid now generally includes digital signal processor (DSP), for handling the sound received by audiphone
Sound, to compensate the hearing loss of user.As is well known in the art, DSP processing by with for adjust be performed
Actual signal processing many kinds of parameters signal processing algorithm control, such as, in each frequency channel of multichannel audiphone
Gain, frequency selectivity filtering algorithm corner frequency or slope, control compressor algorithm flex point and compression ratio parameter.
DSP flexibility is normally used for providing multiple algorithms of different with multi-signal processing parameter.It is for example, a variety of
Algorithm can be provided for noise suppressed, i.e. the decay of undesirable signal and the amplification of desired signal.Desired signal is typically
Voice or music, and undesirable signal can be background sound, restaurant's brouhaha, music (when voice is desired signal),
Traffic noise etc..
Algorithms of different and parameter are generally included, to provide the comfortable and intelligible regeneration in different types of acoustic environment
Sound quality, such as, voice, unclean voice, restaurant's brouhaha, music, traffic noise etc..
The audio signal obtained from alternative sounds environment can possess very different characteristic, for example, average and most loud
Arbitrarily downgrade (SPL) and/or frequency content.So as to which in the audiphone with DSP, the acoustic environment of each species can be with tool
The signal specific Processing Algorithm for having the specific settings of signal processing parameter is associated, and the signal specific Processing Algorithm is in question
Species acoustic environment provide optimum signal quality processing after sound.
So as to, the audiphone based on DSP now is generally provided with multiple unlike signal Processing Algorithms, wherein, each
Algorithm is customized for particular kind of acoustic environment and/or particular user preferences.Signal processing parameter is generally allocator's
It is determined during the initial adaptation phase in office, and by activating Expectation Algorithm and algorithm parameter being arranged on into audiphone
Nonvolatile storage in and/or by Expectation Algorithm and algorithm parameter setting transmit to nonvolatile storage, be programmed into
In audiphone.
The acoustic environment of user can be classified as one of acoustic environment of multiple species by some known audiphones,
Such as, voice, unclean voice, restaurant's brouhaha, music, traffic noise etc..
The classification results obtained can be utilized in audiphone, to automatically select the signal transacting of audiphone spy
Property, for example, being automatically switched to the most suitable signal processing algorithm and parameter in question environment species.It is such to help
Listen device automatically for each hearing aid user optimal sound quality and/or voice can be kept to be appreciated that in muli-sounds environment
Property.
US 2007/0140512A1 and WO 01/76321 discloses the example of classifier methods.
The content of the invention
New hearing aid device system is provided with the geographical position including new hearing aid device system in the species of its determination acoustic environment
Put the audiphone with user feedback.
Acoustic environment in specific geographical area generally remains in identical type over time.So as to it is determined that current sound
Combining geographic location will improve the determination of species during the species of sound environment, i.e. the determination of species can be caused faster, and/or can
To cause the determination of species that there is increased certainty.
A kind of new hearing aid device system is provided, including
(a) the first audiphone, have
First microphone, for the voice signal in response to being received at first microphone in acoustic environment, carry
For the first audio input signal,
First processor, it is configured to handle the first audio input signal according to signal processing algorithm F (Θ), with generation
First hearing compensation audio signal, wherein, Θ is the set of signal processing parameter, and
First output transducer, for based on the first hearing compensation audio signal, there is provided the first sound sense signal,
(b) position detector, it is arranged to determine the geographical position of hearing aid device system,
(c) the first sound environment detector, it is arranged to
Based on the voice signal received by hearing aid device system and the geographical position of identified hearing aid device system, hearing aid is determined
The species of acoustic environment around device system, and
Based on the species of the acoustic environment determined by the first sound environment detector, the first output is supplied to the first processing
Device, it is worth with the first of the set Θ of selection signal processing parameter,
(d) user interface, the user for allowing hearing aid device system make at least one signal processing parameter θ ∈ Θ tune
Section, and wherein
First sound environment detector is arranged to
At least one signal processing parameter θ ∈ Θ made by the user of hearing aid device system regulation is recorded, and
Also based on regulation, the first output is supplied to first processor.
Can based on regulation Bayes's increment preference induce, by first output be supplied to first processor.
Hearing aid device system has signal processing algorithm F (Θ) storehouse, wherein, Θ is algorithm parameter space, including control is used
In the parameter of the selection of the algorithm of execution, for example, noise suppression algorithm can be selected for performing in noise circumstance, and
It can be not selected for performing in quiet environment.
Position detector is included in gps receiver, calendar system, WIFI network interface, mobile telephone network interface extremely
It is few one, for determining the geographical position of hearing aid device system and alternatively determining the speed of hearing aid device system.
First sound environment detector can be configured for by hearing aid device system receive voice signal, determined
Hearing aid device system geographical position and at least one parameter for being selected from by the following group formed determine audiphone system
The species of acoustic environment around uniting：The speed of time, hearing aid device system in date, one day and received by gps receiver
Signal signal intensity.
Acoustic environment at specific geographic position (such as, city square) place can be year after year in a similar manner in 1 year
In and/or changed in a similar manner in one day with repetitive mode day by day, for example, due to traffic, number etc. repetition become
Change, and such change can by allow sound environment detector it is determined that acoustic environment species when include the date and/
Or the time in one day is considered.
When hearing aid device system is in building, the signal intensity of the signal received by gps receiver is substantially reduced, and
So as to which the information on GPS signal strength can be used by sound environment detector, to determine whether hearing aid device system is being built
In thing.
Such as the information on translational speed determined by gps receiver can be used by sound environment detector, with true
Hearing aid device system is determined in means of transport, such as, in automobile.
Audiphone can be any types for being configured to be worn on the head, and with first shift position and orientation, it is all
Such as, the audiphone such as BTE, RIE, ITE, ITC, CIC.
Through the disclosure, term gps receiver is used for the receiver for specifying the satellite-signal of any satellite navigation system,
It is provided on earth or the position of neighbouring any position and temporal information, such as, is safeguarded by U.S. government and can be by appointing
Who freely accesses and is generally designated the satellite navigation system of " GPS- systems ", Russia's worldwide navigation using gps receiver
Satellite system (GLONASS), European Union's Galileo navigation system, Chinese the Big Dipper navigation system, the satellite system of India's area navigation 20
System etc., and also include enhancing GPS, such as, StarFire, Omnistar, India's GPS auxiliary types geostationary track enhanced navigation system
System (GAGAN), European Geostationary navigation overlay service (EGNOS), Japanese multisatellite strengthening system (MSAS) etc..Increasing
In strong GPS, the small change of the signal of the network measure gps satellite of the reference base station based on ground, correction message is sent to GPS
System-satellite, correction message is broadcast back to the earth by it, wherein, enhancing GPS enables receiver and used when calculating their position
Correct to improve the degree of accuracy.The system of International Civil Aviation Organization (ICAO) the type is referred to as satellite-based strengthening system
(SBAS)。
Audiphone may further include one or more orientation sensors, such as, gyroscope, for example, MEMS gyroscope,
Inclination sensor, ball switch etc., the signal that output is used to determine the orientation of the head of the user of wear hearing aid is arranged to,
For example, one or more of the rotation of head yaw angle, the rotation of the head angle of pitch, the rotation of head roll angle or its combination, for example, inclination angle
Or tilt.
Through the disclosure, calendar system is to provide the user with can be by the data of the network access of such as internet
Calendar electronic version system.Well-known calendar system includes such as Mozilla Sunbird, Windows Live
Calendar, Google Calendar, the Microsoft Outlook etc. with swap server.
Through the disclosure, word " inclination " is represented when user stands or sat down, the angular displacement with head normal upright position.
So as to which in the resting position of the head for the people for standing or sitting down, inclination is 0 °, and the resting position of the head in the people to lie down, is inclined
It is tiltedly 90 °.
First sound environment detector can be configured to provide for the first output, with based on selecting to believe with account orientation
The set Θ of number processing parameter the first value, this is that the output signal based on one or more orientation sensors is true with account orientation
Fixed.If lain down for example, user changes into position from standing to doze off, environmental detector can make the first signal transacting
Device correspondingly switches (one or more) signal processing algorithm, for example, the first audiphone can automatically be weakened.
Alternatively, the output signal of one or more orientation sensors can be input into another portion of hearing aid device system
Point, for example, first processor, it is configured for the output signal and the first acoustic environment of one or more orientation sensors
The output of detector, the set Θ of selection signal processing parameter the first value.
Signal processing algorithm can include multiple subalgorithms or subprogram, each be carried out specific in signal processing algorithm
Subtask.As an example, signal processing algorithm can include unlike signal processing subprogram, such as, frequency selectivity filter
Ripple, single or multiple channel compressions, adaptive feedback canceller, speech detection and noise decrease etc..
Moreover, multiple different choices of signal processing algorithm, subalgorithm or subprogram can be grouped together, to form two
Individual, three, four, five or more different preset hearing programmes, user can select in-between according to his/her preference.
Signal processing algorithm will have one or more related algorithm parameters.These algorithm parameters can be generally divided into
It is multiple compared with small parameter set, wherein, the specific part of each such algorithm parameter set and signal processing algorithm or with spy
Sub-routine is related.These parameter sets control their corresponding algorithms or the particular characteristics of subprogram, and such as, wave filter turns
Roll over frequency and slope, the compression threshold of compressor algorithm and ratio, the adaptation rate of adaptive feedback canceller algorithm and detection letter
Number characteristic etc..
During the execution of corresponding signal Processing Algorithm or subprogram, the value of algorithm parameter is preferably in intermediate storage such as
In the volatile data storage area of the processing unit in data RAM areas.The initial value of algorithm parameter is stored in such as EEPROM/ sudden strains of a muscle
In the nonvolatile storage for depositing area or battery back up RAM memory blocks, to allow quilt during these algorithm parameters break in the supply
Keep, the user for interrupting the battery generally by audiphone removes or replaced or the manipulation of ON/OFF switches causes.
Such as the position detector including gps receiver can be included in the first audiphone, for when user is by hearing aid
During its desired operation position that device is worn on the head, in a well-known manner, based on satellite-signal, the geographical position of user is determined
Put.Thus, the current location of user and possible orientation can be provided to such as first based on the data from the first audiphone
Sound environment detector.
Sound environment detector can be arranged to hearing aid parameter and GPS- data being collectively stored in remote service
On device, for example, on the remote server by internet access, can be with the hearing loss (hearing of user
Profile) together, for example, for a variety of GPS locations audiphone setting backup, and/or for other audiphones use
Family is shared in the audiphone setting at a variety of GPS locations.
So as to which sound environment detector can be arranged to regain another made at current GPS location
The audiphone setting of user.Audiphone setting can be according to hearing loss similitude and/or age and/or race and/or ear
Size etc. is grouped, and the user that the audiphone setting of another user can be organized according to as belonging to is chosen.
First sound environment detector can be included in the first audiphone, thus be easy to sound environment detector and hearing aid
Signal transmission between other circuits of device.
Alternatively, such as the position detector including gps receiver can be included in and be set with the hand-held of hearing-aid interconnection
In standby.
Handheld device can be gps receiver, the intelligence electricity such as Iphone, Android phone, Windows phones
Words, such as with gps receiver and calendar system etc., with hearing-aid interconnection.
First sound environment detector can be included in the handheld device.With the available limited computing resource in audiphone
Compared with electric power, the first sound environment detector can benefit from usual larger computing resource available in the handheld device and electricity
Source.
Handheld device can accommodate user interface, and it is arranged to the hearing aid device system for example including the first audiphone
User's control.
Handheld device can have the interface for being used for being connected with the wide area network of such as internet.
Handheld device can access wide area by GSM, IS-95, UMTS, CDMA-2000 etc. mobile telephone network
Net.
For example, by the wide area network of internet, handheld device can access the electronic time management and communication used by user
Instrument, for communicating and for storing the time management related to user and the communication information.Instrument and the information stored are led to
It is frequently located on the remote server by wide-area network access.
Audiphone can include data-interface, and it is used to from handheld device transmit control signal to including the first audiphone
Hearing aid device system other parts.
Audiphone can include data-interface, and it, which is used to transmit the output of one or more orientation sensors to hand-held, sets
It is standby.
Data-interface can be the wireline interface of such as USB interface or the wave point of such as blue tooth interface, for example, blue
Tooth low energy interface.
Audiphone can include COBBAIF, for receiving audio from handheld device and other possible audio signal sources
Signal.
COBBAIF can be wireline interface or wave point.Data-interface and COBBAIF can be combined into single connect
Mouthful, for example, USB interface, blue tooth interface etc..
Audiphone can for example with：Bluetooth low energy data-interface, it is used to hand between audiphone and handheld device
Emat sensor and control signal；And wireline audio interface, for exchanging audio signal between audiphone and handheld device.
First sound environment detector can include fisrt feature extractor, and it is used to determine the first audio input signal
Characterisitic parameter.
Feature extractor can determine the characterisitic parameter of audio input signal, such as, average and maximum sound pressure level (SPL),
Signal power, modal data and other well-known features.Modal data can include discrete Fourier transform coefficient, linear prediction
Coding parameter, cepstrum parameter or respective differences divide cepstrum parameter.
Characterisitic parameter can be output to first environment grader by feature extractor, and it is configured for identified
Characterisitic parameter and geographical position, determine the species of acoustic environment.
First environment grader is arranged to the species of acoustic environment being defined as multiple acoustic environment types or species,
Such as, voice, unclean voice, restaurant's brouhaha, music, traffic noise etc..Classification processing can utilize simple most adjacent
Nearly search, neutral net, hidden Markov model system or another system for being capable of pattern identification.The output of environment classification
Can be comprising a single environment species it is " hard " classify or instruction sound environment belong to corresponding species probability probability
Set.Other outputs can also apply.
The species of identified acoustic environment can be output to the mapping of the first parameter by first environment grader, first ginseng
Number mappings are configured to provide for being used for the output for selecting suitable (one or more) first signal processing algorithm and for by the
The parameter of one computing device.
In this way, the classification results obtained can be utilized in audiphone, to automatically select the letter of audiphone
Number treatment characteristic, for example, being automatically switched to the most suitable algorithm of the acoustic environment for discussing.Such audiphone energy
Enough is that each hearing aid user keeps optimal sound quality and/or speech intelligibility in muli-sounds environment.
As an example, it may be desirable to do not only rely on the grade of ambient noise, and enter one by the ambient noise
Characteristics of signals is walked, is switched between omni-directional and the preset program of directional microphone.Made an uproar in the user of audiphone background being present
In the case of sound with another personal communication in the case of, can identify and the type for ambient noise of classifying is beneficial.Can
Omni-directional operation is selected in the case of to be traffic noise in noise, to allow direction that user reaches with it independently clearly
Hear close traffic.On the other hand, if ambient noise is classified as unclean noise, directive property can be selected
Hearing programmes, to allow user to hear the targeted voice signal with improved signal to noise ratio (SNR) during talk.
Such as microphone can for example be obtained by being used for the analysis of microphone signal and classification using hidden Markov model
The detailed characteristics of signal.Hidden Markov model can change according to short and prolonged time domain, and random and non-stationary is believed
Number modeling.
Geographical position that sound environment detector can be arranged to be determined by position detector with geographical position
The acoustic environment at place determines that species together records.Record can be spaced and/or by the spy between record at regular times
Determine geographic distance execution, and/or by specific events trigger, for example, the conversion of the species of acoustic environment, the change of signal transacting,
Such as, the change of signal handler, the change of signal processing parameter etc..
When hearing aid device system be located at from acoustic environment species precedence record geographical position threshold distance in and/or
When in the region in the precedence record geographical position of the identical recordings of the species with acoustic environment, sound environment detector can be with
Be arranged to increase current sound environment have with or close to the recorded identical type of current geographic position probability,
Or determine that current sound environment has the species of recorded acoustic environment.
First sound environment detector can be configured to consider in from determined geographical positional distance threshold value
, the probability of the generation of the precedence record species of acoustic environment, determine the species of acoustic environment.
Threshold distance can be determined in advance, for example, reflection determines the uncertainty in the geographical position of position detector, example
Such as, less than or equal to the uncertainty of position detector, or the geographical position less than or equal to acoustic environment and the note of species
At average distance between record, or current geographic position less than sports field, master station, City Hall, arenas etc.
The characteristic dimension of notable feature.Threshold distance can be adapted to current environment, for example, in causing the region of such as urban area
Relatively small threshold distance, the region have short distance between the record of different types of acoustic environment, and cause for example to open
Relatively large threshold distance in the region on spacious grassland, the region between the record of different types of acoustic environment have greatly away from
From.
The user interface of hearing aid device system may be configured to make some species of acoustic environment and corresponding geographic area phase
Association.
In the case of in the absence of useful gps signal, what position detector can be may be coupled to based on hearing aid device system
The address of WIFI network, or based on can from field of mobile telephony well-known a variety of GSM- transmitter receipts
Signal by triangulation, determine the geographical position of hearing aid device system.Moreover, position detector can be arranged to access
The calendar system of user, with obtain on user expectation whereabouts information, for example, meeting room, office, buffet, restaurant,
Family etc., and with it is determined that including the information during geographical position.So as to which the information of the calendar system from user can replace
Or information of the supplement on the geographical position of (such as passing through gps receiver) by other means determination.
For example, (one or more) audiphone of hearing aid device system can be automatically switched to fly by sound environment detector
Row mode, i.e. when position detector detects user aboard, (one or more) nothing of (one or more) audiphone
Line electric equipment is closed.
Moreover, when user is in the building of such as high-rise, gps signal may be not present or very weak, make
Obtaining geographical position can not be determined by gps receiver.It is then possible to the letter using the whereabouts on user from calendar system
Breath, there is provided on the information in geographical position, or information from calendar system can supplement the information on geographical position, example
Such as, the instruction of particular conference room can provide the information on which layer in high-rise on hearing aid device system.On height
The information of degree can not generally obtain from gps receiver.
When geographical position can not be determined in addition, for example, when GPS- receivers can not provide geographical position, position inspection
The information from calendar system can automatically be used by surveying device.Not for example from gps receiver and calendar system on
In the case that the information in geographical position can be used for position detector, sound environment detector can be based on the sound letter received
Number, in a conventional manner to sound environment classification；Or audiphone can be configured to for example previously during the adaptation phase or work as
By being operated under the pattern of user's selection when the situation occurs.
User may be dissatisfied to automatically selecting for parameter value, and user interface can be used to perform signal processing parameter
Regulation, for example, the current selection of signal processing algorithm can be changed into another signal processing algorithm by user, for example, with
Family can switch to omni-directional signal Processing Algorithm from directional signal Processing Algorithm.
Sound environment detector is arranged to user's regulation of binding signal process parameter values over time.
Sound environment detector is arranged to the storehouse using signal processing algorithm F (Θ), automatically regulates audiphone system
At least one signal processing parameter θ ∈ Θ in system, wherein, Θ is algorithm parameter space, including control is used for the algorithm of execution
Selection parameter, for example, noise suppression algorithm is selected for performing in noise circumstance, and be not selected for pacifying
Stationary ring performs in border.
Sound environment detector is arranged to
The regulation made by the user of hearing aid device system is recorded, and
Induced based on (Bayes) increment preference, in response to the regulation recorded, change at least one signal processing parameter θ
∈ Θ's automatically adjusts so that when detecting same sound environment next time, performs amended automatically adjust.
Bayesian inference, which is related to collect, to be meaned to assume consistent or inconsistent evidence with given.The confidence level of hypothesis with
The accumulation of evidence and change.By enough evidences, it generally becomes very high or very low.
Before evidence is observed, Bayesian inference uses the Numerical value for the confidence level assumed, and is seen in evidence
After survey, the Numerical value of the confidence level of hypothesis is calculated.
Bayes' theorem adjusts the probability of given fresh evidence in the following manner：
Wherein
H0Represent to be referred to as the null hypothesis inferred before fresh evidence E is changed into available.
P(H0) it is referred to as H0Prior probability,
P(E|H0) be referred to as assuming to assume H0It is true, sees evidence E conditional probability.As the H for being represented as given E0's
During function, it is also known as likelihood function, and
P (E) is referred to as E marginal probability：Fresh evidence E probability is witnessed under all mutual exclusions hypothesis.
It can be calculated as the summation of the full probability of mutual exclusion hypothesis and the product of corresponding conditionses probability：ΣP(E|Hi)P
(Hi)。
P(H0| E) it is referred to as given E H0Posterior probability.
Factor P (E | H0)/P (E) represents the influence for the trust that evidence pair is assumed.If when under consideration being assumed to be is true
When, evidence will be likely to be observed, then the factor will be very big.The prior probability that it will be assumed, which is multiplied by the factor, will obtain given card
According to hypothesis big posterior probability result.So as to, under Bayesian inference, Bayes' theorem measurement fresh evidence will change how much
The trust of hypothesis.
Obtain on Bayes' theorem and the more information of Bayesian inference, c.f.：“Information Theory,
Inference,and Learning Algorithms”by David J.C.Mackay,Cambridge University
Press,2003。
Bayes method for probability theory is the consistent and correlation theory for reasoning under uncertainty.Due to
Consciousness feedback (part) from listener is unknown and generally inconsistent, and it is uncertain that such statistical method needs to handle these
Property.Hereinafter, will be by more detail to the bayes method and particularly Bayes's increment preference abductive approach of audiphone processing
Discuss.
The sound environment detector of new hearing aid device system allows to effectively learn the signal transacting on acoustic environment
The expectation regulation of parameter and individual, change over time, the non-linear and random complex relationship corrected between user's regulation.From
And sound environment detector is considered study sound environment detector.
When each user makes regulation, sound environment detector can update at least one signal processing parameter θ ∈ Θ.Can
Alternatively, renewal can be performed according to specific criteria, for example, the regulation of pre-determined number has been made in user so that only notable
Regulation causes to update.
Sometimes, during the operation of equipment, user is unsatisfied with to the quality of received signal, and so as to be connect by user
Mouth performs (one or more) regulation of hearing aid device system.Learning objective is by sound environment detector that normal mode is slow
Ground is absorbed in model parameter θ.Finally, handling will cause the user for reducing quantity to manipulate.
Only when the knowledge on user preference is available, parameter renewal is performed.When user interface is in the normal operating of equipment
When period is not manipulated, user may be satisfied with to the signal quality transmitted, but this is uncertain.After all, user may
There is no wearable device.However, when user starts to manipulate user interface, it is assumed that user is unsatisfied with the moment.User interface
The starting to be represented as of manipulation stage disagrees the moment.When user manipulates user interface, user may still search for preferably
Regulation.It is next to learn the moment just after user stops manipulating user interface.Now, it is assumed that user finds satisfaction
Regulation；And this is referred to as agreeing to the moment.Disagree and agree to that the moment identifies for collecting passive and active directive data respectively
Situation.
Hereinafter, a kind of illustrative methods suitable for user preference are disclosed.This method is induced based on Bayes's increment preference, but
It is that other method can also.Assuming that disagree the moment for k-th and agree to adjustable signal processing parameter at the moment respectively by
It is arranged to θkdAnd θkc.Further it is assumed that the output of the ambient sound grader during k-th of user's manipulation stage is in CkPlace is substantially
Keep constant.
Obviously, in environmental condition CkUnder, user is compared to θkdIt is preferred that θkc(by (end user) decision-making dk=θkc>θkdTable
Show).Reach the set of all user's decision-makings of k-th of decision-making by Dk-1={ d1,d2,…,dk-1Represent.Then, Bayesian updating
Scheme be used to absorb kth time observation.So that the parameter mapping from type to hearing aid parameter is by probability function p (θ | C, ω) table
Show, wherein, ω represents the parameter for parameter mapping 40.
Then, the Bayesian updating mapped for the parameter of the manipulation based on k-th of user is given by
p(ω|C,Dk)=p (dk|C,ω)×p(ω|C,Dk-1)/p(C)
Chu and Gharamani (Preference Learning with Gaussian Processes, 22nd Int’
L conf on Machine Learning, 2005) in, the Bayesian updating equation is answered in detail, for based on Gauss
The parameter mapping 40 of processing.
New hearing aid device system can be the ears hearing aid device system with two audiphones, an auris dextra for being used for user
An and left ear for being used for user.
So as to which new hearing aid device system can include the second audiphone with second microphone, in response to second
The voice signal received at microphone, there is provided the second audio input signal,
Second processor, it is configured to handle the second audio input signal according to secondary signal Processing Algorithm F (Θ), with
The second hearing compensation audio signal is generated, and
Second output transducer, for providing the second sound sense signal based on the second hearing compensation audio signal.
Because usual binaural listening is lost for two ear differences, except the second audiphone is generally adjusted to compensate for not
Outside the fact that be same as the hearing loss by the hearing loss of the first hearing aid compensation, the circuit of the second audiphone is preferably with first
The circuit of audiphone is identical.
First sound environment detector can be configured for the first and second audio input signals and audiphone system
The geographical position of system, determine the species of the acoustic environment around the user of hearing aid device system.
First sound environment detector can be arranged to the second output being supplied to second processor, for selecting to use
In the secondary signal Processing Algorithm and parameter F (Θ) that are performed by second processor, to generate the second hearing compensation audio letter
Number.
Alternatively, the second audiphone can include the second sound environment measuring similar to the first sound environment detector
Device, and the geographical position of the first and second audio input signals and hearing aid device system is configured for, determine sound ring
The species in border, and for based on the species determined by second sound environmental detector, the second output to be supplied into second processing
Device, with the set Θ of selection signal processing parameter second value.
In ears hearing aid device system, the signal processing algorithm of the first and second signal processors is selected with coordination mode
It is critically important.Due to acoustic environment characteristic at two ears of user it is significantly different, generally occur user two ears
The species of acoustic environment at piece is independently determined difference, and this may cause the undesirable difference of the sound in audiphone
Signal transacting.It is thus preferable to the signal processing algorithm of the first and second processors is chosen based on identical signal, such as,
At handheld device receive voice signal or Zuo Erchu receive voice signal and at auris dextra receive voice signal two
Person or at handheld device receive voice signal and Zuo Erchu receive voice signal and at auris dextra receive sound
Combination of sound signal etc..
As the first sound environment detector, second sound environmental detector can include being used to determine that the second audio is defeated
Enter the second feature extractor of the characterisitic parameter of signal.
Characterisitic parameter can be output to second environment grader by second feature extractor, for based on identified characteristic
Parameter and geographical position, determine the species of acoustic environment.
The species of acoustic environment can be output to the mapping of the second parameter by second environment grader, and it is configured to provide for
For the output for the secondary signal Processing Algorithm for selecting second processor.
As it was previously stated, the method in new hearing aid device system have absorb the user preference that changes over time and/or by with
The ability of the change of the typical sound environment of family experience.The personalization of audiphone can be held during the normal use of audiphone
OK.The user of audiphone in the parameter that these advantages are handled by absorbing audiphone, which is adjusted, to be obtained.Over time, this method is led
Less user during the cycle of user preference is not changed is caused to manipulate.Moreover, method is reliable for inconsistent user behavior
Stable.
User preference for signal processing parameter is in a manner of consistent and be related and according to under uncertainty
The theory of reasoning, is induced during normal use.
New hearing aid device system be capable of learning signal processing parameter expectation regulation and it is personal, change over time, it is non-linear and/
Or the complex relationship between random correction user's regulation.
New hearing aid device system can distinguish the different user preference as caused by alternative sounds environment.Thus, signal transacting is joined
Number can be automatically adjusted according to the consciousness of the user of the most probable parameter setting for actual sound environment.
Medium for storage information/data can use many forms, including but not limited to non-volatile media, volatile
Property medium and transmission medium.Non-volatile media may, for example, be optical storage devices, magnetic memory device or other classes
The memory device of type.Non-volatile media is considered an example of non-transitory medium.Volatile media includes dynamic
State memory, such as, main storage.Volatile media is considered another example of non-transitory medium.Transmission is situated between
Matter includes cable, electric wire and optical fiber.Transmission medium can also use the form of sound or light ripple, such as, in radio wave and infrared
Line number generated according to communication period those.
Signal transacting in new hearing aid device system can be performed by specialized hardware or can be in signal processor
It is performed, or is performed in the combination of specialized hardware and one or more signal processors.
As used herein, term " processor ", " signal processor ", " controller ", " system " etc. are intended to refer to CPU phases
Close entity, hardware, the combination of hardware and software, software or the software being performed.
For example, " processor ", " signal processor ", " controller ", " system " etc. can be, but not limited to be on a processor
Processing, processor, object, executable file, the thread performed, and/or the program of operation.
By way of example, term " processor ", " signal processor ", " controller ", " system " etc. are specified transports on a processor
Both capable application program and hardware processor.One or more " processors ", " signal processor ", " controller ", " system "
Deng or its any combination can be located in processing and/or the thread performed, and one or more " processors ", " at signal
Reason device ", " controller ", " system " etc. or its any combination can be localised on a hardware processor, can be with other
Hardware circuit combines, and/or is distributed between two or more hardware processors, can be combined with other hardware circuits.
Moreover, processor (or similar terms) can be any knot of any component or component that are able to carry out signal transacting
Close.For example, signal processor can be asic processor, FPGA processor, general processor, microprocessor, circuit unit
Or integrated circuit.
A kind of hearing aid device system includes：(a) the first audiphone, there is the first microphone, in response in acoustic environment
In the first microphone at receive voice signal, there is provided the first audio input signal, first processor, it is configured to basis
Signal processing algorithm F (Θ) handles the first audio input signal, to generate the first hearing compensation audio signal, wherein, Θ
It is the set of signal processing parameter, and the first output transducer, for based on the first hearing compensation audio signal, there is provided
First sound sense signal；(b) position detector, it is arranged to determine the geographical position of hearing aid device system；(c) the first sound
Environmental detector, it is configured for the geography of the voice signal and identified hearing aid device system received by hearing aid device system
Position, the species of the acoustic environment around hearing aid device system is determined, and based on the kind determined by the first sound environment detector
Class, the first output is supplied to first processor, with the set Θ of selection signal processing parameter the first value；(d) user interface,
User for allowing audiphone makes at least one signal processing parameter θ ∈ Θ regulation；And (e) non-transitory medium,
For recording at least one signal processing parameter θ ∈ Θ made by the user of audiphone regulation；Wherein, the first acoustic environment
Detector is arranged to, also based on regulation, the first output is supplied into first processor.
Alternatively, regulation is induced based on Bayes's increment preference.
Alternatively, position detector includes gps receiver.
Alternatively, the first sound environment detector is configured for voice signal, the institute received by hearing aid device system
The geographical position of the hearing aid device system of determination and at least one parameter for being selected from by the following group formed determine hearing aid
The species of acoustic environment around device system：The speed of time, hearing aid device system in date, one day and by gps receiver
The signal intensity of the signal of reception.
Alternatively, hearing aid device system further comprises：Non-transitory medium, for the geography that will be determined by position detector
Position is together with the type records in the acoustic environment of the geographical location.
Alternatively, the first sound environment detector is configured to consider with a distance from from identified geographical position
The probability of the generation of the acoustic environment species of precedence record in threshold value, determine the species of acoustic environment.
Alternatively, hearing aid device system further comprises：Non-transitory medium, for storing specific sound environment species and phase
Answer geographic area.
Alternatively, position detector is arranged to automatically access the calendar system of user, to obtain on user's
The information of position, and in addition when position detector not can determine that the geographical position of hearing aid device system, based on user's
The information of position, determine the geographical position of hearing aid device system.
Alternatively, the first sound environment detector is arranged to when position detector detects user aboard,
First audiphone of hearing aid device system is automatically switched to offline mode.
Alternatively, the first audiphone includes：At least one orientation sensor, it is arranged to work as user by the first audiphone
When being worn over its desired operation position, there is provided on the information of the orientation of the head of user, and wherein, the first audiphone is configured to use
In the value of information selection first of the orientation based on the head on user.
Alternatively, position detector is a part for the first audiphone.
Alternatively, hearing aid device system further comprises：Handheld device, it is communicably coupled with the first audiphone, the hand
Holding equipment accommodated position detector.
Alternatively, handheld device also accommodates the first sound environment detector.
Alternatively, handheld device includes user interface.
Alternatively, the first audiphone accommodates the first sound environment detector.
Alternatively, hearing aid device system further comprises：Second audiphone, there is second microphone, in response to
The voice signal received at two microphones, there is provided the second audio input signal, second processor, be configured to according to signal transacting
Algorithm F (Θ) handles the second audio input signal, to generate the second hearing compensation audio signal, and the second output transducing
Device, for based on the second hearing compensation audio signal, there is provided the second sound sense signal；Wherein, the first acoustic environment is examined
The geographical position that device is configured for the first and second audio input signals and hearing aid device system is surveyed, determines acoustic environment
Species.
Alternatively, the first sound environment detector is configured to provide for the second output, with selection signal processing parameter
The second value of set.
Alternatively, the second audiphone includes：Second sound environmental detector, it is configured for the first and second audios
The geographical position of input signal and hearing aid device system, determines the species of acoustic environment, and based on being examined by second sound environment
The species that device determines is surveyed, the second output is supplied to second processor, with the set Θ of selection signal processing parameter second value.
Brief description of the drawings
Accompanying drawing shows the design and effectiveness of embodiment, wherein, similar component is referred to by common reference numerals.These figures are not
It is necessarily drawn to scale.In order to more fully understand how to obtain above and other advantage and target, will be presented on shown in accompanying drawing
Embodiment particularly explanation.These accompanying drawings only show exemplary embodiments, and so as to be not to be construed as limiting its scope.
Fig. 1 shows there is single audiphone and the new hearing aid device system of handheld device, and there is the single audiphone orientation to pass
Sensor, the handheld device have gps receiver, sound environment detector and user interface,
Fig. 2 shows there is single audiphone and the new hearing aid device system of handheld device, and there is the single audiphone orientation to pass
Sensor, sound environment detector, the handheld device have gps receiver and user interface,
Fig. 3 shows the new hearing aid device system with two audiphones and handheld device, and there is two audiphones orientation to pass
Sensor, sound environment detector, the handheld device have gps receiver and user interface, and
Fig. 4 shows the new hearing aid device system with two audiphones and handheld device, and there is two audiphones orientation to pass
Sensor, the handheld device have sound environment detector, gps receiver and user interface.
Embodiment
Various exemplary embodiment is described hereinafter with reference to accompanying drawing.It should be noted that accompanying drawing is not drawn to scale, and similar knot
Structure or the element of function are represented through accompanying drawing by like reference numerals.It should also be noted that accompanying drawing is intended merely to facilitate saying for embodiment
It is bright.They are not intended as the elaborating or as the scope to invention claimed of invention claimed
Limitation.In addition, illustrated embodiment need not have shown all aspects or advantage.With reference to specific embodiment describe aspect or
Advantage is not necessarily limited to the embodiment, and even if not being illustrated or not being expressly recited, can also it is in office what
Realized in his embodiment.
New hearing aid device system is described more fully hereinafter with reference to accompanying drawing, wherein, polytype new hearing aid device system is shown.
New hearing aid device system can be specific with the multi-form not shown in accompanying drawing, and should not be construed as being limited to explain herein
The embodiment and example stated.
Like reference numerals represent the similar component in accompanying drawing.
Fig. 1 schematically shows the new hearing aid device system 10 with single first audiphone 12 and handheld device 30, the list
Individual first audiphone 12 has orientation sensor 44, and the handheld device 30 has gps receiver 48, the and of sound environment detector 14
User interface 45.
First audiphone 12 can be to be configured to wear overhead any types, such as, BTE, RIE, ITE, ITC,
The audiphones such as CIC.
First audiphone 12 include being connected to corresponding A/D converter (not shown) first before wheat after microphone 16 and first
Gram wind 18, for the sound in response to being received at the microphone 16,18 in the acoustic environment around the user of hearing aid device system 10
Sound signal, there is provided respective digital input signal 20,22.Digital input signals 20,22 are input into hearing loss processor 24, its
It is configured to handle digital input signals 20,22 according to signal processing algorithm F (Θ), to generate hearing compensation output signal
26.Hearing compensation output signal 26 is routed to D/A converter (not shown) and output transducer 28, for hearing to be damaged
Lose compensation output signal 26 and be converted to sound sense signal.
New hearing aid device system 10 further comprises the handheld device 30 for accommodating sound environment detector 14, for example, intelligence electricity
Words, the species of the acoustic environment around user for determining hearing aid device system 10.Determine species based on by handheld device
The voice signal that microphone 32 picks up.Determination based on species, sound environment detector 14 are supplied to exporting 34 at audiphone
Device 24 is managed, for selecting (one or more) signal processing algorithm and (one or more) suitable for sorted acoustic environment
Parameter.
Be automatically switched to most suitable one for sorted acoustic environment so as to, hearing aid processor 24 or
Polyalgorithm, optimal sound quality and/or speech intelligibility are thus kept in muli-sounds environment.The signal of processor 24
Processing Algorithm can perform the noise decrease and dynamic range compression and some other signal processing tasks of diversified forms.
First sound environment detector 14 benefit from more than in the first audiphone 12 available resource and power supply it is usual
Available computing resource and power supply in handheld device 30.
Sound environment detector 14 includes feature extractor 36, for the spy for the voice signal for determining to receive from microphone 32
Property parameter.Parameter can be on signal power, modal data and other well-known features.
Sound environment detector 14 further comprises environment classifier 38, for based on the institute exported by feature extractor 36
The characterisitic parameter of determination, determine the species of acoustic environment.Sound classification is multiple environment species by environment classifier 38, such as,
Voice, unclean voice, restaurant's brouhaha, music, traffic noise etc..Classification processing can utilize simply closest search
Rope, neutral net, hidden Markov model system or another system for being capable of pattern-recognition.The output of environment classifier 38
Can be comprising the general of " hard " probability for determining or indicating to belong to the acoustic environment of corresponding species of a single environment species
The set of rate.Other outputs can also apply.
Sound environment detector 14 further comprises parameter mapping 40, is used for for providing from signal processing algorithm and parameter
F (Θ) available storehouse selection (one or more) signal processing algorithm and the output 34 of (one or more) parameter.Parameter maps
40 map to the output of environment classifier 38 set of the parameter θ ∈ Θ for hearing aid sound processor 20.Such ginseng
Several examples are：Whether the amount of noise decrease, the amount of the amount of gain and HF gains, control corresponding signal algorithm are selected for holding
Capable control parameter of algorithm, the corner frequency of wave filter and slope, the compression threshold of compressor algorithm and speed, adaptive feedback
The adaptation rate of cancellation algorithms and detectable signal characteristic etc..Other specification can be included.
Handheld device 30 includes the position detector 41 with gps receiver 42, and it is arranged to determine audiphone system
The geographical position of system 10.Shown handheld device 30 is the smart phone for also having mobile interface 48, and mobile interface 48 includes being used for
With the GSM- interfaces of the interconnection of mobile telephone network and the well-known WIFI interface 48 in field of mobile telephony.In the absence of
In the case of useful gps signal, the position of shown hearing aid device system 10 can be determined that the address of WIFI network, or be based on
The signal of well-known a variety of GSM- transmitter receipts is determined by triangulation from field of mobile telephony.
Shown sound environment detector 14 be arranged to by identified geographical position in corresponding geographical location
Acoustic environment determines type records together.Record can be spaced, and/or by specific between record at regular times
Geographic distance performs, and/or by specific events trigger, for example, the conversion of the species of acoustic environment, the change of signal transacting etc.,
Such as, the change of signal handler, the change of signal processing parameter etc..
When hearing aid device system 10 is located in the region in the geographical position of the record with particular kind of acoustic environment, sound
Sound environmental detector is arranged to increase probability of the current sound environment for the corresponding precedence record species of acoustic environment.
The user interface 45 of hearing aid device system 10 may be configured to distribute to some species of acoustic environment accordingly
Some geographic areas.
Shown sound environment detector 14 is additionally configured to for example access the calendar system of user by mobile interface 48,
To obtain the information of the whereabouts on user, for example, meeting room, office, buffet, restaurant, family etc., and with it is determined that
Include the information during species of acoustic environment.The information of calendar system from user can be replaced or supplemented on being connect by GPS
Receive the information in the geographical position that device determines.
For example, sound environment detector 14 can automatically switch (one or more) audiphone of hearing aid device system 10
To offline mode, i.e. when as indicated in the calendar system of user, user aboard when, (one or more) audiphone
(one or more) wireless device be closed.
Moreover, when user is in the building of such as high-rise, gps signal may be not present or too weak, make
Obtaining geographical position can not be determined by gps receiver.It is then possible to the letter using the whereabouts on user from calendar system
Breath, there is provided on the information in geographical position, or information from calendar system can supplement the information on geographical position, example
Such as, the instruction of particular conference room can provide the information on the floor in high-rise.Information on height is usual
It can not be obtained from gps receiver.
When gps receiver can not provide geographical position, sound environment detector 14 can automatically use and come from calendar
The information of system.In the case of the information on geographical position not from gps receiver and calendar system is available, sound
Environmental detector can be based on the voice signal received, in a conventional manner to sound environment classification；Or audiphone can be by
It is arranged to operate under the pattern for example previously selected by user during the adaptation phase or when the situation occurs.
Audiphone 12 includes one or more orientation sensors 44, such as, gyroscope, for example, MEMS gyroscope, inclination pass
Sensor, ball switch etc., the signal that output is used to determine the orientation of the head of the user of wear hearing aid is arranged to, for example,
One or more of the rotation of head yaw angle, the rotation of the head angle of pitch, the rotation of head roll angle or its combination, such as tilt, i.e. with head
The angular displacement of normal upright position, when user stands or sat down, for example, in resting position, the head for the people for standing or sitting down
Inclination is 0 °, and in resting position, the inclination of the head of the people to lie down is 90 °.
First processor 24 is configured for the first signal processing algorithm with account M8003 line processor 24, should
The defeated of the sound environment detector 14 of output signal 46 and first such as based on one or more orientation sensors 44 is oriented with account
Go out the determination of control signal 34.If lain down for example, user changes into position from standing to doze off, sound environment detector
14 can make the correspondingly changeover program of signal processor 24, for example, the first audiphone 12 can automatically be weakened.
Environment classifier 38 reflects statistical information from audio signal (such as, SNR, RMS) plus position data (coming from GPS)
Be mapped to environmental form, if such as ambiguous, in automobile, in cocktail party, in church.User may be to parameter value
Regulation dissatisfied, and that user interface execution signal processing parameter can be used is automatically selected, for example, user can be by signal
Another signal processing algorithm is changed into the current selection of Processing Algorithm, for example, user can be from directivity signal Processing Algorithm
Switch to omni-directional signal Processing Algorithm.
Sound environment detector is arranged to user's regulation of binding signal process parameter values over time.
Sound environment detector is arranged to the storehouse using signal processing algorithm F (Θ), automatically regulates audiphone system
At least one signal processing parameter θ ∈ Θ in system 10, wherein, Θ is algorithm parameter space, including control is used for the calculation of execution
The parameter of the selection of method, for example, noise suppression algorithm is selected for performing in noise circumstance, and it is not selected for
Performed in quiet environment.
Sound environment detector 14 is arranged to
The regulation made by the user of hearing aid device system by user interface 45 is recorded, and
Induced based on Bayes's increment preference, in response to the regulation recorded, change at least one signal processing parameter θ ∈
Θ's automatically adjusts so that when detecting same sound environment next time, performs amended automatically adjust.
Bayesian inference, which is related to collect, to be meaned to assume consistent or inconsistent evidence with given.The confidence level of hypothesis with
The accumulation of evidence and change.By enough evidences, it generally becomes very high or very low.
Shown hearing aid device system includes sound environment detector, and it is operated with response to around hearing aid device system 10
Acoustic environment, Regulate signal processing parameter θ ∈ Θ.
Environment classifier 38 regards U as inputs, and it is the vector of the correlated characteristic on acoustic environment, it may for example comprise xt
Short-term RMS and SNR estimation.U will also include GPS location.The output of environment classifier is represented by discrete type variable C.Example
Type include voice, noise, the voice in noise, in the car, in church, it is first-class in cocktail party.Environmental form passes through
Parameter mapping 40 is mapped to hearing aid parameter θ.
As described above, sometimes, during the operation of equipment, user is to received signal ytQuality be unsatisfied with, and from
And (one or more) that hearing aid device system is performed by user interface 45 is adjusted.Learning objective will by sound environment detector
Normal mode is slowly absorbed in model parameter θ.Finally, handling will cause the user for reducing quantity to manipulate.
Only when the knowledge of the preference on user is available, parameter renewal is performed.When user interface 45 is in the normal of equipment
When not being manipulated during operation, user may be satisfied with to the signal quality transmitted, but this is uncertain.After all, user
May there is no wearable device.However, when user starts to manipulate user interface, it is assumed that user is unsatisfied with the moment.User connects
Mouthful the manipulation stage start to be represented as disagree the moment.When user manipulates user interface, user may still search for more
Good regulation.It is next to learn the moment just after user stops manipulating user interface 45.Now, it is assumed that user looks for
Adjusted to satisfied；And this is referred to as agreeing to the moment.Disagree and agree to that the moment identifies for collecting passive and actively religion respectively
The situation of derivative evidence.
Method suitable for user preference is in the hearing aid device system 10 induced based on Bayes's increment preference, but its other party
Method is possible.Assuming that the moment is disagreed at k-th and agrees to that the adjustable signal processing parameter at the moment is separately arranged as
θkdAnd θkc.Further it is assumed that the output of the ambient sound grader during k-th of user's manipulation stage is in CkPlace generally remains permanent
It is fixed.
Obviously, in environmental condition CkUnder, user is compared to θkdIt is preferred that θkc(by (end user) decision-making dk=θkc>θkdTable
Show).Reach the set of all user's decision-makings of k-th of decision-making by Dk-1={ d1,d2,…,dk-1Represent.Bayesian updating scheme
It is used to absorb kth time observation.So that the parameter mapping 40 from type to hearing aid parameter is by probability function p (θ | C, ω) table
Show, wherein, ω represents the parameter for parameter mapping 40.
Then, the Bayesian updating mapped for the parameter of the manipulation based on k-th of user is given by
p(ω|C,Dk)=p (dk|C,ω)×p(ω|C,Dk-1)/p(C)
Chu and Gharamani (Preference Learning with Gaussian Processes, 22nd Int’
L conf on Machine Learning, 2005) in, the Bayesian updating equation is answered in detail, for based on Gauss
The parameter mapping 40 of processing.
Except sound environment detector 14 is from addition to the handheld device 30 in Fig. 1 is moved to Fig. 2 the first audiphone 12,
New hearing aid device system 10 shown in Fig. 2 is similar to Fig. 1 new hearing aid device system, and operates in the same manner.In this way,
Microphone output signal 20,22 can be directly connected to sound environment detector 14 so that acoustic environment can be based on by hearing aid
The signal that microphone in device receives is classified, without increasing data transportation requirements.
New hearing aid device system 10 shown in Fig. 3 is the ears hearing aid device system with two audiphones, the first hearing aid
Device 12A is used for the auris dextra of user, and the second audiphone 12B is used for the left ear of user, and handheld device 30 includes GPS receiver
Device 42 and mobile interface 48.
Except the Α of corresponding sound environmental detector 14,14 Β cooperate to provide two audiphones as explained further below
12 Α, signal processing algorithm in 12 Β collaboration selection outside, in the shown Α of first audiphone 12 and the second audiphone 12 Β
Each audiphone similar to shown in Fig. 2 and operate in a similar manner.
Each include ears sound ring in the Α of the first and second audiphone 12,12 Β of ears hearing aid device system 10
The Α of border detector 14,14 Β, the species of the acoustic environment around user for determining ears hearing aid device system 10.Species
It is determined that based on the Α of microphone 20,22 Α, 20 Β, 22 Β output signal.Determination based on species, ears are detected with acoustic environment
The Α of device 14,14 Β will export 34 Α, 34 Β are supplied to the corresponding Α of hearing aid processor 24,24 Β, be applied to sound ring with selection
The signal processing algorithm of the species in border.So as to which the ears Α of sound environment detector 14,14 Β are based on coming from two audiphones
The signal of (that is, two ear use), the species of acoustic environment is determined, thus the Α of hearing aid processor 24,24 Β automatically cooperate switching
To the most suitable algorithm for the species for being used for acoustic environment, protected from there through ears with hearing aid device system 10 in muli-sounds environment
Hold optimal sound quality and/or speech intelligibility.
Inputted except the first sound environment detector 14 only receives from an audiphone 12, and each ears acoustic environment
The Α of detector 14,14 Β are received outside the fact that input from both the Α of audiphone 12,12 Β, the ears sound shown in Fig. 3
The Α of environmental detector 14,14 Β are similar to the sound environment detector 14 shown in Fig. 2.So as in figure 3, in audiphone 12
Transmission signal between Α, 12 Β so that the algorithm that cooperation selection is performed by the Α of signal processor 24,24 Β.
In figure 3, the Α of the first audiphone 12 Α of environment classifier 14 output is transferred to the Β of the second audiphone 12, and
And the Β of the second audiphone 12 Β of environment classifier 14 output is transferred to the Α of the first audiphone 12.Then, first and second
The Α of audiphone 12,12 Β parameter maps 40 Α, 40 Β are based on two input operations of identical, is used for selection processor to produce
The Α of control signal 34,34 Β of algorithm, and because the Α of parameter mapping units 34,34 Β receive identical input, two audiphones
Algorithms selection cooperation in 12 Α, 12 Β.
Transmitted data rates are low, because being only used for the probability of the species of acoustic environment or the set of logical value must be
It is transmitted between the Α of audiphone 12,12 Β.Suitable high latency can be received.Extremely will be according to ginseng by the way that time constant is applied
The variables that the output of number mapping changes, can smoothly can the difference as caused by the stand-by period.As described above, in two audiphones
Signal transacting cooperation it is critically important.If however, allowing the transient period of several seconds, hearing aid device system can pass through only 3- per second
4 transmission operations.Thus, power consumption is kept low.
The Α of sound environment detector 14,14 Β by with reference in a manner of figure 1 and Fig. 2 identicals disclosed above, with reference to by new
What the handheld unit 30 of hearing aid device system 10 provided determines position.
Shown new ears are controlled with hearing aid device system 10 except sound environment detector 14 provides two in Fig. 4
34 Α, 34 Β are exported, one of them 34 Α is connected to the Α of the first audiphone 12, and another 34 Β therein is connected to second
Outside the fact that 12 Β of audiphone, by providing the sound environment detector being similar to shown in Fig. 1 and grasping in a similar manner
The single sound environment detector 14 made, obtain the collaborative signal processing in two Α of audiphone 12,12 Β.Shown acoustic environment
Detector 14 is contained in handheld device 30.
Each Α of audiphone 12,12 Β are similar to the audiphone 12 shown in Fig. 1 and operated in the same manner.
Although specific embodiment has shown and described, it will be understood that, they are not intended to limitation invention claimed,
And, can be so that numerous changes and modifications can be made in the case where not departing from the spirit and scope of invention claimed, this is right
It is obvious in those skilled in the art.So as to which specification and drawings are considered as schematically, rather than limitation
The meaning of property.Invention claimed is intended to covering change, modification and equivalent.
Claims (18)
1. a kind of hearing aid device system, including：
(a) the first audiphone, have
First microphone, first microphone are used for the sound in response to being received at first microphone in acoustic environment
Sound signal, there is provided the first audio input signal,
First processor, the first processor are configured to defeated according to signal processing algorithm F (Θ) processing first audio
Enter signal, to generate the first hearing compensation audio signal, wherein, Θ is the set of signal processing parameter, and
First output transducer, first output transducer are used to be based on the first hearing compensation audio signal, carried
For the first sound sense signal；
(b) position detector, the position detector are arranged to determine the geographical position of the hearing aid device system；
(c) the first sound environment detector, first sound environment detector are arranged to
Based on the voice signal received by the hearing aid device system and the geographical position of the identified hearing aid device system, it is determined that
The species of the acoustic environment around the hearing aid device system, and
Based on the species of the acoustic environment determined by first sound environment detector, the first output is supplied to described
First processor, to select the set Θ of the signal processing parameter the first value；
(d) user interface, the user interface are used to allow the user of the hearing aid device system to make at least one signal transacting
Parameter θ ∈ Θ regulation；And
Wherein, first sound environment detector is arranged to also be based on the regulation, and the described first output is supplied to
The first processor, and first sound environment detector are arranged to the regulation in response to being recorded, and change institute
State automatically adjusting at least one signal processing parameter θ ∈ Θ so that when detecting same sound environment next time, perform modification
Automatically adjusting afterwards.
2. hearing aid device system according to claim 1, wherein, Bayes's increment preference induction based on the regulation will
First output is supplied to the first processor.
3. hearing aid device system according to claim 1, wherein, the position detector includes gps receiver.
4. hearing aid device system according to claim 3, wherein, first sound environment detector is configured for
The voice signal that is received by the hearing aid device system, the geographical position of the identified hearing aid device system and from by
At least one parameter selected in the group formed below determines the species of the acoustic environment around the hearing aid device system：
The speed of time, the hearing aid device system in date, one day and the signal of the signal received by the gps receiver are strong
Degree.
5. hearing aid device system according to claim 1, further comprises：Non-transitory medium, for recording by institute's rheme
Put species of the geographical position together with the acoustic environment in the geographical location of detector determination.
6. hearing aid device system according to claim 5, wherein, first sound environment detector is configured to
The probability that the acoustic environment species of precedence record occurs in the distance threshold from identified geographical position is considered, it is determined that described
The species of acoustic environment.
7. hearing aid device system according to claim 1, further comprises：Non-transitory medium, for storing the sound
Some species of environment and corresponding geographic area.
8. hearing aid device system according to claim 1, wherein, the position detector is arranged to automatically access institute
The calendar system of user is stated, to obtain the information of the position on the user, and in addition when the position detector can not
When determining the geographical position of the hearing aid device system, based on the information of the position on the user, it is determined that described
The geographical position of hearing aid device system.
9. hearing aid device system according to claim 1, wherein, first sound environment detector is arranged to work as institute
When stating position detector and detecting the user aboard, first audiphone of the hearing aid device system is automatically cut
Shift to offline mode.
10. hearing aid device system according to claim 1, wherein, first audiphone includes：At least one orientation sensing
Device, at least one orientation sensor are arranged to when first audiphone is worn over its desired operation position by the user
When putting, there is provided on the information of the orientation of the head of the user, and wherein, first audiphone is configured for closing
In the information of the orientation of the head of the user, first value is selected.
11. hearing aid device system according to claim 1, wherein, the position detector is the one of first audiphone
Part.
12. hearing aid device system according to claim 1, further comprises：Handheld device, the handheld device and described the
One audiphone is communicably coupled, and the handheld device accommodates the position detector.
13. hearing aid device system according to claim 12, wherein, the handheld device also accommodates first acoustic environment
Detector.
14. hearing aid device system according to claim 12, wherein, the handheld device includes the user interface.
15. hearing aid device system according to claim 1, wherein, first audiphone accommodates first acoustic environment
Detector.
16. hearing aid device system according to claim 1, further comprises：
Second audiphone, has
Second microphone, the second microphone are used for the voice signal in response to being received at the second microphone, there is provided
Second audio input signal,
Second processor, the second processor are configured to defeated according to signal processing algorithm F (Θ) processing second audio
Enter signal, to generate the second hearing compensation audio signal, and
Second output transducer, second output transducer are used to provide based on the second hearing compensation audio signal
Second sound sense signal；
Wherein, first sound environment detector is configured for first audio input signal and the second audio is defeated
Enter the geographical position of signal and the hearing aid device system, determine the species of the acoustic environment.
17. hearing aid device system according to claim 16, wherein, first sound environment detector is arranged to carry
For the second output, to select the second value of the set of the signal processing parameter.
18. hearing aid device system according to claim 16, wherein, second audiphone includes：
Second sound environmental detector, the second sound environmental detector are arranged to
The geographical position based on first audio input signal and the second audio input signal and the hearing aid device system
Put, determine the species of the acoustic environment, and
Based on the species determined by the second sound environmental detector, the second output is supplied to the second processing
Device, to select the set Θ of signal processing parameter second value.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| DKPA201370770 | 2013-12-13 | ||
| DKPA201370770 | 2013-12-13 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104717593A CN104717593A (en) | 2015-06-17 |
| CN104717593B true CN104717593B (en) | 2018-03-16 |
Family
ID=53416463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410777867.8A Expired - Fee Related CN104717593B (en) | 2013-12-13 | 2014-12-15 | Position learns audiphone |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP6190351B2 (en) |
| CN (1) | CN104717593B (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101858209B1 (en) | 2014-07-08 | 2018-05-16 | 와이덱스 에이/에스 | Method of optimizing parameters in a hearing aid system and a hearing aid system |
| EP3107314A1 (en) * | 2015-06-19 | 2016-12-21 | GN Resound A/S | Performance based in situ optimization of hearing aids |
| EP3751451A1 (en) * | 2015-08-28 | 2020-12-16 | Sony Corporation | Information processing apparatus, information processing method, and program |
| EP3267695B1 (en) * | 2016-07-04 | 2018-10-31 | GN Hearing A/S | Automated scanning for hearing aid parameters |
| EP3280157B1 (en) * | 2016-08-04 | 2021-04-07 | GN Hearing A/S | Hearing device for receiving location information from wireless network |
| JP6973400B2 (en) * | 2016-09-27 | 2021-11-24 | ソニーグループ株式会社 | Information processing equipment, information processing methods, and programs |
| JP6980695B2 (en) | 2016-12-06 | 2021-12-15 | ヤマハ株式会社 | Specific sound filter processing device and specific sound filter processing method |
| DE102017200599A1 (en) * | 2017-01-16 | 2018-07-19 | Sivantos Pte. Ltd. | Method for operating a hearing aid and hearing aid |
| KR20180125384A (en) * | 2017-05-15 | 2018-11-23 | 한국전기연구원 | Hearing Aid Having Voice Activity Detector and Method thereof |
| KR20180125385A (en) * | 2017-05-15 | 2018-11-23 | 한국전기연구원 | Hearing Aid Having Noise Environment Classification and Reduction Function and Method thereof |
| EP3729828A1 (en) * | 2017-12-20 | 2020-10-28 | Sonova AG | Intelligent,online hearing device performance management |
| EP3621315A1 (en) * | 2018-09-07 | 2020-03-11 | GN Hearing A/S | Methods for operating hearing device processing based on environment and related hearing devices |
| CN111163411B (en) * | 2018-11-08 | 2022-11-18 | 达发科技股份有限公司 | Method for reducing influence of interference sound and sound playing device |
| US11190884B2 (en) * | 2019-06-20 | 2021-11-30 | Samsung Electro-Mechanics Co., Ltd. | Terminal with hearing aid setting, and method of setting hearing aid |
| US11076243B2 (en) * | 2019-06-20 | 2021-07-27 | Samsung Electro-Mechanics Co., Ltd. | Terminal with hearing aid setting, and setting method for hearing aid |
| DE102019213809B3 (en) * | 2019-09-11 | 2020-11-26 | Sivantos Pte. Ltd. | Method for operating a hearing aid and hearing aid |
| DE102019216100A1 (en) * | 2019-10-18 | 2021-04-22 | Sivantos Pte. Ltd. | Method for operating a hearing aid and hearing aid |
| JPWO2021144964A1 (en) * | 2020-01-17 | 2021-07-22 | ||
| DE102020201608B4 (en) * | 2020-02-10 | 2022-10-06 | Sivantos Pte. Ltd. | Method for automatically adjusting a parameter of a signal processing system of a hearing aid |
| DE102020209050B4 (en) | 2020-07-20 | 2022-05-25 | Sivantos Pte. Ltd. | Method for operating a hearing system, hearing system, hearing aid |
| US11689868B2 (en) * | 2021-04-26 | 2023-06-27 | Mun Hoong Leong | Machine learning based hearing assistance system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1813491A (en) * | 2003-06-24 | 2006-08-02 | Gn瑞声达A/S | A binaural hearing aid system with coordinated sound processing |
| CN1822723A (en) * | 2005-02-14 | 2006-08-23 | 西门子测听技术有限责任公司 | Method for adjusting a hearing aid, hearing aid and mobile control unit for the adjustment of a hearing aid |
| CN103222283A (en) * | 2010-11-19 | 2013-07-24 | Jacoti有限公司 | Personal communication device with hearing support and method for providing the same |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS55171199U (en) * | 1979-05-25 | 1980-12-08 | ||
| JP2006254012A (en) * | 2005-03-10 | 2006-09-21 | Matsushita Electric Ind Co Ltd | Volume control system |
| DE602006014572D1 (en) * | 2005-10-14 | 2010-07-08 | Gn Resound As | OPTIMIZATION FOR HEARING EQUIPMENT PARAMETERS |
| WO2009001559A1 (en) * | 2007-06-28 | 2008-12-31 | Panasonic Corporation | Environment adaptive type hearing aid |
| US8611570B2 (en) * | 2010-05-25 | 2013-12-17 | Audiotoniq, Inc. | Data storage system, hearing aid, and method of selectively applying sound filters |
-
2014
- 2014-12-12 JP JP2014252336A patent/JP6190351B2/en not_active Expired - Fee Related
- 2014-12-15 CN CN201410777867.8A patent/CN104717593B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1813491A (en) * | 2003-06-24 | 2006-08-02 | Gn瑞声达A/S | A binaural hearing aid system with coordinated sound processing |
| CN1822723A (en) * | 2005-02-14 | 2006-08-23 | 西门子测听技术有限责任公司 | Method for adjusting a hearing aid, hearing aid and mobile control unit for the adjustment of a hearing aid |
| CN103222283A (en) * | 2010-11-19 | 2013-07-24 | Jacoti有限公司 | Personal communication device with hearing support and method for providing the same |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104717593A (en) | 2015-06-17 |
| JP2015130659A (en) | 2015-07-16 |
| JP6190351B2 (en) | 2017-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104717593B (en) | Position learns audiphone | |
| US9648430B2 (en) | Learning hearing aid | |
| US11501772B2 (en) | Context aware hearing optimization engine | |
| US9094769B2 (en) | Hearing aid operating in dependence of position | |
| EP2884766B1 (en) | A location learning hearing aid | |
| US11290826B2 (en) | Separating and recombining audio for intelligibility and comfort | |
| US10154357B2 (en) | Performance based in situ optimization of hearing aids | |
| US20230056617A1 (en) | Hearing device comprising a detector and a trained neural network | |
| CN107580288A (en) | automatically scanning for hearing aid parameter | |
| US8781142B2 (en) | Selective acoustic enhancement of ambient sound | |
| US9992585B1 (en) | Hearing assistance system incorporating directional microphone customization | |
| Aldaz et al. | Smartphone-based system for learning and inferring hearing aid settings | |
| CN108600907A (en) | Method, hearing devices and the hearing system of localization of sound source | |
| CN106257936B (en) | In-situ fitting system for a hearing aid and hearing aid system | |
| EP3900399B1 (en) | Source separation in hearing devices and related methods | |
| EP2819436B1 (en) | A hearing aid operating in dependence of position | |
| US20230083358A1 (en) | Earphone smartcase with audio processor | |
| US20220377469A1 (en) | System of processing devices to perform an algorithm | |
| DK201370356A1 (en) | A hearing aid operating in dependence of position |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180316 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |