Classifications

• • H—ELECTRICITY
  • H04—ELECTRIC COMMUNICATION TECHNIQUE
  • H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
  • H04R25/00—Deaf-aid sets, i.e. electro-acoustic or electro-mechanical hearing aids; Electric tinnitus maskers providing an auditory perception
  • H04R25/50—Customised settings for obtaining desired overall acoustical characteristics
  • H04R25/505—Customised settings for obtaining desired overall acoustical characteristics using digital signal processing
  • H04R25/507—Customised settings for obtaining desired overall acoustical characteristics using digital signal processing implemented by neural network or fuzzy logic

Definitions

• the invention relates to a method for adapting, in particular fine-tuning, hearing aids, in which parameters of the parameters set on the hearing aid are first evaluated in an evaluation step, in particular by means of psychoacoustic variables, and parameters in need of improvement are adjusted in a subsequent optimization step.
• An object of the invention is to develop a method of the type mentioned at the outset such that the individual adaptation of hearing aids is further improved and their acceptance is increased.
• this object is achieved in that the degree of optimization to be assessed or variables relevant for this within the scope of the assessment step and / or for the
• the optimization step determines the degree of adjustment of the parameter in need of improvement by algorithms or rule sets based on fuzzy logic.
• the method according to the invention is generally carried out following a presetting, the presetting preferably being carried out as a function of the data of the particular hearing impairment, the hearing device selected and / or the hearing environments deemed relevant by the hearing impaired.
• the parameter values to be selected as part of the presetting can be determined, for example, by means of a defined algorithm.
• the method according to the invention can then be carried out for the purpose of fine adjustment, which essentially consists of at least one evaluation step and a subsequent optimization step.
• the current hearing aid setting is assessed by the hearing impaired with regard to different criteria.
• sound images are preferably presented to him acoustically, the sound impression produced individually in the hearing impaired person being evaluated accordingly.
• the required degree of further adjustment or adjustment of parameters in need of improvement is calculated as part of the optimization step.
• an evaluation step is carried out again, in order to achieve an iterative improvement in the degree of optimization of one or more parameters set on the hearing aid, it is useful to repeat the steps described several times, in particular until a certain termination criterion is met is.
• the essential basic idea of the invention is that the evaluation result to be achieved in the context of the evaluation step and / or the degree of adjustment to be determined in the context of the optimization step is calculated on the basis of fuzzy rule sets.
• fuzzy rule sets either only in the calculations required for the evaluation step or only in the calculations required for the optimization step.
• fuzzy rule sets are used both for the evaluation and for the optimization step, since in this way an optimal adaptation of the hearing aid can be achieved.
• fuzzy rule sets makes it possible to process fuzzy information, which is often only available in linguistic form, in a predetermined manner and to standardize and systematize the procedure for adapting hearing aids.
• fuzzy information which is often only available in linguistic form
• fuzzy rule sets that arise and are not clear during an individual fitting process can be taken into account, but also experiences of hearing aid acousticians that are available in linguistic form, so that these Experiences are generally available and automatically taken into account.
• fuzzy input data for the algorithms to be used in the context of the adaptation e.g. description of sound image properties, evaluation of sound images by normal hearing, evaluation of sound images by the hearing impaired, data of hearing impairment, etc. .
• fuzzy input data for the algorithms to be used in the context of the adaptation e.g. description of sound image properties, evaluation of sound images by normal hearing, evaluation of sound images by the hearing impaired, data of hearing impairment, etc. .
• the hearing impaired person assesses the subjective sound impression of various acoustically presented sound images.
• natural sound images corresponding to everyday listening situations are taken into account, with the hearing impaired preferably being offered the listening environments living area, leisure and culture, traffic, work, nature, speakers in undisturbed surroundings and / or speakers in disturbed surroundings.
• the hearing impaired person makes a selection from the sound images or sound image groups offered that corresponds to his or her individual living space.
• the adaptation method according to the invention consisting of evaluation and optimization steps, is then carried out separately for all selected sound images or sound image groups.
• the following assessment features can be taken into account, for example, with regard to the individual sound impression: quality, intelligibility, comfort, naturalness, volume, timbre, noise and / or distortions.
• the evaluation of the evaluation characteristics takes place, for example, by assigning the perceived sound impression to a standardized, fuzzy and only linguistically defined evaluation scale.
• this scale could include the levels extremely bad, very bad, bad, medium, good, very good and extremely good when evaluating quality.
• rating scales preferably comprise approximately 7 levels, although other values can also be realized here.
• the evaluation characteristic "quality" can be calculated from the assessments of speech intelligibility and / or comfort and / or naturalness in the form of fuzzy variables using a fuzzy rule set.
• the assessment of sound images that do not contain speech signals only the assessment of the comfort and naturalness is preferably included in the calculation of the quality. If the sound image additionally or exclusively contains speech signals, the assessment of speech intelligibility should also be included in addition to the quantities mentioned.
• the mentioned characteristic of "naturalness” does not have to be assessed directly by the hearing impaired, but can be calculated from the assessments available in the form of fuzzy variables, for example volume, timbre, noise and distortion, using a fuzzy rule set.
• the result of the fuzzy rule sets for calculating the quality and the naturalness, as a result in turn provide fuzzy variables which can be used, for example, as input variables of the fuzzy rule sets of the optimization step.
• the evaluation step can thus serve to provide information about the degree of optimization of the hearing aid setting that has already been achieved and, on the other hand, to provide knowledge about which transmission properties of the hearing aid or which evaluation characteristics still need to be improved in a subsequent optimization step.
• the degree of adjustment of one or more adjustable parameters of the hearing device is then calculated as part of the optimization step to achieve an improved transmission behavior of the hearing device.
• Input variables of the rule sets of the optimization step that are not available as fuzzy variables, such as physical data of sound images or setting ranges of hearing aid parameters, are fuzzified before the rule sets are used, in order to also enable this sharp data to be processed. This applies in the same way to the corresponding input data of the evaluation step.
• the rule sets of the optimization step which process one or more of the variables mentioned above are preferably composed of several partial rule sets.
• a partial rule set “target formulation” is used at the beginning of the optimization step, which determines at least one assessment feature based on the hearing impairment assessment previously performed and a normal hearing assessment stored in the system, which needs to be improved in terms of its degree of optimization.
• a subsequently applied partial rule set "Basic rules" calculates the degree of adjustment to be determined in a first approximation, taking into account the input variables "Result of the partial rule set target formulation” and "Result of the assessment step (s) carried out by the hearing impaired".
• Results of the partial rule set target formulation and "Result of the assessment step (s) carried out by the hearing impaired.
• Results of the partial rule set target formulation and "Result of the assessment step (s) carried out by the hearing impaired.
• the degree of adjustment to be determined can be calculated using a partial “hearing damage” rule set, in which case the input variables “result (s) of already applied partial rule sets” and “data of the hearing damage” must be taken into account.
• the degree of adjustment to be determined can be calculated in a further approximation by means of a partial rule set "hearing aid properties", the input variables of which are particularly the “result (s) of partial rule sets already used", the "data of the transmission behavior of the hearing aid present before the optimization step” and the "type of parameters that can be set on the hearing aid and their setting ranges” are formed.
• This partial set of rules ensures that the degree of adjustment calculated by the optimization step does not require any settings of the hearing device, such as excessive amplifications, that cannot be implemented.
• the partial rule sets described can be used one after the other, subsequent partial rule sets always taking into account the results of one or more previously applied partial rule sets, or after the partial rule sets have been applied, their partial results are accumulated and compressed by means of suitable algorithms, so that ultimately receives information about how certain parameters of the hearing aid have to be adjusted or readjusted.
• the hearing aid can then be adjusted either automatically by the system or manually.
• defuzzification is recommended since the system can then make the adjustment based on the defuzzified, sharp value.
• the method according to the invention can be used in any modification, it being only necessary to use fuzzy rule sets as part of the evaluation and / or optimization step in order to achieve the improvement of the known matching methods intended with the invention.

Landscapes

• Physics & Mathematics (AREA)
• Engineering & Computer Science (AREA)
• Health & Medical Sciences (AREA)
• General Health & Medical Sciences (AREA)
• Evolutionary Computation (AREA)
• Fuzzy Systems (AREA)
• Mathematical Physics (AREA)
• Software Systems (AREA)
• Artificial Intelligence (AREA)
• Automation & Control Theory (AREA)
• Neurosurgery (AREA)
• Otolaryngology (AREA)
• Acoustics & Sound (AREA)
• Signal Processing (AREA)
• Management, Administration, Business Operations System, And Electronic Commerce (AREA)
• Combined Controls Of Internal Combustion Engines (AREA)

Applications Claiming Priority (2)

Publications (2)

Family

ID=7772710

Family Applications (1)

Country Status (2)

Families Citing this family (5)

Family Cites Families (5)

• 1995
  • 1995-09-20 DE DE1995134981 patent/DE19534981A1/de not_active Ceased
• 1996
  • 1996-07-25 EP EP96112021A patent/EP0765103A3/fr not_active Withdrawn

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events