AU2004240216B2 - Programmable Auditory Prosthesis with Trainable Automatic Adaptation to Acoustic Conditions - Google Patents

Description

21/12 '04 11:53 FAX 81 3 9663 3099

O

0 ci

C)

C)

0 c-i (Nq FB RICE CO.

1 1004

AUSTRALIA

Patents Act 1990 HEARWORKS PTY LTD COMPLETE SPECIFICATION STANDARD PATENT Invention Title: Programmable Auditory Prosthesis with Trainable Automatic Adaptation to Acoustic Conditions The following statement is a full description of this Invention including the best method of performing it known to us:- COMS ID No. SBMI-01048750 Received by !P Austraa: Time 12:08 Date 2004-12-21 21/12 '04 11:54 FAX 61 3 9663 3099 FB RICE CO. 005 0 2 ci

C-)

0L programmable Auditory Prosthesis with Trainable Automatic Adaptation to Acoustic Conditions c-i Technical Field IND 5 The present invention relates to a programmable auditory prosthesis, such as a ci o hearing aid or cochlear implant.

t- H- Backgroand Art Cl Hearing loss, which may be due to many different causes, is generally of two types, conductive and sensorineural. Of these types, conductive hearing loss occurs where the normal mechanical pathways for sound to reach the hair cells in the cochlea are impeded, for example, by damage to the ossicles. Conductive hearing loss may often be helped by use of conventional hearing aid systems, which comprise a microphone, an amplifier and a receiver (miniature speaker) for amplifying detected sounds so that acoustic information does reach the cochlea and the hair cells. Since the elevation of the minimum detectable sound pressure level may vary with the frequency of an acoustic test stimulus, the amplifier may be preceded by, or comprise of, a bank of filters to enable different frequency components of the signal to be amplified by different amounts.

Sensorineural hearing loss occurs whore the hair cells in the cochlea and the attached auditory nerve fibres are damaged or destroyed. Sensorineural hearing loss results in an increase in the minimum detectable sound pressure level, which often varies with the frequency of the test stimulus. However, in contrast to conductive hearing loss, the sound pressure level that is uncomfortably loud at a given test frequency is often approximately the same as for people with normal hearing. The result is a reduction in the dynamic range of sound pressure levels that are audible yet not uncomfortably loud with the impaired car, and this dynamic range of the impaired ear may vary considerably with the frequency of the acoustic test stimulus. For this reason, sensorineural hearing loss is often treated with hearing aid systems that employ non-linear amplification to compress the dynamic range of common sounds towards the dynamic range of the impaired ear. Such systems may use a filter bank that is followed by a bank of compressive amplifiers, so that the dynamic range of the signal is reduced COMS ID No: SSMI-01048750 R.c'cd by IP Australia: Tim (Hm) 12:0 Dato (Y 2004-12-21 21/12 '04 11:54 FAX 61 3 9663 3099 FB RICE CO. 00_ o 3 C-) Sby an amount that is considered appropriate for the dynamic range of the impaired ear in each band.

c-i In many people who are profoundly deaf, the reason for deafness is absence of or destruction of, the hair cells in the cochlea which transduce acoustic signals into nerve impulses. These people are thus unable to derive suitable benefit from hearing o aid systems, no matter how much the acoustic stimulus is amplified, because there is damage to or absence of the mechanism for nerve impulses to be generated from sound in the normal manner. It is for this purpose that cochlear implant systems have been 010 developed. Such systems bypass the hair cells in the cochlea and directly deliver electrical stimulation to the auditory nerve fibres, thereby allowing the brain to perceive a hearing sensation resembling the natural hearing sensation normally delivered to the auditory nerve. US Patent 4,532,930, the contents of which are incorporated herein by reference, provides a description of one type of traditional cochlear implant system.

Cochlear implant systems have typically consisted of two essential components, an external component commonly referred to as a processor unit and an internal implanted component commonly referred to as a stimulator/receiver unit Traditionally, both of these components have cooperated together to provide the sound sensation to a user.

The external component has traditionally consisted of a microphone for detecting sounds, such as speech and environmental sounds, a speech processor that converts the detected sounds, particularly speech, into a coded signal, a power source such as a battery, and an external transmitter coil.

The coded signal output by the speech processor is transmitted transcutaneously to the implanted stimulator/receiver unit situated within a recess of the temporal bone of the user. This transcutaneous transmission occurs via the external transmitter coil which is positioned to communicate with an implanted receiver coil provided with the stimulator/receiver unit. This communication serves two essential purposes, firstly to transcutaneously transmit the coded sound signal and secondly to provide power to the implanted stimulator/receiver unit. Conventionally, this link has been in the form of an RF link, but other such links have been proposed and implemented with varying degrees of success.

COMS rD No. SBM1-01043752 Received by P Australia: Time 12:CS Date 2004-1221 21/12 '04 11:54 FAI 61 3 9663 3099 FB RICE CO. 16007 0 o4 The implanted stimulator/rcceiver unit traditionally includes a receiver coil that 0receives the coded signal and power for the external processor component, and a cstimulator that processes the coded signal and outputs a stimulation signal to an intracochlea electrode assembly which applies the electrical stimulation directly to the auditory nerve producing a hearing sensation corresponding to the original detected sound.

Different users of auditory prostheses require differing outputs from their prosthesis to suit their individual requirements- This is the case even when individual to users may clinically be regarded as having identical hearing loss profiles, are utilising 0- identical prostheses, and when exposed to essentially identical acoustic environments.

Because of this, sound processing schemes for hearing aids and cochlear implants typically contain a number of parameters for which the values can be adjusted to suit the requirements of individual users. Examples of such parameters include the sensitivity to incoming sounds and the variation of the frequency response. Typically, the parameter values are selected either by the prosthesis user in everyday situations (eg. the sensitivity (volume), frequency response), or by the clinician at the time the prosthesis is fitted (eg. the baseline frequency rcsponse and the rate at which the frequency response and sensitivity vary as the input level varies).

In more recent times, there has been a trend to provide auditory prostheses with an increasing number of adjustable parameters that can or must be adjusted to optimise performance. This increase has, however, highlighted the problem that there does not always exist a reliable prescriptive method for selecting the optimum values for the individual user, particularly as some optimum values may vary among individuals who have hearing loss profiles that may clinically be regarded as identical. It is accordingly often necessary for the clinician to make adjustments to the prosthesis based on the user's reported experiences away from the clinic, and the need to return to the clinic for these adjustments can be time consuming and inefficient.

One example of a hearing aid that can receive the impressions of a hearing aid user and take these into consideration during operation is described in US Patent 5604812. This document describes a hearing aid that has a memory that can store socalled "unsharp inputs", or impressions of the hearing aid wearer about prevailing ambient conditions and/or the volume of the output of the hearing aid, prescribable algorithms, hearing loss data and hearing aid characteristic data. A fuzzy logic module COMS I DNo SBMI-D01750 Received byP Australi'a: "mnc (F m) 12:08 Date -V d) 2004-12-21 21/12 '04 11:55 FAI 61 3 9663 3099 FB RICE CO. _008 0 o tuses this data and control signals from an input signal analysis unit to calculate the 0 output setting parameters for the hearing aid. The behaviour rules of the fuzzy logic ci module and/or the prescribable algorithms are calculated from the data stored in the aid's memory by a neural network, with the neural network being typically implemented on a personal computer that is connected to the aid.. A problem with such San aid is the complexity of the processing scheme and an undesirably large drain on the aid's power source, typically an on-board battery. Further to this, such an aid would require a large amount of clinical time to correctly "train", as the user does not have direct control over what is optimal, and "unsharp inputs" are used in the training of the o 10 aid rather than precise and direct inputs.

Another example of a hearing aid that can receive the preferences of a hearing aid user and take these into consideration is described in US Patent 6035050. Such an aid requires the user to identify their current listening environment by selecting one of a limited number of listening situations on a remote unit- Should the user find that their current listening situation does not readily fall within those provided for selection, the benefit of such a aid becomes greatly reduced. Further to this, some listening situations on the remote unit, such as "at work", may consist of a range of different acoustic environments and hence may not be acoustically well-defined. Thus, a neural network may not be able to reliably recognise such listening situations flom an analysis of the microphone signal that, due to practical considerations such as memory limitations and the desirability of fast recognition of a change in listening situation, must be limited to a period of seconds or a few minutes. Therefore, after training it is possible that the neural network may misclassify a listening situation, especially in situations that are not acoustically well-defined, which can result in the application of amplification parameters that are unsuitable for the current listening situation. In such a device focus is placed on attempting to categorise specific acoustic environments, rather than measure the parameters of the environment and use these parameters directly in the processing scheme, as is the case with the present invention. Another disadvantage of this device is that a major lifestyle change, such as a new workplace that is acoustically different to the previous workplace, may require different amplification parameters and hence a return visit to the clinic.

Another example of a hearing aid that can receive the preferences of a hearing aid user and take these into consideration is described in US Patent 6044163. This document describes a hearing aid that is similar to the hearing aid described in US COMS ID No. SDBMIF 04750 Rccived by IP Australia: Time 12.S Dat~ (Y-M d) 2004-12-27 23/12 '08 13:56 FAX 61386184199 FB RICE CO ~0 00 0 6 U patent 6035050. A major difference is that the neural network is not restricted to Selection Of sets of amplification parameters that are stored in a memory, but may 0 directly set the value of individual amplification parameters. The disadvantages of the en hearing aid described in this Patent are similar to those of the hearing aid described in US patent 6035050.

Va Any discussion of documents, acts, materials, devices, articles or the like which 0 has been included in the present specification is solely for the purpose of providing a context for the present invention. It is -not to be taken as an admission that any or all of these matters form part of the prior art base or were common general knowledge in the field relevant to the present invention as it existed before the priority date of each claim ci of this application.

Throughout this specification the word "comprise", or variations such as "Comprises" or "comprising', will be understood to imply the inclusion of a stated element, integer or step, or group Of elements, integers or steps, but not the exclusion of any other element, integer or step, or group of elements, integers or steps.

Sumnmars The Present invention is adapted to providing users with an auditory prosthesis that is adaptable and can adjust its output to the user when that user is exposed to varying ambient acoustic environments.

Further, the Present invention is adapted to providing users with an auditory Prosthesis that can be 'trained' by the individual user to adapt its output to the user's preference in listening conditions encountered away fTon a traditional clinical situation, thereby reducing the clinical time required to fit an optimise a prosthesis to the individual user.

According to a first aspect of the invention, there is provided an auditory prosthesis for rehabilitating the hearing of a user, comprising: a microphone configured to receive sound and to produce a microphone signal corresponding to the received sound; and a sound processor configured to process the microphone signal using a set of variable processing factors to produce an output signal, wherein the prosthesis is COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 '08 13:57 FAX 61386184199 FB RICE

CO

0I010 00 07 U Configured to provide the user with the ability to adjust at least one of the variable Processing factors, and wherein one or more of the variable processing factors are 0 automatically adjusted by the prosthesis based on the adjustmients of the at least one variable processing factor by the user.

In an embodiment of the invention, there is disclosed an auditory prosthesis o ~comfprising: a microphone for receiving soun-d and producing a microphone signal corresponding to the received soundan output device for providing audio signals in a form receivable by a user of the prosthesis; sound processing means operable to receive the microphone signal and carry out processing operations on the microphone signal to produce an output signal int a form suitable to operate the output device, and user control means; wherein the processing means is operable in a first mode in which the processing operationi comprises at least one variable processing factor which is adjustable and/or selectable by a user using the user control means, to a setting which causes the output signal of the sound processing means to be adjusted according to the preference of the user in the current acoustic environment; and a data memory means for storing said setting; wherein the sound processing means is simultaneously operable in a second mode in which at least one of the at least one variable processing factor(s) is automatically adjusted on the basis of one or mnore previously adjusted and/or selected settings Preferably, the at least one variable processing factor(s) are calculated by defined mathematical relationships incorporating coefficients, and wherein the value of the coefficients are calculated to be those values that cause the variable processing factor(s) to best approximate the variable processing factor(s) that occurred as a result of current and previous settings.

Preferably, the user control means is adjustable throughout a substantially continuous range of settings, and wherein a preferred setting is used to calculate the at COMS ID No: ARCS-21823s Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 '08 13:57 FAX 61386184199 F IE&C FB RICE CO Q]thoIi 00 o 7a o least one variable processing factor(s), which alter the processing operations applied to the microphone signal, ci In this invention, it will be appreciated that what is considered an optimal signal by one user may not be considered an optimal signal by another user. The INO present invention ensures that the prosthesis operates in a manner that suits the requirements of the user for the acoustic environments that the user encounters rather o than in some manner predetermined from clinical tests which may not replicate the type of acoustic envirormtents to which the user is routinely or occasionally exposed, and which may not necessarily replicate the subjective reactions of users when in those o environments.

Preferably, the data memory means; stores data sets representative of the characteristics of at least one acoustic environment. Preferably the at least one variable processing factor is automatically adjusted on the basis of the acoustic environments in which at least one variable processing factor was selected.

In a first embodiment the processing means may be adapted when operable in the first mode to offer two or more possible optimal settings of the variable processing COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 21/12 '04 11:56 F.AX 61 3 9663 3099 FB RICE CO, [Mi1 o8

C-)

O factor for selection by the user. In this case, the user is preferably able to compare the 0operation of the prosthesis whea operating in a particular acoustic environment when operating with each setting and then select from these the setting that is best for that particular environment. The setting of the variable processing factor that is selectcd by I the user may then be stored in the data memory means, optionally with data indicative of the characteristics of the particular acoustic environment. This process may be o repeatab]e so as to allow the processing means to monitor whether the user's preference for a particular setting of the variable processing factor changes with time or usage. In ci this way, the user effectively selects or votes for the best setting each time. By oo repeating the process, the number of selections or votes made by the user for each 0, setting may be monitored to allow the prosthesis to select the optimal setting of the variable processing factor or optionally, the optimal setting of the variable processing factor for a particular environment.

In this embodiment, the user may alternate between listening with the offered settings by operating a control means, and can select or vote for a setting of the variable processing factor(s) by operating an indicator means. In one embodiment, the control means may comprise a switch or set of buttons that is actuable by the user, and the indicator means may comprise a switch or button that is actuable by the user.

In another embodiment, the processing operation of the processing means may be adjustable by the user when the processing means is in the first mode of operation.

Rather than offering a discrete selection of possible optimal settings for selection by the user, in this embodiment, the user is able to adjust a control means throughout a substantially continuous range that allows the user to alter the variable processing factor(s) throughout a range of values, and thus alter the processing operation of the processing means. Once the user has adjusted the control means to a setting in the substantially continuous range that is considered by the user an optimal setting for a particular acoustic environment, the user may operate an indicator means, such as a 3o switch or button, leading to storage of that setting and optionally, data indicative of the particular acoustic environment in the data memory means. Actuation of this indicator means may be taken by the processing means as indicating that the particular setting of the control means at that time is considered optimal by the user for the particular acoustic environment in which the user finds themnself.

COM 1D No: SBMI 010137S0 RceiTcd by Australia Time (Hu.e) 12:03 Date 200-4-12-21 21/12 '04 11:56 MAX 61 3 9663 3099 FB RICE CO. [a]012 o9 0In this embodiment, the control means may comprise a rotary wheel control.

0 The rotary wheel control may be mounted on the housing of the prosthesis or alternatively. may be mounted on a remote unit. The control means could also be in the form of a toggle switch or press buttons.

S

IND In the above embodiments, the indicator means may be mounted on a housing Cl of the prosthesis- In another embodiment, the indicator means may be mounted on a remote unit.

1o In a preferred embodiment, the settings selected by the user as being optimal to 0 that user for a plurality of acoustic environments may be stored in the data memory means and optionally, with data indicative of the characteristics of those particular acoustic environments. The processing means operates in the second mode, and may simultaneously be operable in the first mode for a defined period of time or can be considered as operating in the first mode every time the user adjusts the control means and selects what is considered a new optimal setting for that particular environment. In another embodiment, the prosthesis may have a defined training period in which the prosthesis may be operable in the first node of operation- This period may be as long as the user wishes or optionally, the prosthesis may signal to the user when the training 2G period is complete. In this case, the prosthesis may be adapted to indicate the training period is complete once there has been a particular pre-defined number of instances in which the user has not selected a setting which is substantially differunt to the setting already estimated by the prosthesis to be optimal.

In a preferred embodiment, the prosthesis may further comprise a sound analysis meaus. The sound analysis means preferably receives input signals from the microphone and monitors the acoustic environment of the prosthesis user. The sound analysis means preferably provides an output representative of the acoustic environment being monitored at that time.

In a further embodiment, the data memory means may comprise one or more data memory locations. In one embodiment, the data memory means may comprise five data memory locations. In this embodiment, the furst data memory location may contain the audiometric data of the prosthesis user and/or individual data for one or more loudness models used by the prosthesis. The second data memory location may contain characteristic data about the hearing prosthesis. The third data memory COMS 1D N: SBMI-01042752 Riwved by !P Australia: ime (H m) 12.08 Date (Y Md) 2004-12-21 21/12 '04 11:56 FAX 61 3 9063 3099 FD RICE C0, 013 0 location may comprise one or more equations used to predict the optimal processing 0operation of the processing means for an individual user in different acoustic environments. The fourth data memory location may store the optimal sound processing data as selected by the user of the prosthesis. This data may be ordered in different ways, such as sequentially or indexed according to the corresponding acoustic IN environment data supplied by the souud analysis means. The acoustic environment C'l data that corresponds to the optimal sound processing data is optionally stored in the fifth data memory location. Other data memory means having differing numbers of memory locations may be envisaged.

oIn one embodiment the fourth data memory location may store a predefined maximum number of sets of optimal sound processing data- In one embodiment, the memory locaton may store a maximum of 400 data sets. Other maximum numbers can, however, be envisaged. In one embodiment, the data processing means does not utilise all stored data sets but only a predefined number of most recently logged data sets. For example, the data processing means may only utilise the last 256 data sets when determining the optimal value of a variable processing factor. In one embodiment, the newest data set to be stored in the memory location can be stored instead of the oldest data set in the memory location. This first in first out storage system enLres only the most recently logged data is ever stored in the prosthesis at any one time. In another embodiment, older data cannot be overwritten such that once the memory location is full no further data sets can be logged by the prosthesis. In another embodiment, old data is preferably selectively overwritten by the newest sound processing data according to the acoustic environment in which the sound processing data was selected.

In a further embodiment, the prosthesis preferably further comprises a data processing means. The data processing means preferably receives the output of the sound analysis means. Based on the output of the sound analysis means, the data processing means can be adapted to calculate the loudness of sounds present at the microphone. In one embodiment, the data processing means can calculate the loudness of the sounds as they would appear to a person with normal hearing and/or to a person with a degree of hearing impairment. The data processing means can be adapted to calculate other acoustic and psychoacoustic measures of the sounds present at the microphone. The data processing means can use the acoustic and/or psychoacoustic measures as inputs to the one or more equations stored in the third data memory COMS ID Nc $SMI-012427S0 fcci,&j by Austraia Tir-o (Nm) I2:00 Oat Y Md) 2004-12-21 21/12 '04 11:57 FAX 61 3 9663 3099 FB RICE Co. I014 0 Clocation which estimate the optimal sound processing operation of the sound processing 0means for the user in the acoustic environments represented by the sound analysis means.

The data processing means further preferably uses the hearing prosthesis characteristic data stored in the second data memory location and tbe optimal sound Cl processor data generated by the equations to automatically and continuously determine 0the appropriate setting of the processing means to provide the optimal output signal for the user in the current acoustic environment being experienced by the user.

0 In a preferred embodiment, the processing means can include an amplifier means and a gain control means. In this case, the operation of the amplifier means is preferably adjustable when in the first mode to allow the user to optimise the gain at each or every frequency of the amplifier in a particular acoustic environment. Once optimised, the prosthesis operates in a second mode in which the amplification characteristics of the amplifier means automatically adjusts to the optimal level. That is, in such embodiments the variable processing factor is preferably the amplifier gain at each or every frequency.

In a preferred embodiment of the above aspects, the gain at each or every frequency of the amplifier means is calculated through use of an equation having a piedefined forn. In one form the equation can be: G= ai +bI max(LI,c) di (SNR, (1) where i the frequency band number, Gi gain for band i; ai trainable coefficient for band i; h- trainable coefficient for band i; ci trainable coefficient for band i; di= trainable coefficient for band i; 1L sound pressure level at the microphone in band i; SNRi signal to noise ratio in band i: and SNR,, average SNR in all bands.

COMS ID Nc: ZBMI21043750 Received bj IF' A~atraIia: Time (H l2:CS Date (Y-M d) ZZOA*1 2-21 21/12 '04 11:57 F.AX 61 3 9663 3099 FD RICE Co. 015 0 o 12

C-)

C) Other forms of the equation with other acoustic or psychoacoustic parameters, 0such as high-r-order moments of the spectrum of the signal and variations of these moments with time, or other statistical parameters or estimates of the acoustic signal or combinations of these parameters, and optionally with other additional coefficients, can calculate the gain or processing factors other than gain, such as the speed at which the prosthesis reacts to change in the acoustic environment, the choice of the equations Ci used will depend upon the particular application desired, with such applications falling 0within the scope of the present invention.

In the above embodiment, trainable coefficients include a, b, c and d, which as a 0result leads to the amplifier gain G being a variable processing factor.

Thus, in embodiments of the invention, more than one variable processing factor of the processing means can be adjustable by the user using the control means or multiple control means. Similarly, more than one variable processing factor of the processing means can be automatically adjusted as appropriate to any acoustic environment. In this embodiment, the control means or multiple control means can allow the user to adjust one or more of the following example operations of the processor means: the volume of the output signal; (ii) the gain of the output signal at particular frequencies relative to other frequencies, for example the mid frequency gain can be boosted or attenuated with respect to the low or high band frequencies of the output signal; and (iii) a slope control where the low and high frequency band gains are adjusted in opposing directions while the mid band gain is unchangedt In this embodiment, for each of the operations of the processor means the user can select a setting which is optimal for each variable processing factor for the particular acoustic environment that they are in by actuating the indicator means- On multiple operations of the control meaus the user may select a setting which is optimal for one or more variable processing factors for the acoustic envirornent that the user experiences by actuating the indicator means. Thus, more than one operation of the sound processor means, as well as more than one variable processing factor, can be adjusted in combination for each vote or selection. Each time the indicator means is actuated, the gain in each frequency band is preferably recorded along with a data set indicative of the acoustic environment detected by the microphone. This data can be COMS ID No: SSMJ 0104750 R2c,.,,cd by P Aush'alia: Tmc (im) 120 Dac 204-12-21 21/12 '04 11:57 F.AX 61 3 9663 3099 FB RIC.Eq CO.018 0 o13

C-)

used with previous sets of data by the data processing means to calculate the gain 0equation coefficients that best match the logged training data in each band. The recorded gain in each frequency band may be used to calculate the gain equation coefficients.

_In a further embodiment, when operating in the first mode the data processing means preferably does not calculate optimal gain equation coefficients until a predetermined number of selections have been made by the user. In one embodiment, the prosthesis can be adapted to not calculate optimal gain equation coefficients until at oo least fifty selections have been made by the user. In this case, where the predetermined 0, number of selectioDs have not been made, the processing means may when operating in the second mode preferably automatically calculate the output appropriate to the current enviroment based on predetermined coefficients and/or variable processing factors. Once the predetermined number of selections have been made, the processing means may calculate the optimal gain equation coefficients on the basis of previous user preferences when in the first mode of operation, and automatically and continuously calulate the variable processing factors with these equations when in the second mode of operation. In another embodiment, the data processing means preferably does not wait for a predetermined number of selections to be made by the user before calculating the optimal gain equation coefficients, but preferably does not use the calculated coefficients when in the second mode of operation until the variable processing factors that are preferred by the user can be predicted with the equations and calculated coefficients with a certain degree of accuracy.

In an alternative embodiment, equations may not be used to determine the optimal variable processing factors for different environments. The optimally adjusted variable processing factors may be stored in the fourth data memory at locations determined by the corresponding acoustic environment data supplied by the sound analysis means. Prior to the commencement of operation in the first mode, the memory so locations are preferably loaded with processing factors that are delived from empirical observations or prescriptive procedures for the corresponding acoustic environment parameter values. When operating in the first mode the optimally adjusted processing factor(s) may be stored in the fourth data memory at locations determined by the acoustic environment data. The optimally adjusted processing data may simply overwrite the prescribed data, or may be combined with old data written to the same location using a mathematical and/or statistical procedure. When operating in the COMS ID No Z '1 0104'0," Roccivd by: P Australia: Timc )H rn) 12.CS Datc d) 204-12 21 21/12 '04 11:58 FAI 61 3 9663 3099 FB RICE CO, i1017

O

o 14 Sssecond mode, the data supplied by the sound analysis means may be used to index the 0 fourth data memory location, and the value(s) stored at the indexed memory location are preferably used to set the target value(s) of the corresponding processing factor(s) C,1 for the sound processing means. The processing factor(s) applied in the sound processing means may slowly converge towards the target value(s) to avoid the IN possibility of undesirable acoustic effects associated with excessively rapid or CI instantaneous changes of the applied value(s). This alternative embodiment eliminates Sthe need for the fifth data memory location and the equations stored in the third data C memory location, and hence the variation of a processing factor with the acoustic Sto environment is not restricted by the relationships that are predefined in the equations.

0 In another alternative embodiment, the optimally adjusted variable processing factor(s) may be stored in the fourth data memory location and a mathematical and/or statistical procedure is used to calculate the optimal value of the processing factor(s).

This alternative embodiment eliminates the need for the equations stored in the third data memory location, and the storage of acoustic environment data either directly in the fifth data memory location or to index setting data in the fourth data memory location. This alternative embodiment is an appropriate simplification for processing factors that do not vary with the acoustic environment, such as the maximum permissible level of the output of a hearing aid or cochlear implant before the output is uncomfortably loud- In a further embodiment, the user can train the prosthesis to provide optimal variable processing factors for different listening criteria, such as to maximise speech intelligibility, listening comfort, or pleasantness. In one embodiment, an equation similar in form to equation 1 can comprise acoustic environment or psychoacoustic parameters that are highly correlated to a particular listening criterion, such as maximising speech intelligibility or listening comfort. When operable in the second mode the variable processing factors may automatically adapt for the predicted weighted combination of listening criteria as determined by the values of the coefficients and the acoustic environment and/or psychoacoustic parameters used by the equations.

In another embodiment, while operable in the first mode the user may indicate the listening criterion or combination of criteria used to determine the preferred adjustment of the variable processing factor(s) with an indicator means, or optionally COMS ID SClMI.D103750 Received by IP Australia: Ti.-i 12.08 Date (Y-M d) 2004-12 21 21/12 '04 11:58 FAX 61 3 9663 3099 FB RICE CO. I018 o 1 ciq the aid automatically predicts the listening criterion or combination of criteria from the 0values of acoustic environment and/or psychoacoustic parameters that are known to correlate with different listening criteria, and/or from the presence of signals of interest such as speech and music as detected by the prosthesis. The fourth data memory location (otherwise referred to as a first data memory location in claim 76) may be Idivided into sections for different listening criteria, and the optimally adjusted variable Cl processing factor(s) may be stored in a memory location(s) as determined by the 0listening criterion or combination of listening criteria under which they were adjusted by the user. The fifth data memory location (otherwise referred to as a further data oo memory location in claim 77) is optionally also divided into seotions for each different o listening criterion for the storage of the acoustic or psychoacoustic parameter values Cl that correspond to the values stored in the fourth data memory location. With this stucture, the data in each section of the fourth and fifth data memory locations is preferably used to calculate the optimal coefficients of variable processing factor equations similar in form to equation 1 for each listening criterion. When operable in the second mode, the user can indicate the current listening criterion or combination or criteria via an indicator means or alternatively, the listening criterion or weighted combination of criteria can be predicted from the values of acoustic environment and/or psychoacoustic parameters that are known to correlate with different listening criteria 2o and/or the presence of signals of interest such as speech and/or music, or by one or more trainable listening criteria equations that have acoustic environment or psychoacoustic parameters as their input where the optimal coefficients of each listening criterion equation are calculated from the data stored in the corresponding sections of the fourth and fifth data memory locations. Once the listening criterion or weighted combination of listening criteria are either indicated by the user or automatically predicted by the prosthesis, the optimal variable processing factors may be calculated with equations similar in form to equation I for each criterion, and are combined according to the indicated or predicted weighting of the listening criteria.

In an alternative embodiment, when operable in the first mode the listening criterion or combination of criteria indicated by the user via an indicator means or predicted by the prosthesis may be used in addition to the corresponding acoustic environment and!or psychoacoustic parameters to index the location in the fourth data memory where the optimally adjusted variable processing factor(s) are stored. When operable in the second mode, the current values of acoustic environment and/or psychoacoustic parameters and the indicated and/or predicted listening criterion or COM !D NC 0DMI 010437S0 R,=:Zived bt 0' ALSA:O?- Te 12M, Date 2004-12-21 23/12 '08 13:57 F-AX 61388184199 E IE&C FB RICE CO 10012 00 o 16 combination or criteria may be used to index the fourth data memory location to retrieve the optimal variable processing factor(s) for the prevailing conditions. in 0 another embodiment, the listening criterion or combination of criteria mnay not be used to index the fourth data memory location, and the processing factor value(s) stored at each location in the fourth data memory when operable in the first mode may be IND modified according to the weighted combination of listening criteria for the particular acoustic environtment.

In the above embodiments, the listening criterion or combination of criteria is indicated by the user by an indicator means. such as a switch, toggle switch, or set of o pushbuttons on the housing of the prosthesis, or alternatively on a remote unit According to a second aspect of the invention, there is provided an auditory prosthesis for rehabilitating the hearing of a user, comprising: a microphone configured to receive sound and to produce a microphone signal corresponding to the received sound; a sound processor configured to process the microphone signal using a set of variable processing factors to produce an output signal; and a sound analysis configured to receive the microphone signal, and to output a data set representative of the acoustic environment of the prosthesis; wherein the prosthesis is configured to provide the user with the ability to adjust at least one of the variable processing factors of the set, and wherein one or more of the variable processing factors are automatically adjusted by the prosthesis based on the adjustments of the at least one variable processing factor by the user.

In a further embodiment of the invention there is provided arn auditory prosthesis comlprismg: a microphone which receives sound and produces an output signal corresponding to the received sound; an output device that provides audio signals in a form receivable by a user of the prosthesis; a sound processing means operable to receive the microphone signal and carry out a processing operation on the microphone signal to produce an output signal in a form suitable to operate the output device, wherein the processing means is operable in a first mode in which the processing operation comprises at least one variable processing factor which is adjustable by a user to a setting which causes the output COMS ID No: ARC5-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 '08 13:58 FAX 61388184199FERC &COI01 FB RICE CO [a 013 00 o 16a o signal of the processing means to be adjusted according to the preference of the user for the characteristics of the cm-rt acoustic environment; a sound analysis means that also receives the microphone signal and outputs a c-I data set representative of the acoustic environment; a data memory means for storing a plurality of said user adjusted settings; and IND a data processing means that analyses said stored settings and is operable to output control signals to the sound processing means; o wherein, upon receiving control signals from the data processing means, the sound processing means is simultaneously operable in a second mode in which said at least one variable processing factor is automatically adjusted on the basis of the o previously adjusted variable processing factor selected by the user.

COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 21/12 '04 11:59 FAX 61 3 9663 3099 FB RICE CO. Q020 o17 In this aspect, the features of the prosthesis can have the features as described 0above with reference to the first aspect where such features are compatible with the operation of the prosthesis as desen'bed therein.

In this aspect, the hearing prosthesis preferably uses the data memory means to INO store a record of optimal settings as determined by the user of the prosthesis for c,1 different acoustic environments. This record is then used by the data processing means oto calculate coefficients of one or more equations that are used to predict the optimal c,1 setting of the processing operation of the speech processor for that user when exposed oo to a different acoustic environment.

In one embodiment the hearing prosthesis can be adapted to output audio signals to a single ear of the user. In another embodiment, the hearing prosthesis can be adapted to output audio signals to both ears of the user. In yet a further embodiment, signals from the inputs of two prosthesis can be connected to the sound analysis means by wire or wirelessly. The control means for the above embodiments can be located on the housing of each prosthesis, a remote unit for each prosthesis, or the control means for both prostheses can be located on a common remote unit.

In one embodiment of the above aspects, the hearing prosthesis can be a hearing aid. In this embodiment, the output device of the heaing aid can be an earphone that receives the output signal of the processing means and generates amplified sounds that arc delivered into the ear of the user.

In another embodiment, the hearing prosthesis can be an implanted hearing aid device whereby the output device of the prosthesis can be a vibrating mechanism, mechanically coupled to the middle or inner ear.

In auother embodiment of the above aspects, the hearing prosthesis can be a cochlear implant. In this embodiment, the output device can comprise a receiver/stimulator unit that receives encoded stimulation data from the processing means and outputs stimulation signals that are delivered to the cochlea through an electrode array. In this embodiment, the sound processing means can comprise a speech processor that uses a coding strategy to extract speech from the sounds detected by the microphone before or after processing of the microphone signal according to the variable processing factors- In one embodiment, the speech processor of the cochlear COMM, ID Nc SBMI.C10437Z5 Reccived by !P A-*,r.Ai Tinc 12:0S Date (Y-M d) 200-12-21 21/12 '04 11:59 FAl 61 3 9683 3099 FB RICE CO, 2]021 implant can perform an audio spectral analysis of the acoustiu signals and output 0channel amplitude levels. The transformation from the spectrun of the acoustic signal to electrical channel amplitude levels may be determined by a variable processing factor based on previously selected variable processing factors. The threshold and discomfort channel amplitude levels can also be processing factors that are based on IND previous adjustments and selections made by the prosthesis user. The sound processor can also sort the outputs in order of magnitude, or flag the spectral maxima as used in 0the SPEAK strategy developed by Cochlear Lt& oo The receiver/stimulator unit is preferably positioned within a housing that is o implantable within the user- The housing for the receiver/stimulator unit is preferably implantable within a recess in the bone behind the ear posterior to the mastoid.

The speech processing means is, in use, preferably mounted external to the iS body of the user such that the signals are transmitted transeutaneously through the skin of the user. Signals can preferably travel from the processing means to the receiver means and vice versa. The receiver/stimulator unit can include a receiver coil adapted to receive radio frequency (RF) signals from a corresponding transmitter coil worn externally of the body. The radio frequency signals can comprise frequency modulated 2o M signals. While described as a receiver coil, the receiver coil can preferably transmit signals to the transmitter coil which receives the signals.

The transmitter coil is preferably held in position adjacent the implanted location of the receiver coil by way of respective attractive magnets mounted centrally in, or at some other position relative to, the coils.

During use, the microphone is preferably worn on the pinna of the user, however, other suitable locations can be envisaged, such as a lapel of the implantre's clothing. The speech processor preferably encodes the sound detected by the microphone into a sequence of electrical stimuli following given algorithms. The encoded sequence is transferred to the implanted receiver/stimulator unit using the transmitter and receiver coils. The implanted receiver/stimulator unit may demodulate the FM signals and allocates the electrical pulses to the appropriate attached electrode by an algorithm which is consistent with the chosen speech coding strategy.

S !D No: 32MN-01043750 Rc";vcd by IP ?..utra!j: Timc (Iti) 120 Date (U 2 0-12 21 23/12 '08 13:58 FAX 61388184199 FB RICE CO 0~014 00 0 19 The external component housing the processing means frither preferably Uhouses a power supply. The power supply can comprise one or more rechargeable 0 batteries- The transmitter and receiver coils may he used to provide power via en transcutaneous induction to the implanted receiver/stimulator unit and the electrode

IND

While the implant system can rely on external componentry, in another embodiment, the microphone, speech processor and power supl can also be implantable. In th~is embodiment, the implanted components ca beP cotie3 wti hermetically sealed housing or the housing used for the stimulator unit.

c-i According to a third aspect of the invention, there is provided a method of rehabiliating the hearing of a user using an auditory prosthesis, comprising: converting a received sound to a microphonie Signal; processing the microphone signal at a sound processor to generate an output signal utilizing a set of variable processing factors, receiving from the user an adjustment of at least one of the variable processing factors of the set, and automatically adjusting one or more of the variable Processing factors based on the received adjustments of the at least one variable processing factor by the user.

hn an embodiment Of the invention there is provided a method of adjusting an auditory prosthesis by a user in an acoustic environment comprising the steps of: Producing a microphone signal from a microphone that corresponds to sound received by the microphone; Providing audio signals, through an output device, in a form receivable by the user; carrying out a processing operation, through a sound processing means, on the microphone signal so as to produce an output signal in a form suitable to operate the output device wherein the processing means is operable in a first mode in which the processing operation comprises at least one variable processing factor which is adjustable by a -user to a setting which causes the output signal of the sound processing means to be adjusted according to the preference of the user for the characteristics of the current acoustic environment; storing the setting in a data memory means; COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 '08 13:58 F AX 61386184199 F IE&C FB RICE CO 0 00 O1 9a wherei the sound processing men is simultaneously operable inx a second C) mode in which said at least one variable processing factor is automatically adjusted on 0 the basis of the previously adjusted variable processing factor selected by the user.

According to a fourth aspect of the invention, there is provided a method of IND rehabilitating the hearing of a user using an auditory prosthesis, comprising: converting a received sound to a microphone signal; o processing the microphone signal at a sound processor to generate an output signal utilizing a set of variable processing factors, receiving the microphone signal at a sound analysis unit that subsequently o outputs a data set representative of the acoustic environment of the prosthesis; receiving from the user an adjustment of at least one of the variable processing factors of the set;,and automatically adjustg one or more of the variable processing factors based on the received adjustments of the at least one variable processing factor by the user, In an embodiment of the invention there is provided a method of adjusting an auditory prosthesis by a user in an acoustic environment comprising the steps of-, producing a microphone signal from a microphone that corresponds to sound received by the microphone; providing audio signals, through an output device, in a form receivable by the user; COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 21/12 '04 12:00 F.AX 61 3 9663 3099 FB RICE CO. [02.

02 )carrying out a processing operation, through a sound processing means, on the 0microphone signal so as to produce an output signal in a form suitable to operate the output device wherein the processing means is operable in a first mode in which the processing operation comprises at least one variable processing factor which is adjustable by a user to a setting which causes the output signal of the sound processing means to be adjusted according to the preference of the user for the characteristics of the current acoustic environment; receiving the microphone signal at a sound analysis means that subsequently outputs a data set representative of the acoustic environment of the prosthesis; oo storing a plurality of user adjusted settings in a data memory means; 0analysing the stored user adjusted settings and transmitting control signals to the sound processing means; wherein, upon receiving the control signals, the sound processing means is simultaneously operable in a second mode in which said at least one variable processing factor is automatically adjusted on the basis of the previously adjusted variable processing factor selected by the user.

The present invention enables the user of an auditory prosthesis to adjust the operation of the processing means in a simple way while they are 'sing the prosthesis in a normal way. The invention further eventtally allows the user to 'train' the prosthesis to adjust its operation automatically. This has a number of advantages including: greater levels of user satisfaction for both hearing aids and cocblear implants, because of the better control users have over their hearing;, improved listening comfort, and/or speech intelligibility and/or subjective sound quality, because processing parameters are optimised by each user to the personal preference and needs of each user; the ability to generalise the 'training' algorithm to work effectively with future devices that have different, or additional, parameters that require adjustment; because the training is carried out by the user in daily life, expensive clinical time is not required to achieve adjustment of the hearing aid and as adjustment is caried out in real life conditions, the adjustment better relates to actual performance of the device, rather than ideal clinical conditions; and The ability to generalise the training algorithm to work effectively in environments for which no training has been provided by the user, by automatically COMS ID N% C.MI 01043750 Rc.Civo.J 1, 'P Australia. (H:rn 12:.S Date 2201-12-21 21/12 "04 12:00 FAX 61 3 9663 3099 FB RICE CO. Z024 0 applying the general reatiouships between preferred variable processing factors and 0acoustic characteristics that have been established by the training.

Brief Desetiution of the Drawims By way of example only, a preferred embodiment of the invention is now described with reference to the accompanying drawings, in which: Figure 1 is a pictorial representation of a prior art cochlear implant system; oo 0 Figure 2 is a block diagram of one embodiment of an auditory prosthesis according to the present invention; Figure 3 is a flowchart setting one particular mode of operation of the prosthesis of Figure 2; Figure 4 is a graph depicting results of experiments performed on a prototype prosthesis according to one embodiment of the present invettion; Figure 5 is a diagram representing the storage of processing factor values in a 2c two-dimensional memory space; and Figure 6 is a graph showing a relationship between input and output sound pressure levels at various stages of the training process according to one embodiment of the present invention.

Deftiled Description While the present invention is not directed solely to a cochlear implant, it is appropriate to briefly describe the construction of one type of known cochlear implant system with reference to Figure 1.

Known cochlear implants typically consist of two main components, an external component including a speech processor 29, and an internal component including an implanted receiver and stimulator unit 22. The external component includes a microphone 27. The speech processor 29 is, in this illustration, constructed and arranged so that it can fit behind the outer ear 11. Alternative versions may be worn on COMZ ID No. SBMI 010437,, T' Austrara. T11, (11:m) 12:0S Datc (Y M-J) 2'i4 12 21 21/12 '04 12:01 FAX 61 3 9663 3099 FB RICE CO.

0 o22

C-)

C) the body. Attached to the speech processor 29 is a transmitter coil 24 that transmits 0electrical signals to the implanted unit 22 via a radio frequency (RF) link.

Tbe implanted component includes a receiver coil 23 for receiving power and N data from the transmitter coil 24. A cable 21 extends from the implanted receiver and stimulator unit 22 to the cochlea 12 and terminates in an electrode array 20. The o- signals thus received are applied by the array 20 to the basilar membrane 8 and the nerve cells within the cochlea 12 thereby stimulating the auditory nerve 9. The operation of such a device is described, for example, in US Patent No. 4,532,930.

01 o 1 ci As depicted diagrammatically in Figure 1, the cochlear implant electrode array has traditionally been insetted into the initial portion of the scala tympani of the cochlea 12 up to about a full turn within the cochlea.

A block diagram depicting one embodiment of an auditory prosthesis accordiDg to the present invention is depicted generally as 30 in Figure 2. The block diagram depicts features of both a cochlear implant and a hearing aid. Despite this, it will be appreciated that the diagram is representative only and that a prosthesis according to the present invention may not necessarily act as both a cochlear implant and hearing aid.

The auditory prosthesis 30 comprises a microphone 27 which receives sound and produces a microphone signal corresponding to the received sound and an output device that provides audio signals in a form receivable by a user of the prosthesis As can be seen, the output device can comprise an earphone 31 in the case where the prosthesis 30 is a hearing aid. Where the prosthesis 30 is a cochlear implant, the output device comprises an encoder/transmitter unit 32 that outputs encoded data signals to the external transmitter coil 24.

The prosthesis 30 farther has a sound processing unit 33 that is operable to receive the microphone signal provided by the microphone 27 and produce an output signal in a form suitable to operate either the earphone 31 or the encoder/transmitter unit 32.

The nature and fimction of the sound processor 33 will depend on whether the prosthesis 30 is a cochlear implant or a hearing aid. In the case of a hearing aid, the COMS ID N .S 41-0104S750 RcC, :2 FP A:;ltL;. ,im: 12CC D t (Y M.t 04.12 .21 21/12 '04 12:01 F.AX 61 3 9663 3099 FB RICE Co. 1026 o23 ci C) sound processor 33 includeg at least an amplifier whereas in the case of cochlear 0 implant the sound processor 33 will include an amplifier and a speech processor that uses a coding strategy to extract speech from the sounds detected by the microphone 27. In the depicted embodiment, the speech processor of the cochlear implant can perform an audio spectral analysis of the acoustic signals and output channel amplitude _levels. The sound processor can also sort the outputs in order of magnitude, or flag the spectral maxima as used in the SPEAK strategy developed by Cochlear Ltd. Other coding strategies could be employed.

Iu this invention, the sound processor 33 comprises at least one variable processing factor, such as amphfication, which in a first mode of operation is adjustable to a setting in which the output signal of the sound processor 33 is optimised for the preference of the user in at least one acoustic environment. For example, the amplification characteristics of the amplifier can be adjusted by a user of the prosthesis 30 to suit the surrounding background noise in a particular environment.

The prosthesis 30.firther comprises a data memory means 34 for storing at least data indicative of the setting selected by the user and optionally data sets representative of the characteristics of the acoustic environment at the time the setting adjustment and selection is made.

The sound processor 33 when operable in a first mode recalculates the trainable coefficients associated with a corresponding variable processing factor after each adjustment by the user.

The sound processor 33 can be adapted to offer two or more possible optimal settings for selection by the user. One of these settings may be the one automatically calculated by the processor and the other may be a systematic departure therefrom. In this case, the user is preferably able to compare the operation of the prosthesis when 3o operating in a particular acoustic environment when operating with each setting and then select fRom these the setting that is best for that particular enviromnent The setting that is selected by the user can then be stored in the data memory means 34 optionally with data indicative of the particular acoustic environment. This process can be repeatable so as to allow the data processing unit 38 to gather the preferences and to monitor whether the user's preference for a particular environment changes with time or usage. In this way, the user effectively selects or votes for the best setting each time.

COMG ID Nt. ZMI-01042750 R.vcd by rPA7ut;m:r T-nc (Hm) 12.0' Datc Y.M.d) 204 2-21 21/12 '04 12:01 FAX 61 3 9663 3099 FB RICE CO. _0027 0 o 24 0 In this embodiment, the user can alternate between settings by operating a ci control means 36 which can be a toggle switch or a set of pushbuttons for example, and select a setting by operating an indicator means which is represented by switch 35 in ID 5 Figure 2. The switch 35 and control means 36 can be mounted on the housing of the prosthesis 30, or while not depicted, the switch and control means could be mounted on o,1 a remote unil The processing operation of the sound processor 33 can be operable in another 1o manner or mode, in which its operation is adjustable by the user. Rather than offering a c, selection of possible optimal settings for selection by the user, in this arrangement, the user is able to adjust a control means 36 that allows the user to alter the processing operation of the sound processor unit 33. Once the user has adjusted the control means 36 to what is considered by the user an optimal setting for a particular acoustic environment, the user can operate the switch 35, leading to storage of that setting and optionally data indicative of the particular acoustic environment in the data memory means 34. Actuation of the switch 35 is taken by the sound processor unit 33 as indicating that the particular setting of the control means 36 at that time is considered optimal by the user for the particular acoustic environment in which the user finds themself.

The control means 36 can comprise a rotary wheel control that may be mounted on the housing of the prosthesis 30. The switch 35 can also be mounted on the housing of the prosthesis 30. While not depicted, the switch and controller could be mounted on a remote unit.

The settings selected by the user as being optimal to that user for a plurality of acoustic environments are stored in the data memory means 34 optionally with data indicative of those particular acoustic environments. The sound processor 33 continuously operates in the second mode, and can simultaneously be operable in the first mode for a defined period of time or can be considered as operating in that mode every time the user adjusts the control means 36 and selects what is considered a new optimal setting for that particular environment. The training period can be as long as the user wishes or optionally, the prosthesis 30 can signal to the user when the training period is complete. In this case, the prosthesis 30 can be adapted to indicate the training period is complete once there has been a particular pre-defined number of COMC ID 9Mb1-O375O R-lcIccJ L, IP Autarla Timr 12 0 a M d, 004 2-21 21/12 '04 12:02 FAX 61 3 9663 3099 FB RICE CO. M i028 0 o ci

C-)

a) instances in which the user has been unable to select a setting that is substantially 0different from the setting already estimated by the prosthesis to be optimal.

ci The prosthesis 30 further comprises a sound analysis module 37. The sound analysis module 37 receives input signals from the microphone 27 and monitors the acoustic environment of the prosthesis user. The sound analysis module provides o output(s) representative of the type of the acoustic environment being monitored at that time.In the depicted embodiment, the data memory means 34 comprises five data ci memory locations. In this embodiment, the first data memory means 34a contains the io audiometric data of the prosthesis user and/or individual data for one or more loudness models used by the prosthesis 30. The second data memory location 34b contains characteristic data about the hearing prosthesis 30. The third data memory location 34c comprises one or more equations used to predict the optimal processing operation of the sound processor unit 33 for an individual user in different acoustic environments.

The fourth data memory location 34d stores the optimal sound processing data as selected by the user. The acoustic environment data that corresponds to the optimal sound processing data stored in the fourth data memory location is optionally stored in the fifth data memory location 34e.

In the depicted embodiment the fourth data memory location 34d stores a predefmed maximum number of 400 sets of optimal sound processing data. Other maximum numbers can, however, be envisaged. The depicted embodiment uses a data processing unit 38, that is described in more detail below. The unit 38 does not utilise all stored data sets but only a predefined number of most recently logged data sets. In this case, the data processing unit 38 only utilises the last 256 data sets when calculating the optimal equation coefficients- Once the memory location is full, new data sets are stored in the memory location by overwriting the oldest data set in the memory location. This first in first out storage system ensures only the most recently logged data is ever stored in the prosthesis 30 at any one time. In another embodiment, older data cannot be overwritten such that once the memory location is full no further data sets can be logged by the prosthesis. In another embodiment, rules are used to determine the old data that is overwritten by the newest data set, based on the worth of each old data set according to predetermined criteria.

As indicated above, the prosthesis further comprises a data processing unit 38.

The data processing unit 38 receives the output of the sound analysis module 37.

C=CM ID No. NS-01042750 9cci'ccl b, IP Astra@ia: Time 12:CC Datc 2004 2-1 21/12 '04 12:02 FAX 61 3 9663 3099 FB RICE CO. _]029 o26 Based on the output of the sound analysis module 37, the data processing unit 38 may 'be adapted to calculate the loudness of sounds present at the microphone 27. In the ci depicted embodiment the data processing unit 38 may calculate the loudness of the sounds as they would appear to a person with normal hearing and/or to a person with a IND 5 degree of hearing impairment. The data processing unit 38 can be adapted to calculate other acoustic and psychoacoustic measures of the sounds present at the microphone 27. The data processing unit 38 can use the measures of the acoustic conditions as inputs to the one or more equations stored in the third data memory location 34c which estimate the optimal sound processing operation of the sound processor 33 for the user 1o in the acoustic environments determined by the sound analysis module 37.

The data processing unit 38 further uses the hearing prosthesis characteristic data stored in the second data memory location 34b and the optimal sound processor data generated by the equations to automatically and continuously detemine the appropriate setting of the sound processor 33 to provide the optimal output signal for the user in the current acoustic environment being experienced by the user.

As described, the sound processor 33 includes an amplifier means and a control means 36- At all times the data processing unit 38 operates in the second mode, where 2o it automatically calculates the operation of the sound processor 33 that is most likely to provide the optimal amplification characteristics in any environment, as represented by the output of sound analysis unit 37. The prosthesis can also simultaneously operate in the first mode, where operation of the amplifier means is also adjustable by the user varying control 36 to allow the user to optimise the amplification characteristics in the current acoustic environment.

Where the prosthesis 30 is used in any environment, including an environment never previously experienced by the user, the desired gain of the amplifier is calculated through use of an equation having a pre-defined form. In this embodiment, the equation is: Gi ae bi max(Lj,cj) d 4 (SNR,-SNR,) (1) wbere i the band number; Gi gain for band i; ai trainable coefficient for band i; COMS I M 01213. Rccocid 'PAustra a Time 12:00 Date (Y 200 1 '2-21 21/12 '04 12:02 FAX 61 3 9663 3099 FB RICE CO. 030 0 o 27 ci b trainable oeffiient for band i; ci b trainable coefficient for band i; SCi trainable coefficient for band i; di =trainable coefficient for band i; Li sound pressure level at the microphone in band i; SNRi signal to noise ratio in band i; and SNRv average SNTR in all bands.

ci HFor each vote, the following training data is stored for each band i: ~the sound pressure level or SPL, Lj, in dB SPL; o 10 the signal-to-noise ratio in band i minus the average for all bands, that is SNRa- SNR.v, in dB; and the preferred gain Gi also in dB.

Each of the above values are calculated based on the signal at the microphone 27 for a period of time that precedes the actual vote. The values may be calculated based on the signal at the microphone 27 for a time preceding the vote and/or for a time after the vote has been cast. The stored L is the average or RMS value of the SPL over the predetermined period of time and the stored Gi is the average value of the gain over a predetermined period of time prior to the actuation of the voting button plus the user's preferred gain adjustment at the time of actuation of the voting button. The SNR in band i is currently calculated at the SPL at the microphone exceeded 10% of the time in band i minus the SPL at the microphone exceeded 90% of the time in band i over the predetermnnined period of time prior to the vote. This difference may be more accurately described as an estimate of the modulation depth of the signal. It is envisaged that a more accurate SNR or even speech-to-noise ratio estimation techniques may be employed.

The stored training data from many votes, for example up to 256, is used to form a set of simultaneous equations The set of simultaneous equations is solved as a set of linear equations, however since coefficient bi and coefficient ci are potentially multiplied by each other, that is not linear, the simultaneous equations are initially linearised. In other words the value of Ci is set to a predefined value, and the MAX ci] operation performed with this value of ci and the stored L values in order to set the bracketed term, in other words the greater of 1 i and ci the result of which is L'i. Subsequently the optimal values of aj, bi and d are calculated with a direct numerical method wherein the optimal coefficient values are the ones that give the COMS ID No. 5DMI 0104273 Oc.ic.J b, 'P Austra-a: Tlmc 1.0 Da c (Y M-d)2004-12-21 21/12 '04 12;03 FAX 81 3 9663 3099 FB RICE CO. 031

O

o 28 d) least squared error between the stored G, values and Gi values that are calculated with Sthe corresponding L'i and stored SNRj SNRa, values. This is repeated for at least four alternative values of ci, for example 35, 45, 55 and 65 dB SPL, and the set of coefficients that gives the least squared error is selected as the optimal set and applied to the processing of the signal, unless the value of bi is outside the range of -1.0 to or if less than fifty votes have occurred- Other versions may apply the trained

C

coefficients after more or less than fifty votes.

The direct numerical method used is lower-upper (LU) decomposition and back substitution which is a variant of Gaussian elimination the set of simultaneous Sequations are stored as matrices in memory, Ax b, where matrix A contains the stored acoustic environment data, x is a vector that contains the coefficients to be calculated (that is ai, bi and di), and b is a vector that contains the stored Gi values. The LU decomposition method operates on these matrices and solves vector x. Other variants of Gaussian elimination or other direct numerical methods can be used in other applications.

As an example, the training data from four votes may result in the set of simultaneous equations as follows: 25 a (b *35) (d *0) ai (bj *45) (di *0) ai+(b (di*0) a (bi (d *0) In this case, the trained coefficients would be ai 42.5, bi c 35 and di 00.0 This set of coefficients would give OdB error. In practical situations, there may be more variability in people's amplification preferences, or scatter in the training data.

Thus the preferred, user adjusted, Gi values on the left of the equal signs in the above equations would not go up so uniformly in even steps, and the trained coefficient values would be different.

Other forms of the equation with other acoustic or psychoacoustic parameters, such as higher-order moments of the spectrum of the signal and variations of these moments with time, or other statistical parameters or estimates of the acoustic signal or combinations of these parameters, and optionally with other additional coefficients, can COMS ID rN: S Ml 0.10.127SO 5c;& by IP Au~tral;a. 12 03 DatL (Y d) 2004-12-21 21/12 '04 12:03 FAI 61 3 9663 3099 FB RICE CO. [@032 0 o 99 ci calculate the gain or processing factors other than gain, such as the speed at which the 0prosthesis reacts to change in the acoustic environment, the choice of the equations used will depend upon the particular application desired, with such applications falling within the scope of the present invention.

S

IND Ia the above embodiment, trainable coefficients include a, b, c and d, which as a o- result leads to the amplifier gain G being a variable processing factor.

More than one processing factor of the sound processor 33 can be autotnatically calculated using equations similar to equation 1, and then further adjusted by the user io using the control means 36 or multiple control means. in this embodiment, the control 0- means 36 can allow the user to adjust one or more of the following operations of the sound processor 33: the volume of the output signal; (ii) the gain of the output signal at particular frequencies relative to other frequencies, for example the mid frequency gain can be boosted or attenuated with respect to the low or high band frequencies of the output signal; and (iii) a slope control where the low and high frequency band gains are adjusted in opposing directions while the mid band gain is unchanged.

The user can select which setting of the control means 36 is the optimal one for the particular acoustic environment that they are in by actuating the switch 35. Each time the switch 35 is actuated, the gain in each frequency baud is logged along with a data set indicative of the acoustic environment detected by the microphone 27. This data can be used with previous sets of data by the data processing unit 38 to calculate the gain equation coefficients that enable the equations to best predict the preferences of the user from the logged training data in each band. Because the data processing unit 38 has access to the preferred amplification characteristics and the acoustic environments in which they were preferred, the data processing unit 38 can calculate general relationships between the amplification characteristics and the measured aspects of the acoustic environment. The automatically calculated amplification characteristics can thus vary in a complex but defined manner as the acoustic environment changes- For example, in the current embodiment, gain at each frequency varies with the overall input level and the signal-to-noise ratio at said frequency relative to the signal-to-noise ratio at other frequencies.

COMO ID £0MI-01040750 Pocvcd by P ustim(. 'No 1202 Dac Y M-d) 20041 2,21 21/12 '04 12:03 FAX 61 3 9663 3099 FB RICE CO. a]033 030 o c.

U The depicted data processing unit 38 does not calculate optimal gain equation Scoefficients until a predetermined number of selections have been made by the user. In the depicted embodiment, the prosthesis 30 is adapted to not calculate optimal gain equation coefficients until at least fifty selections have been made by the user. In this case, where the predetermined number of selections have not been made, the sound processor 33 will preferably output a signal calculated on the basis of the initial, predefined values of trainable coefficients. These initial, pre-defined values are calculated j- for each user by conventional methods of prescribing prosthesis operation, or by an empirical, trial and error adjustment process. Once the predetermined number of selections have been made, the data processing unit 38 re-calculates the trainable 0 coefficients immediately after every occasion on which the user operates the switch to indicate that the control 36 is in the optimal position. In another embodiment, the data processing unit 38 does not wait until a predetermined number of selections have been made by the user to re-calculate the trainable coefficients, but does not apply these coefficients until the equations can accurately predict the preferences of the aid user with the trained coefficients, and/or the user has made selections in a preferred minimum range of acoustic environments.

In an alternative embodiment, a recursive-averaging equation is used to determine optimal (or trained) values for one or more coefficients.

For example, if the coefficient is the gain below the compression threshold then the optimal, trained setting can be determined using the equation: g(n) g(n- 1) w y AG (2) where n is the vote number g(n) is the trained gain below the CT after vote n w is a weighting factor AG is the adjustment made by the user to the variable processing factor, gain.

y [90 L] [90 CT] (calculated y set to within bounds: 0.0 y Similarly, if the coefficient is the compression ratio then the optimal, trained setting can be determined using the equation: COMS 0!D No. 3BMI-1040750 Reccivcd by IP Auctrati Time 12 05 Date 2004-.2 21 21/12 '04 12:04 F.AX 61 3 9663 3099 FB RICE CO. 034 0 o 31 IiCR(a) I/CR(n 1) q- [w AG (1 2y)] I [90 CT] (3) 0 where CR(n) is the trained CR after vote n The trained values of the coefficients, as calculated with the recursive-averaging °equations above, are then converted into coefficients of the following gain equation: c' G= a +b MAX[L, c) (4) SLo In this example, c CT, b i, and a g(n) [b c].

0 ci The aid user adjusts the output of the aid to his/her preference (in effect the aid user is adjusting the applied gain). After voting, the aid uses this output adjustment to calculate the trained compression ratio and gain below the compression threshold with the above recursive-averaging equations.

Therefore, a single output (or gain) adjustment by the aid user has been used to train two coefficients the CR(n) and the gain below the CT, The CR(n) and g(n) values are used by the aid to continuously and automatically calculate the gain that is applied by the aid, G, with Eq. 4.

Note that Eq. 4 is very similar to the gain equation that is trained in the simultaneous equations method (a b and c CT).

Therefore, Eqs. 2 and 3 replace the simultaneous equations used in the earlier embodiment, and do not require the storage of a finite number of sets of preferred gain and corresponding acoustic environment data in the aid's memory.

The above can be repeated for i frequency channels. For example, there may be three frequency channels (low, mid, high) and the aid user may adjust the aid's output in each of these channels. The output adjustment for each channel would then be used to calculate the trained values of CR(n)i and g(n)j for each channel, and then aj and bi for each channel. The aid would then automatically calculate the applied gain, Gi, for each of the i channels many times per second for the current value of the input level, LL Referring to Fig. 6, the dot shows the preferred output level that was voted for by the aid user for the input level, L, that existed at the time- The solid line shows the COMS ID Nc,. 5DMI 01043750 R tcoiv 'P Austrialia: T 12.CS Datc (Y-M 04 12 21 21/12 '04 12:04 F.A 61 3 9663 3099 FB RICE CO. @035 o 32 (U input-output function before the vote, and AG is the gain adjustment that was voted for 0 by the aid user. The dashed line shows the newly trained input-output function that may exist after this vote using the above recursive-averaging equations.

It will be appreciated that direct user adjustments can be made of the aid's settings (to his/her preference) and used to calculate an optimal set of coefficients with Cl some sort of algorithm. The training algorithms can also be used to calculate the coefficients of other non-gain) equations, as well as the values of variable processing factors that are not calculated with an equation, such as a global volume 0 to adjustment or the T level for an electrode of a cochlear implant.

0 In an alternative embodiment, the data processing unit 38 does not use the equations stored in the third data memory location 34c to determine the optimal variable processing factors for different environments. The optimally adjusted variable processing factors are stored in the fourth data memory location 34d at locations determined by the corresponding acoustic environment data supplied by the sound analysis means 37. Prior to the commencement of operation in the first mode, the fourth data memory locations 34d are loaded with processing factors that are derived from empirical observations or prescriptive procedures for the corresponding acoustic environment parameter values. When operating in the first mode the optimally adjusted processing factor(s) are stored in the fourth data memory 34d at locations determined by the acoustic environment data supplied by the sound analysis means 37.

For example, the fourth data memory location 34d can store a multi-dimensional (or Ndimensional) look-up table or matrix, where each dimension is a different acoustic environment or psychoacoustic parameter, and the parameter values determine the location in the table, matrix, or memory location 34d where the preferred processing factor(s) is stored. Thus the matrix is indexed by rounded values of the N acoustic environment parameters (or optionally the estimated psychoacoustic parameters such as loudness). The optimally adjusted processing data may simply overwrite the prescribed data in the fourth data memory 34d, or may be combined with old data written to the same location using a mathematical and/or statistical procedure. When operating in the second mode, the acoustic environment and psychoacoustic parameter values supplied by the sound analysis means 37 are used to index the table or matrix or fourth data memory location 34d, and the value(s) stored at the indexed location are used to set the target value(s) of the corresponding processing factor(s) for the sound processing means 33. The processing factor(s) applied in the sound processing means 33 may COMC ID I SB t 01048750 Rcccivcd b AUstralia: Ti;ne 12.08 Date (Y M J; 200'-12 21 21/12 '04 12:04 FAX1 61 3 9663 3099 -FS RICE CO. ij036 slowly converge towards the target value(s) to avoid the possibility of undesirable 0 acoustic effects associated with excessively rapid or7 instantaneous changes of the applied value(s). For memory locations that contain a prescribed value of a processing factor, the target value can be adjusted to represent the trend of user-adjusted values for environments located near the current environment in the memory location 34d. For IND this alternative embodiment there is no need for the fifth data memory location 34e and ci the equations stored in the third data memory 34c, and the variation of a processing o factor with acoustic parameters supplied by the sound analysis means 37, or psychoacoustic parameters calculated by the data processing unit 3 8, is not restricted by o 10 the relationships that are predefined in the equations stored in the third data memory o location 34c.

In one embodiment, the hearing prosthesis can be. adapted to output audio signals to a single ear of the user, In another embodiment, the hearing prosthesis can be adapted to output audio signals to both cars of the user.

1n a further embodiment, the trainable hecaring prosthesis can be adapted to accept microphone signals from microphones mounted on each side of the head, end then to output signal to either or both cars.

Figure 3 is a flowchart of the logic behind the operation of the data processing unit 38.

The Start of the flowchart is when the prosthesis 30 is turned on and power is supplied to the pants thereof including the sound processor 33 and data processing unit 38. The end point of the flowchart is wbeD) the user turns the prosthesis 30 off or if the battery for fth power source goes flat. While not depicted, it is possible that another end point could be defined as when a certain number of votes are reached, or if the user stops training the prosthesis 30 for- a predetermined period of time, or if the prosthesis 3o 30 determines from the user adjustments of the control means 36 that it is sufficiently well-trained. Still further, an audiologist or other third party could disable further training of the prosthesis when the user is satisfied with the performance of the prosthesis 30 in a range of different acoustic environments. However, it is desirable that the user can train the prosthesis at any time, so that training does not need to he reenabled if the user encounters a -new environment or if the hearing loss profile of the user changes over time or with slow acclimatisation to the prosthesis.

COMS ID Nc. SOMI-D1042750 Rcsciwcd by P Au&halia. Time 12:08 D~tc (V M d~ 2004 1221 21/12 '04 12:05 FAX 61 3 9663 3099 FB RICE CO. 037 0 o 34 ci 0 Figure 4 is a graph of the results of preliminary experiments performed on a prototype prosthesis according to the present invention- In the experiments, the goal was to train the prosthesis to produce a greater gain than provided by the prosthesis prior to the training in all frequency bands for any input sound. Figure 4 indicates how IN the necessity to adjust the control 36 (RMS control adjustment in dB) following groups C l of 15 votes to get the desired output decreased due to the "training" ability of the 1prosthesis.

Cl SIo Figure 5 is an illustration of the matrix previously referred to using acoustic 0 parameters of equation 1.

Figure 5 is a diagram representing the storage of processing factor values in a two-dimensional memory space that is indexed by the values of two acoustic parameters, the sound pressure level for band i and the signal-to-noise ratio for band i relative to the average for all bands, although this does not exclude the use of a greater number of dimensions from the scope of this invention, and the values shown are illustrative only. The numbers in the memory locations represent gain values, where gain is a variable processing factor. The stored values are prescribed values, except the bold italic values which represent data that differs from the initial prescribed data due to the storage of user-preferred setting data as a result of training by the user- It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments arc, therefore, to be considered in all respects as illustrative and not restrictive.

COMS ID No: SBMI 010437!7 Rece.cl a, rAcAlia: Tinme (tr:m) 12:C3 Datc Y.M-d) 200.1-12-21

Claims (76)

1. 01. An auditory prosthesis for rehabilitating the hearing of a user, comprising: c-i a microphone configured to receive sound and to produce a microphone signal corresponding to the received sound; and IND a sound processor configured to process the microphone signal using a set of variable processing factors to produce an output signal, wherein the prosthesis is o configured to provide the user with the ability to adjust at least one of the variable processing factors, and wherein one or more of the variable processing factors are automatically adjusted by the prosthesis based on the adjustments of the at least one o variable processing factor by the user,
2. 2. An auditory prosthesis according to claim 1, wherein the user operates a control means to adjust the at least one variable processing factor, wherein further the control means is adjustable throughout a substantially continuous range of settings and a preferred setting is used to calculate the at least one variable processing factor which alters the processing operations applied to the microphone signal.
3. 3. An auditory prosthesis according to claim 2 wherein the at least one variable processing factor is calculated by defined mathematical relationships incorporating coefficients, and wherein the value of the coefficients are calculated to be those values that cause the variable processing factor(s) to best approximate the variable processing factor(s) that occurred as a result of current and previous settings.
4. 4. An auditory prosthesis according to any one of the previous claims, further comprising: data memory for storing the at least one variable processing factor adjusted by the user.
5. 5. An auditory prosthesis according to claim 4, wherein the prothesis is configured to provide the user with the ability to repeatedly adjust the at least one variable processing factor.
6. 6, An auditory prosthesis according to claim 5, wherein each adjustment of the at least one variable processing factor by the user is monitored by the sound processor. COMS ID No: ARCS-218238 Received by IP Australia: lime 15:03 Date 2008-12-23 23/12 '08 13:59 FAX 61386184199 FB RICE CC 1 U017 00 o 36
7. 7. An auditory prosthesis according to any one of claims 4 to 6. wherein the at least one variable processing factor comnprises an amplifier gain factor.
8. 8- An auditory prosthesis according to any one of claims 4 to 7, wherein the data futher stores data sets representative of the characteristics of the acoustic INO environment in which the at least one variable processing factor was adjusted by the User.
9. 9. An audio Prosthesis according to claimn 8, wherein the one or more variable processing factors are automatically adjusted on the basis of the acoustic environments in which the at least one variable processing factor was adjusted.
10. An auditory prosthesis according to any one of the previous claims, wherein the prosthesis is configured to provide the user with the ability to adjust the at least one variable processing only during a defined training.
11. 11. An auditory prosthesis according to claim 10 wherein the prosthesis signals to the user when the training period is complete.
12. 12. An auditoryn-prosth 05 is according to claim 10, wherein the prothesis is configured to automatically adjust the one or more variable processing factors only after the at least one variable processing factor has been adjusted by the user a predefined number of instances.
13. 13. An auditory prosthesis according to any one of the previous claims, fiurher comprising: a sound analysis unit configured to receive input signals from the microphone, and to monitor the current acoustic environment in which the prosthesis is operating.
14. 14. An auditory prosthesis according to claim 13, wherein the sound analysis unit provides an output representative of the monitored acoustic environment. An auditory prosthesis according to claim 14, wherein the data memory comprises one or more data memory locations.
15. COMS IDNo: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 '08 13:59 FAX 61386184199 F IE&C FB RICE CO 01)18s 00 0 37 o
16. 16. An auditory prosthesis according to claim 15, wherein the data memory comprises 0 five data memory locations, and wherein a first data Memory location stores audiometric data of the user and/or individual data for one or more loudness models c-i used by the prosthesis.
17. INO17. An auditory Prosthesis according to claim 16, wherein a second data memory location stores characteristic data of the prosthesis.
18. 18. An auditory prosthesis according to claim 16 or claim 17, wherein a third data memory location stores one or more equations used by the sound processor. ci
19. 19. An auditory prosthesis according to any one of claims 16 to 78, wherein a fourth data memnory location stores the at least one variable processing factor adjusted by the 123cr.
20. An auditory prosthesis according to claim 19, wherein the data stored in the fourth data memory location is indexed according to the corresponding acoustic environment data supplied by the sound analysis unit.
21. 21. An auditory prosthesis according to claim 19 or claim 20, wherein a fifth data memory location stores acoustic environment data corresponding to the at least one variable processing factor stored in the fourth data memory location,
22. 22. An auditory prosthesis according to any one of claims 19 to 21, wherein the fourth data memory location stores a predefined maximum number of adjustments of the at least one variable processing factor,
23. 23. Au auditory prosthesis according to any one of claims 14 to 22, futrther comprising a data processing unit configured to receive the output of the sound analysis unit.
24. 24. An auditory prosthesis according to claim 23, wherein the data processing unit is adapted to calculate the loudness of sounds present at the microphone based on the output -from the sound analysis unit,
25. 25. An auditory prosthesis according to claim 23 or claim 24, wherein the data processing unit, based on the output from the sound analysis unit, calculates acoustic COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 '08 13:59 FAX 61386184199 P IE&C FB RICE CO [a]1019 00 o 38 miclrpsho st measures, such as loudness, of the sound received at the mirpoe Cfl
26. 26. An auditory prosthesis according to claimn 25, wherein the data procesing unt is configured to calculate the loudness of the received sound in relation to the loudness of INDthe sound to a person with normnal hearing and/or to a Person with a degree of bearing imfpairment.
27. 27. An auditory Prosthesis according to any One Of claims 5 to 26, wherein te prosthesis is configured to use a Predefined number of the most recent adjustments of the at least one variable processing factor to automatically adjust the one or more ci variable processing factors.
28. 28. An auditory prosthesis accordinge to any One Of the previous claims, wherein the at 1-9 least one variable processing factor comprises a plurality of variable processing factors.
29. 29. An auditory prosthesis according to claim 28, wherein the user operates multiple control means in order to adjust the plurality of variable processing factors.
30. 30. An auditory Prosthesis according to claim 29, wherein the multiple control means are configured to permit the user to adjust one or more of the following operations of the sound processor: the volume of the output signal; (ii) the gain of the output signal at Particular frequencies relative to other frequencies;, and (iii) a slope control where low and bigh frequency band gains are adjusted in opposing directions while a mid-band gain is unchanged.
31. 3 1. An auditory prosthesis according to claim 30, wherein for each of the operations of the sound processor, the user selects a setting for each variable processing factor within the set which is optimal for an acoustic enviroinment that the user experiences.
32. 32. An auditory prosthesis according to claimn 30, wherein for multiple operations of the sound processor, the user selects a setting for one or more variable processing factors within the set which is optimal for an acoustic environment that the user experiences.
33. COMS ID No: ARCS-21823s Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 '08 14:00 FAX 61386184199 1 IE&C FB RICE CO [aI 020 00 An aditry udioryProsthesis according to claim 12, wherein the Predefined n u b ro0n t n e s f f y
34. 34. An auditory prosthesis according to claim 12, wherein until the predefined number of instances has been made, the sound processor processes the mi crophone signal using IND an initial, predefined set of' variable Processing factors.
35. An auditory prosthesis according to claim 34 wherein the initial predefined se of variable processing factors are calculated for each user via an empirical, trial and error adjustment process, ci
36. 36. An auditory prosthesis according to claim 15, wherein the at least one adjusted Variable processing fa~ctor is stored at the one Or more locations as determined by corresponding acoustic environment data supplied by the sound analysis unit.
37. 37. A-n auditory Prosthesis according to claim 36, wherein the adjusted at least one variable processing factor is in a multi-dimensional look- up table where each dimension is a different acoustic environment or psychoacoustie parameter.
38. 38. An auditory prosthesis according to claim 37, wherein the multi-dimensional look- up table is indexed by at least two parameters chosen from acoustic environment parameters or psychoacoustjc parameters.
39. 39, An auditory prosthesis according to claim 38, wherein the parameter values determine the location in the look-up table in which preferred variable processing factors are stored.
40. An auditory prosthesis according to any one of the previous claims, wherein the prosthesis is configured such that the One or more automatically adjusted processing factors converge towards target values.
41. 41. An' auditory Prosthesis according to claim 15, wherein the at least one adjusted variable processing factor is stored at the one or more locations as calculated by a ma2thematicsi and/or statistical procedure. COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 '08 14:00 FAX 61386134199 E IE&C FB RICE CO ZTh021 o0
42. 42. An auditory Prosthesis according to any one of thle previous claimS, wherein, the Protheisis configured to further base the adjustment of the one or more vral 0 processing faictors on values of acoustic environment and/or psychoacoustic parameters that are known to correlate with different listening criteria. IND
43. 43. An auditory prosthesis according to claim 42, wherein the prosthesis automatically Predicts the listening criterion or criteria from the presence of speech and/or music signals of interest detected by the prosthesis.
44. 44. An auditory prosthesis according to claim 42, wherein the user indicates the current listening criteria or criterion via an indicator.
45. An auditory prosthesis for rehabilitating the hearing of a user, comprising: a microphone configured to receive sound and to produce a microphone signa] corresponding to the received sound; a sound processor configured to process the muicrophone signal using a set of variable processing factors to produce an output signal; and a sound analysis configured to receive the microphone signal, and to output a data set representative of the acoustic environment of the prosthesis; 1, wherein the prosthesis is configured to provide the user with the ability to adjust at least one of the variable processing factors of the set, and wherein one or more of the variable processing faictors are automatically adjusted by the prosthesis based on the adjustments of the at least one variable processing factor by the user,
46. 46. An auditory prosthesis according to claim 45, wherein the user operates a control means to adjust the at least one variable processing factor, wherein further the control mneans is adjustable throughout a substantially continuous range of settings and a Preferred setting is used to calculate the at least one variable processing factor which alters the processing operations applied to the microphone signal.
47. 47. An auditory prosthesis according to claim 46 wherein the at least one variable processing factor is calculated by defined mathematical relationships incorporating coefficients, and wherein the value of the coefficients are calculated to be those values that cause the variable processing factor(s) to best approximate the variable processing fecor(s) that occurred as a result of current and previous settings. COMS ID No: ARCS-218238 Received by iP Australia: Time (I-Pm) 15:03 Date 2008-12-23 23/12 '08 14:00 FAX 61386184199 F IE&C FB RICE CO [022 00 o 41
48. 48- An auditory prosthesis according to any one of claims 45 to 47, wherein the Prosthesis fuirther comprises a data memory configured to store said data set output by the sound analysis means and said at least one variable processing factor adjusted by Ri the user. INO
49. 49- An auditory prosthesis according to any one of claims5 45 to 48, wherein the one or more variable processing factors are automatically adjusted on the basis of the acoustic environents in which the at least one variable processing factor wasadut.
50. 50. Arn auditory prosthesis according to any one of claims 45 to 49, further adapted to output audio signals to a single ear of the user.
51. An auditory prosthesis according to any one of claims 45 to 49, further adapted to output audio signals to both ears of the user.
52. 52. An auditory prosthesis according to claim 48, wherein the prosthesis is configured provide the user with the ability to repeatedly adjust the at least one variable processing factor to monitor how the preference of the user for aL particular setting of the at least one variable processing factor changes with time or usage.
53. 53. An auditory prosthesis according to claim 52, wherein the number of adjustments made by the user is monitored by the prosthesis.
54. 54. An auditory prosthesis according to claim 53, wherein the user operates a control means to adjust the at least one variable processing factor.
55. An auditory prosthesis according to any one of claims 45 to 54, wherein the auditory prosthesis is a hearing aid.
56. 56. An auditory prosthesis according to claim 55, further comprising an output device configured to provide the output signals to the user, and wherein the output device is an earphone that receives the output signal of the sound processor and generates amplified sounds that are delivered into the ear of the user.
57. 57. An auditory prosthesis according to any one of claims 45 to 54, wherein the auditory prosthesis is an implanted hearing aid. COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 23/12 'Os 14:01 FAX 61386184199 EB RICE CO Q023 00 o 42
58. 58. An auditory prosthesis according to claim 57, further comprising an output device, 0 and wherein the output device is a vibrating niechanism mechanically coupled to the Cfl middle or inner ear. IND
59. 59. An auditory prosthesis according to any one of claims 45 to 54, wherein the auditory prosthesis is a cochlear implant.
60. An auditory prosthesis according to claim 59, further comprising an output device, ci 10 and wherein the output device comprises a receiver/stinaulator unit that receives 0 encoded stimulation data from the sound processor and outputs stimulation signals that ci are delivered to a cochlea of the user through an electrode array.
61. 61. An auditory prosthesis according to claim 60, wherein the sound processor is configured to use a coding strategy to extract speech from the sounds detected by the microphone before or after processing of the microphone signal according to the set of variable processing factors.
62. 62. An auditory prosthesis according to claim 61, wherein the souand processor performs an audio spectral analysis of the microphone signal to generate output channel amplitude levels.
63. 63. An auditory prosthesis according to claim 62, wherein the transformation from the microphone signal to output channel amplitude levels is based on the set of variable processing factors.
64. 64. An auditory prosthesis according to claim 63, wherein the channel amplitude levels include a threshold level and discomfort level that are based on previous adjustments of the at least one variable processing factor made by the user. An auditory prosthesis according to any one claims 60 to 64, wherein the receiver/Stiuatr unit is positioed within a housing that is imuplantable within the user.
65. COMS ID No: ARCS-218238 Received by IP Australia: Time (H-im) 15:03 Date 2008-12-23 23/12 '08 14:01 FAX 61386184199 F IE&C FB RICE CO [024 00
66. 66- An auditory prosthesis according to claim 65, wherein the housing for the receiver/stimulator unit is imnplantable within a recess in the bone behind the ear 0 Posterior to the mastoid.
67. 67. An auditory prosthesis according to any one of claims 61 to 66, wherein the sound INO Processor encodes the sound detected by the microphone into a sequence of electrical stimuli following given algorithms.
68. 68. An auditory prosthesis according to claim 67, wherein the encoded sequence is transferred to the implanted receiver/stimulator unit using a transmitter a coil and a o receiver coil of the receiver/stimulator unit
69. 69. A method of rehabiliating the hearing of a user using an auditory prosthesis, comprising:, convening a received sound to a microphone signal; processing the microphone signal at a sound processor to generate an output signal utilizing a set of variable processing factors, receiving from the user an adjustment of at least one of the variable processing factors of the set; and automatically adjusting one or more of the variable processing factors based on' the received adjustments of the at least one variable processing factor by the user.
70. The method of claim 69, wherein the step of receiving an adjustment from the user includes: providing the user with a control means, the control means allowing adjustment of the at least one variable processing factor over a substantially continuous range of settings allowing the user to adjust the at least one variable processing factor to a preferred setting.
71. 71. The method of claim 70, wherein the step of automatically adjusting one or more of the variable processing factors includes: calculating the at least one variable processing factor using defined mathematical relationships incorporating coefficients; COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23 23/120 08 14:01 FAX 61386184199 FB RICE CO U025 00 Wherein the coefficients are determined so as to cause the variable Processing fac-tor(s) to approximate the variable processing factor(s) that have occurred as a result 0 Of pre'vious and current Preferred settings selected by the user.
72. 72. A method of rehabilitating the hearing of a user using an auditory prosthesis, IND comprising: Converting a received sound to a microphone signal; o processing the microphone signal at a sound processor to generate an output signal utilizing a set of variable processing factors, .1-1.0 receiving the microphone signal at a sound analysis unit that subsequently outputs a data set representative of the acoustic environment of the prosthesis; receiving from the user an adjustment of at least one of the variable processing factors of the set; and automatically adjusting one or more of the variable processing factors based on the received adjustment of the at least one variable processing factor by the user.
73. 73. The method of claim 72, wherein the step of receiving an adjustment from the user includes: providing the user with a control m-eans, the control means allowing adjustmnent of the at least one variable processing factor over a substantially continuous range oif settings allowing the user to adjust the at least one variable processing factor to a Preferred setting. 23
74. 74. The method of claim 73, wherein the step of automatically adjusting one or more of the variable processing factors includes: calculating the at least one variable processing factor using defined mathematical relationships incorporating coefficients; wherein the coefficients are determined so as to c'ause the variable processing factor(s) to approximate the variable processing factor(s) that have occurred as a result of Previous and current preferred settings selected by the user. An auditory prosthesis for rehabilitating the hearing of a user substantially as hereinbefore described with reference to the accompanying drawings.
75. COMS ID No: ARCS-218235 Received by IP Australia: Time (I-Pm) 15:03 Date 2008-12-23 23/12 '08 14:01 FAX 61386184199 FB RICE CO a 026 00 0 0 'O
76. 76. A method of rehabilitting the hearing of a user using an auditory prosthesis substantially as hereinbefore described with reference to the acvompanying drawings. cN VO ci COMS ID No: ARCS-218238 Received by IP Australia: Time 15:03 Date 2008-12-23