阅读说明：本技术 具有带宽归一化的内侧橄榄耳蜗反射声编码 (Inside olive cochlear reflex acoustic coding with bandwidth normalization ) 是由 恩里克·亚历杭德罗·洛佩兹-波韦达 于 2018-06-28 设计创作，主要内容包括：描述了一种用于双侧听力植入系统中的信号处理的信号处理装置。通道压缩模块基于通道归一化的内侧橄榄耳蜗反射模型使用通道特定动态抑制调整来为每个带通信号产生经抑制调整的带通信号,所述通道归一化的内侧橄榄耳蜗反射模型反映相应的对侧带通信号的带宽能量和选择的参考对侧带通信号的带宽能量。(A signal processing arrangement for signal processing in a bilateral hearing implant system is described. The channel compression module generates a suppression-adjusted band pass signal for each band pass signal using channel-specific dynamic suppression adjustment based on a channel normalized medial olive cochlear reflex model that reflects the bandwidth energy of the corresponding contralateral band pass signal and the bandwidth energy of the selected reference contralateral band pass signal.)

1. A signal processing system for signal processing in a bilateral hearing implant system having left and right side hearing implants, the system comprising for each hearing implant:

at least one sensing microphone configured for sensing a sound environment to form a respective microphone signal output;

a filter bank configured to process the microphone signals to generate a plurality of band pass signals, wherein each band pass signal represents an associated audio frequency band;

a channel compression module configured to form a rejection-adjusted band pass signal for each band pass signal using channel-specific dynamic rejection adjustment based on a channel-normalized medial cochlear reflex model reflecting the bandwidth energy of the corresponding contralateral band pass signal and the bandwidth energy of the selected reference contralateral band pass signal;

a pulse timing and encoding module configured to process the suppression-adjusted band pass signals to form stimulation timing signals; and

a pulse generation module configured to process the stimulation timing signals to form electrode stimulation signals for the hearing implant for perception as sound.

2. The system according to claim 1, wherein the medial olive cochlear reflex model is configured to produce channel-specific dynamic suppression adjustments of equal or greater for lower frequency band pass signals.

3. The system of claim 1, wherein the channel compression module is configured to use a channel-specific dynamic throttling adjustment function

4. The system of claim 3, wherein x and y vary within the interval [0, 1 ].

5. The system according to claim 3, wherein the dynamically determined suppression factor c is inversely related to the bandwidth energy of the respective contralateral band pass signal such that the larger the bandwidth energy of the respective contralateral band pass signal, the smaller the value of the dynamically determined suppression factor c.

6. The system according to claim 1, wherein the bandwidth energy of the respective opposite side band pass signals is based on having two time constants τ_aAnd τ_bThe rms output amplitude integrated over the previous exponential decay time window.

7. The system of claim 1, wherein the pulse timing and encoding module is configured to form stimulation timing signals using Continuous Interleaved Sampling (CIS) or channel-specific sampling sequence (CSSS) based or Fine Structure Processing (FSP) based encoding strategies, or a combination thereof.

8. A method for signal processing in a bilateral hearing implant system having left and right side hearing implants, the method comprising, for each hearing implant:

sensing a sound environment with at least one sensing microphone to form a respective microphone signal output;

processing the microphone signals with a filter bank to generate a plurality of band pass signals, wherein each band pass signal represents an associated audio frequency band;

forming a rejection-adjusted band pass signal for each band pass signal using a channel-specific dynamic rejection adjustment based on a channel normalized inside olive cochlear reflex model reflecting the bandwidth energy of the corresponding contralateral band pass signal and the bandwidth energy of the selected reference contralateral band pass signal;

processing the suppression-adjusted band pass signals with a pulse timing and encoding module to form stimulation timing signals; and

processing the stimulation timing signals with a pulse generation module to form electrode stimulation signals for the hearing implant.

9. The method of claim 8, wherein the medial olive cochlear reflex model is configured to produce equal or greater channel-specific dynamic suppression adjustments for lower frequency band pass signals.

10. The method of claim 8, wherein forming the rejection adjusted band pass signal comprises using a channel specific dynamic rejection adjustment function

11. The method of claim 10, wherein x and y vary within the interval [0, 1 ].

12. The method according to claim 10, wherein the dynamically determined suppression factor c is inversely related to the bandwidth energy of the respective contralateral band pass signal such that the larger the bandwidth energy of the respective contralateral band pass signal, the smaller the value of the dynamically determined suppression factor c.

13. The method of claim 8, wherein the bandwidth energy of the respective opposite side band pass signals is based on having two time constants τ_aAnd τ_bThe rms output amplitude integrated over the previous exponential decay time window.

14. The method of claim 8, wherein the pulse timing and encoding module forms the stimulation timing signal using Continuous Interleaved Sampling (CIS) or channel-specific sampling sequence (CSSS) based or Fine Structure Processing (FSP) based encoding strategies, or a combination thereof.

Technical Field

The present invention relates to hearing implant systems, and in particular to techniques for generating electrical stimulation signals in such systems.

Background

As shown in fig. 1, a normal ear sends sound through outer ear 101 to tympanic membrane 102, which tympanic membrane 102 displaces the bones of middle ear 103 (malleus, incus and stapes) which vibrate the oval window and circular window opening of cochlea 104. Cochlea 104 is a long, narrow duct that is helically wound about the axis of the duct in approximately two and one-half turns. It comprises an upper channel called the scala vestibuli and a lower channel called the scala tympani, connected by the cochlear duct. The cochlea 104 forms an upright spiral cone with a center called the cochlear axis in which the spiral ganglion cells of the acoustic nerve 113 reside. In response to receiving sound transmitted by the middle ear 103, the fluid-filled cochlea 104 acts as a transducer to generate electrical impulses that are transmitted to the cochlear nerve 113, and ultimately to the brain.

Hearing is impaired when there is a problem with the ability to convert external sounds into meaningful action potentials along the neural matrix of the cochlea 104. To improve impaired hearing, hearing prostheses have been developed. For example, when the impairment is related to the operation of the middle ear 103, a conventional hearing aid may be used to provide mechanical stimulation to the hearing system in the form of amplified sound. Or when the injury is associated with the cochlea 104, a cochlear implant with an implanted stimulation electrode can electrically stimulate the auditory nerve tissue with small currents delivered through multiple electrode contacts distributed along the electrode.

Fig. 1 also shows some components of a typical cochlear implant system, including an external microphone that provides an audio signal input to an external signal processor 111, in which different signal processing schemes may be implemented, and the processed signal is then converted into a digital data format, such as a sequence of data frames, for transmission to the implant 108. In addition to receiving the processed audio information, the implant 108 also performs additional signal processing such as error correction, pulse formation, etc., and produces stimulation patterns (based on the extracted audio information) that are sent through the electrode leads 109 to the implanted electrode array 110.

Typically, the electrode array 110 includes a plurality of electrode contacts 112 on its surface that provide selective stimulation of the cochlea 104. Depending on the context, the electrode contacts 112 are also referred to as electrode channels. In cochlear implants today, relatively few electrode channels are each associated with a relatively wide frequency band, each electrode contact 112 addressing a group of neurons with electrical stimulation pulses having a charge derived from the instantaneous amplitude of the signal envelope within that frequency band.

In some encoding strategies, stimulation pulses are applied across all electrode channels at a constant rate, while in other encoding strategies, stimulation pulses are applied at a channel-specific rate. Different specific signal processing schemes may be implemented to generate the electrical stimulation signals. Signal processing methods well known in the cochlear implant field include sequential interleaved sampling (CIS), channel specific sampling sequences (CSS) (as described in U.S. patent No.6,348,070, incorporated herein by reference), Spectral Peak (SPEAK), and Compressive Analog (CA) processing.

Fig. 2 shows the main functional blocks in a typical cochlear implant signal processing system, where band pass signals are processed and encoded to generate electrode stimulation signals to stimulation electrodes in an implanted cochlear implant electrode array. For example, a commercially available Digital Signal Processor (DSP) may be used to perform speech processing according to the 12-channel CIS method. The initial acoustic audio signal input is produced by one or more sensing microphones, which may be omnidirectional and/or directional. The filter bank 201 pre-processes the initial acoustic audio signal using a plurality of band pass filter banks, each associated with a particular audio frequency band-e.g. a digital filter bank having 12 butterworth band pass filters of the order 6 Infinite Impulse Response (IIR) type-such that the acoustic audio signal is filtered into a number M of band pass signals B₁To B_MWherein each signal corresponds to a frequency band of one of the band-pass filters. Each output of the CIS band pass filter can be roughly viewed as a sinusoid modulated by the signal envelope at the center frequency of the band pass filter. This is due to the quality factor of the filter (Q ≈ 3). In the case of a spoken speech segment, the envelope is approximately periodic and the repetition rate is equal to the pitch frequency. Alternatively and without limitation, filter bank 201 may be implemented based on the use of a Fast Fourier Transform (FFT) or a Short Time Fourier Transform (STFT). Each electrode contact within the scala tympani is typically associated with a specific band pass filter of an external filter bank, based on the tensioned tissue of the cochlea.

Fig. 3 shows an example of a short time period of an audio speech signal from a microphone, and fig. 4 shows an acoustic microphone signal decomposed into a set of signals by band-pass filtering by a filter bank. An example of pseudo code for an Infinite Impulse Response (IIR) filter bank based on a direct form II transpose structure is described by Fontaine et al, Brian hearts: online audio Processing Using vector Over Channels (online hearing Processing Using speech Vectorization), Frontiers in Neuroinformatics, 2011 gives; which is incorporated herein by reference in its entirety:

band pass signal B₁To B_M(which may also be considered as frequency channels) are input to a signal processor 202, which signal processor 202 extracts signal specific stimulation information, e.g., envelope information, phase information, timing of requested stimulation events, etc., into N stimulation speech signals S₁To S_NThe N stimulation channel signals representing electrode-specific desired stimulation events. For example, channel specific sampling sequences (CSSs) may be used as described in U.S. patent 6,594,525, which is incorporated herein by reference in its entirety. For example, envelope extraction may be performed using 12 rectifiers and 12 second order IIR type digital butterworth low pass filters.

The pulse timing and encoding module 203 applies a non-linear mapping function (pass) to the amplitude of each band pass envelopeOften logarithmic). This mapping function, e.g. using the instantaneous non-linear compression of the envelope signal (mapping law), is usually adapted to the needs of the individual cochlear implant user during installation of the implant to achieve natural loudness growth. This may be a stimulus event signal S reflecting patient-specific perceptual characteristics applied to each request₁To S_NTo generate a set a of electrode stimulation signals providing an optimal electrical representation of the acoustic signal₁To A_M. A logarithmic function with a shape factor c is typically available as a loudness mapping function, which is typically the same across all band pass analysis channels. In different systems, different specific loudness mapping functions other than logarithmic functions may be used, with only one identical function applied to all channels, or a separate function for each channel, to produce the electrode stimulation signal a output from the pulse timing and encoding module 203₁To A_M。

The pulse generation module 204 applies the set of electrode stimulation signals A₁To A_MSet of output electrode pulses E formed as electrode contacts in an electrode array for implantation₁To E_MThe electrode contacts are used to stimulate adjacent neural tissue. Output electrode pulse E₁To E_MMay be symmetrical biphasic current pulses with an amplitude directly obtained from the compressed envelope signal.

In the particular case of CIS systems, the stimulation pulses are applied in a strictly non-overlapping order. Thus, as a typical CIS feature, only one electrode is active at a time, with a higher overall stimulation rate. For example, assuming a total stimulation rate of 18kpps, where kpps represents 1000 stimulation pulses per second, and a 12-channel filter bank, the stimulation rate per channel is 1.5 kpps. Such a per-channel stimulation rate is usually sufficient to make a sufficient time representation of the envelope signal. The maximum total stimulation rate is limited by the minimum phase duration per pulse. The phase duration cannot be arbitrarily short because the shorter the pulse, the higher the current amplitude of the action potential exciting the neuron, and for various practical reasons, the current amplitude is limited. For a total stimulation rate of 18kpps, the phase duration is 27 μ s near the lower limit.

In the CIS strategy, the signal processors use only the band pass signal envelopes for further processing, i.e. they contain the entire stimulation information. For each electrode channel, the signal envelope is represented as a biphasic pulse train at a constant repetition rate. The performance characteristics of CIS are that the stimulation rate of all electrode channels is equal and has no relation to the center frequency of the individual channels. Typically, the pulse repetition frequency is not a temporal cue for the patient (i.e., it should be high enough that the patient does not perceive a tone having a frequency equal to the pulse repetition frequency). The pulse repetition frequency is typically chosen above twice the envelope signal bandwidth (based on the nyquist theorem).

Another Cochlear implant stimulation strategy that delivers fine temporal structural information is the Fine Structure Processing (FSP) strategy of Med-El tracking the zero crossings of the band pass filtered temporal signal and initiating a Channel Specific Sampling Sequence (CSSS) at each negative to positive zero crossing typically the CSSS sequence is applied only to the first one or two apical-most electrode channels covering the frequency range up to 200 or 330 Hz. in hochmai I, Nopp P, Jolly C, schmidt m, Sch β er H, Garnham C, Anderson I, Med-El Cochlear Implants: State of the Art and Glimpse inter the Future, Trends in amplification, volume 10, 201-219, FSP arrangement is further described.

In addition to the specific processing and encoding methods discussed above, different specific pulse stimulation modes are also possible to deliver stimulation pulses and phased array stimulation using specific electrodes, i.e., monopolar, bipolar, tripolar, multipolar, and phased array stimulation. And also different stimulation pulse shapes, i.e. biphasic, symmetric triphasic, asymmetric triphasic or asymmetric pulse shapes. These different pulse stimulation modes and pulse shapes each provide different benefits, e.g., higher tone selectivity, smaller electrical threshold, higher electrical dynamic range, fewer undesirable side effects such as facial nerve stimulation, etc. Some stimulation arrangements are power hungry, particularly when the adjacent electrodes are used as current sinks (sinks). Up to 10 db of charge may be required compared to a simple single electrode stimulation concept (if the power dissipation pulse shape or stimulation mode is used continuously).

Bilateral stimulation has long been used in hearing aids, but has not until recently been commonly used in hearing implants such as Cochlear Implants (CI). For cochlear implants, binaural stimulation requires a bilateral implant system with two implanted electrode arrays, one in each ear. The left and right side acoustic signals to be entered are similar to those in hearing aids and may simply be the output signals of microphones located near the left and right ears, respectively. Bilateral cochlear implants provide the benefit of bilateral hearing that can allow a listener to locate a sound source in a horizontal plane. It is well known that bilateral hearing also makes speech easier to understand in noise. For example, it is more fully explained In Bronkhorst, A.W. And Pcomp, R., The Effect Of Head-induced interaural Time And Level Differences On Speech Intelligibility In Noise, J.Acoust.Soc.Am.83,1508-1516,1988, which are incorporated herein by reference.

It is well known that for natural hearing, the central nervous system controls sound coding in the cochlea through the medial cochlear (MOC) efferent nerves. MOC efferents innervate the outer hair cells of the cochlea, whose activation inhibits movement of the basal lamina for low and moderate sounds. This inhibition restores the dynamic range of the individual hearing nerve fibers in the noise and may increase the proportion of nerve fibers that fire within their dynamic range. This may improve the neural representation of transient sound features and thus contribute to sound source localization in speech recognition and noise. For more details, reference is made to Guinan JJ., Cochleareferenergation and function (cochlear efferent innervation and function) Current, Opin, Otolaryngol, Head & neutral Surgery, 18(2010), page 447-. It is also possible to amplify the spatial release of masking, see examples Kim s.h., Frisina r.d., Frisina D.R, Effects of age on Speech understating in normal hearing in noise (effect of age on Speech understanding of normal hearing listeners: Relationship between hearing efferent nervous system and Speech understanding in noise.) Speech Communication (2006)48,862.

CI users have a greater difficulty understanding speech in noise than normal-hearing listeners. This may be due in part to their lack of anti-interrogation benefits of medial olive cochlear reflex (MOC), and the ability of CI users to improve hearing in noise using sound processing strategies that simulate the effect of MOC on compression. See patent cooperative treaty publication WO 2015/169649; lopez Poveda et al, loops of the relative efferent reflex in the ear established with the cochlear implants (the role that cochlear implants show in hearing of the contralateral efferent reflex), adv.exp.med.biol.894,105-114,2016; and Lopez-Poveda et al, A bipolar sound encoding linear encoded by the symmetric enteric encoding reflex (bilateral acoustic encoding strategy inspired by the contralateral medial olive cochlea reflex), Ear ear.37, e138-e148,2016; all of which are incorporated herein by reference in their entirety. On the other hand, the ability of a CI user to recognize speech in noise can be considered to depend both on the effective speech-to-noise ratio (SNR) at the output of the CI audio processor and the sensitivity of the CI user to the corresponding electrical stimuli.

Disclosure of Invention

Embodiments of the present invention relate to a signal processing arrangement and a corresponding method for a bilateral hearing implant system with left and right side hearing implants. Each hearing implant includes at least one sensing microphone configured to sense a sound environment to form a respective microphone signal output. The filter bank is configured for processing the microphone signals to generate a plurality of band pass signals, wherein each band pass signal represents an associated audio frequency band. The channel compression module is configured to form a suppression-adjusted band pass signal for each band pass signal using channel-specific dynamic suppression based on a channel-normalized medial olive cochlear reflex model reflecting the bandwidth energy of the corresponding contralateral band pass signal and the bandwidth energy of the selected reference contralateral band pass signal. The pulse timing and encoding module is configured to process the suppression-adjusted band pass signals to form stimulation timing signals. The pulse generation module is configured to process the stimulation timing signals to form electrode stimulation signals for the hearing implant to be perceived as sound.

In further embodiments, the medial olive cochlear reflex model may be specifically configured to produce greater channel-specific dynamic suppression adjustments for lower frequency band pass signals. Channel suppression modules may be configured to use channel specific

Where x represents the amplitude of the input bandpass signal, y represents the amplitude of the adjusted bandpass signal, and c is a dynamically determined suppression factor that determines the amount of suppression adjustment. In these embodiments, x and y may be in the interval [0, 1]]The internal variation, and/or the dynamically determined suppression factor c and the bandwidth energy of the corresponding contralateral band pass signal may be inversely related such that the larger the bandwidth energy of the corresponding contralateral band pass signal, the smaller the value of the dynamically determined suppression factor c. The bandwidth energy of the corresponding opposite side band pass signal can be based on the two time constants τ_aAnd τ_bThe integrated rms output amplitude over the previous time-exponential decay time window. And the pulse timing and encoding module may be configured to form the stimulation timing signal using Continuous Interleaved Sampling (CIS) or channel-specific sampling sequence (CSSS) -based or Fine Structure Processing (FSP) -based encoding strategies, or a combination thereof.

Drawings

Fig. 1 shows a cross-sectional view of a human ear utilizing a typical cochlear implant system designed to deliver electrical stimulation to the inner ear.

FIG. 2 illustrates different functional blocks in a Continuous Interleaved Sampling (CIS) processing system.

Fig. 3 shows an example of a short period of time of an audio speech signal from a microphone.

Fig. 4 shows an acoustic microphone signal decomposed into a set of signals by band-pass filtering of a filter bank.

Fig. 5 shows different functional blocks in a system for signal processing according to an embodiment of the invention.

Fig. 6 shows different logic blocks in a method for signal processing according to an embodiment of the invention.

Detailed Description

Previous signal processing apparatuses using the Medial Olive Cochlear Reflex (MOCR) model are based on a suppression parameter c for each band pass channel, which depends on the output energy E from the corresponding contralateral band pass channel. For a wideband signal, the resulting contralateral rejection pair may be larger for the higher frequency channels than for the lower frequency channels, since the higher frequency channels are wider in frequency and may include more energy than the lower frequency channels. Embodiments of the present invention address the problem of binaural CI sound coding using channel-specific dynamic suppression adjustment based on a channel-normalized MOCR model reflecting the bandwidth energy of the corresponding contralateral band-pass signal and the bandwidth energy of the selected reference contralateral band-pass signal.

Fig. 5 shows different functional blocks in the system, and fig. 6 shows different logic blocks in a method of signal processing for a bilateral hearing implant system with a left-side hearing implant 500 and a right-side hearing implant 501 according to an embodiment of the invention. The system shown in fig. 5 is based on the system of fig. 2 discussed above with respect to CIS-based vocoding. Thus, initially on each side there is at least one sensing microphone configured for sensing the sound environment to form a corresponding microphone signal output, step 601, which is processed by the filter bank 201, step 602, to generate a plurality of band pass signals B₁To B_M(which may also be considered frequency channels), each representing an associated audio frequency band. For example, the filter bank 201 may specifically comprise a high-pass pre-emphasis filter such as a first order Butterworth filter with a 3-dB cutoff frequency of 1.2KHz, followed by 12 sixth order Butterworth band-pass filters with a 3-dB cutoff frequency following a modified logarithmic distribution between 100 and 8500 Hz.

The signal processor channel compression module 502 then processes the bandpass signal B as described above with respect to FIG. 2₁To B_M(e.g., using a 3-dB fourth-order Butterworth low pass with a cut-off frequency of 400HzA filter that performs envelope extraction via full-wave rectification and low-pass filtering) and additionally performs channel-specific dynamic logarithmic suppression adjustment, step 603, which produces an adjusted band-pass signal S based on a channel-normalized medial olive cochlear reflex model, as discussed more fully below₁To S_N。

The pulse timing and encoding module 203 then applies pulse encoding (e.g., using a Continuous Interleaved Sampling (CIS) encoding strategy) and a nonlinear mapping function as described above, step 604, to the adjusted bandpass signal S₁To S_NTo generate an electrode stimulation signal set A₁To A_MThe electrode stimulation signals are formed by the pulse generation module 204 into output electrode pulses E for the electrode contacts in the implanted electrode array₁To E_MStep 605.

In step 603, the MOC module 503 directs the band pass signals B from each filter bank 201₁To B_MThe channel normalized medial olive cochlear reflex model is applied so that the signal processor channel compression module 502 performs channel-specific dynamic suppression adjustments, which may be particularly configured to produce larger channel-specific dynamic suppression adjustments for lower frequency band pass signals. This may be based on, for example, a channel-specific dynamic suppression adjustment function

Where x represents the amplitude of the input bandpass signal, y represents the amplitude of the adjusted bandpass signal, and c is a dynamically determined suppression factor that determines the amount of suppression adjustment. For more details, reference is made to the discussion in Boyd PJ, Effects of programming and mapping on acoustic thresholds and speed characterization with the MED-EL COMBI 40+ cochlear inplant (the effect of programming thresholds and maplaw settings on MED-EL-COMBI 40+ cochlear implant acoustic thresholds and speech recognition), incorporated herein by reference in its entirety, Ear Hear.27,608-618,2006. In another example, the channel-specific dynamic suppression adjustment function y ═ ax^p+ k, where x represents the amplitude of the input bandpass signal, y represents the amplitude of the adjusted bandpass signal, k is a constant, and pIs a dynamically determined suppression factor that determines a suppression adjustment amount. However, it should be understood that the present invention is not limited to these two dynamic suppression adjustment functions, and that other suitable functions may be used. More formally, the rejection factor c (or p) is controlled using the output energy E normalized to the channel bandwidth E':

wherein BW is the channel bandwidth and BW_refIs the reference channel bandwidth.

In an exemplary embodiment, x and y may vary within the interval [0, 1 ]. The dynamically determined suppression factor c and the bandwidth energy of the corresponding contralateral band pass signal may be inversely correlated such that the larger the bandwidth energy of the corresponding contralateral band pass signal, the smaller the value of the dynamically determined suppression factor c. For example, the relationship between the instantaneous value of c and the instantaneous contralateral output energy may be such that the larger the output energy, the smaller the value of c. In one example, the relationship may be given by:

in the formula, c_a、c_bAnd β are constants and E' is a normalized instantaneous output energy E for a given time t, as described in more detail in patent Cooperation treaty publication WO2015/169649 (incorporated herein by reference.) in another example, this relationship can be given by:

in the formula, c_a、c_bAnd β are constants and E' is the normalized instantaneous output energy E for a given time t, however, it should be understood that the invention is not limited to these two functions, and that other suitable functions may be used

In particular, c may vary between about 30 and 1000 for 0 and-20 dB full scale contralateral output energies, respectively (FS: where 0dB FS corresponds to the peak amplitude at unity).

Based on the exponential temporal processes of MOC effect activation and deactivation (see, e.g., Backus and Guinan, Time-course of the human medial olyvocochlear reflexes, J.Acoust.Soc.Am.119,2889-2904,2006, which are all incorporated herein by reference), the bandwidth energy of the corresponding contralateral band pass signal may be based on a signal having two Time constants τ_aAnd τ_bThe rms output amplitude integrated over the previous exponential decay time window. For example, to reflect the time course of activation and deactivation of the natural MOCR, the time constant may be set to τ_a2ms and τ_b＝300ms。

For example, in the embodiment where lane #1 and lane #12 are the lowest and highest frequencies, respectively, when BW is measured_refEqual to the bandwidth of channel #12(BW #12), the overall suppression will be maximum and gradually decrease for lower numbers of channels. Bandwidth normalization, which produces greater rejection, can compromise audibility and reduce intelligibility. Furthermore, for normal hearing listeners, it was observed that contralateral broadband noise at a 60dB sound pressure level increased the hearing threshold by approximately 1 to 9 dB. The inventors have observed that the use of BW_refNormalization to BW #6, BW #7, or BW #8 yields a more reasonable overall suppression, with the effective suppression in the lower frequency channel being equal to or greater than the suppression in the higher frequency channel without significantly compromising audibility.

Embodiments of the present invention may be implemented, in part, in any conventional computer programming language. For example, the preferred embodiments may be implemented in a procedural programming language (e.g., "C") or an object oriented programming language (e.g., "C + +", Python). Alternative embodiments of the invention may be implemented as pre-programmed hardware elements, other related components, or combinations of hardware and software components.