EP1453355A1 - Signal processing in a hearing aid - Google Patents

Abstract

The arrangement has a filter (6) for frequency-dependent amplitude input signal (X) matching and arrangements for adapting filter coefficients (cM) according to the input signal, determining coefficients of a compression gain (g m) describing frequency-dependent amplitude input signal matching according to frequency-dependent signal levels and determining coefficients of a noise suppression (a m) describing frequency-dependent amplitude matching according to input signal noise. The arrangement has a filter for frequency-dependent amplitude matching of an input signal and arrangements for adapting filter coefficients according to the input signal, determining coefficients of a compression gain describing frequency-dependent amplitude matching of an input signal according to frequency-dependent signal levels and determining coefficients of a noise suppression describing frequency-dependent amplitude matching of an input signal according to noise detected in the input signal. The arrangement for adapting filter coefficients determines them from the coefficients of the compression gain and noise suppression. An independent claim is also included for the following: (a) a hearing aid with an arrangement for implementing the inventive method.

Description

The invention relates to a device and a method for signal processing in a hearing aid according to the preambles of the independent claims. The The invention is particularly suitable for improving speech intelligibility by suppressing noise from hearing aids or hearing aids.

STATE OF THE ART

A generic method is known for example from EP 1 067 821 A1, the content of which is hereby incorporated into this application. There will be one Hearing aid described in which a suppression of noise in one Input signal takes place in a main signal path, which is neither a transformation in the frequency range is still divided into subband signals, but only one Suppression filter. A transfer function of the suppression filter is periodically redetermined based on weakening factors that are signal analysis path lying parallel to the main signal path can be determined. The Attenuation factors are used to attenuate signal components in Frequency bands with a significant proportion of noise are used. The Suppression filter is implemented as a transversal filter, its impulse response periodically as a weighted sum of the impulse responses of transverse ones Bandpass filtering is recalculated. In this way, processing with low signal delay possible.

PRESENTATION OF THE INVENTION

It is an object of the invention, an apparatus and a method for signal processing to create in a hearing aid of the type mentioned, which a Realize higher quality and intelligibility of the processed signal.

This object is achieved by a device and a method for signal processing in a hearing aid with the features of claims 1 and 10 and a Hearing aid with the features of claim 20.

• werden Koeffizienten einer Kompressionsverstärkung, welche eine frequenzabhängige Anpassung des Eingangssignals nach Massgabe von frequenzabhängigen Signalpegeln des Eingangssignals beschreiben, bestimmt,
• werden Koeffizienten einer Geräuschunterdrückung, welche eine frequenzabhängige Anpassung des Eingangssignals nach Massgabe von im Eingangssignal detektierten Störgeräuschen beschreiben, bestimmt, und
• werden Koeffizienten eines Filters zur Filterung des Eingangssignals aus den Koeffizienten der Kompressionsverstärkung und den Koeffizienten der Geräuschunterdrückung berechnet..

In the method according to the invention for signal processing in a hearing aid

• coefficients of a compression gain, which describe a frequency-dependent adaptation of the input signal in accordance with frequency-dependent signal levels of the input signal, are determined,
• coefficients of noise suppression, which describe a frequency-dependent adaptation of the input signal in accordance with interference noises detected in the input signal, are determined, and
• coefficients of a filter for filtering the input signal are calculated from the coefficients of the compression gain and the coefficients of the noise suppression.

The term "adaptation of a signal" summarizes both one Reinforcement as well as a weakening is meant.

The invention makes it possible to adjust the amplitude response of the filter changing voice and interference signals as well as the needs of one adjust hearing impaired person, with a delay time for filtering of the input signal is kept small.

Another advantage is that the compression gain is different Gain values for different frequency ranges of the input signal.

Another advantage is that only one controllable filter can be used for both Compression amplification is used as well as noise suppression.

In a preferred embodiment of the invention, the coefficients of the compression gain are determined in a first set of frequency ranges F _n with n = 1..N of the input signal on the basis of signal levels or amplitude components. A signal level is determined from a partial signal of the input signal, which is formed by filtering the input signal and dividing it into partial signals with signal components in only one frequency range in each case. The signal levels are iteratively determined as instantaneous effective values of a signal power in the respective frequency ranges of the input signal. This makes it possible to track the compression gain with a temporal resolution that corresponds to a sampling rate of the input signal.

In a preferred embodiment of the invention, the coefficients a _{m of} the noise suppression are carried out in a second set of frequency ranges Φ _m with m = 1..M of the input signal by determining modulation depths d _m and by determining the coefficients a _m for each of the frequency ranges Φ _{m in} accordance with the corresponding modulation depth d _m . The modulation depths d _{m are determined} from a chronological order of maximum and minimum values of a signal level p _m in the respective frequency range Φ _m . This makes it possible to selectively filter out weakly modulated, i.e. monotonous, background noise. Time constants for adjusting the noise cancellation are preferably in the range of around 50 milliseconds or less.

In a preferred embodiment of the invention, the frequency ranges Φ _m for the noise suppression are small in comparison with the frequency ranges F _n for the compression gain. It therefore comprises at least one frequency range F _n two or more frequency ranges Φ _m . Accordingly, filters for determining portions of the input signal in the frequency ranges Φ _{m have} a longer signal delay or delay than filters for the frequency ranges F _n . This enables a sharp division of the frequency range to suppress interference and, at the same time, rapid adaptation of the compression gain to a changing speech signal. A maximum tolerable delay for the adaptation of coefficients of the compression gain is 5 milliseconds, values below 2.5 milliseconds are preferred. According to the invention, values of less than one millisecond can be achieved.

In a further preferred embodiment of the invention, the filter is not exactly tracked to the newly calculated coefficients in each sampling interval. Instead, it becomes only according to one or more changed coefficients tracked. This allows an adjustment with little computing effort and correspondingly low energy consumption. The adaptation preferably takes place only for the coefficient or coefficients whose change is a predetermined one Exceed the threshold or which is comparatively large or largest. A periodic change of one or a few is also possible Coefficients, or a pseudo-random iteration and adjustment of all Coefficients.

In a further preferred embodiment of the invention, an influence of the Noise suppression in the determination of the coefficients for the Compression gain taken into account. To do this, a means of Determination of noise suppression coefficients Determination of coefficients of the compression gain correction values, which is a signal attenuation caused by noise suppression correspond.

The device according to the invention has the features of claim 10 on. A hearing aid according to the invention has means for performing the method according to the invention.

Further preferred embodiments result from the dependent patent claims out. Features of the process claims are analogous to the Device claims can be combined and vice versa.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following, the subject matter of the invention is based on preferred exemplary embodiments, which are illustrated in the accompanying drawings.

Figur 1

schematisch eine Struktur der Signalverarbeitung;

Figur 2

ein Blockdiagramm einer Berechnung von Verstärkungswerten; und

Figur 3

ein Blockdiagramm einer Berechnung von Abschwächungswerten und Korrekturgrössen gemäss der Erfindung.

Show it:

Figure 1

schematically a structure of the signal processing;

Figure 2

a block diagram of a calculation of gain values; and

Figure 3

a block diagram of a calculation of attenuation values and correction quantities according to the invention.

The reference symbols used in the drawings and their meaning are in the list of reference symbols is summarized. Basically, in the Figures have the same parts with the same reference numerals.

WAYS OF CARRYING OUT THE INVENTION

FIG. 1 schematically shows a structure of the signal processing in a hearing device according to the invention. An input signal X is fed to a controllable filter 6, to a means for determining a compression gain 7 and to a means for determining a noise suppression 8. The controllable filter 6 is designed to form an output signal Y in accordance with filter coefficients c ₁ ..c _M.

The input signal X is fed to a first filter unit 1 by means of determining the compression gain 7. The first filter unit 1 is designed to determine signal components x ₁ ..x _{N of} the input signal X in a first set of frequency ranges F _n with n = 1..N. In signal processing for compression gain 3, parameters or coefficients or adaptation values of the compression gain g ₁ ..g _{M are} calculated from the signal components x ₁ ..x _N. With regard to the amplification function of the hearing aid, these coefficients are also referred to as amplification values. However, other coefficients are also referred to as gain values.

The input signal X is routed to a second filter unit 2 by means of determining the noise suppression 8. The second filter unit 2 is designed to determine signal components y ₁ ..y _{M of} the input signal X in a second set of frequency ranges Φ _m with m = 1..M. In signal processing for noise suppression 4, parameters or coefficients or adaptation values of the noise suppression a ₁ ..a _{M are} calculated from the signal components y ₁ ..y _M. These coefficients are also referred to as attenuation values in view of the noise suppression achieved with them.

The combination unit 5 combines the coefficients of the compression gain g ₁ ..g _M with the coefficients of the noise suppression a ₁ ..a _M and calculates combined logarithmic gain values c ₁ ..c _M as the filter coefficients of the controllable filter 6. The coefficients mentioned are preferably g _i , a _i and c _i logarithmically scaled and in the combination unit 5 essentially a subtraction c _m = g _m -a _m with m = 1..M is carried out.

In a preferred embodiment of the invention, the signal processing for noise suppression 4 transmits the signal processing for compression amplification 3 correction values r ₁ ..r _N , which correspond to a respective signal attenuation in the frequency ranges F ₁ ..F _n caused by the noise suppression.

In a further preferred embodiment of the invention, the first filter unit 1 and the second filter unit 2 are not implemented as separate units, but rather as a combined filter unit. For example, filtering with broad frequency bands is carried out sequentially to determine the signal components x ₁ ..x _N and these filtered signals are further filtered to determine the signal components y ₁ ..y _M.

The invention in the embodiment shown works in summary as follows: The input signal is divided into three signal paths, a main signal path with a controllable filter, a first parallel signal analysis path for the Compression gain and a second parallel signal analysis path for the Noise cancellation.

FIG. 2 shows a block diagram of a calculation of gain values in the signal processing for compression gain 3. Signal levels in N relatively few frequency ranges are calculated for the compression gain. FIG. 2 shows the calculation for one of these N frequency ranges; the same structure is used for the other frequency ranges. A signal power is formed in a block 21 from a signal component x _n in this frequency range, for example as a running sum of squared signal values. In block 22, a signal level p _{n is} formed by logarithmization. The term signal level here means the effective value of the instantaneous signal power in the frequency range F _n , expressed in a logarithmic number range, for example in dB. A modified signal level p _n 'is calculated from the signal level p _n by subtracting 23 a correction value r _n . The determination of correction values r _n is discussed separately below. At least one frequency range Φ _m of noise suppression is assigned to each frequency range F _{n of} the compression gain. A separate function f _{m is} predefined for each of these assigned frequency ranges Φ _m (in FIG. 2 there are three, corresponding to blocks 24, 24 ', 24 "), which calculates an amplification value g _m from the modified signal level p _n ' G m = f m (p n ')

These functions f _m take into account individual hearing loss and audiological experiences. Parameters, amplification values or hearing correction values contained in the functions f _m are preferably user-specific and are stored, for example, in an EPROM of the hearing device. The total number of these functions f _m and the gain values g _m , that is to say over all N frequency ranges F _{n of} the compression gain, is equal to the number M of the frequency ranges Φ _{m of} the noise suppression.

If the aim is to amplify quiet phonemes, i.e. consonants, in a speech signal more than loud phonemes, ie vowels, so that as far as possible all phonemes can be clearly heard in a continuously spoken language for a hearing impaired person, then the signal levels p _{n must} be determined in such a way that Differences between quiet and loud consecutive phonemes are well recorded. In addition, the continuously determined gain values g _m must be applied in time to those signal sections in which the associated phonemes are located, ie the gain values must act on the audio signal X synchronously. Such a synchronous compression amplification, which works so fast in the rhythm of successive phonemes, only gives good results if the number of separate frequency ranges is chosen to be small, e.g. N ≤ 5, preferably N ≤ 3. Otherwise characteristic spectral differences between the frequency ranges become too great for the different phonemes reduced and thus impaired speech intelligibility. Compression amplification with a few, relatively wide frequency ranges is possible with a small processing delay in the order of 1 millisecond, which ideally comes close to the desire for signal processing that is delay-free. In a preferred embodiment of the invention, the compression amplification is carried out jointly for only a single frequency band, that is to say for the entire frequency range of the audio signal. In another embodiment of the invention, two frequency bands are used for this, that is N = 2.

The signal analysis for determining signal levels in frequency ranges f _n for the compression gain is preferably carried out iteratively, current signal levels being determined for each new value of the input signal. Recursive signal analysis methods are preferably used for this. For example, the root mean square of the signal x [k] at the kth sampling time is iteratively as s [k] = s [k-1] + ε · (x 2 [k] - s [k-1]), calculated with 0 <ε << 1 selected.

A corresponding signal level value, eg in dB, then results in p [k] = 10 * log10 (s [k])

The purpose of noise suppression is to attenuate partial signals in frequency ranges of the audio signal in which there are mainly only monotonous noise. For this purpose, in M separate frequency ranges Φ _m differences between temporally successive maximum and minimum values of the signal levels p _m , so-called modulation depths d _m , are determined, where m = 1, ..., M applies.

An iterative determination of the signal levels in time with the sampling rate of the input signal is not necessary for noise suppression. In order to save arithmetic operations, reduced sampling rates are therefore preferably used. The signal level p _{m is formed} segment by segment for segments of a length of approximately 20-30 ms as the current effective value of the signal power in the corresponding frequency range Φ _m . The noise suppression can thus be tracked with a temporal resolution p _m, for example under 50 ms.

Separate are used to determine maximum and minimum values Estimation functions tracked: To do this, every sampling interval stored maximum value either linearly by a small increment or according to an exponential function, or it becomes the current level value accepted if it exceeds this reduced maximum value. Analogous the minimum value is increased by a small increment in each sampling interval or the current level value is adopted if it has the raised minimum value below. The modulation depth is the difference between these two Estimate sizes. A small modulation depth is created with the same Signal energy. To avoid sudden changes in the modulation depth, the difference values determined in this way are preferably still a smoothing subjected. By appropriate selection of the mentioned increments, the sound Extremes with time constants in the range of a few seconds.

For speech in a quiet acoustic environment, the modulation depth takes on values of 30 dB and more. In traffic noise, the low frequency range up to about 500 Hz is often dominated by monotonous noise, so that even when speech signals are present, the modulation depth in this frequency range drops to almost 0 dB. Other interfering noises in turn cover the speech signal in higher frequency ranges. Partial signals are preferably attenuated in frequency ranges Φ _m , in which the modulation depth d _m falls below a critical value of, for example, 15 dB, the extent of the attenuation a _m increasing monotonically and, for example, linearly as the modulation depth decreases.

For the most accurate detection and separation of frequency ranges with different modulation depth is a large number of separate frequency ranges advantageous, e.g. M = 20. For signal processing in so many narrow There is inevitably a long time delay in the frequency ranges Of the order of 10 ms, but with a gradual weakening and occasional raising of the partial signals in these frequency ranges well tolerated is.

The gain values g _{m of} the compression gain 3 and the attenuation values a _{m of} the noise suppression 4 are combined for each frequency range and supplied to the controllable filter 6 in the main signal path as control variables c _m . If necessary, the transfer function of the controllable filter is tracked frequency-specifically in one or a few frequency ranges in each sampling interval of the input signal and is left unchanged in all other frequency ranges.

For the combined use of compression amplification and noise suppression, it is possible to carry out a signal analysis in a relatively large number of frequency ranges Φ _m , as is useful for noise suppression, and then to summarize the results in a suitable manner with regard to the few frequency ranges F _n relevant for compression amplification. The disadvantage of such a sequential approach is that there is a long signal delay in the order of 10 ms for the entire signal processing. In terms of computational effort, the Fast Fourier Transform and the Inverse Fast Fourier Transform seem particularly attractive for such an implementation. The audio signal is successively transformed into individual segments with a duration of approx. 10 ms in the frequency domain, analyzed and modified, and then transformed back into the time domain. The use of segment-by-signal processing, however, has the following further disadvantages: The signal levels p _n are calculated as mean values in a segment, as a result of which a pronounced signal increase at a specific point in time is only recorded with the temporal resolution of a processing segment. The determination of the individual gain values and thus the entire transfer function is also carried out only in time with the successive segments.

Therefore, the filtering of the input signal X is preferably carried out on the basis of a separate and parallel signal analysis for noise suppression as well as for compression amplification. In this case, the coefficients a _m for noise suppression, which are obtained with a delay, are combined with coefficients g _m for compression gain which are obtained more quickly, and several of the coefficients g _m with different functions f _{m are} based on the same, optionally modified, signal level p _n '= p _n -r _n one Frequency range F _{n determined} for compression gain.

The combined and parallel processing takes place as follows: In the lowest signal path, the audio signal passes through a controllable filter 6, which carries out the required frequency-dependent signal modifications. The two upper signal paths each contain a filter unit, which divide the audio signal into partial signals of separate frequency ranges. The first filter unit 1 effects a signal division into only a few, N wide frequency ranges F _n , which can be carried out with a small signal delay. The second filter unit 2 effects a signal division into many, M narrow frequency ranges Φ _m , which results in a long delay time. The frequency ranges are preferably selected so that each frequency range Φ _{m is a} partial range of a frequency range F _n . The frequency ranges for compression gain F _n together preferably cover the same frequency range as the frequency ranges for noise suppression Φ _m . A frequency range for compression amplification covers several frequency ranges for noise suppression. Relationships between the widths of frequency ranges and between the division of frequency ranges are preferably at least approximately logarithmic.

A typical frequency range for the input signal is: 0 to 10 kHz. For example, this is divided into the following frequency ranges for compression gain and noise suppression: Compression gain (Hz) Noise Cancellation (Hz) 0 to 1250 0 to 312.5 312.5 to 625 625 to 937.5 937.5 to 1250 1250 to 2500 1250 to 1562.5 1562.5 to 1875 1875 to 2187.5 2187.5 to 2500 2500 to 10000 2500 to 3125 3125 to 3750 3750 to 4375 4375 to 5000 50000 to 6250 6250 to 7500 7500 to 10000

The sampling rate is 20 kHz, for example, and accordingly Usable bandwidth is half, i.e. 10 kHz. In another embodiment of the According to the invention, these values are 16 kHz and 8 kHz, respectively.

In the signal analysis for noise suppression, the assigned signal level p _m , the modulation depth d _m and the attenuation value a _{m are} determined for each of the M frequency ranges Φ _m , the latter advantageously being expressed in a logarithmic number range. As described above, the modulation depth d _m is determined in accordance with the specification, ie as a function of the time profile of the corresponding signal level p _m , and the coefficients a _{m are determined in} accordance with the corresponding modulation depth d _m . The second filter unit 2 and part of the signal processing for noise suppression 4 thus form a means for determining these variables p _m , d _m and a _m in a second set of frequency ranges of the input signal X.

In the signal analysis for compression amplification, the signal level p _{n is} determined in each of the N frequency ranges F _n in such a way that each signal value of the partial signal x _{n [} k] contributes to an update of the signal level, which leads to a higher temporal resolution than that of the mere one Determination of a segmental mean.

The first filter unit 1 and part of the signal processing for compression amplification 3 thus form a means for determining signal levels in a first set of frequency ranges of the input signal X. After that, amplification values become alle _m for all M frequency ranges G m = f m (p n ') is determined, each modified signal level p _n ', that is to say the level reduced by the correction values r ₁ ..r _N , being used to determine the gain values in all those frequency ranges Φ _{m which} , when combined, result in the frequency range F _n . The correction values r _n take into account any weakening of the signal powers as a result of the noise suppression.

Each of the gain values g _m with m = 1..M is therefore assigned to a frequency range Φ _m . By specifying M different gain values for the narrow frequency ranges Φ _m , the compression gain in the signal processing combined according to the invention can also be realized with a much more flexible transfer function, i.e. with M instead of only N functions f _m , than if only one gain value for each broad frequency range F _n would be set. The gain values g _m are in turn preferably expressed in a logarithmic number range. The functions f _m determine a desired frequency-specific gain according to audiological principles depending on the signal level.

The M gain and attenuation values arrive at the combination 5 of gains and attenuations, where they are combined separately in each frequency range Φ _m , which is done by simple subtraction when using a logarithmic number range:
c _m = g _m - a _m .

The M combined logarithmic gain values c _m arrive at the controllable filter 6, where they are transformed into linear gain values γ _m . The controllable filter 6 with transfer function H (z) can be composed of M parallel filters, the transfer functions H _m (z) of which only have a passband characteristic in the frequency range Φ _m and a blocking characteristic in all other frequency ranges and to achieve the desired frequency-dependent modification of the Audio signal X are each multiplied by the linear gain value γ _m H (z) = γ 1 · H 1 (z) + γ 2 · H 2 (z) + ..... + γ M · H M (Z).

For an update of the controllable filter 6 in time with the sampling rate Audio signal X, this elementary relationship is not suitable because of Computational effort and the associated power consumption of an integrated Circuit would be much too large. It is only suitable for a segment by segment Tracking, but what because of the reduced temporal resolution in the here exemplary embodiment shown is not optimal.

In order to achieve a better temporal resolution, the transfer function H (z) of the controllable filter 6 is preferably updated iteratively in each sampling interval according to H (z) [k] = H (z) [k - 1] + δH (z) [k], the quantity δH (z) [k] stands for the exact updating of the controllable filter 6 in one or at most a few frequency ranges Φ _m . In the case of updating in a single frequency range Φ _{m, the} following applies δH (z) [k] = (γ m [k] - γ m [κ m ]) · H m (Z), where κ _m denotes the sampling interval in which the frequency range Φ _{m was} last updated. Thus, not all but rather only selected coefficients, preferably exactly one, are adapted in the predetermined regular sampling-relevant time intervals, preferably with the sampling rate of the input signal.

In principle, there are various possibilities for the selection of the frequency range Φ _m to be updated in a specific sampling interval. For example, the frequency range Φ _{m can be} updated for which | c _m [k] -c _m [κ _m ] | is maximum or those frequency ranges Φ _m in which these quantities exceed a certain threshold value, for example 1 dB. Yet another possibility is that m simply runs through all values from 1 to M again and again systematically or pseudorandomly.

In a preferred embodiment of the invention, the following facts are taken into account by means of the correction values r ₁ ..r _n : The noise suppression determines attenuation values that only depend on the modulation depths, but not on the signal levels themselves, as is correct for normal hearing people. Hearing-impaired people, whose subjective perception of loudness generally increases in a non-linear manner with the signal level, will consequently perceive a signal attenuation by a fixed value a _m as differently pronounced depending on the signal level. This effect would be corrected automatically in serial processing, i.e. with noise suppression followed by compression amplification. However, since parallel processing is taking place here, the correction values r ₁ ..r _{n are} transmitted from the noise suppression to the compression amplification in order to carry out this correction. In the signal analysis for noise suppression, attenuation-related correction values r _n are therefore determined for the N signal levels of the compression amplification, and the amplification values are calculated using signal levels that are reduced by these correction values. The compression gain is therefore corrected in accordance with the noise suppression. This ensures that the signals that are optimally processed for the normal hearing by means of noise suppression are individually and correctly mapped in the hearing range of every hearing impaired person.

Specifically, this means that for each frequency range Φ _{m, in} addition to the signal power s [k] already present, a signal power u [k] reduced as a result of the frequency-specific noise suppression is also calculated. For the frequency ranges Φ _m contained in a frequency range F _n , the s [k] and the u [k] are added separately. The logarithmic correction quantity r _n which is valid with respect to F _n is obtained from the logarithmic ratio of the two sums.

FIG. 3 shows a block diagram for a corresponding signal processing as it takes place in the signal processing for noise suppression 4 for determining the correction quantities r _n . A case is shown where three frequency ranges Φ _m of noise suppression are included in one frequency range of compression gain. In a block 31, a signal power s [k] on signal path 38 is determined in a known manner and a signal level is derived therefrom in block 32, and a modulation depth d _m therefrom in block 33 and an attenuation value a _m therefrom in block 34. In block 35, the logarithmic attenuation value a _{m is} linearly scaled and the reduced signal power u [k] on signal path 36 is calculated by multiplying it by the signal power s [k].

The reduced signal power u [k] is calculated in parallel for each of the three frequency ranges, that is to say for y _m , y _{m + 1} , y _{m + 2} , and summed in node 37. The signal powers s [k] of the three frequency ranges are summed at summation point 39. The sums are logarithmically scaled in the blocks 40 and 41, respectively, and the correction value r _{n is} formed as the difference in the subtraction 41.

The device according to the invention is preferably at least partially as an analog device Circuit or microprocessor based or using application-specific integrated circuits or a combination of these Techniques implemented.

LIST OF REFERENCE NUMBERS

1

first filter unit

2

second filter unit

3

Signal processing for compression amplification

4

Signal processing for noise suppression

5

combination unit

6

controllable filter

7

Means for determining a compression gain

8th

Means for determining noise cancellation

X

input

Y

output

21

achievement Education

22

Level calculation, logarithmic scaling

23

subtraction

24, 24 ', 24 "

gain function

31

achievement Education

32, 40, 41

Level calculation, logarithmic scaling

33

Modulation depth determination

34

Attenuation value determination

35

linear scaling

36

reduced signal power u [k]

37.39

summation

38

Signal power s [k]

42

subtraction

Claims (20)

Device for signal processing in a hearing aid, comprising a filter (6) for frequency-dependent amplitude adjustment of an input signal (X) and means for adjusting coefficients of this filter (6) in accordance with the input signal (X),
characterized in that the device comprises
a means for determining coefficients of a compression gain g _m which describe a frequency-dependent adaptation of the input signal (X) in accordance with frequency-dependent signal levels x _{n of} the input signal (X),
a means for determining coefficients of noise suppression a _m , which describe a frequency-dependent adaptation of the input signal (X) in accordance with interference noises detected in the input signal (X),
the means for adapting coefficients of the filter (6) determines these coefficients from the coefficients of the compression gain g _m and the coefficients of the noise suppression a _m . Apparatus according to claim 1, wherein the means for determining coefficients of the compression gain g _m comprises means for determining signal levels p _n in a first set of frequency ranges F _n with n = 1..N of the input signal (X) and means for determining the coefficients g _m for the compression gain for each of a second set of frequency ranges Φ _m with m = 1..M of the input signal (X) as a function of an optionally modified signal level p _n assigned to the frequency range Φ _m . Apparatus according to claim 2, wherein the means for determining signal levels p _n iteratively forms these as instantaneous effective values of a signal power in the corresponding frequency range F _n . Device according to one of the preceding claims, wherein the means for determining coefficients of the noise cancellation a _m includes means for determining modulation depths d _m in a second set of frequency ranges Φ _m with m = 1..M of the input signal (X) and a means for Determination of the coefficients a _m for noise suppression for each of the frequency ranges Φ _{m of} the input signal (X) in accordance with the corresponding modulation depths d _m . Device according to one of claims 2 to 4, wherein N <M and at least one of the frequency ranges F _n for compression amplification comprises at least two of the frequency ranges Φ _m for noise suppression. Apparatus according to claim 5, wherein the signal processing for compression amplification 3 is designed to determine each coefficient g _m for compression amplification in each case as g _m = f _m (p _n ), p _{n being} the optionally modified signal level of that frequency range F _n for compression amplification, which encompasses the frequency range äusch _m for noise suppression, and f _{m is} one of M functions which, in their entirety, determine a frequency-dependent compression gain. Apparatus according to claim 6, wherein the combined coefficients a _m and g _{m are} logarithmically scaled and their combination forms a combined logarithmic gain value c _m = g _m -a _m by subtraction. Device according to one of the preceding claims, wherein the means for Adaptation of coefficients of the filter (6) is designed in given time intervals not all, but only selected coefficients adapt. Device according to one of the preceding claims, comprising means (23, 35, 36, 37, 38, 39, 40, 41, 42) for correcting the compression gain (3) by modifying the signal levels p _{n in} accordance with the noise suppression. Method for signal processing in a hearing aid, in which coefficients of a filter (6) for frequency-dependent amplitude adjustment of an input signal (X) are adjusted in accordance with this input signal (X), characterized in that the method has the following steps: Determining coefficients of a compression gain g _m which describe a frequency-dependent adaptation of the input signal (X) in accordance with frequency-dependent signal levels of the input signal (X), Determining coefficients of a noise suppression a _m , which describe a frequency-dependent adaptation of the input signal (X) in accordance with interference noises detected in the input signal, and Calculate the coefficients of the filter (6) from the coefficients of the compression gain g _m and the coefficients a _{m of} the noise suppression. A method according to claim 10, wherein for determining coefficients of the compression gain g _m in a first set of frequency ranges F _n respectively assigned signal levels p _n with n = 1..N of the input signal (X) are determined, and the coefficients of the compression gain g _m for each of a second set of frequency ranges Φ _m with m = 1..M of the input signal (X) as a function of a signal level p _n assigned to the frequency range Φ _m . The method of claim 11, wherein a signal level p _n is iteratively calculated as the instantaneous effective value of a signal power in the corresponding frequency range F _n . Method according to one of claims 10 to 11, wherein to determine coefficients of noise suppression a _m in a second set of frequency ranges Φ _m with m = 1..M of the input signal (X) modulation depths d _m and the coefficients a _m for each the frequency ranges Φ _{m are determined in} accordance with the corresponding modulation depth d _m , the modulation depths d _{m being determined} from a chronological order of maximum and minimum values of a signal level p _m in the respective frequency range Φ _m , and the signal level p _m in a frequency range Φ _m as the effective value the signal power is formed in the corresponding frequency range Φ _m . A method according to claim 13, wherein for each modulation depth d _m that exceeds a predetermined value, the assigned coefficient a _{m is} zero, and for values of the modulation depth d _m below the predetermined value with decreasing modulation depth d _m, the coefficient a _{m increases} monotonously. Method according to claims 10-14, wherein at least one of the frequency ranges F _n for compression amplification comprises at least two of the frequency ranges Φ _m for noise suppression, and each coefficient g _m for compression amplification is determined in each case as g _m = f _m (p _n ), p _n the signal level of the frequency range F _n for compression amplification, which comprises the frequency range Φ _m for noise suppression, and f _{m is} one of M functions which determine a frequency-dependent compression gain in its entirety, and wherein the coefficients a _m and g _{m are} logarithmically scaled and their combination by subtraction forms a combined logarithmic gain value c _m = g _m -a _m . A method according to claims 10-15, wherein the coefficients of the filter (6) in be updated at regular intervals, but with every update not all but only a few of the coefficients are updated, in particular only those coefficients whose change is greatest or one exceeds the specified value. The method according to claim 16, wherein the combined coefficients of the filter (6) c _m in the filter (6) are transformed into linear values γ _m and an iterative, frequency-specific update of a transfer function of the filter (6) according to H (z) [k] = H (z) [k - 1] + Σ _m (γ _m [k] - γ _m [κ _m ]) · H _m (z), where H _m (z) only in the frequency range Φ _m pass and otherwise blocking characteristics κ _m denotes a sampling interval in which the transfer function for the frequency range Φ _m was last updated, and a summation Σ _m in a sampling interval k comprises only one or a few of the total M frequency ranges. Method according to claims 10-17, wherein the determination of coefficients of the compression gain g _{m takes} into account the values of the coefficients of the noise suppression a _m . The method of claim 18, wherein the coefficients of the compression gain are determined from modified signal levels p _n 'instead of the signal levels p _n , where p _n ' = p _n -r _n , and r _{n are} logarithmically scaled correction values caused by the noise suppression Correspond to signal attenuation. Hearing aid, comprising means for performing the method according to one of the Claims 10 to 19.

Images