Patent Application
20070263891
The invention relates to a hearing device. The hearing device can be a hearing aid, a headphone, an earphone, a hearing protection device, a communication device or the like. The hearing device may be worn in or near a user's ear or implanted. The invention relates furthermore to a signal processor for processing audio signals in a hearing device, to a method of operating a hearing device and to a method of fitting a hearing device.
In U.S. Pat. No. 5,278,912 and U.S. Pat. No. 5,488,668, audio compressors for use in hearing aids are disclosed. Such compression systems may work individually on a number of frequency bands.
In EP 0′ 590′ 903 B1 an output limiter for hearing aid amplifiers is disclosed.
In EP 0′ 836′ 363 B1, a loudness limiter is disclosed, by means of which the loudness of a signal transmitted to the human ear by a hearing apparatus can be restricted to a maximum acceptable loudness level.
A goal of the invention is to create a hearing device that has an improved performance in loud environments or at high signal pressure levels (SPLs).
Another goal of the invention is to provide for a signal processor useful in a hearing device, having advanced features and allowing for an improved performance.
Another goal of the invention is to provide for an improved method of operating a hearing device, in particular a method allowing for an improved performance in loud environments or at high signal pressure levels.
Another goal of the invention is to provide for an improved method of fitting a hearing device, in particular a method allowing for an improved performance of the hearing device in loud environments or at high signal pressure levels.
One object of the invention is to decrease the power consumption of a hearing device. The lifetime of a power supply of the hearing device (typically a battery) may be increased.
Another object of the invention is to achieve an improved acoustic performance of a hearing device.
Another object of the invention is to reduce a hearing device user's perception of distortions in a hearing device.
These objects are achieved by a hearing device, a signal processor and by a method of operating a hearing device and by a method of fitting a hearing device according to the patent claims.
The hearing device comprises
In an important embodiment,
In other embodiments, the incoming signal may, e.g., be an electromagnetic signal, e.g., a radio frequency signal. In such a case, said input transducer may convert incoming signals, which already are audio signals, into audio signals in a form suitable for being used within the hearing device.
Due to the importance and advantages of the above-mentioned important embodiment, in the following, the invention and possible embodiments are described with respect to said important embodiment. Nevertheless, the description may be applied to the broader scope of the invention as well. Replacing (mentally) the “SPLs” by “levels” and “input acoustical sound” by “input signals” in the appropriate places in the text below, will yield the corresponding examples and descriptions in the broader scope of the invention.
The invention may be seen in that, for a certain range of input SPLs, the input-SPL-to-output-SPL-characteristic of the hearing device has a negative slope: In that range, the output SPL is the smaller the larger the input SPL is.
Through this, an additional way of dynamics processing is created besides the known expansion, compression and limiting.
This way of dynamics processing is particularly useful in loud environments. In loud environments, the SPL of direct sound, i.e., of sound that propagates as sound waves (acoustic waves) from outside the user's ear canal to the user's ear drum (or other sensing element of the ear), can be considerably high. That direct sound SPL can become comparable to or even higher than sound provided to the user via the hearing device, in which case said sound provided to the user via the hearing device may be reduced or even muted.
In a primary view of the invention, the sound source of said direct sound and the sound source of the audio signals within the hearing device are identical. In that case, said reducing or muting of the sound provided to the user via the hearing device may be done in favour of said direct sound, which then can partially or completely replace said sound provided to the user via the hearing device.
In another possible view of the invention, the sound source of said direct sound and the sound source of the audio signals within the hearing device are not identical. In that case, said reducing or muting of the sound provided to the user via the hearing device may, e.g., be accomplished in order to save energy in the hearing device when the user would anyway not be able to properly perceive the outgoing acoustical sound, because the direct sound is too loud (masking of outgoing acoustical sound from the hearing device by direct sound). For example, if the user attends a loud rock concert and meanwhile receives a telephone call on his mobile phone, the voice on the mobile phone, received in the hearing device by a telephone coil in the hearing device, can be suppressed (optionally in certain frequency bands) whenever the rock music is particularly loud (optionally in said certain frequency bands).
Said direct sound partially propagates through the user's head (bones) and partially propagates through the user's ear canal. The latter part is the larger the less the ear canal is blocked. It can be particularly large, e.g., in case of large vents and in case of a hearing device with an open fitting (open canal device). A vent is a channel-like opening in a part of a hearing device worn in the ear canal, which is usually meant to equalize pressure differences between the inside of the ear canal and the outside. The invention can also be used in conjunction with a closed fitting and/or with a very small or a possibly blocked vent. In that case, the attenuation of direct sound typically is typically about 30 dB to 40 dB.
It is possible to describe “direct sound” as sound that acoustically bypasses the electrical path in the hearing device.
Audio signals are electrical signals (analog and/or digital), which are obtained from acoustical sound (sound waves) through conversion and possibly further processing; i.e., they represent sound.
Said incoming acoustical sound usually is acoustic waves in the environment in which the hearing device user is located.
Said input transducer is typically a mechanical-to-electrical converter, in particular a microphone. The hearing device may, of course, comprise more than one input transducer, e.g., two or three.
Said output transducer is typically an electrical-to-mechanical transducer, in particular a loudspeaker, in the field of hearing aids also named “receiver”.
Said signal processor typically is a digital signal processor (DSP), possibly realized in form or an software-programmed processor, but may be analogue or analogue-digital hybride. In a simple form, the processor is basically an amplifier, wherein “amplifying” may include attenuating as well.
Said audio signals processed by said signal processor are derived from said audio signals, which are derived from said incoming acoustical sound by said input transducer.
Said audio signals transduced by said output transducer are derived from the output of said signal processor.
Said outgoing acoustical sound is to be presented to the user. More precisely, said outgoing acoustical sound is to be presented to a sensing element of an ear of a user of the hearing device. Said sensing element typically is an ear drum; it may be a part of the inner ear, in particular a part of the cochlear (in particular if the hearing device is or comprises an implant).
Said portion of said incoming acoustical sound may be identical with said incoming acoustical sound or, more precisely, with that portion of it, which is transduced by said input transducer. Or, said portion may be defined by a certain frequency range or by a certain class of signals obtained in a classification process, or by others. Classification is well known, e.g., in the field of hearing aids, and allows to classify acoustic events not only with respect to its frequency spectrum, but also, e.g., with respect to its time structure (impulse-like sounds versus constant or repetitive sounds).
Basically the same applies for said portion of said outgoing acoustical sound: It may be identical with said outgoing acoustical sound (as transduced by said output transducer), or it may be a portion defined by a certain frequency range or by a certain class of signals obtained in a classification process, or by others.
In one embodiment, there exists an SPL, above which the output of said least one portion of said outgoing acoustical sound is muted (fully suppressed). This allows to save energy, in particular energy consumed by said output transducer. The energy savings can be particularly high when said portion of said outgoing acoustical sound comprises or consists of low frequencies (typically below 300 Hz or below 100 Hz).
In one embodiment, said input SPL range has a lower limit, referred to as input threshold SPL. In particular, said input SPL range may extend from said input threshold SPL to higher SPLs; the range can be open towards high SPLs.
Said input threshold SPL can be chosen according to various criteria and requirements:
In one embodiment, said input threshold SPL is chosen such that an SPL of a superposition of said outgoing acoustical sound and direct sound near the user's ear drum (or other sensing element of the ear) is below the user's uncomfortable level (UCL). It is also possible to refer to the threshold of pain instead of the UCL as a reference for defining or adjusting the gain curve of the hearing device. Said UCL is typically in the range between 100 dB and 120 dB, but may be in the range between 80 dB and 140 dB. It is possible to choose the input-SPL-to-output-SPL characteristic such that the SPL of said superposition does at no input SPL reach or exceed the UCL, at least not as long as the SPL of direct sound alone does not exceed the UCL. For a suppression of direct sound, it is referred to a special embodiment described below (in the detailed description).
In one embodiment, said input threshold SPL is an input SPL for which the corresponding output SPL (of said portion of outgoing acoustical sound) in dB minus the corresponding SPL of direct sound near the user's ear drum (or other sensing element of the ear) in dB is smaller than or equal to +6 dB (or +3 dB). I.e., the negative slope of the input-SPL-to-output-SPL-characteristic of the hearing device may set in as soon as the direct sound is just a bit softer than the outgoing acoustical sound of the hearing device. Said setting-in of said negative slope may take place at smaller differences (output SPL in dB minus direct sound SPL in dB), e.g., when direct sound and outgoing acoustical sound of the hearing device are equal (0 dB difference) or when the direct sound is somewhat louder than the outgoing acoustical sound (e.g., −6 dB difference), or when the direct sound is even louder than that. And said negative slope may set in at any difference in between the named differences. Such a choice of the inset of said negative slope will usually allow to make good use of the direct sound and to save reasonable amounts of energy. The smaller said input threshold SPL is chosen compared to said SPL of direct sound, the smaller the negative slope of the input-SPL-to-output-SPL-characteristic of the hearing device may be chosen, which allows to avoid a noticeable reduction of loudness with increasing input SPL. This way, the occurrence of disturbing loudness changes near or above said input threshold SPL (i.e., at the inset of the negative slope or within the range of the negative slope) can be prevented.
In one embodiment, said input threshold SPL is chosen equal to an input SPL which is, within 6 dB (or 3 dB), equal to its corresponding output SPL. This way, the inset of said negative slope is chosen in a region where the (SPL-related) gain of the hearing device is about one.
In one embodiment, said input threshold SPL is chosen smaller than or equal to a maximum input SPL (MPI), wherein said input transducer will tend to produce distorted input audio signals for input SPLs of said at least one portion of said incoming acoustical sound above said maximum input SPL. I.e., the input threshold SPL is chosen such that it can be prevented that the user would perceive distortions stemming from exposing said input transducer to too high SPLs. The output SPL for (presumeably) distorted sounds can be muted or chosen so low that distorted portions of sound are masked by other sound portions (of outgoing acoustical sound and/or of direct sound). The perceptions of distortions can be avoided, and an improved acoustical performance is achieved.
Analogously, in one embodiment, said input threshold SPL is chosen such that, as output SPL of said at least one portion of said outgoing acoustical sound, at most a maximum output SPL (MPO) is reached, wherein said output transducer will tend to produce distorted outgoing acoustical sound for output SPLs of said at least one portion of said outgoing acoustical sound above said maximum output SPL. I.e., the input threshold SPL is chosen such that it can be prevented that the user would perceive distortions stemming from demanding too much SPL from said output transducer. Distortion can be avoided, and an improved acoustical performance is achieved.
Combining these two distortion-minimizing embodiments leads towards a virtually distortion-free hearing device.
In one embodiment, said decrease in output SPL is chosen such, that an SPL of a superposition of said outgoing acoustical sound and direct sound near the user's ear drum (or other sensing element of the ear) is, within ±6 dB, in particular within ±3 dB, a constant for input SPLs of said at least one portion of said incoming acoustical sound above said input threshold SPL, as far as the corresponding SPL of said direct sound is not larger than said constant, wherein said direct sound is incoming acoustical sound reaching the user's ear drum (or other sensing element of the ear). In other words, the SPL of the superposition of direct sound and sound output from the hearing device is approximately constant for input SPLs above the input threshold SPL, at least insofar as the direct sound is not too loud, i.e. louder than said constant. If the direct sound is louder than said constant, it is nevertheless possible, namely by means of the special embodiment described below (in the detailed description), to let the SPL of said superposition remain constant at even higher input SPLs, i.e., the size of the interval within which SPL of the superposition is approximately constant can be enlarged.
Usually, for all input SPLs of said at least one portion of said incoming acoustical sound above said input threshold SPL, the corresponding output SPL of said at least one portion of said outgoing acoustical sound is at most as large as, in particular smaller than, the largest output SPL of said at least one portion of said outgoing acoustical sound corresponding to any input SPL of said at least one portion of said incoming acoustical sound equal to or below said input threshold SPL. In other words, usually, the highest output SPL of all output SPL values belonging to an input SPL above the input threshold SPL is smaller than (or, optionally, equal to) the highest output SPL of all output SPL values belonging to an input SPL below the input threshold SPL.
In one embodiment, the output SPL of said at least one portion of said outgoing acoustical sound is decreasing (and optionally remaining constant) with increasing input SPL of said at least one portion of said incoming acoustical sound for all input SPLs of said at least one portion of said incoming acoustical sound above said input threshold SPL. In other words, there is no increase of output SPL above a certain input SPL (the input threshold SPL).
The above-mentioned negative slope of input-SPL-to-output-SPL-characteristics of the hearing device is typically only dependent on the input SPL (or, which is equivalent, on a magnitude depending in the input SPL), but not solely from other magnitudes, like a signal-to-noise ratio or a classification result.
In one embodiment, the hearing device is an open canal device. In that case, the amount of direct sound is rather high, so that by means of the invention, the acoustical performance and the power consumption can be greatly improved. The user may take advantage of direct sound.
In one embodiment, the hearing device is a hearing aid. By means of the invention, the acoustical performance and the power consumption can be greatly improved, since hearing aids mostly have to provide for high gains and high output SPLs, and, at high input SPLs, the invention will allow to reduce the output SPL that has to be provided for by the hearing device.
In one embodiment, said at least one portion of said incoming acoustical sound is obtained by filtering in at least a first frequency range.
In one embodiment, said at least one portion of said outgoing acoustical sound is obtained by filtering in at least a second frequency range.
The above-mentioned negative slope of input-SPL-to-output-SPL-characteristics of the hearing device may be applied individually to various frequency ranges (frequency bands). These may be predetermined or automatically selected frequency bands.
The above-mentioned filtering may be band-pass filtering, highpass filtering, lowpass filtering or others, in which case said frequency range designates frequencies, which may pass; or it may be band-stop filtering or the like, in which case the frequency range designates frequencies, which are rejected.
In one embodiment, said first and second frequency ranges are identical and, in particular, both designate either frequencies, which are rejected or frequencies, which may pass.
In one possible embodiment, in which said portions of incoming/outgoing sound are frequency bands, said signal processor is designed such that there exists, for each of a multitude (typically between 3 and 15) of frequency bands, an input level (SPL) range of the portion of said incoming signal within the corresponding frequency band, in which portion an increase in input level (SPL) of said portion of said incoming signal within said corresponding frequency band results in a decrease in output level (SPL) of the portion of said outgoing acoustical sound within said corresponding frequency band. In other words, a negative slope of the input-SPL-to-output-SPL-characteristics of the hearing device may be applied to each of a multitude of frequency bands. It is possible to select said input level (SPL) ranges and said input threshold levels (SPLs) independently for different frequency bands.
The signal processor for processing audio signals in a hearing device has a non-linear input level-output level characteristic, which comprises an input level range in which an increase in input level corresponds to a decrease in output level. In terms of an input level-gain characteristic of the signal processor, there is an input level range, in which an increase in input level corresponds to a superproportional decrease of the gain (slope<−1).
The advantages of the signal processor correspond to the advantages of the corresponding hearing device.
The method of operating a hearing device comprises the steps of
The method of fitting a hearing device, which comprises an input transducer for transducing incoming acoustic sound into audio sound, comprises the step of adjusting a parameter of a signal processor of said hearing device, which processor is designed such that there exists an input SPL range of at least a portion of said incoming acoustical sound, in which an increase in input SPL of said at least one portion of said incoming acoustical sound results in a decrease in output SPL of at least a portion of said outgoing acoustical sound.
Said parameter may be one of the group comprising
The advantages of the methods correspond to the advantages of corresponding apparatuses.
It should be noted that the invention usually relates to dynamics processing for wanted signal (useful signal; as opposed to noise=unwanted signal) or such signal plus noise. But it usually does not relate to noise (=signals considered to represent noise) alone, although it may in general be used for any kind of signal or noise. In other words, the superproportional gain (or level) decrease is not applied typically exclusively to signals that are considered (prevailingly) noise or unwanted signally.
Further preferred embodiments and advantages emerge from the dependent claims and the figures.
Below, the invention is described in more detail by means of examples and the included drawings. The figures show:
The reference symbols used in the figures and their meaning are summarized in the list of reference symbols. Generally, alike or alike-functioning parts are given the same or similar reference symbols. The described embodiments are meant as examples and shall not confine the invention.
The hearing device 1 in
In the case shown in
Incoming acoustical sound 2, typically from the environment surrounding the user, has near the input transducer 11 a signal pressure level SPLin and is converted into audio signals 21 by the input transducer 11. These audio signals are, after optional further processing, fed into the signal processor 15, in which an amplification takes place as will be discussed below. The term amplification is meant in a general sense and comprises attenuation as well; it may be considered “dynamics processing”, and the signal processor 15 may be considered an amplifier or a dynamics processor.
The signal processor 15 outputs processed audio signals 25, which, after optional further processing, are fed to the output transducer 19, where they are converted into outgoing acoustical sound 3. That outgoing acoustical sound 3 has an SPL depicted as SPLout and impinges on the user's ear drum 90, so that the user can perceive what has been output by the hearing device 1.
A portion of the incoming acoustical sound 2, which propagates to the user's ear drum as acoustical sound, is referred to as “direct sound” 4. That direct sound 4 is mainly sound waves travelling within and through the ear canal 80, but also comprises sound waves conducted through bones of the user's head. The sound waves travelling in air in the ear canal 80 may, e.g., travel through a vent of the hearing device 1, or are hardly or only little influenced by the hearing device 1 when the hearing device 1 is an open-canal device. The SPL of the direct sound 4 near the ear drum 90 is named SPLdirect.
In the ear canal 80 near the ear drum 90, the direct sound 4 and the sound output from the hearing device 1 interfere, and their superposition, which is depicted in
Often, the ear canal 80 is strongly blocked by the hearing device 1, mainly in order to prevent feedback from the outgoing acoustic waves 3 to the input transducer 11. But at least at high input SPLs SPLin, some leakage of direct sound 4 and its perception is unavoidable. In case of large vents and open-canal devices (open fitting), the amount of direct sound 4 is considerable.
The known dynamics processing comprises expansion “Exp.”, linear ranges “Lin.”, compression “Compr.” and limiting “Lim.”. Expansion may help to suppress low signals; in a linear range, amplification is constant; compression reduces the dynamics and may help to adapt to restricted dynamics perception capabilities of the user; and limiting restricts the maximum output, so as to avoid, e.g., exceeding the hearing device's limit for distortion-free output or feeding uncomfortably loud signals to the user.
In the range of expansion “Exp.”, the slope of the input-gain characteristic is greater than zero; in the linear range, the slope is zero; in the compression range, it is between zero and −1; and in the limiting range, it is −1.
In an input SPL range ΔSPLin, which extends from an input threshold SPL named SPLin,thresh to higher input SPLs, SPLout decreases with increasing SPLin (
In
In
The shape of the curves, also outside the range ΔSPLin, may be chosen in various ways. An important parameter (or set of parameters) for the invention describes, where to arrange ΔSPLin, i.e., describes the ΔSPLin, to choose. In particular, the lower input SPL, from which ΔSPLin starts (SPLin,thresh), may be chosen according to various aspects, some of which will be discussed below.
Furthermore, the value of SPLin,thresh has been chosen such that the corresponding output value (and actually all output values SPLout) are below (or at most equal to) a value SPLout,max, above which the output transducer tends to distortions.
Furthermore, the value of SPLin,thresh according to
By means of dotted arrows, an important point of the invention is visualized in
In
Furthermore, it is depicted in
In a simple case of the invention, it is possible to use basically the full audio signal 21 (as output from the input transducer 11) and to amplify or damp that full audio signal 21 in the DSP 15 (cf.
For example, as shown in
Another way of implementing the invention is sketched in
The outputs of the (partial) signal processors 151, 152, 153 is added to a sum signal, which is then fed to the output transducer 19 and converted into output acoustical sound 3.
An even more complex and versatile embodiment is shown in
The controller 17 provides all the (partial) signal processors 151 to 154 with control signals (dashed arrows). Since the controller 17 receives all the above-mentioned audio signals, it is possible to use any algorithm on these for generating said control signals. For example, the embodiment of
Also the embodiment of
As another example, it is also possible to take the input level in the neighboring frequency band(s) into account and, accordingly, feed a part of the corresponding level to the (partial) signal processor 151, 152, and 153, respectively.
It is worthwhile to note that when frequency filtering is used, predetermined frequencies or automatically selected frequencies (and frequency ranges) may be used.
Besides using frequency filtering for splitting up audio signals, it is also possible to use other ways of classifying audio signals. Such are known in the field of hearing aids. E.g., classification in dependence of the time structure of the audio signals.
In a special embodiment of the invention, by means of which the problem of excessive direct sound disturbing the user can be solved, the hearing device will generate phase-reversed sound or “anti-sound”, i.e., acoustical waves with reversed phase. The direct sound will at least partially be suppressed when it superposes with the phase-reversed sound. Accordingly, direct sound will have a decreased SPL. In order to suppress direct sound, an audio signal representative of the direct sound has to be obtained. This may be the audio signal 21 (confer
Nevertheless, it is possible to achieve a negative slope of input-SPL-to-output-SPL-characteristics of the hearing device without the generation of sound (“anti-sound”) for suppressing direct sound through destructive interference, as has been shown above.
One advantage of this special embodiment is, that hearing device users can be protected from excessive environmental sound. In some environments, like, e.g., at rock concerts or at construction sites, the environmental sound may be so loud that the direct sound is disturbing or even painful (SPLdirect>UCL). This may in particular be the case with open-canal devices or hearing device with large vents. Decreasing SPLdirect when SPLdirect is large by adding phase-reversed sound is then very advantageous.
Accordingly, the hearing device may on the one hand work as a hearing aid or communication device or the like, and on the other hand as a hearing protection device (using said reversed-phase sound).
In particular, it is possibly by means of the special embodiment to use the phase-reversed sound for achieving an enlarged SPLin-range, in which SPLsum does not rise. That SPLin-range is enlarged with respect to what can be achieved without anti-phase sound (cf.
The generation and use of phase-reversed sound is described in more detail in EP 1 499 159 A2 and in WO 2005/052911. Therefore, these two documents are herewith incorporated by reference in this application.
