BILATERAL INPUT FOR AUDITORY PROSTHESES

TECHNICAL FIELD

The present invention relates to auditory prostheses, and in particular, to audio signals for auditory prostheses. BACKGROUND OF THE INVENTION

Auditory prostheses are provided to assist or replace the perception of hearing for affected individuals. Such devices include cochlear implants, middle ear implants, brain stem implants, implanted mechanical stimulators, electro- acoustic devices, and other devices providing electrical stimulation, mechanical stimulation, or both.

In the everyday sound environment, the auditory prosthesis recipient listens to the target sound signal, typically speech, in the presence of background noise. In most environments, the locations of the signal and noise sources are not the same. For example, speech is often from in front, as the auditory prosthesis recipient is usually looking at the person talking, while noise is from the side or other locations.

Background noise interferes with speech understanding, and if the level of noise approaches that of the target signal, the auditory prosthesis recipient is unable to effectively distinguish the signal. The signal to noise ratio (SNR) is one measure of this influence of noise upon the signal - a high ratio implies relatively low noise.

Bilateral implantation of auditory prostheses attempts to overcome this problem by providing an auditory prosthesis in each ear, in essence attempting to mimic the normal hearing processes to improve sound, and in particular, speech perception. However, the cost of such a system is more or less double that of a unilateral implant system, with associated surgical risks. This may not be supported by the user's health care arrangements, or the user may not perceive the incremental advantages to justify the additional inconvenience. It is an object of the present invention to deliver an improved quality input signal in an implanted auditory prosthesis system.

SUMMARY OF THE INVENTION

Broadly, the present invention provides an arrangement in which sound signals associated with both ears are received, and the signal from the most appropriate side is used as the basis for stimulation. This may be achieved by providing two complete speech processors, connected by a wired or wireless link, with a selection being made of which signal to use. Alternatively, there may be only a microphone which is provided on one side, and the single processor on the other side decides which sound input to use. This approach may be employed for a user with bilateral implants, with the most appropriate signal being selected for each side.

According to one aspect, the present invention provides a method for providing stimulation signals for an implanted auditory prosthesis, including at least the steps of: assessing first and second signals received at first and second signal receiving devices; determining a signal parameter related to each of the first and second signals; selecting which balance of the first and second signals will be used as the basis for stimulation signals for the auditory prosthesis based on the signal parameter; and generating stimulation signals based upon the selected balance between signals for the implanted auditory prosthesis.

According to another aspect, the present invention provides a system for processing an audio signal for an implanted auditory prosthesis, including first and second signal receiving devices; and processing means adapted to assess first and second signals received respectively at said first and second signal receiving devices, determine a signal parameter of said first and second signals, and select, responsive to said parameter, which balance between the first and second signals is used as a basis for delivering stimuli to the implanted auditory prosthesis.

The present invention also relates to a processing device adapted to receive first and second signals, determine a signal parameter of said first and second signals, and select on of said signals based upon said parameter.

It will be understood that while two signals are referenced, in some implementations there may be more than two signals, for example from multiple microphones on each side, and the present invention is equally applicable to such a situation. The balance of signals used may be to select one or the other, or to deliver some mix of the signals for processing.

The nature of the stimulation signals will vary with the type of device. For an implanted mechanical stimulation device, this may be raw or modified audio data. For a cochlear implant device, it may be detailed electrode and amplitude data. For an electroacoustic device, it may be both electrical and audio data, or separate signals for each type of device.

Similarly, the signal parameter may be determined either at the first and second devices, at one of the devices, or at some other part of the system such as a separate component. For the purposes of this description and claims, the signal parameter may be any measured or determined value which is useful in determining the relative suitability of the signals. For example, the signal to noise ratio may be used, or some other measure of signal strength relative to noise, for example a power ratio or the like, or a measure of noise per se.

Suitable implementations of the present invention accordingly allow the user to have the benefit of selecting for a better audio signal. BRIEF DESCRIPTION OF THE DRAWINGS

An illustrative embodiment of the present invention will be described with reference to the accompanying figures, in which:

Figure 1 is a plan view of an auditory prosthesis recipient in an environment in which there is a target speech signal and noise; Figure 2 is a schematic view of an implementation of the system;

Figure 3 is a block diagram of an automatic sensitivity control arrangement;

Figure 4 is a block diagram showing one implementation of the present invention; and Figure 5 is a block diagram showing an alternative implementation of the present invention.

DETAILED DESCRIPTION

The present invention is applicable to a wide variety of partly or fully implantable auditory prostheses, particularly those which are reliant on signals provided by a microphone or other sound transducer. These prostheses include cochlear implants, middle ear implants, brain stem implants, implanted mechanical stimulators, electro-acoustic devices, or any combination of these, and other fully or partly implanted devices. The devices may have an external processor, or may be partially implanted, with only an external microphone, or even completely implanted including the microphone or other sound transducer. For those devices having an external device, the external device may be continuously, intermittently or occasionally in communication with the implanted device. It is applicable to both unilateral and bilateral implantees.

However, the present invention will be mainly described with reference to an implementation in which a user wears two speech processor devices and has an intracochlear implant. It will be appreciated by those skilled in the art that many other implementations and applications are possible, with suitable modifications to the implementation described.

Figure 1 shows a plan view of a typical sound environment that an auditory prosthesis recipient may encounter. The user 10 is shown seated on a chair. There is a target signal 22, shown here as a speech signal, and a noise signal 24.

The unilaterally implanted auditory prosthesis recipient wears the speech processor and associated microphone, which receives sound signals on the implanted ear, either the left or right. On average, the probability of noise being from the right or left side is equal in the everyday listening environment. When noise is from the same side as the microphone, the signal-to-noise ratio will be worse than if noise was from the opposite ear side. This phenomenon is known as the headshadow effect, which is essentially the attenuation of noise by the head. Bilaterally implanted users take advantage of this effect by consciously listening to the ear with the better SNR, similarly to a person with normal hearing. However, as will be explained below, the present embodiment may have advantages for a bilaterally implanted recipient as well.

To take advantage of this headshadow effect, according to one implementation of the present invention, the auditory prosthesis recipient wears

two speech processors, with one processor situated at each ear. The two processors may be connected by cable for communication, or alternatively by radio frequency or other wireless communication method not requiring direct cable connection. In general overview, the arrangement is shown in figure 2. User 10 has an implanted device 14. A processor 12 is provided on the same side as the device, which for convenience we will call the A side. However, a processor 11 is also provided on the contralateral side, which we will refer to hereafter as the B side. The two processors 11 , 12 are connected by a suitable communications link 16, for example a cable. As will be described in more detail below, in this arrangement processor 12 decides whether the A or B side has experienced the best SNR for the received audio signal. The output of that processor is then used as the basis for stimulation. It will be appreciated that the SNR is only one particular measure which is used in the present implementation, and that the present invention can be implemented using many alternative measures relating to the signal.

Alternatively, there may be a single speech processor and two microphones, with one microphone associated with each ear. In such a case the speech processor would process the signals from each of the microphones. It is emphasised that the substance of the present invention relates to the presence of a microphone or other transducer associated with each side of the user's head, and any suitable arrangement of processors and microphones to achieve this could be used to implement the present invention. In particular, the microphones or other transducers may be external, partially or totally implanted, totally or partially in the ear canal, and associated with processors or not. In a simple implementation, the B side may be a simple microphone, connected by a cable to a speech processor on the A side. ^ι

The present invention may also be applied to a bilaterally implanted user. In this case, the selection of which signal (or which mix of signals) to use can be made commonly for both ears, or on a separate basis for each ear. In this case, it may be that some threshold of noise may need to be passed before the mixing or substitution of signal from the other side microphone is used, as there is a

balance between minimising noise and retaining the inter-ear timing differences for speech interpretation.

The required bandwidth and data rate for transmitting the signal depends on what data is being transmitted and the complexity of the device being used. For example, if the raw audio signal as picked up by the B side microphone is transmitted to the A side, the bandwidth will have to be large enough to cover the approximately 8kHz of the typical cochlear implant audio frequency range at a high enough data rate . The data rate needs to be high enough to ensure that the signals from each processor are very close to being synchronised when received by the A side processor. If the delay between the signals is too large, then if the

B side has the signal with the higher SNR, when the prosthesis processor on side

A comes to process the transmitted signal, the speech percepts heard by the recipient will not be synchronised with the speaker's lip movements.

The signal sent from the B side may be subjected to varying levels of pre- processing. At one extreme is the transmission of raw audio data; at the other may be a fully formed set of stimulation instruction for the prosthesis. The data transmitted may be at any suitable intermediate level. In the latter case, greatly reduced bandwidth may be required.

It is noted that rather than simply using the signal from one side or the other, the present invention can be implemented so that a mix of the signals is used. The changes in the signal parameter, for example the SNR, may be used to control the mix of signals presented to the user.

For example, in one arrangement for an intracochlear implant, the B side processor performs all the signal processing and produces electrode, amplitude (E ₁ A) data pairs for a cochlear implant, being the data that is usually sent to the prosthesis to instruct it on which electrode to stimulate on and at what current level.

An audio compression algorithm could be used to reduce the required bandwidth. For example, US 2006/0235490 assigned to the present applicant, the disclosure of which is hereby incorporated by reference, discloses a suitable coding strategy which could be applied. Other suitable commercial audio compression algorithms could also be used.

Apart from a cable arrangement, a suitable wireless arrangement could be used. As discussed above, the bandwidth required will vary with the implementation chosen. Bluetooth protocol is likely to be suitable for many applications. The person skilled in the art will be aware of the many alternatives available, and can select one with power requirements and bandwidth compatible with this use. Any suitable protocol or arrangement could be used.

It will further be appreciated that the device not specifically associated with the prosthesis may be a relatively simple device, having a microphone and a transmission arrangement to send raw audio data to the processing device, a fully functioned processor as described, or at a level of complexity and processing power in between. The processor could be separate from each of the microphone/processor devices, particularly for example in a cochlear implant which is partially or fully implanted.

The signal-to-noise ratio at each speech processor is preferably independently measured and assessed using a suitable algorithm, for example an automatic sensitivity control (ASC) algorithm. This automatically adjusts the gain of the initial amplifier in the signal pathway, according to the level of background noise. Figure 3 illustrates a prior art ASC arrangement, for example as described in relation to a regular, unilateral arrangement in US patent No. 6,151 ,400, the disclosure of which is hereby incorporated by reference. In this arrangement, the output of the (initial) amplifier 32 is used as an input to the ASC 34. The output of the initial amplifier 32 is the input for the automatic gain control (AGC) amplifier 36. Parameters in the ASC 34 monitor the noise floor, and have pre-set breakpoint level and timing parameters. This allows the gain to be adjusted in response to the ambient noise, and hence in response to the SNR. The perceptual effect of the ASC 34 is a reduction in the loudness of background noise.

This arrangement can be applied to the A and B side signals, as shown in figure 4. On both the A and B side, the respective audio input signal is processed by an initial amplifier 42A, 42B, the output of which forms the input to the respective AGC 46A, 46B. Each ASC 44A, 44B receives the amplifier output, and feeds back a control signal as described. Further, according to the present implementation, a noise floor comparator 48 is provided on the A side. Each ASC

44A, 44B outputs a measure of noise floor on its respective side to the noise floor comparator 48. The characteristics of the ASCs 44A, 44B in this arrangement need to be set to be the same.

The difference between the side A and side B noise floor values is the output of the comparator 48. When the comparator 48 output value is less than or equal to the threshold value, the signal from side A is delivered to the implant.

When the output value from the comparator 48 is above the threshold, the signal from side B is presented. The threshold value of the comparator 48 can be set as appropriate. The default condition is to present side A. It will be appreciated that whilst this is presented as a simple either/or arrangement, the present invention can alternatively be implemented using a variable mixing arrangement. The relative quality of the input signals, for example as measured by SNR, could be used to control the proportion of A and B side signal which is used as the basis for stimulation. The mixing algorithm could take into account, for example, the side for which the stimulation is intended, and have some degree of weighting toward the respective input from that side, in order to preserve some aspects of the signal geometry.

The noise floor comparator 48 has an adjustable time constant, typically in the order of seconds. The background noise level from each ASC 44A, 44B is preferably averaged over a time period, and this averaged value is what is used in the comparator 48. This ensures that the signal delivered to the user is not constantly changing from side to side, which could be distracting for the user.

An alternative arrangement is illustrated in figure 5. In this arrangement, instead of each side having an AGC, the B side could be a simpler device, such as a headset microphone, including an amplifier 52B and ASC function 54B. The speech processor would perform a noise floor comparison 58, and output the selected signal to a shared AGC 56.

Any other suitable SNR assessment algorithm and control arrangement could be used, examples of which are well known to those skilled in the art. It is preferred that relatively slow time constants are used, so that that the selection program function does not switch quickly across the two processors which could be confusing for the auditory prosthesis recipient.

The A side processor according to the implementations described has the additional function of receiving from both processors the measured signal-to- noise ratio In an alternative implementation, this could be arranged to be performed in the implanted device, or elsewhere as part of the prosthesis system, although this is not presently preferred.

It will be appreciated by persons skilled in the art that numerous variations and/or modifications may be made to the invention as shown in the specific embodiments without departing from the spirit or scope of the invention as broadly described. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive. For example, whilst the present invention is described with reference to two microphones, it will be appreciated that the principal could be applied to a larger number of microphones, or to signals which are derived from microphone arrays.