AU2002300314B2 - Apparatus And Method For Frequency Transposition In Hearing Aids - Google Patents

Description

P/00/011 28/5/91 Regulation 3.2(2)

AUSTRALIA

Patents Act 1990 COMPLETE SPECIFICATION STANDARD PATENT Application Number: Lodged: Invention Title: APPARATUS AND METHOD FOR FREQUENCY TRANSPOSITIN IN HEARING AIDS The following statement is a full description of this invention, including the best method of performing it known to us 1 APPARATUS AND METHOD FOR FREQUENCY TRANSPOSITION IN HEARING AIDS TECHNICAL FIELD The present invention relates to hearing aids in which it is desired to modify the frequency content of sound signals, for example in order to improve their audibility and intelligibility.

BACKGROUND ART Numerous frequency-transposition schemes for the presentation of audio signals via hearing-aids for people with sensorineural hearing impairment have been developed and evaluated over many years. In each case the principal aim of the transposition is to improve the audibility and discriminability of certain signals at relatively high frequencies by modifying those signals and presenting them at lower frequencies where hearing-aid users typically have better hearing ability. However, various problems have limited the successful application of such techniques in the past. These problems include technological limitations, distortions introduced into the sound signals by the processing schemes employed, and the absence of methods for identifying suitable candidates and for fitting frequency-transposing hearing aids to them using appropriate objective rules.

The many techniques for frequency transposition reported previously can be subdivided into three broad types: frequency shifting, frequency compression, and reducing the playback speed of recorded audio signals (while discarding portions of the signal in order to preserve the original duration).

Among frequency compression schemes, many linear and non-linear techniques including FFT/IFFT processing, vocoding, and high-frequency envelope transposition followed by mixing with unmodified low-frequency components have been investigated. Since harmonic patterns and formant relations are known to be important in the accurate perception of speech, it is also helpful to distinguish spectrum-preserving techniques from spectrum-destroying techniques. Each of these techniques is summarised briefly below.

At present, the only frequency-transposing hearing instruments available commercially are those manufactured by AVR Ltd., a company based in Israel and Minnesota, USA (see http://www.avrsono.com). An instrument produced previously by AVR, known as the TranSonic, has been superseded recently by the ImpaCt and Logicom-20 devices. All of these frequency-transposition instruments are based on the selective reduction of the playback speed of recorded audio signals. This is achieved by first sampling the input sound signal at a particular rate, and then storing it in an analog memory. When the recorded signal is subsequently read out of the memory, the sampling rate is reduced when frequency-lowering is required. Because the sampling rate can be changed, it is possible to apply frequency lowering selectively. For example, different amounts of frequency-lowering can be applied to voiced and unvoiced speech components. The presence of each type of component in the input signal is determined by estimating the spectral shape: the signal is assumed to be unvoiced when a spectral peak is detected at frequencies above 2.5 kHz, voiced otherwise. In order to maintain the original duration of the signals, parts of the sampled data in the memory are discarded when necessary. US Patent 5014319 (1991) assigned to AVR describes not only the compression of input frequencies frequencies are transposed into lower ranges) but also frequency expansion transposition into higher frequency ranges). Other similar methods of frequency transposition by means of reducing the playback speed of recorded audio signals have also been reported previously FR2364520 1971, DE1762185 1970). As mentioned, a major problem with any of these schemes is that portions of the input signal must be discarded when the playback speed is reduced (to compress frequencies) in order to maintain the original signal duration, which is essential in a real-time assistive listening system such as a hearing-aid. This could result in some important sound information being inaudible to the aid-user.

Linear frequency compression by means of Fourier Transform processing has been investigated by Turner and Hurtig at the University of Iowa, USA (Turner, C. W. and R. R. Hurtig: "Proportional Frequency Compression of Speech for Listeners with Sensorineural Hearing Loss," Journal of the Acoustical Society of America, vol. 106(2), pp. 877-886, 1999), and has led to an international patent application (WO 99/14986, 1999). This real-time algorithm is based on the Fast Fourier Transform (FFT). Input signals are converted into the frequency domain by an FFT having a relatively large number of frequency bins. To achieve frequency lowering, the reported algorithm multiplies each frequency bin by a constant factor (less than 1) to produce the desired output signal in the frequency domain. Data loss resulting from this compression of the spectrum is minimized by linear interpolation across frequencies. The output signal is then converted back into the time domain by means of an inverse FFT (IFFT). One disadvantage of this technique is that it is very inefficient computationally, and would consume an unacceptably large amount of electrical energy if implemented in a wearable hearing-aid. Furthermore, it is possible that the propagation delay of signals processed by this algorithm would be unacceptably long for hearing-aid users, potentially resulting in some interference with their lip-reading ability.

A feature extraction and signal resynthesis procedure and system based on a vocoder have been described by Thomson CSF, Paris (EP1006511).

Information about pitch, voicing, energy, and spectral shape is extracted from the input signal. These features are modified by compressing the formant frequencies in the frequency domain) and then used for synthesis of the output signal by means of a vocoder. a relatively efficient electronic or computational device or technique for synthesizing speech signals). A very similar approach has also been described by Strong and Palmer (US Patent 4051331, 1977). Their signal synthesis is also based on modified speech features.

However, it synthesizes voiced components using tones, and unvoiced components using narrow-band noises. Thus, these techniques are spectrumdestroying rather than spectrum-preserving.

A phase vocoder system for frequency transposition is described in a paper by HJ McDermott and M R Dean 'Speech perception with steeply sloping hearing loss", British Journal of Audiology, vol 34, pp 353-361, December 2000).

A non-real -time implementation is disclosed using a computer program. Digitally recorded speech signals were low pass filtered, down sampled and windowed, and then processed by a FFT. The phase values from successive FFTs were used to estimate a more precise frequency for each FFT bin, which was used to tune an oscillator corresponding to each FFT bin. Frequency lowering was achieved by multiplying the frequency estimates for each FFT by a constant factor.

Another system that can separately compress the frequency range of voiced and unvoiced speech components as well as the fundamental frequency has been described by Sakamoto, K. Goto, et al. ("Frequency Compression Hearing Aid for Severe-To-Profound Hearing Impairments," Auris Nasus Larynx, vol. 27, pp. 327-334, 2000). This system allows independent adjustment of the frequency compression ratio for unvoiced and voiced speech, fundamental frequency, the spectral envelope, and the instrument's frequency response (by the selection of different filters). The compression ratio for either voiced or unvoiced speech is adjustable from 10% to 90% in steps of 10%. The fundamental frequency can either be left unmodified, or compressed with a compression factor either the same as, or lower than, that employed for voiced speech. A problem with each of the above feature-extraction and resynthesis processing schemes is that it is technically extremely difficult to obtain reliable estimates of speech features (such as fundamental frequency and voicing) in a wearable, real-time hearing instrument, especially in unfavourable listening conditions (such as when noise or reverberation is present).

Patent EP0054450 (Lafon, 1984) describes the transposition and amplification of two or three different bands of the frequency spectrum into lowerfrequency bands within the audible range. In this scheme, the number of "image" bands equals the number of original bands. The frequency compression ratio can be different across bands, but is constant within each band. The image bands are arranged contiguously, and transposed to frequencies above 500 Hz. In order to free this part of the spectrum for the image bands, the amplification for frequencies between 500 and 1000 Hz decreases gradually with increasing frequency. Frequencies below 500 Hz in the original signal are amplified with a constant gain.

In US Patent 4419544 (Adelman, 1983), the input signal is subjected to adaptive noise cancelling before filtering into at least two pass-bands takes place.

Frequency compression is then carried out in at least one frequency band.

Other techniques described previously include the modulation of tones or noise bands in the low-frequency range based on the energy present in higher frequencies FR1309425 1961, US3385937 1968), and various types of linear and non-linear transposition of high-frequency components which are then superimposed onto the low-frequency part of the spectrum US Patents 5077800 1991 and 3819875 1974). Another approach (WO 00/75920 2000) describes the superposition of the original input signal with several frequencycompressed and frequency-expanded versions of the same signal to generate an output signal containing several different pitches, which is claimed to improve the perception of sounds by hearing-impaired listeners.

Problems with each of the above described methods for frequency transposition include technical complexity, distortion or loss of information about sounds in some circumstances, and unreliability of the processing in difficult listening conditions in the presence of background noise).

It is an object of the present invention to enable frequency transposition to be carried out efficiently in a wearable hearing-aid using currently available digital technology.

SUMMARY OF THE INVENTION According to a first aspect, the present invention provides a method of sound processing in a hearing aid, including the steps of: receiving an input sound signal; sampling said signal at a constant rate; windowing a block of said samples to produce a data set; applying a Fourier transform to determine the magnitude values for frequency components in said data set; determining a phase value for each said frequency component; estimating a rate of change of phase for each said frequency component using values for phase from previous data sets; Deriving frequencies for each of a set of oscillators from said rate of change of phase values; Deriving amplitude control signals for each of said oscillators from said magnitude values; and Summing output values from said oscillators to produce an output composite signal.

Preferably, the Fourier transform used is a fast Fourier transform FFT).

Preferably, the input sound signal is converted to a digital form prior to step Preferably, there are fewer oscillators than FFT output bins, and the FFT output values are selectively assigned to specific oscillators. The outputs of more than one bin may be combined to provide values for a single oscillator.

In a preferred implementation, if frequency transposition is required, the frequency derived for each oscillator may be reduced (or even increased) either for all oscillators or selectively for each oscillator, depending upon the particular listener's requirements. It is also preferred that the amplitude control signals be derived from the magnitude values using a look up table.

According to another aspect, the present invention provides a sound processor device in a hearing aid, said device receiving an input sound signal and being adapted to sample the input signal at a constant rate, the device further including processing means to window a block of said samples to produce a data set; Fourier transform means to determine the magnitude values for frequency components in said data set; a set of oscillators; and processing means for determining a phase value for each said frequency component and estimating a rate of change of phase for each said frequency component using values for phase from previous data sets, so as to derive frequencies for each of the oscillators from said rate of change of phase values, said processing means further deriving amplitude control signals for each of the oscillators from the magnitude values, the arrangement being such that an output composite signal is produced which is the sum of the outputs of each of the oscillators.

The inventive scheme is based on a relatively simple digital signal processing technique known as the phase vocoder. It enables spectrumpreserving transposition with minimal distortion to be applied to signals without the need for speech-feature estimation. The amount of frequency transposition can be adjusted readily to suit the aid-user's hearing impairment, as can the amount of amplification and/or amplitude compression of the transposed (and amplified) output signals. A further advantage of the invention is to provide a practical and relatively simple means for shifting frequencies in an assistive listening device (such as a hearing-aid) in order to reduce the device's susceptibility to feedback oscillations. The use of phase information to provide a more accurate estimate of frequency content allows for more precise setting of oscillator frequencies.

BRIEF DESCRIPTION OF DRAWINGS An implementation of the invention will now be described with reference to the following drawings, in which: Figure 1 is a block diagram illustrating one approach to implementing the present invention; Figure 2 illustrates a process for allocating FFT bins to specific bands; Figure 3 is a graph illustrating a downward frequency transposition over a restricted range; and Figure 4 is a graph illustrating selective output levels in different frequency bands.

BRIEF DESCRIPTION The present invention is adapted to be implemented in various ways for example, in a DSP device; in software; or even using partly analog devices. The invention will be described with reference to this particular implementation, however, the scope of the present invention extends to alternative implementations. Moreover, it is contemplated that various further specific variations to the scheme to accommodate the requirements of particular users, for example further degrees of customisation, are likely to be included in any product implementing the present invention, specifically such variations as are practiced in the field at present.

Although no detailed implementation will be described, the present invention may have application to areas of sound processing other than hearing aids. It will be understood that some details described in relation to hearing aids below may be of reduced relevance. For example, in an application which is not so constrained as to size, processing power and memory, the suggested uses of reduced numbers of oscillators and look up tables may not be pertinent.

The present invention will be principally described with reference to an implementation for a hearing aid. A block diagram of the inventive soundprocessing scheme is shown in Figure 1. Input signals from a microphone) are sampled, converted to a digital representation, and then subjected periodically to a windowing operation followed by a Fast Fourier Transform (FFT).

The outputs of the FFT are analysed to estimate the magnitude and phase of each frequency component of the input signal. The magnitudes are processed to produce amplitude control signals which are assigned to a number of oscillators.

These oscillators are tuned to appropriate frequencies using information derived from the changes over time in the phase estimates. Frequency shifting and/or frequency-specific amplitude processing may be applied to adjust each oscillator's output signal to suit the hearing characteristics of the listener. The final output signal is constructed by summing the output signals of the oscillators, and subsequently converting the composite signal from digital to analog form. The composite output signal is then conveyed to a suitable transducer, such as the earphone (also known as the receiver) of a hearing-aid.

Each of these functional blocks is described in more detail below.

The processes of pre-amplifying the analog signal obtained from a suitable audio transducer (such as a microphone), filtering that signal to limit its bandwidth, sampling the band-limited signal at a constant rate, and converting the sampled signal into digital form are well known to those skilled in the art, and will not be described further here. According to this implementation of the present scheme, a block of sequential input samples is placed in the memory of a suitable digital signal processing unit (DSP). These samples are windowed by multiplying each sample by a corresponding coefficient. Various windowing functions defining suitable sets of coefficients have been described in the literature. The purpose of the window is to ensure that the subsequent FFT operation produces an acceptable estimate of the short-term spectrum of the input signal without noticeable distortion or other undesirable side-effects. In the development of the inventive scheme, it has been found that a 256-point window, with coefficients defined by the product of a Hamming window and a mathematical sinc function, is suitable when an input sampling rate of 14.4 kHz is used. The window outputs are then stacked and added (using a standard numerical operation also known as folding) to produce a set of 128 windowed input samples. This set of data is then processed by a 128-point FFT. The FFT and subsequent functions of the processing scheme are executed every time a new set of 64 samples has been obtained from the input transducer. Thus, with a sampling rate of 14.4 kHz, the FFT and following processing steps are repeated at intervals of approximately 4.4 ms. However, it will be appreciated that different sampling rates, different types and lengths of the window function and Fourier transform, and different extents of FFT overlap are possible within the scope of the invention, and may improve the system's performance under certain conditions.

The outputs of the FFT comprise a set of complex numbers which together represent approximately the short-term spectrum of the input signal. With a 128point FFT, the first 64 bins contain spectral estimates covering the frequency range 0 7.2 kHz, approximately (for a sampling rate of 14.4 kHz). Ignoring the first and last of these bins, which generally do not contain signals of interest in a sound-processing scheme of this type, the remaining bins each provide information about a substantially contiguous sub-band of the input frequency range, each bin extending over a bandwidth of approximately 112.5 Hz. For example, the first bin of interest contains a complex number which describes the real and imaginary components of the input signal within a bandwidth of approximately 112.5 Hz centred on a frequency of 112.5 Hz. The power of each component of the input signal is estimated for each frequency bin by summing the squares of the real and imaginary parts of the complex estimate. (The power value is equal to the square of the magnitude, and thus represents the level of the corresponding frequency component in the estimated short-term spectrum of the input signal).

A well-known deficiency of the FFT for spectral analysis in general is that the output bins are spaced at constant frequency intervals 112.5 Hz in the present case), and have a constant bandwidth approximately 112.5 Hz). For the purposes of frequency transposition as outlined above, it is highly desirable to obtain a more precise estimate of the frequency content of the input spectrum than is possible using the FFT alone, especially at relatively low frequencies. In the inventive sound-processing scheme. This is achieved by making use of information contained in the phase value represented in each frequency bin at the output of the FFT. This extension of the standard FFT process is embodied in an algorithm sometimes described as a phase vocoder.

First, the phase angle is estimated by calculating the inverse tangent of the quotient of the imaginary and real parts of the complex number in each FFT bin.

In practice, it can be difficult to calculate accurate inverse tangents rapidly enough for use in a real-time sound-processing scheme (such as in a wearable hearing-aid). In the inventive scheme, this problem is overcome by providing a look-up table containing the pre-calculated tangents of a relatively small number 64) of phase values. This table contains discrete samples of the range of possible phase values over only two quadrants for phase values between 7/2 and +7/2 radians). These values correspond to the case where the real part of the complex number from an FFT bin is positive. If the real part is in fact negative, it is first treated as positive, and later the phase estimate is corrected by adding an appropriate constant to the phase angle initially calculated.

The phase value for each FFT bin is estimated by a process of successive approximation. A starting value for the phase angle being sought is selected, and the tangent of that value is obtained from the look-up table. The tangent of the candidate phase value is then multiplied by the imaginary part of the complex number in the FFT bin. The product is compared with the corresponding real part, and the candidate phase value is adjusted up or down according to the difference between the estimated and the actual real part. Next, the new candidate phase value is used to obtain the corresponding tangent from the look-up table. This process is repeated until the candidate phase value has the desired accuracy. As mentioned above, in the inventive processing scheme it has been found that adequate precision can be obtained with a 64-entry look-up table encompassing a phase range of -7r/2 to +7/2 radians. Because multiplications and table look-ups can be carried out very rapidly and efficiently in current DSP devices, the above algorithm is particularly suited for use in a wearable, digital hearing-aid.

To use the phase estimates to improve the resolution of the frequency analysis provided by the FFT, it is necessary to estimate the rate of change of the phase in each FFT bin over time. This is because the rate of phase change in a particular bin is known to be proportional to the difference in frequency between the dominant component contained in that bin and the nominal centre frequency of the bin. In this implementation of the inventive scheme, the rate of phase change for each bin is calculated by subtracting the phase estimates obtained from the immediately previous FFT operation from the current phase estimates.

The phase differences are cumulated over time, and then multiplied by a suitable scaling factor to represent the frequency offset between the input signal component dominating the content of each FFT bin and the corresponding centre frequency of that bin. It will be appreciated that alternative processes to determine the phase estimates may be used, for example, a direct calculation process.

The processing described thus far results in a set of power estimates (representing the square of the magnitude spectrum of the input signal), and a set of precise frequency estimates (representing the dominant components present in the input signal). These sets comprise one power value and one frequency value for each FFT bin. As mentioned previously, these sets would normally contain 62 power and frequency values (assuming that a 128-point FFT is employed). It would be possible to construct the desired output signal from the processing scheme by providing 62 oscillators, one for each FFT bin. The amplitudes of each oscillator's output signal would be determined by the power estimate obtained from the corresponding FFT bin, and the frequency of each oscillator would be determined by the frequency estimate obtained from that bin. Such a direct implementation may be used where processing power and memory are not limited. However, this relatively large number of oscillators is undesirable for a practical real-time implementation of the scheme, such as is required in a wearable hearing-aid.

In one preferred embodiment of the inventive scheme, the number of oscillators is 24( it will be appreciated that the same principles may be applied with more or less oscillators than 24). The information contained in the 62 FFT bins is reduced to 24 bands, with each band assigned to a corresponding oscillator. The frequency sub-ranges covered by the 24 bands are normally (but not necessarily) contiguous. The reduction of the FFT bins to a smaller number of bands may be accomplished in various ways. One practical method is to exploit the fact that less frequency resolution is generally needed in an assistive hearing device at high frequencies than at low frequencies. Thus the contents of several relatively high-frequency FFT bins can be combined into a single processing band. The combining operation is performed by summing the powers in the FFT bins, and by obtaining the required precise frequency estimate from only one of the combined bins. The bin selected for this purpose is the one containing the highest power out of the set of combined bins. For low-frequency FFT bins, each bin is usually assigned separately to a corresponding band for further processing.

An example of this process is illustrated in Figure 2. On the left, the outputs of seven FFT bins are shown, with low-frequency bins at the bottom and high-frequency bins at the top. The complex outputs of the two lowest-frequency bins in the example are processed separately to estimate the precise frequency and amplitude of the signal represented in those bins. Thus, these bins produce estimates for two separate bands. At somewhat higher frequencies, the outputs of two adjacent FFT bins are combined for further processing. The powers in the two bins are added to produce a single amplitude estimate covering the signals in both bins. At the same time, the single bin having the larger power is identified, and the complex outputs from that bin only are passed to the process which estimates the precise frequency for that band. For the three highest-frequency bins shown in the example, a similar operation is applied to produce a single amplitude estimate and a single precise frequency estimate for the output band assigned to those bins. It will be appreciated that, in principle, any number of FFT bins can be combined in this way, and assigned to any (smaller) number of output bands for further processing.

Each of the 24 oscillators generates a sine wave that can be controlled in both amplitude and frequency. The desired amplitude is determined by the power value in the corresponding band, as outlined above. The conversion between the power value and the desired oscillator amplitude may be specified by a look-up table or calculated from an appropriate equation. By such means, any desired amount of amplification or attenuation of the input signal may be achieved at each frequency within the frequency range associated with each band). The desired oscillator frequency may also be specified by a look-up table or calculated from an equation. For example, if no change to the frequencies present in the input signal is required, each of the oscillators is merely tuned to generate the same frequency as that estimated from the input signal in the corresponding band. However, if frequency transposition is required lowering of all input frequencies by one octave), then the frequency estimated from the input signal in each band is multiplied by a suitable factor 0.5) before applying it to tune the corresponding oscillator. Note that both the amplitude control and the frequency control for each oscillator can be specified completely independently of the operation of all other oscillators. Thus it is possible to lower some input frequencies and not others, or to lower each input frequency by a different amount. It is also possible with this scheme to raise input frequencies, although it has generally been assumed that frequency lowering, especially when applied to relatively high input frequencies, is more likely to be beneficial for users of hearing-aids.

Figure 3 illustrates graphically how input frequencies may be shifted downwards over only a restricted range. In the example, input frequencies up to 1000 Hz are conveyed to the output of the hearing-aid without any shifting.

However, frequencies above 1000 Hz are shifted downwards progressively such that an input frequency of 4000 Hz is conveyed to the output after being transposed downwards by one octave, to produce an output frequency of 2000 Hz.

An alternative transposition function (not illustrated) would result in relatively small frequency shifts chosen mainly to reduce the susceptibility of the hearing-aid to feedback oscillations. Such oscillations may occur when the output of a hearing-aid is unintentionally picked up and amplified by the hearing aid itself. Under certain conditions, this feedback process may result in a continuous oscillation of the hearing-aid's amplification circuitry. Oscillation is highly undesirable, partly because it reduces the amount of gain that can be effectively applied by the hearing-aid, at least over some of the input frequency range. The inventive processing scheme can reduce the probability of feedback oscillations occurring by applying a small frequency shift to input signals, at least over the restricted frequency range in which the likelihood of such oscillations is greatest.

Thus, higher gains may be achievable with this type of frequency transposition than can be obtained in practice with a conventional hearing aid. Higher gains with reduced susceptibility to feedback oscillations may improve the perception of speech and other sounds by many hearing-aid users.

Figure 4 shows an example of how input levels may be processed to produce the desired output levels in each frequency band. In the example, input levels up to 60 dB SPL are conveyed to the output with a constant gain of 30 dB.

Higher input levels up to 100 dB SPL are compressed by a progressive reduction in the gain with increasing level. Input levels greater than 100 dB SPL are compressed infinitely further increases in input level result in no change in the output level).

It will be appreciated that Figures 3 and 4 are illustrative only, and that any desired conversion of input to output frequency, and input to output level, is possible in the inventive sound-processing scheme. As mentioned above, the desired conversions are specified as the contents of look-up tables or as constants in an appropriate mathematical equation. In general, one such table or equation must be defined for each frequency band in the system to process the frequency and level data.

A side-effect of the fact that the FFT and subsequent processing operations are performed at a rate much lower than the input (and/or output) sampling rate is that the estimates of power and frequency obtained from the input signal are updated relatively infrequently. For example, in the processing described above, the power and frequency estimates are updated each time 64 new input samples are obtained (and 64 new output samples are produced). To avoid artifacts (such as audible clicks) that may result from these possibly abrupt updates, the power and frequency values are each interpolated between the values that were available immediately before each update, and the newly updated values. The interpolation process adjusts each power and frequency value in a sequence of small steps, with each step being applied at the sampling rate, until the adjusted values equal the newly updated values. In one preferred embodiment of the inventive scheme, the size of the steps in the oscillator amplitude is approximately 0.38 dB, and the size of the steps in the oscillator frequency is approximately These step sizes have been found to result in negligible audible artefacts during the interpolation processes. These interpolation processes are performed independently for each oscillator, and may be conveniently combined with the use of the look-up tables providing the required frequency and amplitude conversions described above.

The composite output signal is produced by summing the output signals from all of the 24 oscillators. Optionally, a volume control for the composite output signal may be provided by multiplying the output samples by a factor derived from the setting of an appropriate manual control. The composite signal is then converted to analog form and amplified to drive a suitable output transducer (such as the earphone of a hearing-aid).

It will be appreciated that, although a substantially digital implementation of the inventive processing scheme has been described above, some or all of the processing stages could be implemented using other techniques, such as analog electronic circuits. For example, the oscillators could be implemented using appropriate analog circuits, possibly resulting in a reduction in the electrical power requirements of the processing scheme, and therefore providing benefits for its practical implementation in a wearable hearing-aid.

The above description of one preferred embodiment of the inventive frequency-transposing sound processing scheme is illustrative only, and additions to the scheme and/or variations in the details of the implementation of each functional block of the scheme are implicitly included in the scope of this disclosure.

Claims (16)

1. A method of sound processsing including the steps of: receiving an input sound signal; sampling said signal at a constant rate; windowing a block of said samples to produce a data set; applying a fourier transform to determine the magnitude values for frequency components in said data set; determining a phase value for each said frequency component; estimating a rate of change of phase for each said frequency component using values for phase from previous data sets; Deriving frequencies for each of a set of oscillators from said rate of change of phase values; Deriving amplitude control signals for each of said oscillators from said magnitude values; and Summing output values from said oscillators to produce an output composite signal.

2. A method according to claim 1, wherein the fourier transform used is a fast fourier transform.

3. A method according to claim 2, wherien the input sound signal is converted to a digital form prior to step

4. A method according to claim 1, wherein there are fewer oscillators than FFT output bins, and the FFT output values are selectively assigned to specific oscillators.

A method according to claim 4, wherein the outputs of more than one bin may be combined to provide values for a single oscillator.

6. A method according to claim 1, wherein the method is adapted to provide frequency transposition, so that the frequency derived at step may be increased or decreased selectively.

7. A method according to claim 6, wherein the method is adapted for use in a hearing prosthesis, and the selective increase or decrease is determined based upon the requirements of the intended user.

8. A method according to claim 1, wherein step is performed using a look up table.

9. A sound processing device, said device operatively receiving an input sound signal and being adapted to sample the input signal at a constant rate, the device further including means to window a block of said samples to produce a data set; fourier transform means to determine the magnitude values for frequency components in said data set; a set of oscillators; and processing means for determining a phase value for each said frequency component and for estimating a rate of change of phase for each said frequency component using phase values from previous data sets, so as to derive frequencies for each of the oscillators from said rate of change of phase values, said processsing means further deriving amplitude control signals for each of the oscillators from the magnitude values, the arrangement being such that an output composite signal is produced which is the sum of the outputs of each of the oscillators.

10. A device according to claim 9, wherein the fourier transform means uses the fast fourier transform.

11. A device according to claim 10, wherein the input sound signal is converted to a digital form prior to step

12. A device according to claim 10, wherein there are fewer oscillators than FFT output bins, and the FFT output values are selectively assigned to specific oscillators.

13. A device according to claim 12, wherein the magnitude and frequency values of more than one bin may be combined to provide values for a single oscillator.

14. A device according to claim 9, wherein the device is adapted to provide frequency transposition, so that the frequency derived for the oscillators may be increased or decreased selectively.

A device according to claim 14, wherein the device is a hearing prosthesis, and the selective increase or decrease is determined based upon the requirements of the intended user.

16. A device according to claim 9, wherein the amplitude control signals for each of the oscillators is derived from the respective magnitude values using a look up table. DATED this 29th day of July 2002 HEARWORKS PTY LTD WATERMARK PATENT TRADE MARK ATTORNEYS BUILDING 1 BINARY CENTRE RIVERSIDE CORPORATE PARK 3 RICHARDSON PLACE NORTH RYDE NSW 2113 AUSTRALIA P21703AU00 PNF/AJF