Electrical audio signal processing systems and devices – Sound effects – Reverberators
Patent
1981-01-27
1984-04-03
Kemeny, E. S. Matt
Electrical audio signal processing systems and devices
Sound effects
Reverberators
G10L 100, H04R 2900
Patent
active
044412029
DESCRIPTION:
BRIEF SUMMARY
This invention relates to signal processing systems of the type which are suitable for use with implanted hearing prostheses which electronically stimulate the auditory nerves to produce auditory-like sensations.
In order to best utilize an auditory prostheses for speech communication, a signal processor is required which codes the speech signal into a stimulus pattern. Such signal processors in the past have fallen into two general categories:
1. Those which stimulate electrodes in regions where, according to the place-pitch theory of hearing, they would be stimulated in a normal hearing person. Stimulation occurs at rates corresponding to the frequency of vibration of that portion of the basilar membrane.
2. Those which stimulate at one or more positions in the cochlea but with a stimulus common to all electrodes and at a rate equal or proportional to the glottis pulse rate of the speech signal.
Whilst speech processors which fall into category 1 provide the formant information of the speech signal, they fail to provide prosodic information. Furthermore, since the spectral energy is distributed over broad peaks, confusing percepts may be heard by the patient. Speech processors which fall into category 2 provide the prosodic information important as an aid to lip reading, but fail to provide the necessary formant frequency information.
It is the aim of the present invention to overcome these disadvantages by providing a speech processor having an output signal in which both formant and prosodic information is present whereby the production of confusing percepts is avoided.
The speech processor according to the invention is particularly adapted for use with the implantable hearing prosthesis and the associated stimulation electrode array described in our Australian Patent Applications Nos. AU-A 41061/78 and AU-A 46563/79, respectively.
As described in the earlier Applications referred to above, the hearing prosthesis includes an electrode array which is implanted within the scala tympani region of the cochlea so as to cause stimulation of the auditory nerve fibers by the application of electrical currents of varying levels of intensity and stimulation frequency. Psychophysical tests are used to determine the sharpness ranking of each electrode and a programmable means within the speech processor is programmed to cause stimulation of selected electrodes depending on the characteristics of at least one of the time varying parameters which essentially describe the speech signal within any defined time period.
Experiments have shown that different spectral colours can be cued by the activation of individual electrodes in the array, and for this reason the determination of the sharpness ranking of each electrode in the array is important. The sharpness ranking may be determined in many ways. In the present case, an indirect procedure has been developed to obtain an indication of the variation of spectral colour amongst electrodes. Stimuli in the form of 300 ms pulse trains with rise-decay times of 50 ms and a pulse rate of 100 pps were delivered to single electrodes at current levels determined by balancing the loudness across the electrodes. Each patient is presented with two pulse trains separated in time by a one second interval and is asked to indicate whether the sensation produced by the second stimulus was duller or sharper than the first. In this way, the sharpness ranking of each electrode can be determined relative to the other electrodes for subsequent use in programming the speech processor.
Our experiments have established that the amplitudes and frequencies of the fundamental voicing component and the second formant component of the speech signal may be successfully used to essentially define the speech signal for utilization by the prosthesis to cause stimulation of the auditory nerves within a patient to produce auditory-like sensations. To achieve this end, the estimated second formant frequency is mapped so that selected segments of the range of second formant frequencies usually experienced a
REFERENCES:
patent: 3989904 (1976-11-01), Rohrer et al.
patent: 4051331 (1977-08-01), Strong
patent: 4063048 (1977-12-01), Kissiah
Mladejousky et al., "A Computer-Based Stimulation . . . ", IEEE Trans. on Biomed. Eng., Jul. 1976, pp. 286-295.
Clark et al., "A Multiple-Electrode Hearing Prosthesis . . . ", Med. Prog. Tech., vol. 5, pp. 127-140, 1977.
Clark Graeme M.
Millar John B.
Patrick James F.
Seligman Peter M.
Tong Yit C.
Kemeny E. S. Matt
The University of Melbourne
LandOfFree
Speech processor does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Speech processor, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Speech processor will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-670269