Method And Apparatus For Providing Electronic Sound Clarification For Aurally Handicapped Persons

Abstract

A method is described for determining the frequency response curve or characteristic to be incorporated into sound amplification apparatus utilized by aurally handicapped persons. An auditory threshold is first determined as a function of frequency utilizing pink or white noise passed through 1/3 octave bandpass filters. Subsequently, comfort level response is determined (in the same manner) with the loudness increased for each nominal frequency band until the patient indicates distinct discomfort, the loudness being then decreased until the patient considers the level to be comfortable. Examination of the curves thus obtained may reveal one or more 1/3 octave frequency bands at which dynamic range, the difference between the threshold and comfort levels, is abnormally reduced. Playback apparatus is then adjusted generally to conform to the comfort level curve. Preferably, these tests and the playback apparatus comprehend a broad frequency range on the order of 100 hz to 10 kHz. The adjustment of the playback apparatus may subsequently be altered in accordance with predetermined principles in order to enhance the aurally handicapped person's capacity to understand the spoken language.

Description

This invention relates to the sound amplification arts and, more particularly, to method and apparatus for clarifying the sounds comprehended by aurally handicapped persons.

Electronic hearing aids have typically been designed to provide amplification over a frequency spectrum of approximately 200 hz to 3 kHz. The use of this particular spectrum has been based upon certain tests carried out following the Second World War which resulted in information being published that concluded such a spectrum to be generally optimum for electronic amplification devices to compensate for hearing losses. Recently, however, more advanced tests have indicated that such a limited spectrum may not be optimum for speech discrimination purposes, naturalness of sounds, comfort, and other factors. Nonetheless, electronic hearing aid devices providing merely a broader amplification spectrum do not provide a radical improvement for several reasons; viz.:

1. The ear's natural sensitivity (or physically impaired sensitivity) may vary considerably over a broad frequency spectrum, and a radical change in sensitivity to sound may be observed at closely adjacent frequencies substantially less than an octave apart;

2. A relatively flat, broadband response amplifies noise or undefined signals indiscriminately, thus reducing one's capacity to listen selectively to wanted signals such as speech.

3. Each person's auditory threshold sensitivity and comfort level response is different, and the frequency response characteristic must therefore be shaped to accommodate each person's (and often each ear's) particular response; and

4. Electronic amplification adjustments are not provided for significant amplitude changes within the broadband frequency gain response; hence, acoustic discrimination over narrow band segments cannot be realized.

Some consideration has been given in the prior art to the broad concept of adjusting the frequency response of amplification apparatus to the threshold frequency response curve of an individual. One may refer, by way of example, to U. S. Pat. No. 1,611,130 to Knudsen et al., U. S. Pat. No. 1,659,965 to Tillyer, U. S. Pat. No. 2,003,875 to Balbi, and U. S. Pat. No. 2,112,569 to Lybarger. I have determined, however, that it is necessary to consider several factors ignored or dealt with incompletely by the prior art in order to achieve meaningful improvement in the ability of the patient to understand and discriminate sounds. First, a tailored response curve for the playback apparatus associated with an individual should generally conform to a comfort level curve rather than the threshold curve. Second, any noted reduction in dynamic range, the difference between the threshold and comfort level curves, should be dealt with by attenuating response of the playback apparatus in those frequency bands at which such reduced dynamic range is observed. Third, both the testing and playback apparatus should be substantially more broadband than is currently thought necessary; and fourth, the tests should be carried out and the playback apparatus adjusted in increments of substantially less than one octave, preferably on the order of one-third octave. Further, in conducting the threshold and comfort level tests, I have determined that more meaningful and repeatable results are achieved if the source sound generator issues pink or white noise of about 1/3 octave band width rather than a pure tone centered within the band under consideration.

It is a broad object of my invention to provide an improved method and apparatus for testing aurally handicapped persons.

It is another object of my invention to provide adjustable test and adjustable playback apparatus which can be altered in accordance with test results to achieve a specially tailored response curve in the playback apparatus affording significant sound discrimination or clarification to the user.

It is a more specific object of my invention to provide such apparatus in which tests are carried out and playback apparatus adjusted in increments of less than one octave and across a broad audio spectrum.

The subject matter of the invention is particularly pointed out and distinctly claimed in the concluding portion of the specification. The invention, however, both as to organization and method of operation, may best be understood by reference to the following description taken in connection with the accompanying drawing of which:

FIG. 1 is a block diagram of the test apparatus utilized in the present invention to obtain the audio information according to an individual's hearing response;

FIG. 2 is a block diagram of the playback apparatus utilized to test the improvement realized in an individual's hearing in accordance with the method of the invention;

FIG. 3 is a graph setting out the response curves of a specific individual; and

2 as 4 is a graph illustrating the adjusted response curve incorporated into the playback apparatus of FIG. 2as a result of the observation recorded in FIG. 3.

FIG. 1 illustrates, in simplified block diagram form, the organization of the test equipment utilized in securing information on the hearing response of a test individual to the extent determined necessary to practice the method of the present invention. A pink or white noise generator 1 is coupled to a 1/3 octave filter set 2 which consists of a plurality of 1/3 octave bandpass filters 3 disposed in parallel and comprehends a normal frequency range of at least 100 hz to 10 kHz. Each of the 1/3 octave filters 3 has the facility for continuously adjusting the amplitude of the signal passing therethrough plus or minus 20 db. The output from the 1/3 octave filter set 2 is impressed on the input terminals of a broadband audio amplifier 4 which drives a high grade speaker 5 at an intensity determined by gain control 9. An individual under test, represented by the ear 6, is seated within an audition room to receive the audio output issuing from the speaker 5. Additionally, a broadband microphone 7 is coupled to an audio level meter 8 for monitoring the intensity of sound issuing from the speaker 5 and intercepted by the ear 6.

By way of example only, the pink noise generator 1 may consist of a General Radio Type 1382 Random Noise Generator, the 1/3 octave filter set 2 may be an Altec-Lansing 9062A Graphic Equalizer or a Hewlett-Packard 8056A Filter Set, the microphone 7 may be an AKG Model D224E, and the audio level meter may be a General Radio Type 1565A. Both the amplifier 4 and the high grade speaker 5 may be conventional broadband frequency units of the type well known in the audio field.

Referring simultaneously to FIGS. 1 and 3, either the threshold sensitivity curve or the comfort level curve may be run first. The sensitivity curve can be ascertained by testing the individual in 1/3 octave steps commencing with 125 hz and continuing through 8,000 hz. These tests are carried out in accordance with well known principles and, in fact, pure tones can be utilized in each instance although it is generally established that narrow band noise, being a complex signal, more closely resembles speech than a pure tone resembles speech and is therefore more valid. Hence, pink or white noise generator signals are passed through the narrow bandpass filters 3 with the bandpass filter centering on the nominal frequency being adjusted for minimum attenuation and all others adjusted for maximum attenuation. Briefly, at each nominal frequency setting, the gain control 9 is gradually increased from a nominal setting until the individual being tested first discerns the sound.

The comfort level curve is determined in a somewhat different manner although the same equipment configuration is utilized. The comfort level test is conducted at intervals of frequency approximating one-third octave commencing at least as low as 100 hz and proceeding upwards to 8 or 10 kHz. For each nominal frequency test, the corresponding 1/3 octave filter 3 is advanced to the maximum signal pass position while all other 1/3 octave filters are adjusted for maximum attenuation. The gain control 9 is then advanced until the individual being tested indicates distinct discomfort after which the gain control 9 is gradually decreased until the individual indicates the sound is no longer intolerably loud and can be accommodated comfortably.

During both the threshold sensitivity test and the comfort level test, the reading of the audio level meter 8 is observed for each test increment and recorded to derive the curves presented in FIG. 3.

Once the comfort level curve has been established, an equalized response curve, FIG. 3, may be configured into the playback apparatus of FIG. 2 by adjusting each 1/3 octave filter 13 of the 1/3 octave filter set 12 to closely approximate the comfort level curve. The playback apparatus of FIG. 2 consists of a broadband microphone 17 coupled to the 1/3 octave filter set 12 which shapes the sound intercepted by the microphone 17 to the equalized response curve, and a broadband amplifier 14 driving a high grade speaker 15. An audio level meter 18 monitors the sound level intercepted by the microphone 17 during playback tests.

Utilizing the equalized response curve shown in FIG. 3, tests indicate a modest improvement in the ability of the individual to distinguish the monosyllabic words in the phonetically balanced word lists included in the well known C.I.D. Auditory Test W-22. The improvement observed, however, was not deemed entirely satisfactory, and the reason was discovered to be narrow dynamic range such as that amounting to only about 5 db, observed at 3,150 hz in the individual whose curves are presented in FIG. 3.

An adjusted comfort level curve, taking into account the individual's dynamic range characteristics, is shown in FIG. 4. The adjusted comfort level curve was derived by boosting the playback system response approximately 5 db for the frequencies below the individual's resonance point at which the dynamic range was previously too restrictive. No boost was made above the individual's resonance point because, as will be apparent from FIG. 3, the dynamic range remains restricted at the higher frequencies. A net effect of the boosted curve is to further attenuate frequencies near the individual's resonance point and thus preclude sound saturation at such frequencies and, hence, degraded speech discrimination. Those skilled in the art will appreciate that the adjusted equalized response curve presented in FIG. 4 can be just as readily achieved, and with some equipment more readily achieved, by attenuating the playback apparatus response at the individual's resonance point.

Utilizing the adjusted equalized response curve, the individual whose curves are presented in FIGS. 3 and 4 scored remarkably higher in the previously referenced C.I.D. Auditory Test W-22. Specifically, the individual's ability to distinguish the spoken words increased from 60 percent to 84 percent, and the words missed were done so much more narrowly than in the response tests carried out with the playback apparatus adjusted for flat response.

It will be understood that the curves presented in FIGS. 3 and 4 are actual test results of an individual and that the curves for each individual are very nearly unique requiring specific attention. Other observed individuals have been deemed to have a plurality of response points at which the dynamic range is too narrow and which must be accommodated in reaching an adjusted equalized response curve in the tailored playback apparatus which will afford the individual the ability to distinguish and clarify sounds. It is fundamentally necessary to understand that the comfort level tests must be carried out in increments of less than an octave, preferably on the order of one-third octave, to detect any points along the spectrum at which abnormally narrow dynamic range occurs. Further, it is necessary to carry out these tests over a broad audio spectrum, and the playback apparatus must be capable of achieving an adjusted equalized response curve closely resembling the adjusted comfort level curve across much of the audio spectrum.

While the principles of the invention have now been made clear in an illustrative embodiment, there will be immediately obvious to those skilled in the art many modifications of structure, arrangement, proportions, the elements, materials, and components, used in the practice of the invention which are particularly adapted for specific environments and operating requirements without departing from those principles.

Claims

I claim:

1. In a method for measuring human auditory deficiency as an aid in prescribing compensatory amplification for an individual through use of selective audio amplification means capable of supplying a complex audio signal to the ear of said individual, and wherein said amplification means includes a plurality of independently adjustable circuits for selectively varying the level of the signal passed by each in adjacent passbands, comprising the steps of

determining the comfort level frequency response of the individual by supplying to the ear of the individual a variable level audio signal in discrete frequency spectrum increments including

1. increasing the level of signals passed by a first of said circuits, with the remaining of said circuits being set for a minimum signal passage level, until audio signals within a first limited frequency spectrum reach a level of discomfort, and then decreasing the level of signals passed by the first of said circuits until audio signals are at a comfort level, and

2. repeating this sequence for each of the remainder of said circuits, and

setting selective audio amplification means to provide an equalized response approximating said comfort level frequency response, and with de-emphasis of a resonance frequency spectrum, said resonance frequency spectrum being defined by a marked deviation of the comfort level frequency response in a frequency spectrum wherein spectral components of speech generally are concentrated, to minimize sound saturation of the ear of the individual being tested and thereby emphasize spectral conponents for clarifying speech discrimination.

2. In a method as in claim 1 including the additional step of

supplying audio signals in the form of monosyllabic words to the ear of said individual through selective audio amplification means adjusted in accordance with said last-defined step to ascertain the number of words correctly identified by the individual.

3. A method as in claim 1 wherein

said de-emphasis of a resonant frequency spectrum is achieved by providing attenuation, and each resonance frequency spectrum is so de-emphasized.

4. A method as in claim 1 wherein

said de-emphasis of a resonance frequency spectrum is achieved by attenuation.

5. A method as in claim 1 wherein

each said limited frequency spectrum is a frequency range of less than one octave.

6. A method as in claim 1 wherein

each said limited frequency spectrum is a frequency range of one-third octave.

7. A method as in claim 1 including

determining a threshold frequency response of the individual by supplying to the ear of the individual a variable level audio signal in discrete frequency spectrum increments including

1. increasing the level of signals passed by a first of said circuits, with the remaining of said circuits being set for a minimum signal passage level, until audio signals within a first limited frequency spectrum determined by the passband of said first circuit are heard by the individual, and

2. repeating this latter sequence for each of the remainder of said circuits, and

said equalized response is attenuated in a spectrum thereof wherein the dynamic range is narrow, said dynamic range being defined as the difference between the threshold frequency response and the comfort level frequency response.

8. A method as in claim 7 wherein

said equalized response curve is attenuated in a spectrum wherein the dynamic range is approximately 10 db and less.

9. In a method for measuring human auditory deficiency as an aid in prescribing compensatory amplification for an individual through use of selective audio amplification means capable of supplying a complex audio signal to the ear of said individual, and wherein said amplification means includes a plurality of independently adjustable circuits for selectively varying the level of the signal passed by each in adjacent passbands, comprising the steps of

determining a threshold frequency response of the individual by supplying to the ear of the individual a variable level audio signal in discrete frequency spectrum increments including

1. increasing the level of signals passed by a first of said circuits, with the remaining of said circuits being set for a minimum signal passage level, until audio signals within a first limited frequency spectrum determined by the passband of said first circuit are heard by the individual, and

2. repeating this sequence for each of the remainder of said circuits,

determining the comfort level frequency response of the individual by again supplying to the ear of the individual a variable level audio signal in discrete frequency spectrum increments including

1. increasing the level of signals passed by a first of said circuits, with the remaining of said circuits being set for a minimum signal passage level, until audio signals within a first limited frequency spectrum reach a level of discomfort, and then decreasing the level of signals passed by the first of said circuits until audio signals are at a comfort level, and

2. repeating this latter sequence for each of the remainder of said circuits, and

setting selective audio amplification means to provide an equalized response approximating said comfort level frequency response, and with de-emphasis of a resonance frequency spectrum, said resonance frequency spectrum being defined by a marked deviation of the comfort level frequency response in a frequency spectrum wherein spectral components of speech generally are concentrated, to minimize sound saturation of the ear of the individual being tested and thereby emphasize spectral components for clarifying speech discrimination.

10. In a method as in claim 9 including the additional step of

supplying audio signals in the form of monosyllabic words to the ear of said individual through selective audio amplifation means adjusted in accordance with said last-defined step to ascertain the number of words correctly identified by the individual.

11. A method as in claim 9 wherein

said de-emphasis of a resonant frequency spectrum is achieved by providing attenuation, and each resonance frequency spectrum is so de-emphasized.

12. A method as in claim 9 wherein

said de-emphasis of a resonance frequency spectrum is achieved by attenuation.

13. A method as in claim 9 wherein

said equalized response is attenuated in a spectrum thereof wherein the dynamic range is narrow, said dynamic range being defined as the difference between the threshold frequency response and the comfort level frequency response.

14. A method as in claim 9 wherein

said equalized response curve is attenuated in a spectrum wherein the dynamic range is approximately 10 db and less.

15. A method as in claim 9 wherein

said audio signals comprise noise signals.

16. A method as in claim 9 wherein

each said limited frequency spectrum is a frequency range of less than one octave.

17. A method as in claim 16 wherein

each said limited frequency spectrum is a frequency range of one-third octave.

18. In a method for measuring human auditory deficiency as an aid in prescribing compensatory amplification for an individual through use of selective audio amplification means capable of supplying a complex audio signal to the ear of said individual, and wherein said amplification means includes a plurality of independently adjustable circuits for selectively varying the level of the signal passed by each in adjacent passbands, comprising the steps of

determing a threshold frequency response of the individual by supplying to the ear of the individual a variable level audio signal in discrete frequency spectrum increments including

1. increasing the level of signals passed by a first of said circuits, with the remaining of said circuits being set for a minimum signal passage level, until audio signals within a first limited frequency spectrum determined by the passband of said first circuit are heard by the individual, and

2. repeating this sequence for each of the remainder of said circuits,

plotting said threshold frequency response as a curve,

determining the comfort level frequency response of the individual by again supplying to the ear of the individual a variable level audio signal in discrete frequency spectrum increments including

1. increasing the level of signals passed by a first of said circuits, with the remaining of said circuits being set for a minimum signal passage level, until audio signals within a first limited frequency spectrum reach a level of discomfort, and then decreasing the level of signals passed by the first of said circuits until audio signals are at a comfort level, and

2. repeating this latter sequence for each of the remainder of said circuits,

plotting said comfort level frequency response as a curve,

examining said comfort level frequency response curve for a resonance spectrum, said resonance spectrum being evidenced by a marked deviation of the comfort level frequency response in a frequency range wherein spectral components of speech generally are concentrated, and

setting selective audio amplification means to provide an equalized response lying between said curves and and approximating said comfort level frequency response curve, but with de-emphasis of said resonance frequency spectrum to minimize sound saturation of the ear of the individual being tested for thereby emphasizing spectral components for clarifying speech discrimination.

19. In a method as in claim 18 including the additional step of

supplying audio signals in the form of monosyllabic words to the ear of said individual through selective audio amplification means adjusted in accordance with said equalized response to ascertain the number of words correctly identified by the individual.

20. A method as in claim 18 wherein

said curves are examined to determined any area of narrow dynamic range, said dynamic range being defined as the difference between the threshold frequency response and the comfort level frequency response, and

said equalized response being attenuated in a spectrum thereof wherein said dynamic range is narrow.

21. A method as in claim 20 wherein

said equalized response curve is attenuated in a spectrum wherein the dynamic range is approximately 10 db and less.

22. In a method for measuring human auditory deficiency as an aid in prescribing compensatory amplification for an individual through use of selective audio amplification means capable of supplying a complex audio signal to the ear of said individual, and wherein said amplification means includes a plurality of independently adjustable circuits for selectively varying the level of the signal passed by each in adjacent passbands, comprising the steps of

determining a threshold frequency response of the individual by supplying to the ear of the individual a variable level audio signal in discrete frequency spectrum increments including

1. increasing the level of signals passed by a first of said circuits, with the remaining of said circuits being set for a minimum signal passage level, until audio signals within a first limited frequency spectrum determined by the passband of said first circuit are heard by the individual, and

2. repeating this sequence for each of the remainder of said circuits,

determining the comfort level frequency response of the individual by again supplying to the ear of the individual a variable level audio signal in discrete frequency spectrum increments including

1. increasing the level of signals passed by a first of said circuits, with the remaining of said circuits being set for a minimum signal passage level, until audio signals within a first limited frequency spectrum reach a level of discomfort, and then decreasing the level of signals passed by the first of said circuits until audio signals are at a comfort level, and

2. repeating this latter sequence for each of the remainder of said circuits, and

setting selective audio amplification means to provide an equalized response approximating said comfort level frequency response, said equalized response being attenuated in a spectrum thereof wherein the dynamic range is narrow, said dynamic range being defined as the difference between the threshold frequency response and the comfort level frequency response.

23. A method as in claim 22 wherein

said equalized response includes de-emphasis of a resonance frequency spectrum to minimize sound saturation of the ear of the individual being tested and thereby emphasize spectral components for clarifying speech discrimination, said resonance frequency spectrum being defined by a marked deviation of the comfort level frequency response in a frequency spectrum wherein spectral components of speech generally are concentrated.

24. In a method for measuring human auditory deficiency as an aid in prescribing compensatory amplification for an individual through the use of selective audio amplification means for supplying a complex audio signal to the ear of said individual, and wherein said amplification means includes a plurality of independently adjustable circuits having adjacent passbands for selectively varying the level of the signal passed thereby and providing said audio signal, comprising the steps of

determining the threshold frequency response of the individual and plotting a curve thereof by setting a first of said circuits for minimum attenuation and the remaining of said circuits for maximum attenuation, increasing the gain of said first circuit until audio signals are heard, and repeating the sequence for each of said circuits,

determining a comfort level frequency response of the individual and plotting a curve thereof by setting a first of said circuits for maximum signal passage and the remaining of said circuits for maximum attenuation, increasing the level of the audio signal until the individual discerns distinct discomfort, reducing the level of said ausio signal to a level of comfort, and repeating these steps for each of the remaining circuits,

examining the curve of said comfort level response and detecting any spectrum area of resonance evidenced by a significant curve deviation in the level of said comfort level response, and

examining the response of said individual to audio signals wherein the spectrum of said audio signals have been de-emphasized at said area of resonance to minimize sound saturation of the individual and thereby emphasize the spectral components of said audio signals required for speech discrimination.

25. A method as in claim 24 wherein

the spacing between said threshold curve and said comfort level curve represents dynamic range, and the level of said audio signals is de-emphasized in a spectrum thereof where said dynamic range is approximately 10 db and less.

26. A method as in claim 25 wherein

said circuits provide bandpass increments of less than one octave.

27. A method as in claim 26 wherein

said circuits provide bandpass increments of one-third octave each.