GB2277683A

GB2277683A - Audiometric method and apparatus

Info

Publication number: GB2277683A
Application number: GB9408240A
Authority: GB
Inventors: Richard Owen Stephens; Keith Maries
Original assignee: University of Leicester
Current assignee: University of Leicester
Priority date: 1993-04-27
Filing date: 1994-04-26
Publication date: 1994-11-09
Anticipated expiration: 2014-04-26
Also published as: GB9408240D0; GB9308720D0; GB2277683B

Abstract

The method comprises measuring the pitch discrimination of a person, animal or audio sensing system by detecting the ability of the person etc. to discriminate the harmonic component of a harmonic transient by applying thereto a series of harmonic transients of different duration. Apparatus for use in such a method comprises a computer 11 adapted to generate harmonic transients at different frequencies and transient pulses 2 - 40 cycles wide. The computer outputs digitised frequency signals to a D to A converter 12 which outputs a tone of desired frequency to headphones 13 via amplifier 14. Also disclosed is a method for optimising earphone performance by measuring the pitch discrimination of the user and adapting the earphone to extend short transients to at least the minimum length that the user can discriminate at the frequency. <IMAGE>

Description

AUDIOMETRIC MEASUREMENTS This invention relates to audiometric measurements, importantly, but not exclusively, to measurements made on the human auditory system.

Conventional audiometric measurements on subjects experiencing hearing difficulties involve audiometric threshold intensity measurement - assessment of the lowest intensity sound the subject is capable of perceiving. Not infrequently, a subject with a substantially normal threshold nevertheless has difficulty understanding speech; often this is wrongly attributed to some mental deficiency, since no measurable hearing loss is apparent.

The present invention provides inter alia a measurement method which is capable of detecting hearing problems which cannot be detected by conventional tests and which give measurements which can be used, again according to the invention, to design or adapt a hearing aid to counter such problems. The invention also provides an audiometric instrument which can be used to carry out the measurement method of the invention, for example for establishing statistics which may be useful in the design of--audio reproduction equipment, telephone systems, acoustic enclosures such as concert halls and theatres, as well as for helping design or adapt hearing aids for the hard of hearing.

The invention comprises, in one aspect, an audiometric method for an audio sensing system comprising measuring the pitch discrimination of the system by detecting the ability of the system to discriminate the harmonic component of a harmonic transient by applying to the system a series of harmonic transients of different duration.

The audio sensing system may comprise an animal auditory system and the ability of the system to so discriminate may be detected by detecting a physiological response to such discrimination, or a learned response (or an instructed response for example in human subjects). In a human subject, the response may be operation of an indicating device such as an electrical switch.

The harmonic transient may comprise a plurality of cycles of an audio frequency tone. In one form of test, the harmonic transients of different duration may be applied randomly, while in another test, or in another part of the same test, the harmonic transients may be applied in order of duration. A two part measurement may for example be carried out in a first part of which harmonic transients are applied in order of increasing duration and in a second part of which the transients are present randomly.

The amplitude of the transients may be predetermined as a function of a measurement of the detection threshold of the system (the standard audiometric test for hearing in a human subject).

The measurement may be made at different harmonic frequencies, for example at 2.0 kHz and 4.0 kHz.

The invention also comprises audiometric apparatus comprising means for measuring the pitch discrimination of the human ear comprising a harmonic generator adapted to generate harmonic sound transients of different duration for assessment of the aural response thereto.

The harmonic transient generater may be adapted to generate harmonic transients at different frequencies, for example at 2.0 kHz and 4.0 kHz, and may be adapted to generate transient pulses between two and forty cycles wide, or, say, between two and thirty cycles wide at 2.0 kHz and between two and forty cycles wide at 4.0 kHz.

The generator may be adapted to generate three transient pulses at each duration, and may be adapted to generate transient pulses at different amplitudes, and may be programmed to operate in an audiometric threshold mode, generating transient pulses at adjustable amplitude.

The apparatus may be realised in a computer programmed to operate as a tone generator outputting for example to headphones, the subject using one or more keys of the computer's keyboard as an indicating device, the computer being programmed to record and even possibly to evaluate responses to a test routine.

The invention also comprises a method for optimising the performance of an ear phone such as a hearing aid or headphones for a personal user by measuring the pitch discrimination of the user and adapting the ear phone to extend short transients to at least the minimum length of transient that the user can discriminate at the frequency of the transient.

One embodiment of audiometric apparatus and audiometric methods according to the invention will now be described with reference to the accompanying drawing, in which the single Figure is a diagrammatic illustration of the apparatus.

The drawing illustrates audiometric apparatus comprising means for measuring the pitch discrimination of the human ear comprising a harmonic transient generator in the form of a microcomputer 11 adapted to generate harmonic transients of different duration for assessment of the aural response thereto.

By suitable programming, the computer is adapted to generate harmonic transients at different frequencies - for the audiometric method to be described below, frequencies of 1.5, 2.0, 3.0 and 4.0 kHz. The computer 11 is also adapted to generate transient pulses at desired pulse between two and forty cycles wide and in particular, pulses at 2.0 kHz between 2 and 30 cycles wide and pulses at 4.0 kHz between 2 and 40 cycles wide, and to generate three such pulses at any duration at desired times or in a random fashion.

The computer 11 outputs digitised frequency signals to an eight-bit D to A converter 12 which in turn outputs a tone of the desired frequency to headphones 13 via an amplifier 14 with controllable gain, controlled from the computer 11. A switch 15 applies the output of the amplifier 14 to the left or right ear piece of the headphones 13.

A standard personal computer with VDU, keyboard and printer can be used, but a compact version dedicated to the task in hand can as illustrated comprise a small liquid crystal display 16, a simple 4 x 1 keypad 17 used to control the test procedure, and a small printer 18 that operates on a paper roll such as used on add-listers or printing calculators.

The apparatus illustrated in Figure 1 can be used to measure the pitch discrimination, of the human auditory system by detecting the ability of the ear to discriminate the harmonic component of a harmonic transient by applying to the ear - through the headphones 13 - a series of harmonic transients of different duration.

The human ear can be represented as a series of tuned band-pass acoustic filters. The auditory hair-cell receptors situated on the basilar membrane of the cochlea are considered to be the elements responsible for the band-pass tuned filter characteristics observed. High frequency units are found at the basal end of the cochlea and low frequency units at the apical end.

Exposure to excessive noise, certain drugs and other factors affect the responsiveness of the hair-cell receptors, by causing an increase in the effective bandwidth of the filters.

For a filter of bandwidth B, the following relation holds nB/fo ~ 1 where fo is the centre frequency of the filter and n the number of cycles required for the filter to reach its peak response.

If the duration At of a transient is less than nt, where t is the period of the carrier frequency of the transient, the frequency characteristics of the filter output are determined by the frequency characteristics of the input, not the filter.

If at > nt or Af < B/fo the response of the filter output is determined by the filter and resolution of the carrier frequency of the transient becomes possible.

In the method of the invention, the subject is presented with a series of harmonic transients of different duration and should only be able to discriminate the harmonic component.

Extensive~observation of normal human ears has shown that n ~ 12 cycles at 2.0 kllz and n si 24 cycles at 4.0 kHz. Subjects with a known history of excessive noise exposure or those experiencing speech discrimination difficulties in noisy environments are found to exhibit elevated discrimination thresholds or frequently no clear threshold at all. There appears to be no relation between audiometric intensity threshold measurement and pitch threshold measurement. Subjects with almost normal audiometric intensity thresholds may have poor discrimination - they may hear speech but fail to extract information from the sounds.

Measurements are carried out on subjects, using the illustrated apparatus according to the following procedure.

I Measurement of hearing threshold The object of this part of the test, which is conventional audiometry, is to standardise the discrimination test by presenting all subjects with harmonic tone pulses of the same relative intensity.

The subject is presented with tone pulses of 40 cycle duration at one of the test frequencies to one ear then the other. The intensity of the tone pulse is reduced gradually until the subject indicates that it is no longer audible - the indication is for example by depressing a key on the keypad 17. The threshold intensity is noted automatically by the computer 11.

The intensity of the pulse is then reduced to a minimum level and then increased until the subject indicates the tone pulse is again audible. The second value of the threshold intensity is noted. If the two values are within 5 dB of each other the higher value is used. if the difference is larger the mean value is calculated and used as a measure of the threshold.

II Measurement of discrimination threshold The tone pulse intensity is set at the level of normal speech (-/+ 60 dB SPL) compensating for any hearing loss revealed in Part I of the test. Thus, a subject with a -20 dB hearing loss would be presented with a tone pulse intensity of +80 dB SPL.

The second part of the test is divided into a familiarisation or learn phase and a measurement or random phase.

In the learn phase the subject is presented with a series of incrementing duration tone pulses starting with a two cycle duration pulse. This initial pulse will sound to all subjects like a sharp click. As the tone pulse duration increases the perceived sound changes to a harmonic "ping" or "cheep". By the time the pulses are of forty cycle duration the harmonic nature of the pulse is clear to all subjects.

When the subject acknowledges that he or she can perceive the difference between a click and a ping or cheep, the learn phase is terminated.

In the random phase, the subject is told that a series of sound pulses will be heard at random intervals, some sounding like clicks, others like pings or cheeps. The subject is asked to depress a key on the keypad 17 whenever a ping or cheep is heard.

The computer 11 generates a series of pulses delivering three pulses of each duration - of duration between 2 and 50 cycles, delivered randomly at random intervals between 2 and 5 sec. The intensities of the pulses are calibrated so that the spectral power at the test frequency is constant for all pulses.

The computer 11 now has a record, which can be printed out or analysed, of the pulses the subject was able to discriminate as being harmonic.

For a single subject, who might be suffering loss of speech discrimination, the result of the test can indicate whether this is the result of abnormal bandwidth in the auditory hair-cell receptor filters.

A hearing aid that merely amplifies sounds, even selectively in certain frequency bands, may be ineffective to correct this type of hearing deficiency.

One that extended short transients by an amount sufficient to bring them within discrimination capability of the subject would be more beneficial.

Such a hearing aid can be designed on the basis of the information derived from the test.

The apparatus and method can also be used for audiometric survey, for example, to test the effect of noise or drugs on the hearing of a population that has been exposed, or to establish statistics that can be used in the design of communications systems such as telephone systems, or in the design of audiotoria or sound reproduction equipment.

The utility of the apparatus and method is not limited to human testing, although it is undoubtedly easier to apply to human subjects because they can be told to press a key when they perceive a transient to be harmonic. The test could be applied, however, to animals that can be trained to have a conditional reflex response to a harmonic transient.

The method and apparatus could also be adapted for the measurement of the frequency bandwidths of electrical or acoustic band-pass filters. The outputs of such filters could be Fourier-analysed for harmonic content on inputs of different duration pulses at different frequencies.

Claims

- CLAIMS

1. An audiometric method for an audio sensing system, comprising measuring the pitch discrimination of the system by detecting the ability of the system to discriminate the harmonic component of a harmonic transient by applying to the system a series of harmonic transients of different duration.

2. A method according to claim 1, in which the audio sensing system comprises an animal auditory system and the ability of the system to discriminate the harmonic component of a harmonic transient is detected by detecting a physiological or learned or instructed response to such discrimination.

3. A method according to claim 2, in which said response is a tougher response effected on recognition of discrimination.

4. A method according to claim 3, in which the auditory system is a human auditory system and the response is operation of an indicating device.

5. A method according to claim 4, in which the indicating device comprises an electrical switch.

6. A method according to any one of claims 1 to 5, in which the harmonic transient comprises a plurality of cycles of an audio frequency tone.

7. A method according to any one of claims 1 to 6, in which the harmonic transients of different duration are applied randomly.

8. A method according to any one of claims 1 to 7, in which the harmonic transients are applied in order of duration.

9. A method according to claim 8, in which a two part measurement is carried out, in a first part of which the harmonic transients are applied in order of increasing duration and in a second part of which the transients are presented randomly.

10. A method according to any one of claims 1 to 7, in which the amplitude of the transients is predetermined as a function of a measurement of the detection threshold of the system.

11. A method according to any one of claims 1 to 10, in which the measurement is made at different harmonic frequencies. -

12. A method according to claim 11, in which the measurement is made at 2.0 kHz and at 4.0 kHz.

13. Audiometric apparatus comprising means for measuring the pitch discrimination of the human ear comprising a harmonic transient generator adapted to generate harmonic sound transients of different duration for assessment of the aural response thereto.

14. Apparatus according to claim 13, of which the harmonic transient generator is adapted to generate harmonic transients at different frequencies.

15. Apparatus according to claim 14, in which the generator is adapted to generate transients at 2.0 kHz and 4.0 kHz.

16. Apparatus according to any one of claims 13 to 15, in which the generator is adapted to generate transient pulses between 2 and 40 cycles wide.

17. Apparatus according to claim 16, in which the generator is adapted to generate transient pulses between 2 and 30 cycles wide at 2.0 kHz and between 2 and 40 cycles wide at 4.0 kHz.

18. Apparatus according to any one of claims 13 to 17, in which the generator is adapted to generate three transient pulses at each duration.

19. Apparatus according to any one of claims 13 to 18, adapted to generate transients at different amplitudes.

20. Apparatus according to any one of claims 13 to 19, programmed to operate in RANDOM mode in which transient pulses of duration between 2 and 50 cycles of the harmonic component frequency are generated at random intervals between 2 and 5 seconds.

21. Apparatus according to claim 20, in which three pulses are generated at each duration.

22. Apparatus according to any one of claims 13 to 21, programmed to operate in a learn mode in which a sequence of transient pulses is generated at increasing (or decreasing) widths over a period of time.

23. Apparatus according to claim 22, in which the pulses increment by 2 cycles between 2 and 50 cycles at a constant interval of 5 seconds.

24. Apparatus according to any one of claims 13 to 23, programmed to operate in an audiometric threshold mode generating transient pulses at adjustable amplitude.

25. A method for optimising the performance of an ear phone such as a hearing aid or headphones for a pesonal user by measuring the pitch discrimination of the user and adapting the ear phone to extend short transients to at least the minimum length of transient that the user can discriminate at the frequency of the transient.