APPLE 1144
`APPLE 1144
`
`INOA
`
`20050094822A1
`
` UN
`
`a») United States
`2) Patent Application Publication (0) Pub. No.: US 2005/0094822 Al
`(43) Pub. Date:
`May5, 2005
`Swartz
`
`(54) LISTENER SPECIFIC AUDIO
`REPRODUCTION SYSTEM
`
`(76)
`
`Inventor: Robert Swartz, Highland Park, IL (US)
`
`Correspondence Address:
`CHARLES G. CALL
`68 HORSE POND ROAD
`WEST YARMOUTH, MA 02673-2516 (US)
`
`(21) Appl. No.:
`
`11/032,367
`
`(22)
`
`Filed:
`
`Jan. 8, 2005
`
`Publication Classification
`
`(51)
`
`Int. GU eacveqcunasners HO4R 5/00; HO4R 29/00
`
`(52)
`
`‘US8Ch ssssnisannaannaaias: 381/56; 381/60
`
`(57)
`
`ABSTRACT
`
`A system for use with an audio reproduction system that
`corrects for distortion caused by the system as well as any
`hearing impairment suffered a
`listener. Test signals are
`introduced into the input of the system to produce test
`sounds that are perceptible bythe listener. Using a pushbut-
`ton, the listener indicates whena test signal of progressively
`increasing volume reaches an audible level. The resulting
`measured values ofthe listener’s threshold of hearing at
`different frequencies is compared with comparable data for
`a normal listener to generate correction values that are used
`to program an equalizer which compensates for not only the
`listener’s hearing impairments but also any distortion pro-
`duced by system components or roomacoustics.
`
`Listener's
`Hearing
`Profile Data
`
`Equalizer
`Gain
`Profile Data
`
`
`
`Audio Prompt
`Instructions
`
`Audio
`Test Signal
`Generator
`
`ee 123
`
`~
`
`\
`
`\
`
`\
`
`|}
`
`/
`
`Audio
`Programming
`Source
`
`Equalizer
`
`/
`
`a
`
`1
`
`

`

`
`Generator
`
`
`Source
`
`Audio Prompt
`Instructions
`
`Audio
`Test Signal
`
`Audio
`Programming
`
`137
`
`121
`
`131
`
`127
`
`*,
`
`ea
`
`Listener's
`
`Hearing
`
`Profile Data
`
`125
`
`Equalizer
`Gain
`Profile Data
`
`Microprocessor
`
`Equalizer
`
`130
`
`124
`
`Display
`
`
`
`109
`
`
`
`
`
`S007‘SsARIA,wONRaGngUoNRoyddyjue
`
`
`
`
`
`fFJOTays
`TV7c8r600/S00TSA
`
`2
`
`

`

`
`
`
`
`
`aanssaid-punas
`
`1000
`
`400
`2
`
`£
`
`
`
`10 15
`
`4 kHz
`
`201
`Fig. 2 (prior art)
`
`
`
`
`
`PatentApplicationPublicationMay5,2005Sheet2of4US2005/0094822Al
`
`3
`
`
`

`

`Click when you
`
`Adjust volume to
`comfortably foud level.
`
`hear the tone.
`
`Fig .3
`
`Fig. 4
`
`
`
`
`
`S007‘SARIMONRANGNUONRayddyjuoEd
`
`
`
`
`
`fFJO€POYS
`TV7c8r600/S00TSA
`
`4
`
`

`

`Patent Application Publication May 5,2005 Sheet 4 of 4
`
`US 2005/0094822 Al
`
`ey"
`
`
`
`505
`
`Play explanatory
`introduction soundfile
`to listener
`
`
`
`
`
`
`
`507
`
`Calculate difference
`
`
`All
`between measured and
`test signals
`normal hearing
`played ?
`thresholds at each test
`signal frequency
`
`
`
`Calculate corrective
`Generate next test
`signal at low level
`equalizergain levelat
`each control frequency
`
`Progressively increase
`test signal volume
`
` Adjust equalizer gains
`
`
`
`Listener
`
`
`hearstest
`
`signal ?
`Repeattest
`procedure ?
`
`
`
`Store lowesttest signal
`No
`level perceived by
`515
`listener
`
`529
`
`Fig. 5
`
`5
`
`

`

`US 2005/0094822 Al
`
`May5, 2005
`
`LISTENER SPECIFIC AUDIO REPRODUCTION
`SYSTEM
`
`FIELD OF THE INVENTION
`
`[0001] This invention relates to audio reproduction sys-
`tems and more particularly to reproduction systems which
`incorporate methods and apparatus for compensating for the
`listener’s hearing impairments.
`
`BACKGROUND OF THE INVENTION
`
`is typically the goal of electronic reproduction
`It
`[0002]
`equipment to reproduced sound waves that accurate repli-
`cate original sound waves picked up by microphone.
`Although modern electronic amplifiers that are highly linear
`across the audible frequency range are now available at low
`cost, microphones and speakers or headsets differ signifi-
`cantly in their ability to faithfully transform sound waves
`into electrical signals and back again across the audible
`spectrum.
`In addition, room acoustics often cause more
`degradation in the quality of reproduced sound reproduction
`than the system components used.
`
`In order to compensate for these and other varia-
`[0003]
`tions, sound systems commonly include adjustable mecha-
`nisms which vary from simple “tone controls” to elaborate
`“graphical equalizers” that may be manually adjusted bythe
`user to obtain a desired soundquality. In addition, a number
`of commercially available advancedstereo systems employ
`automated mechanisms that sample the sound produced by
`the combination of speakers and room acoustics and then
`modify relative gain ofdifferent channels to compensate for
`the characteristics of the speakers and the room acoustics.
`These automated balancing systems are widely used for
`adjusting the relative magnitude of sound produced by
`multiple speakers in a “surround sound” system. For
`example, U.S. Pat, No. 5,666,424 issued to Fosgate et al. on
`Sep. 9, 1997 describes a surround sound system that
`employs a microprocessor to digitally control the gain of
`each channel. The microprocessor receives an input signal
`from a microphone placed at the preferred listening location
`within the listening area for automatically balancing the
`relative channel gains during a calibration process to yield
`the best possible surround sound reproduction ofthe stereo-
`phonic source material. As a visual aid, the microprocessor
`displays menus and messages on a videoscreen, and a visual
`display shows the relative levels of the six axes of control
`signals within the surround sound processor.
`
`[0004] These automatic calibration systems do not com-
`pensate for differences in the hearing abilities of different
`listeners. Everyone’s hearing has a different
`response to
`sound. Those with normal healthy hearing perceive sounds
`al the low and high endsof the audible frequency rangeat
`greatly reduced volume. Genetic causes, certain diseases,
`and exposure to loud noises can further impair hearing in
`different ways. And as we age, our ability to hear high and
`lowfrequencies is reduced even more. The nature and extent
`ofthese degradationsis different for every individual and, as
`a result, the sounds we hear, whether “live” or reproduced,
`are a distorted version of the actual sound pressure waves
`that reach our ears.
`
`loss is typically not uniform over the entire frequency
`spectrum of audible sound, with the loss often being greater
`at higher frequency ranges than at lower frequencies, it has
`become common for a hearing health professional to make
`audiological measurements that will
`indicate the type of
`correction or assistance that will be the most beneficial to
`improve that individual’s hearing capability.
`
`[0006] Various systems for measuring auditory responses
`are known. These systems usually provide for application of
`selected tones, broad-bandnoise, and/or narrow-band noise
`which is variable in frequency and amplitude to aid in
`determining the amount ofhearing loss a person may have.
`To assess hearing thresholds for speech, an audiometer may
`also reproduce live voice or recorded speech at selectable
`calibrated levels. Various controls are used to administer
`varying sound conditions to determine a range of responses
`for the individual. The audiological data which records the
`individual’s hearing response are typically charted or
`graphed, and these charts are then used as the basis for
`adjusting the gain vs. frequency characteristics of a pro-
`grammable hearing aid so that it can better compensate for
`the hearing loss characteristics of the wearer.
`
`[0007] U.S. Pat. No. 5,604,812 issued to Meyer on Feb.
`18, 1997 entitled “Programmable hearing aid with automatic
`adaption to auditory conditions” and the prior art patents
`cited therein describe hearing aids that can be programmed
`by the wearer. The hearing-impaired person can retrieve a
`test program of test tones stored a memoryin the hearing aid
`and can actuate a switch when the desired (appropriate)
`hearing threshold is reached, and thus store a correction
`factor each test tone. These stored correction factors adust
`the signal transfer characteristics of the hearing aid and are
`retained until the wearer again reprograms the hearing aid in
`the same way. U.S. Pat. No. 6,035,050issued to Weinfurtner
`et al. on Mar. 7, 2000 describes a further “Programmable
`hearing aid system and method for determining optimum
`parameter sets in a hearing aid.”
`
`SUMMARY OF THE INVENTION
`
`[1 is an object ofthe present invention to improve
`[0008]
`the quality of sound delivered by a sound reproduction
`system as perceived by a person with impaired hearing.
`
`It isa further object ofthe invention to improve the
`[0009]
`quality of sound heard by a listener by correcting for
`distortion introduced by sound reproduction equipment,
`room acoustics, and any hearing disability suffered by the
`listener.
`
` Itis a still further object of the invention to improve
`(0010)
`the quality of sound producedbyan electronic sound system
`by using the system to delivering audible test sound stimuli
`to the listener, accepting responses from said listener indica-
`tive of the listener’s perception of the test stimuli, translating
`the listener’s responses into measured audiological data
`characterizing listener's hearing, calculating control values
`from said measured audiological data, and adjusting a pro-
`grammable equalizer to preferentially amplify sounddeliv-
`ered to the listenerat different frequencies in response to the
`control values.
`
`[0005] To help restore normal hearing, hearing aids are
`available that use adjustable filtering and automatic gain
`control (AGC) parameters. Since given individual's hearing
`
`[0011] A preferred embodiment of the invention takes the
`form of an automatically programmable distortion correc-
`tion mechanism that can be used with an electronic sound
`
`6
`
`

`

`US 2005/0094822 Al
`
`May5, 2005
`
`reproduction system. The reproduction system typically
`includesan electronic amplifier having an input connectedto
`an audio signal source and an output connected to an
`electro-acoustical transducer for delivering sound waves to
`a listener who may have impaired hearing. The program-
`mable correction mechanism employs a test signal source
`coupled to system input for delivering audible test sound
`stimuli to the listener by way of one or more speakers or
`headphones. The mechanism further includes means for
`accepting responses from said listener that
`indicate the
`listener’s perception of the test sound stimuli, A processor is
`usedto translate these responses into measured audiological
`data that characterize the listener's hearing and tocalculate
`equalization control values from the measured audiological
`data. An equalizer selectively amplifies the sounddelivered
`to the listener at different frequencies in accordance with the
`calculated control values.
`
`In accordance with a further feature of the inven-
`[0012]
`tion,
`the correction mechanism may advantageously store
`audiometric data
`indicative of
`the nominal
`frequency
`response characteristics of a person with normal hearing,
`and the processor may then compare the measured audio-
`logical data indicating the current listener’s hearing ability
`with said audiometric data indicating normal hearing ability,
`and thus produce correction values that produce an output
`which the listener hears in the same way a normal person
`would have heard the originally recorded sounds.
`
`[0013] he present invention may be implementedat low
`cost in existing sound systems such as home entertainment
`systems, television and radiosets, and portable audio play-
`ers. The invention may be used to particular advantage in
`portable audio players which use headphones. Such devices
`typically include the “hardware” needed to implement the
`principle functions needed(generating test signals of differ-
`ent frequencies and amplitudes, accepting indications from
`a listener indicating how these test signals are perceived,
`processing these indications to produce equalization control
`values, and adjusting the transfer characteristics of the
`reproduction system), allowing the invention to be incorpo-
`rated into existing systems merely by adding appropriate
`control software.
`
`(0014] These and other objects, features and advantages of
`the present invention may be better understood by consid-
`ering the following detailed description of a preferred
`embodiment of the invention.
`
`BRIEF DESCRIPTION OF THE DRAWINGS
`
`In the detailed description which follows,frequent
`[0015]
`reference will be made to the attached drawings, in which:
`
`is schematic block diagram of a sound
`1
`[0016] FIG.
`reproduction system embodying the invention;
`
`[0017] FIG. 2 is a chart showing the measured frequency
`response of a normal person’s hearing;
`
`[0018] FIGS. 3 and 4 illustrate a simple control interface
`which maybe used to implement the invention in a portable
`digital audio player; and
`
`illustrating a method for
`[0019] FIG. 5 is a flow chart
`generating test signals, accepting responses, and adjusting
`the gain vs. frequency characteristics of a programmable
`equalizer to improve the performance ofan audio reproduc-
`tion system.
`
`DETAILED DESCRIPTION
`
`[0020] A schematic diagram ofan illustrative implemen-
`tation ofthe invention is shown in FIG. 1. Sound from an
`electronic audio programming signal source 100 is delivered
`to a human ear 103 by a sound reproduction system that
`includes an automatic equalization scheme to compensate
`for:
`
`(1) defects in the reproduction capabilities of
`[0021]
`the programming signal source 100 and ofan elec-
`tronic audio amplifier depicted at 105,
`
`(2) distortion introduced by an electrical-to-
`[0022]
`acoustic transducer illustrated by the loudspeaker
`seen at 107,
`
`(3) distortion introduced by room acoustics or
`[0023]
`other sound transmission media which transmit the
`sound waves 109 to the ear 103, and
`
`[0024]
`
`(4) defects in the listener’s hearing ability.
`
`[0025] The mechanism for correcting for distortions
`caused by the reproduction components, the acoustic trans-
`mission medium andthe listener’s hearing employs a test
`signal generator 121 for introducing audio test signals into
`the system, a mechanism illustrated by the pushbutton 123
`for accepting indications from the listener that specify the
`listener’s perception of those test signals as reproduced by
`the available equipment and room conditions, and a micro-
`processor 125 for storing audiometric data at 127 indicating
`how the listener perceived different test signals and how
`normal listeners would perceive those signals. The micro-
`processor 125 processes this data to produce and store gain
`control values at 129 which are used to adjust the response
`characteristics ofan electronically adjustable equalizer seen
`at 130.
`
`Thetraditional goal of audio reproduction has been
`(0026]
`the accurate reproduction ofthe original sound. A mecha-
`nism for correcting the reproduction system including an
`output speaker 107 as illustrated in FIG. 1 can be imple-
`mented using two identical microphones seen at 131 and
`132. The microphone 131 captures original sound pressure
`waves 133 from a sound source (not shown) and the micro-
`phone 132 captures the reproduced sound 109 from the
`output speaker 107. As the frequency of the input sound
`waves 133 changes from the low to the high end of the
`audible range, the reproduction components should ideally
`exhibit constant gain (i.e. should have a linear response)
`over the entire frequency range. By comparing the outputs
`from microphones 131 and 132, the microprocessor 125 can
`generate equalizer gain profile data stored at 129 to adjust
`the frequency response of the programmable equalizer 130
`in orderto provide the desiredlinear response over the entire
`frequency range. In this way, the sound wavesat 109 should
`accurately replicate the original sound waves at 133.
`
`[0027] A second automatic equalization scheme could be
`used that would eliminate the microphone 131. The test
`signal generator 121 may instead be employed to generate
`an input
`test signal which varies in frequency and has
`constant amplitude of the audio frequency range. As the
`frequency changes, the amplitude of the output sound may
`be detected by the microphone 132 andthe gain profile of
`the equalizer 130 may be adjusted so that the amplitude of
`the signal output of the microphone 132 is constant across
`
`7
`
`

`

`US 2005/0094822 Al
`
`May5, 2005
`
`the audible frequency range. This approach, however, will
`not result in truly linear response unless the response of the
`microphone 132 is linear whereas, using two identical
`microphones as described earlier, any nonlinear properties
`of the twoidentical microphones will cancel out.
`
`Improving the sound reproduction capabilities of
`[0028]
`the components is becomingless important since the quality
`of soundreproduction has increased steadily overthe years.
`This is especially true of electronic audio sources like CD
`players, radios, home entertainment systems, television sets,
`and computer playback equipment which employsolid state
`amplifiers. Speakers and headphones have also improved
`significantly. Given the continual improvement of electronic
`devices, even moderatelypriced equipment nowrivals high
`end equipment in sound reproduction quality. As a result, the
`acoustics of the room in which the equipmentis used often
`has a greater effect on the accuracy of the sound reproduc-
`tion than the electronic amplifiers and speakers. While
`automatic equalization techniques of the type described
`above can also correct for poor roomacoustics, it cannot
`compensate for the distortion in the sound heard by the
`listener because of the listener’s impaired hearing ability
`
`[0029] The human ear is not equally sensitive to all
`frequencies, and hearing everyone’s hearing ability is
`reduced in the low and high frequency ranges. The average
`hearing response over the entire audio range as been mea-
`sured and charted by testing a significant numberof people
`with normal hearing. This work, originally conducted by
`Fletcher and Munson in 1933, with later refinements by
`others, has produced a set of curves seen in FIG, 2 which
`specify the sound pressure levels of tones of different
`frequencies that are perceived as being equally loud by the
`average listener. These curves in FIG, 2 are plotted for each
`10 dB rise in sound level. The lowest curve, seen as a dashed
`line 201 in FIG,2, indicates the “threshold of hearing”—the
`lowest sound level which is perceptible by the average
`person at different frequencies over the audio range. The
`uppermost curve plots sound pressure levels at the “thresh-
`old of pain” at which the average person experiences sig-
`nificant discomfort As seen in FIG, 2,all of the curves are
`lowest in the range from 1
`to 5 kHz, with a dip at about 4
`kHz, indicating that the ear is most sensitive to frequencies
`in this range. The intensity of higher or lower frequency
`tones must be raised substantially in order to create the same
`impression to the average listener as a 4 kHz tone.
`
`[0030] Although it would be possible to provide a human
`listener with a hearing aid or other sound reproduction
`system that compensates for the “normal” degradation in
`hearing ability at high and low frequencies, thereby given
`the listener a more nearly “linear” hearing capability, so that
`the listener would perceive sounds of equal sound pressure
`across the audio range as having the same volume, the result
`would sound very unnatural. Thus, while sound reproduc-
`tion equipment should faithfully reproduce the original
`sound waves and exhibit
`a
`linear
`frequency response,
`mechanisms which correct hearing defects should restore a
`person’s hearing to a normal response ofthe kind depicted
`in FIG, 2 rather than giving the listener a linear response.
`
`[0031] The curves shown in FIG. 2 describe the hearing
`characteristics of a nominal “average” person. When a given
`individual, such as an older person, has impaired hearing
`which reduces that person’s ability to hear sounds at par-
`
`the hearing correction mechanism
`frequencies,
`ticular
`should thus give that individual a hearing response that more
`nearly corresponds to normal hearing. To achieve that goal,
`audiometric data representing the normal hearing character-
`istic, represented by the curve 141 in FIG, 1, are stored in
`the memory device 127. A procedure is then performed, as
`described below in conjunction with FIGS. 3-5, which
`captures audiological data that describes both the hearing
`capability ofthe listener and the response of the reproduc-
`lion system, including room acoustics, as indicated by the
`dashed line curve 142 in FIG, 1. A processor then compares
`the normal hearing data 141 with the measured hearing
`capability of the listener 142 and generates and stores
`equalizer gain profile values as illustrated by the curve at
`153 in FIG. 1. The gain vs. frequency data 153 is then used
`to control the response of the programmable equalizer 130
`which preferentially amplifies signals at the frequenciesat
`which the listener’s hearing is impaired in comparison to
`normal hearing. Expressed as a formula, the equalizer gain
`153 as a function offrequency, G(f), should be proportional
`to the difference between the normal hearing response N(f)
`as illustrated in FIG, 2 and the combined measuredeffect of
`the listener’s hearing response M(f) and the frequency
`response of the reproduction equipment and roomacoustics
`S(f). That is:
`GIN=N()-(MUN*S(P)
`
`[0032] where
`
`frequency
`[0033] G(f=Desired equalizer gain vs.
`where 1.0 (unity gain) means no amplification, 2.0
`means amplification by 2, etc.
`
`[0034] N(f)=Normal person's perceived amplitude
`vs.
`frequency in response to constant amplitude
`pressure waves over entire frequency range
`
`[0035] M(f)=Measuredperceived amplitude vs. fre-
`quency of individual user in response to constant
`amplitude pressure waves over entire frequency
`range
`
`[0036] S(f)=Effective gain vs. frequency of reproduc-
`tion equipment and room acoustics where 1.0(unity
`gain) means the room reproduction equipment and
`room acoustics neither increases or decreases the
`amplitude at that frequency
`
`* means the measured response M(f) is mul-
`[0037]
`liplied by the system gain S(f)
`
`invention improves the quality of
`[0038] The present
`sound perceived by the listener by, in effect, administering
`a combined reproduction system test and an audiological
`hearing test to obtain data describing the combined response
`of the reproduction system,
`the room acoustics, and the
`listener’s hearing ability over the normal audible frequency
`range. This testing can be automated in a variety of ways,
`and in many existing audio reproduction systems, can be
`performed by a microprocessor or digital signal processor
`(DSP) which is already present in the system. For example,
`portable music players such as the iPod® marketed by Apple
`Computer of Cupertino, Calif. and the Nomad® manufac-
`tured by Creative Labs,
`Inc. of Stillwater, OK include
`built-in digital processors which can be programmed to
`perform the necessary audiological testing, including test
`stimuli generation, playing and displaying instructional
`
`8
`
`

`

`US 2005/0094822 Al
`
`May5, 2005
`
`prompts to the listener, accepting indications from the
`listener indicating the listener’s perception of the sounds
`produced by the stimuli, processing the results to create gain
`control values, and using those gain control values to adjust
`the frequency response characteristics of the device.
`
`[0039] A-simple switching device such as the pushbutton
`seen at 123 which can be pressed by the listener to indicate
`the times at whicha test tones have reached a predetermined
`volume level, such as the threshold of hearing or a level
`perceived to have the same volume as that of a reference
`signal. The system may include stored spoken instructions
`seen at 137 in FIG. 2 that may be played to prompt the
`listener to perform certain actions.
`In addition, or in the
`alternative, the instructions may be displayed on an avail-
`able display screen illustrated at 139 in FIG, 1.
`
`[0040] FIGS. 2 and 3 illustrate the way in which an
`audiological
`test procedure might be implemented on a
`portable playback device like the Apple iPod®. Using the
`conventional menu system provided by the device,
`the
`listener may initiate the test procedure which may begin
`with an audio instruction and a display illustrated in FIG. 3
`requesting the user to set the device’s volume setting to a
`comfortably load level. A sequence ofdifferent test signals
`(tones, spread-spectrum noise, speech, etc.) is then pro-
`duced, and the listener is instructed to activate a control,
`such as the device’s “click wheel”401 as soon as eachsignal
`is first heard. Some orall ofthe test signals may be delivered
`to one ear al time through the headphonestoseparately test
`the hearing ofeach ear. In this way, the microprocessor can
`determine and store the listener’s hearing threshold level at
`predeterminedtest frequenciesacross the audio spectrum for
`both ears.
`
`[0041] A more detailed explanation of an illustrative
`audiological test procedure is depicted by the flow chart seen
`in FIG. 5. Beginning at 503,
`the device first plays (or
`displays) introductory information to the listener explaining
`the procedure to be followed as indicated at 505. Then
`system then plays a sequence of test signal stimuli of
`progressively increasing volume until the last test signal has
`been produced as indicated by the test at 507. Each test
`signal has a different frequency content and begins at a low
`level as seen at 509. The volume ofeach test signal pro-
`gressively increases as seen at 511 until a level is reached at
`which the listener first hears each signal as shown at 513.
`Wheneach signal reaches the listener's hearing threshold,
`the user operates an input device (such as the pushbutton
`switch seen at 123 in FIG, 1), and the microprocessor stores
`a value indicative ofthe listener’s threshold for that signal
`as seen at 515, and then continues with the next test signal.
`
`[0042] When the listener’s responses to all ofthe test
`stimuli have been recorded, the microprocessor compares
`the listener’s calculated hearing threshold at each frequency
`with a stored value indicating the normal person’s hearing
`thresholdat that frequency (i.c. the values represented by the
`lowest curve in FIG, 2). After the deviation from normal
`hearing is determined at 521, the microprocessorcalculates
`equalizer gain values calculated at 523 that are used at 525
`to adjust the equalizer gain settings. The audible test signals
`producedby the generator 121 may take the form ofselected
`tones, broad-bandnoise, and/or narrow-band noise which is
`variable in frequency and amplitude.
`‘To assess hearing
`thresholds for speech,
`the generator may reproduce live
`
`voice or recorded speechat selectable calibrated levels. The
`same kind of test signals that are used for audiological
`testing by health care professionals can be used to advantage
`to determine the listener’s hearing response.
`
`[0043] After the equalizer has been set, the procedure may
`be performed again to “fine tune” the equalizer settings.
`Note also that two channel stereo systems and multi-channel
`surround sound systems may perform a separate procedure
`for each channel. Thus, for example, a portable music player
`which delivers stereo sound to two earphones may perform
`separate procedures for each ear. In this way, the “balance”
`of the reproduction is automatically adjusted so that more
`nearly “normal hearing” is providedto each ear, even though
`the two ears may be subject to different degrees of hearing
`impairment, or be impaired in different ways. In a multiple
`speaker system, such as a surround sound system, the same
`arrangement may be used with a separate procedure per-
`formed for each channel. Once the listener’s hearing has
`been measured using one or more speakers, the remaining
`speakers may be tested to adjust their relative level (bal-
`ance). Systems that employ special purpose speakers, such
`as a woofer, which are intended to reproduce sound only
`over a limited frequency range may be tested over that range
`only. The control settings generated for each individual
`listener may be separately stored so that, when a given
`listener desires to use the reproduction system, the listener
`need onlyidentify herself and the system will then use that
`listener’s stored equalization values.
`
`[0044] Devices which are readily connected to a personal
`computer, such as portable device players which can
`exchange music files with a personal computer, may offload
`all or part ofthe testing procedure to the personal computer.
`A hearing test conducted on a personal computer or audi-
`ometer can yield audiological test data in a standard form
`that can then betransferred to a numberofdifferent devices
`to program equalizers in cach device to correct for a par-
`ticular person’s hearing. Such audiological test data may be
`conveniently stored on an Internet server where it may be
`automatically accessed when an individual user identifies
`herself.
`
`Soundreproduction systems now commonly use a
`[0045]
`digital signal processor (DSP) to digitize, compress, and
`decompress audio signals. These DSPs often execute equal-
`ization and filtering operations, such as Finite Impulse
`Response (FIR) filtering, and offer
`the listener graphic
`equalizers that can be manually adjusted to provide a desired
`frequency response. In such systems, the present invention
`maybe used to automate the adjustment of such equalizers
`to compensate for a listener’s hearing disabilities. The
`programmed equalization contemplated by the present
`invention can also be performed by the audio signal pro-
`cessors on a personal computer “sound card,” by audio
`amplifiers in cellular telephones, by audio amplifiers built
`into stereo headphones or into “sound card” processing
`circuitry connected between a PC USB port and head-
`phones, and by a wide variety of other sound processing
`devices which would benefit from the automatic, listener
`specific mechanism for compensating for differences in
`individual hearing abilities.
`CONCLUSION
`
`[1 is to be understood that the methods and appa-
`[0046]
`ratus which have been described above are merely illustra-
`
`9
`
`

`

`US 2005/0094822 Al
`
`May5, 2005
`
`tive applications of the principles of the invention. Numer-
`ous modifications may be made by those skilledin the art
`without departing from the true spirit and scope of the
`invention.
`
`Whatis claimedis:
`1. Apparatus for adjusting the frequency response char-
`acteristics of a soundreproduction system, said reproduction
`system including a source of audio frequency electrical
`signals, at
`least one electronic amplifier, and at
`least one
`electro-acoustical transducer, said amplifier having an input
`connected to receive saidelectrical signals from said source
`and having an output connected to said transducer to pro-
`duce sound waves corresponding to said electrical signals
`which are perceived by a human listener, said apparatus
`further comprising:
`
`an electrical audio test signal generator connected to the
`input of said amplifier to produce a sequence of dif-
`ferent test sound waveshaving different volumelevels
`and different frequency content which can be heard by
`said listener,
`
`inpul means for accepting indications from said listener
`indicative ofthe perceived intensity ofat least some of
`said different test waves as perceived bysaid listener,
`
`a processor responsive to said indications for translating
`said indications into a plurality of control values, and
`
`an adjustable equalizer connected between said source of
`audio frequency programming signals and the input of
`said amplifier, said equalizer being coupled to said
`processor to receive said control values andpreferen-
`tially amplify said programming signals at ditferent
`frequencies in accordance with said control values.
`2. Apparatus for adjusting the frequency response char-
`acteristics of a sound reproduction system as set forth in
`claim 1 wherein at
`least selected ones of said audio test
`signals produced by said test signal generator have an
`intensity that progressively increases over time and wherein
`at least some ofsaid indications from said listener indicate
`
`the time at which the sound waves produced by said selected
`ones of said audio test signals are first perceptible by said
`listener.
`
`3. Apparatus for adjusting the frequency response char-
`acteristics of a sound reproduction system as set forth in
`claim 2 wherein said processor translates said times into
`intermediate values indicating the lowest sound level per-
`ceived by said listenerat particular frequencies and wherein
`at
`least some of said control values are related to the
`difference betweensaid intermediate values and the lowest
`sound level which is perceptible by the average person at
`said particular frequencies.
`4. Apparatus for adjusting the frequency response char-
`acteristics of a sound reproduction system as set forth in
`claim 1 wherein at least some of said control values are
`related to the difference between the lowest sound level
`perceived bysaidlistener at particular test sound frequencies
`and the lowest sound level which is perceptible by the
`average person at said test sound frequencies.
`5. A m