United States Patent
`
`[19]
`
`[11] Patent Number:
`
`5,027,410
`
`Williamson et a1.
`[45] Date of Patent:
`Jun. 25, 1991
`
`[54] ADAPTIVE, PROGRAMMABLE SIGNAL
`PROCESSING AND FIL'I'ERING FOR
`HEARING AIDS
`
`[75]
`
`Inventors: Malcolm J. Williamson; Kenneth L.
`Cummins; Kurt E. Hecox, all of
`Madison, Wis.
`
`[73] Assignee:
`
`Wisconsin Alumni Research
`Foundation, Madison, Wis.
`
`[21] Appl. No.: 269,937
`
`[22] Filed:
`
`Nov. 10, 1988
`
`[51]
`Int. Cl.5 ............................................. HMR 25/00
`
`[52] US. Cl. .................
`381/68.4; 364/724.01
`[58] Field of Search ....................... 381/68.4, 68, 68.2,
`381/71, 73.1, 94, 104, 106, 107, 99; 333/166,
`167, 168, 173, 14; 328/167; 364/724.01, 724.08,
`724.09
`
`[56]
`
`References Cited
`U.S. PATENT DOCUMENTS
`
`
`3,180,936 4/1965 Schroeder
`...... 381/94
`3,403,224 9/1968 Schroeder
`.......... 381/98
`
`3,509,558 4/1970 Cancro .........
`340/349 AD
`3/1971 Gharib et a1. .............. 381/72
`3,571,529
`
`3/1971 Shigeyama et a1.
`.
`3,784,749
`381/106
`
`3,803,357 4/1974 Sacks ...................
`330/279
`
`.
`3,855,423 12/1974 Brendzel et a1.
`381/68.2
`
`3,872,290
`3/1975 Crooke et a1.
`381/106
`3,894,195
`7/1975 ,Kryter ................................... 369/48
`(List continued on next page.)
`
`FOREIGN PATENT DOCUMENTS
`
`7/1987 Australia .
`67671/87
`.
`237203 9/ 1987 European Pat. Off.
`2407613 6/ 1979 France ............................ 364/724. l 9
`60-21612 2/1985 Japan ..................................... 381/94
`2184629A 6/ 1987 United Kingdom .
`
`OTHER PUBLICATIONS
`
`Tavares, “Nature and Application of Digital Filters,”
`The Engineering Journal (The Engineering Institute of
`Canada), vol. 50, No. 1, Jan. 1967, pp. 23-27.
`Brochure entitled “The Heritage” by Zenith Hearing
`Aid Sales Corporation, (publication, date unknowp)._
`
`Rabiner et al., “Terminology in Digital Signal Process-
`ing,” IEEE Trans. Audio Electro. Acoust., vol.
`AV—20, pp. 322—337, Dec. 1972.
`Bader et
`al., “Programmgesteuertes Rauschfilter,”
`Fernseh und Kino Technik, 1974, No. 8, pp. 231—233 (in
`German). Accompanying English translation (A Pro-
`gram Controlled Noise Filter).
`Barford, “Automatic Regulation Systems with Rele-
`vance to Hearing Aids,” Scandinavian Audiology Sup-
`plement, (6/ 1978), pp. 335—378.
`’
`Mangold et
`al.,
`“Programmerbart Filter Hjalper
`Horselskade,” Elteknik med Aktuell Elektronik, 1977,
`(List continued on next page.)
`
`Primary Examiner—Forester W. Isen
`Attorney, Agent, or Firm—Foley & Lardner
`
`[57]
`
`ABSTRACT
`
`A hearing aid system utilizes digital signal processing to
`correct for the hearing deficit of a particular user and to
`maximize the intelligibility of the desired audio signal
`relative to noise. An analog signal from a microphone is
`converted to digital data which is operated on by a
`digital signal processor, with the output of the digital
`signal processor being converted back to an analog
`signal which is amplified and provided to the user. The
`digital signal processor includes a time varying spectral
`filter having filter coefficients which can be varied on a
`quasi-real time basis to spectrally shape the signal to
`match the hearing deficit of the user and to accommo-
`date ambient signal and noise levels. The coefficients of
`the spectral filter are determined by estimating the en-
`ergy in several frequency bands within the frequency
`range of the input signal, and using those energy esti-
`mates to calculate desired gains for the frequency bands
`and corresponding spectral filter coefficients. The spec-
`tral energy analysis may be carried out using pairs of
`high pass and low pass digital filters in cascade relation,
`with the output of each low pass filter being provided to
`the next pair of high pass and low pass filters. The rate
`at which output data is provided from the filters in each
`pair may be reduced from the sample rate of input data
`by one half for succeeding pairs of filters in the cascade
`to thereby reduce the computation time required.
`
`IO
`
`58 Claims, 10 Drawing Sheets
`RECEIVER
`SPEAKER
`
`
`
`39
`POWER AMP
`AND VOLUM
`
`
`
`
`CONTROL
`
`
` 38
`LOW PASS
`PRE-EMPHASIS
`
`
`
`FILTER
`AGC AMPLIFIER
`37
`AND LOW PASS
`
`
`
`
`FILTER
`3‘
`EARPIECE
`
`
`
`‘2
`PROCESSOR
`AC COUPLING
`
`
`Z-POLE
`Z-POLE
`
`27
`
`LOW PASS
`LOW PASS
`
`
`FILTER/AMP
`FILTER
`u
`52
`AC COUPLING
`
`AC COUPLIN
`
`
`
`I/o CONTROL
`rs
`53
`AND TIMING
`57
`GAIN RANGING
`LOGIC
`AMPLIFIER
`
`
`
`83-h J” an
`
`no CONVERT
`O/A CONVERT
`LINEAR
`LINEAR
`‘8
`
`DIGITAL
`56
`SIGNAL
`PROCESSOR
`
` AND PROM
`
`
`5'
`
`A
`
`
`
`
`
`£45330
`
`’
`
`APPLE 1015
`
`1
`
`APPLE 1015
`
`

`

`Page 2
`__________________________-———————————-—-—-—
`
`5,027,410
`
`U.S. PATENT DOCUMENTS
`
`.
`
`.
`..
`
`3,927,279 12/1975 Nakamura et a1.
`................... 381/68
`
`..... 381/68
`3,989,897 11/1976 Carver
`.. 381/68.4
`3,989,904 11/1976 Rohrer et al.
`
`..... 381/47
`4,025,721
`5/1977 Graupe et al.
`.. 38l/68.4
`4,051,331
`9/1977 Strong et a1.
`.
`..... 381/72
`4,061,875 12/1977 Freifeld et 81.
`1/1978 Flannigan et a1.
`4,071,695
`381/106
`
`3/1978 Hamilton .......
`4,079,334
`330/279
`.. 381/68.2
`4,099,035
`7/1978 Yanick
`
`381/106
`4,112,254 9/1978 Blackmer
`
`9/1979 Beard .....
`. 369/48
`4,169,219
`
`1/1980 Graupe et al.
`4,185,168
`381/68
`
`2/1980 Moser ............
`.. 381/68
`4,187,413
`381/68.4
`2/1980 Graupe et al.
`4,188,667
`
`2/1981 Orban ........
`4,249,042
`381/106
`
`4,297,527 10/1981 Pate .......
`381/107
`
`4,366,349 12/1982 Adelman
`381/68.2
`
`4,396,806
`8/1983 Anderson ..
`381/103
`
`4,409,435 10/1983 Ono ...............
`.. 381/68.2
`1/1984 Mansgold et al.
`4,425,481
`381/68.2
`
`
`
`...... 381/68
`4,454,609
`6/1984 Kates
`
`381/68.2
`4,508,940 4/1985 Steeger ..
`4,548,082 10/1985 Engebretson et al.
`.. 73/585
`
`
`4,622,440 11/1986 Slavin ...............
`381/68.l
`,. 381/94
`4,628,529 12/1986 Borth et al.
`..
`4,630,304 12/1986 Borth et a1.
`381/94
`
`......
`4,630,305 12/1986 Borth et a1.
`381/94
`4/1987 Henrickson et a1.
`.. 381/31
`4,661,981
`
`4,696,044 9/1987 Waller, Jr.
`.......
`381/98
`
`4,700,361 10/1987 Todd et al.
`381/94
`4,701,953 10/1987 White .......
`381/46
`
`4,723,294 2/1988 Taguchi
`381/94
`4,731,850
`381/68.2
`3/1988 Levitt ct a1.
`.
`
`5/1988 Kroeger et al.
`.
`4,747,143
`381/47
`
`..
`4,783,818 11/1988 Graupe et a1.
`381/71
`
`4,791,672 12/1988 Nunley et a1.
`..
`381/68
`
`381/68.4
`4,792,977 12/1988 Anderson et al.
`7/1989 Kates ...............
`. 381/68.4
`4,852,175
`4,887,299 12/1989 Cummins et a1. .................. 38l/68.4 -
`
`OTHER PUBLICATIONS
`
`No. 15, pp. 64—66 (In Swedish). Accompanying English
`
`translation Programmable Filter Helps Hearing Im—
`paried People.
`Braida et a1., “Hearing Aids—A Review of Past Re-
`search,” ASHA Monographs, No. 19, 1979, pp. 54—56,
`section entitled Characteristics of Compression Ampli-
`fiers.
`
`Marigold et a1., “Programmable Hearing Aid with Mul-
`tichannel Compression,” Scandinavian Audiology 8,
`1979, pp. 121—126.
`.
`Mangold et a1., "Multichannel Compression in a Porta—
`ble Programmable Hearing Aid,” Hearing Aid Journal,
`Apr. 1981, pp. 6, 29, 30, 32.
`Walker et a1., “Compression in Hearing Aids: An Anal-
`ysis, A Review and Some Recommendations”, National
`Acoustics Laboratories, NAL Report, No. 30, Jun.
`1982, Australian Government Publishing Service.
`McNally, “Dynamic Range Control of Digital Audio
`Signals,“ .1. Audio Eng. Soc., vol. 32, No. 5, May 1984,
`pp. 316—326.
`Williamson, “Gisting Analysis" Rome Air Develop-
`ment Center Final Technical Report RADC—TR—8-
`4—130, Jun. 1984.
`Stikvoort, “Digital Dynamic Range Compressor for
`Audio,” J. Audio Eng. Soc., vol. 34, No. i, Jan/Feb.
`1986, pp. 3—9.
`White, “Compression Systems for Hearing Aids and
`Cochlear Prostheses,” Veterans Administration Journal
`of Rehabilitation Research and Deve10pment, vol. 23,
`No. 1, 1986, pp. 25—39.
`Cummins et a1., “Ambulatory Testing of Digital Hear-
`ing Aid Algorithms,” RESNA 10th Annual Confer-
`ence, San Jose, Calif., 1987, pp. 398-400.
`F. .1. Bloom, “High-Quality Digital Audio in the Enter-
`tainment Industry: An Overview of Achievements and
`Challenges,” IEEE ASSP Magazine, Oct. 1985, pp.
`2—25.
`“TMS320 First—Generation Digital Signal Processors,”
`brochure published by Texas Instruments, Jan. 1987.
`P. 0, Vaidyanathan, “Quadrature Mirror Filter Banks,
`M—Band Extensions and Perfect Reconstruction Tech-
`niques,” IEEE ASSP Magazine, Jul. 1987, pp. 4—20.
`
`2
`
`

`

`US. Patent
`
`June 25, 1991
`
`Sheet 1 of 10'
`
`5,027,410
`
`28
`
`20
`
`30
`
`24
`
`22
`
`2I
`
`27
`
`F IG.
`
`1
`
`30
`
`3I
`
`40
`
`RECEIVER-
`SPEAKER
`
`39
`
`38'
`
`37
`
`MIC
`
`TELE
`COIL
`
`32
`
`33
`
`44
`
`45
`
`POWER AMP.
`ANO VOLUME
`CONTROL
`Low PAss
`F'LTER
`AGC AMPLIFIER
`FILTER MPO
`AND LOW PASS
`CONTROLS
`EAR'PlECE
`34
`FILTER
`____ _____A___________________________
`AC COUPLING
`2
`PROCESSOR
`58
`36
`2-POLE
`43
`LOW PASS
`F lLTER/AMP
`Ac COUPLING
`.
`GAIN RANCINC
`
`27
`
`52
`
`5'
`
`53
`
`USER MODES/
`TEST CONTROL
`\__
`l/O CONTROL
`AND TIMING
`
`Z‘POLE
`LOW PASS
`FILTER
`
`S7
`
`'
`
`
`firm 3 I
`AMPLIFIER
`
`AMPLIFIER
`LEENE'ATR
`'2 B” —
`
`30 dB GAIN
`“NEAR
`D/A CONVERT.
`
`A/D CONVERT.
`
`47
`
`FIG. 2
`
`48
`
` DHHTAL
`SIGNAL
`
`
`PROCESSOR
`
`
`AND PROM
`
`
`
`50
`
`56
`
`3
`
`

`

`US. Patent
`
`June 25, 1991
`
`Sheet 2 of 10
`
`'
`
`5,027,410
`
`30 MICROPHONE
`
`SELECTABLE
`PRE/DE-EMPHASIS
`FILTERING
`
`65
`
`
`
`
`
`TIME VARYING
`SPECTRAL
`SHAPING
`
`FILTERING
`
`66
`
`DIGITAL TO
`ANALOG
`CONVERSION
`
`ANTI-IMAGING
`(LOW PASS)
`FILTERING
`
`DRIVER
`AMPLIFICATION
`
`68
`
`69
`
`70
`
`SPEAKER
`
`40
`
`60
`
`6|
`
`62
`
`63
`
`6d
`
`
`
`
`
`
`PREAMPLIFICATION
`AND
`PRE-EMPHASIS
`(HIGH PASSI
`
`
`
`
`
`
`
`
`SLOW
`AUTOMATIC
`
`
`GAIN
`
`
`CONTROL '
`
`
`
`
`
`
`ANTI-ALIASI NG
`ILOW PASS)
`FILTERING
`
`
`AND
`AMPLIFICATION
`
`
`
`
`
`
`
`ANALOG
`TO
`DIGITAL
`CONVERSION
`
`
`
`
`SELECTABLE
`HIGH PASS
`F l LTERING
`
`FIG?)
`
`4
`
`

`

`US. Patent V
`
`June 25, 1991
`
`Sheet 3 of 10
`
`5,027,410
`
`FoldB)
`
`
`
`SLOPE R2
`
`
` SLOPE RI
`
`H4) \~
`
`
`
`SLOPE R0
`\
`
`20 LOG G(T)
`
`
`
`SLOPE=O
`
`
`
`
`
`
`
`
`SLOPEzRO-l
`R0>I
`
`»
`
`HG. 5
`
`20 LOG EtT)
`
`/2
`
`SLOPE:R2-l
`R2<l
`
`5
`
`

`

`US. Patent
`
`June 25, 1991
`
`Sheet 4 of‘10
`
`5,027,410
`
`
`
`TIME
`
`6
`
`

`

`U.S. Patént
`
`June 25, 1991
`
`Sheet 5 of 10
`
`5,027,410
`
`CALCULATE
`COEFFICIENTS (l6)
`
`
`
`
`167
`
`SPECTRAL
`FILTER
`
`FIG. 8
`
`7
`
`

`

`US. Patent
`
`June 25, 1991
`
`Sheet 6 of 10
`
`5,027,410
`
`INPUT (0-8 KHZ}
`
`I73
`
`rnoH
`PASS
`4—8 KHZ
`
`LOW
`PASS
`0-4 KHZ
`
`02
`
`SECOND RANGE mGNAL
`
`HlGH
`PASS
`2-4 KHZ
`
`LOW
`PASS
`0—2 KHZ
`
`I75
`
`I74
`
`THIRD RANGE SIGNAL
`
`HIGH
`PASS
`h2_KHZ
`
`LOW
`PASS
`O-IKHZ
`
`I77
`
`'75
`
`FOURTH RANGE smNAL
`
`
`
`Lovv
`PASS
`0-500 H2
`
`[78
`
`'
`
`'
`
`I79
`
`4 KHZ
`—3 KHZ
`
`,
`
`2 KHZ
`-4 KHZ
`
`l KHZ
`-2 KHZ
`
`500 HZ
`4 KHZ
`
`0—500 HZ
`
`F7l(3.
`
`E3
`
`8
`
`

`

`US. Patent
`
`June 25,1991
`
`Sheet 7 of 10
`
`5,027,410
`
`180
`
`I8!
`
`I82
`
`
`
`
`
`
`INIT IALIZATION
`CODE
`
`MAIN
`PROGRAM
`
`INTERRUPT
`ROUTINE
`
`
`
`
`
`
`
`
`FIG. I0
`
`
`
`dB ENERGY CALCULATIONS
`FROM CHANNEL ENERGIES
`
`
`
`
`GAIN CALCULATION
`FOR EACH BAND
`
`CONVERT GAINS FROM
`dB TO LINEAR
`
`CALCULATE FILTER
`COEFFICIENTS
`
`TRACK NOISE AND
`PEAKS IN EACH BAND
`
`
`
`
`'85
`
`I
`
`I86
`
`I87
`
`I88
`
`I89
`
`
`
`
`
`I90
`
`I9|
`
`192
`
`
`
`RECALCULATE KNEES
`OF
`l/O CURVE
`
`
`
`
`CHECK MODE SWITCHES
`AND RESET PARAMETERS
`
`
`
`
`WAIT UNTIL MILLISEC
`COUNTER IS ZERO
`
`
`
`
` I93
`
`RESET MILLISEC
`COUNTER
`
`FIG.
`
`II
`
`9
`
`

`

`US. Patent
`
`June 25, 1991
`
`Sheet 8 of 10
`
`5,027,410
`
`200
`
`I
`
`20
`
`202
`
`203
`
`204
`
`CONTEXT SAVE
`FOR MAIN PROGRAM
`
`SAMPLE OUTPUT
`AND INPUT
`
`GAIN RANGE CODE
`
`DC FILTER
`
`PRE/DE EMPHASIS
`FILTER
`
`OCTAVE BAND
`ANALYSIS
`
`RECTIFY
`
`LOW PASS FILTER
`
`STORE ENERGY
`ESTIMATES
`FOR MAIN PROGRAM
`
`FIFO DELAY
`
`SPECTRAL FILTER
`
`DECREMENT
`MILLISEC COUNTER
`
`CONTEXT RESTORE
`FOR MAIN PROGRAM
`
`RETURN TO MAIN
`PROGRAM UNTIL
`NEXT
`INTERRUPT
`
`FIG.
`
`I2
`
`10
`
`205
`
`206
`
`207
`
`208
`
`209
`
`2|0
`
`2”
`
`I 2|2
`
`2|3
`
`10
`
`

`

`US. Patent
`
`June 25, 1991
`
`Sheet 9 of 10
`
`5,027,410
`
`mm
`
`
`
` .>>On_“3m__r.........1.
`
`
`fl........N«m......fl
`_II“"EN0__a$2_
`
`m_.9u.
`
`mm
`
`mmZE
`
`mo<z_-_._.z<
`
`m<3<-_._.z<
`
`EN5_m<1¢:m-mma
`
`
`
`H”5,59,.Nm
`
`_m
`
`om
`
`11
`
`
`
`
`

`

`US. Patent
`
`June 25, 1991
`
`Sheet 10 of 10
`
`5,027,410
`
`Acoustic feedback
`
`Speech
`
`Microphone
`xltl
`
`
`
`Hearing
`aid
`
`Oatput
`Signal
`
`
`
`
`signal
`processing
`
`Filter
`estimator
`
`
`
`' Microphone
`xlt)
`
`
`
`12
`
`l lI I lI ll
`
`'
`
`
`
`Oatput
`Signal
`
`Hearing
`aid
`signal
`processing
`
`ll lI| ll l ll
`
`Speech
`signal+
`
`12
`
`

`

`1
`
`5,027,410
`
`ADAPTIVE, PROGRAMMABLE SIGNAL
`PROCESSING AND FILTERING FOR HEARING
`AIDS
`
`FIELD OF THE INVENTION
`
`This invention pertains generally to the field of audio
`signal processing and particularly to hearing aids.
`BACKGROUND OF THE INVENTION
`
`The nature and severity of hearing loss among hear—
`ing impaired individuals varies widely. Some individu—
`als with linear impairments, such as that resulting from
`conductive hearing loss, can benefit from the linear
`amplification provided by conventional hearing aids
`using analog signal processing. Such aids may have the
`capacity for limited spectral shaping of the amplified
`signal using fixed low pass or high pass filters to com-
`pensate for broad classes of spectrally related hearing
`deficits. However, many types of hearing loss, particu-
`larly those resulting from inner ear problems, can result
`in non-linear changes in an individual’s auditory system.
`Individuals who suffer such problems may experience
`limited dynamic range such that the difference between
`the threshold hearing level and the discomfort level is
`relatively small. Individuals with loudness recruitment
`perceive a relatively small change in the intensity of
`sound above threshold as a relatively large change in
`the apparent loudness of the signal. In addition, the
`hearing loss of such individuals at some frequencies may
`be much greater than the loss at other frequencies and
`the spectral characteristics of this type of hearing loss
`can differ significantly from individual to individual.
`Conventional hearing aids which provide pure linear
`amplification inevitably amplify the ambient noise as
`well as the desired signal, such as speech or music, and
`thus do not improve the signal to noise ratio. The ampli-
`fication may worsen the signal to noise ratio where an
`individual’s hearing has limited dynamic range because
`the noise will be amplified above the threshold level
`while the desired speech signal may have to be clipped
`or compressed to keep the signal within the most com-
`fortable hearing range of the individual.
`Although hearing impaired individuals often have
`unique and widely varying hearing problems, present
`hearing aids are limited in their ability to match the
`characteristics of the aid to the hearing deficit of the
`individual. Moreover, even if an aid is relatively well
`matched to an individual’s hearing deficit under certain
`conditions, such as a low noise environment where
`speech is the desired signal, the aid may perform poorly
`in other environments such as one in which there is high
`ambient noise level or relatively high signal intensity
`level.
`
`SUMMARY OF THE INVENTION
`
`invention, digital
`In accordance with the present
`signal processing is utilized in a hearing aid system
`which is both programmable to fit the hearing deficit of
`a particular user and adaptive to the sound environment
`to maximize the intelligibility and quality of the audio
`signal provided to the user. Background noise levels are
`reduced in either a fixed or an adaptive manner to en-
`hance the signal to noise ratio of the desired signal, such
`as speech. The effective dynamic range of the user is
`expanded by maintaining high sensitivity for low inten-
`sity sound while providing long term automatic gain
`compression and output limiting control to insure that
`
`ll)
`
`15
`
`20
`
`25
`
`30
`
`35
`
`45
`
`50
`
`55
`
`60
`
`65
`
`13
`
`2
`the sound signal does not exceed the comfort level of
`the wearer. The majority of normal sound signals, such
`as speech, are thereby provided to the user at levels
`which will best fit the available dynamic range of the
`user’s ear. The audio signal provided to the user is also
`spectrally shaped to match and compensation for the
`specific spectral deficiency characteristics of the user’s
`ear. The signal processing hearing aid further has sev-
`eral modes selectable at the user’s choice which change
`the signal processing characteristics of the hearing aid
`to best accomodate the sound environment, such as the
`ambient noise level or the volume of the speech or
`music which the user wishes to listen to.
`The signal processing hearing aid includes a micro-
`phone preferably located near or at the ear of the
`wearer, associated analog filtering and amplifying cir-
`cuits, an analog to digital converter for converting the
`analog signal to digital data, a digital signal processor
`which operates on the digital data, a digital to analog
`converter for converting the processed data back to
`analog signal form, and analog filters and amplifiers
`which drive a receiver or speaker in an ear piece worn
`by the user. The signal from the microphone preferably
`receives pre-amplification and high pass filtering for
`pre-emphasis and is subjected to relatively slow auto-
`matic gain control to adjust the gain level to accommo-
`date slowly varying sound levels. Anti-aliasing low pass
`filtering of the analog signal is performed before analog
`to digital conversion. In digital form, the signal data
`may be subjected to selectable high pass filtering and
`pre- and de-emphasis filtering if desired in combination
`with time varying spectral shaping digital filtering. The
`spectral shaping filtering is performed in accordance
`with prescribed spectral characteristics matching the
`hearing deficit of the particular user for whom the hear-
`ing aid is prescribed. In addition, the parameters of the
`spectral filter are variable to adjust the amplification so
`that the signal level is best matched to the expressed
`preference of the individual user, preferably with ex—
`pansion of low level signals, normal amplification of
`intermediate level signals, and compression of high
`level signals. The processed digital data is then con-
`verted back to analog form and anti-imaging low pass
`filtering is performed on the signal before it is amplified
`and delivered to the speaker. The digital signal proces-
`sor preferably has a programmable read only memory
`which can be programmed with the desired spectral
`shaping characteristics and variable amplification char-
`acteristics that fit the user.
`The Spectral filter of the digital signal processor has
`filter parameters which can be varied to provide a non-
`linear input-output characteristic in several frequency
`ranges. The input-output characteristics preferably in-
`clude several piecewise linear sections. For example, a
`first section may have a slope greater than one to pro-
`vide expansion of low level signals. At a first knee point,
`the slope of the input-output characteristic changes to a
`one to one or linear input—output relationship which is
`maintained up to a second knee. The range of output
`levels between the two knees preferably corresponds to
`that chosen by the user, usually a best fit to the dynamic
`range of the user’s hearing so that most of second knee,
`the slope of the input-output characteristic is less than
`one to provide compression to reduce the effect of
`over-range signals and minimize loudness discomfort to
`the user. An estimate of the level of background noise is
`preferably made from the energy envelope of the input
`
`13
`
`

`

`5,027,410
`
`3
`signal in various frequency ranges. This estimate of the
`noise is used to adjust the position of the first knee up or
`down and/or change the expansion ratio of the first
`section, with the calculated gain in the various fre-
`quency ranges being used to reduce the noise compo-
`nent of the amplified signal being supplied to the user.
`The slopes of the input-output curve above and below
`the knees may be changed and the initial position of the
`upper and lower knees may be changed in different
`modes of operation of the hearing aid to best accommo-
`date the preference of the user as to the desired charac-
`teristics of the perceived sound, such as intelligibility,
`loudness or quality. For example, one set of slopes and
`knee values may be utilized in one mode while a second
`set of slopes and knee values may be used in another
`mode.
`The time constants of the non-linear gain functions
`over which the gain at various frequencies remains
`substantially unchanged is an important characteristic
`which affects system performance. The longer the time
`constant,
`the less compression of short
`term level
`changes is achieved. However, the shorter the time
`constant, the more distortion is introduced for a given
`expansion or compression ratio. In the system of the
`present invention, different time constants may be used
`for the energy analysis in the different frequency bands.
`Preferred values for the time constants range from 4
`milliseconds (ms) to 8 ms for the lowest frequency
`bands to 0.5 ms to 1 ms for the highest frequency band.
`Time constants in these ranges allow compression up to
`about 3.3 to 1 and expansion down to about 1 to 2 while
`keeping distortion at an acceptable level. The accept-
`able level of distortion depends upon the user, and more
`compression and/or expansion are acceptable to some
`users.
`.
`
`10
`
`15
`
`20
`
`25
`
`30
`
`35
`
`4
`of the spectral filter are updated in this manner to best
`accommodate the filter to the incoming signal. Because
`only estimation of the energy in the signal in each of the
`frequency bands is carried out, less computation is nec-
`essary than would be required for filtering the full signal
`in each of the frequency bands. A particularly signifi-
`cant advantage of the present system is that the input
`signal data passes through only one processing block,
`the time-varying spectral filter. In such a system there is
`less opportunity for quantization noise to enter the sig-
`nal than in prior systems which split the input signal
`into several frequency bands which are operated on
`separately and then recombined to form the output
`signal, and the present system is less subject to distor-
`tion than such prior systems.
`The energy analysis in each of the frequency bands is
`preferably carried out by operating on the input signal
`data and dividing it into two halves by a high-pass/low-
`pass pair of filters. Each of these filters contains half the
`band-width of the signal, so the rate at which the com-
`putations must be carried out can be reduced from the
`rate for the computations required to analyze the entire
`input signal frequency range. The high-pass half of the
`signal, containing the higher frequencies, is one octave
`wide and the energy in it can be estimated by a simple
`rectify and low-pass filter operation. The low-pass half
`of the signal is again filtered by a high-pass/low-pass
`pair of filters. Because the sampling rate has been
`halved, the cut off points of the digital filters are halved
`in frequency. The output from a high-pass filter can be
`rectified and low-pass filtered to estimate the energy in
`the frequency band and the output of the low-pass filter
`can again be filtered by a high-pass/low-pass pair of
`filters. In this way, it is possible to successively calcu-
`late the energy in portions of the input signal in nar-
`rower and narrower frequency bands, with lower and
`lower sampling rates. The high-pass and low-pass filters
`can be implemented by simple digital filters having e.g.,
`coefficients — l, 2, —1 and l, 2, 1. Such filters are fairly
`shallow with 12 dB per octave rolloff, but have only
`three simple integer taps each and the sum of the two
`filters is flat across the spectrum. More complicated
`filters also can be used. The number of operations
`needed in the energy analysis circuit is very small,
`mainly because so much of the filtering is done on
`streams of data which have been decimated to succes-
`sively slower rates. Consequently, the computation time
`required to determine the coefficients is reduced and the
`amount by which the input signal must be delayed to
`match the coefficients is also reduced, allowing the
`processing system to function in a real time manner,
`with a processing delay which is inperceptible to the
`listener.
`Further objects, features, and advantages of the in-
`vention will be apparent from the following detailed
`description when taken in conjunction with the accom-
`panying drawings.
`DESCRIPTION OF THE DRAWINGS
`
`45
`
`50
`
`In a preferred embodiment, the time varying spectral
`filter is a digital filter having filter coefficients which
`can be varied on a quasi-real time basis to accomplish
`nonlinear amplification within the spectrum of the hear-
`ing aid to best accommodate ambient signal levels and
`noise levels. A single spectral filter is utilized which
`receives the digital data corresponding to the input
`signal after a time delay sufficient to accommodate the
`time required to calculate the coefficients which match
`the data being processed by the digital spectral filter.
`The digital signal processor carries out the computation
`of the spectral filter coefficients by first band pass filter-
`ing the input signal digital data to provide several sets of
`digital data corresponding to the portions of the signal
`lying within certain frequency ranges, e.g., 0 to 500 Hz,
`500HztolkHz, lkHztoZkHz,2kHzto4kI-Iz,and
`4 kHz to 8 kHz, assuming that the frequency content of
`the input signal is limited to approximately 8 kHz. The
`energy in each of the frequency range limited signals is
`then estimated, such as by taking the absolute value of 55
`the data and then low pass filtering it, and this energy
`estimate is then utilized as described above to determine
`an appropriate gain for the portion of the signal con-
`tained within that frequency range. Each frequency
`band may also have a baseline gain which is set to shape
`the frequency response of the system to compensate for
`spectrally related hearing deficiences of a particular
`user. The calculated values for the gains are then used
`by the system to calculate the filter coefficients, such as
`in a finite impulse response filter implementation, for
`the time varying spectral filter. The coefficients of the
`filter are then changed and the delayed input signal data
`is then provided to the spectral filter. The coefficients
`
`65
`
`In the drawings:
`FIG. 1 is an illustrative view showing the major com-
`ponents of the adaptive signal processing hearing aid of
`the present invention as worn by a user.
`FIG. 2 is a schematic block diagram of the hardware
`components of the adaptive signal processing hearing
`aid of the invention.
`
`14
`
`14
`
`

`

`5
`FIG. 3 is a signal flow diagram showing the opera-
`tions performed on the signals from the microphone to
`the speaker in the hearing aid of the invention.
`FIG. 4 is a graph showing the gain function charac-
`teristics for determining the gains to be used in calculat-
`ing the digital spectral filter coefficients of the digital
`signal processor within the hearing aid.
`FIG. 5 is a graph showing the relationship between
`estimated energy within a frequency band in the input
`signal and the gain for that frequency band to be used in
`calculating the coefficients of the digital spectral filter.
`FIG. 6 is a graph similar to that of FIG. 4 showing
`the effect of a change in the lower knee level as a result
`of changes in the background noise level in the signal.
`FIG. 7 is a graph illustrating the changes in the ampli-
`tude envelope (or signal energy) of a typical signal
`within a specified frequency range and the manner in
`which the noise and peak levels of the signal are esti—
`mated.
`FIG. 8 is a schematic block diagram showing an
`implementation for the time varying spectral filter func-
`tion in accordance with the present invention.
`FIG. 9 is a schematic block diagram illustrating a
`preferred implementation of the energy analysis func-
`tion of the time varying spectral filter.
`FIG. 10 is a flow chart showing the program blocks
`in the programming of the digital signal processor
`which carry out the time varying filter processing of
`FIG. 8.
`FIG. 11 is a flow chart showing the main program
`portion of the, processing system of FIG. 10.
`FIG. 12 is a flow chart showing the interrupt routine
`program of the processing system of FIG. 10.
`FIG. 13 is a schematic block diagram showing the
`hardware components of the ear piece portion of the
`hearing aid system of the present invention.
`FIG. 14 is a schematic block diagram showing one of
`adaptive suppression of acoustic feedback.
`,
`FIG. 15 is a schematic block diagram showing an-
`other form of adaptive suppression of acoustic feed-
`back.
`
`.
`
`DESCRIPTION OF THE PREFERRED
`EMBODIMENT
`
`An illustrative view of one style of an adaptive, pro-
`grammable signal processing hearing aid in accordance
`with the present invention is shown generally in FIG. 1,
`composed of an ear piece 20 and a body aid or pocket
`processing unit 21 which are connected by a wiring set
`22. It is, of course, apparent that the healing aid can be
`incorporated in various standard one piece packages,
`including behind-the-ear units and in—the-ear units, de-
`pending on the packaging requirements for the various
`components of the aid and power requirements. As
`explained further below, the pocket processing unit 21
`includes a power on-'off button 24 and mode control
`switches 27. The mode switches 27 can optionally pro-
`vide selection by the user of various operating strategies
`for the system which suit'the perceived preference of
`the user. The mode switches allow the user to select the
`mode which best suits his subjective perception of the
`sound from the aid. As explained further below, the
`hearing aid system is programmable to adapt the signal
`processing functions carried out in each of the modes to
`the hearing deficit of the user for whom the hearing aid
`is prescribed. A volume control dial 28 is also provided
`on the ear piece 20 to allow user control of the overall
`volume level.
`
`15
`
`5,027,410
`
`6
`A hardware block diagram of the ear piece unit 20
`and pocket processor unit 21 is shown in FIG. 2. The
`ear piece includes a microphone 30 which can be of
`conventional design (e.g., Knowles EK3027 or Lectret
`SA-leO). The ear piece may also optionally include a
`telecoil 31 to allow direct coupling to audio equipment.
`The output signal from the microphone 30 or telecoil is
`provided to an analog pre-amplifier/pre-emphasis cir-
`cuit 32 which amplifies the output of the microphone
`(or telecoil) and provides some high pass filtering (e.g.,
`6 dB per octave) to provide a frequency spectrum flat-
`tening effect on the incoming speeéh signal which nor-
`mally has a 6 dB per octave amplitude roll off. This
`pre-emphasis serves to make the voiced and unvoiced
`portions of speech more equal in amplitude, and thus
`better suited to subsequent signal processing. In particu-
`lar, the pre-emphasis reduces the dynamic range of the
`speech signal and so reduces the number of bits needed
`in the analog to digital converter. The output of the
`pre-amplifier/pre—emphasis circuit
`is provided to an
`automatic gain control circuit and low pass filter 33.
`The automatic gain control (AGC) circuit attempts to
`maintain the long-term root-mean-square (RMS) input
`level at or below a specified value to minimize dynamic
`range requirements for the analog to digital converter
`which is used to convert the analog signal to a digital
`signal. Preferably, RMS inputs below 70—75 dB SPL (at
`4 kHz) are amplified linearly with about 40 dB gain,
`resulting in a 45 mV RMS signal level (e.g., 0.125 V
`peak to peak for a 4 kHz sine wave) which will be pro-
`vided to the analog to digital converter. Inputs between
`75 dB and 95 dB are maintained at the 45 mV level for
`the long term average. Inputs above 95 dB preferably
`have a gain less than 15 dB, and will be hard-clipped at
`the one volt peak to peak level. However, it is apparent
`that the total gain received by the listener can be se-
`lected either more or less than these values depending
`on the subsequent digital signal processing and the ana-
`log output stage.
`To minimize the interaction between speech modula-
`tion (syllabic) and the AGC circuit, the attack time is
`preferably approximately 300 milliseconds (msec) and
`the release time is approximately 2.5 seconds. This long
`term AGC function is desirable to allow the total gain
`to the user to be automatically adjusted to provide a
`comfortable listening level in situations where the user
`can control the signal level but not the noise level, for
`example, in using the car radio, watching television in a
`noisy environment, and so forth. The time-constants are
`chosen long enough so that the AGC is not affected by
`syllabic changes in speech level.
`The output of the automatic gain control circuit is
`provided on signal lines 34 (forming part of the connect-
`ing line 22) to the main body or pocket processor unit.
`21. The ear piece also receives an output signal on lines
`36 from the pocket processor. This output signal
`is
`received by a maximum power output control circuit 37
`which is adjusted by the fitter. The signal then is pro-
`vided to a low pass filter 38 and a power amplifier and
`volume control circuit 39 and finally to the'receiver
`transducer or speaker 40 (e.g., Knowles CI-1762) for
`conversion to a corresponding sound. The analog out-
`put power amplifier 39 (e.g., an LTC 551 from LTI,
`Inc.) determines the overall system gain and maximum
`power