INOA
`
`S$ 20050089181
`
`ao) United States
`(2) Patent Application Publication (0) Pub. No.: US 2005/0089181 A1
`(43) Pub. Date:
`Apr. 28, 2005
`Polk, JR.
`
`(54)
`
`MULTI-CHANNEL AUDIO SURROUND
`SOUND FROM FRONT LOCATED
`LOUDSPEAKERS
`
`(76)
`
`Inventor:
`
`MatthewS. Polk JR., Gibson Island,
`MD(US)
`
`Correspondence Address:
`STERNE, KESSLER, GOLDSTEIN & FOX
`PLLC
`
`1100 NEW YORK AVENUE, N.W.
`WASHINGTON, DC 20005 (US)
`
`Appl. No.:
`
`10/692,692
`
`Filed:
`
`Oct. 27, 2003
`
`Publication Classification
`
`(51)
`
`Write OL” spesssccoscccnssseseasicceis HOAR 5/00; HO4R 5/02
`
`(52
`
`USSG: stascncussunauainad 381/300; 381/27
`
`ABSTRACT
`(57)
`Asurround sound reproduction system uses a seriesof filters
`and a system of main and sub-speakers to produce phantom
`rear surround sound channels or a phantom surround sound
`effect from a loudspeaker system or pair of loudspeaker
`systems located in front of the listener. The sound system
`includes left and right surround input signals, and left and
`right front input signals. Left andright sub-speakers, andleft
`and right main speakers are located in front ofa listening
`location. Spacing between respective main and sub-speakers
`is approximatelyequal to ear spacing for an average person.
`The input to the left sub-speaker comprises the right sur-
`round signal subtracted from the left surround signal each
`signal having previously passed through a front-to-back
`filter anda series of high and lowpassfilters. The input into
`the left main speaker comprisestheleft front signal addedto
`the left surround signal after the left surround signal has
`passed througha front-to-backfilter. The input into the right
`sub-speaker comprises the left surround signal subtracted
`from the right surroundsignal each signal having previously
`passedthrougha front-to-backfilter anda series ofhigh and
`low pass filters. The input
`into the right main speaker
`comprises the right front signal added tothe right surround
`signal after the right surround signal has passed through a
`front-to-backfilter.
`
`
`AF >
`RS —
`Signal Inputs
`y7 iF
`Ls
`
`
`
`Left Front{LettFront
`Left
`
`
`
`Frontto
`Back Filter
`
`BackFilter
`
` Frontto
`
`
`
`
`APPLE 1137
`APPLE 1137
`
`1
`
`

`

`CL
`
`APPARENT
`SOURCE P|
`
`aS7
`
`
`
`
`
`PTJOTPAYSONT‘87AdyUONRINGNEUONRIYddyJAE
`
`Prior Art
`Fig. 1
`
`TV1816800/S007SN
`
`2
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 2 of 24
`
`US 2005/0089181 Al
`
`Left
`Surround
`
`Signal Inputs
`
`RF >
`Right Front
`
`RS—,
`Right
`Surround
`
`
`
`BackFilter
`
`
`
`
`Front to
`Back Filter
`
`pe
`
`
`
`
`
`
` Front to
`
`
` Low Pass
`
`
`
`Filter
`
`
`
`
`
`Low Pass
`
`Filter
`
`
`
`RSS
`
`| | | |
`
`Pi
`
`LMS
`
`Lss
`
`t+At
`
`t
`
`3
`
`

`

`Patent Application Publication Apr. 28, 2005 Sheet 3 of 24
`
`US 2005/0089181 Al
`
`Input Signals
`
`Signal Modification and Combination
`
`(RS'- Ls}
`
`Fig. 2a
`
`4
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 4 of 24
`
`US 2005/0089181 Al
`
`Ls
`
`Left
`Surround
`
`lr
`
`
`
`Signal Inputs
`Cc
`
`RF —
`Right Front
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`RS
`
`Right
`Surround
`
`
`Front to
`BackFilter
`
`
`
`
`
`
`
`
`Lss
`
`t+At
`
`RSS
`
`——~
`|
`a,
`|
`
`|
`/\
`
`5
`
`

`

` UM-DECIBELS pceeT
`
`
`Tony EIN M3
`“EERSPUNayns
`
`
`
`
`Ee
`
`Somesz
`
`
`rr Tae PACH
`
`HesiTANGPiAApefeeteteae
`|||LeerRa" LLLLNsst
`
`
`
`
`
`
`\ H\HvateNySAWBSWSS
`
`“6JERfl
`EH ||FH
`TRANSFORMA t
`CHAEHe
`
`NEY
`
`Pay
`CECACEA
`TV1816800/S007SO
`
`
`
`a2
`o3 0405 oF
`1 12
`oO
`ind
`2
`S
`id
`FREQUENCY - kHz
`Fig. 2) Transformation of sound pressure level from the free Geld to the humas eardrum in
`thhclpane foc foeaBe 1S)imervale ofazimuth €. Caves show
`average valucs based onca dam from (2? studiet. (After Ref. 6).
`Fig. 3
`
`
`
`
`
`
`
` TIONOFSOUNDPRESSURELEVELFROMFREEFIELDTOEARDA
`
`
`
`Ne
`
`ey
`i)
`
`
`
`FTJOSWAYSSOOT‘SZAdyUONRINGNGUOHwIddyJOE
`
`
`
`
`
`
`
`
`
`6
`
`

`

`
`
`
`
`£7J99WIYSSOOT‘SZAdyUOHRIyGngUONRIyddyjude
`
`
`
`
`
`TV1816800/S007SN
`
`Fig. 4
`Near Ear - Front to Back Frequency ResponseDifferences
`
` —— 0-180 deg.
`
`--- -15-1465d g.
`— - -30-150d g.
`— - -45-135d g.!
`— — 60-120d g.|
`——.75-105d g.|
`
`| |
`
`
`
`
`
`7
`
`

`

`-- = .15-165d 2.
`i — - -30-150 deg.|
`|- - - 45-135 deg.
`/— — 60-120d g.|
`6.0
`
`|_——75-105dg.|
`8.0
`
`
`
`
`
`
`
`
`
`—— 0-180 deg.
`
`Fig. 5
`Far Ear-Front to Back Frequency ResponseDifferences
`
`
`
`PTILPAYSS007‘87AdyUONRINGNEUONRINddyJAE
`
`-10.0
`
`-12.0
`
`Frequency, Hz
`
`TV1816800/S007SN
`
`8
`
`

`

`Fig. 6
`Front to Back Frequency ResponseDifferences
`Far Ear vs Near Ear
`
`
`
`
`
`S
`er
`
`££
`
`S
`SF
`
`S
`S
`©
`S
`s
`S S ss
`AS
`AAS
`a> ? ae
`Frequency, Hz
`
`©
`S
`S
`Ss
`S
`S
`ro o oe
`
`9
`gs
`g*
`
`S
`SS
`ww
`
`
`
`
`----18-465d g.
`
`|. -30-150d g.
`|_ . .45-135 deg.
`
`— —60-120d g.
`|__75-105d g..
`
`
`
`
`
`FTJORDIYSSHOT‘SZAdyUONRANQngUONRIyddyJag
`
`
`
`
`
`TV1816800/S007SN
`
`9
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 9 of 24
`
`US 2005/0089181 Al
`
`Signal Inputs
`
`
`
`
`Front to
`Backfilter
`45-135 deg.
`
`LMs
`
` Left
`
`Surround
`
`
`
`
`
`
`
`Surround
`
`
`
`Front to
`Backfilter
`
`RMS
`45-135 deg.
`
`OY
`
`| |
`
`45 de
`
`135 deg.
`
`{~
`
`S
`
`.ay me u
`
`7")
`aa
`
`Fig. 7
`
`10
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 10 of 24
`
`US 2005/0089181 Al
`
`LS
`
`Left
`Surround
`
`Front to Back Filter
`45-135 deg.
`
`Low
`
`P
`
`+
`
`1
`
`/
`
`Signal Inputs
`
`RS
`
`:
`
`Low Pass
`Filter
`
`Invert
`
`Low Pass
`Filter
`
`Right
`Surround
`
`Front to Back Filter
`45-135 deg.
`
`=
`
`saat
`Filter
`
`nm
`
`8
`
`9
`
`=
`
`&)
`(>
`rT a
`12
`(anne LSE
`
`Ree
`
`L]
`fs
`
`10
`
`o
`
`©
`
`+(
`
`:
`
`Ll
`os
`
`LMS
`
`LSS
`
`—
`
`—
`
`\
`
`\
`
`(
`\
`
`\
`
`\
`
`\
`
`Le
`LL
`
`\
`
`\
`
`\
`
`\
`
`\
`
`RSL
`“
`
`“
`
`/
`
`RSS
`
`_
`
`y
`
`/
`
`/
`
`\
`|
`
`/
`
`|
`|
`—
`
`CLA
`
`<P
`
`RMS
`
`Re
`
`|
`
`S74
`
`|
`Cie,
`Approximate Range of Perceived
`Sound Locations from Input Signals
`LS and RS
`
`NX ~
`
`~ ne
`
`4
`
`A“
`
`-”
`
`Fig. 8
`
`11
`
`11
`
`

`

`
`
`PTJOTLANSSOOT‘SZAdyWONRINGNgUONRIddyjog
`
`
`
`
`
`TV1816800/S007SO
`
`Fig. 9
`Left to Right Frequency Response Differences
`
`—— 150 to-150
`
`[....30to-30
`| . .45 to-45
`|
`| - -60to-60
`'— — 120 to-120|
`aae
`
`S
`D>
`&
`&
`&
`&
`£
`S
`ef FF LFS SS gs
`s
`S
`Ss
`3
`S Ss
`
`Frequency, Hz
`
`12
`
`12
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 12 of 24
`
`US 2005/0089181 Al
`
`LS
`[— LF
`Signal Inputs
`RF al
`RS
`
`
`
`
`Right[_LeftFront|[FightFront)(Plott
`Left Front
`Right Front
`Left
`
`
`
`Ss
`
`4
`
`Front to
`
`‘
`
`Frontto
`
`«
`
`(2)
`
`5
`
`
`
`
`<7?essFilter
`
` Low Pass
`
`
`
`
`Filter
`
`Low Pass
`Filter
`
`7
`
`
`Low Pass
`Filter
`
`+
`
`WE
`
`7
`
`1
`
`1
`
`13
`
`Low Pass
`Filter
`
`Left-Right
`Filter
`
`CoP
`V/A
`
`LMS
`
`~~
`
`LSS
`
`13
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 13 of 24
`
`US 2005/0089181 Al
`
`LF
`
`Signal Inputs
`g
`Pp
`
`LS
`
`Left
`Surround
`
`1
`
`Front to
`Back Filter |+
`
`UY
`
`+
`
`--3
`
`5
`
`4
`
`6
`
`an
`RF
`Right Front
`
`RS
`
`Right
`Surround
`
`
`
`Front to
`Back Filter
`
`
`
`Low Pass
`
`
`
`| Filter
`
`Low Pass
`
`Filter .
`
`| | I |
`
`Py
`
`14
`
`LMS
`
`LSS
`
`t
`
`trAt
`
`RSS
`
`14
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 14 of 24
`
`US 2005/0089181 Al
`
`Signal Inputs
`Be
`—
`g
`f~
`p
`c
`
`(tatFon)[FiFon)
`Right
`Right Front
`Left Front
`Left
`
`Ls
`
`LF
`
`RF
`
`1
`
`Frontto
`Back
`
`V
`
`+
`
`S
`
`
`
`Front to
`BackFilter
`
`Filter
`+
`
`Pew Paes
`Filter
`
` Low Pass
`
`Filter
`
`Filter
`
`
`5 ; LowPass
`LowPass ;
`es) == se@re
`
`(aa=\
`
`SE ¢eMat aN
`
`ope RSE
`
`PCS
`
`RMS
`
`At
`
`Lss
`
`t i t+At
`|
`cla
`
`_ RSS
`
`|
`
`Le S
`
`Re
`
`15
`
`15
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 15 of 24
`
`US 2005/0089181 Al
`
`LS
`
`Left
`Surround
`
`Low Pass
`Filte
`a
`+
`
`[~ LF
`
`Signal Inputs
`
`AF —)
`Right Front
`
`RS —
`Right
`Surround
`
`:
`
`5
`
`4
`
`Front to
`BackFilter
`
`7
`
`Low Pass
`Filter
`| =
`
`ie
`
`Low Pass
`Filter
`
`5
`
`Low Pass
`it
`Filter
`
`+
`
`10
`
`
`
`
`SL<S
`bm oD
`
`17
`
`ip
`
`- LSs
`
`RMS
`
`At‘max
`
`t+At
`
`RSS
`
`16
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 16 of 24
`
`US 2005/0089181 Al
`
`Input Signals
`
`LS
`
`LF
`
`RF
`
`RS
` Signal Modification and Combination
`
`(LF-RF’)
`|
`|!
`|
`(RF’-LF’)
`
`(Ls’ : RS’)
`
`(LS’+LF) |
`
`RSS
`
`| | |
`
`Py
`
`t+At
`
`LSS
`
`Le
`- |
`
`SeRe
`
`Fig. 13a
`
`17
`
`17
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 17 of 24
`
`US 2005/0089181 Al
`
`-_—_— OTT
`-” = Approximate Rangeof
`a Perceived SoundLocations
`from Input Signals LF and RF
`
`7
`
`se
`
`RSE
`
`PFS
`
`™
`
`“~
`
`N“
`
`-
`
`|
`- om
`|
`
`7S
`
`RMS
`“NX
`
`/
`
`/
`
`/
`
`/
`|
`
`Lvs
`iss
`/
`
`—=
`
`(
`
`‘“
`
`S
`
`Le
`
`ola
`
`Re
`
`\ ‘
`
`.
`PRSL /
`
`\
`
`\
`
`\
`
`rss
`
`a be
`
`\
`
`7
`
`\
`
`\
`
`Approximate RangeofPerceived
`SoundLocations from Input Signals
`LS and RS
`
`/
`
`/
`
`18
`
`18
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 18 of 24
`
`US 2005/0089181 Al
`
`
`Front to
`
`
`
`Low Pass
`Filter
`
`
`
`Back Filter See e eee ee eee ew ese dew es
`
` +
`
`19
`
`19
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 19 of 24
`
`US 2005/0089181 Al
`
`PFS
`
`——_—
`_
`—
`- = Approximate Range of
`Perceived Sound Locations
`“a”
`from Input Signals LF and RF
`
`ee rse~”)
`w
`
`“
`
`7—_—
`
`/
`
`/
`
`|
`LL
`
`iMS ~
`iss}
`4
`
`_-_ i — ~ RMS
`
`|
`
`CLA
`
`Fig. 16
`
`20
`
`20
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 20 of 24
`
`US 2005/0089181 Al
`
`RS
`RF =
`Signal Inputs
`i LF
`LS
`
`
`
`Left
`
`Rc[RightFront]
`Right
`Right Front
`Left Front
`Surround
`
`
`1
`
`Rear
`Center
`
`+
`
`+
`
`!
`
`4
`
`
`
`Front to
`<7?
`Center Front
`VC
`Front to
`
`
`
`
`
`
`Back Fitter [+™XS : 8 as BackFilter
`
`
`
`Low Pass
`Filter
`
`+
`
`<)
`
`+
`
`8
`
`High-Pass
`:
`Filter
`
`T2
`A
`
`Lss
`
`Bay
`
`15
`
`Lse
`
`LMS
`
`t
`
`
`
`7
`
`1
`13
`
`
`
`
`
`
`
`
`
`
`Invert
`
`——
`
`8
`
`Low Pass
`
`12
`on
`
`=
`HighPass
`Filter
`
`.
`Filter
`
`Right-Left
`Filter
`
`16
`RSE
`
`RMS
`
`Filter
`RSS
`
`:
`Filter
`
`Left-Right
`Filter
`=
`
`21
`
`21
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 21 of 24
`
`US 2005/0089181 Al
`
`LF
`
`i
`
`Signal Inputs
`g
`Pp
`
`RF —
`
`Right Front
`
`
`
`
`
`Center Front
`to BackFilter
`
`RSS
`
`| | | |
`
`Psy
`
`
`
`Lss
`
`22
`
`22
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 22 of 24
`
`US 2005/0089181 Al
`
`Signal Inputs
`
`Right
`
`Dn
`
`fos
`
`a
`
`oe,
`
`Filter
`
`Filter
`
`— —
`
`e Giavetou Range ofPerceivedSound
`Locations from Input Signals LFand RF
`
`~
`
`RMS
`
`\
`
`\
`
`CLA
`
`LSE?)
`
`4
`
`/
`
`/
`
`LNS
`/
`Lss
`|
`/
`Le
`
`_
`
`a
`
`LF
`RF — RS ~
`Left
`[
`Surround
`Left Front
`aFront
`as
`Le
`[Invert|[Invert] Je
`Pae} |LowPass
`Hayeass
`LowPass
`ve
`or
`yeelaSePFSL
`rN
`—_—_———_—o _— —
`Aipccuinats Range of PerceivedSound
`Locations from InputSignalsLSandRS
`ee inc
`nt
`|
`~
`I
`—~
`
`0
`
`z,
`\
`\
`
`\
`
`\
`RSS
`\
`SLD
`
`PSSL
`
`_
`
`Re
`
`ig
`et
`
`|
`
`Fig. 19
`
`23
`
`23
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 23 of 24
`
`US 2005/0089181 Al
`
`Input Signals
`
`LS
`
`LF
`
` v
`
`eS eee
`
`"
`
` (RS’
`
`
`\ ‘ RSS
`
`
`
`
`Approximate Range of
`Perceived Sound
`Locations from Input
`Signals LF and RF
`
`Lss
`
`Fig. 20
`
`24
`
`

`

`Patent Application Publication Apr. 28,2005 Sheet 24 of 24
`
`US 2005/0089181 Al
`
`Original 2-ChannelInput
`Signals
`
`L
`
`R
`
`Two Channel to Multi-Channel
`Conversion
`
`LS
`
`LF
`
`RF
`
`RS
`
`
`‘Signal Modification and Combination
`
`(LS'- RS’)
`
`LSS
`
`(LS'+LF)
`
`(RS'+RF)
`
`(AS'-LS’) LMS
`
`25
`
`25
`
`

`

`US 2005/0089181 Al
`
`Apr. 28, 2005
`
`MULTI-CHANNEL AUDIO SURROUND SOUND
`FROM FRONT LOCATED LOUDSPEAKERS
`
`BACKGROUND OF THE INVENTION
`
`[0001]
`
`1. Field of the Invention
`
`[0002] This invention relates generally to the reproduction
`of sound in multichannel systems generically known as
`“surround-sound” systems and more specifically to the
`application of psychoacoustic principles in the design of a
`loudspeaker system for reproducing a surround sound expe-
`rience from loudspeakers located only in front of thelistener.
`
`[0003]
`
`2. Background Art
`
`It has long been recognizedthatit is possible to use
`[0004]
`interaural crosstalk cancellation (IACC) and headrelated
`transfer functions (HRTF) to expand the perceived sound-
`stage of a two channel audio system or to create the illusion
`of sounds coming from phantom locations independent of
`the actual location of the loudspeakers. Through the 1970's
`and 1980's a number of audio components were available
`for purchase which used IACC to expand the perceived
`soundstage. However, until the availability of inexpensive,
`powerful digital signal processing (DSP) more accurate
`generation of phantom sound sources at specific locations
`was very difficult and costly due to the complexity of
`accurate HRTF synthesis.
`
`availability of DSP and
`recently the
`[0005] More
`improvedfiltering algorithms has made it possible to create
`a phantom sound source in almost any location using just a
`single pair of loudspeakers typically located in front ofthe
`listener. Using variations of the same techniquesit is pos-
`sible to create several phantom sound sources at the same
`time from a single pair of loudspeakers typically located in
`front of the listener. This technique has many practical
`applications. For example, the experience of having front,
`rear and center speakers as in a complete 5.1 surround sound
`audio system can be simulated using a single pair of loud-
`speakers or headphones.
`[0006] These techniques are based on the way in which
`human beings process sounds received by their ears to
`determine the location of the sources of those sounds.
`In
`
`general, we hear the direction of sounds based on two
`primary mechanisms,
`Interaural Time Delays (ITD) and
`Interaural Level Differences (ILD). ITDrefers to the addi-
`tional time required for a sound locatedto one side ofthe
`listeners head toarrive at the opposite side ear as compared
`to the time required to reach the near side ear. The ITD of
`a soundallowsthe listener to determinethe lateral direction
`
`of a sound with great precision. ILD refers to the difference
`in perceived intensity between the listeners two ears for a
`sound arriving from a particular location. For example, a
`sound located to the listeners left would appear generally
`louder in the left ear as compared to the right ear due to a
`reduction in loudness as the sound passes across the listen-
`er’s head. Overall
`intensity differences between the ear
`reinforce lateral localization of sounds through ITD’s.
`In
`addition, sounds arriving from a particular direction produce
`a complicated frequency response pattern at each ear which
`is characteristic of that specific directional
`location. The
`combination of these characteristic directional frequency
`response curves and the ITD’s associated with sounds
`
`arriving from that direction are referred to as Head
`[0007]
`Related Transfer Functions (HRTF). The frequency response
`
`component of the HRTF’s is quite complex and somewhat
`different for eachindividual. It is the detailed structure ofthe
`
`HRTFfrequency response at each earthat allows the listener
`to determine the elevation of a sound and whether it is in
`
`front or behind. For example, a sound source located 60
`degrees to the left and in front ofthe listener has the same
`ITD (approx. 300 ms) as a sound source located at 60
`degrees left and behind the listener. However, the asymme-
`try ofthe outer ear produces very different HRTF’s for those
`two sound source locations thereby allowing the listener to
`
`determine both thelateral location and front versus
`[0008]
`back. A similar mechanism allowsthe listener to determine
`the approximate elevation of a soundsource. In general the
`mechanismfor determining lateral location of sounds based
`on ITD’s operates in the frequency range of approximately
`150 Hz to 1,200 Hz. The mechanisms for localizing sounds
`based on the frequency response of HRTF’s operates from
`approximately 500 Hz to above 12,000 Hz.
`[0009] Based on these principles various methods have
`been devisedfor canceling interaural crosstalk in loudspeak-
`ers, generating phantom sound sources from monaural sig-
`nals using synthetic or measured HRTF’s and for using
`HRTF’s to create phantom rear channels for an audio
`surround sound system from only a front pair of speakers.
`
`In general, methods using HRTF’s to create phan-
`[0010]
`tom sound sources, whether for simulation of a surround
`soundaudio system or other application, have a number of
`practical limitations. Accurate representation of HRTP’s is
`very computation intensive andit is therefore difficult
`to
`obtain sufficient accuracy using practical and cost efficient
`DSP methods. For example, U.S. Pat. No. 6,173,061, which
`describes a method for phantom sound source generation
`using HRTF’s, acknowledges the need for more efficient
`sound processing algorithms and seeks to address this prob-
`lem. Additionally,
`the specific HRTF’s used in prior art
`methods are selected on the basis of assumptions regarding
`the characteristics of the loudspeakers employed, the spe-
`cific positional relationship between the loudspeakers and
`the listener, and the variation of actual HRTF’s from listener
`to listener. Given the highly specific and detailed nature of
`HRTP’s, those skilledin the art will recognize that changes
`in the loudspeaker characteristics or locations combined
`with movement of the listener away from the assumed
`listening location can easily destroy the phantom sound
`source illusion. Also, the actual HRTF’s of somelisteners
`maybe too different from the HRTF’s employed
`[0011]
`in the device for the illusion to work. For example,
`U.S. Pat. No. 4,893,342 and its related patents describe
`methods for increasing the positional flexibility of an HRTF
`based method by limiting the frequency range of the HRTF
`representations to a range of approximately 600 Hz to 10
`kHz and methods for determininglistener tolerant HRTP’s.
`[0012] Some known methods for creating phantom sound
`locations and sourcesrely on the use ofbinaurally recorded
`signals or other specially recorded signals as inputs. These
`methods may be subject to the above described limitations
`and will also function properly only when using input
`signals made with the specified recording scheme. For
`example, U.S. Pat. No. 4,199,658 describes such a method
`based on the use of binaurally recorded signals as inputs.
`[0013] Finally, most known methods for creating phantom
`rear channel sound sources seek to reproduce the illusion
`
`26
`
`26
`
`

`

`US 2005/0089181 Al
`
`Apr. 28, 2005
`
`loudspeakers are located at specific locations
`that actual
`listener. Such methods are disclosed,
`for
`behind the
`example, in U.S. Pat. No. 6,052,470 and its related patents
`which describe various methods for using HRTF’s to create
`the illusion ofa pair ofspeakers located behind the listener.
`However, those skilled in the art generally agree that in rear
`channel sound reproduction for an audio surround sound
`system, diffuse localization is preferable to the type of
`specific localization provided by actual rear located direct
`radiator loudspeakers. Furthermore, as will be understood by
`those skilled in the art, audio surround sound systems
`composed offront andrear pairs of speakers are not effective
`in localizing sounds in the general areas
`
`directly to the left and right of a listener located
`[0014]
`centrally between the two pairs of speakers.
`
`there exists a need for methods for
`[0015] Therefore,
`creating phantom rear surround sound channels which
`require less complicated signal processing, which are more
`tolerant of loudspeaker characteristics, loudspeaker place-
`ment, listener location andlistener to listener HRTF varia-
`tions, whichare effective when using commonlyavailable
`recordings and which are capable ofdiffuse localization of
`rear channel soundsin an audio surround sound system over
`a range of locations aroundthe listener.
`
`BRIEF SUMMARY OF THE INVENTION
`
`is an object of this invention to
`it
`[0016] Therefore,
`provide a device and methodfor producing phantom rear
`surround soundchannelsor a phantom surround sound effect
`from a loudspeaker system or pair of loudspeaker systems
`located in front of the listener. An additional object of this
`invention is to permit implementation using simple analog
`filters or simple DSP.
`It
`is another object of the present
`invention to be more tolerant of loudspeaker characteristics,
`loudspeaker placement,
`listener location and listener to
`listener variation. Yet another object of this invention is to
`create effective surround sound reproduction when using
`commonly available audio surround sound recordings. A
`further object ofthis invention is to generate phantom sound
`sources that are perceived
`
`as originating from a range ofdifferent locations
`[0017]
`around or behind the listener including the general areas
`directly to the left and right of the listener.
`
`[0018] U.S. Pat. Nos, 4,489,432; 4,497,064; 4,569,074
`and 4,630,298 disclose a method for using an arrangement
`of main and sub-speakers in a stereo sound reproduction
`system to cancel [ACandto producea realistic acoustic field
`extending beyond the loudspeaker locations using signals
`from commonly available stereo recordings. ‘The disclosures
`of these patents are incorporatedherein in their entirety by
`reference. For example, prior art FIG. 1 (FIG. 10 of U.S.
`Pat. No. 4,489,432) shows specifically how an arrangement
`of main and sub-speakers can be used to create a phantom
`sound source outside the boundaries of the loudspeaker
`locations fromtwo input signals. Basedon the disclosures of
`US. Pat. Nos. 4,489,432; 4,497,064; 4,569,074 and 4,630,
`298it will be apparent tothose skilled in the art that a system
`constructed in accordance with these disclosures is capable
`of creating phantom sound sources anywhere in front of the
`listener more or less independent of the loudspeaker loca-
`tions according to the localization information contained in
`the two recorded signals used as inputs. The methods
`
`described in these patents are also capable of creating a
`stable sound image whenno localization information exists
`in the two recorded signals usedas inputs.
`
`In accordance with one embodimentofthe present
`[0019]
`invention, in an audio reproduction system having at least
`four inputs for accepting at least four audio input signals, for
`example,
`left
`front,
`right
`front,
`left surround and right
`surround channel signals, a right main speaker and a left
`main speaker are providedrespectively at right andleft main
`speaker locations along a speaker axis which are equidis-
`tantly spaced from the principle listening location. The
`principle listening location LL is generally defined as a
`spatial position for accommodating a listener’s head facing
`the main speakers along a central listening axis and having
`a right ear location and a left ear location along anear axis,
`with the right and left ear locations separated by a maximum
`interaural sounddistanceofAt,,,,,. and the principle listening
`locationis specifically defined as the point on the ear axis
`equidistant to the right and left ears. The central listening
`axis CLA is defined as a line passing through the principle
`listening location and a point on the speaker axis equidistant
`from the right and left main speakers. A right sub-speaker
`and a left sub-speaker are provided at right and left sub-
`speaker locations substantially on the speaker axis of the left
`and right main speakers and which are equidistantly spaced
`from the principle listening location LL. By careful location
`of the sub-speakers relative to the main speakers, use of
`proper modifications and combinations of the left and right
`surround signals to create driving signals for the main and
`sub-speakers, and appropriate filtering of the component
`parts of said driving signals, a listener located in the prin-
`ciple listening location LL perceives a surround sound
`experience from speakers located only in front of the lis-
`tener.
`
`BRIEF DESCRIPTION OF THE
`DRAWINGS/FIGURES
`
`[0020] FIG. 1 is a diagram illustrating an apparent source
`location as produced by the arrangement disclosed in U.S.
`Pat. No. 4,489,432, FIG, 10.
`
`FIG.2 is a diagram showinga first embodiment of
`(0021]
`the present invention,
`
`[0022] FIG. 2a is a diagram showing the signal combi-
`nations of a first embodiment of the present invention.
`
`[0023] FIG. 2bis a diagram showingthe addition of a fifth
`audio input signal to the first embodiment of the present
`invention.
`
`(0024] FIG. 3 shows a family of frequency response
`curves of sounds incident from various angular directions.
`
`[0025] FIG. 4 shows a family of frequency response
`curves showing frequency response differences between
`sounds incident from in front ofa listener and behind the
`listener, at the near ear of the listener.
`
`[0026] FIG. 5 shows a family of frequency response
`curves showing frequency response differences between
`sounds incident from in front of a listener and behind the
`listener, at the far ear of the listener.
`
`[0027] FIG. 6 shows a family of frequency response
`curves representing the differences between the front-to-
`back curves for the near ear shown in FIG. 4 and the
`
`27
`
`27
`
`

`

`front-to-back curves for the far ear shown in FIG, 5 for each
`[0045] FIG. 2 and FIG, 2a showafirst preferred embodi-
`mirror image front to back pair of soundlocations.
`ment of the present
`invention. Referring to FIG, 2, four
`audio signal inputs, for example only and not by way of
`limitation, corresponding to signal channels of a surround
`sound system are provided. It
`is understoodthat these may
`be any four audio input signals. However, for purposes of
`clarity and consistency these signals will be referred to
`herein as left surround signal LS;left front signal LF; right
`front signal RF;and right surroundsignal RS. Left andright
`loudspeaker enclosures, LSE and RSE are also provided.
`Left loudspeaker enclosure LSE contains at least one left
`main speaker LMSandat least one left sub-speaker LSS.
`Right loudspeaker enclosure RSE contains at least one right
`main speaker RMS and at least one right sub-speaker RSS.
`As is well known by those skilled in the art unmodified
`audio signals reproduced by a pair ofof loudspeakers, such
`as in a
`typical stereo audio system, are perceived by a
`istener sitting in front of the speakers as originating from a
`range of sound locations between the two loudspeakers.
`Therefore, sounds produced only by main left and right
`loudspeakers LMS and RMSare perceived by a listener
`locatedat principle listening location LL as originating from
`a range of sound locations approximately between and
`bounded by the actual
`locations of left and right main
`loudspeakers LMS and RMS.
`
`[0030] FIG. 9 shows a family of curves calculated by
`subtracting the frequency response shown in FIG. 3 for
`sounds arriving froma particular direction at the listener’s
`nearest ear from the frequency response for soundsarriving
`from the same direction at the listener’s farthest ear.
`
`[0031] FIG. 10 is a diagram showing a second embodi-
`ment ofthe present invention.
`
`[0032] FIG. 11 is a diagram showing a third embodiment
`of the present invention.
`
` [0036] FIG. 14 is a diagram showing approximate per-
`
`
`US 2005/0089181 Al
`
`Apr. 28, 2005
`
`[0028] FIG. 7 is a schematic diagram showing perceived
`rear sounds at a point location behindthelistener.
`
`[0029] FIG. 8 is a schematic diagram showing perceived
`apparent sound locations over a broad range of locations
`begin the listener when utilizing the present invention.
`
`[0033] FIG. 12 is a diagram showing a fourth embodiment
`of the present invention,
`
`[0034] FIG. 13 is a diagram showinga fifth embodiment
`of the present invention.
`
`[0035] FIG, 13a is a diagram showing the signal combi-
`nations ofa fifth embodiment ofthe present invention.
`
`ceived sound locations in front ofthe listener and apparent
`perceived sound locations to the rear ofthe listener when
`using a fifth embodiment of the present invention.
`
`[0037] FIG. 15 is a diagram showing a sixth embodiment
`of the present invention.
`
`[0038] FIG. 16 is a diagram showing approximate per-
`ceived sound locations in front of the listener when using a
`sixth embodiment of the present invention.
`
`[0039] FIG. 17 is a diagram showing an seventh embodi-
`ment of the present invention.
`
`[0040] FIG. 18 is a diagram showing approximate appar-
`ent perceived sound locations to the rear ofthe listener when
`using an seventh embodiment of the present invention.
`
`[0041] FIG. 19 is a diagram showing a eighth embodi-
`ment of the present invention.
`
`[0042] FIG. 20 is a diagram showing the signal combi-
`nations of a ninth embodiment of the present invention.
`
`[0043] FIG. 21 is a diagram showing a tenth embodiment
`of the present invention.
`
`DETAILED DESCRIPTION OF THE
`INVENTION
`
`Preferred embodiments of the present invention are
`[0044]
`now described with reference to the figures where like
`reference characters/numbers indicate identical or function-
`ally similar elements. While specific configurations and
`arrangements are discussed, it should be understood that this
`is donefor illustrative purposesonly. A person skilled in the
`relevant art will
`recognize that other configurations and
`arrangements can be used without departing from thespirit
`and scope ofthe invention.
`
`
`
`[0046] As shown in FIG. 2, a listener located at principle
`listening location LLhas a left car Le and a right ear Re. The
`midpoint between the left ear Le and the right car Re is
`located along a central listening axis CLA. Asnotedin U.S.
`Pat. No. 4,489,432, incorporated in its entirety by reference
`herein,
`the right and left ear locations are separated by a
`maximum interaural sound distance of At,,.,. As also
`explained in U.S. Pat. No. 4,489,432, and shownin FIG, 2,
`sound distance t
`is the time for sound from the left main
`
`speaker LMS toreach the left ear Le and sounddistance t+At
`is the time for sound from the left main speaker LMS to
`reach the right ear Re. Similarly, sounddistance t is also the
`time required for sound from right main speaker RMS to
`reach right ear Re and sound distance t+At is also the time
`for sound from the right main speaker RMS toreach theleft
`ear Le In similar fashion, t+At is also the time for sound from
`the right sub-speaker RSS to reach the right ear Re, and the
`time for sound fromthe left sub-speaker LSS to reach the
`left ear Le.
`
`[0047] Referring again to FIG.2,left surround signal LS
`passes through front-to-backfilter 1 and is combined with
`left front signal LF in adder 3. The combinedsignal is then
`transmittedto left main speaker LMS. Similarly, right sur-
`roundsignal RS passes through front-to-backfilter 2 and is
`combined with right front signal RF in adder 4. The com-
`binedsignal is then transmitted to right main speaker RMS.
`
` Front-to-back filters 1 and 2 modify the surround
`[0048]
`signals LS and RS such that, at the listeners ears and over a
`certain frequency range,
`they will approximate the fre-
`quency responseofsoundsignalsas if they originated from
`the rear of the listener, even thoughthey are being projected
`from the front ofthe listener, This modification is explained
`with reference to FIGS. 3 to 6. FIG. 3 shows a family of
`frequency response curves
`representing the
`frequency
`response at the ear drum ofa listener relative to free field
`conditions for soundsarriving from different angular sound
`locations in the horizontal plane. FIG. 4 shows another
`family of frequency response curves calculated by subtract-
`
`28
`
`28
`
`

`

`US 2005/0089181 Al
`
`Apr. 28, 2005
`
`ing the frequency response from FIG,3 for sounds arriving
`at the listener’s nearest ear of sound locations in front of the
`
`listener from the frequency response for sounds arriving
`from a mirror image sound location behind the listener. For
`example, referring to FIG. 3, subtracting the curve for
`sounds arriving at
`the listeners left ear at an angle of 45
`degrees, in front of the listener, from the curve for sounds
`arrivingattheleft ear at an angle of 135 degrees, behind the
`listener, produces the curve labeled “45-135 deg.” in FIG, 4.
`Thus, with the front-to-back filters 1 and 2 of FIG. 2 having
`the approximate characteristics of, for example, the front-
`to-back frequency response curve from FIG. 4 labeled
`“45-135 deg.” and left and right main speakers LMS and
`RMSlocated approximately 45 degrees to either side of
`central listening axis CLA,a listener located at the principle
`listening location LL will perceive approximately the same
`frequency response for surround signals LS and RSat the ear
`drum ofthe respective nearest ear as if these sounds were
`originating at sound locations behind the listener mirror
`imaged to the actual sound locations of LMS and RMS in
`front ofthe listener from which the surroundsignals LS and
`RS are actually emanating.
`
`[0049] FIG. 5 shows a similar family of front-to-back
`frequency response curves calculated by subtracting the
`frequency response shown in FIG.3 for sounds arriving at
`the listener's farthest ear of sound locations in front ofthe
`listener, from the frequency response for sounds arriving
`from mirror image sound locations behind the listener.
`Application of front-to-back filters with these characteristics
`to sounds arriving at the listener’s farthest ear from actual
`sound locations in front of the listener will duplicate the
`frequency response at the listener’s farthest ear drum of a
`soundarriving from a mirror image sound location behind
`the listener. For example, referring again to FIG. 3, sub-
`tracting the curve for soundsarriving at the listeners left ear
`at an angle of minus 45 degrees,in front ofthe listener, from
`the curve for sounds arriving at the left ear at an angle of
`minus 135 degrees, behind the listener, produces the curve
`labeled “45-135
`
`deg.” in FIG. 5. Thus, with front-to-back filters 1
`[0050]
`and 2 of FIG, 2 having the approximate characteristics of,
`for example the front-to-back frequency response curve
`from FIG. 5 labeled “45-135 deg.” andleft and right main
`speakers LMS and RMS,located approximately 45 degrees
`to either side of central listening axis CLA,a listener located
`at principle listening location LL will perceiv