JP2012019506

Patent Translate
Powered by EPO and Google
Notice
This translation is machine-generated. It cannot be guaranteed that it is intelligible, accurate,
complete, reliable or fit for specific purposes. Critical decisions, such as commercially relevant or
financial decisions, should not be based on machine-translation output.
DESCRIPTION JP2012019506
A vehicle audio system having a speaker incorporating a headrest. An audio signal processor
comprising two speakers (1R, 1L) built into a headrest of a vehicle, a protective cap for each of
the two speakers, and an audio signal processor for receiving an audio input signal Is such that
when an audio output signal is output by the two speakers, the audio output signal is perceived
by the user sitting on a seat where the headrest with the speaker is mounted as a virtual sound
field. An audio signal processor, configured to generate an audio output signal, and a database
containing cap compensation information, the audio signal processor generating a virtual sound
field for the user taking into account the cap compensation information Vehicle audio system
Temu. [Selected figure] Figure 2
Vehicle audio system with headrest built-in speaker
[0001]
The present invention relates to a vehicle audio system in which speakers are incorporated in the
head rest of a vehicle seat and to a method of generating a virtual sound field for a user sitting in
the vehicle seat.
[0002]
2. Related Art Speakers that generate acoustic signals in convertible cars are exposed to various
weather conditions.
10-05-2019
1
The speakers must be protected against rain and direct sunlight. Furthermore, the noise during
driving can be very strong in convertible vehicles. As a result, the driver uses much higher audio
signal levels as compared to non-convertible vehicles. The strong audio signals emitted by the
speakers in a convertible car can be annoying to other people in the surrounding area.
Furthermore, strong current is required to output advanced audio signals, resulting in high
demands on the vehicle battery.
[0003]
SUMMARY There is a need, therefore, to provide a vehicle audio signal in which the audio output
by the vehicle audio system is not or hardly perceived by people outside the vehicle. Further,
there is a need to provide a vehicle audio system that requires less battery power. Furthermore,
the speakers should be protected against the weather, which is particularly important for
convertible vehicles.
[0004]
This need is met by the features of the independent claims. In the dependent claims, preferred
embodiments of the invention are described.
[0005]
According to a first aspect of the present invention, there is provided a vehicle audio system
comprising two speakers incorporated in a headrest of a vehicle. Furthermore, a protective cap is
provided on each of the two speakers in the headrest above each speaker, the protective cap
extending in the direction in which the sound of each speaker is emitted. The audio signal
processor of this audio system receives an audio input signal, and when the audio output signal
is output by two speakers, the audio by the user sitting on a seat provided with a headrest with
two speakers as a virtual sound field An audio output signal is generated for the two speakers
such that the output signal is perceived. In the virtual sound field, the spatial perception of music
is obtained, where the user has the impression of listening to the audio signal from any point in
space, not from the direction of the position of the loudspeakers. The audio system further
comprises a database containing cap compensation information, which is emitted by the two
loudspeakers, making it possible to compensate for the effect of the protective cap on the audio
output signal perceived by the user as a virtual sound field. According to the invention, the audio
10-05-2019
2
signal processor is configured to generate a virtual sound field for the user taking into account
cap compensation information. The cap mounted on the speaker has a dual function. First of all,
the cap protects the loudspeakers integrated in the headrest against the environment, such as
rain or any other environment-related influences. As an example, the cap avoids direct exposure
of the speaker to the sun and further helps to avoid rapid aging of the components of the
speaker. In addition, the protective cap helps protect the speaker against particles (eg, spiders,
mosquitos or any other debris) that fly with the air flow in the vehicle while driving. Because the
speaker is located near the ear, the amount of signal can be lowered compared to if the speaker
is mounted somewhere further from the head. Furthermore, by providing a protective cap on
each speaker, the emitted sound is directed downward. This helps to keep the sound in the
vehicle so that people outside the vehicle are less disturbed by the sound emitted from the audio
system. Although protective grids mounted parallel to the top surface of the loudspeaker are
known in the art, these grids do not extend mainly parallel to the sound emitted from the
loudspeaker but extend vertically.
The main components of the protective cap extend in the direction in which the sound field from
the speaker is emitted and act as a reflector or guide for the emitted sound field, so that the
protective cap of the invention emits the sound emitted by the speaker Has a strong influence on
the field. As an example, the virtual sound field improves the listening impression to the user,
since the virtual surround sound can be simulated to the user / driver. Perception is induced in
the virtual surround sound so that the user has the impression that there are more sound sources
than the ones actually used to generate the sound field. In a virtual surround field, the user
perceives the sound as having the impression that the sound is coming from where it does not
actually come from. The protective cap helps to guide or focus the sound field to the user's ear
and helps minimize the emission of the sound field to the surroundings. Thus, much lower signal
levels can be used for the system of the present invention as compared to prior art audio
systems.
[0006]
The invention may be used in connection with convertible vehicles. However, the invention is not
limited to this application. By way of example, the invention may also be used in closed vehicles
such as trucks, or in motor boats or planes. The protective cap is mounted on the speaker and
mainly extends in the direction in which the sound is emitted at least partially, thus affecting the
emitted sound field and thus the virtual sound field perceived by the user. In order to provide a
virtual sound field, the influence of the protective cap on the sound field generated by the two
speakers in the headrest has to be determined and compensated. Cap compensation information
is used for this purpose.
10-05-2019
3
[0007]
The virtual sound field as perceived by the user depends on the position of the user's head, ie the
position of the ear. The human auditory system localizes the sound source and the localization
depends on the way in which the acoustic signal travels from the loudspeaker to the human ear.
In one embodiment of the invention, a predetermined common position of the head is used. The
position of the head relative to the headrest is relatively fixed for the driver of the vehicle and
has only slight variations. When the correct setting of the headrest is considered, the relative
position of the headrest and the head is the same for people of different heights, so usually the
height of the user sitting on the seat with the headrest and the two speakers is , I do not play a
part. In another embodiment, an image sensor for tracking the user's head is provided, and the
database further includes head positional relationship data. The audio signal processor then
generates a virtual sound field for the user taking into account the cap compensation information
and the tracked head position.
[0008]
Head related data may include a set of binaural room impulse responses (BRIR) determined for
the user with respect to various possible head positions. Acoustic signals for both ears are
usually intended to be reproduced using headphones and simulate the actual position of the
sound at which the sound is produced if the recorded audio signals for both ears are reproduced
on the headphones Listening experience is obtained. When a normal stereo signal is reproduced
using headphones, the listener perceives the signal in the center of the head. If an acoustic signal
recorded for both ears is reproduced by headphones, the position from where the signal was
originally recorded is simulated. In this case, the acoustic signal is not output by the headphones
but by a pair of speakers mounted on the headrest. Since the perceived acoustic signal is
dependent on the position of the listening user's head, the position of the user's head is tracked
using an image sensor.
[0009]
Further crosstalk cancellation may be performed for virtual sound field generation for the user,
in which the acoustic signal emitted by one speaker reaches the intended ear while the acoustic
signal of this speaker is , Against the other ear, and vice versa. However, in the present invention,
10-05-2019
4
one speaker may be positioned next to one ear on the side of the headrest, and the other speaker
may be mounted on the other side positioned next to the other ear So, for example, the
component of the acoustic signal that is suppressed in the crosstalk cancellation of the right
speaker relative to the left ear is very small. The binaural room impulse response (BRIR)
determined for various head positions may be predetermined in the vehicle using a dummy head.
Different BRIRs can be obtained by mounting a microphone on each ear of the dummy head at
different head positions and placing the dummy head on the vehicle. According to this
embodiment, the transfer function associated with the head and the in-vehicle influence on the
signal path from the loudspeaker to the ear can be determined. If reflections are neglected, it is
also possible to use a head related transfer function instead of BRIR. As an example, the position
of the head may be tracked by the image sensor by determining translation in three different
directions, and further, possible rotations of the head may be tracked. The set of predetermined
binaural impulse responses may then include BRIR for various possible translations and
rotations. In a vehicle environment, it may be sufficient to consider fewer degrees of freedom
only for translation (left or right and back, front) and one rotation, ie head rotation to the left and
right. The user-specific binaural sound signal at a given common head position without head
tracking is then determined by determining the convolution of the source signal with binaural
room impulse responses to the head position. It can be done. The acoustic signal from a sound
source such as a CD, DVD or radio may be a 1.0, 2.0, 5.1 or 7.1 signal, and the acoustic signal for
both ears from the user may be a 2-channel signal (each 1 channel signal) for the speaker.
[0010]
Preferably, the protective cap mounted on each speaker is designed such that the audio output
signal emitted by the two speakers is directed to the area of the cabin provided below the
headrest. This helps to keep the sound inside the vehicle and to minimize the sound emitted
outside the vehicle. The exact form of the protective cap may also depend on other parameters,
such as the movement of air in the driving vehicle. Furthermore, the design of the headrest and
the cabin can influence the exact form of the protective cap. By mounting the speakers on the
side of the headrest, the safety function of the headrest is not degraded. By mounting the
speakers on the side, it can be avoided that the head hits the speakers in case of an accident.
[0011]
Preferably, the vehicle audio system comprises a pair of speakers (i.e., one for the driver, one for
the person sitting next to the driver) at each headrest of the front seat. For each of the two, a
virtual sound field may be generated by a corresponding pair of speakers. Because the speakers
10-05-2019
5
are mounted very close to the ear, high sound levels of emitted sound may not be necessary. As a
result, the sound field emitted for the first user (e.g. the driver) does not disturb the other user
seated in the other front seat. However, in another embodiment, the audio system is such that the
audio signal processor performs cross sound field suppression and the sound field emitted by the
other pair of speakers for the other user is for the first user Cross sound field suppression can be
designed to be performed which is suppressed and vice versa.
[0012]
The two speakers are preferably satellite speakers emitting audio signals in a frequency range of
more than 100 MHz to 15,000 Hz or 18,000 Hz, as the space for housing the speakers in the
headrest is limited. Woofers, which are additional speakers for lower frequencies, may be
mounted elsewhere in the vehicle cabin.
[0013]
The invention relates to a method of generating a virtual sound field for a user sitting in a vehicle
seat, which method comprises the step of providing cap compensation information,
compensating for the effect of a protective cap on an audio output signal. , And processing an
audio input signal input to the audio signal processor, such that an audio output signal perceived
as a virtual sound field by a user is generated taking into account cap compensation information.
And processing. As discussed above, it is possible to generate a user-specific sound field at low
signal levels.
[0014]
Preferably, processing the audio output signal comprises the steps of: generating a user-specific
binaural sound signal for the user; and generating the crosstalk-canceled user-specific sound
signal. Performing a crosstalk cancellation on the user, wherein the user-specific audio signal for
both ears is a crosstalk-canceled user-specific audio signal of the first user of the user; The
crosstalk-canceled user-specific acoustic signal is processed so as to be suppressed to the user's
second ear, and vice versa, when output by one of the two loudspeakers to the user's ear Also
includes steps, which are also the same. In addition, the crosstalk canceled user-specific acoustic
signal is processed with cap compensation information to generate an audio output signal. For
generation of the virtual sound field, either a fixed head position can be used or the position of
10-05-2019
6
the head can be tracked using an image sensor. If loudspeakers are used in the two headrests of
the vehicle, it is possible to further perform cross sound field suppression; in this cross sound
field suppression, the audio output signal output by a pair of speakers of one headrest is another
It is restrained for the user sitting in the seat of the vehicle with which one headrest is equipped.
[0015]
For example, the present invention provides the following items. (Item 1) Two speakers (1R, 1L)
incorporated in a headrest (20) of a vehicle, and a protective cap (40) for each of the two
speakers, the protective cap being the headrest (20) is provided on each speaker (1R, 1L) so as to
extend in the direction in which the sound of each speaker is emitted, a protective cap (40), and
an audio signal processor (60) for receiving an audio input signal. And the audio signal processor
(60) controls the audio by the user sitting on a seat where the headrest with the speakers is
mounted as a virtual sound field when an audio output signal is output by the two speakers The
audio output signals for the two speakers (1R, 1L) so that the output signal is perceived An audio
signal processor (60) configured to generate a database (80) including cap compensation
information, the cap compensation information being perceived by the user as the virtual sound
field A database (80) capable of compensating the effect of the protective cap (40) on the audio
output signal emitted by the two speakers (1R, 1L), the audio signal processor (60) comprising A
vehicle audio system configured to generate the virtual sound field for the user taking into
account the cap compensation information. (Item 2) The image sensor (70a, 70b) tracking the
head of the user, the database 80 further includes data related to the position of the head, and
the audio signal processor The vehicle audio system of any of the above items, configured to
generate a virtual sound field for the user to further take into account the tracked head position.
3. The vehicle audio of any of the above items, wherein the head related data comprises a set of
binaural room impulse responses determined for the user with respect to various possible head
positions. system. (Item 4) The protective cap (40) mounted on each speaker is designed such
that the audio output signal emitted by the two speakers is directed to the area of the cabin
provided under the headrest Vehicle audio system according to any of the above items. (Item 5)
The vehicle audio system according to any one of the above items, wherein the speaker (1R, 1L)
is provided on each side (21, 22) of the headrest.
6. The audio signal processor (60) processes the audio input signal such that the audio signal
processor generates a user-specific binaural sound signal for the user from the audio input
signal. Configured to generate the virtual sound field for the user, the audio signal processor
performing crosstalk cancellation on acoustic signals emitted by the two speakers of the headrest
Vehicle audio system as described. The audio signal processor (60) is configured to provide cross
sound field suppression in which sound fields emitted by other speakers to another user are
10-05-2019
7
suppressed for each ear of the user The vehicle audio system according to any of the above
items. (Item 8) The vehicle audio system according to any of the above items, wherein the two
speakers (1R, 1L) are designed to be optimized to output an audio signal in a frequency range
higher than 100 Hz. (Item 9) A method of generating a virtual sound field for a user sitting in a
vehicle seat, the virtual sound field comprising two speakers (1R, 1L) incorporated in a headrest
(20) of the vehicle seat And each speaker (1R, 1L) is mounted on a headrest above each speaker
and protected by a protective cap (40) extending in the direction in which the sound of each
speaker is emitted, the method comprising Providing information, said cap compensation
information being able to compensate for the effect of said protective cap on the audio output
signal output by said two speakers perceived by said user as said virtual sound field The audio
output signal perceived by the user as the virtual sound field takes into account the cap
compensation information Comprising a step of processing the audio input signal to be
generated by putting the method. 10. The method of claim 1, wherein processing the audio input
signal comprises: generating a user-specific binaural sound signal for the user; and generating
the crosstalk-canceled user-specific sound signal. Performing crosstalk cancellation on the user,
wherein the user-specific audio signal for both ears is the crosstalk-canceled user-specific audio
signal for the first ear of the user. When output by one of the two speakers, the crosstalk
canceled user-specific acoustic signal is suppressed to the user's second ear and the crosstalkcanceled user-specific An acoustic signal is output by the other of the two speakers to the second
ear of the user When the crosstalk canceled user-specific acoustic signal is processed to be
suppressed to the first ear of the user, and cap compensation to generate the audio output signal.
Processing the crosstalk-cancelled user-specific acoustic signal using information. A method
according to any of the above items.
11. The method of claim 1, wherein the audio output signal is generated using a fixed position of
the head of the user in the headrest. 12. The method of claim 12, further comprising the step of
tracking the position of the user's head, wherein the audio output signal is generated taking into
account the cap compensation information and the tracked head position. The method described
in any of the above. (Item 13) Two vehicle seats are provided, and the headrest of each vehicle
seat incorporates two speakers, and the processing of the audio input signal is an audio output
output by the speakers in the headrest of one vehicle seat The method according to any of the
above items, further comprising the step of performing cross sound field suppression in which
the signal is suppressed for each ear of the user sitting in the other vehicle seat.
[0016]
The invention consists of:-two loudspeakers integrated in the headrest of the vehicle;-a protective
cap for each of the two loudspeakers mentioned above, in the headrest, on each loudspeaker and
the sound of each loudspeaker A protective cap, provided so as to extend in the direction in
which the audio is emitted-an audio signal processor for receiving an audio input signal, the
10-05-2019
8
audio signal processor being a speaker when the audio output signal is output by two speakers
An audio signal processor, configured to generate audio output signals for the two speakers such
that an audio output signal is perceived by a user sitting on a seat provided with a headrest
having a virtual sound field. , A database containing cap compensation information The database
comprises: a database capable of compensating for the effect of the protective cap on the audio
output signal emitted by the two speakers, perceived by the user as a virtual sound field; A
vehicle audio system configured to generate a virtual sound field for the user taking into account
cap compensation information.
[0017]
The invention will be discussed in more detail with reference to the accompanying drawings.
[0018]
FIG. 1 is a drawing illustrating installation of a speaker in a headrest.
FIG. 2 is a block diagram of the audio system of the present invention.
FIG. 3 is a more detailed diagram of an audio signal processor for processing audio signals. FIG. 4
is a diagram showing generation of virtual headphones. FIG. 5 is a diagram illustrating the
generation of a user-specific virtual sound field using cross sound field cancellation. FIG. 6 is a
flow chart showing the steps of generating a virtual sound field for one user.
[0019]
DETAILED DESCRIPTION FIG. 1 is a schematic view of a user 10 sitting in a vehicle seat (not
shown) having a headrest 20. The headrest comprises two sides 21 and 22 in which the speakers
1 R, 1 L are integrated, the two speakers emitting acoustic signals to the user 10. The vehicle
seat and headrest shown in FIG. 1 are preferably located on a convertible car. A protective cap 40
is attached over each speaker to protect the speakers against the environment. The inner surface
of the protective cap has a concave shape and directs the sound emitted from the loudspeaker
towards the lower part of the car and towards the user's ear. The protective cap is designed in
such a way that in case of an accident the safety requirements for the other occupants are
fulfilled. Furthermore, the air flow in the case of a driving vehicle influences the shape of the
10-05-2019
9
protective cap. The outer surface of the protective cap can be covered by the same material as
the headrest. As can be seen from FIG. 1, the two speakers are positioned very close to the user's
two ears. The two speakers 1R, 1L shown in FIG. 1 are not necessarily the only speakers provided
in the vehicle. Every headrest in a vehicle can have a pair of speakers incorporated as shown in
FIG. Because the space in the headrest is limited, the speaker is a small speaker, such as a
satellite component of a speaker system that further includes a woofer, for example, somewhere
in the vehicle. Taking into account the size of the loudspeaker, the loudspeaker is preferably
designed for the higher frequency, not for the lower frequency. By way of example, the speakers
1R, 1L can be adapted in particular to a frequency range above approximately 100 Hz. Because
of the short distance between the loudspeakers and the user's ear, the vehicle in which the
acoustic system shown in FIG. 1 is integrated is convertible and a low signal level output audio
signal is sufficient even while driving. is there.
[0020]
The virtual sound field generated by the audio output signals of the two speakers will be
described in more detail in connection with FIGS. In FIG. 2 an audio system is schematically
shown which is incorporated in the two headrests of the front seat of the vehicle. The user A
shown in the left part of the figure sits on a seat with a headrest 20 in which two speakers 1L
and 1R are integrated, the protective cap is only shown in phantom for the sake of clarity. The
user B sits on the adjacent seat, and the two speakers 2L and 2R are incorporated in the headrest
20. The speakers shown in FIG. 2, designated 1L, 1R and 2L and 2R, may be the same speakers
as shown in FIG. The audio system comprises an audio signal source 50 which provides the audio
signal output by the speaker. The audio signal may be from a CD, DVD or hard disk where the
audio signal may be stored in digital form. The audio system further comprises a signal processor
50 for processing the audio signal received from the audio signal source 50 before the audio
signal is output to the left and right pair of speakers. The signal processor processes the received
audio input signal such that the audio output signals provided to the speakers 1L, 1R, 2L and 2R
are perceived by the users A and B sitting on a seat as a virtual sound field . In a virtual sound
field, the user is provided a spatial auditory representation of the audio signal. Preferably, the
virtual sound field is a virtual ambient sound, the virtual ambient sound being distributed from
the various speakers in an impression that some speakers are equipped and in fact there is no
speaker at all Give the user the impression of Before the audio signal is output, the audio signal
emitted by the speaker 1L is transmitted to the left ear, while the audio signal component
(indicated by the broken line) output to the right ear by the speaker 1L is suppressed This virtual
ambient sound is possible if it is possible to process the audio signal as it is. In the same way, the
audio output signal of the speaker 1R is specified for the right ear and the audio output signal
from the right speaker 1R for the user A's left ear is suppressed. In the illustrated embodiment, it
is conceivable that both users A and B listen to the audio signal from the same source.
10-05-2019
10
The same is done for user B and a virtual sound field is provided for user B using speakers 2L
and 2R. The system of virtual sound fields for each user is also called virtual headphones and this
concept is disclosed in more detail in FIG.
[0021]
The system for user A is shown in more detail, with two speakers 1 L and 1 R and an audio signal
being transmitted to the user's two ears. As can be seen from FIG. 2, for the sake of clarity, the
two speakers are not shown in the actual position of the speakers, but more clearly show the
transmission of the acoustic signal from the two speakers to the left ear and the right ear Is
represented by a larger distance to allow for A spatial auditory representation of the audio signal
is obtained by emitting signals for both ears, and the signals for both ears emitted by the speaker
1L are carried to the left ear, while those for the two ears emitted by the speaker 1R. The signal
is carried to the right ear. For this purpose, crosstalk cancellation is necessary, and the audio
signal emitted from the speaker 1L is suppressed to the right ear (1L-R in FIG. 4), and the audio
output signal of the speaker 1R is on the left ear It is suppressed (1R-L). As can be seen from FIG.
4, the transmission path from the speaker to the ear depends on the position of the head. In one
embodiment of the invention, a fixed head position is taken as the basis of the calculation. In
another embodiment, a camera is provided for each user, such as camera 70A for user A and
camera B for user B as shown in FIG. The camera 70A and the camera 70B can determine the
position of the user's head. As an example, the camera may determine the position of the head by
using pattern recognition techniques, in which any other predetermined part of the face or head
is recognized in the image. The motion of the head can be inferred from the motion of the
recognized part of the image. The camera may determine the three dimensional translation of the
head in addition to the three different possible rotations. The signal processor 60 is connected to
the database 80 and the binaural room impulse response (BRIR) for different head translations
and rotational positions is database as will be described in more detail later in connection with
FIGS. 3 and 5 It is stored in 80. If the position of the head is not tracked, a fixed general head
position is used and the binaural room impulse response to this head position is used in the
database. The BRIR takes into account the transition path from the loudspeaker to the tympanic
membrane and the possible reflection of the audio signal in the car. The user-specific binaural
sound for the user A from the audio source may be generated by first generating a user-specific
binaural sound signal and then performing crosstalk cancellation, the crosstalk In cancellation,
the signal path from the left speaker to the right ear and from the right speaker to the left ear is
suppressed.
10-05-2019
11
User-specific binaural sound signals are obtained by determining the convolution of the audio
input signal with binaural room impulse responses. The crosstalk cancellation is then obtained by
calculating a new filter for crosstalk cancellation, which crosstalk cancellation again depends on
the position of the tracked head using the crosstalk cancellation filter obtain. A more detailed
analysis of cross-talk cancellation that relies on head rotation can be found in the Audio
Engineering Society Paper 6541 presented at 119 <th> Convention, October 2005, 7-10. Lentz, I
.; Assenmacher and J.A. Sokoll, "Perfomance of Spatial Audio Using Dynamic Cross-Talk
Cancellation" and J.A. Audio Eng. Soc. , Vol. 54, no. 4, "Dynamic Crosstalk Cancellation for
Binaural Synthesis" in April 2006, pages 283-294. Explained by Lentz. Crosstalk cancellation is
obtained by determining the convolution of the user-specific binaural sound signal with the
newly determined crosstalk cancellation filter. After processing with this new filter, crosstalkcancelled user-specific audio signals are obtained for each of the speakers and, when output to
the user, an audio signal from a direction determined by the position from the speakers As well
as providing spatial recognition of the audio signal, the user has the impression of listening to
the audio signal from any other point in space.
[0022]
In FIG. 3, a more detailed view of the signal processing performed in signal processor 60 is
shown. The audio input signal from the audio signal source 50 is a stereo signal or a 5.1
surround signal. The audio signal may be a multi-channel audio signal of any format. In FIG. 3,
various computational steps are symbolized by various modules to facilitate understanding of the
present invention. However, it should be understood that the processing is performed by a single
processing unit, which preferably performs the various computational steps symbolized by the
modules in FIG. In the following, signal processing is considered by means of head tracking in
which the movement of the user's head is taken into account. However, the processing steps
shown in FIG. 3 can also be performed using a fixed position of the head. In a first module which
receives the position of the head as symbolized by the arrow coming from the image sensor 70,
the binaural room impulse response to that position is extracted from the database 80. In the
first processing unit 61, the multi-channel audio signal is converted into an audio signal for both
ears, and the audio signal for both ears gives a listener a 3D impression when output by
headphones. This user-specific binaural acoustic signal is obtained by determining the
convolution of the multi-channel audio input signal with the corresponding BRIR of the tracked
head position. The user-specific binaural sound signal is then further processed in module 62
where a crosstalk cancellation filter is calculated. The crosstalk cancellation filter is used to
determine crosstalk cancellation by determining the convolution of the user-specific audio signal
for both ears by the crosstalk cancellation filter. The output of the module 62 is a crosstalkcancelled user-specific audio signal, and the crosstalk-canceled user-specific audio signal is
output by the speaker using the headphones. Give the listener the same impression as the
10-05-2019
12
listener listening to the Since crosstalk cancellation also depends on the position of the head,
corresponding head position information is also output to module 62. At module 63, the effect of
the protective cap is determined.
For this purpose, the database may comprise predetermined filters for various head positions
which may be predetermined using dummy heads sitting on corresponding vehicles. The various
filters may be determined using measurements of the signal emitted by the speaker, with the
protective cap attached. These measurements make it possible to determine the effect of the
protective cap on the emitted sound. With respect to the various filters used in modules 61 and
62, database 80 comprises filters for various head positions, and the signals emitted from
module 62 are from module 62 with cap compensation filters for the corresponding head
positions. It is used to determine the convolution of the emitted crosstalk-canceled user-specific
acoustic signal. The cap-compensated audio output signal is thus determined for each addressed
loudspeaker and output to the corresponding loudspeaker. The sound emitted by the two
speakers creates a virtual sound field for the user.
[0023]
If the position of the head is not tracked, then only filters for the middle head position have to be
provided for the generation of binaural sound, crosstalk cancellation and cap compensation
respectively. Signal processing as shown in FIG. 3 may be performed for each of users A and B
shown in FIG. Since the signal level of the sound field emitted by the two pairs of loudspeakers is
very low, the interference of the sound field created for user A is low for user B and vice versa.
However, in another embodiment, it is possible to perform cross sound field cancellation where
the sound field generated for user A is suppressed for user B. Such an embodiment is disclosed in
more detail in connection with FIG.
[0024]
Besides the signal processing steps shown in FIG. 3, cross sound field cancellation can be
performed, the basic principle is shown in FIG. The two speakers 1L, 1R for the first user A
generate user-specific acoustic signals for the first user A, and the two speakers 2L, 2R generate
user-specific acoustic signals for the second user B. Generate Two cameras 70a and 70b are
provided to determine the position of the heads of two users A and B respectively. The first
speaker 1L outputs an audio signal which is heard by the left ear and the right ear of the listener
A indicated by AL and AR under normal conditions. Signals 1L, AL corresponding to the signal
10-05-2019
13
emitted from the speaker 1L to the left ear of the listener A are shown by bold lines and are not
suppressed. The other acoustic signal 1L, AR, shown in dashed lines with respect to the right ear
of the user A, is suppressed. In the same way, the signals 1R, AR reach the right ear, while the
signals 1R, AL for the left ear are suppressed. However, since the signals from the speakers 1L
and 1R are perceived by the listener B, cross sound field cancellation in which the sound field is
suppressed can be performed. The signals to be suppressed from the loudspeaker 1L are shown
as 1L, BR and 1L, BL. In the same way, the signal emitted by the speaker 1 R is suppressed for
the listener B's ears. In the same way, the acoustic signals emitted from the speakers 2L and 2R
are suppressed for the user A. Thus, in principle cross sound field cancellation works in the same
way as crosstalk cancellation.
[0025]
In the embodiment shown in FIG. 3, three convolutions are performed in the signal path. Filtering
for auralisation crosstalk cancellation and cap compensation may be performed one after
another. In another embodiment, three different filtering operations may be combined into one
convolution using a pre-determined filter. Cross talk cancellation using head tracking is described
in "Dynamic Crosstalk Cancellation for Binaural Synthesis in Virtual Reality Environments," J.
Audio Eng. Soc. , Vol. 54, no. 4, April 2006, pages 283-294. It is explained in more detail by
Lentz. The same principle is used in cross sound field cancellation, but the signals emitted from
the other two speakers are suppressed.
[0026]
In FIG. 6, the various steps of determining a user specific virtual sound field without cross sound
field cancellation are summarized. The method starts at step 100. At step 110, the user's head,
for example User A or User B, is tracked. Depending on the position of the tracked head, it is
possible to determine the binaural sound signal in step 120 by calculating the convolution of the
audio input signal by BRIR with the determined head position. At step 130, crosstalk cancellation
is performed at module 62 as described in connection with FIG. In step 140, the influence of the
protective cap on the emitted acoustic signal is taken into account. After the compensation step
140, the signal is output after module 63 in response to the signal output. If cross sound field
cancellation is performed, this cross sound field cancellation may be performed after step 140.
The signal is then output at step 150 to obtain a user-specific virtual sound field.
[0027]
10-05-2019
14
The speaker mounted on the headrest does not have to be positioned with the outer side of the
speaker parallel to the side of the headrest. The speakers can also be placed at a slight angle to
the outer surface of the headrest. Furthermore, the shape of the protective cap is governed by
the need to protect the speaker, the need to avoid noise generated by the air flow traveling
around the cap, and the need to obtain a design that is acceptable to the user.
[0028]
10 User 20 Headrest 21 Side 22 Side 40 Protective Cap 50 Audio Signal Source 60 Signal
Processor 61 Module 62 Module 70 Image Sensor 70A Camera 70B Camera 80 Database
10-05-2019
15