close

Вход

Забыли?

вход по аккаунту

?

DESCRIPTION JP2015524224

код для вставкиСкачать
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 JP2015524224
Abstract: The present invention discloses a headset communication method and headset in a
strong noise environment. The method reduces mid-high frequency noise entering the ear canal
with an earplug, diverts middle-low frequency noise with an external connection cavity in parallel
with the ear canal, and acoustics in the ear canal with an internal microphone. The ambient noise
signal that has leaked is acquired, the ambient noise signal is picked up by the external
microphone, the noise component in the internal microphone signal is removed using the
external microphone signal as a reference signal, and the voice component is retained. After the
sound pressure level is reduced and compensated for the signal acquired by the external
microphone by using the acoustic signal compression technique, the sound pressure range is
received by the human ear. Compressing into a suitable range and then reproducing the signal,
along with the receiver signal received by the headset, through the headset handset. The
technical solution of the present invention can fully realize the hearing protection, speech
enhancement and three-dimensional environment monitoring functions in a high noise
environment. [Selected figure] Figure 1
Communication method and headset of headset in strong noise environment
[0001]
The present invention relates to the field of acoustics, and more particularly to a method and
headset for headset communication in a high noise environment.
[0002]
With the progress of society and economic development, the situation where human beings are
forced to face a strong noise environment is, for example, operation next to a large machine (eg
11-04-2019
1
loom lathe, air compressor, blower, etc.) in industrial production. It has come to be found
everywhere in the roaring of modern transportation vehicles (eg motorcycles, trains, planes etc),
putting on a construction site or a military battlefield.
[0003]
Strong noise causes a series of serious problems.
First, high-intensity noise makes people feel tired and produces negative emotions, causing
serious damage to the nervous system, blood circulation system, endocrine system, digestive
system, vision, hearing, intelligence, etc. Therefore, hearing protection in a strong noise
environment is an essential measure.
Next, when performing voice communication using a headset in a strong noise environment, the
voice signal of the speaker may be completely wiped out by environmental noise, and a voice call
may not be realized normally. This can have a major impact on production activities, daily life,
military operations, etc., and can cause huge losses to individuals, groups, and even the nation. As
can be seen, maintaining a clear and stable voice communication function in a strong noise
environment is of great significance, which is also a high interest issue that people are studying.
Also, in some high noise environments, it is required to maintain sufficient sensitivity to the
sound of the surrounding environment in order to monitor and accurately respond to real-time
changes in the surrounding environment while receiving hearing protection. If it does not, it will
not be able to sense a possible danger signal. For example, in a battlefield environment, soldiers
can obviously fall into situations where they can not make their own decisions or dangerous
situations if they can not hear the surrounding sound.
[0004]
In view of this, the present invention provides a headset communication method in a strong noise
environment that can guarantee the headset and the speech quality of the headset, in order to
overcome or at least partially solve the above problems.
[0005]
In order to achieve the above object, the technical solution of the present invention is realized as
11-04-2019
2
follows.
The present invention discloses a method of communicating a headset in a strong noise
environment, which reduces medium-frequency noise entering the ear canal by means of an
earplug tightly coupled to the wearer's ear canal, as well as the ear Noise reduction in the entire
band with respect to an acoustic signal entering the ear canal by diverting medium and low
frequency noise entering the ear canal by the external connection cavity extending out from the
plug and forming a parallel branch with the ear canal The internal microphone of the headset
acquires the audio signal in the ear canal and the environmental noise signal leaked into the ear
canal, and the external microphone of the headset acquires the environmental noise signal and
the audio signal propagated through the air Using the signal acquired by the external
microphone as a reference signal and removing the noise component in the signal acquired by
the internal microphone to retain the voice component. After the sound pressure level is reduced
and compensated for the signals acquired by the external microphones on both sides of the
headset, the sound pressure range of the processed signal is And C. recomposing the processed
signal, along with the receiver signal received by the headset, through the handset of the headset,
after compression to a range suitable for receipt by the human ear.
[0006]
The acoustic impedance of the external connection cavity may be significantly smaller than the
acoustic impedance of the ear canal, and a sound absorbing material may be provided on the
inner wall of the external connection cavity.
[0007]
Taking the signal acquired by the external microphone as a reference signal and removing the
noise component in the signal acquired by the internal microphone to retain the audio
component, it is possible to obtain the audio signal of the headset within the low frequency
range. The control parameter α is specified by the statistical energy ratio between the signal
acquired by the external microphone and the signal acquired by the internal microphone, the
weight of the adaptive filter is updated by the feedback output signal, and the control parameter
α The weight update rate of the adaptive filter is controlled, and adaptive filtering processing is
performed on the signal acquired by the external microphone to obtain an adaptive filter output
signal, and the adaptive filter output signal is obtained from the signal acquired by the internal
microphone. Obtaining an output signal by subtracting the output signal from the transmitter
signal of the headset May be included.
[0008]
11-04-2019
3
The method includes performing single-channel audio processing and frequency spread
spectrum processing on the output signal, and using a signal subjected to single-channel audio
processing and frequency spread spectrum processing as a transmitter signal of a headset. And
may be further included.
[0009]
The invention further discloses a headset, which is closely coupled to the wearer's ear canal and
extends from the ear plug capable of reducing medium frequency noise entering the ear canal.
An external connection cavity that composes the ear canal and a parallel branch, and shunts lowfrequency noise entering the ear canal; an internal microphone for acquiring audio signals in the
ear canal and environmental noise signals leaked into the ear canal; An external microphone for
acquiring a noise signal and an audio signal propagated through the air, a signal acquired by the
internal microphone and a signal acquired by the external microphone, and a signal acquired by
the external microphone as a reference signal Audio signal processing means for obtaining a
transmitting side signal of the headset after removing the noise component in the signal acquired
by the internal microphone and retaining the audio component; For reducing and compensating
the sound pressure level of the signal acquired by the external microphone, and then
compressing the sound pressure range of the processed signal into a range suitable for reception
by the human ear The acoustic dynamic compression means includes: a receiver for reproducing
the signal processed by the acoustic dynamic compression means processing together with the
receiver signal received by the headset.
[0010]
The acoustic impedance of the external connection cavity may be significantly smaller than the
acoustic impedance of the ear canal, and a sound absorbing material may be provided on the
inner wall of the external connection cavity.
[0011]
The audio signal processing means receives a signal acquired by the internal microphone and a
signal acquired by the external microphone, and a statistic of a signal acquired by the external
microphone within a low frequency range and a signal acquired by the internal microphone A
voice detection module for specifying the control parameter α and outputting the control
parameter α by the dynamic energy ratio, the output signal fed back as a reference signal for
weight update, and the control parameter α as a control parameter for weight update rate An
adaptive filter for performing an adaptive filtering process on the received external microphone
signal to output an adaptive filter output signal; and an adaptive filter output signal received
from a signal acquired by the received internal microphone. And a noise reduction module for
obtaining an output signal by subtraction. .
11-04-2019
4
[0012]
The audio signal processing means may further include a post-processing module for performing
single-channel audio processing and frequency spread spectrum processing on the output signal.
[0013]
The audio signal processing means and the acoustic dynamic compression means may be
integrated in one DSP chip.
[0014]
The number of the internal microphones is one, and is located on the left ear side or the right ear
side of the headset. The number of the external microphones is two, and the left ear side and the
right ear of the headset are respectively. Located on the side, the audio signal processing means
receives signals acquired by the internal microphone and the external microphone on the same
ear side, and the acoustical dynamic compression means are for signals acquired by the two
external microphones. The sound pressure level may be reduced and compensated.
[0015]
The earplugs closely coupled to the ear canal as in the present invention reduce mid-high
frequency noise entering the ear canal and by the external connection cavity extending from the
earplugs to form a parallel branch with the ear canal, the inside of the ear canal The method of
shunting the low-frequency noises entering into can reduce the noise in the whole band to the
sound signal entering into the ear canal, and can further protect the hearing.
While acquiring an audio signal in the ear canal and an environmental noise signal leaked into
the ear canal by an internal microphone of the headset, an environmental noise signal and an
audio signal propagated through air are acquired by an external microphone of the headset,
Using the signal acquired by the external microphone as a reference signal, after removing the
noise component in the signal acquired by the internal microphone and retaining the audio
component, the technical solution of obtaining the transmitting side audio signal of the headset is
first At the acoustics level the signal in the ear canal acquired by the internal microphone has a
high signal-to-noise ratio by the action of the earplugs and the external connection cavity, and at
the electronic level it is the signal acquired by the internal and external microphones. On the
other hand, adaptive filtering processing is performed to greatly improve the clarity of the
11-04-2019
5
transmitter signal and to emphasize speech.
At the same time, after the sound pressure level is reduced and compensated for the signal
acquired by the external microphone of the headset using acoustic dynamic compression
technology, the sound pressure range is suitable for reception by the human ear After
compression to the range, the technical method of reproducing the signal whose sound pressure
has been compressed through the headset receiver together with the audio signal received by the
headset reduces strong noise to human hearing. Monitoring of the environment can be realized
by avoiding the damage of as well as appropriately increasing the sound with low sound pressure
level to identify useful information to the wearer therein.
As can be understood from the above, the technical solution of the present invention can realize
hearing protection, speech enhancement and three-dimensional environment monitoring in a
high noise environment.
[0016]
FIG. 5 is a structural diagram of a headset to which a method of communicating a headset in a
strong noise environment according to an embodiment of the present invention is applied.
FIG. 5 is a schematic view of a sound distribution structure of a headset according to an
embodiment of the present invention.
FIG. 5 is a schematic view of the sound distribution principle of the headset according to an
embodiment of the present invention.
FIG. 5 is a structural schematic view of an in-ear portion of a headset according to an
embodiment of the present invention.
FIG. 5 is a structural block diagram of a speech enhancement portion of a headset according to
an embodiment of the present invention.
11-04-2019
6
FIG. 5 is a structural block diagram of a headset to which a three-dimensional environment
monitoring function is further added.
5 is a flowchart of an acoustical dynamic compression algorithm according to an embodiment of
the present invention. 5 is a graph of acoustic dynamic compression effects according to an
embodiment of the present invention.
[0017]
In order to make the objectives, technical solutions and advantages of the present invention
clearer, embodiments of the present invention will be described in more detail with reference to
the drawings.
[0018]
The method of headset communication in a strong noise environment according to an
embodiment of the present invention includes several points:
[0019]
(1) The ear plug tightly coupled to the wearer's ear canal reduces high-frequency noise entering
the ear canal and is extended from the ear plug into the ear canal by the external connection
cavity forming a parallel branch with the ear canal The low-frequency noise that enters is
diverted, thereby providing full band noise reduction for the acoustic signal that enters the ear
canal.
Here, noise reduction in all bands can be realized by using a method combining passive noise
reduction technology and sound shunting technology, and an audio signal with a high SN ratio
can be provided to the processing of (2). .
[0020]
(2) An internal microphone of the headset acquires a voice signal in the ear canal and an
environmental noise signal leaked into the ear canal, and an external microphone of the headset
transmits the environmental noise signal and a voice signal propagated through the air Using the
11-04-2019
7
signal acquired and obtained by the external microphone as a reference signal, noise components
in the signal acquired by the internal microphone are removed to retain the voice component,
and then the transmitting side signal of the headset is obtained.
Here, emphasis is given to the sound at two levels: acoustics level (earplugs, external connection
cavity and internal microphone) and electronic level (applicable to the signals of internal
microphone and external microphone) Then, the transmitter side signal with high clarity and
naturalness is obtained.
[0021]
(3) After performing sound pressure level reduction and compensation processing on a signal
acquired by an external microphone of the headset using an acoustic dynamic compression
technique, the processed signal (ie, sound pressure level The sound pressure range of the signal
subjected to the reduction and compensation processing is compressed within the range suitable
for reception by the human ear, and then the processed signal is combined with the receiver
signal received by the headset, and the headset Play through the handset. Here, acoustic dynamic
compression techniques are used to project the intensity range of environmental noise onto the
hearing area of the human ear, thereby avoiding any damage that may be caused to the human
ear by the instantaneous extreme sound, while the background Noise is also shown completely
on the wearer's ear.
[0022]
As can be seen, the above method effectively combines the noise shunting, in-ear microphone,
acoustic signal processing techniques and acoustical dynamic compression techniques to provide
hearing protection, speech enhancement and stereo environment in a strong noise environment.
Monitoring can be realized.
[0023]
FIG. 1 is a structural diagram of a headset to which a method of communicating a headset in a
strong noise environment according to an embodiment of the present invention is applied.
As shown in FIG. 1, the headset comprises an internal microphone 101, two external
11-04-2019
8
microphones 102 on both ears, an external receiver 103 on both ears, an external connection
cavity 104 on both audio sides, and a low power consumption DSP chip. The internal microphone
101 is used to acquire an audio signal having a high S / N ratio in the ear canal and is used to
guarantee the clarity of the call, and the two external microphones 102 on the binaural side.
Enables a three-dimensional environment monitoring to be realized, and a three-dimensional
realistic sound field can be reproduced, and an external microphone 102 on the same ear side as
the internal microphone 101 further provides environmental noise as a reference for speech
enhancement. The binaural receiver 103 reproduces the environmental noise reference signal
together with the receiver side signal, and the binaural side external connection cavity 104 is
shown by a dotted line in the figure to realize sound shunting, and all bands Low power
consumption DSP chip 105 provides voice enhancement processing and acoustic dynamic
compression processing at the electronic level, and dry cell 106 provides DSP chip 105. Feed the
[0024]
In order to explain the technical solution of the present invention in more detail, hereinafter, a
multifunctional headset having a hearing protection function, a voice communication function
and a three-dimensional environment monitoring function in a strong noise background
according to the present invention will be described. Specifically, the following three areas will be
described.
[0025]
1. Hearing protection FIG. 2 is a schematic view of the sound distribution structure of a headset
according to an embodiment of the present invention. The structure of a human ear is
schematically shown in FIG. 2 and includes an ear canal 202 and an eardrum 201. Also shown
schematically in FIG. 2 is the diverted structure of the headset according to an embodiment of
the present invention, which includes a perforated earplug 203 that can be tightly coupled to the
wearer's ear canal, and extends from the earplug 203 perforations. , An external connection
cavity 204 that can be configured with a parallel branch with the ear canal of the wearer. The
external connection cavity 204 communicates with the bore of the earplug 203 via the
connection tube 205. A sound absorbing material 206 is provided on the inner wall of the
external connection cavity 204. The acoustic impedance of the external connection cavity 204 is
significantly smaller than the acoustic impedance of the ear canal. Environmental noise 207 is
diverted at the ear canal opening, and most of the noise 208 is diverted to the external
connection cavity 204 since the acoustic impedance of the external connection cavity 204 is
much smaller than the acoustic impedance of the ear canal, and a small portion of noise 209
11-04-2019
9
Enters the ear canal. Earplugs 203 can reduce mid-high frequency noise entering the ear canal,
and external connection cavity 204 can divert most mid-low frequency noise, thus achieving
noise reduction in all bands. Can.
[0026]
FIG. 3 is a schematic diagram of the sound splitting principle of a headset according to an
embodiment of the present invention. Referring to FIG. 3, one pipe or one cavity is closely
coupled to the ear canal opening, which corresponds to the addition of one parallel branch to the
ear canal when viewed from the ear canal opening, the ear canal The noise that has leaked to the
entrance of is partially diverted to the branch. The smaller the acoustic impedance Zm of the
branch with respect to the ear canal acoustic impedance Ze, the more acoustic energy will enter
the added pipe or cavity, thereby reducing the noise entering the ear canal. Referring to FIG. 3,
the external noise Pa first penetrates the equivalent acoustic impedance Zs of the headset to
reach the ear canal opening, and the residual noise is the sound pressure Pe at the ear canal
opening. If there is no bifurcation branch, all residual noise whose sound pressure at the ear
canal is Pe enters the ear canal, passes through the acoustic impedance Ze of the ear canal,
reaches the tympanic membrane, and evokes hearing . This is a passive noise reduction process.
On the other hand, when the split branch is incorporated, the residual noise whose sound
pressure is Pe is partially split by the split branch, resulting in the sound pressure entering the
ear canal. Here, P'e represents the sound pressure that enters the ear canal of residual noise, and
if Zm is much smaller than Ze, then As can be seen, the effect of the sound shunting is directly
determined by the value of the ratio of Zm to Ze, the smaller the value of the ratio, the greater
the shunting effect. The diverted pipe or cavity mainly exists as one capacitive element, and its
impedance is Here, Ca is the acoustic capacitance of the additional cavity, Va is its volume, c0 is
the speed of sound in air, ω is the angular frequency, and ρ0 is the air density. As can be seen,
the larger the volume of the additional cavity and the smaller its acoustic impedance, the more
pronounced the diversion effect.
[0027]
In the embodiment of the present invention, a noise reduction amount of 30 dB or more can be
realized in the entire band by using a method combining passive noise reduction technology and
sound division technology. Referring to FIG. 2, the structure of this portion is configured to
include earplug 203 that can be closely coupled to the canal, and external connection cavity 204
extended from earplug 203. Earplugs 203 made of rubber or other elastic acoustically resistant
material can effectively block medium high frequency noise. When the residual noise reaches the
11-04-2019
10
entrance of the ear canal 202, it will face two parallel paths between the ear canal 202 and the
external connection cavity 204. When the acoustic impedance of the external connection cavity
204 is set much lower than the acoustic impedance of the ear canal 202, most of the acoustic
energy flows into the external connection cavity 204 and then undergoes multiple random
reflections to absorb the sound absorbing material on the cavity wall. Will be absorbed by By this
method, the acoustic energy entering the ear canal 202 is reduced to achieve the effect of noise
reduction in the entire band. In addition, the dimension of the external connection cavity 204
may be quantitatively controlled so as to cause resonance at a specific frequency. As a result, a
stronger muffling effect occurs near the frequency. By designing and controlling the internal
acoustic structure of the external connection cavity and the distribution of the sound absorbing
material attached to the internal cavity, the strength and band range of the resonant silencer can
be adjusted to achieve the highest noise reduction effect over the entire band .
[0028]
2. Speech enhancement The speech enhancement scheme used in the embodiment of the present
invention includes two parts, the first part performs speech enhancement at the acoustics level,
and the speech enhancement algorithm at the electronic level to the SN ratio To provide a good
main signal and a noise reference signal with high correlation with the main signal, and the
second part is further voice-emphasized on the signal using the acoustic signal processing
method of the tip. And post-processing to improve the speech signal-to-noise ratio and improve
the intelligibility and comfort of the transmitter voice. Hereinafter, the speech enhancement
methods at the acoustics level and the electronic level will be respectively described. As clarified
by the research, when a person's ear canal is blocked from the open air and one closed cavity is
formed, synchronized vibration occurs in air in the ear canal when the person is speaking, The
vibration includes a strong audio signal. Thus, in the headset provided by the present invention,
the in-ear microphone acquires audio signals in the ear and leaked residual noise signals, and the
external microphone acquires environmental noise signals. The noise of the internal microphone
signal is transmitted using the adaptive filtering method at the electronic level by simultaneously
transmitting both the internal microphone and external microphone signals to the audio signal
processing means, using the internal microphone signal as the main signal and the external
microphone signal as the reference signal. The signal is adaptively removed to retain the speech
component, and finally, frequency spectrum compensation is performed on the adaptively
filtered speech signal to obtain a transmit side speech signal with high intelligibility and
naturalness. Specifically, refer to FIG. 4 and FIG.
[0029]
11-04-2019
11
FIG. 4 is a structural diagram of an in-ear portion of a headset according to an embodiment of the
present invention. The structure of a human ear is schematically shown in FIG. 4 and includes an
ear canal 202 and an eardrum 201. Also shown schematically in FIG. 4 is the structure of the inear portion of the headset according to an embodiment of the present invention, an earplug 203
that can be tightly coupled with the ear canal 202, and audio signals in the ear canal and leakage
into the ear canal. An internal microphone 404 for acquiring a residual noise signal, and an
external microphone 405 for acquiring an environmental noise signal. Referring to FIG. 4, the
internal microphone 404 is located at a portion of the earplug 203 that can enter the ear canal
when the headset is worn, and the external microphone 405 is the earplug 203 when the headset
is worn. Located outside the ear canal. The headset according to the embodiment of the present
invention receives the signal acquired by the internal microphone 404 and the signal acquired by
the external microphone 405, and uses the signal acquired by the external microphone 405 as a
reference signal by the internal microphone 404. The apparatus further comprises audio signal
processing means for obtaining a transmission side signal of the headset after removing noise
components in the acquired signal and retaining the audio components. The audio signal
processing means is not shown in FIG. The audio signal processing means may be respectively
connected to the internal microphone 404 and the external microphone 405, and the position
may be arranged in a rational part of the headset according to the actual situation, which is an
implementation of the present invention. It does not affect the implementation of the example.
[0030]
When a person is speaking, an audio signal is transmitted to the ear canal through the ear canal,
and when making a sound, the muscles in the ear canal vibrate and at the same time produce an
air vibration, i.e. produce an audio. When the ear canal opening is open, the vibration (source) of
the gas in the ear canal radiates to a large space (large load), the gas amplitude is small, and the
sound energy is weak while the ear canal opening is blocked. If so, the vibration (source) of the
gas in the ear canal only acts on a very small space (small load) in the ear canal, the gas's
amplitude is large, the sound energy is strong, and the external noise is passive. Soundproofed,
the energy transmitted to the ear canal is reduced and the SN is greatly improved. Thus, referring
to FIG. 4, the internal microphone 404 acquires audio signals in the ear and leaked
environmental noise signals, and the external microphone 405 acquires environmental noise
signals and audio signals propagated through air. Do. The sound signal acquired by the internal
microphone 404 already has a relatively good signal-to-noise ratio, as environmental noise is
already greatly attenuated when it enters the ear canal through blocking of the earplug and
shunting of the externally connected cavity. The relatively pure external noise signal acquired by
the external microphone 405 can be provided as a good external noise reference signal for noise
reduction at the next electronic level. Spatially, because the distance between the internal
11-04-2019
12
microphone 404 and the external microphone 405 is relatively close, it is ensured that the
acquired external noise signal has a good correlation, which further reduces the noise signal at
the electronic level What can be done is secured.
[0031]
After speech enhancement at the acoustic level, acoustic signal processing techniques at the
electronic level are further used to further improve the signal-to-noise ratio of the speech signal
as well as improve the naturalness and intelligibility of the speech signal. Specifically, reference
is made to FIG.
[0032]
FIG. 5 is a structural block diagram of the speech enhancement portion of the headset according
to an embodiment of the present invention. As shown in FIG. 5, the headset includes an internal
microphone 404, an external microphone 405, and an audio signal processing means 506.
[0033]
Specifically, the audio signal processing means 506 receives the signal s1 acquired by the
internal microphone 404 and the signal s2 acquired by the external microphone 405, and the
statistical energy ratio between s2 and s1 within the low frequency range. , Specifies the control
parameter α, updates the reference signal using the voice detection module 5061 for outputting
the control parameter α, and the feedback output signal y as a weight, and outputs the control
parameter α output by the voice detection module 5061. Adaptive filtering processing is
performed on the received signal s2 of the external microphone 405 using the weight update
rate control parameter as the control parameter of the weight update rate, and an adaptive filter
5062 for outputting an adaptive filter output signal s3 The adaptive filter received from the
received signal s1 By subtracting the force signal s3, it includes a noise reduction module 5063
for obtaining an output signal y, and post-processing module 5064 for performing a singlechannel audio processing and frequency spread spectrum processing on the output signal y, a.
The output signal of the post-processing module is the transmitter signal of the headset.
[0034]
11-04-2019
13
Voice detection module 5061: If a voice signal is present, the internal microphone 404 will
capture many voice signals in the ear canal, but if the wearer of the headset speaks loudly it will
be propagated through the air and be an external microphone The audio signal acquired by 405
can not be ignored either. If the signal of the external microphone 405 is directly updated as a
reference signal and the adaptive filter is updated, the sound may be damaged. Therefore, in the
present invention, the speech detection module 5061 is added, and the speech detection module
5061 outputs the control parameter α. The control parameter α is mainly used to weight the
convergence step size of the adaptive filter. The value of the control parameter α is mainly
specified by calculating the statistical energy ratio of the external microphone to the internal
microphone in the low frequency range, and the numerical range of α is 0 ≦ α ≦ 1.
[0035]
Adaptive Filter 5062 and Noise Reduction Module 5063: The adaptive filter 5062 is an FIR filter
of order P (P ≧ 1), the filter weights are and in one embodiment of the present invention P = 64
It becomes. The input signal of the adaptive filter 5062 is s2 (n), where n is the discrete time
number, and the output signal of the adaptive filtering is s3 (n), and s3 (n) and s1 (n) Subtraction
processing is performed between the two to obtain a signal y (n) after cancellation, y (n) is fed
back to the adaptive filter to update the filter weight, and the update speed is controlled by the
parameter α. In the case of α = 1, that is, when s1 (n) and s2 (n) are all noise components, the
adaptive filter 5062 rapidly converges on the transfer function H_noise from the external
microphone 405 to the internal microphone 404 of noise. , S3 (n) to be the same as s1 (n) so that
y (n) after cancellation becomes smaller, thereby removing noise. When α = 0, that is, when s1
(n) and s2 (n) are all target speech components, the adaptive filter 5062 stops the update,
whereby the adaptive filter 5062 transmits the audio from the external microphone 405 to the
internal microphone 404. Since the convergence does not occur in the transfer function
H_speech up to the point s3 (n) is different from s1 (n), the speech component after subtraction
is not canceled and the speech component remains at the output y (n). If 0 <α <1, that is, if both
the speech and noise components are present in the signal collected by the microphone, then the
speech is to ensure that noise is eliminated and the speech component is retained. The amount of
components and noise components control the update rate of adaptive filter 5062. The transfer
function H_noise from the noise external microphone 405 to the internal microphone 404 is
similar to the transfer function H_speech from the external microphone 405 to the internal
microphone 404 of voice, so even if the adaptive filter 5062 converges to H_noise, Α is used to
constrain the weight of the adaptive filter 5062 as it will cause some damage to the speech. The
constraint made in this example is that if α = 1, ie if all the collected signals are noise
components, then the adaptive filter 5062 is not constrained and the noise is completely
11-04-2019
14
eliminated. , If α = 0, ie if all the collected signals are speech components, then the adaptive
filter 5062 will be completely constrained and speech will remain completely, if 0 <α <1 ie If
both the speech and noise components are present in the signal collected at the microphone, the
adaptive filter 5062 will be partially constrained, some noise will be removed, and the speech
will remain completely.
In this way, the effects of reducing noise and well protecting voice are achieved.
[0036]
Post-processing module 5064: The post-processing module 5064 first performs two-channel
audio enhancement processing on the signal y output from the noise reduction module 5063 to
further improve the SN ratio of the audio signal, Then, frequency spectrum spreading is
performed on the single-channel processed signal to improve the intelligibility and intelligibility
of the output speech signal. Single channel speech enhancement and frequency spread spectrum
are not described in detail here, as conventional mature schemes can be used.
[0037]
As can be seen from the above, the headset according to the embodiment of the present
invention acquires an audio signal in the ear using an internal microphone and acquires an audio
signal having a high S / N ratio in terms of speech enhancement, and an acoustic level While
providing ambient noise with an external microphone, providing suitable requirements for
electronic-level speech enhancement, and at the electronic level, mainly an internal microphone
signal, and an external microphone signal. The background noise is further removed by an
adaptive filtering scheme. Such a scheme provides an original signal with a sufficient S / N ratio
even in the case of extreme noise as compared to the conventional method of emphasizing
speech using a close-talking microphone, and detects the speech signal. And because it can be
used as a basis for judgment, the intelligibility and intelligibility of the transmitter voice are
guaranteed.
[0038]
It should be described that in the present embodiment, the sound shunting technique is adopted
to shunt most of the noise that has reached the ear canal opening by the external connection
cavity, so it passes through the earplugs, passes through the sound shunting, and then enters the
11-04-2019
15
ear canal. Incoming environmental noise is already greatly attenuated and the audio signal
acquired by the internal microphone 404 already has a high signal-to-noise ratio.
[0039]
3.
Three-Dimensional Environment Monitoring FIG. 6 is a structural block diagram of a headset in
which a three-dimensional environment monitoring function is further added to FIG. 6 and 4, the
headset according to the present embodiment includes the structure of the headset shown in FIG.
5, and in addition, the headset according to the present embodiment has an external microphone
405 'on the other ear side. And an acoustic dynamic compression means 601 and a receiver 602,
wherein the external microphone 405 and the external microphone 405 'are located on the both
ears side, and the external microphone 405 and the internal microphone 404 are located on the
same ear side. The acoustic dynamic compression means 601 receives the signal s2 acquired by
the external microphone 405 and the signal s3 acquired by the external microphone 405 ′, and
acquires the signal s2 acquired by the two external microphones using the acoustic dynamic
compression technology. The receiver 602 is used by the acoustic dynamic compression means
601 to process the entire sound pressure range of the signals s 2 and s 3 within the range
suitable for reception by the human ear. The signal s2 'and s3' which, together with the
reception-side signal L received by the headset, to play in the wearer's ear.
[0040]
In the present embodiment, the number of internal microphones is one, located on the left ear
side or the right ear side, and the number of external microphones is two, located on the left ear
side and the right ear side, respectively. The audio signal processing means receives the signals
acquired by the internal microphone and the external microphone on the same ear side. The
acoustic dynamic compression means performs sound pressure level reduction and
compensation processing on the signals acquired by the two external microphones.
[0041]
In the embodiment of the present invention, in terms of three-dimensional environment
11-04-2019
16
monitoring, environmental noise is acquired by the external microphone 405 and the external
microphone 405 ′ and transmitted to the acoustic dynamic compression means 601, and the
acoustic dynamic compression means 601 first Alternatively, estimation is performed on the
energy magnitude of the signal in the frequency domain, and then the gain is adjusted according
to the energy magnitude. Specifically, it gives small (less than 1) gain to a signal of large energy
and gives large (greater than 1) gain to a signal of small energy. This modulation reduces strong
noise that can damage human hearing in the environment and moderately increases sound with
low sound pressure levels, allowing the wearer to figure out its useful information become. The
overall sound pressure dynamic range is compressed within the range that can be received by
the human ear, provided that environmental acoustic information is not compromised. For
example, when the environmental noise sound pressure range is 20 dB to 160 dB, the sound
pressure range can be compressed to 40 dB to 90 dB after processing.
[0042]
As understood from this, in this embodiment, after the sound pressure level is reduced and
compensated for the signals acquired by the external microphones on both sides of the headset,
the sound pressure range is received by the human ear. After being compressed into a suitable
range, the processed signal is reproduced via the headset handset together with the receiver
signal received by the headset.
[0043]
FIG. 7 is a flow chart of an acoustic dynamic compression algorithm according to an embodiment
of the present invention.
FIG. 8 is a graph of acoustic dynamic compression effects according to an embodiment of the
present invention. Referring to FIGS. 7 and 8, in the aspect of three-dimensional environment
monitoring, environmental sound is acquired by the external microphone 405 and transmitted to
the acoustic dynamic compression means 601. As shown in FIG. 7, the process of the acoustic
dynamic compression means 601 mainly includes: Fourier transform (FFT), 2. Calculation of
sound pressure of feature points Determination of gain values of feature points; Differential
calculation of gain array over frequency domain, 5. Determination of frequency spectrum after
compensation; Includes several parts, such as the inverse Fourier transform (IFFT). Referring to
FIG. 8, as a result after processing, a noise area ASZ (ambient sound zone) is projected to a
person's ear hearing area AZ (audibility zone). The signal processed by the acoustic dynamic
compression means 601 is reproduced by the receiver 602 in the wearer's ear, along with the
audio signal received by the headset. In this way, the wearer's hearing can not be damaged, and
11-04-2019
17
the wearer can monitor the sound of the surrounding environment.
[0044]
In the embodiment of the present invention, the audio signal processing means 506 and the
acoustic dynamic compression means 601 are integrated in one DSP chip. The headset according
to an embodiment of the present invention further comprises a dry cell for powering the DSP
chip. DSP chips powered by dry batteries can have low power consumption and can guarantee
strong durability.
[0045]
Summarizing the above, in the embodiment of the present invention, effective combination of
sound shunting technology, in-ear microphone technology and acoustic dynamic compression
technology, effective hearing protection function in a strong noise environment, clear voice
communication function and stereoscopic environment monitoring Provided a multifunctional
headset that can provide functions. The headset according to this embodiment has the following
advantages as compared to the conventional headset.
[0046]
(1) In terms of hearing protection, a special sound shunting technology achieves a noise
reduction of over 30 dB in all bands, and it is passive without using an active noise reduction
technology that has a complicated structure and high energy consumption. Noise reduction
techniques reduce the power of the headset and greatly improve the durability of the headset. (2)
In terms of voice communication, an in-ear microphone acquires a voice signal in the ear, and
voice signal processing means further removes background noise, and the method uses a close
talk microphone or a bone conduction microphone. Compared to the general method of speech
enhancement, even in the case of extreme noise, the original signal still having a sufficient signalto-noise ratio can be provided and used for speech signal detection and determination. (3) In
terms of environmental monitoring, using acoustic dynamic compression technology, the
acoustic dynamic compression algorithm of the tip realizes projecting the intensity range of
environmental noise to the hearing area of the human ear, which is instantaneous Background
noise is also completely indicated to the wearer's ear, while extreme sound pressure avoids
possible damage to the human ear.
11-04-2019
18
[0047]
What has been described above is merely a preferred embodiment of the present invention and is
not intended to limit the protection scope of the present invention. All changes, equivalent
replacements, improvements, etc. made within the spirit and principle of the present invention
shall fall within the protection scope of the present invention.
11-04-2019
19
Документ
Категория
Без категории
Просмотров
0
Размер файла
35 Кб
Теги
description, jp2015524224
1/--страниц
Пожаловаться на содержимое документа