CN104575486B - Sound leakage protection method and system based on the principle of acoustic masking - Google Patents
Sound leakage protection method and system based on the principle of acoustic masking Download PDFInfo
- Publication number
- CN104575486B CN104575486B CN201410822950.2A CN201410822950A CN104575486B CN 104575486 B CN104575486 B CN 104575486B CN 201410822950 A CN201410822950 A CN 201410822950A CN 104575486 B CN104575486 B CN 104575486B
- Authority
- CN
- China
- Prior art keywords
- sound
- interference
- signal
- voice signal
- unit
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
Abstract
The present invention provides a kind of sound leakage protection system based on the principle of acoustic masking, which includes: sound monitoring unit, sound interference unit and middle control unit.The present invention also provides a kind of sound leakage protection methods based on the principle of acoustic masking, comprising: is acquired using sound monitoring unit to voice signal, and obtains A weighted sound pressure level；Judge whether the voice signal is voice signal, if the voice signal is voice signal and the A weighted sound pressure level is more than preset threshold value in a period of time, issues alarm signal；Middle control unit carries out analysis and feature extraction to the voice signal, generates interference signal, and the interference signal is sent to sound interference unit；Sound interference unit carries out sound interference to each region using the interference signal.The masking signal and target voice that the present invention is handled in real time have the higher degree of correlation, and masking effect more preferably, can preferably realize the protection to general privacy.
Description
Technical field
The present invention relates to speech signal processing technologies, and in particular to sound leakage protection method based on the principle of acoustic masking
And system.
Background technique
As an important carrier of people's information exchange and transmission, it is most that there are occasions for phonetic sound, and the information content of carrying is also most
Horn of plenty.And it will cause the destructions of language privacy for the leakage of sensitive sound information.When people talk in places such as rooms
Words, produce a large amount of acoustic information.Inside plant condition and security situation are directly related to the safety of speech content.In order to
Language privacy and safety are improved, needs to carry out acoustic intelligence leakage protection.
The a part of sound leakage protection system as inside plant, can effectively to produced by people's daily life, deposit
The acoustic intelligence of storage and processing carries out security protection.Existing sound leakage protection system, including sound collector, central processing unit and
Sound jammer.But sound jammer only includes voice interference unit in existing system, does not include vibration interference device, then existing
System can not prevent laser eavesdropping, pipe vibration eavesdropping and wall vibrations eavesdropping etc., and voice signal can still leak.And it is existing
System uses traditional white noise or pink noise masking signal, but since this masking signal is related to masked source signal
It spends low, is then easy to by restoring masked source signal after some simple processing.
Summary of the invention
In view of the drawbacks of the prior art, the present invention provides a kind of sound leakage protection method based on the principle of acoustic masking and is
System obtains sound masking signal similar to the target speech signal, and be directed to language by carrying out Speech processing to voice signal
Sound signal generates space acoustic jamming and vibration interference, so that system masking effect is good, it is not easy to be reduced out targeted voice signal.
In a first aspect, the present invention provides a kind of sound leakage protection system based on the principle of acoustic masking, and the system comprises: sound
Sound monitoring unit, sound interference unit and middle control unit；
Sound monitoring unit is connected with the central control management unit, and is differentiated for sound collection, speech detection, sound pressure level
And sound-light alarm；
Sound interference unit is connected with the central control management unit, transaudient dry for carrying out air-borne sound interference and vibration
It disturbs；
Middle control unit is respectively connected with the sound interference unit, for providing people with the sound monitoring unit
Machine interface, and control the sound monitoring unit and sound interference unit collaborative work.
Preferably, the sound monitoring unit, is specifically used for:
Sound signal collecting, weighted filtering, octave analysis, sound pressure level measurement and speech detection.
Preferably, the sound interference unit includes: spatial sound interference module, pipe vibration interference module, glass vibration
Dynamic interference module and wall vibration interference module；
Spatial sound interference module is installed on the outside of room enclosure structure, for by generating voice masking signal, interference
Air sound leakage；
Pipe vibration interference module is installed on the pipeline across room enclosure structure, fresh air and air conditioning, for leading to
Cross generation vibration interference signal, prevent because pipe vibration it is transaudient caused by indoor sensitive acoustic intelligence leakage；
Glass vibration interference module is installed on the window-glass inside room, generates vibration interference signal for passing through,
Indoor sensitivity acoustic intelligence leakage caused by preventing because of laser eavesdropping；
Wall vibration interference module is installed on the wall at setting position, for preventing by generating vibration interference signal
Because of sensitive acoustic intelligence leakage indoor caused by vibrating wiretap.
Preferably, the system also includes network units, for connecting the sound interference unit and the middle keyholed back plate reason
Unit, so that the sound interference unit and the middle control unit are communicated by network protocol.
Second aspect, the present invention provides a kind of sound leakage protection methods based on the principle of acoustic masking, which comprises
Voice signal is acquired using sound monitoring unit, and obtains A weighted sound pressure level；
Judge whether the voice signal is voice signal, if voice signal institute for voice signal and in a period of time
A weighted sound pressure level is stated more than preset threshold value, then issues alarm signal；
Middle control unit carries out analysis and feature extraction to the voice signal, generates interference signal, and will be described dry
It disturbs signal and is sent to sound interference unit；
Sound interference unit carries out sound interference to each region using the interference signal.
Preferably, the method also includes:
Middle control unit regularly sends heartbeat signal to sound interference unit, if not receiving the heart in the predetermined time
The response signal for jumping signal then determines the sound interference unit broken string, and issues alarm signal.
Preferably, described that voice signal is acquired using sound monitoring unit, and A weighted sound pressure level is obtained,
Include:
Data acquisition is carried out to voice signal, obtains instantaneous sound pressure level intensity value；
Octave analysis is carried out to the voice signal, obtains the sound pressure level strength in different frequency bands；
Weighted filtering is carried out to the voice signal, obtains A weighted sound pressure level value.
It is preferably, described to judge whether the voice signal is voice signal, comprising:
The voice signal is divided into several frames；
The short-time energy of every frame voice signal and the short-time zero-crossing rate of every frame voice signal is calculated；
Using double threshold method, judge whether the voice signal is voice signal.
Preferably, the middle control unit carries out analysis and feature extraction to the voice signal, generates interference signal,
And the interference signal is sent to sound interference unit, comprising:
The voice signal is analyzed, the temporal signatures and frequency domain character of the voice signal are extracted；
Based on the temporal signatures and frequency domain character, reversion and scramble are carried out to the voice signal, generate masking sound letter
Number；
The sound masking signal is sent to the spatial sound interference module in sound interference unit, and controls pipe vibration
Interference module, glass vibration interference module and wall vibration interference module are started to work.
Preferably, the sound interference unit carries out sound interference to each region using the interference signal, comprising:
Spatial sound interference module carries out sound interference using the sound masking signal received；
Pipe vibration interference module, glass vibration interference module and wall vibration interference module, according to middle control unit
Control, carry out vibration interference.
As shown from the above technical solution, the present invention provides a kind of sound leakage protection method based on the principle of acoustic masking and is
System obtains sound masking signal similar to the target speech signal, and be directed to language by carrying out Speech processing to voice signal
Sound signal generates space acoustic jamming and vibration interference, so that system masking effect is good, it is not easy to be reduced out targeted voice signal.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these figures.
Fig. 1 is the structural schematic diagram for the sound leakage protection system based on the principle of acoustic masking that one embodiment of the invention provides；
Fig. 2 be another embodiment of the present invention provides S3C2410 minimum system structural schematic diagram；
Fig. 3 is the flow diagram for the sound leakage protection method based on the principle of acoustic masking that one embodiment of the invention provides；
Fig. 4 be another embodiment of the present invention provides sound monitoring unit workflow schematic diagram.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
As shown in Figure 1, the structure of the sound leakage protection system based on the principle of acoustic masking provided for one embodiment of the invention
Schematic diagram, the system include sound monitoring unit 101, sound interference unit 102 and middle control unit 103.
Sound monitoring unit 101, is connected with the central control management unit, and sentences for sound collection, speech detection, sound pressure level
Other and sound-light alarm.
Sound interference unit 102, is connected with the central control management unit, transaudient for carrying out air-borne sound interference and vibration
Interference.
Middle control unit 103, is respectively connected with the sound interference unit, for providing with the sound monitoring unit
Man-machine interface, and control the sound monitoring unit and sound interference unit collaborative work.
Specifically, sound monitoring unit 101 are installed on interior using Distributed Design, pass through acquisition current indoor
Voice signal simultaneously calculates A weighted sound pressure level, issues acousto-optic hint when the sound pressure level intensity exceeds a specified threshold value, if indoor sound
Intensity continuous of arbitrarily downgrading is more than that threshold value then sends alarm signal to middle control management system, to control the work of sound interference system.For
The specific aim of protection is improved, sound monitoring unit 101 uses speech recognition algorithm, when indoor sound pressure level intensity is more than threshold value
When, judge whether current sound is voice signal, does not issue warning note then if other non-voice-like interferences.Then sound is supervised
It surveys unit 101 and is mainly used for sound signal collecting, weighted filtering, octave analysis, sound pressure level measurement and speech detection.
Since sound wave is mechanical wave, the vibration of wall, pipeline, door and window etc., and then expansion sound can be caused in transmission process
The range that information is propagated, therefore sound interference unit 102 includes space acoustic jamming unit and vibrates acoustic jamming unit two major classes, and
Sound interference unit 102 specifically includes:
(1) spatial sound interference module is installed on the outside of room enclosure structure, generates voice masking signal for passing through,
Prevent air sound leakage.
(2) pipe vibration interference module is installed on the pipeline across room enclosure structure, fresh air and air conditioning, uses
In by generating vibration interference signal, prevent because of pipe vibration it is transaudient caused by indoor sensitive acoustic intelligence leakage.
(3) glass vibration interference module is installed on the window-glass inside room, for by generating vibration interference letter
Number, indoor sensitivity acoustic intelligence leakage caused by preventing because of laser eavesdropping.
(4) wall vibration interference module is installed on the wall at setting position, generates vibration interference signal for passing through,
Indoor sensitive acoustic intelligence leakage caused by preventing because of vibration wiretap.
Wherein, spatial sound interference module is for carrying out the transaudient interference in space, and pipe vibration interference module, glass vibration
Interference module and wall vibration interference module are used to vibrate transaudient interference, and vibrating transaudient interference is for laser eavesdropping, pipe
Road vibration eavesdropping and wall vibrations are eavesdropped and the interference of implementation.
And middle control unit 103 provides man-machine interface and cooperates with sound monitoring unit 101 and 102 work of sound interference unit
Make.Middle control management system uses standardized network interface, connect respectively with sound monitoring unit 101 and sound interference unit 102:
Four road sound collections/monitoring input (major function is sound collection, monitoring, alarm), four mouthfuls of (16 tunnel) sound interference outputs
(main control air acoustic jamming and vibration interference).When indoor sound pressure level intensity is more than defined threshold and is voice signal, sound
Sound monitoring unit 101 issues alarm alert signal to middle control unit 103, and middle control unit 103 receives each road prompt letter
Macro or mass analysis is carried out after number, when it is necessary to turn on the masking sound, control life is issued to sound interference unit 102 by network protocol
It enables, opens air sound masking interference module and pipe vibration interference module, glass vibration interference module, wall vibrations interfere mould
Block.Continue to be monitored current indoor sound pressure level strength simultaneously to stop emitting after it restores normal sound pressure level strength and do
Disturb sound.
It should be noted that since sound interference unit 102 and middle control unit 103 are led to by network protocol
Letter, therefore the system further includes network unit, for connecting sound interference unit 102 and middle control unit 103, so that sound
Interference unit 102 and middle control unit 103 are communicated by network protocol.
In the present embodiment, which can monitor the sound pressure level strength in room, when the preset value is exceeded, issue alarm, and
Air-borne sound interference noise relevant to voice is generated, and for the vibration noise of the weak links such as door, window, pipeline, to reach
The purpose of sensitive sound information inside to protection room.Sound monitoring unit uses the relevant interference signal of voice, can also be real
When collection room in voice signal handled, be converted into masking signal.The masking signal and target voice handled in real time has more
The high degree of correlation, masking effect more preferably, can preferably realize the protection to general privacy.
In the following, illustrating a kind of sound leakage protection system based on the principle of acoustic masking by a specific embodiment.Its
In, middle control unit uses S3C2410 embedded platform, runs built-in Linux operating system.Sound monitoring unit uses
MSP430 series monolithic is as coprocessor, and sound interference unit is using WT588D as sound module.When main control platform is adopted
After collecting the target voice of current indoor and carrying out feature extraction matching, obtained best masking sound source information is sent to masking
Module occurs for sound, and masking sound occurs module and selects corresponding masking sound in the pre-stored masking sound source, carries out interference transmitting.
During the launch process, it is communicated always with main control platform, obtains the sound status information of current indoor, including sound pressure level strength
Change with target voice, to carry out Adaptive matching.To prevent masking sound module from unexpected broken string, main control platform and masking occurs
Heartbeat signal is periodically transmitted between sound module.And the working condition of the sound module of adjustment masking in real time.
Wherein, S3C2410 processor is based on ARM920T processor core, using 32 micro-controls of 0.18um manufacturing process
Device processed.ARM920T realizes memory management unit (Memory Management Unit, abbreviation MMU), on piece middle line
(Advanced Microcontroller Bus Architecture, AMBA) and Harvard framework Harvard cache hierarchy
Structure.This structure has independent 16KB instruction cache Cache and 16KB data Cache, each is by 8 word lengths
Row is constituted.
In order to reduce the cost of whole system, S3C2410 provides following internal unit abundant: LCD controller, NAND
Flash controller, sdram controller, 3 channel UART, 4 channel DMA, Timer of 4 tunnels with PWM, I/O mouthfuls, RTC, 8 10, tunnels
ADC, Touch Screen interface, IIC-BUS interface, IIS-BUS interface, 2 usb hosts, 1 USB device, SD host and
MMC interface, 2 road SPI.By providing a series of complete system peripherals, S3C2410 greatly reduce whole system at
This, eliminates as the needs of system configuration additional devices.
Wherein, middle control unit uses embedded system, can carry out flexible customization according to application.Embedded system
It is low in energy consumption, small in size, specificity is strong.The integrated portion in the chip of task that many in computer can be completed by board, thus
Be conducive to Embedded System Design to tend to minimize.Speed and system reliability are executed in order to improve, it is soft in embedded system
Part solidifies in the single chip microcomputer itself.Desktop Linux OS is carried out cutting modification by the embedded OS of product operation
It forms, makes it to run in embedded computer systems.Built-in Linux had both inherited open source code unlimited on network
Resource, and the characteristic with embedded OS.
Specifically, the minimum system 28M RAM and 256M Flash composition for the S3C2410 that middle control unit uses,
Its block diagram is as shown in Figure 2.
And since sound interference unit is communicated by network protocol with middle control unit, network list need to be increased
Member.In the present embodiment, network unit uses CS8900A chip, which is 10M network chip.It is used with middle keyholed back plate platform
The User Datagram Protocol (User Datagram Protocol, abbreviation UDP) of standard is communicated.To guarantee sound interference list
Connection between first and middle control unit, middle control unit regularly send heartbeat signal and receive the sound of sound interference unit
It answers, if a certain interference module sends wire break alarm signal without feedback response.
Based on above system, as shown in figure 3, anti-for the sound leakage based on the principle of acoustic masking that one embodiment of the invention provides
The flow diagram of maintaining method, this method comprises:
301, voice signal is acquired using sound monitoring unit, and obtains A weighted sound pressure level.
302, judge whether the voice signal is voice signal, if the voice signal is voice signal and a period of time
The interior A weighted sound pressure level is more than preset threshold value, then issues alarm signal.
Wherein, the distinguishable voice of sound monitoring system and ambient noise, are monitored only for voice signal；When indoor language
When speech is arbitrarily downgraded more than preset threshold, system alarm；Type of alarm is light prompt or auditory tone cues.
303, middle control unit carries out analysis and feature extraction to the voice signal, generates interference signal, and by institute
It states interference signal and is sent to sound interference unit.
304, sound interference unit carries out sound interference to each region using the interference signal.
Optionally, this method further include:
Middle control unit regularly sends heartbeat signal to sound interference unit, if not receiving the heart in the predetermined time
The response signal for jumping signal then determines the sound interference unit broken string, and issues alarm signal.
In the present embodiment, as shown in figure 4, step 301 specifically comprises the following steps:
401, sound monitoring unit carries out data acquisition to voice signal, obtains instantaneous sound pressure level intensity value.
Wherein, the measurement of sound pressure level includes instantaneous sound pressure grade, equivalent continuous sound pressure level, peak sound pressure level etc., in this method
The middle standard using instantaneous sound pressure level intensity value as measurement indoor sound pressure level intensity.
402, octave analysis is carried out to the voice signal, obtains the sound pressure level strength in different frequency bands.
Specifically, resolution capability not instead of single-frequency of the human ear to sound, frequency band.Octave analysis is mainly used for
Signal is divided by frequency characteristic, is allowed to meet the frequency resolution of human ear.To measure voice signal in different frequencies
Energy distribution in band, to show the sound pressure level strength in current indoor different frequency bands.
403, weighted filtering is carried out to the voice signal, obtains A weighted sound pressure level value.
In this step, since response of the human ear to different frequency sounds is different, in order to keep instrument measurements directly anti-
The subjective loudness for reflecting people is felt, usually by the sound pressure level measured by a rectangular window, by each frequency component in sound wave
It is weighted processing.The filter is referred to as " weighting network ".Common weighted filter includes tetra- kinds of A, B, C, D, and wherein A is counted
The frequency decay of power meets human ear, therefore when measuring indoor sound pressure level, indicates usually using A weighted sound pressure level.
In the present embodiment, sound monitoring unit needs to judge whether the voice signal is voice signal in step 302, and
Speech recognition algorithm used in this step is used in such a way that short-term average energy method and short-term average zero-crossing rate method combine
Double threshold method detects voice signal.Include:
A01, the voice signal is divided into several frames.
The short-time zero-crossing rate of A02, the short-time energy that every frame voice signal is calculated and every frame voice signal.
Wherein, if n-th frame voice signal xn(m) short-time energy EnIt indicates, shown in calculation formula such as formula (1):
If voice signal xn(m) short-time zero-crossing rate ZnIt indicates, shown in calculation formula such as formula (2):
Wherein, sgn [] is sign function, i.e., as shown in formula (3):
A03, using double threshold method, judge whether the voice signal is voice signal.
In this step, a higher thresholding is chosen according to the voice short-time energy of voice signal and short-time zero-crossing rate
Amp1, zcr1, and according to the average energy of the ambient noise of estimation and Average zero-crossing rate determine a lower thresholding amp2,
Zcr2 determines that voice starts if the short-time energy of the voice signal of detection is greater than higher thresholding.If short after a period of time
Shi Nengliang is less than lower thresholding, then determines that voice terminates.If voice insufficient certain section of time span after starting, short-time energy is with regard to small
In compared with low threshold, then determining that section of sound just for noise section.For short-time zero-crossing rate, judgment criterion is analogous in short-term
Energy determining method.Judged in conjunction with short-time energy and short-time zero-crossing rate, judging result can be made more accurate.
In the present embodiment, the detailed process of step 303 includes:
B01, the voice signal is analyzed, extracts the temporal signatures and frequency domain character of the voice signal.
In this step, speech waveform is the continuous function of time, it will be noted that a sound to another sound gradually
Transition.The characteristic of voice signal changes over time.For example, the excitation of voiced sound and voiceless sound is different, it is transitioned into voiceless sound from voiced sound,
To change excitation accordingly；The amplitude of voice signal has significant variation with the time；Even if being both voiced sound, fundamental frequency
Also it is different.These time-varying characteristics of voice signal can clearly observe in waveform.But voice signal
Characteristic to change with time be that comparison is slow, can substantially think the voice signal in every 10-30ms short time interval
Characteristic it is substantially changeless.Therefore, one section of progress Short Time Fourier Analysis can be therefrom intercepted, the short of voice is obtained
Shi Pu.
There is spectrum analysis function to be closely related by the perception of voice signal and human auditory system, different phonetic
It is to be determined by sound channel different shape and size, this just determines that they have different spectral characteristics.There is sound channel one group to be total to
Vibration frequency, referred to as formant.Different voices corresponds to a different set of formant parameter, and spectral characteristic is just main to be reflected
The frequency bandwidth of the different location of these formants and each peak.
B02, it is based on the temporal signatures and frequency domain character, reversion and scramble is carried out to the voice signal, generate masking
Acoustical signal.
In this step, best sound masking signal need to be chosen by interfering indoor sensitive sound signal, that is, reach same
Signal transmission power is minimum under conditions of masking effect.
Good sound masking signal should be higher with the degree of correlation of masked sound source, that is, uses the relevant masking noise of voice
Interference can be realized under the condition of low noise emission intensity.It is not difficult to judge directly in real time using speaker (target sound source)
There should be best masking effect as masking sound source after voice is transformed, however implement sound masking, human speech of speaking for room
Sound itself has certain confidentiality, is not suitable for as masking sound source, without the meaning of one's words and not noticeable in order to meet masking sound
Requirement so that masking sound will not become new interference sound source, obtain after most preferably sheltering sound source, need to continue to shelter sound source sound
Sound signal is handled.This method uses audio signal processing method, is inverted to masking sound source, generation and target after scramble
The most similar sound masking signal of voice.
B03, the sound masking signal is sent to the spatial sound interference module in sound interference unit, and controls pipeline
Vibration interference module, glass vibration interference module and wall vibration interference module are started to work.
And step 304, detailed process is as follows:
Spatial sound interference module carries out sound interference using the sound masking signal received；
Pipe vibration interference module, glass vibration interference module and wall vibration interference module, according to middle control unit
Control, carry out vibration interference.
A kind of sound leakage protection method based on the principle of acoustic masking is present embodiments provided, by carrying out language to voice signal
Sound signal processing, obtain sound masking signal similar to the target speech signal, and for voice signal generation space acoustic jamming and
Vibration interference, so that system masking effect is good, it is not easy to be reduced out targeted voice signal.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that；It still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (9)
1. a kind of sound leakage protection system based on the principle of acoustic masking, which is characterized in that the system comprises: sound monitoring is single
Member, sound interference unit and middle control unit；
Sound monitoring unit is connected with the central control management unit, and for sound collection, speech detection, sound pressure level differentiates and sound
Light alarm；
The sound monitoring unit obtains A weighted sound pressure level for being acquired to voice signal；Judge the sound
Whether signal is voice signal, if the voice signal is voice signal and the A weighted sound pressure level is more than preset threshold
Value, then issue alarm signal；
Sound interference unit is connected with the central control management unit, for carrying out air-borne sound interference and the transaudient interference of vibration；Institute
It states sound interference unit to be communicated with the middle control unit always, obtains the sound status information of current indoor, including
Sound pressure level strength and target voice variation, to carry out Adaptive matching；
Middle control unit is respectively connected with the sound interference unit, for providing human-machine interface with the sound monitoring unit
Mouthful, and control the sound monitoring unit and sound interference unit collaborative work；The middle control unit is used for institute
Predicate sound signal is analyzed, and the temporal signatures and frequency domain character of the voice signal are extracted；Based on the temporal signatures and frequency
Characteristic of field carries out reversion and scramble to the voice signal, generates sound masking signal；And the sound masking signal is sent to sound
Sound interferes unit.
2. system according to claim 1, which is characterized in that the sound monitoring unit is specifically used for:
Sound signal collecting, weighted filtering, octave analysis, sound pressure level measurement and speech detection.
3. system according to claim 1, which is characterized in that the sound interference unit includes: spatial sound interference mould
Block, pipe vibration interference module, glass vibration interference module and wall vibration interference module；
Spatial sound interference module is installed on the outside of room enclosure structure, and for generating voice masking signal, interference air-borne sound is let out
Leakage；
Pipe vibration interference module is installed on the pipeline across room enclosure structure, fresh air and air conditioning, for generating vibration
Dynamic interference signal, prevent because pipe vibration it is transaudient caused by indoor sensitive acoustic intelligence leakage；
Glass vibration interference module is installed on the window-glass inside room, for generating vibration interference signal, is prevented because swashing
Indoor sensitive acoustic intelligence leakage caused by light eavesdropping；
Wall vibration interference module is installed on the wall at setting position, for generating vibration interference signal, prevents from stealing because of vibration
The sensitivity acoustic intelligence leakage of interior caused by device for tone frequencies.
4. system according to claim 1, which is characterized in that the system also includes network units, described for connecting
Sound interference unit and the middle control unit, so that the sound interference unit and the middle control unit pass through network
Agreement is communicated.
5. a kind of sound leakage protection method based on the principle of acoustic masking, which is characterized in that the described method includes:
Voice signal is acquired using sound monitoring unit, and obtains A weighted sound pressure level；
Judge whether the voice signal is voice signal, if the voice signal is voice signal and the A in a period of time
Weighted sound pressure level is more than preset threshold value, then issues alarm signal；
Middle control unit carries out analysis and feature extraction to the voice signal, generates interference signal, and the interference is believed
Number it is sent to sound interference unit；
The middle control unit carries out analysis and feature extraction to the voice signal, generates interference signal, and will be described dry
Disturb signal and be sent to sound interference unit and specifically include: the voice signal is analyzed, extract the voice signal when
Characteristic of field and frequency domain character；Based on the temporal signatures and frequency domain character, reversion and scramble are carried out to the voice signal, generated
Sound masking signal；The sound masking signal is sent to the spatial sound interference module in sound interference unit, and controls pipeline
Vibration interference module, glass vibration interference module and wall vibration interference module are started to work；
Sound interference unit carries out sound interference to each region using the interference signal；The sound interference unit always with institute
It states middle control unit to be communicated, obtains the sound status information of current indoor, including sound pressure level strength and the target language change of tune
Change, to carry out Adaptive matching.
6. according to the method described in claim 5, it is characterized in that, the method also includes:
Middle control unit regularly sends heartbeat signal to sound interference unit, if not receiving the heartbeat letter in the predetermined time
Number response signal, then determine sound interference unit broken string, and issue alarm signal.
7. according to the method described in claim 5, it is characterized in that, described adopt voice signal using sound monitoring unit
Collection, and obtain A weighted sound pressure level, comprising:
Data acquisition is carried out to voice signal, obtains instantaneous sound pressure level intensity value；
Octave analysis is carried out to the voice signal, obtains the sound pressure level strength in different frequency bands；
Weighted filtering is carried out to the voice signal, obtains A weighted sound pressure level value.
8. according to the method described in claim 5, it is characterized in that, described judge whether the voice signal is voice signal,
Include:
The voice signal is divided into several frames；
The short-time energy of every frame voice signal and the short-time zero-crossing rate of every frame voice signal is calculated；
Using double threshold method, judge whether the voice signal is voice signal.
9. according to the method described in claim 5, it is characterized in that, the sound interference unit is using the interference signal to each
Region carries out sound interference, comprising:
Spatial sound interference module carries out sound interference using the sound masking signal received；
Pipe vibration interference module, glass vibration interference module and wall vibration interference module, according to the control of middle control unit
System carries out vibration interference.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410822950.2A CN104575486B (en) | 2014-12-25 | 2014-12-25 | Sound leakage protection method and system based on the principle of acoustic masking |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410822950.2A CN104575486B (en) | 2014-12-25 | 2014-12-25 | Sound leakage protection method and system based on the principle of acoustic masking |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104575486A CN104575486A (en) | 2015-04-29 |
| CN104575486B true CN104575486B (en) | 2019-04-02 |
Family
ID=53091390
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410822950.2A Active CN104575486B (en) | 2014-12-25 | 2014-12-25 | Sound leakage protection method and system based on the principle of acoustic masking |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104575486B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI839796B (en) | 2022-08-01 | 2024-04-21 | 逸陞有限公司 | Sound monitoring system |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104992710A (en) * | 2015-06-18 | 2015-10-21 | 惠州Tcl移动通信有限公司 | Method for finding article based on intelligent device and intelligent device |
| CN105185370B (en) * | 2015-08-10 | 2019-02-12 | 电子科技大学 | A kind of sound masking door |
| CN106558303A (en) * | 2015-09-29 | 2017-04-05 | 苏州天声学科技有限公司 | Array sound mask device and sound mask method |
| CN106601230B (en) * | 2016-12-19 | 2020-06-02 | 苏州金峰物联网技术有限公司 | Logistics sorting place name voice recognition method and system based on continuous Gaussian mixture HMM model and logistics sorting system |
| US11205439B2 (en) | 2019-11-22 | 2021-12-21 | International Business Machines Corporation | Regulating speech sound dissemination |
| CN111614829B (en) * | 2020-05-21 | 2021-07-23 | 维沃移动通信有限公司 | Electronic equipment and anti-eavesdropping method |
| CN116320123B (en) * | 2022-08-11 | 2024-03-08 | 荣耀终端有限公司 | Voice signal output method and electronic equipment |
| CN116524891A (en) * | 2023-05-08 | 2023-08-01 | 拓达世纪信息产业有限公司 | AI sound masking system and secret meeting room thereof |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102110441A (en) * | 2010-12-22 | 2011-06-29 | 中国科学院声学研究所 | Method for generating sound masking signal based on time reversal |
| CN102543066A (en) * | 2011-11-18 | 2012-07-04 | 中国科学院声学研究所 | Target voice privacy protection method and system |
| CN202383998U (en) * | 2011-12-22 | 2012-08-15 | 国家保密科学技术研究所 | Sound masking system |
| CN203273230U (en) * | 2013-04-28 | 2013-11-06 | 北京天大清源通信科技股份有限公司 | Fixing clamp and audio interference device with same |
| CN104142635A (en) * | 2013-05-10 | 2014-11-12 | 北京天大清源通信科技股份有限公司 | Sound masking system with monitoring center |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8477958B2 (en) * | 2001-02-26 | 2013-07-02 | 777388 Ontario Limited | Networked sound masking system |
| JP5732937B2 (en) * | 2010-09-08 | 2015-06-10 | ヤマハ株式会社 | Sound masking equipment |
| JP5644359B2 (en) * | 2010-10-21 | 2014-12-24 | ヤマハ株式会社 | Audio processing device |
| JP5761259B2 (en) * | 2013-06-24 | 2015-08-12 | ヤマハ株式会社 | Conversation leakage prevention device |
| CN103886858B (en) * | 2014-03-11 | 2016-10-05 | 中国科学院信息工程研究所 | A kind of sound masking signal generating method and system |
| RU2556272C2 (en) * | 2014-04-03 | 2015-07-10 | Закрытое акционерное общество "Защита электронных технологий" | Vibratory noise unit for acoustic voice data leak channel |
-
2014
- 2014-12-25 CN CN201410822950.2A patent/CN104575486B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102110441A (en) * | 2010-12-22 | 2011-06-29 | 中国科学院声学研究所 | Method for generating sound masking signal based on time reversal |
| CN102543066A (en) * | 2011-11-18 | 2012-07-04 | 中国科学院声学研究所 | Target voice privacy protection method and system |
| CN202383998U (en) * | 2011-12-22 | 2012-08-15 | 国家保密科学技术研究所 | Sound masking system |
| CN203273230U (en) * | 2013-04-28 | 2013-11-06 | 北京天大清源通信科技股份有限公司 | Fixing clamp and audio interference device with same |
| CN104142635A (en) * | 2013-05-10 | 2014-11-12 | 北京天大清源通信科技股份有限公司 | Sound masking system with monitoring center |
Non-Patent Citations (2)
| Title |
|---|
| 《时间反转语音掩蔽的语音信号可懂度的客观评价方法》;王玥等;《网络新媒体技术》;20120531;第1卷(第2期);第54-59页 * |
| 《语音的掩蔽声实验研究》;蒋斌等;《声学技术》;20111231;第30卷(第6期);第275-276页 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI839796B (en) | 2022-08-01 | 2024-04-21 | 逸陞有限公司 | Sound monitoring system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104575486A (en) | 2015-04-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104575486B (en) | Sound leakage protection method and system based on the principle of acoustic masking | |
| US11120821B2 (en) | Vowel sensing voice activity detector | |
| CN107371085B (en) | Safety protection method and device and intelligent sound box | |
| Assmann et al. | The perception of speech under adverse conditions | |
| Sachs et al. | Rate versus level functions for auditory‐nerve fibers in cats: tone‐burst stimuli | |
| CN102884575A (en) | Voice activity detection | |
| CN105096946B (en) | Awakening device and method based on voice activation detection | |
| CN103169505A (en) | Method and device for Doppler ultrasound pickup analysis processing | |
| JP2010537253A (en) | System and method for noise activity detection | |
| CN113170268A (en) | Method and device for detecting probability silent fault | |
| Zhang et al. | Estimate the noise effect on automatic speech recognition accuracy for mandarin by an approach associating articulation index | |
| Garg | Environmental NoiseControl: A Review | |
| CN103519785A (en) | Multi-dimensional voice disorder measuring system based on multi-dimensional real-time voice modeling and method thereof | |
| Marchi et al. | Audio onset detection: A wavelet packet based approach with recurrent neural networks | |
| CN116524891A (en) | AI sound masking system and secret meeting room thereof | |
| Wu et al. | A pitch-based method for the estimation of short reverberation time | |
| Large et al. | The noise characteristics of'compliant'wind farms that adversely affect its neighbours | |
| Kwon et al. | A sensor network for real-time acoustic scene analysis | |
| Dai et al. | An improved model of masking effects for robust speech recognition system | |
| Sugimoto et al. | Measurement of infrasound generated by wind turbine generator | |
| Zhang | An objective evaluation method and experiment on the impact of road traffic noise on the soundscape: The case of West Lake | |
| CN104142635A (en) | Sound masking system with monitoring center | |
| Cao et al. | Acoustics recognition of excavation equipment based on MF-PLPCC features and RELM | |
| Pierzga et al. | Analysis of low-frequency acoustic signals emitted by low-power vertical axis wind turbine VAWT | |
| Fang et al. | Research on Signal Extraction and Classification for Ship Sound Signal Recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |