CN102037664A - Method and device for reducing audio frequency interference - Google Patents
Method and device for reducing audio frequency interference Download PDFInfo
- Publication number
- CN102037664A CN102037664A CN2008801293036A CN200880129303A CN102037664A CN 102037664 A CN102037664 A CN 102037664A CN 2008801293036 A CN2008801293036 A CN 2008801293036A CN 200880129303 A CN200880129303 A CN 200880129303A CN 102037664 A CN102037664 A CN 102037664A
- Authority
- CN
- China
- Prior art keywords
- pulse
- audio
- noise
- noise pulse
- model
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000012545 processing Methods 0.000 claims abstract description 29
- 238000005516 engineering process Methods 0.000 claims abstract description 15
- 230000001629 suppression Effects 0.000 claims abstract description 3
- 230000001413 cellular effect Effects 0.000 claims description 35
- 230000009467 reduction Effects 0.000 claims description 30
- 230000003595 spectral effect Effects 0.000 claims description 4
- 238000011084 recovery Methods 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 2
- 238000004891 communication Methods 0.000 description 10
- 238000001228 spectrum Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 8
- 238000003860 storage Methods 0.000 description 5
- 230000014759 maintenance of location Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 230000005236 sound signal Effects 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 238000001914 filtration Methods 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005670 electromagnetic radiation Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B1/00—Details of transmission systems, not covered by a single one of groups H04B3/00 - H04B13/00; Details of transmission systems not characterised by the medium used for transmission
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03G—CONTROL OF AMPLIFICATION
- H03G3/00—Gain control in amplifiers or frequency changers
- H03G3/20—Automatic control
- H03G3/30—Automatic control in amplifiers having semiconductor devices
- H03G3/34—Muting amplifier when no signal is present or when only weak signals are present, or caused by the presence of noise signals, e.g. squelch systems
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Noise Elimination (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
A method for reducing audio frequency interference is disclosed. The audio frequency is interfered by a noise pulse, which is a repeatability similarity pulse and includes a center pulse and an extending pulse. The method includes: generating a templet pulse to represent the noise pulse; and implementing the following steps (A), (B), (C) or the combination of them to suppress the noise pulse: (A) subtracting the templet pulse from the audio frequency to generate a first processing audio frequency; (B) at least spanning the center pulse of the noise pulse so as to interpolate the first processing audio frequency or the audio frequency to generate a second processing audio frequency; and/or (C) making a noise suppression technology process the second processing audio frequency or the audio frequency.
Description
Reduce the method and its device of audio disturbances
Technical field
The present invention is to reduce by thousand method and devices disturbed on a kind of, and the method and device that audio thousand is disturbed is reduced especially with regard to a kind of.Background technology
With the fast development of communication science and technology, various mobile phones (mobile phone) or cellular telephone (cell phone) are popularized already, communication protocol is wirelessly transferred used in its all kinds of transport communication protocol used, and substantial amounts of large amplitude electromagnetic pulse is produced during communication, at registration networking, receive, during transmitting procedure and handover step, it is controlled or transmits data-signal, once the voice signal in these mobile phones is transmitted, tonepulse circuit will be received on audio-frequency amplifier and line, so as to cause the sound equipment distortion that perceives of staccato form to produce, commonly referred to as it is global system for mobile communications abnormal sound (GSM buzz), and with the frequency near about 200 hertz for global system for mobile communications signal.
Many audio frequency apparatuses such as Car Stereo device, phone, recorder, portable audio player and hearing assistor etc. can all be disturbed by the tonepulse of these mobile phones or cellular telephone just in the communications, when being continued to increase using the number of users of cellular telephone, this problem will become readily apparent from, for top consumption terminal audio product and hearing assistor, the definition of sound equipment is even more the emphasis of product quality.
Although reducing cellular telephone transimission power, this problem can be mitigated by changing the transport communication protocol and covering the voicefrequency circuit, and these settling modes can not be realized in the case where not changing existing hardware design, and generally extremely difficult and costliness.In addition, for communication science and technology now, designing a noiseless voicefrequency circuit, the noise jamming caused by transmitting electromagnetic pulse using various communications protocols can be altogether dispensed with and be nearly impossible.Mobile phone or cellular telephone disturbed almost to the thousand of these audios nowhere-do not exist, interference is occurred often among any phase of audio transmission path, when the audio volume control of these audio frequency apparatuses is disturbed, and it is the settling mode for filtering out the distortion that the noise does not result in the original audio again to be not present.
Certainly, a kind of simpler settling mode is that mobile phone should not be used around these audio frequency apparatuses
Or cellular telephone, but this and do not conform to convention.For example, if use cellular telephone in the car, it will sufficiently close to the automobile audio apparatus and cause interference, and if when automotive interior is using cellular telephone and movement simultaneously, it disturbs the automobile audio apparatus by inevitable thousand.In addition, such as bright one people for having dysacousis, also inevasible that phone can be caused very close to his hearing assistor when it is communicated using mobile phone or cellular telephone.
Therefore in communication science and technology now, need it is a kind of without disadvantages mentioned above but can effectively reduce the method and its device of mobile phone or cellular telephone to the impulse disturbances caused by audio frequency apparatus, applicant is in view of missing in known technology, by carefully experiment and research, visualize the present invention in line with the spirit work with perseverance reduces the method and its device of audio disturbances one by one, disadvantages mentioned above can be overcome, the cylinder for the present invention will illustrate below.The content of the invention
To overcome the defect described in the section of background technology one present in known technology, the present invention proposes a kind of method and its device for reducing noise pulse to the interference of audio frequency apparatus, these noise pulses are as produced by the interference sources such as mobile phone or cellular telephone, its concept is mainly first found out as the model pulse of the noise pulse caused by mobile phone or cellular telephone, the model pulse is reduced from being influenceed by the noise pulse in the audio of the audio frequency apparatus containing the noise pulse waveform again, then treated audio is given with interpolation method ~ just, such as autoregression interpolation (auto-regressive interpolation), then situation is regarded again, one noise reduction techniques are bestowed to revised audio, such as spectral subtraction removes technology noise reduction techniques, above step is when implementing, short burst (short burst) or long burst (long burst) will be categorized as depending on the duration of the noise pulse and there are different possible combinations.
According to the design of the present invention, propose a kind of method for reducing audio disturbances, the audio is by one or a succession of similar noise pulse thousand is disturbed, and the noise pulse includes a center pulse and an extension pulse, it is characterised in that：A model pulse is produced to represent the noise pulse；And the following step (A) of implementation,（B), (C) or its combination are to suppress the noise pulse：（A) position of the noise pulse reduces the model pulse or the center pulse in the model pulse and produces one first processing audio from the audio；（B) at least across the noise pulse the center pulse and the interpolation first processing audio or the audio and two second processing audios for producing；Or (C) handles the second processing audio or the audio with a noise reduction techniques.
It is preferred that the method that this kind provided by the present invention reduces audio disturbances, it is characterised in that by the sound
Frequency division is segmented into multiple interval such as frames, and in each frame the step of the implementation suppression noise pulse.
It is preferred that the method that this kind reduction audio thousand provided by the present invention is disturbed, it is characterised in that one or more consecutive noise pulse is categorized as into long burst or short burst by its duration.
It is preferred that the method that this kind provided by the present invention reduces audio disturbances, it is characterised in that：If the noise pulse is the short burst, implementation steps (A),（B) or its combination.
It is preferred that the method that this kind provided by the present invention reduces audio disturbances, it is characterised in that：If the noise pulse is length burst, implementation steps (A),（B)、 （C) or its combination.
It is preferred that the method that this kind provided by the present invention reduces audio disturbances, wherein the generation step also include step：Confirm an amplitude of a position and the noise pulse of the noise pulse in the audio, thereby according to scaling of the magnitude determinations to the model pulse, the model pulse or the center pulse of the model pulse are then subtracted on the relevant position of the audio.
It is preferred that the method that this kind provided by the present invention reduces audio disturbances, it is characterised in that the generation step also includes step：Thus multiple features that estimating the noise pulse has produce the model pulse.
It is preferred that the method that this kind provided by the present invention reduces audio disturbances, it is characterised in that the interpolation is to utilize an autoregression pattern, a currently commercially obtainable recovery technology or its combination.
It is preferred that the method that this kind provided by the present invention reduces audio disturbances, it is characterised in that the noise reduction techniques are that a spectral subtraction removes technology, a currently commercially obtainable noise reduction techniques or its combination.
It is preferred that the method that this kind reduction audio thousand provided by the present invention is disturbed, it is characterised in that the noise pulse is as produced by a mobile phone or a cellular telephone.
Preferably, the method that this kind reduction audio thousand provided by the present invention is disturbed, it is characterized in that the audio is the audio that an audio frequency apparatus is sent, the audio frequency apparatus is a microphone, a sound equipment, a phone, a recorder, a portable audio player or a hearing assistor.
A kind of device for implementing the method for this kind reduction audio disturbances as the aforementioned.
According to the design of the present invention, a kind of device of reduction audio disturbances is proposed, it is characterised in that including：One input, to receive an audio；One memory, wherein plate pulse as to produce one noise pulse of ^/or storage, the model pulse represent one of a succession of similar noise pulse, and the noise pulse includes a center pulse and an extension pulse；And a noise suppressor, it is connected with the input, to implement the steps of (A；)、 （B)、 （) or its combination is to suppress the noise pulse C：（A) subduction should from the audio
Model pulse and produce one first processing audio；（B) at least across the noise pulse the center pulse and interpolation this first processing audio or the audio and produce a second processing audio；^/or (C) handle the second processing audio or the audio with a noise reduction techniques.
It is preferred that this kind provided by the present invention reduces the device of audio disturbances, it is characterised in that also including a detector, the appearance to confirm a noise pulse.
It is preferred that the device that this kind reduction audio thousand provided by the present invention is disturbed, it is characterised in that the noise suppressor also includes an amplitude determination device, the amplitude to determine the pulse.
Preferably, this kind provided by the present invention reduces the device of audio disturbances, it is characterized in that the noise suppressor also includes a position pulse position detector to detect the noise pulse, and the pulse position detector also includes an intercorrelation detector.
Brief summary, the present invention is minimized the influence of the noise pulse by way of the noise pulse to be decomposed into a center pulse and an extension pulse, wherein the center pulse is the electromagnetic radiation product for coming from the cellular telephone or other interference sources, the extension pulse then comes from the audio frequency apparatus, particularly the feature of electric capacity (capacitance) characteristic of the audio frequency apparatus.But by the center pulse stereotyping and when subtracting the stereotyping center pulse, the voicefrequency circuit characteristic of the audio frequency apparatus is not considered.Summary of drawings
By way of the embodiment of the present invention is as example, and with reference to it with accompanying drawing, of the invention will be apparent from other features and advantage by described below above, wherein：
Fig. 1 is schematic diagram of the typical case as the noise pulse produced by the interference sources such as mobile phone or cellular telephone；
Fig. 1 is the device to suppress a reduction audio disturbances of noise pulse from audio；Fig. 3 is a kind of implementing procedure figure of the method for reduction audio disturbances proposed by the invention；Fig. 4 is the noise pulse schematic diagram for being categorized as short burst；
Fig. 5 is the noise pulse schematic diagram for being categorized as long burst；And
Fig. 6 is the spectrum diagram of noise pulse.
The better embodiment of the present invention
Referring to Fig. 1, Fig. 1 is a typical schematic diagram as the noise pulse produced by the interference sources such as mobile phone or cellular telephone, the noise pulse τ shown in Fig. 1 includes two major parts, a center pulseτιAnd one extension pulse tau2.Producing the situation of noise pulse shown in Fig. 1 is, when an audio frequency apparatus, such as：When microphone, sound equipment, phone, recorder, portable audio player or hearing assistor are operated, when namely receiving audio or playing audio, just there are mobile phone or cellular telephone also being communicated or information transfer around these audio frequency apparatuses, it is now rule or a smooth normal audio originally in audio frequency apparatus, by because by the impulse disturbances produced by mobile phone or cellular telephone, and start to be mingled with the noise pulse as shown in Fig. 1 in the normal audio of script, including center pulseτιAnd extension pulse tau2。
Noise pulse τ as shown in Fig. 1, its center pulseτιIt is to be communicated or produced electromagnetic wave during information transfer is excited by mobile phone or cellular telephone, and extends pulse tau2It is then relevant with the capacitance characteristic of audio frequency apparatus.Therefore, noise pulse τ characteristic will be relevant with the characteristic of the voicefrequency circuit of the electromagnetic wave activity and audio frequency apparatus of the cellular telephone, the length for the tonepulse that mobile phone or cellular telephone are transmitted is different with the specification of phone, in general it is equal to 25 samples for the sample frequency of 44 KHzs, the amplitude of pulse is especially directly proportional to the distance between the cellular telephone and the audio frequency apparatus.
Please continue to refer to Fig. 2, for for the device for the reduction audio disturbances for suppressing noise pulse from audio.Such as the device 20 of reduction audio disturbances shown in fig. 2, one input 10 of the device 20 of this reduction audio disturbances is connected with a voicefrequency circuit 1, this voicefrequency circuit 1 is the voicefrequency circuit being arranged in various audio frequency apparatuses, this audio frequency apparatus or voicefrequency circuit 1 are by surrounding as the interference caused by the electromagnetic pulse of mobile phone or cellular telephone, if the audio is analog form, the output of the voicefrequency circuit 1 is provided to an analog-digital converter 2, if the audio has been digital form, for example come from a CD, then the analog-digital converter 2 will be omitted, and the output of the voicefrequency circuit 1 will be provided directly to a first in first out (FIFO) Slow storages 3;The output of the analog-digital converter 2 is provided to the first in first out (FIFO) Slow storages 3, and the output of the first in first out (FIFO) Slow storages 3 is then connected to a digital signal processor (Digital Signal Processor) 4;One hardware electromagnetic pulse detector 5 has the output for being connected to the digital signal processor 4.The analog-digital converter 2 also changes an output to a memory 6, and the memory 6 can also be accessed by the digital signal processor 4.The one of the digital signal processor 4
Output is also supplied with a digital analog converter 7, and the digital analog converter 7 is connected to an output 11, and the output 11 will be combined with some such as loudspeakers or earphone transducer form.In fact, also can be by the combination of analog-digital converter 2 disclosed in fig. 2, first in first out (FIFO) Slow storages 3, digital signal processor 4 and digital analog converter 7 etc., it is considered as a wave filter 22 or a noise suppressor 24, its removable noise pulse being present in input 10, and the audio that the audio of output 11 is disturbed for noiseless pulse.Wherein digital signal processor 4 doubles as amplitude of the amplitude determination device to determine the pulse, and the position as pulse position detector to detect the noise pulse simultaneously.
It is a kind of implementing procedure figure of reduction audio disturbances method proposed by the invention please continue to refer to Fig. 3.The method that this disclosed kind reduction audio thousand is disturbed in figure 3, it includes：Step 30, it is multiple frames by audio segmentation；Step 32, a model pulse is produced to represent the noise pulse；Step 34, the noise pulse is categorized as a short burst or a length happens suddenly；Step 36, if the noise pulse is length burst, implementation steps (A),（B)、 （C) or its combination；Step 38, if the noise pulse is the short burst, implementation steps (A),（B) or its combination；Step (A) produces one first processing audio to reduce the model pulse from the audio；Step (B) at least across the noise pulse the center pulse and interpolation this first processing audio or the audio and produce a second processing audio；Step (C) handles the second processing audio or the audio with a noise reduction techniques.Wherein, the step of being represented in Fig. 3 with dotted line (dashed line), if step 30,34 and 38 are washability step, user when implementing a kind of method of reduction audio disturbances proposed by the invention, can with the implementation of unrestricted choice step 30,34 and 38 whether.
When implementing above method, first an audio is obtained from audio frequency apparatus just in operation, this audio frequency apparatus is because by mobile phone around just in use or in communication or the impulse disturbances of cellular telephone, so that containing noise pulse in audio, this noise pulse is divided into two parts, respectively center pulse and extension pulse, wherein center pulse is being communicated or produced electromagnetic impulse during information transfer is excited by mobile phone or cellular telephone, and it is then relevant with the capacitance characteristic of audio frequency apparatus to extend pulse.
Then decide whether to implement above method by the way of based on frame (frame-based), if intending by the way of based on frame, audio signal first must be divided into several frames, each frame is about 40/wsec length.If being not intended to by the way of based on frame implement above method, the process described above can be neglected.Described above is the step 30 shown in Fig. 3.Illustrated below with decision by the way of based on frame (frame-based) to implement the above method.
A model pulse being enough instead of this noise pulse is produced followed by following mode, is
Build and represent plate pulse as the typical pulse as produced by mobile phone or cellular telephone, device shown in Fig. 2 is set with a zero value audio, it is a cellular telephone pulse now only to pass through the voicefrequency circuit 1 and provide to the signal of the digital signal processor 4, its expression can disturb those noise pulses of normal audio in normal operating, the features such as amplitude, the cycle of the pulse are calculated, a model pulse is thus produced and is stored in memory 6.When the device is in normal use, according to scaling of the magnitude determinations to model pulse, just after scale is zoomed in or out it can be made to be consistent with noise pulse model pulse, the center pulse of noise pulse and the length of extension pulse are constant for the audio devices of specification known to the cellular telephone and one of specification known to one, but noise pulse intensity (amplitude) by the distance dependent with mobile phone or cellular telephone and the audio frequency apparatus.The digital signal processor 4 is then programmed to the method described using the model pulse progress stored in the memory 6 at this, and produces the model pulse as described simultaneously, and is updated when needing.
Certainly also have other modes to set up a model pulse, such as it is direct using the detector of an outside to estimate model pulse that model pulse, usage history experience are set up, using previous model pulse, using exemplarily pulse, the noise pulse that is obtained using the former frame exemplarily mode such as pulse of the noise pulse measured by when the audio in audio frequency apparatus is by blank.But particularly note that, when setting up model pulse, it it is not absolutely required to after the step of receiving an audio, if using such as model pulse that usage history experience is set up, using previous model pulse, using the noise pulse measured by when the audio in audio frequency apparatus is blank, exemplarily the mode such as pulse obtains model pulse, then i.e. executable before an audio step is received.Described above is the step 32 shown in Fig. 3.
Noise pulse is searched in frame and recognizes the starting point of each noise pulse.If it was found that more than one noise pulse, determine the frequency of occurrences of the noise pulse in frame and the noise pulse in this frame is categorized as long burst (long burst) accordingly or short (short burst) is happened suddenly, generally the noise pulse for being continued above 30 sec is categorized as to grow burst, the noise pulse that 30/ sec is not continued above is then categorized as short burst.Fig. 4 and Fig. 5 are referred to, the noise pulse schematic diagram of short burst is respectively categorized as and is categorized as the noise pulse schematic diagram of long burst.
If the noise pulse in this frame is categorized as short burst, implementation steps (A),（) or step (A) and the combination step of (B) (step (A) and (B) will in be described in detail hereinafter) B, i.e., implementation steps (A), (B) successively, implementation steps (B), (A) successively, only implementation steps (A), the only mode such as implementation steps (B).Usual only implementation steps (A) or one of step of (B) can effectively suppress to be categorized as the noise pulse of short burst, particularly work as
The extension pulsion phase of noise pulse is when obvious, you can the implementation of omited steps (B).
If the noise pulse in this frame is categorized as growing burst, implementation steps (A),（B)、（C) or step (A), (B) and (C) combination step (step (A),（B) and (C) will in hereinafter be described in detail), i.e. may be selected only apply implementation steps (A),（B) or (C) one of step or two steps therein, and implement when, the precedence of step can be replaced according to actual state.Another common situation is：If the noise pulse in this frame has been categorized as growing burst, it is finished the step of based on short burst after, it is proposed that should implementation steps (C), or leading implementation steps (C) again, then implement be based on short burst the step of.
Present invention use three below step (A),（B)、 （C) or (A),（) and the combination of (C) suppresses the noise pulse carried in audio B.
Illustrate step (A) first, the position of each noise pulse reduces the model pulse and produces one first mode for handling audio from the audio.Be in step (A) using equation as described below remove included in audio as the noise pulse caused by cellular telephone：
xno,se(") = b{n)ypulse{n) + xoriginal{n) + en (1)
Wherein：
xnoiseTo contain the audio signal as the noise pulse caused by the cellular telephone；
For the known model pulse of the noise pulse of the cellular telephone；
E is a white noise (white noise) defeated people；
^ ■ original are the original not disturbed audio.
If it is known that the cellular telephone disturbs the exact position in audio, it is possible to if determining scale factor b, then can be defined by scale factor b, to model pulse《Carry out scaling Huo Shrink small, the center pulse of model pulse or plate pulse is then subtracted on the relevant position of audio, when model pulse ypulse is by the audio signal X containing noise pulse.After middle subduction, you can preferably to restore primary signal x0 al.This primary signal； C.„.g,.„fl/Also the first processing audio can be temporarily regarded as.It is described above be Fig. 3 shown in the step of (A；).
When carrying out step (A), in need it can know noise pulse corresponding exact position in audio, then the center pulse of model pulse or model pulse after proportional zoom could be subtracted on this position, generally there are many method for detecting to can be used to determine the exact position of the noise pulse, such as：1) hardware electromagnetic wave detector, 2) signal amplitude the detecting of maximum threshold, 3) detecting of the maximum threshold of signal slope, 4)
Intercorrelation/be consistent filtering detector, 5) bayesian ladder detector, and the 6) device of autoregression detector etc..All detectors are not included above, but for application in real time, hardware electromagnetic wave detector is probably most convenient and maximally effective settling mode.The software detected using the maximum threshold detecting ^ of signal amplitude or the maximum threshold of signal slope realizes to be also simple, efficient and effectively bright settling mode.
Then illustrate step (B), at least across the noise pulse the center pulse and the interpolation first processing audio or the audio and produce a second processing audio.For noise pulse, generally by mobile phone or originally),<Therefore ^ can utilize the individual simple interpolative modes of ^ " to assume at this, the information between 25 ^ samples is loss, and carries out interpolation.
" all there is a fixed value b; and have confirmed that the position of the noise pulse; then now using such as single channel autoregression pattern (single channel auto-regressive (AR) model); with the position of a low order time (p=20 on typical case) recovery/interpolation center pulse, with the value of interpolated value substitution center pulse assuming that the scaling function b () in equation (1) can be reduced to for all：
(")=Σ i oise (" -0+(2) are wherein by ^noise：
.For the observation noise signal disturbed comprising the cellular telephone：
E is a white noise vocal input.
For the interpolation of single channel least square autoregression (LSAR), the answer of equation (2) is：
For the second processing audio tried to achieve by interpolation.
Actually, interpolation employed in step (B) is not limited only to AR patterns, other various business are available for recovering the interpositioning of audio, can be applied in step (B), simulate because of the audio that the generation of center pulse is lost.
Then illustrate step (C), the second processing audio or the audio are handled with a noise reduction techniques.Pointed noise reduction techniques are current and following commercially available noise reduction techniques in step (C), such as：The group of frequency blind subduction (spectrum subtraction) technology, power Pass subduction (power spectrum subtraction) technologies and other current and following commercially available noise reduction techniques
Close.Refer to Fig. 6, for the spectrum diagram of noise pulse, the frequency spectrum or power spectrum that can be set up in the following manner in each frame, as directly set up using the detector of an outside using estimated spectral, usage history experience frequency spectrum, using previous frequency spectrum, won using the frequency measured by when the audio in audio frequency apparatus is by blank exemplarily consult frequently, the mode such as the frequency spectrum that is obtained using former frame.
Aforesaid equation (1) and (2) also apply to the audio of multichannel, to suppress the noise pulse that multi-channel audio is carried.
Claims (1)
- Claims1st, a kind of to reduce the method disturbed of audio thousand, the audio is disturbed by one or a succession of similar noise pulse, and the noise pulse includes a center pulse and an extension pulse, it is characterised in that：A model pulse is produced to represent the noise pulse；AndThe following step (A) of implementation,（B)、 （) or its combination is to suppress the noise pulse C：(A) position of the noise pulse reduces the model pulse or the center pulse in the model pulse and produces one first processing audio from the audio；(B) at least across the noise pulse the center pulse and interpolation this first processing audio or the audio and produce a second processing audio；SJ or(C) the second processing audio or the audio are handled with a noise reduction techniques.2nd, according to the method described in claim 1, it is characterised in that：It is multiple intervals by the audio segmentation, and in each interval the step of the implementation suppression noise pulse.3rd, method according to claim 2, it is characterised in that the interval is a frame.4th, according to the method described in claim 1, it is characterised in that：One or more consecutive noise pulse is categorized as long burst or short burst by its duration.5th, method according to claim 4, it is characterised in that：If the noise pulse is the short burst, implementation steps (A),（B) or its combination.6th, method according to claim 4, it is characterised in that：If the noise pulse is length burst, implementation steps (A),（B)、 （C) or its combination.7th, according to the method described in claim 1, it is characterised in that the generation step also includes step：Confirm an amplitude of a position and the noise pulse of the noise pulse in the audio, thereby according to scaling of the magnitude determinations to the model pulse, the model pulse or the center pulse of the model pulse are then subtracted on the relevant position of the audio. 8th, according to the method described in claim 1, it is characterised in that the generation step also includes step：Thus multiple features that estimating the noise pulse has produce the model pulse.9th, according to the method described in claim 1, it is characterised in that：The interpolation is to utilize an autoregression pattern, a currently commercially obtainable recovery technology or its combination, and to be a spectral subtraction remove technology, a currently commercially obtainable noise reduction techniques or its combination to the noise reduction techniques.10th, according to the method described in claim 1, it is characterised in that：The noise pulse is as produced by a mobile phone or a cellular telephone, and the audio is the audio that an audio frequency apparatus is sent, the audio frequency apparatus is a microphone, a sound equipment, a phone, a recorder, a portable audio player or a hearing assistor.11st, a kind of device for the method for implementing this kind reduction audio disturbances according to claim 1.12nd, it is a kind of to reduce the device of audio disturbances, it is characterised in that including：One input, to receive an audio；One memory, wherein plate pulse as to produce and/or store a noise pulse, the model pulse represent one of a succession of similar noise pulse, and the noise pulse includes a center pulse and an extension pulse；AndOne noise suppressor, is connected with the input, to implement the steps of (A),（B)、 （) or its combination is to suppress the noise pulse C：(A) the model pulse is reduced from the audio and one first processing audio is produced；(B) at least across the noise pulse the center pulse and interpolation this first processing audio or the audio and produce a second processing audio； H(C) the second processing audio or the audio are handled with a noise reduction techniques.13rd, device according to claim 12, it is characterised in that also including a detector, the appearance to confirm the noise pulse.14th, device according to claim 12, it is characterised in that the noise suppressor also includes One amplitude determination device, the amplitude to determine the pulse.15th, the device described in root Ju claim 12, it is characterised in that the noise suppressor, which also includes a pulse position detector, the position to detect the noise pulse, and the pulse position detector, also includes an intercorrelation detector.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| PCT/CN2008/000980 WO2009140792A1 (en) | 2008-05-21 | 2008-05-21 | Method and device for reducing audio frequency interference |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102037664A true CN102037664A (en) | 2011-04-27 |
Family
ID=41339715
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2008801293036A Pending CN102037664A (en) | 2008-05-21 | 2008-05-21 | Method and device for reducing audio frequency interference |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN102037664A (en) |
| WO (1) | WO2009140792A1 (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019071127A1 (en) * | 2017-10-05 | 2019-04-11 | iZotope, Inc. | Identifying and removing noise in an audio signal |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3444198B2 (en) * | 1997-09-16 | 2003-09-08 | 株式会社デンソー | Noise suppression device and speech recognition system using the device |
| JP2002533964A (en) * | 1998-12-18 | 2002-10-08 | テレフオンアクチーボラゲツト エル エム エリクソン（パブル） | Noise suppression in mobile communication systems. |
| DE10017646A1 (en) * | 2000-04-08 | 2001-10-11 | Alcatel Sa | Noise suppression in the time domain |
| CA2326948A1 (en) * | 2000-11-24 | 2002-05-24 | Tet Hin Yeap | Noise/interference suppression system |
-
2008
- 2008-05-21 WO PCT/CN2008/000980 patent/WO2009140792A1/en active Application Filing
- 2008-05-21 CN CN2008801293036A patent/CN102037664A/en active Pending
Also Published As
| Publication number | Publication date |
|---|---|
| WO2009140792A1 (en) | 2009-11-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3295681B1 (en) | Acoustic echo cancelling system and method | |
| CN106463106B (en) | Wind noise reduction for audio reception | |
| US8972251B2 (en) | Generating a masking signal on an electronic device | |
| WO2016056167A1 (en) | Echo cancellation device, echo cancellation program, and echo cancellation method | |
| US9343073B1 (en) | Robust noise suppression system in adverse echo conditions | |
| CN104243732A (en) | Use of vibration sensor in acoustic echo cancellation | |
| US10755728B1 (en) | Multichannel noise cancellation using frequency domain spectrum masking | |
| EP3662651B1 (en) | Mitigating impact of double talk for residual suppressors | |
| JP2007003702A (en) | Noise eliminator, communication terminal, and noise eliminating method | |
| CN105657110B (en) | Echo cancellation method and device for voice communication | |
| US20110228946A1 (en) | Comfort noise generation method and system | |
| US7656933B2 (en) | Method and device for the suppression of periodic interference signals | |
| CN108028983A (en) | Nonlinear acoustics echo cancellor based on transducer impedance | |
| CN101233561B (en) | Enhancement of speech intelligibility in a mobile communication device by controlling the operation of a vibrator of a vibrator in dependance of the background noise | |
| JP5153389B2 (en) | Acoustic signal processing device | |
| US8254590B2 (en) | System and method for intelligibility enhancement of audio information | |
| WO2019239977A1 (en) | Echo suppression device, echo suppression method, and echo suppression program | |
| JP2008309955A (en) | Noise suppresser | |
| EP2490218B1 (en) | Method for interference suppression | |
| EP2230664B1 (en) | Method and apparatus for attenuating noise in an input signal | |
| US20040242157A1 (en) | Device and method for supressing periodic interference signals | |
| CN102037664A (en) | Method and device for reducing audio frequency interference | |
| CN104660835A (en) | Apparatus And Method For Eliminating Echo For Hands Free System | |
| CN109994104B (en) | Self-adaptive call volume control method and device | |
| JP4527654B2 (en) | Voice communication device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110427 |
|
| WD01 | Invention patent application deemed withdrawn after publication |