CN115954012A - Periodic transient interference event detection method - Google Patents
Periodic transient interference event detection method Download PDFInfo
- Publication number
- CN115954012A CN115954012A CN202310198975.9A CN202310198975A CN115954012A CN 115954012 A CN115954012 A CN 115954012A CN 202310198975 A CN202310198975 A CN 202310198975A CN 115954012 A CN115954012 A CN 115954012A
- Authority
- CN
- China
- Prior art keywords
- transient
- transient interference
- interference
- noise
- frame
- 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.)
- Granted
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D30/00—Reducing energy consumption in communication networks
- Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks
Abstract
S1, estimating a non-transient signal power spectrum and a conditional speech existence probability according to a time-frequency domain signal received by a microphone: s2, estimating a transient noise power spectrum and an optimal gain spectrum; s3, judging whether the transient interference has periodicity according to the conditional existence probability of the transient interference: s31, calculating the average value of the optimal gain function in the designated frequency band according to the optimal gain frequency spectrum of the transient interference noise power spectrum, S32, judging whether the current frame contains transient interference or not according to the average value of the optimal gain function, and S33, judging whether a periodic transient interference event occurs or not. The invention utilizes the conditional probability of the transient interference obtained in the transient interference enhancing process to obtain the conditional probability of the frame transient interference, and carries out periodic transient interference event judgment on the average value of the conditional probability of the transient interference of the continuous frames, thereby improving the judgment accuracy and providing a technical basis for utilizing the transient interference to control the voice equipment.
Description
Technical Field
The invention belongs to the technical field of digital signal processing, and particularly relates to a periodic transient interference event detection method.
Background
In the field of speech signal processing, transient noise is generally regarded as an interference signal, and transient noise suppression is widely used to improve speech quality, i.e. speech intelligibility and intelligibility. However, transient noise may also be useful information in situations where a voice call is not required. For example: the range hood is started by periodic transient interference formed by ignition of the gas stove, people are prevented from forgetting to turn on the range hood, and kitchen oil smoke pollution is reduced. In addition, under the condition that the refrigerator door is forgotten to be closed or is not tightly closed, the refrigerator can send out prompt tones, the refrigerator door can be automatically closed by utilizing the periodic transient sound, and energy conservation and emission reduction are realized. Therefore, a method for detecting a periodic transient interference event is needed to determine the operating state of the device.
Disclosure of Invention
The invention discloses a method for detecting a periodic transient interference event, which is used for detecting the working state of voice equipment by utilizing transient noise.
The invention discloses a periodic transient interference event detection method, which is characterized by comprising the following steps:
s1, setting a microphone to receive signals
y(n)=x(n)+t(n)
Where n is the time scale, x (n) is the speech signal, and t (n) is the transient noise.
The time-frequency domain signal received by the microphone can be obtained by short-time Fourier transform
Y (k, l) = X (k, l) + T (k, l), where k is the frequency scale and l is the time domain frame index;
estimating a non-transient signal power spectrum lambda d (k, l) and conditional speech presence probability p (k, l);
s2, estimating the power spectrum lambda of the transient noise t (k, l) and optimal gain spectrum G (k, l)
The power spectrum of the transient noise is expressed as
λ t (k,l)=|G(k,l)*Y(k,l)| 2
Wherein G (k, l) represents the optimal gain spectrum of the power spectrum of the estimated transient interference noise, and the calculation formula is
In the formula G min Is a predetermined spectral gain, G, in the absence of transient noise H1 Is the gain in the presence of transient noise and is expressed as
Wherein e is a natural constant, ξ (k, l) is the prior signal-to-noise ratio of transient noise, α t represents a weighting factor, and α t belongs to [0,1], and γ (k, l) is the posterior signal-to-noise ratio, v (k, l) represents the relation parameters of the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and the calculation formulas are respectively:
where Y (k, l) is the time-frequency domain signal received by the microphone in step S1, λ d (k, l) is the non-transient signal power spectrum;
s3, judging whether the transient interference has periodicity according to the conditional existence probability of the transient interference:
s31, firstly, calculating the average value of the optimal gain function in the specified frequency band according to the optimal gain spectrum G (k, l) of the transient interference noise power spectrum, namely
Middle bin of the formula start To specify the starting point of the band, bin end Is the end of the band.
Calculating the average value of the optimal gain function of the continuous m frames taking the current frame as the terminal point
m is the set average optimal frame number;
s32. If
Greater than a predetermined transient interference probability threshold G Th Judging the current frame as transient interference, otherwise, considering the current frame not containing transient interference noise;
when the current frame contains the transient interference noise, carrying out periodic transient interference event detection, entering step S33, otherwise, repeating step S32;
s33, judging whether a periodic transient interference event occurs or not;
s331, constructing an event judgment circulating window for storing the difference T between the current transient interference frame number and the last transient interference frame number d The expression is
T d = T current - T last
In the formula T current For the current number of transient interference frames, T last The number of the last transient interference frames;
constructing an event judgment circulation window to circularly store all T between the current transient interference frame number and the last transient interference frame number d Frame, the window length of the cycle window is L, for each current frame in the cycle window, the difference value of the current L frame is calculated;
T d (l)= T current - T last
l=1,2…L
s332, setting the period time of the periodic transient interference event as T 0 Further converting the cycle time into the cycle frame number T
T=T 0 /（len/F s ）
In the formula, len is the length of each frame signal, and Fs is the signal sampling frequency;
setting and judging minimum threshold value T of periodic transient interference event min And a maximum threshold value T max And satisfy T min <T<T max ；T min ，T max Respectively representing a threshold minimum value and a threshold maximum value, wherein the unit is a frame number;
s333, judging the current difference value stored in the circulating window, if the difference value is more than or equal to the threshold minimum value and less than or equal to the threshold maximum value, carrying out period counting,
if T is min ≤T d (l) ≤T max Adding 1 to the Count value;
if the period count of the L frames in the circulation window is larger than the specified periodic transient interference event threshold T judge If so, it is determined that there is periodicityA transient interference event occurs.
Preferably, the step S331 further includes determining adjacent interference, setting an interference threshold δ T, when Td is greater than δ T, determining that there is no adjacent interference, and if it is determined that there is adjacent interference, not performing subsequent steps and repeating the step S331.
Preferably, the minimum threshold value Tmin = T-TR and the maximum threshold value Tmax = T + TR of the periodic transient interference event are determined, and TR is the set tolerance frame number.
Preferably, the periodic transient interference event detection method as claimed in claim 1, wherein the step S1 is to estimate the non-transient signal power spectrum by using MCRA algorithm
In the formula alpha d Is a first smoothing parameter, and α d ∈[0,1]，
Representing time-varying smoothing parameters, p (k, l) is conditional speech existence probability and is expressed as
p(k,l)=α p *p(k,l-1)+( 1-α p )*I(k,l)
In the formula alpha p Is a second smoothing parameter, and alpha p ∈[0,1]I (k, l) is an identification function, when I (k, l) =1, the k-th frequency point marking the l-th frame contains a voice signal, otherwise, the k-th frequency point does not contain the voice signal, wherein delta is an empirical threshold value, S r (k, l) is a power spectrum ratio expressed by
S r （k,l）= S（k,l）/ S min (k, l), where S (k, l) is the time-domain smoothed power of the power spectrum of the noisy speech, S min (k, l) isThe minimum power value is searched, and the expressions are sequentially
S（k,l）=α S *S（k,l-1）+(1-α S ) *| Y（k,l）| 2
S min （k,l）=min{ S min （k,l-1）, S（k,l）}
In the formula alpha S Is a third smoothing parameter, and alpha S ∈[0,1]。
The invention utilizes the conditional probability of the transient interference obtained in the transient interference enhancing process to obtain the conditional probability of the frame transient interference, and carries out periodic transient interference event judgment on the average value of the conditional probability of the transient interference of the continuous frames, thereby improving the judgment accuracy and providing a technical basis for utilizing the transient interference to control the voice equipment.
Drawings
FIG. 1 is a schematic flow chart diagram illustrating a method for detecting a periodic transient interference event according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating the periodic transient interference event determination according to an embodiment of the present invention;
FIG. 3 is a comparison graph of a microphone received signal and an enhanced transient noise spectrum in accordance with an embodiment of the present invention.
Detailed description of the preferred embodiments
The following provides a more detailed description of the present invention.
The following detailed description of the embodiments of the present invention is provided with reference to the accompanying drawings, but the present invention is not limited thereto.
The flow of one embodiment of the present invention is shown in fig. 1.
S1, estimating non-transient signal power spectrum lambda according to MCRA algorithm d (k, l) and conditional speech presence probability p (k, l)
Setting the microphone to receive signals as
y(n)=x(n)+t(n)
Where n is the time scale, x (n) is the speech signal, and t (n) is the transient noise.
The time-frequency domain signal received by the microphone can be obtained through short-time Fourier transform
Y(k,l)=X(k,l)+T(k,l)
Where k is the frequency scale and l is the time domain frame index.
The power spectrum of the non-transient signal estimated by the MCRA algorithm is
In the formula of alpha d Is a first smoothing parameter, and alpha d ∈[0,1]，
Representing time-varying smoothing parameters, p (k, l) is conditional speech existence probability and is expressed as
p(k,l)=α p *p(k,l-1)+( 1-α p )*I(k,l)
In the formula of alpha p Is a second smoothing parameter, and alpha p ∈[0,1]When I (k, l) =1, marking the k frequency point of the l frame to contain a voice signal, otherwise not containing, wherein delta is an empirical threshold value, S r (k, l) is a power spectrum ratio expressed by
S r （k,l）= S（k,l）/ S min (k, l), where S (k, l) is the time-domain smoothed power of the noisy speech power spectrum, S min (k, l) is the minimum power value searched, and the expressions are sequentially
S（k,l）=α S *S（k,l-1）+(1-α S ) *| Y（k,l）| 2
S min （k,l）=min{ S min （k,l-1）, S（k,l）}
Where α S is the third smoothing parameter and α S ∈ [0,1], the smaller α S, the faster the change in PSD to capture speech or background noise.
In the step, according to the characteristics that the transient interference of the voice and the background noise have different change rates, and the transient interference changes rapidly along with time compared with the slower voice and the pseudo static background noise, and the like, the power spectrum of the transient interference is estimated by adopting the MCRA algorithm, optimization can be performed by adjusting the flat factor of the MCRA algorithm, and the power spectrum density change of the voice or the background noise can be captured more rapidly.
S2, estimating the power spectrum lambda of the transient noise based on an OM-LSA algorithm as shown in figure 2 t (k, l) and optimal gain spectrum G (k, l)
The transient noise power spectrum is expressed as
λ t (k,l)=|G(k,l)*Y(k,l)| 2
Wherein G (k, l) represents the optimal gain spectrum of the power spectrum of the estimated transient interference noise, and the calculation formula is
In the formula G min Is the spectral gain in the absence of transient noise, G min Is a predetermined value, G H1 Is the gain in the presence of transient noise and is expressed as
Wherein e is a natural constant, ξ (k, l) is the prior signal-to-noise ratio of transient noise, α t represents a weighting factor, and α t belongs to [0,1], and γ (k, l) is the posterior signal-to-noise ratio, v (k, l) represents the relation parameters of the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and the calculation formulas are respectively:
where Y (k, l) is the time-frequency domain signal received by the microphone in step S1, λ d (k, l) is the non-transient signal power spectrum.
The method can enhance the transient interference by using an OM-LSA algorithm, and inhibit the speech signal and background noise of the non-transient interference.
S3, judging whether the transient interference has periodicity according to the conditional existence probability of the transient interference:
fig. 2 is a flowchart illustrating a specific example of the periodic glitch event determination.
S31, firstly, according to the optimal gain spectrum of the transient interference noise power spectrum, calculating the average value of the optimal gain function in the specified frequency band, namely
Middle bin of the formula start To specify the starting point of the band, bin end The designated frequency band is the frequency band at which the presence of the transient interference event needs to be determined in the present invention.
The average value of the optimal gain function is calculated for m consecutive frames ending at the current frame, where m is the set average optimal frame number, and for example, m =3 may be taken,
s32. If
Greater than a predetermined transient interference probability threshold G Th Judging the current frame as transient interference, otherwise, considering the current frame not containing transient interference noise;
when the current frame is determined to contain the transient interference noise, carrying out periodic transient interference event detection, entering the step S33, otherwise, repeating the step S32;
s33, when the current frame is determined to contain the transient interference noise, periodic transient interference event detection is carried out, and the specific steps are as follows:
s331, constructing an event judgment circulating window for storing the difference T between the current transient interference frame number and the last transient interference frame number d The expression is
T d = T current - T last
In the formula T current For the current number of transient interference frames, T last The number of the last transient interference frames;
the event judging circulating window circularly stores all T between the current transient interference frame number and the last transient interference frame number d Frame, the window length of the cyclic window is L, for every current frame in the cyclic window the difference value of current L frame is calculated
T d (l)= T current - T last
l=1,2…L
To avoid inter-neighbor interference, an interference threshold δ may be set T When T is d Greater than delta T If the interference between adjacent cells does not exist, the subsequent steps are continued, otherwise, if the interference between adjacent cells does exist, T is determined d (l) =0, do not perform subsequent steps and continue with this step S331;
s332, setting the period time of the periodic transient interference event as T 0 Further converting the cycle time into the number of cycle frames T
T=T 0 /（len/F s ）
Where len is the length of each frame signal and Fs is the signal sampling frequency.
Setting a minimum threshold value T for judging the periodic transient interference event according to the value of the periodic frame number T min And a maximum threshold value T max The specific setting mode is that the number of periodic frames T is respectively increased and decreased by one tolerance frame number TR to obtain the minimum threshold value T min And a maximum threshold value T max ；
A minimum threshold value T is set according to T min And a maximum threshold value T max The purpose is to expand the value of the cycle frame number T to a valueRange, thereby increasing the accuracy of the cycle count in step S333, therefore T min ，T max Should not be too large, in this embodiment it is a set empirical value, such as T min =T-10，T max T +10, i.e. the tolerance frame number TR is 10;
s333, judging the current difference value stored in the circulating window, if the difference value is more than or equal to the threshold minimum value and less than or equal to the threshold maximum value, carrying out period counting,
if T is min ≤T d (l) ≤T max
Count value Count plus 1
If the period count of the L frames in the circulation window is larger than the specified periodic transient interference event threshold T judge For example, if the count is greater than 10, it is determined that a periodic transient interference event occurs.
The invention is to detect the periodicity of transient interference, and utilize the conditional probability of transient interference obtained in the process of enhancing transient interference to calculate the conditional probability of frame transient interference, and judge the transient interference by calculating the average value of the conditional probability of frame transient interference for continuous multi-frames, and when the average value is larger than a given threshold, the current frame is determined as the transient interference. And if the current frame is transient interference, carrying out periodic transient interference event detection.
The method comprises the steps of establishing an event judgment circulating window, storing a difference value between the current transient interference frame number and the last transient interference frame number, determining the period of an event through prior information, setting the minimum value and the maximum value of a judgment threshold according to the period, judging the circulating window, counting the period if the difference value of the circulating window is within the range of the judgment threshold, and judging the occurrence of a periodic transient interference event when the period count value is larger than the set threshold, so that the working state of equipment is judged.
The embodiment is as follows:
the pollution of kitchen oil smoke can seriously harm human health, and the use of the range hood during cooking is an effective measure for reducing the pollution of the kitchen oil smoke, but people may forget to turn on the range hood.
The method for detecting the periodic transient interference event provided by the embodiment is applied to the intelligent range hood, the range hood is awakened to be started by detecting the periodic transient interference formed by ignition of the gas stove, and the situation that people forget to start the range hood is avoided.
The periodic transient interference event detection method of the embodiment utilizes an OM-LSA algorithm to enhance transient interference and suppress non-transient interference speech signals and background noise. The parameter settings of this embodiment are as follows: first smoothing parameter alpha d =0.7, second smoothing parameter α p =0.3, third smoothing parameter α s =0.85, weight factor α t =0.8, empirical threshold δ =1.8,g min =0.02。
As shown in fig. 3, the abscissa is time, the ordinate is frequency domain, the upper part of fig. 3 is a spectrogram of a microphone received signal, which contains a transient interference signal and other background noise signals, and the lower part of the spectrogram is an enhanced transient noise spectrogram.
As shown in the spectrogram in the lower part of fig. 3, a part of the spectrogram with a color close to light indicates no energy, i.e., no signal, a part with a color close to red indicates the presence of a voice signal, and the darker the color, the greater the signal energy. Within the time interval of 13s to 15.5s, a plurality of vertical lines with colors deeper than the front and rear areas exist, and each vertical line indicates that the energy at the moment of the abscissa suddenly increases and accords with the characteristics of transient signals, so that the transient interference exists.
In this embodiment, in order to detect the periodicity of the transient interference, the conditional probability of the transient interference is obtained by using the conditional existence probability of the transient interference obtained in the transient interference enhancing process, an average value of the conditional probabilities of the transient interference of consecutive 3 frames is determined to perform transient interference judgment, and when the average value is greater than a given threshold, the current frame is determined to be the transient interference. If the current frame is transient interference, periodic transient interference event detection is carried out, an event judgment circulating window is firstly constructed and used for storing the difference value between the current transient interference frame number and the last transient interference frame number, then the period of the event is determined through prior information, the minimum value and the maximum value of a judgment threshold are set according to the period, finally the circulating window is judged, if the difference value of the circulating window is in the range of the judgment threshold, period counting is carried out, and finally, when the period counting value is larger than the set threshold, the occurrence of the periodic transient interference event is judged, so that the working state of the equipment is judged. The parameters of this example are set as follows:
designating a starting point bin of a band start =15, assigned end of band bin end =25, transient interference probability threshold G Th =0.3, interference threshold δ t =5，
The window length of the circulating window L =20,T 0 =100,len=32,F S =16000,T judge =10。
The transient interference signals are printed through the serial port to indicate that the transient interference signals are detected currently, if a plurality of transient interference signals are detected continuously, the existence of ignition sound of the gas stove can be judged, and the awakening words are sent to automatically start the range hood.
The foregoing is directed to preferred embodiments of the present invention, wherein the preferred embodiments are not obviously contradictory or subject to any particular embodiment, and any combination of the preferred embodiments may be combined in any overlapping manner, and the specific parameters in the embodiments and examples are only for the purpose of clearly illustrating the inventor's invention verification process and are not intended to limit the scope of the invention, which is defined by the claims and the equivalent structural changes made by the description and drawings of the present invention are also intended to be included in the scope of the present invention.
Claims (4)
1. A method for periodic transient interference event detection, comprising the steps of:
s1, setting a microphone to receive signals
y(n)=x(n)+t(n)
In the formula, n is time scale, x (n) is a voice signal, and t (n) is transient noise;
the time-frequency domain signal received by the microphone can be obtained by short-time Fourier transform
Y (k, l) = X (k, l) + T (k, l), where k is a frequency scale and l is a time-domain frame index;
estimating non-transient signal power spectrum lambda d (k, l) and stripsThe speech existence probability p (k, l);
s2, estimating a transient noise power spectrum lambda t (k, l) and optimal gain spectrum G (k, l)
The transient noise power spectrum is expressed as
λ t (k,l)=|G(k,l)*Y(k,l)| 2
Wherein G (k, l) represents the optimal gain spectrum of the power spectrum of the estimated transient interference noise, and the calculation formula is
In the formula G min Is a predetermined spectral gain, G, in the absence of transient noise H1 Is the gain in the presence of transient noise and is expressed as
Wherein e is a natural constant, ξ (k, l) is the prior signal-to-noise ratio of the transient noise, α t represents a weighting factor, and α t belongs to [0,1], and γ (k, l) is the posterior signal-to-noise ratio, v (k, l) represents the relation parameters of the prior signal-to-noise ratio and the posterior signal-to-noise ratio, and the calculation formulas are respectively:
where Y (k, l) is the time-frequency domain signal received by the microphone in step S1, λ d (k, l) is the non-transient signal power spectrum;
s3, judging whether the transient interference has periodicity according to the conditional existence probability of the transient interference:
s31, firstly, calculating the average value of the optimal gain function in the specified frequency band according to the optimal gain spectrum G (k, l) of the transient interference noise power spectrum, namely
Middle bin of the formula start Bin as a starting point of a band end Is the end of the frequency band;
calculating the average value of the optimal gain function of the continuous m frames taking the current frame as the terminal point
m is the set average optimal frame number;
s32. If
Greater than a predetermined transient interference probability threshold G Th If not, the current frame is considered not to contain transient interference noise;
when the current frame is determined to contain the transient interference noise, carrying out periodic transient interference event detection, entering the step S33, otherwise, repeating the step S32;
s33, judging whether a periodic transient interference event occurs or not;
s331, constructing an event judgment circulating window for storing the difference T between the current transient interference frame number and the last transient interference frame number d The expression is
T d = T current - T last
In the formula T current For the current number of transient interference frames, T last The number of the last transient interference frames;
constructing an event judgment circulation window to circularly store all T between the current transient interference frame number and the last transient interference frame number d Frame, the window length of the cycle window is L, for each current frame in the cycle window, the difference value of the current L frame is calculated;
T d (l)= T current - T last
l=1,2…L
s332, setting the period time of the periodic transient interference event as T 0 Further converting the cycle time into the cycle frame number T
T=T 0 /（len/F s ）
In the formula, len is the length of each frame signal, and Fs is the signal sampling frequency;
setting and judging minimum threshold value T of periodic transient interference event min And a maximum threshold value T max And satisfy T min <T<T max ；T min ，T max Respectively representing a threshold minimum value and a threshold maximum value, wherein the unit is a frame number;
s333, judging the current difference value stored in the circulating window, if the difference value is more than or equal to the threshold minimum value and less than or equal to the threshold maximum value, carrying out period counting,
if T is min ≤T d (l) ≤T max Adding 1 to the Count value;
if the period count of the L frames in the circulation window is larger than the specified periodic transient interference event threshold T judge Then, it is determined that a periodic transient interference event occurs.
2. The method of claim 1, wherein the step S331 further includes determining adjacent interference, and setting an interference threshold δ T When T is d Greater than delta T If it is determined that there is inter-neighbor interference, the subsequent step is not performed and the step S331 is repeated.
3. The method of claim 1, wherein the minimum threshold T of the periodic jamming event is determined min = T-TR and maximum threshold value T max T + TR, TR being the set number of tolerated frames.
4. The method of claim 1, wherein the step S1 is performed by estimating a non-transient signal power spectrum using an MCRA algorithm
In the formula alpha d Is a first smoothing parameter, and alpha d ∈[0,1]，
Represents time-varying smoothing parameters, p (k, l) is conditional speech existence probability, and the expression is
p(k,l)=α p *p(k,l-1)+( 1-α p )*I(k,l)
In the formula alpha p Is a second smoothing parameter, and alpha p ∈[0,1]When I (k, l) =1, marking the k frequency point of the l frame to contain a voice signal, otherwise not containing, wherein delta is an empirical threshold value, S r (k, l) is the power spectrum ratio, and the expression is
S r （k,l）= S（k,l）/ S min (k, l), where S (k, l) is the time-domain smoothed power of the noisy speech power spectrum, S min (k, l) is the minimum power value searched, and the expressions are sequentially
S（k,l）=α S *S（k,l-1）+(1-α S ) *| Y（k,l）| 2
S min （k,l）=min{ S min （k,l-1）, S（k,l）}
In the formula alpha S Is a third smoothing parameter, and S ∈[0,1]。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310198975.9A CN115954012B (en) | 2023-03-03 | 2023-03-03 | Periodic transient interference event detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310198975.9A CN115954012B (en) | 2023-03-03 | 2023-03-03 | Periodic transient interference event detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115954012A true CN115954012A (en) | 2023-04-11 |
| CN115954012B CN115954012B (en) | 2023-05-09 |
Family
ID=85892921
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310198975.9A Active CN115954012B (en) | 2023-03-03 | 2023-03-03 | Periodic transient interference event detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115954012B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116312545A (en) * | 2023-05-26 | 2023-06-23 | 北京道大丰长科技有限公司 | Speech recognition system and method in a multi-noise environment |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102084416A (en) * | 2008-02-21 | 2011-06-01 | 史诺有限公司 | Audio visual signature, method of deriving a signature, and method of comparing audio-visual data |
| CN103310800A (en) * | 2012-03-06 | 2013-09-18 | 中国科学院声学研究所 | Voiced speech detection method and voiced speech detection system for preventing noise interference |
| CN103456310A (en) * | 2013-08-28 | 2013-12-18 | 大连理工大学 | Transient noise suppression method based on spectrum estimation |
| CN104157295A (en) * | 2014-08-22 | 2014-11-19 | 中国科学院上海高等研究院 | Method used for detecting and suppressing transient noise |
| CN106170929A (en) * | 2014-02-10 | 2016-11-30 | 奥迪马科斯公司 | There is the communication system of the noise immunity of improvement, method and apparatus |
| CN107507630A (en) * | 2017-07-17 | 2017-12-22 | 嘉兴开泽电子设备有限公司 | A kind of non-cooperation voice communication receives data dead time section recognition methods |
| CN108922554A (en) * | 2018-06-04 | 2018-11-30 | 南京信息工程大学 | The constant Wave beam forming voice enhancement algorithm of LCMV frequency based on logarithm Power estimation |
| CN110739005A (en) * | 2019-10-28 | 2020-01-31 | 南京工程学院 | real-time voice enhancement method for transient noise suppression |
| US20200388267A1 (en) * | 2019-06-05 | 2020-12-10 | Harman International Industries, Incorporated | Voice echo suppression in engine order cancellation systems |
| CN114527321A (en) * | 2022-02-24 | 2022-05-24 | 国网北京市电力公司 | Anti-transient interference voltage sag detection method and device suitable for SSTS |
| US20220301582A1 (en) * | 2016-01-25 | 2022-09-22 | China Academy Of Telecommunications Technology | Method and apparatus for determining speech presence probability and electronic device |
| US20220375479A1 (en) * | 2021-05-04 | 2022-11-24 | Cypress Semiconductor Corporation | Retransmission Softbit Decoding |
-
2023
- 2023-03-03 CN CN202310198975.9A patent/CN115954012B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102084416A (en) * | 2008-02-21 | 2011-06-01 | 史诺有限公司 | Audio visual signature, method of deriving a signature, and method of comparing audio-visual data |
| CN103310800A (en) * | 2012-03-06 | 2013-09-18 | 中国科学院声学研究所 | Voiced speech detection method and voiced speech detection system for preventing noise interference |
| CN103456310A (en) * | 2013-08-28 | 2013-12-18 | 大连理工大学 | Transient noise suppression method based on spectrum estimation |
| CN106170929A (en) * | 2014-02-10 | 2016-11-30 | 奥迪马科斯公司 | There is the communication system of the noise immunity of improvement, method and apparatus |
| CN104157295A (en) * | 2014-08-22 | 2014-11-19 | 中国科学院上海高等研究院 | Method used for detecting and suppressing transient noise |
| US20220301582A1 (en) * | 2016-01-25 | 2022-09-22 | China Academy Of Telecommunications Technology | Method and apparatus for determining speech presence probability and electronic device |
| CN107507630A (en) * | 2017-07-17 | 2017-12-22 | 嘉兴开泽电子设备有限公司 | A kind of non-cooperation voice communication receives data dead time section recognition methods |
| CN108922554A (en) * | 2018-06-04 | 2018-11-30 | 南京信息工程大学 | The constant Wave beam forming voice enhancement algorithm of LCMV frequency based on logarithm Power estimation |
| US20200388267A1 (en) * | 2019-06-05 | 2020-12-10 | Harman International Industries, Incorporated | Voice echo suppression in engine order cancellation systems |
| CN110739005A (en) * | 2019-10-28 | 2020-01-31 | 南京工程学院 | real-time voice enhancement method for transient noise suppression |
| US20220375479A1 (en) * | 2021-05-04 | 2022-11-24 | Cypress Semiconductor Corporation | Retransmission Softbit Decoding |
| CN114527321A (en) * | 2022-02-24 | 2022-05-24 | 国网北京市电力公司 | Anti-transient interference voltage sag detection method and device suitable for SSTS |
Non-Patent Citations (1)
| Title |
|---|
| 李茂: ""基于矩阵补全的天波雷达瞬态干扰抑制算法"", 《电子与信息学报》 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116312545A (en) * | 2023-05-26 | 2023-06-23 | 北京道大丰长科技有限公司 | Speech recognition system and method in a multi-noise environment |
| CN116312545B (en) * | 2023-05-26 | 2023-07-21 | 北京道大丰长科技有限公司 | Speech recognition system and method in a multi-noise environment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115954012B (en) | 2023-05-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108831499B (en) | Speech enhancement method using speech existence probability | |
| EP0790599B1 (en) | A noise suppressor and method for suppressing background noise in noisy speech, and a mobile station | |
| CN103854662B (en) | Adaptive voice detection method based on multiple domain Combined estimator | |
| CN103238183B (en) | Noise suppression device | |
| JP4279357B2 (en) | Apparatus and method for reducing noise, particularly in hearing aids | |
| EP2201563B1 (en) | Multiple microphone voice activity detector | |
| CN103456310B (en) | Transient noise suppression method based on spectrum estimation | |
| US4630304A (en) | Automatic background noise estimator for a noise suppression system | |
| CN110232933B (en) | Audio detection method and device, storage medium and electronic equipment | |
| EP3998557B1 (en) | Audio signal processing method and related apparatus | |
| CN110739005B (en) | Real-time voice enhancement method for transient noise suppression | |
| CN109410977B (en) | Voice segment detection method based on MFCC similarity of EMD-Wavelet | |
| CA2404030A1 (en) | Relative noise ratio weighting techniques for adaptive noise cancellation | |
| CN115954012B (en) | Periodic transient interference event detection method | |
| US20030220786A1 (en) | Communication system noise cancellation power signal calculation techniques | |
| JP2004254322A5 (en) | ||
| WO1989008910A1 (en) | Voice activity detection | |
| EP1279163A1 (en) | Speech presence measurement detection techniques | |
| CN112242147A (en) | Voice gain control method and computer storage medium | |
| CN112951259A (en) | Audio noise reduction method and device, electronic equipment and computer readable storage medium | |
| CN112309417A (en) | Wind noise suppression audio signal processing method, device, system and readable medium | |
| Hirszhorn et al. | Transient interference suppression in speech signals based on the OM-LSA algorithm | |
| CN111292758A (en) | Voice activity detection method and device and readable storage medium | |
| WO2017128910A1 (en) | Method, apparatus and electronic device for determining speech presence probability | |
| CN110689905B (en) | Voice activity detection system for video conference system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |