Summary of the invention
The embodiment of the invention discloses a kind of picture covering methods applied to group, can effectively prevent the group of group
The privacy photo issued on group session interface is sniffed by other people, reduces the risk that privacy photo is revealed.
Wherein, a kind of picture covering method applied to group, the group are created in social client, the side
Method includes:
The social client detection is for the either objective picture input issued on the cluster conversation interface of the group
Covering instruction;
The social client is instructed according to the covering, prompts to be selected from the group in cluster conversation circle
Publication has any two user of voice signal on face, to form the first user couple;
First user couple is associated by the social client with the Target Photo, and in group's meeting
It talks about and covers the Target Photo using preset non-privacy picture on interface;Wherein, first user is to as removing
The first foundation of the non-privacy picture of the Target Photo is covered.
As an alternative embodiment, the social client is in cluster conversation circle in the embodiment of the present invention
After covering the Target Photo using preset non-privacy picture on face, the method also includes:
The social client detection has covered the described non-of the Target Photo on the cluster conversation interface
The removal of privacy picture instructs;
The social client is instructed according to the removal, prompts to choose any two user from the group, with shape
At second user pair;
The social client judge the second user pair with as removing the institute for having covered the Target Photo
First user of the first foundation of non-privacy picture is stated to whether identical;
If they are the same, the social client will cover the described non-of the Target Photo on the cluster conversation interface
Privacy picture is removed, to show the Target Photo again.
As an alternative embodiment, in the embodiment of the present invention, the social client is by first user couple
After being associated with the Target Photo, and preset non-privacy picture covering institute is utilized on the cluster conversation interface
Before stating Target Photo, the method also includes:
Social client prompt selected from the cluster conversation interface first user to comprising one
User publication the first voice signal, and prompt selected from the cluster conversation interface first user to comprising
The second sound signal of another user publication;
The social client synthesize to first voice signal of selection and the second sound signal
To verifying voice signal;
The verifying voice signal is associated by the social client with the Target Photo, and execute it is described
The step of covering the Target Photo using preset non-privacy picture on the cluster conversation interface;Wherein, the verifying sound
Sound signal is as the second foundation for removing the non-privacy picture for having covered the Target Photo;
The social client judges that the second user pair has covered the Target Photo with as removing
First user of first foundation of the non-privacy picture is to after identical and the social client is by the group
The non-privacy picture for having covered the Target Photo on session interface is removed, to show the Target Photo again
Before, the method also includes:
Social client prompt selected from the cluster conversation interface second user to comprising one
User publication third voice signal, and prompt selected from the cluster conversation interface second user to comprising
The falling tone sound signal of another user publication;
The social client synthesize to the third voice signal of selection and the falling tone sound signal
To synthetic video signal;
The social client judges the synthetic video signal and has covered the Target Photo as removing
Whether the verifying voice signal of the second foundation of the non-privacy picture matches, if matching, executes described by the group
The non-privacy picture for having covered the Target Photo on group session interface is removed, to show the target figure again
The step of piece.
As an alternative embodiment, in the embodiment of the present invention, the social client to selection described first
Voice signal and the second sound signal carry out synthesis and are verified voice signal, including:
The social client determines being aligned between first voice signal and the second sound signal chosen
Point;Wherein, the snap point refers to the starting position that first voice signal is synthesized with the second sound signal;
The social client synthesizes first voice signal with the second sound signal according to the snap point
To verify voice signal.
As an alternative embodiment, in the embodiment of the present invention, the social client determines described the chosen
Snap point between one voice signal and the second sound signal, including:
The social client calculates the first duration and the rising tone message for first voice signal chosen
Number the second duration;Wherein, first duration indicates the time of the sound go of first voice signal;When described second
The time of the long sound go for indicating the second sound signal;
The social client calculates the difference between first duration and second duration;
The social client judges whether the difference is less than or equal to default value, if so, to first sound
Any voice signal in signal and the second sound signal carries out the scaling on the period, identical to obtain final duration
The first voice signal and second sound signal, then with identical first voice signal of the final duration and second sound
The first audio frame of signal is as snap point.
As an alternative embodiment, the social client believes first sound in the embodiment of the present invention
Number and the second sound signal in any voice signal carry out the period on scaling, including:
If the first duration of first voice signal is shorter relative to the second duration of the second sound signal, described
Social client determines that the difference accounts for the ratio X of the first duration of first voice signal according to the difference;
The social client calculates the audio frame number Y of first voice signal;
The social client calculates amplification coefficient Z, the Z=X* (Y/ (Y-1));
The social client according to the amplification coefficient, in first voice signal other than first audio frame
Other audio frames carry out equal proportion amplification so that the final duration of amplified first voice signal and described second
Second duration of voice signal is identical.
As an alternative embodiment, in the embodiment of the present invention, it is described if the difference is greater than the default value
Method further includes:
The social client is using identical default sample frequency to first voice signal and the rising tone
Sound signal is sampled respectively, obtains the first set of samples and the second set of samples;
The social client according to the default sample frequency, first set of samples, second set of samples and
Cross-correlation weight generates cross-correlation group;Wherein, the cross-correlation weight and the difference are positively correlated, and include in the cross-correlation group
There are multiple numerical value;
Multiple numerical value in the cross-correlation group are compared by the social client, find out maximum numerical value;
The social client uses the corresponding audio frame position of the maximum numerical value as snap point.
As an alternative embodiment, the social client is according to the default sample in the embodiment of the present invention
Frequency, first set of samples, second set of samples and cross-correlation weight generate cross-correlation group, including:
Wherein, Sn[t] indicates cross-correlation group, and x [m] indicates m-th of sampled data in first set of samples, y [m-t]
Indicate (m-t) a sampled data in second set of samples, t indicates the offset of time, and t is integer, value be from 0 to
M, WtIndicate window function, wherein n=l*f, l is cross-correlation weight, and f is the default sample frequency.
As an alternative embodiment, the social client judges the synthetic video in the embodiment of the present invention
Signal with as removing the verifying sound for having covered the second foundation of the non-privacy picture of the Target Photo
Whether signal matches, including:
The social client judges the synthetic video signal and has covered the Target Photo as removing
Whether the snap point between the verifying voice signal of the second foundation of the non-privacy picture is identical;
If they are the same, the social client judges corresponding first multi-C vector of the vocal print feature of the synthetic video signal
Whether the second multi-C vector corresponding with the verifying vocal print feature of voice signal matches, if matching, determines the synthetic sound
Sound signal matches with the verifying voice signal;If mismatching, determine that the synthetic video signal and the verifying sound are believed
Number mismatch;
Wherein, corresponding first multi-C vector of the vocal print feature of the synthetic video signal by mel-frequency cepstrum coefficient,
Linear prediction residue error, the first-order difference of the mel-frequency cepstrum coefficient, the linear prediction residue error a scale
Divide, energy, the first-order difference of the energy and Gammatone filter cepstrum coefficient collectively constitute.
In the embodiment of the present invention, social client can detecte any mesh issued on the cluster conversation interface for group
The covering instruction of piece of marking on a map input, and prompt has appointing for voice signal from the publication on cluster conversation interface is selected in the group
Anticipate two users, to form the first user couple;First user couple can be associated by social client with Target Photo, and
Preset non-privacy picture coverage goal picture is utilized on cluster conversation interface, the first user has covered to as removing
First foundation of the non-privacy picture of Target Photo.As it can be seen that implementing the embodiment of the present invention, even if in the use where social client
In the case where family equipment (such as mobile phone) non-screen locking, it also can effectively prevent the privacy issued on the cluster conversation interface of group from shining
Piece is sniffed by other people, reduces the risk that privacy photo is revealed.
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 the described embodiment is only a part of the embodiment of the present invention, instead of all the embodiments.Based on this
Embodiment in invention, every other reality obtained by those of ordinary skill in the art without making creative efforts
Example is applied, shall fall within the protection scope of the present invention.
It should be noted that the term " includes " of the embodiment of the present invention and " having " and their any deformation, it is intended that
Be to cover it is non-exclusive include, for example, containing the process, method, system, product or equipment of a series of steps or units not
Those of be necessarily limited to be clearly listed step or unit, but may include be not clearly listed or for these processes, side
The intrinsic other step or units of method, product or equipment.
The embodiment of the invention discloses a kind of picture covering methods applied to group, even if where social client
In the case where user equipment (such as mobile phone) non-screen locking, the privacy issued on the cluster conversation interface of group also can be effectively prevented
Photo is sniffed by other people, reduces the risk that privacy photo is revealed.Attached drawing is combined below to be described in detail.
Referring to Fig. 1, Fig. 1 is that a kind of process of the picture covering method applied to group disclosed by the embodiments of the present invention is shown
It is intended to.In picture covering method applied to group shown in Fig. 1, group is created in social client, social client
End can include but is not limited to wechat, QQ;Social client may be mounted on the user equipmenies such as cell phone, tablet computer.
As shown in Figure 1, should may comprise steps of applied to the picture covering method of group:
101, social client detection covering for the either objective picture input issued on the cluster conversation interface of group
Lid instruction.
As an alternative embodiment, social client execute step 101 after and social client execute step
Before rapid 102, following steps can also be performed:
Social client acquires the color information of the facial image of the currently used person of social client;
Social client carries out binary conversion treatment to the color information of the facial image of currently used person;
The facial image of currently used person after binary conversion treatment is divided into multiple block of pixels by social client, and to each
The corresponding pixel value of all pixels carries out or operation in block of pixels, obtain each block of pixels or operation result composition it is currently used
The down-sampling picture of the facial image of person;
Obtained down-sampling picture is divided into multiple pixel regions by social client, by institute in each pixel region
Have pixel or operation result summation, obtain the feature for forming each pixel region of facial image of the currently used person
Information;
Social client currently makes according to the judgement of the characteristic information of each pixel region of the facial image of currently used person
Whether the facial image of user and the facial image of the pre-stored legitimate user of social client match, if matching, execute
Step 102;If mismatching, terminate this process.Wherein, this embodiment can accurately identify working as social client
Preceding user just executes subsequent picture covering when being the pre-stored legitimate user of social client, prevent illegal user wantonly
Triggering social client covers the either objective picture issued on the cluster conversation interface of group, so as to promote figure
Legitimacy, the reliability of piece covering.
As another optional embodiment, social client is according to each pixel of the facial image of currently used person
The characteristic information in region judges the face of the facial image of currently used person Yu the pre-stored legitimate user of social client
After image matches and before social client execution step 102, following steps can also be performed:
Social client instructs the corresponding fingerprint image that touches to pre-process the covering, and pretreatment includes respectively to touching
Image segmentation, image enhancement, image binaryzation and the micronization processes of fingerprint image obtain input refinement fingerprint image;
Social client refines the minutiae point that takes the fingerprint in fingerprint image in input, and carries out to input refinement fingerprint image
Tracking, and the sampled point in input refinement fingerprint image on crestal line is extracted, and extract the sampled point of input refinement fingerprint image
Convex closure, generate containing fingerprint minutiae, all crestal lines up-sampling point and sampled point convex closure input fingerprint characteristic;
Social client identification input fingerprint characteristic and the fingerprint characteristic of the pre-stored legitimate user of social client are
It is no to match, if matched, just execute step 102;If mismatched, terminate this process.Wherein, this embodiment can be with
It is just held when the currently used person for more accurately identifying social client is the pre-stored legitimate user of social client
The subsequent picture covering of row prevents illegal user from triggering social client wantonly and appoints to what is issued on the cluster conversation interface of group
One Target Photo is covered, so as to promote legitimacy, the reliability of picture covering.
102, social client is instructed according to the covering, prompts to have from selecting to issue on cluster conversation interface in group
Any two user of voice signal, to form the first user couple.
In the embodiment of the present invention, social client can be instructed according to the covering, be selected from group in cluster conversation
Publication has each user of voice signal on interface, and pops up the first user to selection interface, and first user is to selection interface
On include from the mark for selecting the publication on cluster conversation interface in group and having voice signal each user;Correspondingly, society
Client is handed over to can detecte the currently used person of social client from first user to any two selected in selection interface
The mark of user, so that realization has any two of voice signal to use from the publication on cluster conversation interface is selected in group
Family, to form the first user couple.
Wherein, publication has each user of voice signal to refer to currently on cluster conversation interface on cluster conversation interface
The publication that can be viewed has each user of voice signal.
103, the first user couple is associated by social client with Target Photo, and utilizes on cluster conversation interface
Preset non-privacy picture coverage goal picture;Wherein, the first user is to as removing the non-hidden of coverage goal picture
First foundation of one's personal scheme or attempt piece.
Wherein, the first user couple and Target Photo are associated by social client refers to social client by the first user
The corresponding relationship of the mark of two users of centering is associated with Target Photo.
In the method depicted in fig. 1, though user equipment (such as mobile phone) non-screen locking where social client feelings
It under condition, also can effectively prevent the privacy photo issued on the cluster conversation interface of group from being sniffed by other people, reduce privacy
The risk that photo is revealed.In addition, in the method depicted in fig. 1, can prevent illegal user from triggering social client wantonly
End covers the either objective picture issued on the cluster conversation interface of group, so as to promote the legal of picture covering
Property, reliability.
Referring to Fig. 2, the process of Fig. 2 another picture covering method for being applied to group disclosed by the embodiments of the present invention is shown
It is intended to.As shown in Fig. 2, should may comprise steps of applied to the picture covering method of group:
201, social client detection covering for the either objective picture input issued on the cluster conversation interface of group
Lid instruction.
As an alternative embodiment, social client execute step 201 after and social client execute step
Before rapid 202, following steps can also be performed:
Social client acquires the color information of the facial image of the currently used person of social client;
Social client carries out binary conversion treatment to the color information of the facial image of currently used person;
The facial image of currently used person after binary conversion treatment is divided into multiple block of pixels by social client, and to each
The corresponding pixel value of all pixels carries out or operation in block of pixels, obtain each block of pixels or operation result composition it is currently used
The down-sampling picture of the facial image of person;
Obtained down-sampling picture is divided into multiple pixel regions by social client, by institute in each pixel region
Have pixel or operation result summation, obtain the feature for forming each pixel region of facial image of the currently used person
Information;
Social client currently makes according to the judgement of the characteristic information of each pixel region of the facial image of currently used person
Whether the facial image of user and the facial image of the pre-stored legitimate user of social client match, if matching, execute
Step 202;If mismatching, terminate this process.Wherein, this embodiment can accurately identify working as social client
Preceding user just executes subsequent picture covering when being the pre-stored legitimate user of social client, prevent illegal user wantonly
Triggering social client covers the either objective picture issued on the cluster conversation interface of group, so as to promote figure
Legitimacy, the reliability of piece covering.
As another optional embodiment, social client is according to each pixel of the facial image of currently used person
The characteristic information in region judges the face of the facial image of currently used person Yu the pre-stored legitimate user of social client
After image matches and before social client execution step 202, following steps can also be performed:
Social client instructs the corresponding fingerprint image that touches to pre-process the covering, and pretreatment includes respectively to touching
Image segmentation, image enhancement, image binaryzation and the micronization processes of fingerprint image obtain input refinement fingerprint image;
Social client refines the minutiae point that takes the fingerprint in fingerprint image in input, and carries out to input refinement fingerprint image
Tracking, and the sampled point in input refinement fingerprint image on crestal line is extracted, and extract the sampled point of input refinement fingerprint image
Convex closure, generate containing fingerprint minutiae, all crestal lines up-sampling point and sampled point convex closure input fingerprint characteristic;
Social client identification input fingerprint characteristic and the fingerprint characteristic of the pre-stored legitimate user of social client are
It is no to match, if matched, just execute step 202;If mismatched, terminate this process.Wherein, this embodiment can be with
It is just held when the currently used person for more accurately identifying social client is the pre-stored legitimate user of social client
The subsequent picture covering of row prevents illegal user from triggering social client wantonly and appoints to what is issued on the cluster conversation interface of group
One Target Photo is covered, so as to promote legitimacy, the reliability of picture covering.
202, social client is instructed according to the covering, prompts to have from selecting to issue on cluster conversation interface in group
Any two user of voice signal, to form the first user couple.
In the embodiment of the present invention, social client can be instructed according to the covering, be selected from group in cluster conversation
Publication has each user of voice signal on interface, and pops up the first user to selection interface, and first user is to selection interface
On include from the mark for selecting the publication on cluster conversation interface in group and having voice signal each user;Correspondingly, society
Client is handed over to can detecte the currently used person of social client from first user to any two selected in selection interface
The mark of user, so that realization has any two of voice signal to use from the publication on cluster conversation interface is selected in group
Family, to form the first user couple.
Wherein, publication has each user of voice signal to refer to currently on cluster conversation interface on cluster conversation interface
The publication that can be viewed has each user of voice signal.
203, the first user couple is associated by social client with Target Photo, and utilizes on cluster conversation interface
Preset non-privacy picture coverage goal picture;Wherein, the first user is to as removing the non-hidden of coverage goal picture
First foundation of one's personal scheme or attempt piece.
204, shifting of the social client detection for the non-privacy picture of the picture of coverage goal on cluster conversation interface
Except instruction.
205, social client prompts to choose any two user from group according to instruction is removed, and is used with forming second
Family pair.
In the embodiment of the present invention, social client can be according to this according to instruction is removed, and pop-up second user is to selection circle
Face, the second user is to the mark including all users in group in selection interface;Correspondingly, social client can detecte
The currently used person of social client from the second user to the mark of any two user selected in selection interface, thus real
Any two user is selected, from group now to form second user pair.
206, the second user pair that social client judgement is chosen with as removing the non-of coverage goal picture
Whether the first user of the first foundation of privacy picture is to identical;If they are the same, step 207 is executed;If not identical, terminate this stream
Journey.
207, social client removes the non-privacy picture of the picture of coverage goal on the cluster conversation interface,
With displaying target picture again.
In the method depicted in fig. 2, only when any two user that social client is selected from group is formed
Second user pair it is associated with Target Photo and on cluster conversation interface publication have the first user of voice signal to identical
When, just the non-privacy picture of the picture of coverage goal on the cluster conversation interface is removed, with displaying target figure again
Piece;, whereas if second user pair and Target Photo that any two user that social client is selected from group is formed
It is associated and publication on cluster conversation interface have voice signal the first user it is not identical when, on the cluster conversation interface
The non-privacy picture of coverage goal picture will not be removed, and Target Photo is still capped.As it can be seen that implementing described in Fig. 2
Method, even if also can effectively prevent group in the case where user equipment (such as mobile phone) non-screen locking where social client
The privacy photo issued on the cluster conversation interface of group is sniffed by other people, reduces the risk that privacy photo is revealed.In addition,
In the method depicted in fig. 2, it can prevent illegal user from triggering social client wantonly on the cluster conversation interface of group
The either objective picture of publication is covered, so as to promote legitimacy, the reliability of picture covering.
Referring to Fig. 3, the process of Fig. 3 another picture covering method for being applied to group disclosed by the embodiments of the present invention is shown
It is intended to.As shown in figure 3, should may comprise steps of applied to the picture covering method of group:
301, social client detection covering for the either objective picture input issued on the cluster conversation interface of group
Lid instruction.
302, social client is instructed according to the covering, prompts to have from selecting to issue on cluster conversation interface in group
Any two user of voice signal, to form the first user couple.
303, the first user couple is associated by social client with Target Photo, prompts to choose from cluster conversation interface
Out the first user to comprising user publication the first voice signal, and prompt selects the from cluster conversation interface
One user to comprising another user publication second sound signal.
Wherein, the first user to comprising a user publication the first voice signal can be the first user to comprising
The voice signal or environmental sound signal that one user issues on cluster conversation interface, the first user to comprising another user
First voice signal of publication can be the first user to comprising the voice issued on cluster conversation interface of another user
Signal or environmental sound signal, the embodiment of the present invention are not construed as limiting.
304, social client carries out synthesis to the first voice signal and second sound signal of selection and is verified sound
Sound signal.
As an alternative embodiment, social client is social before executing step 304 in present invention implementation
Client can first judge whether the first voice signal and second sound signal are voice signal, if the first voice signal with
And second sound signal is voice signal, just executes step 304.
For example, social client can take following manner accurately to judge whether the first voice signal is voice
Signal, i.e.,:
Social client carries out Fast Fourier Transform (FFT) to the first voice signal and obtains frequency-region signal;
Social client calculates spectrum amplitude angle value according to frequency-region signal;
Social client calculates probability density according to spectrum amplitude angle value;
Social client calculates the spectrum entropy of the first voice signal according to probability density;
Social client determines whether the first voice signal is voice signal according to spectrum entropy;
Wherein, social client determines whether the first voice signal is voice signal according to spectrum entropy, may include:
Social client calculates the energy of the first voice signal;
Social client determines whether the first voice signal is voice letter according to the energy and spectrum entropy of the first voice signal
Number, i.e., social client can calculate the product of the energy of the first voice signal and the spectrum entropy of the first voice signal, and multiply to this
Product carries out extraction of square root operation, obtains the corresponding extraction of square root value of the product;And social client may determine that the product is corresponding
Whether extraction of square root value is greater than preset threshold, if so, determining that the first voice signal is voice signal, if not, it is determined that first
Voice signal is not voice signal.
305, social client will verify voice signal and Target Photo is associated, using pre- on cluster conversation interface
If non-privacy picture coverage goal picture;Wherein, the first user is to as the non-privacy for removing coverage goal picture
First foundation of picture;Verifying voice signal as be used to remove the second of the non-privacy picture of coverage goal picture according to
According to.
306, shifting of the social client detection for the non-privacy picture of the picture of coverage goal on cluster conversation interface
Except instruction.
307, social client prompts to choose any two user from group according to instruction is removed, and is used with forming second
Family pair.
308, social client judge second user pair with as the non-privacy picture for removing coverage goal picture
The first foundation the first user to whether identical;If they are the same, step 309- step 311 is executed;If not identical, terminate this stream
Journey.
309, social client prompt selected from cluster conversation interface second user to comprising a user issue
Third voice signal, and prompt selected from cluster conversation interface second user to comprising another user publication
Falling tone sound signal.
310, social client is synthesized to obtain synthetic sound to the third voice signal and falling tone sound signal of selection
Sound signal.
311, social client judges synthetic video signal and as the non-privacy figure for removing coverage goal picture
Whether the verifying voice signal of the second foundation of piece matches, if matching, executes step 312;If mismatching, terminate this stream
Journey.
312, social client removes the non-privacy picture of the picture of coverage goal on the cluster conversation interface,
With displaying target picture again.
As an alternative embodiment, in above-mentioned steps 304, social client to the first voice signal of selection with
And second sound signal carries out synthesis and is verified voice signal, including:
Social client determines the snap point between the first voice signal and second sound signal chosen;Wherein, it is aligned
Point refers to the starting position that the first voice signal is synthesized with second sound signal;In other words, the first voice signal and the rising tone
It if sound signal will synthesize, needs to find and is synthesized since which audio frame, this audio frame is it can be understood that be snap point;
First voice signal and second sound signal are synthesized verifying voice signal according to snap point by social client.
As an alternative embodiment, social client determines the first voice signal and second sound signal chosen
Between snap point, including:
Social client calculates the first duration of the first voice signal chosen and the second duration of second sound signal;
Wherein, the first duration indicates the time of the sound go of the first voice signal;The sound of second duration expression second sound signal
Duration;
Social client calculates the difference between the first duration and the second duration;
Social client judges whether the difference is less than or equal to default value, if so, to the first voice signal and second
Any voice signal in voice signal carries out the scaling on the period, to obtain identical first voice signal of final duration
And second sound signal, then with the first audio frame of identical first voice signal of final duration and second sound signal work
For snap point.
In the embodiment of the present invention, if the difference is less than or equal to default value, illustrate two voice signals (i.e. first
Voice signal and second sound signal) input when gap it is smaller, at this time can be to one of voice signal (such as first
Voice signal) carry out the period on scaling, such as to the longer voice signal of duration carry out the period on compression (be namely commonly called as
F.F.), and/or the amplification (slow-motion being namely commonly called as) on the period is carried out to the shorter voice signal of duration, so that two
The final duration of voice signal is identical, then is aligned using the first audio frame of two voice signals as snap point.
Wherein, the value range of the default value can be 0 to 0.1 second.
As an alternative embodiment, social client is to any in the first voice signal and second sound signal
Voice signal carries out the scaling on the period, including:
If the first duration of the first voice signal is shorter relative to the second duration of second sound signal, social client root
Determine that the difference accounts for the ratio X of the first duration of the first voice signal according to the difference;
Social client calculates the audio frame number Y of the first voice signal;
Social client calculates amplification coefficient Z, Z=X* (Y/ (Y-1));
Social client is according to amplification coefficient, to other audio frames in the first voice signal other than first audio frame
Carry out equal proportion amplification so that the final duration of amplified first voice signal and second sound signal second when
Length is identical.
For example, the first voice signal is 1 second, has 100 audio frames, then each audio frame 0.01 second, second sound
Signal is 1.1 seconds, needs to be amplified to the first voice signal 1.1 seconds.First frame is motionless, amplifies subsequent 99 frame, and first determination is put
Big coefficient Z be 0.1* (100/ (100-1))=0.101, i.e., 10.1%;Subsequent 99 frame, every frame need to amplify at this time
10.1%, amplified every frame is 0.01* (1+10.1%)=0.01101, and the length of this 99 frame is 1.09 seconds after amplification, then plus
On do not move first frame 0.01 second, be just 1.1 seconds, i.e., the final duration and second of amplified first voice signal
Second duration of voice signal is identical.
In the embodiment of the present invention, if the difference is greater than default value, illustrate two voice signals (i.e. the first sound letter
Number and second sound signal) input when gap it is larger, if at this time or to one of voice signal carry out the period
On scaling, then scale after will cause more serious distortion, subsequent verification will appear problem, it is possible to use cross-correlation
Algorithm determines snap point.That is, this method further includes when the difference is greater than default value:
Social client using identical default sample frequency to the first voice signal and second sound signal respectively into
Row sampling, obtains the first set of samples and the second set of samples;
Social client is according to default sample frequency (such as 8000Hz to 10000Hz), the first set of samples, the second set of samples
And cross-correlation weight generates cross-correlation group;Wherein, cross-correlation weight and the difference are positively correlated that (such as the cross-correlation weight can be with
It is 1.5 times of the difference), it include multiple numerical value in cross-correlation group;
Multiple numerical value in cross-correlation group are compared by social client, find out maximum numerical value;
Social client uses the corresponding audio frame position of maximum numerical value as snap point.
Wherein, social client is raw according to default sample frequency, the first set of samples, the second set of samples and cross-correlation weight
At cross-correlation group, including:
Wherein, Sn[t] indicates cross-correlation group, and x [m] indicates m-th of sampled data in first set of samples, y [m-t]
Indicate (m-t) a sampled data in second set of samples, t indicates the offset of time, and t is integer, value be from 0 to
M, WtIndicate window function, wherein n=l*f, l is cross-correlation weight, and f is default sample frequency.
Wherein, social client, which uses the corresponding audio frame position of maximum numerical value as snap point, to be:
After social client finds the maximum numerical value, instead it can release m is how many according to above-mentioned formula (1),
It is exactly which sampled data, which the audio frame where then determining the sampled data again is, and uses the audio frame
As snap point.
In the embodiment of the present invention, social client is after getting the first voice signal and second sound signal, not
It is the two voice signals are verified one by one, but the two voice signals is carried out synthesis and are verified sound letter
Number, it is then subsequent again to match synthetic video signal with verifying voice signal when needing to remove non-privacy picture, and sound
After sound signal synthesis, it can generate and more can verify that parameter (such as whether snap point is identical, whether vocal print feature matches), phase
Than improving the safety of verifying in verifying two voice signals one by one.
As an alternative embodiment, social client judges synthetic video signal and conduct in above-mentioned steps 311
Whether the verifying voice signal for removing the second foundation of the non-privacy picture of coverage goal picture matches, including:
Social client judge synthetic video signal with as removing the non-privacy picture of coverage goal picture
Whether the snap point between the verifying voice signal of the second foundation is identical;Wherein, it includes snap point corresponding two that snap point is identical
A mixed audio frame (a mixed audio frame belongs to synthetic video signal, another mixed audio frame belongs to verifying voice signal)
Mixing content frame is identical and the frame time of the corresponding two mixed audio frames of snap point is also identical;
If they are the same, social client judges corresponding first multi-C vector of the vocal print feature of synthetic video signal and verifying sound
Whether corresponding second multi-C vector of the vocal print feature of sound signal matches, if matching, determines synthetic video signal and verifying sound
Signal matches;If mismatching, determine that synthetic video signal and verifying voice signal mismatch;
Wherein, corresponding first multi-C vector of the vocal print feature of synthetic video signal is by mel-frequency cepstrum coefficient, linear
Predict cepstrum coefficient, the first-order difference of the mel-frequency cepstrum coefficient, the first-order difference of the linear prediction residue error, energy
Amount, the first-order difference of the energy and Gammatone filter cepstrum coefficient collectively constitute.Wherein, implement above-mentioned embodiment party
The accuracy of Sound Match can be improved in formula.
In the method depicted in fig. 3, though user equipment (such as mobile phone) non-screen locking where social client feelings
It under condition, also can effectively prevent the privacy photo issued on the cluster conversation interface of group from being sniffed by other people, reduce privacy
The risk that photo is revealed.In addition, in the method depicted in fig. 3, can prevent illegal user from triggering social client wantonly
End covers the either objective picture issued on the cluster conversation interface of group, so as to promote the legal of picture covering
Property, reliability.
Referring to Fig. 4, Fig. 4 is that a kind of variation at the cluster conversation interface of social client disclosed by the embodiments of the present invention is shown
It is intended to.In cluster conversation interface shown in Fig. 4, different head portraits respectively represents user different in group, and identical head
As representing the same user in group.As shown in the A in Fig. 4, social client can detecte currently making for social client
The covering for the either objective picture that user issues on the cluster conversation interface for group by touching operation input instructs;Such as
Shown in B in Fig. 4, social client can be instructed according to the covering, be prompted from being selected on cluster conversation interface in group
Publication has any two user of voice signal, to form the first user couple;Wherein, the currently used person of social client can be with
There is voice signal any two user to carry out making hook for publication on cluster conversation interface, two users for being made hook indicate quilt
Two users selected, the two users can form the first user couple;As shown by c in fig 4, social client is by
After one user couple and Target Photo are associated, mesh can be covered using preset non-privacy picture on cluster conversation interface
It marks on a map piece;Wherein, the first user is to as the first foundation for removing the non-privacy picture of coverage goal picture.In this way,
Even if also can effectively prevent the group of group in the case where user equipment (such as mobile phone) non-screen locking where social client
The privacy photo issued on group session interface is sniffed by other people, reduces the risk that privacy photo is revealed.
Those of ordinary skill in the art will appreciate that all or part of the steps in the various methods of above-described embodiment is can
It is completed with instructing relevant hardware by program, which can be stored in a computer readable storage medium, storage
Medium include read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory,
RAM), programmable read only memory (Programmable Read-only Memory, PROM), erasable programmable is read-only deposits
Reservoir (Erasable Programmable Read Only Memory, EPROM), disposable programmable read-only memory (One-
Time Programmable Read-Only Memory, OTPROM), the electronics formula of erasing can make carbon copies read-only memory
(Electrically-Erasable Programmable Read-Only Memory, EEPROM), CD-ROM (Compact
Disc Read-Only Memory, CD-ROM) or other disc memories, magnetic disk storage, magnetic tape storage or can
For carrying or any other computer-readable medium of storing data.
A kind of picture covering method applied to group disclosed by the embodiments of the present invention is described in detail above, this
Apply that a specific example illustrates the principle and implementation of the invention in text, the explanation of above example is only intended to
It facilitates the understanding of the method and its core concept of the invention;At the same time, for those skilled in the art, think of according to the present invention
Think, there will be changes in the specific implementation manner and application range, in conclusion the content of the present specification should not be construed as pair
Limitation of the invention.