CN106790950A - The recognition methods of malicious call and device - Google Patents
The recognition methods of malicious call and device Download PDFInfo
- Publication number
- CN106790950A CN106790950A CN201510810521.8A CN201510810521A CN106790950A CN 106790950 A CN106790950 A CN 106790950A CN 201510810521 A CN201510810521 A CN 201510810521A CN 106790950 A CN106790950 A CN 106790950A
- Authority
- CN
- China
- Prior art keywords
- malicious call
- voice
- call
- malicious
- label
- 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 49
- 238000004458 analytical method Methods 0.000 claims abstract description 43
- 230000005611 electricity Effects 0.000 claims description 15
- 230000005540 biological transmission Effects 0.000 claims description 2
- 238000001228 spectrum Methods 0.000 description 91
- 238000001914 filtration Methods 0.000 description 14
- 230000008859 change Effects 0.000 description 8
- 238000005516 engineering process Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 238000003860 storage Methods 0.000 description 8
- 230000005713 exacerbation Effects 0.000 description 7
- 238000010606 normalization Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000011982 device technology Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/436—Arrangements for screening incoming calls, i.e. evaluating the characteristics of a call before deciding whether to answer it
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/66—Substation equipment, e.g. for use by subscribers with means for preventing unauthorised or fraudulent calling
- H04M1/663—Preventing unauthorised calls to a telephone set
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/66—Substation equipment, e.g. for use by subscribers with means for preventing unauthorised or fraudulent calling
- H04M1/663—Preventing unauthorised calls to a telephone set
- H04M1/665—Preventing unauthorised calls to a telephone set by checking the validity of a code
Landscapes
- Engineering & Computer Science (AREA)
- Computer Security & Cryptography (AREA)
- Signal Processing (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Telephonic Communication Services (AREA)
Abstract
Recognition methods and device the invention discloses a kind of malicious call, cloud server receive other side's voice of the certain time length that user equipment is gathered during receiving calls;Other side's voice is carried out into speech analysis, the phonetic feature point of other side's voice is obtained;The phonetic feature point of other side's voice is matched with the phonetic feature point in the malicious call voice feature data storehouse for pre-building;When the match is successful, the prompting of malicious call is returned to user equipment;So that prompting of the user equipment according to the malicious call, shows the prompting of the malicious call on call interface.Improve user makes telephonic security.
Description
Technical field
The invention belongs to mobile device technology field, specifically, it is related to a kind of identification side of malicious call
Method and device.
Background technology
In current social using phone without recipient permit, carry out mandatory illegal phone advertisement,
Phone is publicized and faxed, and they not only also often dial harassing and wrecking at dead of night, continuously on daytime
Phone, disturbs the normal living environment of people and working environment, due to harassing call unbridledly
It is much to implement harassing and wrecking words by the method for the networking telephone to move, existing public telephone network cannot be known again
The source of other harassing call, it is also difficult to differentiate harassing call and normal telephone, and due to current
Telephone set can only be by by the preset trusted telephone number of user oneself, isolating the side of other all phones
Method prevents harassing call, and this just greatly limit the purposes of telephone set, therefore harassing call is difficult Du
Absolutely, utilized by saboteur in society and bad commercial competition, carry out illegal harassing and wrecking activity.
The content of the invention
In view of this, the recognition methods this application provides a kind of malicious call and device, it is possible to achieve logical
The matching for crossing phonetic feature point is intercepted to malicious call.
The embodiment of the present invention provides a kind of recognition methods of malicious call, is performed positioned at cloud server side,
Including:
Cloud server receives other side's language of the certain time length that user equipment is gathered during receiving calls
Sound;
Other side's voice is carried out into speech analysis, the phonetic feature point of other side's voice is obtained;
By in the phonetic feature point of other side's voice and the malicious call voice feature data storehouse for pre-building
Phonetic feature point is matched;
When the match is successful, the prompting of malicious call is returned to user equipment;So that user equipment is according to institute
The prompting of malicious call is stated, the prompting of the malicious call is shown on call interface.
Alternatively, described method also includes:
Cloud server receives the voice for being designated malicious call that user equipment sends, described to be designated evil
Anticipate phone voice be user equipment during receiving calls, according to user instruction, detecting user
Selection malicious call option when, by other side's incoming identification be malicious call, and gather certain time length evil
The voice of meaning phone;
Voice to malicious call carries out speech analysis, obtains the phonetic feature point of malicious call;
The phonetic feature point of the malicious call is saved in malicious call voice feature data storehouse.
Alternatively, if carrying the label of the malicious call in the voice of the malicious call;The evil
Anticipate the label of label including intermediary's phone of phone, the label of sale call, the label of fraudulent call, disturb
Disturb the label or self-defined phone label of phone;
Then the voice to malicious call carries out speech analysis, after obtaining the phonetic feature point of malicious call,
Including:
The label carried in voice according to the malicious call, sets up the phonetic feature of the malicious call
Put the corresponding relation with the label;
The phonetic feature point of the malicious call is saved in malicious call with the corresponding relation of the label
Voice feature data storehouse.
Alternatively, if carrying the caller ID of the malicious call in the voice of the malicious call;
Then the voice to malicious call carries out speech analysis, after obtaining the phonetic feature point of malicious call,
Including:
The caller ID carried in voice according to the malicious call, sets up the voice of the malicious call
The corresponding relation of characteristic point and the caller ID;
The phonetic feature point of the malicious call and the corresponding relation of the caller ID are saved in malice
Call voice property data base.
Alternatively, the voice to malicious call carries out speech analysis, obtains the phonetic feature point of malicious call,
Including:
Obtain the power spectrum of the voice of the malicious call;
The power spectrum that will be obtained obtains the short-time rating spectrum of noisy speech by filtering process;
Short-time rating spectrum according to the noisy speech for being obtained, noisy speech is asked for by the way of frame is average
Medium duration power spectrum;
The medium duration power spectrum of the noisy speech to being obtained carries out asymmetric filtering and shelters anti-noise, with
Obtain the medium duration power spectrum of clean speech;
The medium duration power spectrum of medium duration power spectrum, noisy speech according to the clean speech and contain
Make an uproar voice short-time rating spectrum obtain clean speech short-time rating spectrum;
Short-time rating spectrum to clean speech carries out energy normalized treatment, to eliminate multiplying property noise;
Short-time rating spectrum to eliminating the clean speech of multiplying property noise the loudness such as carries out and aggravates
The short-time rating of the clean speech after reciprocity loudness exacerbation is composed into row index nonlinear operation;
Short-time rating to having carried out the clean speech after index nonlinear operation is composed and carries out Fourier inversion
Change, to ask for cepstrum coefficient;
Mean normalization treatment is carried out to cepstrum coefficient, the phonetic feature point of the malicious call is obtained.
Alternatively, after cloud server receives the voice of the malicious call that user equipment sends, also include:
By the speech recognition of the malicious call into corresponding text information;
Set up right between the text information of the malicious call and the phonetic feature point of the malicious call
Should be related to.
Alternatively, the malicious call voice feature data storehouse includes:The incoming call number of multiple malicious calls
Code information, the phonetic feature point information of multiple malicious call, the label information of the voice of multiple malicious calls,
The corresponding text information of voice of multiple malicious calls;
Also include in the malicious call voice feature data storehouse:The caller ID of malicious call and malice electricity
The phonetic feature point of corresponding relation, malicious call between the phonetic feature point of words is corresponding with label to close
System, the corresponding relation of the phonetic feature point of malicious call and text information, the caller ID of malicious call and
The corresponding relation of text information, the corresponding relation of the label of malicious call and text information, malicious call
One-to-one relationship, the incoming call number of malicious call between caller ID and phonetic feature point and label
Code, phonetic feature point, the one-to-one relationship between label and text information.
Alternatively, the prompting of the malicious call includes the label information of the malicious call, to use
Label information of the family equipment in the prompting of the malicious call, shows on the call interface of user equipment
Show the malicious call corresponding to the label.
The embodiment of the present invention also provides a kind of identifying device of malicious call, positioned at cloud server side, bag
Include:
Receiver module, the other side for receiving the certain time length that user equipment is gathered during receiving calls
Voice;
Speech analysis module, for other side's voice to be carried out into speech analysis, the voice for obtaining other side's voice is special
Levy a little;
Matching module, for the phonetic feature point of other side's voice is special with the malicious call voice for pre-building
The point of the phonetic feature in database is levied to be matched;
Sending module, for when the match is successful, the prompting of malicious call being returned to user equipment;So that
User equipment shows carrying for the malicious call according to the prompting of the malicious call on call interface
Show.
Alternatively, described device also includes:
The receiver module, is additionally operable to receive the voice for being designated malicious call that user equipment sends, institute
State be designated malicious call voice be user equipment during receiving calls, according to user instruction,
Detect user selection malicious call option when, by other side's incoming identification be malicious call, and gather one
The voice of timing malicious call long;
The speech analysis module, is additionally operable to carry out speech analysis to the voice of malicious call, obtains malice
The phonetic feature point of phone;
Preserving module, for the phonetic feature point of the malicious call to be saved in into malicious call phonetic feature
In database.
Alternatively, if carrying the label of the malicious call in the voice of the malicious call;The evil
Anticipate the label of label including intermediary's phone of phone, the label of sale call, the label of fraudulent call, disturb
Disturb the label or self-defined phone label of phone;
Described device also includes:
Module is set up, for the label carried in the voice according to the malicious call, the malice is set up
The phonetic feature point of phone and the corresponding relation of the label;
The preserving module, is additionally operable to the phonetic feature point of the malicious call is corresponding with the label
Relation is saved in malicious call voice feature data storehouse.
Alternatively, if carrying the caller ID of the malicious call in the voice of the malicious call;
It is described to set up module, the caller ID carried in the voice according to the malicious call is additionally operable to, build
Found the phonetic feature point of the malicious call and the corresponding relation of the caller ID;
The preserving module, is additionally operable to the phonetic feature point and the caller ID of the malicious call
Corresponding relation is saved in malicious call voice feature data storehouse.
Alternatively, the speech analysis module specifically for:
Obtain the power spectrum of the voice of the malicious call;
The power spectrum that will be obtained obtains the short-time rating spectrum of noisy speech by filtering process;
Short-time rating spectrum according to the noisy speech for being obtained, noisy speech is asked for by the way of frame is average
Medium duration power spectrum;
The medium duration power spectrum of the noisy speech to being obtained carries out asymmetric filtering and shelters anti-noise, with
Obtain the medium duration power spectrum of clean speech;
The medium duration power spectrum of medium duration power spectrum, noisy speech according to the clean speech and contain
Make an uproar voice short-time rating spectrum obtain clean speech short-time rating spectrum;
Short-time rating spectrum to clean speech carries out energy normalized treatment, to eliminate multiplying property noise;
Short-time rating spectrum to eliminating the clean speech of multiplying property noise the loudness such as carries out and aggravates
The short-time rating of the clean speech after reciprocity loudness exacerbation is composed into row index nonlinear operation;
Short-time rating to having carried out the clean speech after index nonlinear operation is composed and carries out Fourier inversion
Change, to ask for cepstrum coefficient;
Mean normalization treatment is carried out to cepstrum coefficient, the phonetic feature point of the malicious call is obtained.
Alternatively, the speech analysis module, is additionally operable to the speech recognition of the malicious call into correspondence
Text information;
It is described to set up module, it is additionally operable to set up the text information and the malicious call of the malicious call
Corresponding relation between phonetic feature point.
Alternatively, the malicious call voice feature data storehouse includes:The incoming call number of multiple malicious calls
Code information, the phonetic feature point information of multiple malicious call, the label information of the voice of multiple malicious calls,
The corresponding text information of voice of multiple malicious calls;
Also include in the malicious call voice feature data storehouse:The caller ID of malicious call and malice electricity
The phonetic feature point of corresponding relation, malicious call between the phonetic feature point of words is corresponding with label to close
System, the corresponding relation of the phonetic feature point of malicious call and text information, the caller ID of malicious call and
The corresponding relation of text information, the corresponding relation of the label of malicious call and text information, malicious call
One-to-one relationship, the incoming call number of malicious call between caller ID and phonetic feature point and label
Code, phonetic feature point, the one-to-one relationship between label and text information.
Alternatively, the prompting of the malicious call includes the label information of the malicious call, to use
Label information of the family equipment in the prompting of the malicious call, shows on the call interface of user equipment
Show the malicious call corresponding to the label
The embodiment of the present invention is based on malicious call voice feature data storehouse, is set when cloud server receives user
Other side's voice of the standby certain time length gathered during receiving calls;Other side's voice is carried out into voice point
Analysis, obtains the phonetic feature point of other side's voice;By the phonetic feature point and the evil for pre-building of other side's voice
Phonetic feature point in meaning call voice property data base is matched;When the match is successful, set to user
The standby prompting for returning to malicious call;So that prompting of the user equipment according to the malicious call, on call circle
The prompting of the malicious call is shown on face.So that user hangs up the telephone malicious call in time, prevent from being taken in
Being deceived, therefore can improve user makes telephonic security.
Brief description of the drawings
Accompanying drawing described herein is used for providing further understanding of the present application, constitutes of the application
Point, the schematic description and description of the application is used to explain the application, does not constitute to the application's
Improper restriction.In the accompanying drawings:
Fig. 1 is a kind of collocation method in the phonetic feature storehouse of malicious call of the offer of the embodiment of the present application
Schematic flow sheet;
Fig. 2 is a kind of schematic flow sheet of the recognition methods of malicious call of the offer of the embodiment of the present application;
Fig. 3 is a kind of structural representation of the identifying device of malicious call of the offer of the embodiment of the present application.
Specific embodiment
Embodiments of the present invention are described in detail below in conjunction with drawings and Examples, thereby to the present invention
How application technology means can fully understand to solve technical problem and reach the implementation process of technology effect
And implement according to this.
In a typical configuration, computing device include one or more processors (CPU), input/
Output interface, network interface and internal memory.
Internal memory potentially includes the volatile memory in computer-readable medium, random access memory
And/or the form, such as read-only storage (ROM) or flash memory (flash RAM) such as Nonvolatile memory (RAM).
Internal memory is the example of computer-readable medium.
Computer-readable medium includes that permanent and non-permanent, removable and non-removable media can be by
Any method or technique realizes information Store.Information can be computer-readable instruction, data structure,
The module of program or other data.The example of the storage medium of computer includes, but are not limited to phase transition internal memory
(PRAM), static RAM (SRAM), dynamic random access memory (DRAM), other
The random access memory (RAM) of type, read-only storage (ROM), electrically erasable is read-only deposits
Reservoir (EEPROM), fast flash memory bank or other memory techniques, read-only optical disc read-only storage
(CD-ROM), digital versatile disc (DVD) or other optical storages, magnetic cassette tape, tape magnetic
Disk storage or other magnetic storage apparatus or any other non-transmission medium, can be used for storage can be counted
The information that calculation equipment is accessed.Defined according to herein, computer-readable medium does not include non-temporary computer
Readable media (transitory media), such as data-signal and carrier wave of modulation.
Some vocabulary are such as have invoked in the middle of specification and claim to censure specific components.This area skill
Art personnel are, it is to be appreciated that hardware manufacturer may call same component with different nouns.This explanation
Book and claim not in the way of the difference of title is used as distinguishing component, but with component in function
On difference be used as distinguish criterion.Such as " the bag in specification in the whole text and claim mentioned in
Containing " it is an open language, therefore " include but be not limited to " should be construed to." substantially " refer to can
In the error range of reception, those skilled in the art can solve the technology in the range of certain error and ask
Topic, basically reaches the technique effect.Additionally, " coupling " one word is herein comprising any directly and indirect
Electric property coupling means.Therefore, if a first device is coupled to a second device described in text, represent
The first device can directly be electrically coupled to the second device, or by other devices or coupling means
The second device is electrically coupled to indirectly.Specification subsequent descriptions are implementation preferable implementation of the invention
Mode, so the description is for the purpose of illustrating rule of the invention, to be not limited to the present invention
Scope.Protection scope of the present invention ought be defined depending on the appended claims person of defining.
Also, it should be noted that term " including ", "comprising" or its any other variant be intended to contain
Lid nonexcludability is included, so that commodity or system including a series of key elements not only include those
Key element, but also other key elements including being not expressly set out, or also include for this commodity or
The intrinsic key element of system.In the absence of more restrictions, being limited by sentence "including a ..."
Fixed key element, it is not excluded that also exist in the commodity or system including the key element it is other it is identical will
Element
The embodiment of the present invention is based on malicious call voice feature data storehouse, is set when cloud server receives user
Other side's voice of the standby certain time length gathered during receiving calls;Other side's voice is carried out into voice point
Analysis, obtains the phonetic feature point of other side's voice;By the phonetic feature point and the evil for pre-building of other side's voice
Phonetic feature point in meaning call voice property data base is matched;When the match is successful, set to user
The standby prompting for returning to malicious call;So that prompting of the user equipment according to the malicious call, on call circle
The prompting of the malicious call is shown on face.So that user hangs up the telephone malicious call in time, prevent from being taken in
Being deceived, therefore can improve user makes telephonic security.
Therefore, before the recognition methods of malicious call provided in an embodiment of the present invention is discussed in detail, this hair
The bright embodiment illustratively collocation method in the phonetic feature storehouse of malicious call first:
Fig. 1 is a kind of collocation method in the phonetic feature storehouse of malicious call of the offer of the embodiment of the present application
Schematic flow sheet, as shown in figure 1, including:
101st, during receiving calls, phone label option is shown on call interface;
The malicious call of the present embodiment for example includes intermediary's phone, sale call, fraudulent call, harassing and wrecking electricity
Words or self-defined phone etc., it is corresponding, in the present embodiment, corresponding electricity is provided with to each malicious call
Words label, wherein, phone label option for example including the label of intermediary's phone, the label of sale call,
The option of the label of fraudulent call, the label of harassing call or self-defined phone label.
102nd, when user selects any label according to other side's voice in phone label option, other side is come
Electricity is designated the malicious call corresponding to the label of user's selection;
Specifically, for example, when user determines that incoming call is intermediary's phone according to other side's voice, can be in phone
Intermediary's phone label is selected in label option, other side's incoming identification is designated intermediary's phone;And for example,
When user determines that incoming call is fraudulent call according to other side's voice, swindle can be selected in phone label option
Phone label, fraudulent call is designated by other side's incoming identification, and differ an illustration here.
103rd, the voice to malicious call carries out the recording of certain time length;
Alternatively, the voice of malicious call is carried out in step 103 after the recording of certain time length, is also wrapped
Include:According to the phone label that user selects, user's selection is added in the voice of the malicious call recorded
Phone label;
Alternatively, the voice of malicious call is carried out in step 103 after the recording of certain time length, is also wrapped
Include:Caller ID is added in the voice of the malicious call recorded.
104th, the voice of malicious call is sent to cloud server;
105th, cloud server carries out speech analysis to the voice of malicious call, obtains the language of malicious call
Sound characteristic point;
Specifically, phonetic feature point can be carried out using speech recognition technology when step 105 is implemented
Extract, specifically include:
Obtain the power spectrum of the voice of the malicious call;
The power spectrum that will be obtained obtains the short-time rating spectrum of noisy speech by filtering process;
Short-time rating spectrum according to the noisy speech for being obtained, noisy speech is asked for by the way of frame is average
Medium duration power spectrum;
The medium duration power spectrum of the noisy speech to being obtained carries out asymmetric filtering and shelters anti-noise, with
Obtain the medium duration power spectrum of clean speech;
The medium duration power spectrum of medium duration power spectrum, noisy speech according to the clean speech and contain
Make an uproar voice short-time rating spectrum obtain clean speech short-time rating spectrum;
Short-time rating spectrum to clean speech carries out energy normalized treatment, to eliminate multiplying property noise;
Short-time rating spectrum to eliminating the clean speech of multiplying property noise the loudness such as carries out and aggravates
The short-time rating of the clean speech after reciprocity loudness exacerbation is composed into row index nonlinear operation;
Short-time rating to having carried out the clean speech after index nonlinear operation is composed and carries out Fourier inversion
Change, to ask for cepstrum coefficient;
Mean normalization treatment is carried out to cepstrum coefficient, the phonetic feature point of the malicious call is obtained.
Speech recognition technology of the present invention any can realize that phonetic feature point lifts with existing
Speech recognition technology, the present invention this is not limited in any way.
106th, the phonetic feature point of malicious call is saved in malicious call voice feature data storehouse.
In a kind of optional implementation method, if carrying the malice electricity in the voice of the malicious call
The label of words;Then step 106 includes when implementing:
The label carried in voice according to the malicious call, sets up the phonetic feature of the malicious call
Put the corresponding relation with the label;
The phonetic feature point of the malicious call is saved in malicious call with the corresponding relation of the label
Voice feature data storehouse.
In a kind of optional implementation method, if carrying the malice electricity in the voice of the malicious call
The caller ID of words;Then step 106 includes when implementing:
The caller ID carried in voice according to the malicious call, sets up the voice of the malicious call
The corresponding relation of characteristic point and the caller ID;
The phonetic feature point of the malicious call and the corresponding relation of the caller ID are saved in malice
Call voice property data base.
In a kind of optional implementation method, cloud server receives the malicious call that user equipment sends
After voice, can also be by the speech recognition of the malicious call into corresponding text information;Set up described
Corresponding relation between the phonetic feature point of the text information of malicious call and the malicious call, and by institute
The corresponding relation stated between the text information of malicious call and the phonetic feature point of the malicious call is preserved
To in malicious call voice feature data storehouse.
In a kind of optional implementation method, the embodiment of the present invention also includes:
During receiving calls, other side's voice of certain time length is gathered;
Other side's voice is sent to cloud server, so that other side's voice is carried out voice point by cloud server
Analysis, obtains the phonetic feature point of other side's voice, and the phonetic feature point of other side's voice is white with what is pre-build
Phonetic feature point in list voice feature data storehouse is matched, and when matching unsuccessful, returns to other side
Phone whether be malicious call selection prompting;
Whether it is the selection prompting of malicious call according to what cloud server fed back, if detecting user's selection
During malicious call option, the malicious call option that user selects is sent to cloud server, so that high in the clouds
Be saved in the phonetic feature point of other side's voice in malicious call voice feature data storehouse by server.
Wherein, the foundation in the white list voice feature data storehouse for pre-building is, for example, as follows:User equipment
Contact phone in white list is received, the call voice of contact person in white list is gathered and is sent to high in the clouds clothes
Business device, cloud server carries out phonetic feature point extraction to the call voice of contact person in white list, will be white
The phonetic feature point of the call voice of contact person is saved in white list voice feature data storehouse in list.
It follows that the collocation method in the phonetic feature storehouse of the malicious call for providing according to embodiments of the present invention
Included but is not limited in the malicious call voice feature data storehouse for obtaining:The caller ID of multiple malicious calls
Information, the phonetic feature point information of multiple malicious call, the label information of the voice of multiple malicious calls,
The corresponding text information of voice of multiple malicious calls;
Further, the collocation method in the phonetic feature storehouse of malicious call provided in an embodiment of the present invention is obtained
Malicious call voice feature data storehouse in can also include:The caller ID and malicious call of malicious call
Phonetic feature point between corresponding relation, the corresponding relation of the phonetic feature point of malicious call and label,
The phonetic feature point of malicious call and the corresponding relation of text information, the caller ID and word of malicious call
The corresponding relation of information, the corresponding relation of the label of malicious call and text information, the incoming call of malicious call
One-to-one relationship, the caller ID of malicious call between number and phonetic feature point and label, language
Sound characteristic point, the one-to-one relationship between label and text information.
Technical scheme is described in detail below by way of specific embodiment.
Fig. 2 is a kind of schematic flow sheet of the recognition methods of malicious call of the offer of the embodiment of the present application,
As shown in Fig. 2 including:
201st, cloud server receives the right of the certain time length that user equipment is gathered during receiving calls
Fang Yuyin;
202nd, other side's voice is carried out into speech analysis, obtains the phonetic feature point of other side's voice;
Specifically, step 202 for example includes when realizing:
Obtain the power spectrum of other side's voice;
The power spectrum that will be obtained obtains the short-time rating spectrum of noisy speech by filtering process;
Short-time rating spectrum according to the noisy speech for being obtained, noisy speech is asked for by the way of frame is average
Medium duration power spectrum;
The medium duration power spectrum of the noisy speech to being obtained carries out asymmetric filtering and shelters anti-noise, with
Obtain the medium duration power spectrum of clean speech;
The medium duration power spectrum of medium duration power spectrum, noisy speech according to the clean speech and contain
Make an uproar voice short-time rating spectrum obtain clean speech short-time rating spectrum;
Short-time rating spectrum to clean speech carries out energy normalized treatment, to eliminate multiplying property noise;
Short-time rating spectrum to eliminating the clean speech of multiplying property noise the loudness such as carries out and aggravates
The short-time rating of the clean speech after reciprocity loudness exacerbation is composed into row index nonlinear operation;
Short-time rating to having carried out the clean speech after index nonlinear operation is composed and carries out Fourier inversion
Change, to ask for cepstrum coefficient;
Mean normalization treatment is carried out to cepstrum coefficient, the phonetic feature point of other side's voice is obtained.
203rd, by the phonetic feature point of other side's voice and the malicious call voice feature data storehouse for pre-building
In phonetic feature point matched;
Specifically, step 202 for example includes when realizing:
By in the phonetic feature point of other side's voice and the malicious call voice feature data storehouse for pre-building
When phonetic feature point is matched, if the feature point number being consistent is more than or equal to default threshold value,
Can determine that the match is successful, otherwise it fails to match.
204th, when the match is successful, the prompting of malicious call is returned to user equipment;So that user equipment
According to the prompting of the malicious call, the prompting of the malicious call is shown on call interface.
If the prompting of the malicious call includes label information, user equipment is according to the malicious call
The label information that includes of prompting, the corresponding malicious call of the label information is shown on call interface
Label.For example, the prompting of the malicious call of cloud server feedback includes fraudulent call label, then exist
The prompting of fraudulent call is shown as on call interface;Again for example, the malicious call of cloud server feedback
Prompting includes advertisement telephone label, then the prompting of advertisement telephone is shown as on call interface.
The evil that the collocation method in the phonetic feature storehouse based on malicious call provided in an embodiment of the present invention is obtained
Meaning call voice property data base, gathers when cloud server receives user equipment during receiving calls
Certain time length other side's voice;Other side's voice is carried out into speech analysis, the voice for obtaining other side's voice is special
Levy a little;By in the phonetic feature point of other side's voice and the malicious call voice feature data storehouse for pre-building
Phonetic feature point is matched;When the match is successful, the prompting of malicious call is returned to user equipment;With
Make user equipment according to the prompting of the malicious call, carrying for the malicious call is shown on call interface
Show.So that user hangs up the telephone malicious call in time, preventing from having dust thrown into the eyes, therefore can improve user makes
Telephonic security.
Fig. 3 is a kind of structural representation of the identifying device of malicious call of the offer of the embodiment of the present application,
Positioned at cloud server side, as shown in figure 3, including:
Receiver module 31, for receiving the certain time length that user equipment is gathered during receiving calls
Other side's voice;
Speech analysis module 32, for other side's voice to be carried out into speech analysis, obtains the language of other side's voice
Sound characteristic point;
Matching module 33, for by the phonetic feature point of other side's voice and the malicious call language for pre-building
Phonetic feature point in sound property data base is matched;Specifically for by the phonetic feature point of other side's voice
When being matched with the phonetic feature point in the malicious call voice feature data storehouse for pre-building, if being consistent
When the feature point number of conjunction is more than or equal to default threshold value, then can determine that the match is successful, otherwise matching is lost
Lose;
Sending module 34, for when the match is successful, the prompting of malicious call being returned to user equipment;
So that prompting of the user equipment according to the malicious call, shows the malicious call on call interface
Prompting.
The receiver module 31, is additionally operable to receive the voice for being designated malicious call that user equipment sends,
It is described be designated malicious call voice be user equipment during receiving calls, according to user instruction,
It is malicious call by other side's incoming identification when user's selection malicious call option is detected, and gather
The voice of the malicious call of certain time length;
The speech analysis module 32, is additionally operable to carry out speech analysis to the voice of malicious call, obtains
The phonetic feature point of malicious call;
Described device also includes:Preserving module 35, for the phonetic feature point of the malicious call to be protected
It is stored in malicious call voice feature data storehouse.
If carrying the label of the malicious call in the voice of the malicious call;The malicious call
The label of label including intermediary's phone, the label of sale call, the label of fraudulent call, harassing call
Label or self-defined phone label;
Described device also includes:
Module 36 is set up, for the label carried in the voice according to the malicious call, sets up described
The phonetic feature point of malicious call and the corresponding relation of the label;
The preserving module 35, is additionally operable to the phonetic feature point and the label of the malicious call
Corresponding relation is saved in malicious call voice feature data storehouse.
If carrying the caller ID of the malicious call in the voice of the malicious call;It is described to set up mould
Block 36, is additionally operable to the caller ID carried in the voice according to the malicious call, sets up the malice
The phonetic feature point of phone and the corresponding relation of the caller ID;
The preserving module 35, is additionally operable to the phonetic feature point of the malicious call and the incoming call number
The corresponding relation of code is saved in malicious call voice feature data storehouse.
The speech analysis module 32 specifically for:
Obtain the power spectrum of the voice of the malicious call;
The power spectrum that will be obtained obtains the short-time rating spectrum of noisy speech by filtering process;
Short-time rating spectrum according to the noisy speech for being obtained, noisy speech is asked for by the way of frame is average
Medium duration power spectrum;
The medium duration power spectrum of the noisy speech to being obtained carries out asymmetric filtering and shelters anti-noise, with
Obtain the medium duration power spectrum of clean speech;
The medium duration power spectrum of medium duration power spectrum, noisy speech according to the clean speech and contain
Make an uproar voice short-time rating spectrum obtain clean speech short-time rating spectrum;
Short-time rating spectrum to clean speech carries out energy normalized treatment, to eliminate multiplying property noise;
Short-time rating spectrum to eliminating the clean speech of multiplying property noise the loudness such as carries out and aggravates
The short-time rating of the clean speech after reciprocity loudness exacerbation is composed into row index nonlinear operation;
Short-time rating to having carried out the clean speech after index nonlinear operation is composed and carries out Fourier inversion
Change, to ask for cepstrum coefficient;
Mean normalization treatment is carried out to cepstrum coefficient, the phonetic feature point of the malicious call is obtained.
The speech analysis module 32, is additionally operable to the speech recognition of the malicious call into corresponding text
Word information;
It is described to set up module 36, it is additionally operable to set up the text information and malice electricity of the malicious call
Corresponding relation between the phonetic feature point of words.
The malicious call voice feature data storehouse includes:The caller ID information of multiple malicious calls,
The phonetic feature point information of multiple malicious calls, the label information of the voice of multiple malicious calls, multiple evils
The corresponding text information of voice of meaning phone;
Also include in the malicious call voice feature data storehouse:The caller ID of malicious call and malice electricity
The phonetic feature point of corresponding relation, malicious call between the phonetic feature point of words is corresponding with label to close
System, the corresponding relation of the phonetic feature point of malicious call and text information, the caller ID of malicious call and
The corresponding relation of text information, the corresponding relation of the label of malicious call and text information, malicious call
One-to-one relationship, the incoming call number of malicious call between caller ID and phonetic feature point and label
Code, phonetic feature point, the one-to-one relationship between label and text information.
The prompting of the malicious call includes the label information of the malicious call, so that user equipment root
According to the label information in the prompting of the malicious call, the mark is shown on the call interface of user equipment
Sign corresponding malicious call.
Fig. 3 shown devices can perform the method described in Fig. 1 and embodiment illustrated in fig. 2, and it realizes former
Reason and technique effect are repeated no more.
1a, a kind of recognition methods of malicious call, perform positioned at cloud server side, including:
Cloud server receives other side's language of the certain time length that user equipment is gathered during receiving calls
Sound;
Other side's voice is carried out into speech analysis, the phonetic feature point of other side's voice is obtained;
By in the phonetic feature point of other side's voice and the malicious call voice feature data storehouse for pre-building
Phonetic feature point is matched;
When the match is successful, the prompting of malicious call is returned to user equipment;So that user equipment is according to institute
The prompting of malicious call is stated, the prompting of the malicious call is shown on call interface.
2a, the method as described in 1a, also include:
Cloud server receives the voice for being designated malicious call that user equipment sends, described to be designated evil
Anticipate phone voice be user equipment during receiving calls, according to user instruction, detecting user
Selection malicious call option when, by other side's incoming identification be malicious call, and gather certain time length evil
The voice of meaning phone;
Voice to malicious call carries out speech analysis, obtains the phonetic feature point of malicious call;
The phonetic feature point of the malicious call is saved in malicious call voice feature data storehouse.
3a, the method as described in 2a, if carrying the malicious call in the voice of the malicious call
Label;The label of the malicious call includes label, the label of sale call, the swindle of intermediary's phone
The label of phone, the label of harassing call or self-defined phone label;
Then the voice to malicious call carries out speech analysis, after obtaining the phonetic feature point of malicious call,
Including:
The label carried in voice according to the malicious call, sets up the phonetic feature of the malicious call
Put the corresponding relation with the label;
The phonetic feature point of the malicious call is saved in malicious call with the corresponding relation of the label
Voice feature data storehouse.
4a, the method as described in 2a, if carrying the malicious call in the voice of the malicious call
Caller ID;
Then the voice to malicious call carries out speech analysis, after obtaining the phonetic feature point of malicious call,
Including:
The caller ID carried in voice according to the malicious call, sets up the voice of the malicious call
The corresponding relation of characteristic point and the caller ID;
The phonetic feature point of the malicious call and the corresponding relation of the caller ID are saved in malice
Call voice property data base.
5a, the method as described in 2a, the voice to malicious call carry out speech analysis, obtain malice electricity
The phonetic feature point of words, including:
Obtain the power spectrum of the voice of the malicious call;
The power spectrum that will be obtained obtains the short-time rating spectrum of noisy speech by filtering process;
Short-time rating spectrum according to the noisy speech for being obtained, noisy speech is asked for by the way of frame is average
Medium duration power spectrum;
The medium duration power spectrum of the noisy speech to being obtained carries out asymmetric filtering and shelters anti-noise, with
Obtain the medium duration power spectrum of clean speech;
The medium duration power spectrum of medium duration power spectrum, noisy speech according to the clean speech and contain
Make an uproar voice short-time rating spectrum obtain clean speech short-time rating spectrum;
Short-time rating spectrum to clean speech carries out energy normalized treatment, to eliminate multiplying property noise;
Short-time rating spectrum to eliminating the clean speech of multiplying property noise the loudness such as carries out and aggravates
The short-time rating of the clean speech after reciprocity loudness exacerbation is composed into row index nonlinear operation;
Short-time rating to having carried out the clean speech after index nonlinear operation is composed and carries out Fourier inversion
Change, to ask for cepstrum coefficient;
Mean normalization treatment is carried out to cepstrum coefficient, the phonetic feature point of the malicious call is obtained.
6a, the method as described in 5a, cloud server receive the language of the malicious call that user equipment sends
After sound, also include:
By the speech recognition of the malicious call into corresponding text information;
Set up right between the text information of the malicious call and the phonetic feature point of the malicious call
Should be related to.
7a, the method as any one of 2a-6a, the malicious call voice feature data storehouse Zhong Bao
Include:Phonetic feature point information, the multiple of the caller ID information of multiple malicious calls, multiple malicious calls
The corresponding text information of voice of the label information of the voice of malicious call, multiple malicious calls;
Also include in the malicious call voice feature data storehouse:The caller ID of malicious call and malice electricity
The phonetic feature point of corresponding relation, malicious call between the phonetic feature point of words is corresponding with label to close
System, the corresponding relation of the phonetic feature point of malicious call and text information, the caller ID of malicious call and
The corresponding relation of text information, the corresponding relation of the label of malicious call and text information, malicious call
One-to-one relationship, the incoming call number of malicious call between caller ID and phonetic feature point and label
Code, phonetic feature point, the one-to-one relationship between label and text information.
8a, the method as described in 7a, the prompting of the malicious call include the mark of the malicious call
Label information, so that label information of the user equipment in the prompting of the malicious call, in user equipment
Call interface on show malicious call corresponding to the label.
9b, a kind of identifying device of malicious call, positioned at cloud server side, including:
Receiver module, the other side for receiving the certain time length that user equipment is gathered during receiving calls
Voice;
Speech analysis module, for other side's voice to be carried out into speech analysis, the voice for obtaining other side's voice is special
Levy a little;
Matching module, for the phonetic feature point of other side's voice is special with the malicious call voice for pre-building
The point of the phonetic feature in database is levied to be matched;
Sending module, for when the match is successful, the prompting of malicious call being returned to user equipment;So that
User equipment shows carrying for the malicious call according to the prompting of the malicious call on call interface
Show.
10b, the device as described in 9b, also include:
The receiver module, is additionally operable to receive the voice for being designated malicious call that user equipment sends, institute
State be designated malicious call voice be user equipment during receiving calls, according to user instruction,
Detect user selection malicious call option when, by other side's incoming identification be malicious call, and gather one
The voice of timing malicious call long;
The speech analysis module, is additionally operable to carry out speech analysis to the voice of malicious call, obtains malice
The phonetic feature point of phone;
Preserving module, for the phonetic feature point of the malicious call to be saved in into malicious call phonetic feature
In database.
11b, the device as described in 10b, if carrying the malice electricity in the voice of the malicious call
The label of words;The label of the malicious call includes label, the label of sale call, the swindleness of intermediary's phone
Deceive the label of phone, the label of harassing call or self-defined phone label;
Described device also includes:
Module is set up, for the label carried in the voice according to the malicious call, the malice is set up
The phonetic feature point of phone and the corresponding relation of the label;
The preserving module, is additionally operable to the phonetic feature point of the malicious call is corresponding with the label
Relation is saved in malicious call voice feature data storehouse.
12b, such as 10b described devices, if carrying the malicious call in the voice of the malicious call
Caller ID;
It is described to set up module, the caller ID carried in the voice according to the malicious call is additionally operable to, build
Found the phonetic feature point of the malicious call and the corresponding relation of the caller ID;
The preserving module, is additionally operable to the phonetic feature point and the caller ID of the malicious call
Corresponding relation is saved in malicious call voice feature data storehouse.
13b, the device as described in 10b, the speech analysis module specifically for:
Obtain the power spectrum of the voice of the malicious call;
The power spectrum that will be obtained obtains the short-time rating spectrum of noisy speech by filtering process;
Short-time rating spectrum according to the noisy speech for being obtained, noisy speech is asked for by the way of frame is average
Medium duration power spectrum;
The medium duration power spectrum of the noisy speech to being obtained carries out asymmetric filtering and shelters anti-noise, with
Obtain the medium duration power spectrum of clean speech;
The medium duration power spectrum of medium duration power spectrum, noisy speech according to the clean speech and contain
Make an uproar voice short-time rating spectrum obtain clean speech short-time rating spectrum;
Short-time rating spectrum to clean speech carries out energy normalized treatment, to eliminate multiplying property noise;
Short-time rating spectrum to eliminating the clean speech of multiplying property noise the loudness such as carries out and aggravates
The short-time rating of the clean speech after reciprocity loudness exacerbation is composed into row index nonlinear operation;
Short-time rating to having carried out the clean speech after index nonlinear operation is composed and carries out Fourier inversion
Change, to ask for cepstrum coefficient;
Mean normalization treatment is carried out to cepstrum coefficient, the phonetic feature point of the malicious call is obtained.
14b, the device as described in 13b:
The speech analysis module, is additionally operable to believe the speech recognition of the malicious call into corresponding word
Breath;
It is described to set up module, it is additionally operable to set up the text information and the malicious call of the malicious call
Corresponding relation between phonetic feature point.
15b, the device as any one of 10b-14b, the malicious call voice feature data storehouse
Include:The caller ID information of multiple malicious calls, the phonetic feature point information of multiple malicious calls,
The corresponding text information of voice of the label information of the voice of multiple malicious calls, multiple malicious calls;
Also include in the malicious call voice feature data storehouse:The caller ID of malicious call and malice electricity
The phonetic feature point of corresponding relation, malicious call between the phonetic feature point of words is corresponding with label to close
System, the corresponding relation of the phonetic feature point of malicious call and text information, the caller ID of malicious call and
The corresponding relation of text information, the corresponding relation of the label of malicious call and text information, malicious call
One-to-one relationship, the incoming call number of malicious call between caller ID and phonetic feature point and label
Code, phonetic feature point, the one-to-one relationship between label and text information.
16b, the device as described in 15b, the prompting of the malicious call include the malicious call
Label information, so that label information of the user equipment in the prompting of the malicious call, sets in user
The malicious call corresponding to the label is shown on standby call interface.
Described above has shown and described some preferred embodiments of the invention, but as it was previously stated, should manage
The solution present invention is not limited to form disclosed herein, is not to be taken as the exclusion to other embodiment,
And can be used for various other combinations, modification and environment, and can in invention contemplated scope described herein,
It is modified by the technology or knowledge of above-mentioned teaching or association area.And those skilled in the art are carried out changes
Dynamic and change does not depart from the spirit and scope of the present invention, then all should be in the protection of appended claims of the present invention
In the range of.
Claims (10)
1. a kind of recognition methods of malicious call, performs positioned at cloud server side, it is characterised in that
Including:
Cloud server receives other side's language of the certain time length that user equipment is gathered during receiving calls
Sound;
Other side's voice is carried out into speech analysis, the phonetic feature point of other side's voice is obtained;
By in the phonetic feature point of other side's voice and the malicious call voice feature data storehouse for pre-building
Phonetic feature point is matched;
When the match is successful, the prompting of malicious call is returned to user equipment;So that user equipment is according to institute
The prompting of malicious call is stated, the prompting of the malicious call is shown on call interface.
2. the method for claim 1, it is characterised in that also include:
Cloud server receives the voice for being designated malicious call that user equipment sends, described to be designated evil
Anticipate phone voice be user equipment during receiving calls, according to user instruction, detecting user
Selection malicious call option when, by other side's incoming identification be malicious call, and gather certain time length evil
The voice of meaning phone;
Voice to malicious call carries out speech analysis, obtains the phonetic feature point of malicious call;
The phonetic feature point of the malicious call is saved in malicious call voice feature data storehouse.
3. method as claimed in claim 2, it is characterised in that if in the voice of the malicious call
Carry the label of the malicious call;The label of the malicious call includes the label of intermediary's phone, pushes away
Sell label, the label of fraudulent call, the label of harassing call or the self-defined phone label of phone;
Then the voice to malicious call carries out speech analysis, after obtaining the phonetic feature point of malicious call,
Including:
The label carried in voice according to the malicious call, sets up the phonetic feature of the malicious call
Put the corresponding relation with the label;
The phonetic feature point of the malicious call is saved in malicious call with the corresponding relation of the label
Voice feature data storehouse.
4. method as claimed in claim 2, it is characterised in that cloud server receives user equipment
After the voice of the malicious call of transmission, also include:
By the speech recognition of the malicious call into corresponding text information;
Set up right between the text information of the malicious call and the phonetic feature point of the malicious call
Should be related to.
5. the method as any one of claim 2-4, it is characterised in that the malicious call
Voice feature data storehouse includes:The language of the caller ID information of multiple malicious calls, multiple malicious calls
Sound characteristic point information, the label information of the voice of multiple malicious calls, the voice correspondence of multiple malicious calls
Text information;
Also include in the malicious call voice feature data storehouse:The caller ID of malicious call and malice electricity
The phonetic feature point of corresponding relation, malicious call between the phonetic feature point of words is corresponding with label to close
System, the corresponding relation of the phonetic feature point of malicious call and text information, the caller ID of malicious call and
The corresponding relation of text information, the corresponding relation of the label of malicious call and text information, malicious call
One-to-one relationship, the incoming call number of malicious call between caller ID and phonetic feature point and label
Code, phonetic feature point, the one-to-one relationship between label and text information.
6. a kind of identifying device of malicious call, positioned at cloud server side, it is characterised in that including:
Receiver module, the other side for receiving the certain time length that user equipment is gathered during receiving calls
Voice;
Speech analysis module, for other side's voice to be carried out into speech analysis, the voice for obtaining other side's voice is special
Levy a little;
Matching module, for the phonetic feature point of other side's voice is special with the malicious call voice for pre-building
The point of the phonetic feature in database is levied to be matched;
Sending module, for when the match is successful, the prompting of malicious call being returned to user equipment;So that
User equipment shows carrying for the malicious call according to the prompting of the malicious call on call interface
Show.
7. device as claimed in claim 6, it is characterised in that also include:
The receiver module, is additionally operable to receive the voice for being designated malicious call that user equipment sends, institute
State be designated malicious call voice be user equipment during receiving calls, according to user instruction,
Detect user selection malicious call option when, by other side's incoming identification be malicious call, and gather one
The voice of timing malicious call long;
The speech analysis module, is additionally operable to carry out speech analysis to the voice of malicious call, obtains malice
The phonetic feature point of phone;
Preserving module, for the phonetic feature point of the malicious call to be saved in into malicious call phonetic feature
In database.
8. device as claimed in claim 7, it is characterised in that if being taken in the voice of the malicious call
Label with the malicious call;The label of the malicious call includes the label of intermediary's phone, promotes
The label of phone, the label of fraudulent call, the label of harassing call or self-defined phone label;
Described device also includes:
Module is set up, for the label carried in the voice according to the malicious call, the malice is set up
The phonetic feature point of phone and the corresponding relation of the label;
The preserving module, is additionally operable to the phonetic feature point of the malicious call is corresponding with the label
Relation is saved in malicious call voice feature data storehouse.
9. device as claimed in claim 2, it is characterised in that:
The speech analysis module, is additionally operable to believe the speech recognition of the malicious call into corresponding word
Breath;
It is described to set up module, it is additionally operable to set up the text information and the malicious call of the malicious call
Corresponding relation between phonetic feature point.
10. device as claimed in any one of claims 7-9, it is characterised in that the malicious call
Voice feature data storehouse includes:The language of the caller ID information of multiple malicious calls, multiple malicious calls
Sound characteristic point information, the label information of the voice of multiple malicious calls, the voice correspondence of multiple malicious calls
Text information;
Also include in the malicious call voice feature data storehouse:The caller ID of malicious call and malice electricity
The phonetic feature point of corresponding relation, malicious call between the phonetic feature point of words is corresponding with label to close
System, the corresponding relation of the phonetic feature point of malicious call and text information, the caller ID of malicious call and
The corresponding relation of text information, the corresponding relation of the label of malicious call and text information, malicious call
One-to-one relationship, the incoming call number of malicious call between caller ID and phonetic feature point and label
Code, phonetic feature point, the one-to-one relationship between label and text information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510810521.8A CN106790950A (en) | 2015-11-20 | 2015-11-20 | The recognition methods of malicious call and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510810521.8A CN106790950A (en) | 2015-11-20 | 2015-11-20 | The recognition methods of malicious call and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106790950A true CN106790950A (en) | 2017-05-31 |
Family
ID=58886075
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510810521.8A Pending CN106790950A (en) | 2015-11-20 | 2015-11-20 | The recognition methods of malicious call and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106790950A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109547621A (en) * | 2018-12-11 | 2019-03-29 | 平安科技(深圳)有限公司 | A kind of information prompting method, device, electronic equipment and storage medium |
CN110351415A (en) * | 2019-06-26 | 2019-10-18 | 厦门快商通信息咨询有限公司 | Determination method, apparatus, equipment and the storage medium of harassing call number |
CN110602302A (en) * | 2019-08-15 | 2019-12-20 | 厦门快商通科技股份有限公司 | Voice intercepting method and device for telephone robot and storage medium |
CN110619893A (en) * | 2019-09-02 | 2019-12-27 | 合肥工业大学 | Time-frequency feature extraction and artificial intelligence emotion monitoring method of voice signal |
CN111246008A (en) * | 2020-01-10 | 2020-06-05 | 三星电子(中国)研发中心 | Method, system and device for realizing telephone assistant |
-
2015
- 2015-11-20 CN CN201510810521.8A patent/CN106790950A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109547621A (en) * | 2018-12-11 | 2019-03-29 | 平安科技(深圳)有限公司 | A kind of information prompting method, device, electronic equipment and storage medium |
CN110351415A (en) * | 2019-06-26 | 2019-10-18 | 厦门快商通信息咨询有限公司 | Determination method, apparatus, equipment and the storage medium of harassing call number |
CN110602302A (en) * | 2019-08-15 | 2019-12-20 | 厦门快商通科技股份有限公司 | Voice intercepting method and device for telephone robot and storage medium |
CN110619893A (en) * | 2019-09-02 | 2019-12-27 | 合肥工业大学 | Time-frequency feature extraction and artificial intelligence emotion monitoring method of voice signal |
CN111246008A (en) * | 2020-01-10 | 2020-06-05 | 三星电子(中国)研发中心 | Method, system and device for realizing telephone assistant |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106790950A (en) | The recognition methods of malicious call and device | |
US20200251117A1 (en) | Method of defending against inaudible attacks on voice assistant based on machine learning | |
CN106686191A (en) | Processing method for adaptively identifying harassing call and processing system thereof | |
CN106782495A (en) | The recognition methods of malicious call and device | |
US9837079B2 (en) | Methods and apparatus for identifying fraudulent callers | |
CN106790949A (en) | The collocation method and device in the phonetic feature storehouse of malicious call | |
US20190325345A1 (en) | Bot-based data collection for detecting phone solicitations | |
CN103685185B (en) | Mobile equipment voiceprint registration, the method and system of certification | |
Li et al. | Behaviour profiling on mobile devices | |
CN102045712A (en) | Anti-monitoring method and system for mobile terminal | |
CN101252748A (en) | System and method for realizing privacy SMS on mobile terminal | |
Wang et al. | Ghosttalk: Interactive attack on smartphone voice system through power line | |
CN102231277A (en) | Method for protecting mobile terminal privacy based on voiceprint recognition | |
CN106453061A (en) | Method and system for recognizing internet fraud behavior | |
CN106850931A (en) | The method and mobile intelligent terminal of Barassment preventing telephone | |
CN101674353A (en) | Incoming call filter method | |
CN107770777B (en) | Method for identifying recorded fraud calls | |
CN102316203A (en) | Method for preventing crank call based on cloud computing, device and terminal | |
CN105472093B (en) | Mobile terminal call answers method for authenticating and mobile terminal | |
US9571640B1 (en) | Systems and methods for detecting calls from illegitimate calling parties | |
CN104410973B (en) | A kind of fraudulent call recognition methods of playback and system | |
CN104023110A (en) | Voiceprint recognition-based caller management method and mobile terminal | |
CN104184653B (en) | A kind of method and apparatus of message screening | |
Giura et al. | Is it really you? User identification via adaptive behavior fingerprinting | |
CN109688273B (en) | Reminding method, reminding device, terminal and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170531 |