CN106341539A - Automatic evidence obtaining method of malicious caller voiceprint, apparatus and mobile terminal thereof - Google Patents

Abstract

The present invention provides a method and device for automatically obtaining evidence of the voiceprint of a malicious caller, which determines that the incoming number corresponding to the incoming call is a malicious call; obtains the voice data of the caller during the call, and extracts the voiceprint of the caller from the voice data Features: judging whether the caller's voiceprint feature has been stored in the malicious voiceprint database, if not stored, then storing the caller's voiceprint feature in the malicious voiceprint database. In this way, as long as the incoming call is identified as a malicious call, the caller's voiceprint features will be stored in the malicious voiceprint database, and the malicious caller will be identified when he calls the user again with another phone number, prompting the user in time, improving This increases security and saves user time. The invention also provides a mobile terminal.

Description

Automatic evidence collection method, device and mobile terminal of voiceprint of malicious caller

technical field

The present invention relates to the technical field of call security, in particular, the present invention relates to a method, device and mobile terminal for automatic evidence collection of voiceprints of malicious callers.

Background technique

In recent years, malicious telephone behaviors such as telephone fraud, telemarketing, and telephone harassment have become more and more rampant, and telecom users have been disturbed. Some mobile phone security software, such as 360 Mobile Guard, will uniformly identify these phone numbers reported by many users as malicious calls as malicious numbers, and prompt the called user that the phone number is a malicious number when the phone number is called, making the called party Users can effectively distinguish and avoid being cheated or wasting time, which ensures user safety and saves user time to a certain extent. However, if these malicious callers change their phone numbers and make malicious phone calls again, since the phone number was not effectively identified at the beginning, the called user has the risk of being cheated and will waste user time, security and efficiency Both need to be improved.

Contents of the invention

The purpose of the present invention is to at least solve one of the above-mentioned technical defects, especially the technical defect of poor safety.

The present invention provides a method for automatically obtaining evidence of a voiceprint of a malicious caller, comprising the following steps:

Determine that the incoming call number corresponding to the incoming call is a malicious call;

Obtain the voice data of the caller during the call, and extract the voiceprint features of the caller from the voice data;

It is judged whether the voiceprint feature of the caller has been stored in the malicious voiceprint database, and if not stored, the voiceprint feature of the caller is stored in the malicious voiceprint database.

In one of the embodiments, the process of determining that the incoming number corresponding to the incoming call is a malicious incoming call includes:

Obtain the incoming call number corresponding to the incoming call;

Determine that the incoming number is a malicious call, and determine its malicious type.

In one embodiment, the malicious types are more than one, including at least one of fraudulent, promotional, and harassment types.

In one of the embodiments, there are more than one malicious voiceprint databases, each malicious voiceprint database corresponds to a respective malicious type, and each malicious voiceprint database stores pre-existing voiceprint features corresponding to malicious types.

In one of the embodiments, the pre-stored voiceprint feature is associated with a phone number.

In one of the embodiments, the judging whether the caller's voiceprint features have been stored in the malicious voiceprint database includes:

Acquiring the pre-stored voiceprint features associated with the caller number from the malicious voiceprint database, and judging whether the caller's voiceprint features match the pre-stored voiceprint features.

In one of the embodiments, the process of determining that the incoming number corresponding to the incoming call is a malicious incoming call includes:

Judging whether the incoming call number has been stored in the malicious number blacklist, and if so, determining that the corresponding incoming call is a malicious incoming call.

In one of the embodiments, the blacklist of malicious numbers is stored locally or on a server.

In one of the embodiments, the caller's voiceprint is characterized by a sound wave spectrum carrying speech information with characteristic words or phrases.

In one of the embodiments, the process of acquiring the voice data of the caller during the call, and extracting the voiceprint features of the caller from the voice data includes:

When detecting that a caller generates a voice stream, record and store at least one piece of voice data;

Analyzing the at least one piece of voice data to extract the voiceprint feature of the caller.

In one of the embodiments, the analyzing the at least one piece of voice data to extract the voiceprint feature of the caller includes:

Analyzing the at least one piece of speech data, extracting at least one set of characteristic words or words in the speech data;

The voiceprint features of the caller are extracted according to the feature words or words.

In one of the embodiments, the voice data is voice data with a set duration.

In one embodiment, after the voiceprint feature of the caller is successfully extracted, or after it is confirmed that the voiceprint feature of the caller cannot be extracted, the voice data is deleted.

In one of the embodiments, storing the voiceprint features of the caller in the malicious voiceprint database includes:

Storing the caller's voiceprint feature mapped with the caller number into the malicious voiceprint database.

In one of the embodiments, the malicious voiceprint database is established on a server.

The present invention also provides an automatic evidence collection device for the voiceprint of a malicious caller, which includes:

An identification module, configured to determine that the incoming number corresponding to the incoming call is a malicious incoming call;

The analysis module is used to obtain the voice data of the caller during the call, and extract the voiceprint feature of the caller from the voice data;

The storage module is used to judge whether the voiceprint feature of the caller has been stored in the malicious voiceprint database, and if not, store the voiceprint feature of the caller in the malicious voiceprint database.

In one of the embodiments, the identification module is used for:

Obtain the incoming call number corresponding to the incoming call;

Determine that the incoming number is a malicious call, and determine its malicious type.

In one embodiment, the malicious types are more than one, including at least one of fraudulent, promotional, and harassment types.

In one of the embodiments, there are more than one malicious voiceprint databases, each malicious voiceprint database corresponds to a respective malicious type, and each malicious voiceprint database stores pre-existing voiceprint features corresponding to malicious types.

In one of the embodiments, the pre-stored voiceprint feature is associated with a phone number.

In one of the embodiments, the storage module is used for:

Acquiring the pre-stored voiceprint features associated with the caller number from the malicious voiceprint database, and judging whether the caller's voiceprint features match the pre-stored voiceprint features.

In one of the embodiments, the identification module is used for:

Judging whether the incoming call number has been stored in the malicious number blacklist, and if so, determining that the corresponding incoming call is a malicious incoming call.

In one of the embodiments, the blacklist of malicious numbers is stored locally or on a server.

In one of the embodiments, the caller's voiceprint is characterized by a sound wave spectrum carrying speech information with characteristic words or phrases.

In one of the embodiments, the analysis module is used for:

When detecting that a caller generates a voice stream, record and store at least one piece of voice data;

Analyzing the at least one piece of voice data to extract the voiceprint feature of the caller.

In one of the embodiments, the analyzing the at least one piece of voice data to extract the voiceprint feature of the caller includes:

Analyzing the at least one piece of speech data, extracting at least one set of characteristic words or words in the speech data;

The voiceprint features of the caller are extracted according to the feature words or words.

In one of the embodiments, the voice data is voice data with a set duration.

In one of the embodiments, the device further includes a deletion module; the deletion module is used to delete the caller's voiceprint feature after the analysis module successfully extracts the caller's voiceprint feature, or after confirming that the caller's voiceprint feature cannot be extracted. voice data.

In one of the embodiments, the storage module is used for:

Storing the caller's voiceprint feature mapped with the caller number into the malicious voiceprint database.

In one of the embodiments, the malicious voiceprint database is established on a server.

The present invention also provides a mobile terminal, which includes:

touch-sensitive display;

one or more processors;

memory;

one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the one or more processors, the one or more program programs are configured to: execute The automatic forensics method for the voiceprint of a malicious caller described in any one of the embodiments.

The above-mentioned automatic evidence collection method, device and mobile terminal for the voiceprint of a malicious caller determine that the incoming number corresponding to the incoming call is a malicious call; obtain the voice data of the caller during the call, and extract the voiceprint of the caller from the voice data Features: judging whether the caller's voiceprint feature has been stored in the malicious voiceprint database, if not stored, then storing the caller's voiceprint feature in the malicious voiceprint database. In this way, as long as the incoming call is identified as a malicious call, the caller's voiceprint features will be stored in the malicious voiceprint database, and the malicious caller will be identified when he calls the user again with another phone number, prompting the user in time, improving This increases security and saves user time.

Moreover, by collecting the voiceprint characteristics of the malicious caller, the malicious caller will be prompted as a malicious call even if the normal daily call is made, which can effectively curb the malicious caller from making malicious phone calls. If the incoming call of the malicious caller is no longer reported as a malicious call by the user within a preset investigation period, the voiceprint feature of the caller in the malicious voiceprint database can also be deleted.

Additional aspects and advantages of the invention will be set forth in part in the description which follows, and will become apparent from the description, or may be learned by practice of the invention.

Description of drawings

The above and/or additional aspects and advantages of the present invention will become apparent and easy to understand from the following description of the embodiments in conjunction with the accompanying drawings, wherein:

Fig. 1 is the flow chart of the method for automatically obtaining evidence of a malicious caller's voiceprint of an embodiment;

Fig. 2 is the block diagram of the automatic forensic collection device of the malicious caller's voiceprint of an embodiment;

Fig. 3 shows a block diagram of a partial structure of a mobile phone related to a terminal provided by an embodiment of the present invention.

detailed description

Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary only for explaining the present invention and should not be construed as limiting the present invention.

Those skilled in the art will understand that unless otherwise stated, the singular forms "a", "an", "said" and "the" used herein may also include plural forms. It should be further understood that the word "comprising" used in the description of the present invention refers to the presence of said features, integers, steps, operations, elements and/or components, but does not exclude the presence or addition of one or more other features, Integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Additionally, "connected" or "coupled" as used herein may include wireless connection or wireless coupling. The expression "and/or" used herein includes all or any elements and all combinations of one or more associated listed items.

Those skilled in the art can understand that, unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this invention belongs. It should also be understood that terms, such as those defined in commonly used dictionaries, should be understood to have meanings consistent with their meaning in the context of the prior art, and unless specifically defined as herein, are not intended to be idealized or overly Formal meaning to explain.

Those skilled in the art can understand that the "terminal" and "terminal equipment" used here not only include wireless signal receiver equipment, which only has wireless signal receiver equipment without transmission capabilities, but also include receiving and transmitting hardware. A device having receiving and transmitting hardware capable of performing bi-directional communication over a bi-directional communication link. Such equipment may include: cellular or other communication equipment, which has a single-line display or a multi-line display or a cellular or other communication equipment without a multi-line display; PCS (Personal Communications Service, personal communication system), which can combine voice, data Processing, facsimile and/or data communication capabilities; PDA (Personal Digital Assistant, Personal Digital Assistant), which may include radio frequency receiver, pager, Internet/Intranet access, web browser, notepad, calendar and/or GPS (Global Positioning System (Global Positioning System) receiver; a conventional laptop and/or palmtop computer or other device having and/or including a radio frequency receiver. As used herein, a "terminal", "terminal device" may be portable, transportable, installed in a vehicle (air, sea, and/or land), or adapted and/or configured to operate locally, and/or In distributed form, the operation operates at any other location on Earth and/or in space. The "terminal" and "terminal equipment" used here can also be a communication terminal, an Internet terminal, a music/video player terminal, such as a PDA, a MID (Mobile Internet Device, a mobile Internet device) and/or a music/video player Functional mobile phones, smart TVs, set-top boxes and other devices.

Those skilled in the art can understand that the remote network device used here includes, but is not limited to, a computer, a network host, a single network server, a set of multiple network servers, or a cloud formed by multiple servers. Here, the cloud is composed of a large number of computers or network servers based on cloud computing (Cloud Computing), wherein cloud computing is a kind of distributed computing, a super virtual computer composed of a group of loosely coupled computer sets. In the embodiment of the present invention, the communication between the remote network equipment, the terminal equipment and the WNS server can be realized through any communication method, including but not limited to, mobile communication based on 3GPP, LTE, WIMAX, based on TCP/IP, UDP protocol Computer network communication and short-distance wireless transmission based on bluetooth and infrared transmission standards.

The method and device for automatically obtaining evidence of a malicious caller's voiceprint described below can be applied to mobile terminals, such as mobile phones and tablet computers, in short, terminals with communication functions. In the following description, terminals are used as examples.

FIG. 1 is a flow chart of an embodiment of an automatic forensics method for a voiceprint of a malicious caller.

The present invention provides a method for automatically obtaining evidence of a voiceprint of a malicious caller, comprising the following steps:

Step S100: Determine that the incoming number corresponding to the incoming call is a malicious incoming call.

A blacklist of malicious numbers can be pre-established on the local machine or server. When the terminal receives an incoming call, it obtains the incoming call number and judges whether the incoming call number has been stored in the malicious number blacklist. If so, it determines that the corresponding incoming call is a malicious call. If the blacklist of malicious numbers is stored in the local machine, it will be identified locally; if the blacklist of malicious numbers is stored in the server, it will be identified on the server, and the server will feed back the identification result to the terminal.

The blacklist of malicious numbers can be a blacklist established by the user. For example, when the user receives some malicious calls, the phone number is added to the blacklist. At this time, the blacklist of malicious numbers can be stored locally, and the user can also share it. The malicious numbers are blacklisted to the server. Of course, the blacklist of malicious numbers can also be a blacklist established by the server. For example, the server blacklists the phone numbers reported by many users. At this time, the blacklist of malicious numbers can be stored on the server, and the user can also request to obtain it from the server. The blacklist of malicious numbers.

The malicious number blacklist can be a blacklist, or can be divided into multiple malicious number lists according to malicious types. For example, malicious types can be divided into fraud types, sales types, harassment types, etc., and malicious number blacklists can be divided into fraud blacklists, sales blacklists, and harassment blacklists. The terminal recognizes which malicious number blacklist the incoming call number belongs to, and then prompts accordingly. For example, if it is identified that the number of the incoming call belongs to the fraud blacklist, the user is prompted that the incoming call is a fraudulent call. Therefore, when the malicious type is further subdivided, the terminal can obtain the corresponding incoming call number of the incoming call, determine that the incoming call number is a malicious incoming call, and determine its malicious type. The specific process is: after the terminal obtains the corresponding incoming call number of the incoming call, it compares the incoming call number in multiple blacklists of malicious numbers. The phone number is a malicious caller number, and its malicious type is determined.

Step S200: Obtain the voice data of the caller during the call, and extract the voiceprint features of the caller from the voice data.

Because each person's vocal organs, such as vocal cords, oral cavity, nasal cavity, tongue, teeth, lips, lungs, etc., present in various poses and with different expressions when they pronounce. Due to various reasons such as age, personality, language habits, etc., coupled with different pronunciation volumes and pronunciation frequencies, even small differences will cause the sounds produced by these organs to have their own characteristics, thus forming each Individual's unique voiceprint (Voiceprint) can be observed by spectrogram.

Voiceprint recognition is the process of identifying the identity of the speaker from a certain segment of voice. Similar to fingerprints, the speech features and pronunciation habits of each person's speech are almost unique. Speech recognition is generic recognition, which determines what is said (what is said). Voiceprint recognition is personality recognition, which determines the identity of the speaker (who said it).

There are two key issues in voiceprint recognition, one is feature extraction, and the other is pattern recognition.

The task of feature extraction is to extract and select acoustic or language features that have strong separability and high stability for the speaker's voiceprint. Although most voiceprint recognition systems currently use features at the acoustic level, the features that characterize a person's characteristics should be multi-level, including: (1) acoustic features related to the anatomical structure of the human pronunciation mechanism (such as Spectrum, cepstrum, formant, pitch, reflection coefficient, etc.), nasal, breathy, hoarse, laughter, etc.; (2) Semantics, rhetoric, Pronunciation, speech habits, etc.; (3) Personal characteristics or characteristics such as rhythm, rhythm, speed, intonation, and volume influenced by parents. From the perspective of using mathematical methods to model, the current features that can be used in the voiceprint automatic recognition model include: (1) acoustic features (cepstrum); (2) lexical features (speaker-related word n-gram, phoneme n -gram); (3) prosodic features (using the pitch and energy "posture" described by n-gram); (4) language, dialect and accent information; (5) channel information (which channel to use); and so on.

For pattern recognition, there are mainly these categories of methods: (1) Template matching method: use dynamic time warping (DTW) to align training and test feature sequences, mainly for the application of fixed phrases (usually text-related tasks) ; (2) Nearest neighbor method: keep all feature vectors during training, and find the nearest K training vectors for each vector during recognition, and identify them accordingly. Usually, the amount of model storage and similar calculation is very large; (3 ) Neural network method: There are many forms, such as multi-layer perception, radial basis function (RBF), etc., which can be explicitly trained to distinguish speakers from their background speakers. The amount of training is large, and the model can be generalized Not good; (4) Hidden Markov Model (HMM) method: Usually single-state HMM, or Gaussian Mixture Model (GMM), is a more popular method, and the effect is better; (5) VQ clustering method ( Such as LBG, K-means): the effect is better, the algorithm complexity is not high, and it can receive better results when combined with the HMM method; (6) polynomial classifier method: has higher accuracy, but the model storage And the amount of calculation is relatively large.

The main point of the template matching method is to extract the corresponding feature vector from each speaker's training sentence to describe the behavior of each speaker during the training process; in the test phase, use the same method from the speaker's test voice signal There are mainly dynamic time warping methods and vector quantization methods for extracting test templates. In the following description, the template matching method is taken as an example.

Since each person's voiceprint is unique, malicious callers can be identified by analyzing the caller's voice to obtain the caller's voiceprint features. The caller's voiceprint features carry speech information with characteristic words or words Sound spectrum. For example, when detecting that the caller is speaking (generating a voice stream), the terminal can record and store at least one piece of voice data; then analyze at least one piece of voice data to extract the voiceprint feature of the caller. The voice data can be voice data with a set duration, for example, the set duration is 10 seconds, and the recording starts when it is detected that the caller is speaking, and the recording is 10 seconds.

Obtaining the voice data of the caller during the call, and extracting the voiceprint features of the caller from the voice data can be achieved through the following process: parsing at least one piece of voice data, extracting at least one group of feature words or words in the voice data; Or words to extract the voiceprint features of the caller.

Feature words or words are pre-set, this is because usually scammers, salesmen, harassers will say specific words or words in the call, such as "hello", "hello", "excuse me", "you "Yes", "Mr", "Ms", "What" and other characteristic words or words, by identifying these characteristic words or words in the voice data, the efficiency of voiceprint recognition can be effectively improved. For example, the sentence "Hello, are you Mr. Tang Bohu" is recognized in the voice clip, and the terminal extracts "Hello", "Excuse me", "Are you", "Sir", "are you" These five groups of feature words or words, and then extract the voiceprint features of the caller according to these feature words or words.

If there is no voiceprint feature corresponding to the first voiceprint feature of the caller acquired this time in the following malicious voiceprint database, after storing the first voiceprint feature of the caller this time, if The caller uses another phone number to call other users again, and usually also speaks the above-mentioned characteristic words or words, then the terminal can extract the second voiceprint feature of the caller according to the characteristic words or words, and the caller's first voice The fingerprint feature is compared with the second voiceprint feature of the caller to identify malicious calls.

The acquired voice data will be stored or temporarily stored in the storage medium of the terminal. When the voiceprint feature of the caller is successfully extracted, or after confirming that the voiceprint feature of the caller cannot be extracted, the terminal can delete the voice segment to save terminal storage space.

Step S300: Determine whether the caller's voiceprint features have been stored in the malicious voiceprint database, and if not, store the caller's voiceprint features in the malicious voiceprint database.

After extracting the caller's voiceprint features of the caller, the terminal judges whether the caller's voiceprint features have been stored in the malicious voiceprint database. The malicious voiceprint database stores pre-stored voiceprint features associated with phone numbers. These pre-stored voiceprint features are collected and stored by the server in many previous malicious calls. After collection, the pre-stored voiceprint features and corresponding phone numbers Establish a good mapping relationship. Therefore, the malicious voiceprint database is established on the server.

There may be more than one malicious voiceprint database, for example, there may be more than one malicious voiceprint database, and each malicious voiceprint database corresponds to its own malicious type. For example, there are more than one malicious type, which can be divided into fraud type, sales type, harassment type, etc., then the malicious voiceprint database can be divided into fraud database, sales database, and harassment database. When the terminal obtains the phone number of a malicious call, in the above-mentioned process of identifying a malicious call, the malicious type can be determined according to the phone number, and then when judging whether the caller's voiceprint features have been stored in the malicious voiceprint database, The determined malicious type can be matched and judged in the corresponding malicious voiceprint database, which improves the efficiency.

Therefore, judging whether the voiceprint feature of the caller has been stored in the malicious voiceprint database can be realized through the following process: obtain the pre-stored voiceprint feature associated with the caller number from the malicious voiceprint database, and determine whether the voiceprint feature of the caller is consistent with Pre-stored voiceprint feature matching.

If it is divided into multiple malicious voiceprint databases corresponding to the malicious type according to the malicious type, when the terminal stores the voiceprint feature of the caller, it will be stored in the corresponding malicious voiceprint database according to the malicious type. When storing, the voiceprint feature of the caller that has a mapping relationship with the caller number should be stored in the malicious voiceprint database, that is, the voiceprint feature of the caller and the caller number are stored after the mapping relationship is established. Since the malicious voiceprint database is established on the server, after the terminal acquires the voiceprint features of the caller, it uploads the voiceprint features of the caller to the server for storage.

Of course, if the malicious caller no longer conducts malicious phone calls, the voiceprint feature of the malicious caller can be deleted from the malicious voiceprint database. For example, if the investigation period is preset (for example, 2 months), if the caller’s voiceprint features are no longer collected during the investigation period, it means that the malicious caller’s calls will no longer be reported as malicious by the user within the investigation period. If there is an incoming call, the voiceprint feature of the caller in the malicious voiceprint database can also be deleted.

The above-mentioned automatic evidence collection method for the voiceprint of a malicious caller determines that the incoming number corresponding to the incoming call is a malicious call; obtains the voice data of the caller during the call, and extracts the voiceprint characteristics of the caller from the voice data; Whether the voiceprint feature of the caller has been stored in the malicious voiceprint database, if not stored, the voiceprint feature of the caller is stored in the malicious voiceprint database. In this way, as long as the incoming call is identified as a malicious call, the caller's voiceprint features will be stored in the malicious voiceprint database, and the malicious caller will be identified when he calls the user again with another phone number, prompting the user in time, improving This increases security and saves user time.

Moreover, by collecting the voiceprint characteristics of the malicious caller, the malicious caller will be prompted as a malicious call even if the normal daily call is made, which can effectively curb the malicious caller from making malicious phone calls. If the incoming call of the malicious caller is no longer reported as a malicious call by the user within a preset investigation period, the voiceprint feature of the caller in the malicious voiceprint database can also be deleted.

The malicious voiceprint database formed through the participation of many users, in addition to identifying and prompting the user through the voiceprint characteristics of the caller when a malicious call is made, can also have the following expansion functions. For example, with the continuous development of the real-name system for SIM card processing, the follow-up "ID card-mobile number" database (DB1) system will be improved day by day. (DB2) improvement, through the mutual access of the two databases, the following data query platforms can be implemented in the future:

1. Fraud record inquiry platform:

When a fraudster commits a fraudulent act by phone, the cloud (server) can query the mobile phone number for identity authentication through the voiceprint feature of the database DB2, and then confirm the fraudster's true identity through the query of the database DB1.

2. Financial credit inquiry platform:

Combining databases DB1 and DB2, a complete credit evaluation platform can be integrated to jointly establish a system guarantee for maintaining social integrity.

A device corresponding to the above-mentioned automatic evidence collection method for the voiceprint of a malicious caller is described below, and the device is applied to a mobile terminal. Fig. 2 is a block diagram of an automatic forensic collection device for voiceprints of malicious callers according to an embodiment. An automatic evidence collection device for voiceprints of malicious callers, which includes: an identification module 100 , an analysis module 200 and a storage module 300 .

The identification module 100 is used to determine that the incoming number corresponding to the incoming call is a malicious call; the analysis module 200 is used to obtain the voice data of the caller during the call, and extract the voiceprint characteristics of the caller from the voice data; the storage module 300 judges Whether the voiceprint feature of the caller has been stored in the malicious voiceprint database, if not stored, the voiceprint feature of the caller is stored in the malicious voiceprint database.

The identification module 100 determines that the incoming call number corresponding to the incoming call is a malicious incoming call.

A blacklist of malicious numbers can be pre-established on the local machine or server. When the terminal receives an incoming call, the identification module 100 obtains the incoming call number to determine whether the incoming call number has been stored in the malicious number blacklist, and if so, determines that the corresponding incoming call is a malicious call . If the blacklist of malicious numbers is stored in the local machine, it will be identified locally; if the blacklist of malicious numbers is stored in the server, it will be identified on the server, and the server will feed back the identification result to the terminal.

The blacklist of malicious numbers can be a blacklist established by the user. For example, when the user receives some malicious calls, the phone number is added to the blacklist. At this time, the blacklist of malicious numbers can be stored locally, and the user can also share it. The malicious numbers are blacklisted to the server. Of course, the blacklist of malicious numbers can also be a blacklist established by the server. For example, the server blacklists the phone numbers reported by many users. At this time, the blacklist of malicious numbers can be stored on the server, and the user can also request to obtain it from the server. The blacklist of malicious numbers.

The malicious number blacklist can be a blacklist, or can be divided into multiple malicious number lists according to malicious types. For example, malicious types can be divided into fraud types, sales types, harassment types, etc., and malicious number blacklists can be divided into fraud blacklists, sales blacklists, and harassment blacklists. The terminal recognizes which malicious number blacklist the incoming call number belongs to, and then prompts accordingly. For example, if it is identified that the number of the incoming call belongs to the fraud blacklist, the user is prompted that the incoming call is a fraudulent call. Therefore, when subdividing the malicious type, the identification module 100 can acquire the corresponding incoming call number, determine that the incoming call number is a malicious incoming call, and determine its malicious type. The specific process is: after the identification module 100 obtains the corresponding incoming call number of the incoming call, the incoming call number is compared in a plurality of blacklists of malicious numbers. Then determine that the phone number is a malicious caller number, and determine its malicious type.

The analysis module 200 acquires the voice data of the caller during the call, and extracts the voiceprint features of the caller from the voice data.

Because each person's vocal organs, such as vocal cords, oral cavity, nasal cavity, tongue, teeth, lips, lungs, etc., present in various poses and with different expressions when they pronounce. Due to various reasons such as age, personality, language habits, etc., coupled with different pronunciation volumes and pronunciation frequencies, even small differences will cause the sounds produced by these organs to have their own characteristics, thus forming each Individual's unique voiceprint (Voiceprint) can be observed by spectrogram.

Voiceprint recognition is the process of identifying the identity of the speaker from a certain segment of voice. Similar to fingerprints, the speech features and pronunciation habits of each person's speech are almost unique. Speech recognition is generic recognition, which determines what is said (what is said). Voiceprint recognition is personality recognition, which determines the identity of the speaker (who said it).

There are two key issues in voiceprint recognition, one is feature extraction, and the other is pattern recognition.

The task of feature extraction is to extract and select acoustic or language features that have strong separability and high stability for the speaker's voiceprint. Although most voiceprint recognition systems currently use features at the acoustic level, the features that characterize a person's characteristics should be multi-level, including: (1) acoustic features related to the anatomical structure of the human pronunciation mechanism (such as Spectrum, cepstrum, formant, pitch, reflection coefficient, etc.), nasal, breathy, hoarse, laughter, etc.; (2) Semantics, rhetoric, Pronunciation, speech habits, etc.; (3) Personal characteristics or characteristics such as rhythm, rhythm, speed, intonation, and volume influenced by parents. From the perspective of using mathematical methods to model, the current features that can be used in the voiceprint automatic recognition model include: (1) acoustic features (cepstrum); (2) lexical features (speaker-related word n-gram, phoneme n -gram); (3) prosodic features (using the pitch and energy "posture" described by n-gram); (4) language, dialect and accent information; (5) channel information (which channel to use); and so on.

For pattern recognition, there are mainly these categories of methods: (1) Template matching method: use dynamic time warping (DTW) to align training and test feature sequences, mainly for the application of fixed phrases (usually text-related tasks) ; (2) Nearest neighbor method: keep all feature vectors during training, and find the nearest K training vectors for each vector during recognition, and identify them accordingly. Usually, the amount of model storage and similar calculation is very large; (3 ) Neural network method: There are many forms, such as multi-layer perception, radial basis function (RBF), etc., which can be explicitly trained to distinguish speakers from their background speakers. The amount of training is large, and the model can be generalized Not good; (4) Hidden Markov Model (HMM) method: Usually single-state HMM, or Gaussian Mixture Model (GMM), is a more popular method, and the effect is better; (5) VQ clustering method ( Such as LBG, K-means): the effect is better, the algorithm complexity is not high, and it can receive better results when combined with the HMM method; (6) polynomial classifier method: has higher accuracy, but the model storage And the amount of calculation is relatively large.

The main point of the template matching method is to extract the corresponding feature vector from each speaker's training sentence to describe the behavior of each speaker during the training process; in the test phase, use the same method from the speaker's test voice signal There are mainly dynamic time warping methods and vector quantization methods for extracting test templates. In the following description, the template matching method is taken as an example.

Since each person's voiceprint is unique, the analysis module 200 can identify malicious callers by analyzing the voice of the caller to obtain the voiceprint features of the caller. The acoustic spectrum of speech information. For example, when the terminal detects that the caller is speaking (generates a voice stream), it can record and store at least one piece of voice data; then the analysis module 200 analyzes the at least one piece of voice data to extract the voiceprint feature of the caller. The voice data can be voice data with a set duration, for example, the set duration is 10 seconds, and the recording starts when it is detected that the caller is speaking, and the recording is 10 seconds.

The analysis module 200 acquires the voice data of the caller during the call, and extracting the voiceprint features of the caller from the voice data can be realized through the following process: the analysis module 200 parses at least one section of voice data, and obtains at least one set of characteristic words or Word; Acquire the voiceprint feature of the caller according to the feature word or word.

Feature words or words are pre-set, this is because usually scammers, salesmen, harassers will say specific words or words in the call, such as "hello", "hello", "excuse me", "you "Yes", "Mr", "Ms", "What" and other characteristic words or words, by identifying these characteristic words or words in the voice data, the efficiency of voiceprint recognition can be effectively improved. For example, the sentence "Hello, are you Mr. Tang Bohu" is recognized in the voice data, and the terminal extracts "Hello", "Excuse me", "Are you", "Sir", "are you?" These five groups of feature words or words, and then extract the voiceprint features of the caller according to these feature words or words.

If there is no voiceprint feature corresponding to the first voiceprint feature of the caller acquired this time in the following malicious voiceprint database, after storing the first voiceprint feature of the caller this time, if The caller uses another phone number to call other users again, and usually also speaks the above-mentioned characteristic words or words, then the terminal can extract the second voiceprint feature of the caller according to the characteristic words or words, and the caller's first voice The fingerprint feature is compared with the second voiceprint feature of the caller to identify malicious calls.

The apparatus can also include a deletion module. The acquired voice data will be stored or temporarily stored in the storage medium of the terminal. When the voiceprint feature of the caller is successfully extracted, or after confirming that the voiceprint feature of the caller cannot be extracted, the deletion module can delete the voice data to save the terminal storage.

The storage module 300 judges whether the voiceprint feature of the caller has been stored in the malicious voiceprint database, and if not stored, stores the voiceprint feature of the caller in the malicious voiceprint database.

After the analysis module 200 extracts the caller's voiceprint features of the caller, the storage module 300 judges whether the caller's voiceprint features have been stored in the malicious voiceprint database. The malicious voiceprint database stores pre-stored voiceprint features associated with phone numbers. These pre-stored voiceprint features are collected and stored by the server in many previous malicious calls. After collection, the pre-stored voiceprint features and corresponding phone numbers Establish a good mapping relationship. Therefore, the malicious voiceprint database is established on the server.

There may be more than one malicious voiceprint database, for example, there may be more than one malicious voiceprint database, and each malicious voiceprint database corresponds to its own malicious type. For example, there are more than one malicious type, which can be divided into fraud type, sales type, harassment type, etc., then the malicious voiceprint database can be divided into fraud database, sales database, and harassment database. When the identification module 100 obtains the phone number of a malicious call, in the above-mentioned process of identifying a malicious call, the analysis module 200 can determine the malicious type according to the phone number, and then the storage module 300 determines whether the voiceprint feature of the caller has been stored. When in the malicious voiceprint database, the storage module 300 can match and judge the determined malicious type in the corresponding malicious voiceprint database, which improves the efficiency.

Therefore, the analysis module 200 can determine whether the voiceprint feature of the caller has been stored in the malicious voiceprint database can be realized through the following process: obtain the pre-stored voiceprint feature associated with the caller number from the malicious voiceprint database, and determine the caller's voiceprint Whether the features match the pre-stored voiceprint features.

If it is divided into a plurality of malicious voiceprint databases corresponding to malicious types according to malicious types, when storing the voiceprint features of the caller, it will be stored in the corresponding malicious voiceprint database according to its malicious type. When storing, the voiceprint feature of the caller that has a mapping relationship with the caller number should be stored in the malicious voiceprint database, that is, the voiceprint feature of the caller and the caller number are stored after the mapping relationship is established. Since the malicious voiceprint database is established on the server, the storage module 300 uploads the voiceprint features of the caller to the server for storage after acquiring the voiceprint features of the caller.

Of course, if the malicious caller no longer conducts malicious phone calls, the deletion module can delete the voiceprint feature of the malicious caller from the malicious voiceprint database. For example, if the investigation period is preset (for example, 2 months), if the caller’s voiceprint features are no longer collected during the investigation period, it means that the malicious caller’s calls will no longer be reported as malicious by the user within the investigation period. If there is an incoming call, the deletion module can also delete the voiceprint feature of the caller in the malicious voiceprint database.

In the above-mentioned automatic forensics device for the voiceprint of a malicious caller, the identification module determines that the incoming number corresponding to the incoming call is a malicious call; the analysis module obtains the voice data of the caller during the call, and extracts the voiceprint characteristics of the caller from the voice data ; The storage module judges whether the voiceprint feature of the caller has been stored in the malicious voiceprint database, and if not, stores the voiceprint feature of the caller in the malicious voiceprint database. In this way, as long as the incoming call is identified as a malicious call, the caller's voiceprint features will be stored in the malicious voiceprint database, and the malicious caller will be identified when he calls the user again with another phone number, prompting the user in time, improving This increases security and saves user time.

Moreover, by collecting the voiceprint characteristics of the malicious caller, the malicious caller will be prompted as a malicious call even if the normal daily call is made, which can effectively curb the malicious caller from making malicious phone calls. If the incoming call of the malicious caller is no longer reported as a malicious call by the user within a preset investigation period, the voiceprint feature of the caller in the malicious voiceprint database can also be deleted.

The embodiment of the present invention also provides a mobile terminal. As shown in FIG. 3 , for convenience of description, only the parts related to the embodiment of the present invention are shown. For specific technical details not disclosed, please refer to the method part of the embodiment of the present invention. The terminal can be any terminal device including mobile phone, tablet computer, PDA (Personal Digital Assistant, personal digital assistant), POS (Point of Sales, sales terminal), vehicle-mounted computer, etc. Taking the terminal as a mobile phone as an example:

Fig. 3 shows a block diagram of a partial structure of a mobile phone related to a terminal provided by an embodiment of the present invention. Referring to FIG. 3 , the mobile phone includes: a radio frequency (Radio Frequency, RF) circuit 1510, a memory 1520, an input unit 1530, a display unit 1540, a sensor 1550, an audio circuit 1560, a wireless fidelity (wireless fidelity, WiFi) module 1570, and a processor 1580 , and power supply 1590 and other components. Those skilled in the art can understand that the structure of the mobile phone shown in FIG. 3 does not constitute a limitation to the mobile phone, and may include more or less components than shown in the figure, or combine some components, or arrange different components.

The following is a specific introduction to each component of the mobile phone in conjunction with Figure 3:

The RF circuit 1510 can be used for sending and receiving information or receiving and sending signals during a call. In particular, after receiving the downlink information from the base station, it is processed by the processor 1580; in addition, the designed uplink data is sent to the base station. Generally, the RF circuit 1510 includes, but is not limited to, an antenna, at least one amplifier, a transceiver, a coupler, a low noise amplifier (Low Noise Amplifier, LNA), a duplexer, and the like. In addition, RF circuitry 1510 may also communicate with networks and other devices via wireless communications. The above-mentioned wireless communication can use any communication standard or protocol, including but not limited to Global System of Mobile Communication (Global System of Mobile communication, GSM), General Packet Radio Service (General Packet Radio Service, GPRS), Code Division Multiple Access (Code Division Multiple Access) , CDMA), Wideband Code Division Multiple Access (WCDMA), Long Term Evolution (Long Term Evolution, LTE), email, Short Messaging Service (Short Messaging Service, SMS), etc.

The memory 1520 can be used to store software programs and modules, and the processor 1580 executes various functional applications and data processing of the mobile phone by running the software programs and modules stored in the memory 1520 . The memory 1520 can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, at least one application program required by a function (such as a voiceprint playback function, an image playback function, etc.) and the like; the storage data area can store Data created based on the use of the mobile phone (such as audio data, phonebook, etc.), etc. In addition, the memory 1520 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage devices.

The input unit 1530 can be used to receive input numbers or character information, and generate key signal input related to user settings and function control of the mobile phone. Specifically, the input unit 1530 may include a touch panel 1531 and other input devices 1532 . The touch panel 1531, also referred to as a touch screen, can collect touch operations of the user on or near it (for example, the user uses any suitable object or accessory such as a finger or a stylus on the touch panel 1531 or near the touch panel 1531). operation), and drive the corresponding connection device according to the preset program. Optionally, the touch panel 1531 may include two parts, a touch detection device and a touch controller. Among them, the touch detection device detects the user's touch orientation, and detects the signal brought by the touch operation, and transmits the signal to the touch controller; the touch controller receives the touch information from the touch detection device, converts it into contact coordinates, and sends it to the to the processor 1580, and can receive and execute commands sent by the processor 1580. In addition, the touch panel 1531 can be implemented in various types such as resistive, capacitive, infrared, and surface acoustic wave. In addition to the touch panel 1531 , the input unit 1530 may also include other input devices 1532 . Specifically, other input devices 1532 may include but not limited to one or more of a physical keyboard, function keys (such as volume control keys, switch keys, etc.), trackball, mouse, joystick, and the like.

The display unit 1540 may be used to display information input by or provided to the user and various menus of the mobile phone. The display unit 1540 may include a display panel 1541. Optionally, the display panel 1541 may be configured in the form of a liquid crystal display (Liquid Crystal Display, LCD), an organic light-emitting diode (Organic Light-Emitting Diode, OLED), or the like. Furthermore, the touch panel 1531 may cover the display panel 1541, and when the touch panel 1531 detects a touch operation on or near it, it transmits to the processor 1580 to determine the type of the touch event, and then the processor 1580 determines the type of the touch event according to the The type provides a corresponding visual output on the display panel 1541 . Although in FIG. 3 , the touch panel 1531 and the display panel 1541 are used as two independent components to realize the input and input functions of the mobile phone, in some embodiments, the touch panel 1531 and the display panel 1541 can be integrated to form a mobile phone. Realize the input and output functions of the mobile phone.

The handset may also include at least one sensor 1550, such as a light sensor, motion sensor, and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display panel 1541 according to the brightness of the ambient light, and the proximity sensor may turn off the display panel 1541 and/or when the mobile phone is moved to the ear. or backlight. As a kind of motion sensor, the accelerometer sensor can detect the magnitude of acceleration in various directions (generally three axes), and can detect the magnitude and direction of gravity when it is stationary, and can be used to identify the application of mobile phone posture (such as horizontal and vertical screen switching, related Games, magnetometer attitude calibration), vibration recognition related functions (such as pedometer, tap), etc.; as for other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc. repeat.

The audio circuit 1560, the speaker 1561, and the microphone 1562 can provide an audio interface between the user and the mobile phone. The audio circuit 1560 can transmit the electrical signal converted from the received audio data to the speaker 1561, and the speaker 1561 converts it into a voiceprint signal for output; The circuit 1560 converts the received audio data into audio data, and then outputs the audio data to the processor 1580 for processing, and sends the audio data to another mobile phone through the RF circuit 1510, or outputs the audio data to the memory 1520 for further processing.

WiFi is a short-distance wireless transmission technology. The mobile phone can help users send and receive emails, browse web pages, and access streaming media through the WiFi module 1570. It provides users with wireless broadband Internet access. Although Fig. 3 shows the WiFi module 1570, it can be understood that it is not an essential component of the mobile phone, and can be completely omitted as required without changing the essence of the invention.

The processor 1580 is the control center of the mobile phone. It uses various interfaces and lines to connect various parts of the entire mobile phone. By running or executing software programs and/or modules stored in the memory 1520, and calling data stored in the memory 1520, execution Various functions and processing data of the mobile phone, so as to monitor the mobile phone as a whole. Optionally, the processor 1580 may include one or more processing units; preferably, the processor 1580 may integrate an application processor and a modem processor, wherein the application processor mainly processes operating systems, user interfaces, and application programs, etc. , the modem processor mainly handles wireless communications. It can be understood that the foregoing modem processor may not be integrated into the processor 1580 .

The mobile phone also includes a power supply 1590 (such as a battery) for supplying power to various components. Preferably, the power supply can be logically connected to the processor 1580 through the power management system, so that functions such as charging, discharging, and power consumption management can be realized through the power management system.

Although not shown, the mobile phone may also include a camera, a Bluetooth module, etc., which will not be repeated here.

In the embodiment of the present invention, the processor 1580 included in the terminal also has the following functions: determine that the incoming call number corresponding to the incoming call is a malicious incoming call; obtain the voice data of the caller during the call, and extract the caller number from the voice data judge whether the voiceprint feature of the caller has been stored in the malicious voiceprint database, and if not, store the voiceprint feature of the caller in the malicious voiceprint database. That is to say, the processor 1580 has the function of executing the method for automatically obtaining evidence of a malicious caller's voiceprint in any of the above-mentioned embodiments, which will not be repeated here.

It should be understood that although the various steps in the flow chart of FIG. 1 are displayed sequentially as indicated by the arrows, these steps are not necessarily executed sequentially in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some of the steps in Figure 1 may include multiple sub-steps or multiple stages, these sub-steps or stages are not necessarily executed at the same time, but may be executed at different times, and the execution order is not necessarily performed sequentially, but may be performed alternately or alternately with at least a part of other steps or sub-steps or stages of other steps.

The above descriptions are only part of the embodiments of the present invention. It should be pointed out that those skilled in the art can make some improvements and modifications without departing from the principles of the present invention. It should be regarded as the protection scope of the present invention.

Claims (10)

1. a kind of automatic evidence-collecting method of malice caller's vocal print is it is characterised in that comprise the steps:

Determine that the corresponding caller ID of incoming call is that malice is sent a telegram here；

Obtain the speech data of caller in communication process, extract the vocal print feature of caller from described speech data；

Judging whether the vocal print feature of described caller has stored in malice voice print database, if not storing, storing institute The vocal print feature stating caller is in described malice voice print database.

2. the automatic evidence-collecting method of malice caller's vocal print according to claim 1 is it is characterised in that described determination is sent a telegram here Call the process that corresponding caller ID is malice incoming call to include:

Obtain the corresponding caller ID of incoming call；

Determine that caller ID is malice incoming call, and determine its malice type.

3. the automatic evidence-collecting method of malice caller's vocal print according to claim 2 is it is characterised in that described malice type More than one, including swindle type, promote at least one of type, harassing and wrecking type.

4. the automatic evidence-collecting method of malice caller's vocal print according to claim 3 is it is characterised in that described malice vocal print Database is more than one, and each malice voice print database corresponds to respective malice type, and each malice voice print database is stored with The vocal print feature that prestores of corresponding malice type.

5. the automatic evidence-collecting method of malice caller's vocal print according to claim 4 is it is characterised in that the described vocal print that prestores Feature is associated with telephone number.

6. the automatic evidence-collecting method of malice caller's vocal print according to claim 5 is it is characterised in that described in described judgement Whether the vocal print feature of caller has stored in malice voice print database includes:

Obtain, from described malice voice print database, the vocal print feature that prestores associating with described caller ID, judge described caller Vocal print feature whether mate with the described vocal print feature that prestores.

7. the automatic evidence-collecting method of malice caller's vocal print according to claim 1 is it is characterised in that described determination is sent a telegram here Call the process that corresponding caller ID is malice incoming call to include:

Judge whether described caller ID has stored in malice number blacklist, if then determining corresponding incoming call as malice Incoming call.

8. the automatic evidence-collecting method of malice caller's vocal print according to claim 7 is it is characterised in that described malice number Blacklist is stored in local or server.

9. a kind of automatic evidence-collecting device of malice caller's vocal print is it is characterised in that include:

Identification module, for determining that the corresponding caller ID of incoming call is that malice is sent a telegram here；

Analysis module, for obtaining the speech data of caller in communication process, extracts caller's from described speech data Vocal print feature；

Memory module, whether the vocal print feature for judging described caller has stored in malice voice print database, if not having Storage is had then to store the vocal print feature of described caller in described malice voice print database.

10. a kind of mobile terminal it is characterised in that comprising:

Touch-sensitive display；

One or more processors；

Memory；

One or more application programs, wherein said one or more application programs are stored in described memory and are configured It is by one or more of computing devices, one or more of programs are configured to: execution is according to claim 1～8 The automatic evidence-collecting method of the malice caller's vocal print described in any one.