Telephone fraud prevention and control method based on multi-mode voice modification and related equipment

Abstract

The invention provides a telephone fraud prevention and control method based on multi-mode voice modification and related equipment. The method comprises the following steps: step 1, receiving network data packets, and screening out RTP voice packets suspected of telephone fraud from the network data packets; 2, analyzing the RTP voice packet, and determining a voice coding type adopted by original effective load data in the RTP voice packet; 3, generating safety load data adopting the voice coding type, replacing original effective load data in the RTP voice packet with the safety load data, and assembling a new RTP voice packet; and step 4, the new RTP voice packet is sent back. According to the invention, the occurrence of network fraud events can be effectively prevented.

Description

Technical Field

[0001] The present invention relates to the field of communication technology, and in particular to a telephone fraud prevention and control method based on multimodal voice modification and related equipment. Background Art

[0002] Voice over Internet Protocol (VoIP) is a voice call technology that uses the Internet Protocol (IP) to achieve voice calls and multimedia conferences, that is, to communicate through the Internet. Specifically, VOIP refers to voice chats over IP networks. This voice chat method is different from the traditional telecom operator telephone network for network transmission calls. Instead, it converts the chat voice into RTP voice data packets based on the RTP protocol standard and then transmits it over the Internet. Among them, RTP is the Real-time Transport Protocol. RTP is used to provide end-to-end real-time transmission services for various multimedia data that need to be transmitted in real time on IP networks, such as voice, images, fax, etc. RTP is a protocol running on the transport layer, usually based on the UDP protocol, but also supports the TCP protocol. The RTP data packet consists of two parts, one is the RTP Header, which occupies a minimum of 12 bytes and a maximum of 72 bytes; the other part is the RTP Payload, which is used to encapsulate the actual data load.

[0003] VoIP can be used on many Internet-accessible devices, including VoIP phones, smartphones, and personal computers, to make calls and send text messages over cellular networks and Wi-Fi. However, with the development of Internet communications, there are more and more cases of Internet fraud through Internet calls. Summary of the invention

[0004] The present invention aims to provide a telephone fraud prevention and control method and related equipment based on multimodal voice modification. When it is known that the initiator of an Internet call is a fraudster, the voice data of the fraudster is modified, thereby protecting the user at the receiving end of the Internet call from the fraudster's deception, and ultimately effectively preventing the occurrence of Internet fraud incidents.

[0005] In a first aspect, the present invention provides a telephone fraud prevention and control method based on multimodal voice modification, comprising: Step 1: Receive network data packets and filter out RTP voice packets suspected of telephone fraud from the network data packets; Step 2: parsing the RTP voice packet to determine the voice coding type used by the original payload data in the RTP voice packet; Step 3: Generate security payload data using the voice coding type, use the security payload data to replace the original payload data in the RTP voice packet and assemble it into a new RTP voice packet; Step 4: Send back the new RTP voice packet.

[0006] Furthermore, step 2 specifically includes: The RTP voice packet is parsed, and the voice coding type adopted by the original payload data in the RTP voice packet is determined according to the payload type PT value obtained by the parsing.

[0007] Further, the voice coding type adopted by the original payload data in the RTP voice packet is determined according to the payload type PT value obtained by parsing, specifically including: If the payload type PT value is less than 96, the voice coding type used by the original payload data in the RTP voice packet is directly determined according to the payload type PT value; If the payload type PT value is greater than or equal to 96 and falls within the range of the four dynamic payload types of AMRNB, AMRWB, EVS and ILBC, the voice coding type used by the original payload data in the RTP voice packet is further determined according to the dynamic payload type judgment rule.

[0008] Furthermore, the dynamic load type judgment rule specifically includes: Determine the length of the original payload data in the RTP voice packet. If the length is 6 bytes, it is considered that the original payload data in the RTP voice packet adopts EVS voice encoding and is an EVS silent frame; if the length is 7 bytes, continue to determine whether the first four bits of the original payload data are all 1. If so, further determine according to the frame type FT value that the original payload data in the RTP voice packet adopts AMRNB voice encoding or AMRWB voice encoding, and is an AMRNB silent frame or AMRWB silent frame.

[0009] Furthermore, the dynamic load type judgment rule specifically includes: If it is a non-silent frame and its length is less than 32 bytes, it is further determined whether the first four bits of the original payload data are all 1. If not, it is considered that the original payload data in the RTP voice packet adopts EVS voice coding; if so, it is further determined according to the frame type FT value that the original payload data in the RTP voice packet adopts AMRNB voice coding or AMRWB voice coding; If the length is greater than 32 bytes, then continue to determine whether the length is equal to 38 bytes. If so, it is considered that the original payload data in the RTP voice packet adopts ILBC voice encoding; if not, then continue to determine whether the values ​​of the first four bits of the original payload data are all 1. If so, and the frame type FT value conforms to AMRWB voice encoding, then it is determined that the original payload data in the RTP voice packet adopts AMRWB voice encoding; otherwise, it is determined that the original payload data in the RTP voice packet adopts EVS voice encoding.

[0010] Furthermore, step 3 also includes selecting a modification mode, wherein the modification mode includes silent modification and non-silent modification; Correspondingly, step 3 specifically includes: If silence modification is selected, silence data of the voice coding type is generated as security payload data.

[0011] If non-silence modification is selected, audio data of the voice coding type is generated as security payload data.

[0012] Furthermore, step 2 also includes: Parsing the RTP voice packet to obtain basic feature information of the RTP voice packet and storing it in a database for record analysis; The original payload data in the RTP voice packet is audio-decoded according to the voice coding type, and the audio-decoded data is stored in a designated location.

[0013] Furthermore, in step 1, the network data packet is received using the original socket; correspondingly, in step 4, the new RTP voice packet is sent back using the original socket.

[0014] In a second aspect, the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the program.

[0015] In a third aspect, the present invention provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in the first aspect.

[0016] Beneficial effects of the present invention: (1) The present invention firstly selects RTP voice data packets suspected of telephone fraud from captured network data packets, and then accurately identifies the voice coding type of the payload of the RTP voice data packet, replaces the original payload data with the security payload data, and then sends the RTP voice packet back, thereby preventing the receiving end user from being deceived. The present invention can effectively prevent the occurrence of Internet telephone fraud incidents, thereby effectively regulating the behavior of Internet users and improving the level of digital social governance. It can be applied to occasions such as telephone fraud and other network public hazard detection and governance.

[0017] (2) Based on the accurate identification of the voice coding type of the payload data in the RTP voice data packet, the present invention can decode different types of audio such as PCMA, PCMU, G723, G729, G726, EVS, GSM, AMRNB, AMRWB, ILBC and other voice coding types, and then store the decoded audio to the corresponding location. At the same time, the basic information of the voice packet such as the five-tuple information is stored in the database for filing and analysis, providing a data analysis basis for the existing network fraud detection technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 One of the flow charts of a telephone fraud prevention and control method based on multimodal voice modification provided by an embodiment of the present invention; Figure 2 A schematic diagram of the composition of an RTP voice packet provided in an embodiment of the present invention; Figure 3 A schematic diagram of the RTP message format provided in an embodiment of the present invention; Figure 4 AMR audio format schematic diagram provided by an embodiment of the present invention; Figure 5 A schematic diagram of the AMRNB audio operation mode provided by an embodiment of the present invention; Figure 6 A schematic diagram of the AMRWB audio operation mode provided by an embodiment of the present invention; Figure 7 A flow chart of silent frame determination in a dynamic load determination function provided in an embodiment of the present invention; Figure 8 A flow chart of non-silent frame determination in a dynamic load determination function provided in an embodiment of the present invention; Fig. 9 A modification mode determination flow chart provided by an embodiment of the present invention; Fig.10 A schematic diagram of an audio modification process provided by an embodiment of the present invention; Fig.11 A schematic diagram of a mute modification process provided by an embodiment of the present invention; Fig.12 A second flowchart of a telephone fraud prevention and control method based on multimodal voice modification provided by an embodiment of the present invention; Fig.13 A structural block diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the embodiment of the present invention will be clearly described below in conjunction with the drawings in the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] like Figure 1 As shown, the embodiment of the present invention provides a telephone fraud prevention and control method based on multimodal voice modification, comprising the following steps: S101 receives network data packets through raw sockets and filters out RTP voice packets suspected of telephone fraud from the network data packets; Specifically, Figure 2 As shown in the figure, the RTP voice packet is composed of the Ethernet frame header, IP header, UDP header, RTP header and RTP payload. The RTP header and RTP payload are used as the payload part of the UDP message. Therefore, by analyzing the sequence of the RTP voice packet, the RTP voice packet can be screened out according to the characteristics of the IP message, TCP message and RTP message.

[0021] It should be noted that the embodiment of the present invention is carried out on the premise that the sender of the target RTP voice packet is determined to be a telephone fraud implementer. Therefore, when implementing this step, the existing telephone fraud detection method can be combined to determine whether the RTP voice packet is a voice packet suspected of telephone fraud. Since this process is not a key improvement point of the present invention, it will not be repeated here. In addition, it can be understood that the method of receiving the network data packet includes but is not limited to the original socket.

[0022] S102: parsing the RTP voice packet to determine the voice coding type used by the original payload data in the RTP voice packet; Specifically, Figure 3The figure shows the header format of the RTP protocol, which consists of the RTP protocol version number V, fill flag P, extension flag X, CSRC counter CC, mark M, payload type PT, sequence number, timestamp, synchronization source (SSRC) identifier and special source (CSRC) identifier. Among them, the payload type PT uniquely identifies the voice coding type (hereinafter referred to as voice type) of the media carried in the RTP network data packet. Among them, when the PT value is 0-95, the payload type is a fixed payload type, and when the PT value is 96-127, the payload type is a dynamic payload type.

[0023] Specifically, voice types such as PCMA, PCMU, G723, G729, G726, GSM, etc. all use fixed payload types, while voice types such as AMRNB, AMRWB, EVS, ILBC, etc. use dynamic payload types. Based on this, it can be understood that the multimodal voice in the present invention refers to the RTP voice packet using the above-mentioned multiple voice types, that is, the voice type of the received RTP voice packet is uncertain and dynamic.

[0024] Therefore, it can be understood that by parsing the RTP voice packet, the voice coding type used by the original payload data in the RTP voice packet can be determined according to the payload type PT value obtained by parsing. Specifically, if the payload type PT value is less than 96, the voice coding type used by the original payload data in the RTP voice packet is directly determined according to the payload type PT value; for example, when the PT value is 0, the voice coding type is PCMU; when the PT value is 8, the voice coding type is PCMA.

[0025] However, due to the existence of dynamic load type, if the specific voice type is simply judged according to the PT value in the RTP protocol header format, an erroneous result may be obtained. Therefore, when the PT value is greater than or equal to 96, it is also necessary to further determine the voice coding type adopted by the original effective load data in the RTP voice packet according to the dynamic load type judgment rule. In the present embodiment, the dynamic load type judgment rule is mainly constructed according to the header format characteristics of the specific voice type to achieve unique identification of the voice type.

[0026] In the embodiment of the present invention, dynamic load judgment is mainly performed on four voice types, namely, AMRNB, AMRWB, EVS, and ILBC. The premise of the following judgment function is that only the above four dynamic load voice types exist in the network.

[0027] In this embodiment, if Figure 4As shown, AMR audio consists of CMR (codec mode request), ToC (The table of contents), and SpeechData. ToC occupies 6 bits and consists of F, FT, and Q. CMR is the coding mode request field. CMR=15 means that the receiver does not need to encode or decode the data and can keep it as it is. In this embodiment, the CMR value of AMR audio is 15 (that is, the four bits occupied by this field are all 1). F is used to mark whether it is the last frame, 0 indicates that the frame is the last frame; FT represents the frame type; Q refers to the quality of the frame, 0 indicates that the corresponding frame is damaged, and 1 indicates that it is not damaged. If the frame has been damaged, it can be processed by directly discarding the frame.

[0028] Since the frame length of AMR audio is 20 milliseconds, according to Figure 5 and Figure 6 The rate matching table of AMRNB and AMRWB can be obtained as follows: ;

[0029] Mode refers to the rate mode of voice, which can be determined by FT according to the rate matching table. The rate of AMRNB or AMRWB is obtained through FT to determine a unique mode value. Byte is the total number of bytes of AMR audio, which is also the number of payload bytes of the RTP protocol.

[0030] From the above conclusions, it can be concluded that the silent frame (SID) of AMRNB and AMRWB audio is 7 bytes, and the maximum transmission byte of AMRNB is 32. In addition, the silent frame (SID) of EVS audio is 6 bytes, and ILBC is usually fixed to 38 bytes. Therefore, when these four dynamic loads exist at the same time, the above judgment principle can be used for judgment.

[0031] In this embodiment, the audio whose payload type PT value of the RTP voice packet is greater than or equal to 96 is first used as a judgment premise. The payload type PT value greater than or equal to 96 indicates that the voice type carried by the RTP voice packet is a dynamic payload type, that is, the dynamic payload judgment function needs to be triggered. Here, EVS, ILBC, AMRNB, and AMRWB are used as examples. When receiving RTP voice packets, they can be divided into two situations: silent frames and non-silent frames. Among them, if the length of the original payload data in the RTP voice packet is 6 bytes or 7 bytes, the silent frame judgment process is adopted; otherwise, the non-silent frame judgment process is adopted.

[0032] like Figure 7As shown in the figure, it is the dynamic load judgment process of the silent frame; for the judgment of the silent frame, it can be determined whether it conforms to the byte number of the AMR or EVS silent frame according to the byte number of the silent frame. If the byte number of the RTP payload length is 6, it can be determined that the frame is a silent frame of EVS voice. If the byte number of the RTP payload length is 7, it conforms to the silent frame length of AMR voice. Then, it is further determined according to the CMR field of AMR: that is, whether the first four bits of the RTP payload data are all 1. If it conforms, it is determined according to the AMR rule by the FT value whether it is a silent frame of AMRNB or AMRWB.

[0033] In this implementation, for the dynamic load determination process of the non-silent frame, as shown in FIG. Figure 8 As shown. Since the maximum payload length byte number of AMRNB voice is 32, if the RTP payload length byte number is greater than 32, the AMRNB voice type can be excluded. If the RTP effective payload length byte number is less than or equal to 32, it is preliminarily determined that the frame meets the byte number of AMRNB, AMRWB or EVS. By judging whether the CMR field of AMR is 15, it can be distinguished whether the frame is AMR or EVS. If the CMR field meets the requirements, it can be determined that the frame meets the voice characteristics of AMRNB or AMRWB, and then the FT value is used to determine whether the frame specifically belongs to AMRNB or AMRWB. If the FT value does not meet the characteristics of AMRNB voice or AMRWB voice, the non-silent frame is wrong and an error is reported. If the CMR field does not meet the requirements, it can be determined that the frame meets the voice characteristics of EVS.

[0034] In this embodiment, if the number of bytes in the RTP payload length is greater than 32, it can be preliminarily determined that the frame may conform to the ILBC voice type, the AMRWB voice type, or the EVS voice type. If the number of bytes in the RTP payload length is 38, it can be determined that the frame conforms to the characteristics of the ILBC voice type. If it does not conform to the voice characteristics of ILBC, it is further determined whether the frame conforms to the characteristics of the AMR voice type by whether the CMR field is 15. If it conforms, the FT value is matched to AMRWB to determine whether the voice type of the frame is the AMRWB voice type. If the FT value does not conform to the operating mode of AMRWB, it can be determined that the frame conforms to the characteristics of the EVS voice type.

[0035] S103: Generate security payload data using the voice coding type, use the security payload data to replace the original payload data in the RTP voice packet and assemble into a new RTP voice packet; It can be understood that during the modification process, the voice can be modified to be silent or non-silent (also known as voice content modification or audio modification) according to specific needs, such as Fig. 9 If it is a silent modification, silent data of the same voice type is generated according to the voice type, and the silent modification is achieved by replacing the RTP payload data with the silent data; if it is an audio modification, audio data of the same voice type is generated according to the voice type, and the voice modification is achieved by replacing the RTP payload data with the audio data.

[0036] For the convenience of description, it is assumed that the original effective payload data in the RTP voice packet is recorded as the first payload data, and the generated security payload data is recorded as the second payload data. If a voice that needs to be modified includes multiple RTP voice packets, if the length of the second payload data is not less than the sum of the lengths of the first payload data of all RTP voice packets of the voice, then only a single modification is required; conversely, if the length of the second payload data is less than the sum of the lengths of the first payload data of all RTP voice packets of the voice, then a cyclic modification can be performed until the first payload data is replaced.

[0037] Specifically, for audio modifications, see Fig.10 , the top is the RTP voice packet, where the first payload data is the payload part of the RTP voice packet; the bottom is the second payload data consisting of audio data. When the device receives the first RTP voice packet, the first payload data of the first RTP voice packet is replaced with data(1) in the second payload data. When the device receives the second RTP voice packet, the first payload data of the second RTP voice packet is replaced with data(2) in the second payload data, and so on, until all RTP voice packets of the voice are replaced. During the replacement process, the data length of data(n) is determined by the payload length of the nth RTP voice packet.

[0038] For mute modifications, see Fig.11 , similar to audio modification, the above is the RTP voice packet, where the first payload data is the payload part of the RTP voice packet, and the following is the second payload data composed of silent data. It can be understood that the silence of different payload types of voice is different, for example, the silent data of PCM16 is 0x00, and the silent data of PCMA is 0xD5. When the device receives the first RTP voice packet, the first payload data of the first RTP voice packet is replaced with the silent data (1) in the second payload data. When the device receives the second RTP voice packet, the first payload data of the second RTP voice packet is replaced with the silent data (2) in the second payload data, and so on, until all RTP voice packets of the voice are replaced.

[0039] It should be noted that the modification in the embodiment of the present invention is performed in the form of borrowing packets, that is, after the communication device receives the RTP voice packet, the packet header information of the original voice packet is retained, and only the payload content of the voice packet is modified.

[0040] S104: Sending back the new RTP voice packet using the original socket.

[0041] The telephone fraud prevention and control method provided in the embodiment of the present invention adopts a multimodal voice modification method. For ordinary users who receive fraudulent calls, the voice content heard by the user is a normal anti-fraud publicity prompt tone or silence, rather than the original voice of the telephone fraud perpetrator, thereby preventing the user from being deceived.

[0042] Based on the above embodiments, Fig.12 As shown, in the embodiment of the present invention, it also includes: parsing the RTP voice packet to obtain basic feature information of the RTP voice packet and storing it in a database for filing and analysis.

[0043] Specifically, the basic feature information includes five-tuple information (source IP, destination IP, source port, destination port, voice type) and time information.

[0044] Based on the above embodiments, in the embodiments of the present invention, Fig.12 As shown, it also includes: performing audio decoding on the original effective load data in the RTP voice packet according to the voice coding type, and storing the audio decoded data to a specified location.

[0045] Specifically, considering that PCM16bit is one of the most commonly used audio standards, in this embodiment, the decoding process corresponding to each voice coding type is used to uniformly decode the original payload data in the RTP voice packet into a PCM16bit raw code format. The audio data can be stored in a persistent medium such as a hard disk to facilitate repeated listening of the voice.

[0046] Fig.13 An example of a physical structure diagram of an electronic device is shown in FIG. Fig.13As shown, the electronic device may include: a processor 1301, a communication interface 1302, a memory 1303 and a communication bus 1304, wherein the processor 1301, the communication interface 1302 and the memory 1303 complete mutual communication through the communication bus 1304. The processor 1301 may call the logic instructions in the memory 1303 to execute the telephone fraud prevention and control method, which includes: step 1: receiving a network data packet, screening out an RTP voice packet suspected of telephone fraud from the network data packet; step 2: parsing the RTP voice packet, determining the voice coding type used by the original payload data in the RTP voice packet; step 3: generating security payload data using the voice coding type, replacing the original payload data in the RTP voice packet with the security payload data and assembling into a new RTP voice packet; step 4: sending back the new RTP voice packet.

[0047] In addition, when the logic instructions in the above-mentioned memory 1303 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0048] An embodiment of the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the methods provided by the above-mentioned method embodiments, for example, including: Step 1: Receive a network data packet, and filter out RTP voice packets suspected of telephone fraud from the network data packet; Step 2: Parse the RTP voice packet to determine the voice encoding type used by the original payload data in the RTP voice packet; Step 3: Generate security payload data using the voice encoding type, use the security payload data to replace the original payload data in the RTP voice packet and assemble it into a new RTP voice packet; Step 4: Send the new RTP voice packet back.

[0049] An embodiment of the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the methods provided by the above-mentioned method embodiments are implemented, for example, including: Step 1: receiving a network data packet, and screening out an RTP voice packet suspected of telephone fraud from the network data packet; Step 2: parsing the RTP voice packet, and determining the voice encoding type used by the original payload data in the RTP voice packet; Step 3: generating security payload data using the voice encoding type, using the security payload data to replace the original payload data in the RTP voice packet and assembling it into a new RTP voice packet; Step 4: sending back the new RTP voice packet.

[0050] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0051] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A telephone fraud prevention and control method based on multimodal voice modification, characterized in that: include: Step 1: Receive network data packets and filter out RTP voice packets suspected of telephone fraud from the network data packets; Step 2: parsing the RTP voice packet to determine the voice coding type used by the original payload data in the RTP voice packet; Step 3: Generate security payload data using the voice coding type, use the security payload data to replace the original payload data in the RTP voice packet and assemble it into a new RTP voice packet; Step 4: Send back the new RTP voice packet.

2. According to claim 1, a telephone fraud prevention and control method based on multimodal voice modification is characterized in that: Step 2 specifically includes: The RTP voice packet is parsed, and the voice coding type adopted by the original payload data in the RTP voice packet is determined according to the payload type PT value obtained by the parsing.

3. According to claim 2, a telephone fraud prevention and control method based on multimodal voice modification is characterized in that: The voice coding type used by the original payload data in the RTP voice packet is determined according to the payload type PT value obtained by parsing, specifically including: If the payload type PT value is less than 96, the voice coding type used by the original payload data in the RTP voice packet is directly determined according to the payload type PT value; If the payload type PT value is greater than or equal to 96 and falls within the range of the four dynamic payload types of AMRNB, AMRWB, EVS and ILBC, the voice coding type used by the original payload data in the RTP voice packet is further determined according to the dynamic payload type judgment rule.

4. According to claim 3, a telephone fraud prevention and control method based on multimodal voice modification is characterized in that: The dynamic load type judgment rules specifically include: Determine the length of the original payload data in the RTP voice packet. If the length is 6 bytes, it is considered that the original payload data in the RTP voice packet adopts EVS voice encoding and is an EVS silent frame; if the length is 7 bytes, continue to determine whether the first four bits of the original payload data are all 1. If so, further determine according to the frame type FT value that the original payload data in the RTP voice packet adopts AMRNB voice encoding or AMRWB voice encoding, and is an AMRNB silent frame or AMRWB silent frame.

5. According to claim 4, a telephone fraud prevention and control method based on multimodal voice modification is characterized in that: The dynamic load type judgment rule specifically includes: If it is a non-silent frame and its length is less than 32 bytes, it is further determined whether the first four bits of the original payload data are all 1. If not, it is considered that the original payload data in the RTP voice packet adopts EVS voice coding; if so, it is further determined according to the frame type FT value that the original payload data in the RTP voice packet adopts AMRNB voice coding or AMRWB voice coding; If the length is greater than 32 bytes, then continue to determine whether the length is equal to 38 bytes. If so, it is considered that the original payload data in the RTP voice packet adopts ILBC voice encoding; if not, then continue to determine whether the values ​​of the first four bits of the original payload data are all 1. If so, and the frame type FT value conforms to AMRWB voice encoding, then it is determined that the original payload data in the RTP voice packet adopts AMRWB voice encoding; otherwise, it is determined that the original payload data in the RTP voice packet adopts EVS voice encoding.

6. A telephone fraud prevention and control method based on multimodal voice modification according to any one of claims 1 to 5, characterized in that: Step 3 also includes selecting a modification mode, wherein the modification mode includes silent modification and non-silent modification; Correspondingly, step 3 specifically includes: If silence modification is selected, silence data of the voice coding type is generated as security payload data; If non-silence modification is selected, audio data of the voice coding type is generated as security payload data.

7. A telephone fraud prevention and control method based on multimodal voice modification according to claim 6, characterized in that: Step 2 also includes: Parsing the RTP voice packet to obtain basic feature information of the RTP voice packet and storing it in a database for record analysis; The original payload data in the RTP voice packet is audio-decoded according to the voice coding type, and the audio-decoded data is stored in a designated location.

8. A telephone fraud prevention and control method based on multimodal voice modification according to claim 6, characterized in that: In step 1, the network data packet is received using the original socket; correspondingly, in step 4, the new RTP voice packet is sent back using the original socket.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the method according to any one of claims 1 to 8 is implemented.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images