Wire telephone eavesdropping detection method and device

Abstract

The invention relates to the technical field of wire telephone communication, and provides a wire telephone eavesdropping detection method and device, and the method comprises the steps: collecting the double-side call voice of a current call based on a wire telephone; separating the two-side call voice to obtain an inner voice and an outer voice, and performing time axis synchronization on the inner voice and the outer voice to obtain a synchronous two-way voice; and determining an eavesdropping detection result of the current call based on the acoustic characteristics of the synchronous two-way voice and the time conflict judgment result. The method provided by the invention comprises the following steps: collecting double-side call voices of a current call based on a wire telephone; performing separation to obtain inner voice and outer voice, and performing time axis synchronization on the inner voice and the outer voice to obtain synchronous bidirectional voice; the wiretap detection result of the current call is determined based on the acoustic characteristics of the synchronous bidirectional voice and the time conflict judgment result, comprehensive, effective and reliable wire telephone wiretap detection is realized, and the safety of the call based on the wire telephone is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of wired telephone communication, and in particular, to a method and device for detecting wiretapping of a wired telephone. Background Art

[0002] When a wired telephone is in an idle state, the line is always connected. Since the wired telephone line may be connected to a public telephone network, such as an international line, there is a security risk of wiretapping using the wired telephone line. Because there is a microphone inside the wired telephone, once the microphone is activated, the wired telephone will directly leak voice through the line, that is, hands-free wiretapping is achieved. At the same time, a wiretapping device may also be installed on the line to conduct wiretapping through the line. Moreover, due to work needs, wired telephones are often deployed in work areas, which makes the detection of wiretapping of wired telephones particularly important and complex. The existing wiretapping detections mainly include parallel-wire wiretapping detection and signaling-based wiretapping detection.

[0003] However, parallel-wire wiretapping detection cannot detect signal coupling or cross-program control, and signaling-based wiretapping detection cannot prevent abnormal operation of internal telephones or wiretapping during normal dialing. Moreover, the communication process of a wired telephone is connected to multiple program-controlled switches, telephone lines, and the telephone itself. That is, the current wiretapping detection methods for wired telephones are difficult to effectively prevent wiretapping using the line in all links, and thus the effect of wiretapping detection is not ideal. Summary of the Invention

[0004] The present invention provides a method and device for detecting wiretapping of a wired telephone to solve the defect that the existing wiretapping detection for a wired telephone is difficult to effectively prevent wiretapping using the line in all links.

[0005] The present invention provides a method for detecting wiretapping of a wired telephone, including: Collecting bilateral call voices of the current call based on the wired telephone; Separating the bilateral call voices to obtain inner voices and outer voices, and synchronizing the inner voices and the outer voices on the time axis to obtain synchronized bidirectional voices; Determining a wiretapping detection result of the current call based on the acoustic features and time conflict judgment results of the synchronized bidirectional voices; The time conflict judgment result is used to reflect whether there is a time conflict between the inner voices and the outer voices in the synchronized bidirectional voices on the time axis.

[0006] According to the method for detecting wiretapping of a wired telephone provided by the present invention, the determining the wiretapping detection result of the current call based on the acoustic features and time conflict judgment results of the synchronized bidirectional voices includes: Compare the acoustic features of the synchronous two-way voice with the acoustic feature threshold to obtain a feature comparison result; When the feature comparison result is normal, analyze the call mode of the current call based on the time conflict judgment result; Obtain the wiretapping detection result based on the call mode; The time conflict judgment result includes the time conflict frequency and the conflict duration.

[0007] According to a wiretapping detection method for a wired telephone provided by the present invention, the comparing the acoustic features of the synchronous two-way voice with the acoustic feature threshold to obtain a feature comparison result includes: Extract the inner acoustic features of the inner voice and the outer acoustic features of the outer voice in the synchronous two-way voice respectively; Compare the inner acoustic features with the inner acoustic feature threshold, and compare the outer acoustic features with the outer acoustic feature threshold to obtain the feature comparison result; The acoustic feature threshold includes the inner acoustic feature threshold and the outer acoustic feature threshold.

[0008] According to a wiretapping detection method for a wired telephone provided by the present invention, the feature comparison result includes an inner comparison result and an outer comparison result; After the comparing the acoustic features of the synchronous two-way voice with the acoustic feature threshold to obtain a feature comparison result, it further includes: When the inner comparison result or the outer comparison result is that the acoustic feature is outside the acoustic feature threshold, determine that the wiretapping detection result is that there is wiretapping.

[0009] According to a wiretapping detection method for a wired telephone provided by the present invention, the collecting the bilateral call voice of the current call based on the wired telephone includes: Determine the time reference point of the current call; Based on the time reference point, perform parallel sampling on the current call to obtain the bilateral call voice.

[0010] According to a wiretapping detection method for a wired telephone provided by the present invention, the synchronizing the inner voice and the outer voice on the time axis to obtain a synchronous two-way voice includes: Convert the inner voice and the outer voice to the time axis starting from the time reference point to obtain the synchronous two-way voice.

[0011] The present invention also provides a wiretapping detection device for a wired telephone, including: A collection unit that collects the bilateral call voice of the current call based on the wired telephone; A synchronization unit separates the two-sided call voice to obtain an inner voice and an outer voice, and synchronizes the inner voice and the outer voice on the time axis to obtain synchronized two-way voice; A detection unit determines the wiretapping detection result of the current call based on the acoustic features and time conflict judgment result of the synchronized two-way voice; The time conflict judgment result is used to reflect whether there is a time conflict between the inner voice and the outer voice in the synchronized two-way voice on the time axis.

[0012] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the wiretapping detection method of the wired telephone as described in any one of the above.

[0013] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the wiretapping detection method of the wired telephone as described in any one of the above.

[0014] The present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the wiretapping detection method of the wired telephone as described in any one of the above.

[0015] The wiretapping detection method and device for a wired telephone provided by the present invention collect the two-sided call voice of the current call based on the wired telephone; separate the two-sided call voice to obtain an inner voice and an outer voice, synchronize the inner voice and the outer voice on the time axis to obtain synchronized two-way voice; and determine the wiretapping detection result of the current call based on the acoustic features and time conflict judgment result of the synchronized two-way voice, realizing comprehensive, effective, and reliable wiretapping detection of the wired telephone, thereby improving the security of calls made using the wired telephone. Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0017] Figure 1 is one of the flow schematic diagrams of the wiretapping detection method for a wired telephone provided by the present invention; Figure 2 is a schematic diagram of the wiretapping risk points for a wired telephone provided by the present invention; Figure 3 is a schematic diagram of the voice during a call using a wired telephone without wiretapping; Figure 4 It is a voice schematic diagram for telephone eavesdropping based on forged voice by recording and external playback provided by the present invention; Figure 5 It is one of the voice schematic diagrams for off-hook eavesdropping provided by the present invention; Figure 6 It is the second voice schematic diagram for off-hook eavesdropping provided by the present invention; Figure 7 It is a schematic diagram of the call voice acquisition process of a wired telephone provided by the present invention; Figure 8 It is the second schematic diagram of the process of the eavesdropping detection method for a wired telephone provided by the present invention; Figure 9 It is the third schematic diagram of the process of the eavesdropping detection method for a wired telephone provided by the present invention; Figure 10 It is a schematic diagram of the internal and external voice comparison for unilateral eavesdropping provided by the present invention; Figure 11 It is a schematic diagram of the internal and external voice comparison for recording and external playback provided by the present invention; Figure 12 It is a schematic diagram of the structure of the eavesdropping detection device for a wired telephone provided by the present invention; Figure 13 It is a schematic diagram of the structure of an electronic device provided by the present invention. Detailed implementation manners

[0018] To make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without making creative efforts fall within the scope of protection of the present invention.

[0019] In view of the above problems, the present invention provides an eavesdropping detection method for a wired telephone to achieve real-time and comprehensive eavesdropping detection based on the wired telephone and improve the security of communication based on the wired telephone. Figure 1 It is the first schematic diagram of the process of the eavesdropping detection method for a wired telephone provided by the present invention. As Figure 1 shown, the method includes: Step 110, collect bilateral call voices of the current call based on the wired telephone; Here, the current call based on the wired telephone may refer to a real-time call between two telephone terminals through a wired telephone line. In addition, the bilateral call voices here refer to the voice signals sent by both parties during the call.

[0020] Specifically, by accessing a professional audio acquisition device or software in the communication line of a wired telephone or at the call interface, the voice data of both parties in the call can be collected in real time to obtain the bilateral call voice of the current call. It should be noted that wired telephones currently mainly include analog telephones and IP telephones. Among them, for the voice data acquisition of analog telephone types, it can be achieved by collecting the voice signals of both parties and uniformly converting the voice signals into digital PCM (Pulse Code Modulation) voice data to obtain the bilateral call voice of the current call. For the voice data acquisition of IP telephones, the collected voice data of both parties can be directly used as the bilateral call voice.

[0021] Step 120: Separate the bilateral call voice to obtain the inner voice and the outer voice, and synchronize the inner voice and the outer voice on the time axis to obtain synchronized two-way voice; Here, the inner voice refers to the call voice of the local user during the call, and the outer voice refers to the call voice of the remote user during the call.

[0022] Specifically, through voice separation technologies, such as algorithms like side tone cancellation, blind source separation, and independent component analysis, the collected bilateral call voice can be separated into the inner voice and the outer voice. Then, the separated inner voice and outer voice can be aligned on the time axis to ensure that they are consistent in time for subsequent eavesdropping detection.

[0023] Step 130: Determine the eavesdropping detection result of the current call based on the acoustic features and time conflict judgment result of the synchronized two-way voice; The time conflict judgment result is used to reflect whether there is a time conflict between the inner voice and the outer voice in the synchronized two-way voice on the time axis.

[0024] Here, the acoustic features of the synchronized two-way voice refer to the acoustic characteristics of the inner voice and the outer voice, such as volume, voice length, sound frequency, etc., which are used to analyze the voice characteristics of both parties in the call. In addition, the time conflict judgment result here refers to the result obtained after performing time conflict analysis on the synchronized two-way voice, which is used to reflect whether there is a conflict between the inner voice and the outer voice of both parties in the call at the same time point, that is, whether there is voice overlap. The eavesdropping detection result here can include the existence and non-existence of eavesdropping.

[0025] Specifically, the acoustic features of the synchronized two-way voice can be extracted, including volume, voice length, sound frequency, etc. In addition, by checking whether there is a time conflict in the synchronized two-way voice, the time conflict judgment result of the current call can be obtained. For example, if there is both an inner voice or an outer voice during a certain period of time, that is, there is an overlap between the inner voice and the outer voice.

[0026] Further, the eavesdropping detection result of the current call can be determined by analyzing the acoustic features of the synchronous two-way voice and the time conflict judgment result. For example, the eavesdropping detection result can be calculated by applying corresponding algorithms to the acoustic features of the synchronous two-way voice and the time conflict judgment result; alternatively, the eavesdropping detection result can be output by analyzing the acoustic features of the synchronous two-way voice and the time conflict judgment result using a trained eavesdropping detection model. Here, the eavesdropping detection model can be trained based on various normal voice calls and abnormal eavesdropping voice samples, and the initial model can be constructed based on models such as CNN (Convolutional Neural Network), Transformer, BERT (Bidirectional Encoder Representations from Transformers). In addition, when the eavesdropping detection result indicates the existence of eavesdropping, an alarm message can be sent and the line of the current call can be interrupted.

[0027] In one embodiment, Figure 2 is a schematic diagram of the risk points of telephone eavesdropping based on a wired telephone provided by the present invention. As Figure 2 shown, in a call based on a wired telephone, the two parties of the call (the outer telephone and the inner telephone) are connected by a wired line, and there may be multiple program-controlled switches in the middle. Thus, in practical applications, there may be hang-up eavesdropping using the wired line in a call based on a wired telephone, and the risk points exist in the line between the outer telephone and the program-controlled switch, on the program-controlled switch, the line between the program-controlled switch and the inner-line telephone, and on the inner-line telephone. In addition, due to work needs, wired telephones are often deployed in work areas, and the eavesdropping of wired telephones also includes hang-up eavesdropping or line eavesdropping of the working environment sound.

[0028] It should be noted that the acoustic features of the synchronous two-way voice can be used to reflect whether the acoustic features of the inner and outer voices of the current call conform to the acoustic features in the case of a non-eavesdropped call. When the acoustic features of the synchronous two-way voice are quite different from the acoustic features in the case of a non-eavesdropped call, it can be considered that there is hang-up eavesdropping or unilateral eavesdropping in the current call. Thus, the eavesdropping detection of the working environment can be realized through the acoustic features of the synchronous two-way voice, and the detection of unilateral eavesdropping and hang-up eavesdropping at each risk point on the wired line can be carried out.

[0029] In addition, when the acoustic features of the synchronous two-way voice conform to the acoustic features in the case of a non-eavesdropped call, the call mode can be analyzed through the time conflict judgment result of the current call, so as to implement eavesdropping detection by forging voices at the recording playback end at each risk point on the wired line.

[0030] Therefore, by performing eavesdropping detection based on the acoustic features of the synchronous two-way voice and the time conflict judgment result, full-node detection covering the mainstream eavesdropping of wired telephones can be achieved, improving the comprehensiveness of wired telephone eavesdropping detection. Moreover, the eavesdropping detection based on the acoustic features and the time conflict judgment result ensures the reliability of the eavesdropping detection result through the means of detecting from the features.

[0031] The method provided by the embodiments of the present invention collects the bilateral call voices of the current call based on a wired telephone; separates the bilateral call voices to obtain the inner voice and the outer voice, synchronizes the inner voice and the outer voice on the time axis to obtain synchronous two-way voice; determines the eavesdropping detection result of the current call based on the acoustic features of the synchronous two-way voice and the time conflict judgment result, realizing comprehensive, effective and reliable wired telephone eavesdropping detection, and further improving the security of calls based on wired telephones.

[0032] Based on any of the above embodiments, step 130 includes: Comparing the acoustic features of the synchronous two-way voice with the acoustic feature threshold to obtain a feature comparison result; When the feature comparison result is normal, analyzing the call mode of the current call based on the time conflict judgment result; Obtaining the eavesdropping detection result based on the call mode; The time conflict judgment result includes the time conflict frequency and the conflict duration.

[0033] Here, the feature comparison result can be used to reflect whether the acoustic features of the inner voice and the outer voice of the current call are normal call voices without eavesdropping.

[0034] Specifically, first, the feature comparison result of the current call can be obtained by comparing the acoustic features of the synchronous two-way voice with the acoustic feature threshold. For example, the inner acoustic features of the synchronous two-way voice can be compared with the inner acoustic feature threshold, and the outer acoustic features can be compared with the outer acoustic feature threshold, and the comparison results can be used as the feature comparison result.

[0035] Among them, the acoustic feature threshold may include an inner acoustic feature threshold and an outer acoustic feature threshold. The acoustic feature threshold can be obtained by extracting the acoustic features of normal calls without eavesdropping. Here, the acoustic feature threshold can be adjusted correspondingly based on the actual settings of the wired telephone. For example, customized configurations can be set according to the user's usage scenario, and it is also applicable to telephone conferences, video conferencing systems, etc., which can prevent abnormal transmission of voices through the line. The embodiments of the present invention do not make specific limitations.

[0036] Then, the acoustic features of the synchronous two-way voice can be compared. When the result of the feature comparison is normal, it does not necessarily mean that there is no eavesdropping in the current call. Therefore, when the result of the feature comparison is normal, that is, when the acoustic features of the inner voice and the outer voice both conform to the acoustic features of a normal call, the call mode of the current call can be analyzed through the time conflict judgment result and the synchronous two-way voice. It should be noted that the time conflict judgment result here can be used to reflect the time conflict frequency and conflict duration of the inner voice and the outer voice of the current call on the time axis. Among them, the time conflict frequency represents the number of times of time conflicts between the inner and outer voices in the current call; the conflict duration can represent the duration of each time conflict between the inner and outer voices in the current call. It should be noted that in normal call behaviors without eavesdropping, overlapping bilateral calls are generally less. Then, when the time conflict frequency and conflict duration of the inner voice and the outer voice of the current call on the time axis are greater than a certain threshold, it can be judged that there is eavesdropping in the current call.

[0037] It can be understood that if the time conflict judgment result of the current call indicates that both the time conflict frequency and the conflict duration exceed the preset threshold, it means that the call mode of the current call is an abnormal mode, and there may be call eavesdropping; if the time conflict judgment result of the current call indicates that both the time conflict frequency and the conflict duration do not exceed the preset threshold, the call mode of the current call can be obtained as a normal mode, such as a reporting mode, etc.

[0038] Finally, the eavesdropping detection result can be obtained through the call mode. For example, if the call mode is an abnormal mode, the eavesdropping detection result is that there is eavesdropping; if the call mode is a normal mode, such as a reporting mode, the eavesdropping detection result is that there is no eavesdropping.

[0039] In one embodiment, Figure 3 is a voice schematic diagram of a call based on a wired telephone without eavesdropping provided by the present invention. As Figure 3 shown, after the inner-end voice (inner voice) and the outer-end voice (outer voice) of the call are synchronized on the time axis, the inner-end voice and the outer-end voice are in a "zipper shape", and there is basically no time conflict between the voices of the two call parties, and the acoustic features are normal.

[0040] Figure 4 This is a voice schematic diagram of telephone eavesdropping based on forged voice by recording and playing outwards provided by the present invention. As Figure 4 shown, the inner-end voice (inner-side voice) not only includes the original inner-end voice represented by the green square, but also includes the outer-end voice represented by the grid square recorded by the eavesdropping device. In order to make the eavesdropping more concealed, the eavesdropper intentionally simulates the voice to counter the detection of eavesdropping by recording and playing outwards, that is, the grid square on the outer-end voice axis in Figure 4 . It should be noted that since the inner-end voice is the bilateral call voice before eavesdropping by recording and playing outwards, the inner-end voice has a complete call sequence and the remaining space gap is extremely small. On the other hand, due to the extremely high real-time nature of voice calls, usually at the millisecond level, it is very difficult for the eavesdropper to predict the next state of the outer-end voice to be eavesdropped and perfectly construct the gap to simulate a bilateral conversation. Therefore, the presented result is that there is always a time conflict between the eavesdropped voice and the bilateral call voice at the inner end.

[0041] The method provided by the embodiment of the present invention, when the feature comparison result is normal, analyzes the call mode of the current call based on the time conflict judgment result including the time conflict frequency and conflict duration to obtain the eavesdropping detection result, realizing the effective detection of telephone eavesdropping based on forged voice by recording and playing outwards, and greatly improving the comprehensiveness of wired telephone eavesdropping detection.

[0042] Based on any of the above embodiments, comparing the acoustic features of the synchronous two-way voice and the acoustic feature threshold to obtain the feature comparison result includes: Respectively extracting the inner acoustic features of the inner-side voice and the outer acoustic features of the outer-side voice in the synchronous two-way voice; Comparing the inner acoustic features with the inner acoustic feature threshold, and comparing the outer acoustic features with the outer acoustic feature threshold to obtain the feature comparison result; The acoustic feature threshold includes the inner acoustic feature threshold and the outer acoustic feature threshold.

[0043] Specifically, the medial acoustic features of the medial voice and the lateral acoustic features of the lateral voice in the synchronous two-way voice can be extracted separately. For example, features such as the volume, duration, and voice frequency of the medial voice and the lateral voice can be extracted separately to obtain the medial acoustic features and the lateral acoustic features. Then, the medial acoustic features can be compared with the medial acoustic feature threshold, and the lateral acoustic features can be compared with the lateral acoustic feature threshold. For example, algorithms can be used to calculate the medial acoustic features and the medial acoustic feature threshold, and algorithms can be used to calculate the lateral acoustic features and the lateral acoustic feature threshold. For another example, the medial acoustic feature threshold and the lateral acoustic feature threshold obtained based on normal calls can be used as parameters of the eavesdropping detection model, and the medial acoustic features and the lateral acoustic features can be judged separately through the eavesdropping detection model to obtain the feature comparison result.

[0044] Then, the comparison result is used as the feature comparison result, that is, the feature comparison result of the medial voice and the lateral voice is obtained. Among them, the acoustic feature threshold includes the medial acoustic feature threshold and the lateral acoustic feature threshold. The medial acoustic feature threshold and the lateral acoustic feature threshold can be benchmark values obtained by extracting features from the medial voice and the lateral voice in multiple normal calls and performing statistics, and can be used to measure whether the acoustic features of the medial and lateral voices in the current call are normal.

[0045] Based on any of the above embodiments, the feature comparison result includes a medial comparison result and a lateral comparison result; After comparing the acoustic features of the synchronous two-way voice with the acoustic feature threshold to obtain the feature comparison result, the following further includes: When the medial comparison result or the lateral comparison result is that the acoustic feature is outside the acoustic feature threshold, it is determined that the eavesdropping detection result is that there is eavesdropping.

[0046] Here, the medial comparison result represents the feature comparison result of the medial voice, and the lateral comparison result represents the feature comparison result of the lateral voice.

[0047] Specifically, when the medial comparison result or the lateral comparison result in the feature comparison result is abnormal, that is, when the lateral comparison result is that the acoustic feature is outside the acoustic feature threshold, it is determined that the eavesdropping detection result of the current call is that there is eavesdropping. For example, when the medial comparison result is that the volume and duration of the medial voice are lower than the medial acoustic feature threshold, it can be directly determined that the medial call is abnormal and there is a hang-up eavesdropping of ambient sound. For another example, when the volume and duration of the collected lateral voice are lower than the lateral acoustic feature threshold, it can be directly determined that there is no call on the outside and it is a unilateral call leak.

[0048] In one embodiment, Figure 5 is one of the voice schematic diagrams of the hang-up eavesdropping provided by the present invention, asFigure 5 As shown, only the inner-end sound (inner-side voice) contains a voice sequence, and there is no voice sequence in the outer-end sound (outer-side voice). It should be noted that when wiretapping the ambient sound of the inner-end phone, it is only necessary to determine whether the acoustic characteristics of the inner-end sound are normal. It should also be noted that due to the distance between the speaker and the sound pickup device, generally, the call sounds collected in the presence of a wiretap are quite different from the call sounds normally collected through a phone microphone, that is, the wiretap cannot collect high-definition sounds at the microphone level. Therefore, the abnormal situation reflected by the feature comparison result of the inner-side voice can be used to detect the wiretapping of the ambient sound during a call hang-up, and it is a unilateral wiretap.

[0049] In another embodiment, Figure 6 is the second voice schematic diagram of the call hang-up wiretapping provided by the present invention. As Figure 6 shown, only the inner-end sound (inner-side voice) contains a voice sequence and contains the sounds at both ends, and there is no voice sequence in the outer-end sound (outer-side voice). It should be noted that generally, when wiretapping, in order to prevent themselves from being discovered, wiretappers adopt silent wiretapping. In addition, unilateral wiretapping is mainly applicable to the case of recording and playing out. At this time, the acoustic characteristics of the outer-side voice can be directly compared to obtain the feature comparison result of the outer-side voice. Therefore, the abnormal situation reflected by the feature comparison result of the outer-side voice can be used to determine whether there is a unilateral wiretapping of recording and playing out in the current call.

[0050] The method provided by the embodiments of the present invention determines the wiretapping detection result as wiretapping exists by respectively judging the acoustic characteristics of the inner-side voice and the outer-side voice. When the inner-side comparison result or the outer-side comparison result is that the acoustic characteristics are outside the acoustic characteristic threshold, it covers the detection of call hang-up wiretapping based on ambient sound and the detection of unilateral wiretapping of recording and playing out.

[0051] Based on any of the above embodiments, step 110 includes: Determine the time reference point of the current call; Based on the time reference point, perform parallel sampling on the current call to obtain the bilateral call voice.

[0052] Here, the time reference point refers to a unified clock source. For example, the start time point of the current call can be used as the time reference point.

[0053] Specifically, the start time point of the current call can be used as the time reference point. By using a unified time reference point, the two parties of the current call are sampled in parallel in real time through an audio acquisition device or software to obtain the bilateral call voice. In detail, wired telephones can be divided into analog telephones and IP telephones (digital telephones).

[0054] Figure 7It is a schematic diagram of the call voice acquisition process of the wired telephone provided by the present invention. As Figure 7 shown, for analog telephone voice acquisition, a unified clock source and sampling frequency can be adopted to synchronously acquire two-way voice and obtain bilateral call voices. During a voice call, the voice data of analog voice has the characteristic of high real-time performance. Therefore, for the acquisition of voice data of analog telephones, a unified clock source can be adopted, and the sampling of two-way analog signals can be realized by wiring in parallel. The voice of traditional analog telephones is a 4 kHz signal, so the input PCM data (RX, Receive, receive) and the output PCM data (TX, Transmit, transmit) can be sampled uniformly at a 2-fold sampling rate of 8 kHz. The received voice and the transmitted voice start recording on the same time axis, and the time axis here includes coordinate index points. Due to the high real-time performance of the analog line of the voice call, by starting sampling simultaneously with the same clock source, it can be ensured that the error of two-way voice does not exceed 1 / 8 kHz = 0.125 μs, and the time length of normal call voice is at least 0.5 s or more, so it fully meets the synchronous acquisition of bilateral voices and ensures the accuracy of eavesdropping detection based on bilateral call voices.

[0055] In addition, for the call voice acquisition of digital telephones, the synchronization of data transmission and reception can be achieved through unified reception. If the collected digital signal is encoded, the digital signal is first decoded into the original PCM data, that is, bilateral call voices are obtained. The RX and TX of digital voice communication are based on mature digital communications such as E1 / T1 or RTP. Therefore, in order to realize the synchronous acquisition of bilateral call voices, a unified reception program is also adopted to receive bilateral call voices at the same time. Considering that digital signals such as E1 / T1 adopt time-division communication with the characteristic of low delay, and its delay is basically equal to 0.125 μs, other digital voices can perform data sampling based on a unified time reference point, which can prevent the time difference problem caused by network jitter. And, for individual timeout packets, they can be directly discarded to ensure that the reception rates of bilateral voices are consistent. Therefore, the delay of the acquisition synchronization of digital voices in a normal voice system is lower than 0.1 s, so it can meet the synchronous acquisition of bilateral call voices and ensure the accuracy of eavesdropping detection based on bilateral call voices.

[0056] In addition, the collected bilateral call voices can be stored through a unified time axis, and the judgment of eavesdropping detection is realized by scanning a sliding window.

[0057] Based on any of the above embodiments, the time axis synchronization of the inner voice and the outer voice to obtain synchronous two-way voice includes: Converting the inner voice and the outer voice to a time axis starting from a time reference point to obtain the synchronous two-way voice.

[0058] Specifically, during the voice collection process, voice data is sampled based on a unified time reference point. Thus, after separating the bilateral call voices collected, the separated inner voice and outer voice can be converted to a time axis starting from the time reference point to obtain synchronous two-way voices, ensuring that the time interval between the synchronous two-way voices conforms to the time interval during the actual call, thereby ensuring the accuracy and reliability of eavesdropping detection.

[0059] Based on any of the above embodiments, Figure 8 is the second flowchart of the eavesdropping detection method for a wired telephone provided by the present invention. As Figure 8 shown, the method includes: First, perform real-time voice collection. For a wired telephone of the IP telephone type, the analog signal can be converted to digital and the IP packet can be received. Then, perform bilateral voice PCM extraction. Specifically, separate the bilateral call voices collected in the previous step to obtain the inner voice and the outer voice, that is, obtain the inner PCM voice data and the outer PCM voice data. Further, perform time alignment on the bilateral PCM voice data to obtain synchronous two-way voices. Then, perform call mode recognition on the synchronous two-way voices to determine whether an abnormal call mode is detected. When an abnormal call mode is detected currently, alarm and communication blocking are performed.

[0060] In another embodiment, Figure 9 is the third flowchart of the eavesdropping detection method for a wired telephone provided by the present invention. As Figure 9 shown, first, perform voice extraction and time synchronization on the current call, that is, implement the collection of bilateral call voices, separate the bilateral call voices, and perform time alignment on the separated inner voice data PCM1 and outer voice data PCM2. Further, determine whether the inner voice is a normal hands-free or microphone pickup. If not, it means that there is ambient sound eavesdropping in the current call, which is handset hanging eavesdropping or line eavesdropping. That is, the inner voice presents an abnormal call state, and the call can be ended by alarm.

[0061] In addition, determine that the outer voice has no sound or the sound frequency is abnormal. If it is abnormal, there is unilateral eavesdropping in the current call, which is an eavesdropping behavior of unilateral leakage of the inner voice, and the call can be ended by alarm. Figure 10 is the schematic diagram of the comparison between the inner and outer voices of unilateral eavesdropping provided by the present invention. As Figure 10 shown, if the sound volume of the outer voice is much smaller than the normal volume, there is unilateral eavesdropping in the current call.

[0062] When the inner voice is normal and the outer voices are all normal, bilateral voice call mode analysis is performed to determine whether the internal and external call modes are abnormal. In the case of abnormality, it indicates that there is recording and external playback during the current call, and the internal bilateral voices are sent externally. If the internal and external call modes are abnormal, an alarm can be issued to end the call. Figure 11 This is a schematic diagram of the comparison between the inner and outer voices for recording and external playback provided by the present invention. As Figure 11 shown, the acoustic characteristics of the inner voice and the outer voices are within the normal threshold range, but there is a time conflict between the outer voice and the inner voice on the time synchronization axis, that is, both the inner voice and the outer voice exist in the same time period.

[0063] Finally, the current call is collected and wiretapping detection is performed in real time until the call ends.

[0064] It should be noted that based on the identification of wiretapping behavior by judging the call mode, by adopting the method of real-time monitoring of the call voice on the wired line, various call modes can be identified after synchronizing the call voice, which can effectively prevent any wiretapping behavior on the line and the device. It is a detection means based on features, ensuring the reliability of the wiretapping detection result. Moreover, it can support various existing analog telephones, E1 digital trunks, and IP telephones, and can support the external voice access end. In addition, based on the identification of wiretapping behavior, it can prevent the detection of wiretapping after any future attack on a communication device or system and entering the line, and has strong practical capabilities.

[0065] The method provided by the embodiment of the present invention realizes the detection of various wiretapping behaviors through the above three detection mechanisms. The way that these three detection mechanisms act simultaneously can more completely realize the detection of wiretapping behavior based on the wired line and avoid the behavior of wiretapping a telephone based on the wired line.

[0066] Based on any of the above embodiments, Figure 12 This is a schematic structural diagram of a wiretapping detection device for a wired telephone provided by the present invention. As Figure 12 shown, the device includes: A collection unit 1210 that collects bilateral call voices of the current call based on the wired telephone; A synchronization unit 1220 that separates the bilateral call voices to obtain an inner voice and an outer voice, and synchronizes the inner voice and the outer voice on the time axis to obtain synchronized bidirectional voices; A detection unit 1230 that determines the wiretapping detection result of the current call based on the acoustic characteristics and time conflict judgment result of the synchronized bidirectional voices; The time conflict judgment result is used to reflect whether there is a time conflict between the inner voice and the outer voice in the synchronized bidirectional voices on the time axis.

[0067] The device provided by the embodiment of the present invention collects bilateral call voices of the current call based on a wired telephone; separates the bilateral call voices to obtain an inner voice and an outer voice, synchronizes the inner voice and the outer voice on the time axis to obtain synchronized two-way voices; and determines the wiretapping detection result of the current call based on the acoustic features and the time conflict judgment result of the synchronized two-way voices, realizing comprehensive, effective and reliable wiretapping detection of the wired telephone, thereby improving the security of calls made based on the wired telephone.

[0068] Based on any of the above embodiments, the detection unit is specifically configured to: Compare the acoustic features of the synchronized two-way voices with an acoustic feature threshold to obtain a feature comparison result; When the feature comparison result is normal, analyze the call mode of the current call based on the time conflict judgment result; Obtain the wiretapping detection result based on the call mode; The time conflict judgment result includes the time conflict frequency and the conflict duration.

[0069] Based on any of the above embodiments, the detection unit is further specifically configured to: Extract the inner acoustic features of the inner voice and the outer acoustic features of the outer voice in the synchronized two-way voices respectively; Compare the inner acoustic features with an inner acoustic feature threshold, and compare the outer acoustic features with an outer acoustic feature threshold to obtain the feature comparison result; The acoustic feature threshold includes the inner acoustic feature threshold and the outer acoustic feature threshold.

[0070] Based on any of the above embodiments, the feature comparison result includes an inner comparison result and an outer comparison result; The detection unit is further specifically configured to: When the inner comparison result or the outer comparison result indicates that the acoustic features are outside the acoustic feature threshold, determine that the wiretapping detection result is that there is wiretapping.

[0071] Based on any of the above embodiments, the acquisition unit is specifically configured to: Determine the time reference point of the current call; Based on the time reference point, perform parallel sampling on the current call to obtain the bilateral call voices.

[0072] Based on any of the above embodiments, the synchronization unit is specifically configured to: Convert the inner voice and the outer voice to the time axis starting from the time reference point to obtain the synchronized two-way voices.

[0073] Figure 13 An entity structure schematic diagram of an electronic device is exemplified, as Figure 13 shown. The electronic device may include: a processor 1310, a communications interface 1320, a memory 1330, and a communication bus 1340. Among them, the processor 1310, the communications interface 1320, and the memory 1330 complete mutual communication through the communication bus 1340. The processor 1310 may call logic instructions in the memory 1330 to execute a wiretap detection method for a landline phone. The method includes: collecting bilateral call voices of a current call based on the landline phone; separating the bilateral call voices to obtain an inner voice and an outer voice, synchronizing the inner voice and the outer voice on the time axis to obtain synchronized two-way voices; determining a wiretap detection result of the current call based on acoustic features and a time conflict judgment result of the synchronized two-way voices; the time conflict judgment result is used to reflect whether there is a time conflict between the inner voice and the outer voice in the synchronized two-way voices on the time axis.

[0074] In addition, when the logic instructions in the foregoing memory 1330 can be implemented in the form of software functional units and sold or used as an independent product, 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 a part of this technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The foregoing storage medium includes: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk, or an optical disc that can store program codes.

[0075] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the wire telephone eavesdropping detection method provided by the above-mentioned various methods. The method includes: collecting bilateral call voices of a current call based on a wire telephone; separating the bilateral call voices to obtain an inner voice and an outer voice, synchronizing the inner voice and the outer voice on the time axis to obtain synchronized two-way voices; determining the eavesdropping detection result of the current call based on the acoustic features and time conflict judgment result of the synchronized two-way voices; the time conflict judgment result is used to reflect whether there is a time conflict between the inner voice and the outer voice in the synchronized two-way voices on the time axis.

[0076] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the wire telephone eavesdropping detection method provided by the above-mentioned various methods. The method includes: collecting bilateral call voices of a current call based on a wire telephone; separating the bilateral call voices to obtain an inner voice and an outer voice, synchronizing the inner voice and the outer voice on the time axis to obtain synchronized two-way voices; determining the eavesdropping detection result of the current call based on the acoustic features and time conflict judgment result of the synchronized two-way voices; the time conflict judgment result is used to reflect whether there is a time conflict between the inner voice and the outer voice in the synchronized two-way voices on the time axis.

[0077] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative labor.

[0078] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment 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 essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing 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.

[0079] 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 them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A wiretap detection method for a wired telephone, characterized in that: include: Collect the audio of both sides of the current call based on the wired telephone; Separating the bilateral conversation voices to obtain inner voice and outer voice, and synchronizing the inner voice and outer voice on a time axis to obtain synchronized bidirectional voice; Determining a wiretap detection result of the current call based on the acoustic characteristics of the synchronous two-way voice and the time conflict judgment result; The time conflict judgment result is used to reflect whether there is a time conflict between the inner voice and the outer voice in the synchronous two-way voice on the time axis.

2. The wiretap detection method for a wired telephone according to claim 1, characterized in that: The step of determining the wiretap detection result of the current call based on the acoustic features of the synchronous two-way voice and the time conflict judgment result includes: Comparing the acoustic features of the synchronous two-way speech with the acoustic feature threshold to obtain a feature comparison result; When the feature comparison result is normal, analyzing and obtaining the call mode of the current call based on the time conflict judgment result; Based on the call mode, obtaining the wiretap detection result; The time conflict judgment result includes the time conflict frequency and the conflict duration.

3. The wiretap detection method for a wired telephone according to claim 2, characterized in that: The comparing the acoustic features of the synchronous bidirectional speech with the acoustic feature threshold to obtain a feature comparison result includes: Respectively extracting inner acoustic features of the inner speech and outer acoustic features of the outer speech in the synchronous bidirectional speech; Comparing the inner acoustic feature with the inner acoustic feature threshold, and comparing the outer acoustic feature with the outer acoustic feature threshold, to obtain the feature comparison result; The acoustic feature threshold includes the inner acoustic feature threshold and the outer acoustic feature threshold.

4. The wiretap detection method for a wired telephone according to claim 3, characterized in that: The feature comparison results include inner comparison results and outer comparison results; The step of comparing the acoustic features of the synchronous bidirectional speech with the acoustic feature threshold to obtain a feature comparison result further includes: When the inner comparison result or the outer comparison result is that the acoustic feature is outside the acoustic feature threshold, the eavesdropping detection result is determined to be the presence of eavesdropping.

5. The wiretap detection method for a wired telephone according to any one of claims 1 to 4, characterized in that: The collecting of the two-way call voice of the current call based on the wired telephone includes: Determining a time reference point of the current call; Based on the time reference point, the current call is sampled in parallel to obtain the bilateral call voice.

6. The wiretap detection method for a wired telephone according to claim 5, characterized in that: The step of synchronizing the inner voice and the outer voice on a time axis to obtain a synchronized two-way voice comprises: The inner voice and the outer voice are converted to a time axis with a time reference point as a start node to obtain the synchronous two-way voice.

7. A wiretap detection device for a wired telephone, characterized in that: include: A collection unit, collecting the voice of both sides of the current call based on the wired telephone; A synchronization unit, which separates the two-way call voice to obtain an inner voice and an outer voice, and synchronizes the inner voice and the outer voice on a time axis to obtain a synchronized two-way voice; A detection unit, which determines a wiretap detection result of the current call based on the acoustic characteristics of the synchronous two-way voice and the time conflict judgment result; The time conflict judgment result is used to reflect whether there is a time conflict between the inner voice and the outer voice in the synchronous two-way voice on the time axis.

8. 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 computer program, the wiretap detection method for a wired telephone according to any one of claims 1 to 6 is implemented.

9. 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 wiretap detection method for a wired telephone as claimed in any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the wiretap detection method for a wired telephone as claimed in any one of claims 1 to 6 is implemented.

Images