Voice call data processing method and device, electronic equipment, and storage medium

Abstract

The application provides a voice call data processing method and device, electronic equipment and storage medium, and belongs to the technical field of data processing. The method is applied to a first device and includes the following steps: receiving an encrypted voice frame sequence of a second device; the second device is a wearable device connected to the first device through Bluetooth; decrypting the encrypted voice frame sequence to obtain a target voice frame sequence, and converting the target voice frame sequence to obtain a target text sequence; determining a sensitive level corresponding to the target voice frame sequence based on the target text sequence; and determining an encryption strategy of the target voice frame sequence based on an environmental risk coefficient corresponding to the wearable device and the sensitive level corresponding to the target voice frame sequence. The voice call data processing method and device, electronic equipment and storage medium provided by the application can improve the security of voice call data.

Description

Technical Field

[0001] This application belongs to the field of data processing technology, and more specifically, relates to a voice call data processing method and apparatus, electronic device, and storage medium. Background Technology

[0002] With the increasingly widespread use of wearable devices (such as smart glasses and smartwatches) and mobile phones, the need for secure protection of voice call data is becoming increasingly prominent. In existing technologies, wearable devices and mobile phones mostly transmit voice data via Bluetooth. However, during transmission, unencrypted or only minimally encrypted raw voice streams are often used directly, making them vulnerable to eavesdropping on the Bluetooth link and subsequent data leakage. Furthermore, the identification of sensitive information in voice call data largely relies on direct voice matching, which is susceptible to environmental noise and semantic ambiguity, resulting in low recognition accuracy and ultimately, inadequate overall security for voice call data. Summary of the Invention

[0003] The purpose of this application is to provide a voice call data processing method, apparatus, electronic device, and storage medium to improve the security of voice call data.

[0004] A first aspect of this application provides a voice call data processing method, applied to a first device, comprising: Receive the encrypted voice frame sequence from the second device; the second device is a wearable device connected to the first device via Bluetooth, the wearable device is a device worn by the first user, and the encrypted voice frame sequence is a voice frame sequence generated by processing and encrypting the voice call data generated during the call between the first user and the second user through the wearable device; The encrypted voice frame sequence is decrypted to obtain the target voice frame sequence, and the target voice frame sequence is then converted to obtain the target text sequence. Determine the sensitivity level of the target speech frame sequence based on the target text sequence; The encryption strategy for the target speech frame sequence is determined based on the environmental risk coefficient corresponding to the wearable device and the sensitivity level corresponding to the target speech frame sequence.

[0005] A second aspect of this application provides a voice call data processing apparatus, applied to a first device, comprising: The voice receiving module is used to receive the encrypted voice frame sequence of the second device; the second device is a wearable device connected to the first device via Bluetooth, the wearable device is a device worn by the first user, and the encrypted voice frame sequence is a voice frame sequence generated by processing and encrypting the voice call data generated during the call between the first user and the second user through the wearable device; The decryption and conversion module is used to decrypt the encrypted voice frame sequence to obtain the target voice frame sequence, and to convert the target voice frame sequence to obtain the target text sequence. The sensitivity assessment module is used to determine the sensitivity level of the target speech frame sequence based on the target text sequence. The encryption module is used to determine the encryption strategy for the target speech frame sequence based on the environmental risk coefficient corresponding to the wearable device and the sensitivity level corresponding to the target speech frame sequence.

[0006] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described voice call data processing method.

[0007] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described voice call data processing method.

[0008] The beneficial effects of the voice call data processing method, apparatus, electronic device, and storage medium provided in this application are as follows: This application ensures the initial security of voice data during Bluetooth transmission by receiving encrypted voice frame sequences transmitted by a wearable device, preventing theft during transmission; the target voice frame sequence is decrypted and converted into a text sequence, realizing the parsing of voice content and providing a foundation for subsequent sensitive analysis; the sensitivity level is determined based on the text sequence, accurately identifying the privacy sensitivity of the voice content, making protection more targeted; the encryption strategy is determined by combining the environmental risk coefficient and the sensitivity level, dynamically adjusting the encryption strength according to the actual environmental security status and content sensitivity, avoiding resource waste caused by over-encryption and preventing security risks caused by insufficient encryption, ultimately ensuring the security of voice call data while also considering processing efficiency and adaptability. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 A schematic flowchart illustrating a voice call data processing method provided in an embodiment of this application; Figure 2 This is a structural block diagram of a voice call data processing apparatus provided in an embodiment of this application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0011] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0013] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating a voice call data processing method provided in an embodiment of this application, which can be executed by an electronic device. The method may include: S101: Receive the encrypted voice frame sequence from the second device; the second device is a wearable device connected to the first device via Bluetooth, the wearable device is a device worn by the first user, and the encrypted voice frame sequence is a voice frame sequence that is processed and encrypted during the voice call data generated by the first user through the wearable device to the second user.

[0014] In this embodiment, the first device refers to the main control device that executes the data processing method, which can be a smartphone, tablet, or other device with strong computing power. The second device is a wearable device connected to the first device via Bluetooth, such as smart glasses or Bluetooth headsets, worn by the first user. It directly collects the user's voice call data, performs preliminary encryption, and then sends it to the first device. The encrypted voice frame sequence is an ordered data sequence obtained by processing the voice call data through frame segmentation (dividing continuous voice into short segments), encoding, and then processing it through an encryption algorithm. Encryption is used to prevent the call data collected by the wearable device from being stolen or tampered with during transmission between the wearable device and the first device via Bluetooth.

[0015] In this embodiment, the first device receives a sequence of encrypted voice frames from the second device. These voice frame sequences originate from a real-time conversation between the first user and the second user via a wearable device, and have already undergone preliminary processing (such as framing and encoding) and encryption by the wearable device. This encryption is to protect the security of the data during Bluetooth transmission.

[0016] For example, the first user initiates a WeChat voice call through the touch button on the smart glasses. The built-in microphone of the smart glasses collects the first user's voice data in real time. The content of the call is as follows: "Now we are synchronizing the 2025 flagship mobile phone R&D plan. The screen will be a 6.8-inch 2K screen, and the supply chain cost will be controlled at 1,200 yuan."

[0017] The smart glasses perform dual-microphone beamforming and adaptive filtering on the raw voice data to remove ambient noise. Then, they segment the data into 10 voice frames at 20ms per frame. The smart glasses use the initial session key (AES-256 key) generated during the Bluetooth pairing process with the phone to encrypt each of the 10 voice frames, forming an encrypted voice frame sequence. The smart glasses then transmit this encrypted voice frame sequence to the first user's phone in real-time via Bluetooth 5.3, with a transmission latency of ≤50ms, meeting the requirements for real-time calls.

[0018] S102: Decrypt the encrypted voice frame sequence to obtain the target voice frame sequence, and convert the target voice frame sequence to obtain the target text sequence.

[0019] In this embodiment, the target speech frame sequence is the original speech data segment sequence restored after decryption of the encrypted speech frame sequence. The target text sequence is the text content obtained by converting the target speech frame sequence through speech recognition technology, arranged in speech time sequence, and used to analyze the sensitivity of the call content.

[0020] In this embodiment, the first device invokes a pre-negotiated session key with the second device to decrypt the received encrypted voice frame sequence frame by frame, restoring the target voice frame sequence containing the original voice features. Then, speech-to-text (ASR) technology is used to convert the voice frame sequence into a target text sequence for subsequent analysis of content sensitivity. The voice signal is converted to text because the accuracy of sensitive information recognition (such as keyword matching) in text form is far higher than that of direct speech recognition, providing a precise analytical medium for subsequent sensitivity level determination.

[0021] For example, after the first user's mobile phone receives the encrypted voice frame sequence, it calls the initial session key (consistent with the smart glasses) stored in the SE secure element, and uses the AES-256 algorithm to decrypt frame by frame to restore the target voice frame sequence containing the original voice features; the mobile phone starts the built-in ASR speech recognition algorithm to convert the target voice frame sequence into a target text sequence; the mobile phone can perform grammatical error correction and semantic completion on the target text sequence to ensure the accuracy of subsequent sensitive word matching.

[0022] S103: Determine the sensitivity level corresponding to the target speech frame sequence based on the target text sequence.

[0023] In this embodiment, the sensitivity level is a classification (such as low, medium, high) based on the privacy sensitivity of the voice call content (target text sequence), reflecting the potential risks of data leakage.

[0024] Based on the transformed target text sequence, the sensitivity level (e.g., low, medium, high) of the voice call content is determined through pre-defined sensitive word matching and semantic analysis to assess the privacy importance of the content. For example, if the text contains sensitive words such as new product development plans or patient medical record numbers, it is judged as highly sensitive; if it is just casual conversation, it is judged as low sensitive.

[0025] S104: Determine the encryption strategy for the target speech frame sequence based on the environmental risk coefficient corresponding to the wearable device and the sensitivity level corresponding to the target speech frame sequence.

[0026] In this embodiment, the environmental risk coefficient is an indicator used to measure the security of the physical environment in which the wearable device is located. The environmental risk coefficient can be factors such as the strength of surrounding network signals and the number of suspicious devices (such as unfamiliar Bluetooth devices, Wi-Fi hotspots), with a higher value indicating a more dangerous environment. The encryption strategy is a security processing scheme determined based on the environmental risk coefficient and the content sensitivity level, including encryption algorithm selection, key length, key update frequency, and additional security checks, with the aim of providing security protection that matches the risk level while ensuring communication efficiency.

[0027] This embodiment can determine an appropriate encryption strategy by using preset decision logic, combined with sensitivity level and environmental risk coefficient, so that the encryption strength matches the actual risk level and balances security and processing efficiency.

[0028] For example, assuming that "R&D plan" is classified as highly sensitive in the generated target text sequence, the mobile phone collects environmental parameters through its built-in network and Bluetooth modules. The current network signal strength is -70dBm (weak signal, easily intercepted), and there are 35 active Bluetooth devices within 10 meters (dense device density, high risk of eavesdropping). The calculated environmental risk coefficient is 0.7 (high risk). Combining the high sensitivity level and the high environmental risk coefficient, the mobile phone determines that the encryption algorithm is SM4-CTR and the transmission channel is a 5G private network. When the mobile phone transmits voice data to the second user, it will follow this encryption strategy to ensure that confidential information such as the R&D plan is not leaked.

[0029] As can be seen from the above, this embodiment ensures the initial security of voice data during Bluetooth transmission by receiving encrypted voice frame sequences transmitted by wearable devices, preventing theft during transmission; after decryption, the target voice frame sequence is obtained and converted into a text sequence, realizing the parsing of voice content and providing a foundation for subsequent sensitivity analysis; based on the text sequence, the sensitivity level can be determined, accurately identifying the privacy sensitivity of the voice content, making the protection more targeted; combining the environmental risk coefficient and the sensitivity level to determine the encryption strategy, the encryption strength can be dynamically adjusted according to the actual environmental security status and content sensitivity, avoiding the waste of resources caused by over-encryption and preventing the security risks caused by insufficient encryption, ultimately ensuring the security of voice call data while taking into account processing efficiency and adaptability.

[0030] In one embodiment of this application, decrypting an encrypted voice frame sequence to obtain a target voice frame sequence includes: The target voice frame sequence is obtained by decrypting each frame of data in the encrypted voice frame sequence based on the private key in the determined session key. The session key is a key generated by the first device and the second device during the Bluetooth pairing process. The first device and the second device store the same session key. The session key includes a public key and a private key. The private key is used to encrypt the voice frame sequence after the voice call data has been processed.

[0031] In this embodiment, the session key refers to the key jointly generated by the first and second devices during the Bluetooth pairing phase. It is only valid within this pairing session and is used for data encryption and decryption between the two parties to ensure secure communication between the devices. The private key is a component of the session key and is a secret key held only by the paired first and second devices. It is confidential and is mainly used for encrypting and decrypting voice frame sequences. The public key is another component of the session key and is used for key exchange or to verify data encrypted with the private key. In this embodiment, it works in conjunction with the private key to form a complete encryption and decryption system.

[0032] In this embodiment, when the first device and the second device pair via Bluetooth, they generate an identical session key through the Bluetooth protocol's key negotiation mechanism. This key is stored in both devices simultaneously, serving as the basis for subsequent encryption and decryption operations. Before sending the voice frame sequence, the wearable device uses the private key from the session key to encrypt each frame of voice data, forming an encrypted voice frame sequence, which is then transmitted to the first device. Upon receiving the encrypted voice frame sequence, the first device uses the private key stored locally in the same session key to decrypt each frame of data in the encrypted voice frame sequence, ultimately restoring the original target voice frame sequence. The session key includes a public key and a private key, with the private key specifically used for the encryption and decryption of the voice frame sequence. The asymmetric encryption ensures that only the successfully paired first and second devices can encrypt and decrypt the data, preventing unauthorized devices from accessing the voice call content.

[0033] For example, a first user makes a call with a second user through smart glasses. The smart glasses collect the voice call data and divide it into frames. They then use a stored private key to encrypt each frame, generating an encrypted voice frame sequence, which is transmitted to the first device (mobile phone) via Bluetooth. Upon receiving the encrypted voice frame sequence, the mobile phone reads the same session key stored during pairing with the second device from its built-in secure element and extracts the private key as the decryption key. The mobile phone decrypts each frame in the encrypted voice frame sequence using the extracted private key, frame by frame. For example, the encrypted fifth frame is decrypted using the private key and restored to the original data containing voice features (the target voice frame). After decrypting all frames sequentially, a continuous target voice frame sequence is obtained.

[0034] As can be seen from the above, this embodiment generates a session key during the Bluetooth pairing phase, enabling the first device and the second device to share a key pair containing a public key and a private key. The private key is used to decrypt each frame of data, ensuring that only the paired device can decrypt the voice frame. This strengthens the authentication between devices, prevents unauthorized access, and improves the security and privacy of voice data transmission.

[0035] In one embodiment of this application, it further includes: The session key is updated based on a preset update time threshold to obtain a new session key; In response to a voice call duration being less than or equal to a preset update time threshold, the system decrypts each frame of data in the encrypted voice frame sequence based on the private key in the session key to obtain the target text sequence. In response to a voice call duration exceeding a preset update time threshold, the system decrypts each frame of data in the encrypted voice frame sequence based on the private key in the new session key to obtain the target text sequence.

[0036] In this embodiment, the preset update time threshold refers to the pre-set validity period of the session key, which is the criterion for determining the time to trigger a key update. The preset update time threshold can be configured according to security requirements. The new session key is a new key pair that is renegotiated and generated by the first device and the second device after the preset update time threshold is reached, replacing the original key. It has the same structure as the initial session key (including a public key and a private key) and is used for subsequent encryption and decryption operations.

[0037] In this embodiment, a pre-set update time threshold for the session key is used as the trigger condition for key updates. During a voice call, the relationship between the call duration and the preset update time threshold is monitored in real time. When the call duration does not exceed the threshold, the first device continues to use the private key in the initial session key to decrypt encrypted voice frames. When the call duration exceeds the threshold, the first device automatically generates a new session key and switches to using the private key in the new key for decryption. This key update mechanism ensures that the key does not remain unchanged during long calls, reducing the risk of overall data leakage due to key breaches through periodic rotation.

[0038] For example, in a long business meeting scenario, the first device is a mobile phone and the second device is smart glasses. Bluetooth pairing has been completed and an initial session key K1 has been generated. The preset update time threshold is set to 30 minutes.

[0039] At the start of the meeting, the mobile phone and smart glasses initiate a timed key update mechanism based on a preset update time threshold. The initial session key is K1 (including the private key). ).

[0040] When the meeting has lasted 25 minutes (call duration ≤ 30 minutes), the mobile phone receives the encrypted voice frame sequence transmitted by the smart glasses, using the private key of the original session key K1. Decrypt frame by frame to obtain the target text sequence. When the meeting reaches 30 minutes, a key update is automatically triggered, generating a new session key K2 (containing the private key). The data is then synchronized to the phone and smart glasses. When the meeting continues for 40 minutes (call duration > 30 minutes), the phone receives a new sequence of encrypted voice frames transmitted from the smart glasses and automatically switches to the private key of the new session key K2. Decrypt frame by frame to obtain the target text sequence.

[0041] As can be seen from the above, this embodiment updates the session key periodically by setting a preset update time threshold. When the call duration does not exceed the threshold, the original key is used for decryption; otherwise, a new key is used. This avoids the long-term use of the same key, reducing the risk of voice data leakage due to key breach. Simultaneously, the decryption key is dynamically adapted according to the call duration, ensuring the security of long-term voice data transmission while also balancing the stability and efficiency of the decryption operation.

[0042] In one embodiment of this application, it further includes: The adjustment coefficient corresponding to the preset update time threshold is determined based on the sensitivity level of the target speech frame sequence. The preset update time threshold is adjusted based on the adjustment coefficient to obtain a new preset update time threshold, and the session key is updated based on the new preset update time threshold to obtain a new session key.

[0043] In this embodiment, the adjustment coefficient is a proportional coefficient determined based on the sensitivity level of the target speech frame sequence. It is used to correct the preset update time threshold, thereby achieving dynamic control of the key update frequency. The new preset update time threshold is a time parameter obtained by correcting the initial preset update time threshold with the adjustment coefficient. It serves as a new triggering condition for session key updates and directly determines the interval duration of key updates.

[0044] In this embodiment, the sensitivity level of the target speech frame sequence is used as the basis for setting corresponding adjustment coefficients for different sensitivity levels. For example, high sensitivity corresponds to a small coefficient, and low sensitivity corresponds to a large coefficient, thus achieving a direct correlation between the level and the coefficient. The initial preset update time threshold is corrected using the adjustment coefficient to obtain a new threshold (the higher the sensitivity level, the smaller the new threshold). Then, the session key update is triggered based on the new threshold, making key updates for high-sensitivity data more frequent and the update cycle for low-sensitivity data longer, thus balancing security and resource consumption.

[0045] For example, assuming a medical consultation scenario, the first device is a mobile phone and the second device is smart glasses. The initial preset update time threshold is 60 minutes, and the sensitivity level corresponds to the adjustment coefficient: high sensitivity 0.3, medium sensitivity 0.6, and low sensitivity 1.0.

[0046] The patient (first user) speaks with a doctor through smart glasses. The phone decrypts the voice frame sequence and converts it to text, recognizing content such as "cancer diagnosis results," which is classified as high sensitivity. Based on the high sensitivity level, the phone sets an adjustment coefficient of 0.3, adjusting the initial threshold of 60 minutes: 60 × 0.3 = 18 minutes, resulting in a new preset update time threshold of 18 minutes. The system updates the session key based on the new 18-minute threshold, generating a new key and synchronizing it to the phone and smart glasses. Subsequent key updates occur at 18-minute intervals. If the call content changes to a casual greeting (classified as low sensitivity), the adjustment coefficient becomes 1.0, the new threshold reverts to 60 minutes, and the key updates at 60-minute intervals.

[0047] As can be seen from the above, this embodiment determines the adjustment coefficient based on the sensitivity level of the voice frame sequence, adjusts the preset update time threshold to the new threshold, and then updates the session key accordingly. This makes the key update frequency match the content sensitivity; highly sensitive content corresponds to a shorter update cycle, enhancing security; and low-sensitivity content has a longer cycle, reducing resource consumption and balancing security and efficiency.

[0048] In one embodiment of this application, determining the sensitivity level corresponding to the target speech frame sequence based on the target text sequence includes: The target text sequence is segmented into multiple text fragments. Each text fragment is matched with a sensitive word in a preset scenario sensitive word library to obtain the weight value corresponding to each text fragment. The preset scenario sensitive word library is built based on the sensitive words involved in different types of call scenarios. The types of sensitive words include core sensitive words and regular sensitive words, and different types of sensitive words correspond to different weight values. The weight values ​​corresponding to each text segment are summed to obtain the total weight value corresponding to the target text sequence; The sensitivity level of the target speech frame sequence is determined based on the total weight value.

[0049] In this embodiment, the preset scenario sensitive word library is a pre-built database containing sensitive words for different call scenarios. Sensitive words are categorized by scenario and distinguished between core and regular types, providing a reference standard for sensitivity assessment. Core sensitive words are words with high sensitivity and a high risk of leakage (such as ID card numbers, payment passwords, etc.), corresponding to higher weight values, and playing a key role in determining the sensitivity level. Regular sensitive words are words with low sensitivity (such as home addresses, phone numbers, etc.), with lower weight values ​​than core sensitive words, affecting the sensitivity level but with a relatively minor effect. The weight value is a quantitative score assigned to a text segment that matches a sensitive word, reflecting the sensitivity of that segment; the weight values ​​differ between core and regular sensitive words. The total weight value is the sum of the weight values ​​of all text segments, which can measure the overall sensitivity of the target text sequence and is directly used to determine the sensitivity level.

[0050] In this embodiment, the target text sequence is decomposed into multiple semantically complete text segments, providing basic units for subsequent sensitive word matching and solving the problem of low accuracy in matching whole text segments. A preset scenario sensitive word library is constructed, which covers sensitive words in different call scenarios (such as business, private, medical, etc.), and the sensitive words are divided into core sensitive words and regular sensitive words, with different weight values ​​for the two categories (core sensitive words have higher weight). Based on the preset scenario sensitive word library, the text segments are matched with the core and regular sensitive words in the library, and corresponding weight values ​​are assigned according to the matching results. The weight values ​​of all text segments are accumulated to obtain the total weight, and the sensitivity level is then determined based on the total weight, transforming the level determination from qualitative to quantitative, improving accuracy and consistency.

[0051] For example, assuming a financial business call scenario, the first device is a bank account manager's mobile phone, and the preset scenario sensitive word library is the financial business library: core sensitive words such as bank card password, transfer amount, etc., correspond to a weight of 3 points, and regular sensitive words such as account balance, wealth management products, etc., correspond to a weight of 1 point; sensitivity level classification: total weight ≥ 6 points is high sensitivity, 3-5 points is medium sensitivity, and < 3 points is low sensitivity.

[0052] The target text sequence in the call between the account manager and the customer is: "Your bank card password needs to be changed. Your current account balance is 50,000 yuan, and the transfer amount is 100,000 yuan." After word segmentation of the text sequence, the following text fragments are obtained: ["your", "bank card password", "needs to be changed", "current", "account balance", "50,000 yuan", "transfer amount", "100,000 yuan"].

[0053] The sensitive word database was used to match: "bank card password" matched a core sensitive word (+3 points), "account balance" matched a regular sensitive word (+1 point), "transfer amount" matched a core sensitive word (+3 points), and the remaining segments had no match (+0 points). The total weight was 3+1+3=7 points. Based on the total weight, the sensitivity level of the target speech frame sequence was determined to be high sensitivity.

[0054] As can be seen from the above, this embodiment processes text sequences through word segmentation, matches the segments with a preset scenario-sensitive word library and assigns weights, and accumulates the total weights to determine the sensitivity level. The differentiated weighting of core and conventional sensitive words, combined with scenario adaptability, makes the sensitivity level determination more accurate, provides a reliable basis for subsequent encryption strategies, and improves the targeted nature of voice data protection.

[0055] In one embodiment of this application, it further includes: Obtain the network signal strength of the environment where the wearable device is located; obtain the number of active Bluetooth devices within a preset range around the wearable device; obtain the number of active Wi-Fi hotspots within a preset range around the wearable device; The network signal strength, the number of active Bluetooth devices, and the number of active Wi-Fi hotspots were standardized to obtain the network signal strength value, the number of active Bluetooth devices value, and the number of active Wi-Fi hotspots value, respectively. Substituting the network signal strength value, the number of active Bluetooth devices, and the number of active Wi-Fi hotspots into the first formula, we obtain the environmental risk coefficient corresponding to the wearable device. The first formula is:

[0056] in, This indicates the environmental risk coefficient corresponding to wearable devices. Indicates the network signal strength value. This indicates the number of active Bluetooth devices. This indicates the number of active Wi-Fi hotspots. , and All of these represent weighting coefficients.

[0057] In this embodiment, network signal strength refers to the strength of the wireless communication signal received by the wearable device. Unstable signals may increase the risk of data transmission being interfered with or eavesdropped on. Active Bluetooth devices / active Wi-Fi hotspots refer to Bluetooth devices and Wi-Fi hotspots that are active and detectable within a preset range around the wearable device. An excessive number of these devices indicates the presence of more potential sources of eavesdropping or interference. Standardization is the process of converting raw environmental parameters of different types and magnitudes into dimensionless values ​​within a uniform range (e.g., 0~1), eliminating dimensional differences between parameters and making multi-factor weighted calculations feasible.

[0058] In this embodiment, the network signal strength of the environment where the wearable device is located, the number of active Bluetooth devices within a preset range, and the number of active Wi-Fi hotspots within a preset range can be obtained by the first device to provide raw data for risk assessment.

[0059] The collected raw data is normalized to obtain dimensionless network signal strength (S), number of active Bluetooth devices (N), and number of active Wi-Fi hotspots (W), ensuring that the parameters can be directly used in the calculation.

[0060] Substitute the standardized three parameters into the first formula. By using a weighted summation method, the standardized values ​​of three environmental factors—network signal strength, number of active Bluetooth devices, and number of active Wi-Fi hotspots—are integrated into a comprehensive environmental risk coefficient R. The weighting coefficients are... , and Each factor is individually represented to reflect its impact on environmental risk, ultimately assessing the safety risk level of the environment in which wearable devices operate. Weighting coefficients. , and It can be set based on experience.

[0061] For example, assuming a business negotiation scenario, the first device is a mobile phone, the second device is smart glasses, and the preset range is 10 meters; weighting coefficient: =0.2 (network signal strength) =0.3 (Bluetooth device) =0.5 (Wi-Fi hotspot); Standardization rules: Network signal strength (-110dBm corresponds to 1.0, -50dBm corresponds to 0.0), Number of Bluetooth devices (0 devices correspond to 0.0, 50 devices correspond to 1.0), Wi-Fi hotspots (0 devices correspond to 0.0, 30 devices correspond to 1.0).

[0062] The phone's sensors detected a network signal strength of -90dBm, 30 active Bluetooth devices, and 20 active Wi-Fi hotspots within 10 meters.

[0063] Standardized processing: Network signal strength value S=(-90-(-110)) / (-50-(-110))=0.33; Bluetooth device quantity value N=30 / 50=0.6; Wi-Fi hotspot quantity value W=20 / 30≈0.67.

[0064] Substituting into the first formula, we get: R = 0.2 × 0.33 + 0.3 × 0.6 + 0.5 × 0.67 ≈ 0.066 + 0.18 + 0.335 = 0.581, which gives us the environmental risk coefficient R ≈ 0.58.

[0065] As can be seen from the above, this embodiment calculates the environmental risk coefficient using a weighted formula after acquiring network signal strength, the number of active Bluetooth devices, and the number of Wi-Fi hotspots, and after standardization processing. By integrating environmental parameters from multiple dimensions and combining them with weighted differential assessment, environmental security risks are objectively quantified, providing accurate environmental basis for subsequent encryption strategies and improving the adaptability and reliability of voice data protection.

[0066] In one embodiment of this application, determining the encryption strategy for the target voice frame sequence based on the environmental risk coefficient corresponding to the wearable device and the sensitivity level corresponding to the target voice frame sequence includes: Determine the sensitivity values ​​corresponding to the sensitivity levels based on the sensitivity level benchmark library; Substituting the environmental risk coefficient and sensitivity value into the second formula, we obtain the encryption requirement assessment value of the target voice frame sequence; The second formula is:

[0067] in, This represents the assessment value of encryption requirements. This indicates the environmental risk coefficient corresponding to wearable devices. Indicates the sensitive value. and All represent weighting coefficients; The encryption strategy for the target speech frame sequence is determined based on the encryption requirement assessment value.

[0068] In this embodiment, the sensitivity level benchmark library is a reference library used to convert sensitivity levels into sensitivity values, containing the mapping relationship between different sensitivity levels and their corresponding sensitivity values ​​(e.g., low sensitivity corresponds to 10-30, medium sensitivity corresponds to 31-60, etc.). The sensitivity value is the specific numerical value obtained after converting the sensitivity level, allowing the originally qualitative sensitivity to participate in mathematical calculations and providing quantitative parameters for encryption requirement assessment. The encryption requirement assessment value is a comprehensive index calculated using the second formula, reflecting the degree of encryption protection required for the voice frame sequence.

[0069] In this embodiment, based on a preset sensitivity level quantization benchmark library, the sensitivity level (e.g., low, medium, high) of the target speech frame sequence is converted into a corresponding sensitivity value Lq, realizing the conversion from qualitative level to quantitative value. The environmental risk coefficient R and the sensitivity quantization value Lq are substituted into the second formula. Where R reflects the security risks of the objective environment, and Lq reflects the sensitivity of the speech content itself. and These are the weighting coefficients for environmental risk and content sensitivity, used to adjust the relative importance of these factors in the encryption requirement assessment. The second formula transforms these two key factors—environmental security and content sensitivity—into directly comparable encryption requirement values ​​E. A higher E value indicates that the audio frame sequence requires stronger encryption protection, ultimately providing a quantitative basis for determining the appropriate encryption strategy.

[0070] For example, assuming a confidential corporate call scenario, the first device is a mobile phone, and the sensitivity level quantification benchmark is: high sensitivity Lq=0.8, medium sensitivity Lq=0.5, low sensitivity Lq=0.2; weighting coefficients: =0.4 (environmental risk) =0.6 (Sensitive Quantization); Encryption strategy mapping: E≥0.7 corresponds to advanced strategy (SM4 algorithm + 128-bit key), 0.4≤E<0.7 corresponds to intermediate strategy (AES-256 algorithm), and E<0.4 corresponds to basic strategy (AES-128 algorithm).

[0071] After preliminary processing, the target speech frame sequence was classified as highly sensitive, and the sensitivity quantification value Lq was determined to be 0.8 based on the benchmark library; the environmental risk coefficient R was 0.65.

[0072] Substituting into the second formula, we calculate the encryption requirement assessment value: E = 0.4 × 0.65 + 0.6 × 0.8 = 0.26 + 0.48 = 0.74. Since E = 0.74 ≥ 0.7, we match the advanced encryption strategy: using the SM4 encryption algorithm and a 128-bit key to encrypt the target voice frame sequence.

[0073] As can be seen from the above, this embodiment obtains the sensitivity value through the sensitivity level benchmark library, combines it with the environmental risk coefficient, and calculates the encryption requirement assessment value using the formula, thereby determining the encryption strategy. By integrating both environmental risk and content sensitivity factors, the encryption requirement is weighted and assessed, making the strategy more aligned with actual security needs and improving the accuracy and effectiveness of voice data encryption.

[0074] Corresponding to the voice call data processing method in the above embodiments, Figure 2 This is a structural block diagram of a voice call data processing apparatus provided according to an embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The voice call data processing device 20 is applied to the first device and includes: a voice receiving module 21, a decryption and conversion module 22, a sensitivity judgment module 23, and an encryption module 24.

[0075] The voice receiving module 21 is used to receive the encrypted voice frame sequence of the second device; the second device is a wearable device connected to the first device via Bluetooth, the wearable device is a device worn by the first user, and the encrypted voice frame sequence is a voice frame sequence generated by processing and encrypting the voice call data generated during the call between the first user and the second user through the wearable device. The decryption and conversion module 22 is used to decrypt the encrypted voice frame sequence to obtain the target voice frame sequence, and to convert the target voice frame sequence to obtain the target text sequence. Sensitivity determination module 23 is used to determine the sensitivity level of the target speech frame sequence based on the target text sequence; The encryption module 24 is used to determine the encryption strategy of the target speech frame sequence based on the environmental risk coefficient corresponding to the wearable device and the sensitivity level corresponding to the target speech frame sequence.

[0076] In one embodiment of this application, the decryption conversion module 22 is specifically used for: The target voice frame sequence is obtained by decrypting each frame of data in the encrypted voice frame sequence based on the private key in the determined session key. The session key is a key generated by the first device and the second device during the Bluetooth pairing process. The first device and the second device store the same session key. The session key includes a public key and a private key. The private key is used to encrypt the voice frame sequence after the voice call data has been processed.

[0077] In one embodiment of this application, the decryption conversion module 22 is further configured to: The session key is updated based on a preset update time threshold to obtain a new session key; In response to a voice call duration being less than or equal to a preset update time threshold, the system decrypts each frame of data in the encrypted voice frame sequence based on the private key in the session key to obtain the target text sequence. In response to a voice call duration exceeding a preset update time threshold, the system decrypts each frame of data in the encrypted voice frame sequence based on the private key in the new session key to obtain the target text sequence.

[0078] In one embodiment of this application, the voice call data processing device 20 further includes: an adjustment module, specifically used for: The adjustment coefficient corresponding to the preset update time threshold is determined based on the sensitivity level of the target speech frame sequence. The preset update time threshold is adjusted based on the adjustment coefficient to obtain a new preset update time threshold, and the session key is updated based on the new preset update time threshold to obtain a new session key.

[0079] In one embodiment of this application, the sensitivity determination module 23 is specifically used for: The target text sequence is segmented into multiple text fragments. Each text fragment is matched with a sensitive word in a preset scenario sensitive word library to obtain the weight value corresponding to each text fragment. The preset scenario sensitive word library is built based on the sensitive words involved in different types of call scenarios. The types of sensitive words include core sensitive words and regular sensitive words, and different types of sensitive words correspond to different weight values. The weight values ​​corresponding to each text segment are summed to obtain the total weight value corresponding to the target text sequence; The sensitivity level of the target speech frame sequence is determined based on the total weight value.

[0080] In one embodiment of this application, the voice call data processing device 20 further includes: a first computing module, specifically used for: Obtain the network signal strength of the environment where the wearable device is located; obtain the number of active Bluetooth devices within a preset range around the wearable device; obtain the number of active Wi-Fi hotspots within a preset range around the wearable device; The network signal strength, the number of active Bluetooth devices, and the number of active Wi-Fi hotspots were standardized to obtain the network signal strength value, the number of active Bluetooth devices value, and the number of active Wi-Fi hotspots value, respectively. Substituting the network signal strength value, the number of active Bluetooth devices, and the number of active Wi-Fi hotspots into the first formula, we obtain the environmental risk coefficient corresponding to the wearable device. The first formula is:

[0081] in, This indicates the environmental risk coefficient corresponding to wearable devices. Indicates the network signal strength value. This indicates the number of active Bluetooth devices. This indicates the number of active Wi-Fi hotspots. , and All of these represent weighting coefficients.

[0082] In one embodiment of this application, the encryption module 24 is further configured to: Determine the sensitivity values ​​corresponding to the sensitivity levels based on the sensitivity level benchmark library; Substituting the environmental risk coefficient and sensitivity value into the second formula, we obtain the encryption requirement assessment value of the target voice frame sequence; The second formula is:

[0083] in, This represents the assessment value of encryption requirements. This indicates the environmental risk coefficient corresponding to wearable devices. Indicates the sensitive value. and All represent weighting coefficients; The encryption strategy for the target speech frame sequence is determined based on the encryption requirement assessment value.

[0084] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of the modules in the aforementioned device embodiments, for example... Figure 2 The functions of the voice receiving module 21, decryption and conversion module 22, sensitivity judgment module 23, and encryption module 24 are shown.

[0085] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0086] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0087] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store information such as a preset scene sensitive word library and encrypted voice frame sequences.

[0088] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation method described in the voice call data processing method provided in the embodiments of this application, or they can execute the implementation method of the electronic device described in the embodiments of this application, which will not be repeated here.

[0089] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0090] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0091] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0092] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0093] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed may be indirect couplings or communication connections through some interfaces or units, or they may be electrical, mechanical, or other forms of connection.

[0094] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0095] Furthermore, the functional modules in the various embodiments of this application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0096] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A voice call data processing method, applied to a first device, characterized in that, include: Receive a sequence of encrypted voice frames from the second device; The second device is a wearable device connected to the first device via Bluetooth. The wearable device is worn by the first user. The encrypted voice frame sequence is a voice frame sequence that is processed and encrypted from the voice call data generated during the call between the first user and the second user through the wearable device. The encrypted voice frame sequence is decrypted to obtain the target voice frame sequence, and the target voice frame sequence is converted to obtain the target text sequence; The target text sequence is segmented into multiple text fragments. Each text fragment is matched with a sensitive word in a preset scenario sensitive word library to obtain a weight value for each text fragment. The preset scenario sensitive word library is constructed based on sensitive words involved in different types of call scenarios. The types of sensitive words include core sensitive words and regular sensitive words, and different types of sensitive words correspond to different weight values. The weight values ​​corresponding to each text segment are summed to obtain the total weight value corresponding to the target text sequence; The sensitivity level corresponding to the target speech frame sequence is determined based on the total weight value; The system obtains the network signal strength of the environment in which the wearable device is located; the system obtains the number of active Bluetooth devices within a preset range around the wearable device; and the system obtains the number of active Wi-Fi hotspots within a preset range around the wearable device. The network signal strength, the number of active Bluetooth devices, and the number of active Wi-Fi hotspots are standardized respectively to obtain the network signal strength value, the number of active Bluetooth devices value, and the number of active Wi-Fi hotspots value, respectively. Substituting the network signal strength value, the number of active Bluetooth devices, and the number of active Wi-Fi hotspots into the first formula, the environmental risk coefficient corresponding to the wearable device is obtained; The first formula is: in, This indicates the environmental risk coefficient corresponding to wearable devices. Indicates the network signal strength value. This indicates the number of active Bluetooth devices. This indicates the number of active Wi-Fi hotspots. , and All represent weighting coefficients; The sensitivity value corresponding to the sensitivity level is determined based on the sensitivity level benchmark library; Substituting the environmental risk coefficient and the sensitivity value into the second formula, the encryption requirement assessment value of the target voice frame sequence is obtained; The second formula is: in, This represents the assessment value of encryption requirements. This indicates the environmental risk coefficient corresponding to wearable devices. Indicates the sensitive value. and All represent weighting coefficients; The encryption strategy for the target voice frame sequence is determined based on the encryption requirement assessment value.

2. The voice call data processing method as described in claim 1, characterized in that, The process of decrypting the encrypted voice frame sequence to obtain the target voice frame sequence includes: The target voice frame sequence is obtained by decrypting each frame of data in the encrypted voice frame sequence based on the private key in the determined session key. The session key is a key generated by the first device and the second device during the Bluetooth pairing phase. The first device and the second device store the same session key. The session key includes a public key and a private key. The private key is used to encrypt the voice frame sequence after the voice call data has been processed.

3. The voice call data processing method as described in claim 2, characterized in that, Also includes: The session key is updated based on a preset update time threshold to obtain a new session key; In response to a voice call duration being less than or equal to the preset update time threshold, each frame of data in the encrypted voice frame sequence is decrypted based on the private key in the session key to obtain the target text sequence. In response to a voice call duration exceeding the preset update time threshold, the private key in the new session key is used to decrypt each frame of data in the encrypted voice frame sequence to obtain the target text sequence.

4. The voice call data processing method as described in claim 3, characterized in that, Also includes: The adjustment coefficient corresponding to the preset update time threshold is determined based on the sensitivity level corresponding to the target speech frame sequence. The preset update time threshold is adjusted based on the adjustment coefficient to obtain a new preset update time threshold, and the session key is updated based on the new preset update time threshold to obtain a new session key.

5. A voice call data processing device, applied to a first device, characterized in that, include: A voice receiving module is used to receive encrypted voice frame sequences from a second device. The second device is a wearable device connected to the first device via Bluetooth. The wearable device is worn by the first user. The encrypted voice frame sequence is a voice frame sequence that is processed and encrypted from the voice call data generated during the call between the first user and the second user through the wearable device. The decryption and conversion module is used to decrypt the encrypted voice frame sequence to obtain the target voice frame sequence, and to convert the target voice frame sequence to obtain the target text sequence. The sensitivity detection module is specifically used for: The target text sequence is segmented into multiple text fragments. Each text fragment is matched with a sensitive word in a preset scenario sensitive word library to obtain a weight value for each text fragment. The preset scenario sensitive word library is constructed based on sensitive words involved in different types of call scenarios. The types of sensitive words include core sensitive words and regular sensitive words, and different types of sensitive words correspond to different weight values. The weight values ​​corresponding to each text segment are summed to obtain the total weight value corresponding to the target text sequence; The sensitivity level corresponding to the target speech frame sequence is determined based on the total weight value; The first calculation module is specifically used for: The system obtains the network signal strength of the environment in which the wearable device is located; the system obtains the number of active Bluetooth devices within a preset range around the wearable device; and the system obtains the number of active Wi-Fi hotspots within a preset range around the wearable device. The network signal strength, the number of active Bluetooth devices, and the number of active Wi-Fi hotspots are standardized respectively to obtain the network signal strength value, the number of active Bluetooth devices value, and the number of active Wi-Fi hotspots value, respectively. Substituting the network signal strength value, the number of active Bluetooth devices, and the number of active Wi-Fi hotspots into the first formula, the environmental risk coefficient corresponding to the wearable device is obtained; The first formula is: in, This indicates the environmental risk coefficient corresponding to wearable devices. Indicates the network signal strength value. This indicates the number of active Bluetooth devices. This indicates the number of active Wi-Fi hotspots. , and All represent weighting coefficients; The encryption module is specifically used for: The sensitivity value corresponding to the sensitivity level is determined based on the sensitivity level benchmark library; Substituting the environmental risk coefficient and the sensitivity value into the second formula, the encryption requirement assessment value of the target voice frame sequence is obtained; The second formula is: in, This represents the assessment value of encryption requirements. This indicates the environmental risk coefficient corresponding to wearable devices. Indicates the sensitive value. and All represent weighting coefficients; The encryption strategy for the target voice frame sequence is determined based on the encryption requirement assessment value.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 4.

Images