TTS service agent system, device and medium

Abstract

The invention discloses a TTS service agent system, device and medium, and the system comprises a security isolation module which comprises an intelligent agent and a security gateway, and the security gateway is connected with the intelligent agent; the security gateway receives the authentication material sent by the intelligent agent so as to verify the intelligent agent; if the verification is passed, the security gateway and the intelligent agent generate the same temporary shared key based on an elliptic curve algorithm, the intelligent agent sends a text encrypted by the temporary shared key to the security gateway, and the security gateway sends the text to a transfer service module; the transfer service module is used for receiving the text sent by the security gateway, connecting the Internet service module and sending the text to the Internet service module; and the internet service module is used for converting the received text into voice, sending the voice to the transfer service module and transmitting the voice back to the intelligent agent. The security of the intelligent agent can be ensured, and the intelligent agent is prevented from being attacked by the outside.

Description

Technical Field

[0001] This invention relates to the field of aviation voice, and more particularly to a TTS service agent system, device and medium. Background Technology

[0002] TTS (Text-to-Speech) is an AI model system that converts text information into speech output. In aviation voice systems, it is mainly used for flight announcements, weather reports, and instruction transmission. The intelligent agent is the client in the TTS service system, which is used to receive instructions from the pilot and perform voice broadcasts.

[0003] Existing aviation voice transmission systems directly call public network TTS services, with the agent directly connecting to the Internet to provide TTS services. This exposes multiple open ports and does not involve text encryption. During the text transmission to the TTS service, the agent may be vulnerable to attacks from public network attackers, seriously threatening the continuity of aviation text-to-speech services. Summary of the Invention

[0004] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention proposes a TTS service proxy system that can ensure the security of intelligent agents and prevent them from being attacked by external forces.

[0005] The present invention also proposes devices and media having the above-mentioned TTS service agent system.

[0006] A TTS service proxy system according to a first aspect of the present invention includes:

[0007] Security isolation module, relay service module, internet service module;

[0008] The security isolation module includes an intelligent agent and a security gateway, with the security gateway connected to the intelligent agent. The security gateway receives authentication materials sent by the intelligent agent to verify the intelligent agent. If the verification is successful, the security gateway and the intelligent agent generate the same temporary shared key based on the elliptic curve algorithm. The intelligent agent sends text encrypted with the temporary shared key to the security gateway, and the security gateway sends the text to the relay service module.

[0009] The relay service module is connected to the security isolation module at one end and to the Internet service module at the other end. It is used to receive text sent by the security gateway, connect to the Internet service module, and send the text to the Internet service module.

[0010] The Internet service module is used to convert the received text into speech and send the speech to the relay service module, thereby transmitting it back to the intelligent agent.

[0011] According to an embodiment of the present invention, a TTS service proxy system has at least the following beneficial effects: The system adopts a three-level security isolation architecture. In the text-to-speech service, the agent is first verified. After the agent (i.e., the client) is verified, the encrypted text is decrypted through the security gateway and sent to the relay service module and the Internet service module to execute the text-to-speech service. In the transmission between the agent and the security gateway, the text is encrypted with a temporary shared key, and the agent only opens the port for transmitting the encrypted text during transmission. This port is only connected to the security gateway, and the agent is isolated from the external Internet, which ensures the security of the agent and avoids attacks on the agent from the outside world.

[0012] According to some embodiments of the present invention, the authentication materials sent by the intelligent agent include: the intelligent agent's device certificate and the intelligent agent's public key corresponding to the device certificate;

[0013] The security gateway and the intelligent agent generate the same temporary shared key based on the elliptic curve algorithm, including:

[0014] The security gateway obtains its public key based on preset elliptic curve parameters and its own private key.

[0015] The security gateway obtains a temporary shared key factor based on the agent's public key and the gateway's private key, and sends the gateway's public key to the agent; the security gateway obtains a temporary shared key based on the temporary shared key factor;

[0016] The agent obtains a temporary shared key factor based on the gateway public key and its own agent private key; the agent obtains a temporary shared key based on the temporary shared key factor.

[0017] According to some embodiments of the present invention, obtaining the temporary shared key based on the temporary shared key factor includes:

[0018] The temporary shared key is obtained using the Elliptic Curve ECDH protocol based on the elliptic curve parameters and the temporary shared key factor.

[0019] Alternatively, a temporary shared key can be obtained by using an application key derivation function based on a preset key length and the temporary shared key factor.

[0020] According to some embodiments of the present invention, the Internet service module sending the voice to the relay service module includes:

[0021] The Internet service module sends the voice message to the relay service module using a segmentation strategy, specifically including:

[0022] Detect the RTT latency, PLR packet loss rate, and available bandwidth of the connected internet;

[0023] Based on the RTT delay, determine whether the latency of the connected internet is severe; if not, send the voice directly to the relay service module.

[0024] If so, the voice is divided into blocks according to the RTT latency, the PLR ​​packet loss rate, and the available bandwidth, and the voice blocks are sent to the relay service module one by one.

[0025] According to some embodiments of the present invention, the Internet service module sends the voice message to the relay service module using a chunking strategy, including:

[0026] The formula for the size of the speech block is as follows:

[0027]

[0028] Among them, B max The maximum preset block size is L, where RTT is the RTT delay. f This is the preset QoS level coefficient.

[0029] According to some embodiments of the present invention, the Internet service module sending the voice to the relay service module includes:

[0030] The Internet service module transmits the voice message to the relay service module using bandwidth allocated by a hierarchical token bucket algorithm, specifically as follows:

[0031] B i (t)=min(β i +r i ·t,R i )

[0032] Among them, B i (t) represents the bandwidth allocated to the Internet service module during the time period t preceding the traffic mutation, β i For the preset emergency limit, r i For the preset continuous flow, R i The maximum flow rate within the time period t preceding the flow rate mutation.

[0033] According to some embodiments of the present invention, the relay service module connects to the Internet service module in the following ways:

[0034] If the relay service module fails to connect to the Internet service module, an exponential backoff strategy is used to reconnect. The reconnection interval is calculated as follows:

[0035] t k =min(α)k ·t0,T max )

[0036] Where α is the preset backoff factor, t0 is the preset base interval, and T max is the preset maximum interval, and k is the number of times to reconnect.

[0037] According to some embodiments of the present invention, the relay service module connecting the Internet service module further includes:

[0038] If the relay service module fails to connect to the internet service module more than N times, then the relay service module connects to another internet service module; where N is calculated using the following formula:

[0039]

[0040] PLR is the detected packet loss rate. This rounds up the result of (2 + 0.1PLR).

[0041] An electronic device according to a second aspect of the present invention includes:

[0042] Memory, used to store programs;

[0043] A processor for executing a program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to deploy a system as described in any one of the first aspects.

[0044] According to a third aspect of the present invention, a storage medium stores computer-executable instructions for deploying a system as described in any one of the first aspects.

[0045] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description, claims, and drawings. Attached Figure Description

[0046] The accompanying drawings are provided to further understand the technical solutions of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the technical solutions of the present invention, and do not constitute a limitation on the technical solutions of the present invention.

[0047] Figure 1 This is a structural block diagram of a TTS service proxy system provided in an embodiment of the present invention;

[0048] Figure 2This is a structural block diagram of a TTS service proxy system provided in an embodiment of the present invention. Detailed Implementation

[0049] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0050] It should be understood that in the description of the embodiments of the present invention, "multiple" (or "amounts") means two or more, "greater than," "less than," and "exceeding" are understood to exclude the stated number, while "above," "below," and "within" are understood to include the stated number. If "first," "second," etc., are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance, or implicitly indicating the number of indicated technical features, or implicitly indicating the order of the indicated technical features.

[0051] like Figure 1 , Figure 2 As shown, this embodiment of the invention provides a TTS service proxy system, including:

[0052] Security isolation module, relay service module, internet service module;

[0053] The security isolation module includes an agent and a security gateway. The security gateway is connected to the agent. The security gateway receives authentication materials sent by the agent to verify the agent. If the verification is successful, the security gateway and the agent generate the same temporary shared key based on the elliptic curve algorithm. The agent sends text encrypted with the temporary shared key to the security gateway, and the security gateway sends the text to the relay service module.

[0054] The relay service module connects to the security isolation module on one end and the internet service module on the other end. It is used to receive text sent by the security gateway, connect to the internet service module, and send text to the internet service module.

[0055] The Internet service module is used to convert received text into speech and send the speech to the relay service module, which then transmits it back to the intelligent agent.

[0056] This invention employs a three-tiered security isolation architecture. In the text-to-speech service, the agent is first verified. Once the agent (i.e., the client) passes verification, the encrypted text is decrypted by the security gateway and sent to the relay service module and the internet service module to execute the text-to-speech service. During transmission between the agent and the security gateway, the text is encrypted, and the agent only opens one data transmission port, which connects exclusively to the security gateway. This isolates the agent from the external internet, ensuring its security and preventing external attacks.

[0057] It's easy to understand why intelligent agents are prohibited from directly connecting to the internet.

[0058] This embodiment reduces the number of exposed ports of the intelligent agent from the existing 22 to a single port (8080 of the HTTPS protocol), reducing the attack surface by 95.5%, greatly reducing the risk of network attacks, effectively following the principle of minimum exposure, and building a solid security defense for aviation voice services; when applied to civil aviation flight risk briefings, it makes the entire briefing system comply with the communication security requirements of DO-278A (the "Software Integrity Assurance Considerations for Communications, Navigation, Surveillance and Air Traffic Management Systems" issued by the U.S. Air Transport Technical Committee in 2011), improving usability.

[0059] In one embodiment, the authentication materials sent by the agent include: the agent's device certificate and the agent's public key corresponding to the device certificate; wherein, the agent's public key is obtained by: calculating the agent's public key according to the preset elliptic curve parameters and its own agent private key, specifically: X_client=x*G, where X_client is the agent's public key, x is the agent's private key, and G is the elliptic curve base point;

[0060] The security gateway and the intelligent agent generate the same temporary shared key based on the elliptic curve algorithm, including:

[0061] The security gateway obtains the gateway public key based on the elliptic curve parameters and its own gateway private key; where the elliptic curve parameters include the elliptic curve base point G, the order p of the prime field in which the elliptic curve is located, and the elliptic curve expression;

[0062] The security gateway obtains a temporary shared key factor S_gateway based on the agent's public key X_client and the gateway's private key y, specifically: S_gateway = y * X_client, and sends the gateway's public key to the agent; the security gateway obtains a temporary shared key based on the temporary shared key factor.

[0063] The agent obtains a temporary shared key factor S_client based on the gateway public key Y_gw and its own agent private key x, specifically S_client = x * Y_gw; the agent obtains a temporary shared key based on the temporary shared key factor.

[0064] By using the Elliptic Curve Diffie-Hellman (ECDH) protocol to generate single-use session keys (i.e., temporary shared keys) with short validity periods (25-35 seconds), and through dynamic key distribution and encryption of temporary server endpoint addresses, this method not only ensures the security of data transmission encryption but also cleverly solves the security risks caused by proxy devices opening multiple ports in traditional technologies. It achieves end-to-end secure communication and can be quickly deployed without large-scale modifications to the existing aviation network architecture.

[0065] It's easy to understand that for security gateways and smart agents, the process of transmitting voice data is similar to that of sending text data; the only difference is the type of data being transmitted.

[0066] In one embodiment, the elliptic curve parameters are elliptic curve parameters as defined in the standard, including:

[0067] Base point G of the elliptic curve;

[0068] The order p of the prime field containing the elliptic curve;

[0069] The selected elliptic curve (e.g., NIST P-256);

[0070] The security gateway obtains its public key based on preset elliptic curve parameters and its own private key, specifically as follows:

[0071] Y_gw = y * G, where Y_gw is the gateway public key, y is the gateway private key, and G is the elliptic curve base point.

[0072] It should be noted that the "*" in y*G represents the dot product operation on an elliptic curve, and the "*" in other expressions has the same meaning.

[0073] The security gateway and the intelligent agent generate identity proofs for the same temporary shared key based on the elliptic curve algorithm:

[0074] According to the associative and commutative laws of dot product operations on elliptic curves, we have:

[0075] S_client=x*Y_gw=x*(y*G)=(x*y)*G

[0076] S_gateway=y*X_client=y*(x*G)=(y*x)*G=(x*y)*G

[0077] Therefore, S_client = S_gateway = S, where S is the temporary shared key factor.

[0078] In one embodiment, obtaining the temporary shared key based on the temporary shared key factor includes:

[0079] The temporary shared key K_session is obtained using the Elliptic Curve ECDH protocol based on the elliptic curve parameters and the temporary shared key factor. Specifically, K_session = HASH(x·G·ymodp), where x is the agent's private key, G is the elliptic curve base point, y is the gateway's private key, p is the order of the prime number field where the elliptic curve is located, and HASH() is the SHA-256 hash function.

[0080] Alternatively, the application key derivation function KDF can be used to obtain the temporary shared key K_session based on the preset key length [OtherInfo] and the temporary shared key factor S. Specifically, the temporary shared key factor S is transformed and compressed using the SHA-256 hash function (HASH) according to the key length [OtherInfo], i.e., K_session = KDF(S, [OtherInfo]).

[0081] It should be noted that the connected intelligent agent and the security gateway obtain the temporary shared key in the same way, which is pre-set; this further strengthens the secure generation of keys and improves the security of transmission.

[0082] In one embodiment, the generation of the same temporary shared key by the security gateway and the intelligent agent based on the elliptic curve algorithm further includes:

[0083] After successful verification, the security gateway sends the gateway public key to the agent;

[0084] After the security gateway generates a temporary shared key, it sends text encrypted with the temporary shared key to the agent through a preset encryption port. The agent can only receive the transmitted text if the temporary shared key generated by the agent is the same as the temporary shared key generated by the security gateway.

[0085] It is easy to understand that there are several ways to obtain a temporary shared key based on the temporary shared key factor. Only if the security gateway and the intelligent agent generate the same temporary shared key will the generated temporary shared key be the same.

[0086] It should be noted that temporary shared keys are valid only once and are used only for the specific session currently established; they also have a short validity period (e.g., 25 to 35 seconds) and need to be renegotiated after expiration.

[0087] In one embodiment, the port number of the encrypted port for transmitting data is 8080, which is known only to the security gateway and the intelligent agent, and cannot be known to the outside world.

[0088] like Figure 2As shown, in one embodiment, the relay service module includes: an authentication unit, a proxy cluster, a protocol conversion engine, and a QoS controller. After the security gateway sends data to the relay service module, the relay service module authenticates the security gateway, performs clustering and protocol conversion on each authenticated security gateway, and sends the data transmitted by each security gateway to the corresponding Internet service module according to the QoS level of each security gateway.

[0089] In one embodiment, the Internet service module sending voice to the relay service module includes:

[0090] The Internet service module sends voice messages to the relay service module using a chunked strategy, specifically including:

[0091] Detect the RTT latency, PLR packet loss rate, and available bandwidth of the connected internet;

[0092] Based on the RTT latency, determine whether the latency of the connected internet is severe; if not, send the voice directly to the relay service module.

[0093] If so, the voice is divided into blocks based on RTT latency, PLR packet loss rate, and available bandwidth, and the blocks are sent to the relay service module one by one.

[0094] In one embodiment, the Internet service module sends voice messages to the relay service module using a chunking strategy, including:

[0095] The formula for the size of speech blocks is as follows:

[0096]

[0097] Among them, B max = 4096 bytes, RTT is the RTT delay, L f The preset QoS level coefficient is 0.8 for P0 level and 1.2 for P1 level.

[0098] It should be noted that QoS level coefficient is an abbreviation for Quality of Service, which refers to a series of requirements and guarantees for the quality of data transmission in network communication, including multiple indicators such as bandwidth, latency, and packet loss rate, and is used to measure the performance and stability of network services.

[0099] The QoS levels are divided as follows:

[0100]

[0101] In this embodiment, the TTS service proxy system is mainly used for pre-flight clearance briefings and pre-flight weather reports, which belong to the P1 level routine broadcasts. It creatively establishes a QoS hierarchical control method that includes priority division of P0 and P1 levels and corresponding bandwidth guarantees, latency requirements, and retry mechanisms. It also adopts a hierarchical token bucket algorithm to perform fine-grained bandwidth allocation and traffic control for different levels of traffic. This enables the system to prioritize the voice transmission quality of urgent and critical instructions under limited network resources, while reasonably arranging routine broadcast tasks. It realizes differentiated and fine-grained QoS management for different types of text-to-speech services, fully meets the diverse and multi-level text-to-speech service needs in aviation scenarios, and improves overall service efficiency and user experience.

[0102] For different network environments (such as 5G, fiber optic, satellite links, etc.) and QoS levels (P0 level emergency release command, P1 level regular weather broadcast), an innovative dynamic block size calculation formula was designed. Based on real-time network RTT, packet loss rate and other status parameters, the voice data block size is adaptively adjusted, which not only avoids audio stuttering caused by network jitter, but also significantly improves bandwidth utilization. It ensures that the audio stream can be transmitted stably, efficiently and with low latency under complex and ever-changing network conditions, especially in special scenarios such as satellite links. It achieves excellent performance with zero service interruption when the end-to-end latency is ≤100ms for P0 level and ≤300ms for P1 level in 5G network, and ≤0.1% packet loss rate in fiber optic network.

[0103] In one embodiment, the voice chunk size when the Internet service module uses a 5G / fiber optic network is as follows:

[0104]

[0105] RTT stands for Round Trip Time, which is the current network round trip time. Typical values ​​for 5G are 5-30ms, and for fiber optics are 1-10ms.

[0106] In one embodiment, the Internet service module sending voice to the relay service module includes:

[0107] The Internet service module sends voice messages to the relay service module using bandwidth allocated by a hierarchical token bucket algorithm, specifically as follows:

[0108] B i (t)=min(β i +r i ·t,R i )

[0109] Among them, B i (t) represents the bandwidth allocated to the Internet service module during the time period t preceding the traffic mutation (8MB for P0, 2MB for P1), β iFor the preset burst limits (5Mbps for P0 level, 2Mbps for P1 level), r i For the preset continuous flow, R i The maximum flow rate within the time period t preceding the flow rate mutation.

[0110] In one embodiment, the relay service module connects to the Internet service module including:

[0111] If the relay service module fails to connect to the Internet service module, an exponential backoff strategy is used to reconnect. The reconnection interval is calculated as follows:

[0112] t k =min(α) k ·t0,T max )

[0113] Where α is the preset backoff factor, t0 is the preset base interval, and T max is the preset maximum interval, and k is the number of times to reconnect.

[0114] The retry interval calculation method has been optimized. Based on the typical RTT characteristics of network type (5G or fiber), the basic interval is reasonably shortened, the backoff factor is adjusted, and a maximum interval that better meets the real-time requirements of aviation scenarios is set. At the same time, a service switching strategy that dynamically determines the threshold N of consecutive retry failures based on the current packet loss rate is introduced. When network anomalies occur, retry and fault switching can be performed quickly and intelligently, effectively avoiding voice service interruptions caused by fluctuations in Internet service, significantly reducing the outage rate to 0.7%, greatly improving the system's fault tolerance and reliability, and ensuring the continuity and stability of aviation text-to-speech service.

[0115] In one embodiment, t0 = 0.1s, α = 1.5, T max =10s.

[0116] In one embodiment, the reconnection interval corresponding to the backoff factor α is shown in the table below:

[0117] Avoidance factor α Mean recovery time (ms) Retry success rate 1.2 85 98.2％ 1.5 102 99.1％ 1.8 145 98.7％

[0118] In one embodiment, the relay service module connecting to the Internet service module further includes:

[0119] If the relay service module fails to connect to the internet service module more than N times, then the relay service module will connect to another internet service module; where N is calculated using the following formula:

[0120]

[0121] PLR is the detected packet loss rate. This rounds up the result of (2 + 0.1PLR).

[0122] In one embodiment, when the TTS service connection is interrupted due to network fluctuations, the relay service immediately triggers a retry mechanism (1s→2s→4s interval). The connection is successfully restored on the third retry (after 4s). The crew, i.e., the user using the intelligent agent, only perceives a 2.3-second voice delay and there is no service interruption.

[0123] In one embodiment, the Internet service module uses Edge-TTS service. Through a carefully designed TTS protocol adapter, full compatibility with the native interface parameters and protocols of various mainstream third-party TTS cloud services (such as edge_tts, Azure TTS, etc.) is achieved. This ensures that intranet clients can seamlessly access and use various TTS services without requiring additional modifications to the client or existing TTS services, greatly reducing system integration costs and complexity, improving the versatility and scalability of the invention, and providing a convenient and efficient bridge for the aviation industry and other industries to widely utilize abundant cloud TTS resources.

[0124] This invention also provides an electronic device, which includes, but is not limited to:

[0125] Memory, used to store programs;

[0126] The processor is used to execute programs stored in memory. When the processor executes programs stored in memory, it is used to deploy the aforementioned TTS service agent system.

[0127] The processor and memory can be connected via a bus or other means.

[0128] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs and non-transitory computer-executable programs, as described in the embodiments of this invention. The processor deploys the aforementioned system by running the non-transitory software programs and instructions stored in the memory.

[0129] The memory may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function; the data storage area may store the system deployed thereon. Furthermore, the memory may include high-speed random access memory and non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include memory remotely located relative to the processor, which can be connected to the processor via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0130] The non-transitory software programs and instructions required to implement the above-described terminal selection method are stored in memory and deployed when executed by one or more processors.

[0131] This invention also provides a storage medium storing computer-executable instructions for deploying the aforementioned system.

[0132] In one embodiment, the storage medium stores computer-executable instructions that are executed by one or more control processors.

[0133] The embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0134] It will be understood by those skilled in the art that all or some of the steps and systems in the methods disclosed above can be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components can be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include computer storage media (or non-transitory media) and communication media (or transient media). As is known to those skilled in the art, the term computer storage media includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, as is known to those skilled in the art, communication media typically include computer-readable instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0135] This document describes embodiments of the invention, including preferred embodiments known to the inventors for carrying out the invention. Variations of these embodiments will become apparent to those skilled in the art upon reading the foregoing description. The inventors encourage those skilled in the art to adopt such variations as appropriate, and the inventors intend to practice embodiments of the invention in ways other than those specifically described herein. Therefore, the scope of the invention includes all modifications and equivalents of the subject matter set forth in the appended claims, as permitted by applicable law. Furthermore, the scope of the invention covers any combination of the foregoing elements in all possible variations thereof, unless otherwise indicated herein or otherwise clearly contradicted by the context.

Claims

1. A TTS service proxy system, characterized in that, include: Security isolation module, relay service module, internet service module; The security isolation module includes an intelligent agent and a security gateway, with the security gateway connected to the intelligent agent. The security gateway receives authentication materials sent by the intelligent agent to verify the intelligent agent. If the verification is successful, the security gateway and the intelligent agent generate the same temporary shared key based on the elliptic curve algorithm. The intelligent agent sends text encrypted with the temporary shared key to the security gateway, and the security gateway sends the text to the relay service module. The relay service module is connected to the security isolation module at one end and to the Internet service module at the other end. It is used to receive text sent by the security gateway, connect to the Internet service module, and send the text to the Internet service module. The Internet service module is used to convert the received text into speech and send the speech to the relay service module, thereby transmitting it back to the intelligent agent.

2. The TTS service proxy system according to claim 1, characterized in that, The authentication materials sent by the intelligent agent include: the intelligent agent's device certificate and the intelligent agent's public key corresponding to the device certificate; The security gateway and the intelligent agent generate the same temporary shared key based on the elliptic curve algorithm, including: The security gateway obtains its public key based on preset elliptic curve parameters and its own private key. The security gateway obtains a temporary shared key factor based on the agent's public key and the gateway's private key, and sends the gateway's public key to the agent; the security gateway obtains a temporary shared key based on the temporary shared key factor; The agent obtains a temporary shared key factor based on the gateway public key and its own agent private key; the agent obtains a temporary shared key based on the temporary shared key factor.

3. The TTS service proxy system according to claim 2, characterized in that, The step of obtaining the temporary shared key based on the temporary shared key factor includes: The temporary shared key is obtained using the Elliptic Curve ECDH protocol based on the elliptic curve parameters and the temporary shared key factor. Alternatively, a temporary shared key can be obtained by using an application key derivation function based on a preset key length and the temporary shared key factor.

4. The TTS service proxy system according to claim 1, characterized in that, The Internet service module sends the voice to the relay service module in the following ways: The Internet service module sends the voice message to the relay service module using a segmentation strategy, specifically including: Detect the RTT latency, PLR packet loss rate, and available bandwidth of the connected internet; Based on the RTT delay, determine whether the latency of the connected internet is severe; if not, send the voice directly to the relay service module. If so, the voice is divided into blocks according to the RTT latency, the PLR ​​packet loss rate, and the available bandwidth, and the voice blocks are sent to the relay service module one by one.

5. A TTS service proxy system according to claim 4, characterized in that, The Internet service module sends the voice message to the relay service module using a segmentation strategy, including: The formula for the size of the speech block is as follows: Among them, B max The maximum preset block size is L, where RTT is the RTT delay. f This is the preset QoS level coefficient.

6. A TTS service proxy system according to claim 4, characterized in that, The Internet service module sends the voice to the relay service module in the following ways: The Internet service module transmits the voice message to the relay service module using bandwidth allocated by a hierarchical token bucket algorithm, specifically as follows: B i (t)=min(β i +r i ·t,R i ) Among them, B i (t) represents the bandwidth allocated to the Internet service module during the time period t preceding the traffic mutation, β i For the preset emergency limit, r i For the preset continuous flow, R i The maximum flow rate within the time period t preceding the flow rate mutation.

7. A TTS service proxy system according to claim 1, characterized in that, The relay service module connects to the Internet service module in the following ways: If the relay service module fails to connect to the Internet service module, an exponential backoff strategy is used to reconnect. The reconnection interval is calculated as follows: t k =min(α k ·t0,T max ) Where α is the preset backoff factor, t0 is the preset base interval, and T max is the preset maximum interval, and k is the number of times to reconnect.

8. A TTS service proxy system according to claim 7, characterized in that, The relay service module connecting to the Internet service module also includes: If the relay service module fails to connect to the internet service module more than N times, then the relay service module connects to another internet service module; where N is calculated using the following formula: PLR is the detected packet loss rate. This rounds up the result of (2 + 0.1PLR).

9. An electronic device, characterized in that, include: Memory, used to store programs; A processor for executing a program stored in the memory, wherein when the processor executes the program stored in the memory, the processor is configured to deploy a system as described in any one of claims 1 to 8.

10. A storage medium, characterized in that, The system contains computer-executable instructions for deploying the system as described in any one of claims 1 to 8.

Images