Management controller, voice command transmission system and voice command transmission method

Abstract

The invention provides a management controller, a voice instruction transmission system and a voice instruction transmission method, which are applied to the technical field of servers, and the management controller comprises an integrated circuit built-in audio bus controller which is used for obtaining voice instruction data through an integrated circuit built-in audio bus interface, extracting a check code from the voice instruction data, and transmitting the check code to the integrated circuit built-in audio bus controller; verifying the voice instruction in the voice instruction data by using the verification code; the direct execution memory is used for storing the voice instruction passing the verification; the voice recognition and conversion module is connected with the direct execution memory and is used for reading a voice instruction from the direct execution memory, converting the voice instruction into an executable management controller control instruction, writing the executable management controller control instruction into the direct execution memory, and sending the executable management controller control instruction to the direct execution memory; therefore, the executable management controller control instruction can be executed. Therefore, the instruction execution convenience of the management controller is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of servers, and in particular to a management controller, a voice command transmission system and a voice command transmission method. Background Art

[0002] The management controller is an embedded management system used to monitor the operation of the server. Currently, operators usually issue instructions to the management controller through a terminal interface to control the server. However, issuing instructions through the terminal interface requires operators to manually perform a series of operations on the terminal interface, which is not convenient enough.

[0003] It can be seen that how to improve the convenience of managing controller instruction execution is a problem that technical personnel in this field need to solve. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a management controller, a voice command transmission system and a voice command transmission method, which can improve the convenience of executing the commands of the management controller. The specific scheme is as follows:

[0005] In a first aspect, the present invention provides a management controller, comprising:

[0006] An integrated circuit built-in audio bus controller, used to obtain voice command data through an integrated circuit built-in audio bus interface, extract a check code from the voice command data, and use the check code to check the voice command in the voice command data;

[0007] A direct execution memory for storing the voice commands that pass the verification;

[0008] A speech recognition and conversion module connected to the direct execution memory is used to read speech instructions from the direct execution memory, convert the speech instructions into executable management controller control instructions, and write the executable management controller control instructions into the direct execution memory so as to execute the executable management controller control instructions.

[0009] Optionally, the speech recognition and conversion module is further used for:

[0010] Recording the storage address of the voice command in the direct execution memory;

[0011] Accordingly, writing the executable management controller control instruction into the direct execution memory includes:

[0012] The executable management controller control instruction is written into the direct execution memory based on the storage address, covering the voice instruction.

[0013] Optionally, the integrated circuit built-in audio bus controller further includes a decryption module for:

[0014] Decrypting the voice command data to obtain decrypted voice command data;

[0015] Correspondingly, extracting a verification code from the voice command data, and using the verification code to verify the voice command in the voice command data includes:

[0016] A verification code is extracted from the decrypted voice instruction data, and the voice instruction in the voice instruction data is verified using the verification code.

[0017] Optionally, a command execution module is also included, which is used to:

[0018] An executable management controller control instruction is read from the direct execution memory and executed.

[0019] Optionally, a two-wire serial bus connected to an audio codec is included for:

[0020] The audio codec is configured to process the signal collected by the microphone through the audio codec and transmit the voice command data to the audio bus controller built into the integrated circuit.

[0021] Optionally, the speech recognition and conversion module is specifically used for:

[0022] The voice command is converted into an executable management controller control command by using a voice conversion network model, wherein the training process of the voice conversion network model includes:

[0023] Acquire an initial voice data set, wherein the initial voice data in the initial voice data set are voice instructions of a management controller recorded for different personnel;

[0024] Using a preset processing strategy to process the initial speech data in the initial speech data set to obtain processed speech data, wherein the preset processing strategy includes one or more processing strategies of pitch modulation, timbre modulation, and background sound modulation;

[0025] Adding label information to the processed voice data to obtain a training sample set, wherein the label information is an executable management controller control instruction corresponding to the processed voice data;

[0026] The deep learning network model is trained using the training sample set. During the training process, the training loss is calculated using the output results of the deep learning network model and the label information of the input processed speech data, and the deep learning network model is updated based on the training loss to obtain an updated deep learning network model. When the training end condition is met, the current updated deep learning network model is determined as the speech conversion network model.

[0027] In a second aspect, the present invention discloses a management controller voice command transmission system, comprising the aforementioned management controller, an audio codec connected to the integrated circuit built-in audio bus controller, and a microphone connected to the audio codec; wherein,

[0028] The microphone is used to collect voice commands;

[0029] The audio encoder is used to convert the voice instructions output by the microphone into a format corresponding to the integrated circuit built-in audio bus, so as to transmit it to the integrated circuit built-in audio bus controller through the integrated circuit built-in audio bus interface.

[0030] In a third aspect, the present invention discloses a management controller voice command transmission method, comprising:

[0031] Using the built-in audio bus controller of the integrated circuit, obtaining voice command data through the built-in audio bus interface of the integrated circuit, extracting a verification code from the voice command data, and verifying the voice command in the voice command data using the verification code;

[0032] storing the verified voice command in the direct execution memory;

[0033] The voice instructions are read from the direct execution memory and converted into executable management controller control instructions, and the executable management controller control instructions are written into the direct execution memory so as to execute the executable management controller control instructions.

[0034] In a fourth aspect, the present invention discloses a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the aforementioned management controller voice instruction transmission method are implemented.

[0035] In a fifth aspect, the present invention discloses a computer program product, including a computer program / instruction, which, when executed by a processor, implements the steps of the aforementioned management controller voice instruction transmission method.

[0036] It can be seen from the above scheme that the present invention provides a management controller, including: an integrated circuit built-in audio bus controller, used to obtain voice command data through the integrated circuit built-in audio bus interface, and extract a verification code from the voice command data, and use the verification code to verify the voice instructions in the voice command data; a direct execution memory, used to store voice instructions that pass the verification; a voice recognition and conversion module connected to the direct execution memory, used to read voice instructions from the direct execution memory, and convert the voice instructions into executable management controller control instructions, and write the executable management controller control instructions into the direct execution memory so as to execute the executable management controller control instructions.

[0037] It can be seen that the beneficial effects of the present invention are: the integrated circuit built-in audio bus controller of the management controller obtains voice command data through the integrated circuit built-in audio bus interface, and extracts the verification code from the voice command data to verify the voice command, and the voice command that passes the verification is stored in the direct execution memory, and the voice recognition and conversion module connected to the direct execution memory reads the voice command from the direct execution memory and converts it into an executable management controller control instruction, and stores it back to the direct execution memory, waiting for the executable management controller control instruction to be executed. In this way, by setting the integrated circuit built-in audio bus controller, the direct execution memory and the voice recognition and conversion module in the management controller, the voice command can be transmitted, verified and converted, and the executable management controller control instruction can be obtained and then executed. In this way, voice control based on the management controller can be realized, and data integrity and accuracy can be guaranteed, thereby improving the convenience of executing the management controller instructions.

[0038] Correspondingly, a management controller voice instruction transmission system, a management controller voice instruction transmission method, a computer-readable storage medium, and a computer program product provided by the present invention also have the above-mentioned technical effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0040] Figure 1 A schematic diagram of a management controller structure provided by an embodiment of the present invention;

[0041] Figure 2 A circuit connection diagram of a management controller voice command transmission system provided by an embodiment of the present invention;

[0042] Figure 3 A schematic diagram of I2S signal transmission with CRC added provided by an embodiment of the present invention;

[0043] Figure 4 A schematic diagram of direct execution memory storage provided by an embodiment of the present invention;

[0044] Figure 5 A flow chart of a method for transmitting voice commands to a management controller provided by an embodiment of the present invention;

[0045] Figure 6 A structural diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0047] The terms "including" and "having" in the specification of the present invention and the above-mentioned drawings, as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device including a series of steps or units is not limited to the listed steps or units, but may include steps or units that are not listed.

[0048] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0049] Next, a management controller provided by an embodiment of the present invention is introduced in detail. Figure 1 A management controller is provided in an embodiment of the present invention, and the management controller includes:

[0050] The integrated circuit built-in audio bus controller 11 is used to obtain voice command data through the integrated circuit built-in audio bus interface, extract a check code from the voice command data, and use the check code to check the voice command in the voice command data;

[0051] A direct execution memory 12, used for storing the voice commands that pass the verification;

[0052] The speech recognition and conversion module 13 connected to the direct execution memory is used to read the speech instructions from the direct execution memory, convert the speech instructions into executable management controller control instructions, and write the executable management controller control instructions into the direct execution memory so as to execute the executable management controller control instructions.

[0053] The management controller may be one of a BMC (Baseboard Management Controller), an Integrated Light-Out (iLO) system, an Integrated Dell Remote Access Control (IDRAC), or a processing chip.

[0054] In an optional implementation, the check code may be a cyclic redundancy check. The embodiment of the present invention may extract the check code from the voice command data, calculate the voice command based on a preset polynomial to obtain a calculated check code, and compare the calculated check code with the check code extracted from the voice command data. If they are consistent, the check passes, otherwise, the check fails. The check code extracted from the voice command data is the check code calculated by the sender of the voice command data based on the preset polynomial. The integrated circuit built-in audio bus controller 11 may also be used to store the voice command that has passed the check into a direct execution memory.

[0055] In an optional embodiment, the integrated circuit built-in audio bus controller also includes a decryption module, which is used to: decrypt the voice command data to obtain decrypted voice command data; correspondingly, extract a verification code from the voice command data, and use the verification code to verify the voice commands in the voice command data, including: extracting the verification code from the decrypted voice command data, and using the verification code to verify the voice commands in the voice command data.

[0056] That is, in the embodiment of the present invention, the decryption module can also be implemented in the audio bus controller built into the integrated circuit. The encrypted voice command data is encrypted, that is, the voice command data is encrypted during the transmission process, thus further improving the data security.

[0057] The executable management controller control instruction is an executable management controller control instruction, that is, an instruction that can be executed by a processor to remotely manage and monitor server hardware. For example, the executable management controller control instruction may be an executable baseboard management controller control instruction.

[0058] Among them, the flash (i.e., flash memory) in the direct execution memory 12 is divided according to a preset sector size such as 4KB to obtain multiple sectors, each of which can be used as an independent storage unit to read, write and erase data, and multiple fixed-size sectors can store a voice command or an executable management controller control instruction. XIP flash is divided according to the 4KB sector size to provide greater flexibility and reliability to meet the needs of different application scenarios. In addition, an end identifier can be added to the executable management controller control instruction to ensure that the complete executable management controller control instruction is read.

[0059] In an optional embodiment, the speech recognition and conversion module 13 is also used to: record the storage address of the voice instruction in the direct execution memory; accordingly, writing the executable management controller control instruction into the direct execution memory includes: writing the executable management controller control instruction into the direct execution memory based on the storage address, overwriting the voice instruction.

[0060] The embodiment of the present invention can delete the voice instruction in the direct execution memory based on the storage address of the voice instruction in the direct execution memory, and write the converted executable management controller control instruction into the position corresponding to the storage address in the direct execution memory.

[0061] In an optional embodiment, the speech recognition and conversion module is specifically used to: use a speech conversion network model to convert speech instructions into executable management controller control instructions, wherein the training process of the speech conversion network model includes: obtaining an initial speech data set, wherein the initial speech data in the initial speech data set are management controller voice instructions recorded for different personnel; using a preset processing strategy to process the initial speech data in the initial speech data set to obtain processed speech data, wherein the preset processing strategy includes one or more processing strategies of pitch modulation, timbre modulation, and background sound modulation; adding label information to the processed speech data to obtain a training sample set, wherein the label information is the executable management controller control instruction corresponding to the processed speech data; using the training sample set to train a deep learning network model, during the training process, using the output result of the deep learning network model and the label information of the input processed speech data to calculate the training loss, and updating the deep learning network model based on the training loss to obtain an updated deep learning network model, and when the training end condition is met, the current updated deep learning network model is determined as the speech conversion network model.

[0062] In an embodiment of the present invention, the recorded management controller voice instructions may include device information management instructions, server status monitoring management instructions, remote control management instructions, maintenance management instructions, and visual console interface instructions. The device information management instructions may include instructions for obtaining server information: querying and recording detailed information of the server, including model, manufacturer, production date, production and technical information of each component, chassis information, motherboard information, etc. Instructions for obtaining BMC information: querying the information of the BMC itself, such as server host name, IP address, BMC firmware version, etc. The server status monitoring management instructions may include: instructions for detecting the temperature, voltage and other health status of each component of the server (such as CPU, memory, hard disk, fan, frame, etc.), fan speed adjustment instructions, adjusting the fan speed in real time according to the situation of the temperature collection point, ensuring that the server will not overheat, and controlling the overall power consumption within a reasonable range. The remote control management instructions may include: remote power on and off: sending instructions through the BMC to realize remote power on or off of the server. Remote restart: when necessary, sending a restart instruction through the BMC to restore the normal operation of the server. Remote firmware update: uploading and installing a new firmware version through the BMC to fix known problems or improve performance. Remote system installation: When the remote PC is connected to the network of the BMC management port, the operating system iso image in the PC disk can be directly used for installation. Maintenance management instructions may include: Log management instructions: query, download or clear the BMC log file for troubleshooting or performance analysis. User management instructions: add, delete or modify BMC user accounts and permissions. BIOS management instructions: access and modify BIOS settings such as boot sequence and hardware configuration through BMC. Alarm management instructions: view, confirm or clear alarm information generated by BMC. Visual console interface instructions may include BMC Web interface login instructions: log in to the BMC Web interface through the IP address configured by BMC in the LAN to achieve visual management. View device list instructions: view the usage status of all devices in the device list interface. View device details instructions: view detailed information such as CPU usage, hardware temperature, memory usage, network transceiver rate, disk usage, IP address, etc. of a single device in the device details interface.

[0063] Furthermore, the management controller also includes a command execution module, which is used to:

[0064] An executable management controller control instruction is read from the direct execution memory and executed.

[0065] Furthermore, the management controller also includes a two-wire serial bus connected to the audio codec, which is used to configure the audio codec so that the audio codec processes the signal collected by the microphone and transmits the voice command data to the audio bus controller built into the integrated circuit. That is, the two-wire serial bus initializes and configures the audio codec so that the audio codec processes the signal collected by the microphone and transmits the voice command data to the audio bus controller built into the integrated circuit. The audio codec may also include an encryption module for encrypting the voice command data.

[0066] It can be seen that the beneficial effects of the present invention are: the integrated circuit built-in audio bus controller of the management controller obtains voice command data through the integrated circuit built-in audio bus interface, and extracts the verification code from the voice command data to verify the voice command, and the voice command that passes the verification is stored in the direct execution memory, and the voice recognition and conversion module connected to the direct execution memory reads the voice command from the direct execution memory and converts it into an executable management controller control instruction, and stores it back to the direct execution memory, waiting for the executable management controller control instruction to be executed. In this way, by setting the integrated circuit built-in audio bus controller, the direct execution memory and the voice recognition and conversion module in the management controller, the voice command can be transmitted, verified and converted, and the executable management controller control instruction can be obtained and then executed. In this way, voice control based on the management controller can be realized, and data integrity and accuracy can be guaranteed, thereby improving the convenience of executing the management controller instructions.

[0067] Furthermore, an embodiment of the present invention discloses a management controller voice command transmission system, comprising the management controller disclosed in the aforementioned embodiment, an audio codec connected to the integrated circuit built-in audio bus controller, and a microphone connected to the audio codec; wherein the microphone is used to collect voice commands; the audio encoder is used to convert the voice commands output by the microphone into a format corresponding to the integrated circuit built-in audio bus, so as to transmit them to the integrated circuit built-in audio bus controller through the integrated circuit built-in audio bus interface.

[0068] See also Figure 2 As shown, Figure 2A circuit connection diagram of a management controller voice command transmission system provided by an embodiment of the present invention. Taking BMC as an example, it can be any of the aforementioned other management controllers. The embodiment of the present invention ensures the accuracy of data by adding CRC check in I2S data transmission; further ensures the integrity of data by adding encryption and decryption modules in CODEC module (i.e., audio codec) and I2Scontroller (i.e., integrated circuit built-in audio bus controller); adds XIP module (i.e., direct execution memory) in BMC, which can be XIP type Nor Flash, shortening the execution time of instructions; adds voice recognition and conversion module in BMC, realizing the execution of voice control BMC instructions. Among them, I2S (i.e., Inter-IC Sound, integrated circuit built-in audio bus) is a serial interface standard applied to digital audio transmission, used to solve the problem of transmitting audio data between integrated circuits. The I2S protocol defines the transmission format, timing and control signal of audio data. During the work of audio acquisition and output, the communication between AD and the main chip is all carried out through I2S. I2S is a digital audio interface that can realize the transmission of voice signals. CODEC is a codec chip used for audio data conversion, realizing the mutual conversion between analog signals and digital signals. Use I2C (two-wire serial bus) to initialize and configure the CODEC. XIP (eXecute In Place) means that the CPU reads the program code directly from the memory and executes it without reading it into the memory. The application can run directly in the flash memory without reading the code into the system RAM. The command execution module can be a hardware module, and the command execution can also be implemented by software. The software obtains the instructions in XIP and executes them in sequence. Figure 2 All modules involved are hardware modules.

[0069] like Figure 2 As shown, the microphone is connected to the CODEC module to transmit voice commands. The CODEC module converts the input analog signal into an output storable digital signal. At the same time, the CODEC integrates a hardware encryption module to encrypt the converted digital signal. After that, the encrypted signal is transmitted to the I2S controller through the I2S interface. The hardware decryption module integrated in the I2S controller decrypts the encrypted signal to obtain the correct voice command, which is directly stored in the nor flash that can be executed on site. The speech recognition and conversion model obtains the voice signal from the XIP memory, and after processing, obtains the BMC executable instruction. The instruction overwrites the stored voice instruction, and then the BMC obtains the executable instruction from the XIP memory in turn to realize the BMC remote voice command control. The embodiment of the present invention mainly includes I2S signal transmission, hardware encryption and decryption module, XIP storage module, and speech recognition and conversion module.

[0070] Among them, I2S signal transmission involves three transmission lines: SCK serial clock signal; WS channel selection signal, also called LRCK, used to switch left and right channel data; SD serial data signal, which is the actual audio data. In the I2S data transmission process, the left alignment mode is selected, and the CRC check code is placed in the last 8 bits of the data transmission. When the I2S controller receives the decrypted voice signal, the CRC check code is used to verify the accuracy of the signal transmission. For further information, see Figure 3 As shown, Figure 3 A schematic diagram of I2S signal transmission with CRC added provided by an embodiment of the present invention.

[0071] The embodiment of the present invention provides a hardware encryption and decryption module, and the encryption and decryption module used in the system can be SM2 implemented in hardware language. In the SM2 public key encryption algorithm, the sender encrypts the message into a ciphertext with the public key of the receiver, and the receiver decrypts the received ciphertext with his own private key to restore it to the original message. Keys are divided into public keys and private keys. The public key is given to others and can be made public within a certain range, while the private key is kept by oneself and must be kept confidential. The public key can be calculated from the private key; it is quite difficult to calculate the private key from the public key, and it is impossible at this stage. SM2 is an elliptic curve public key cryptography algorithm. SM2 is asymmetric encryption based on ECC. Because the algorithm is based on ECC, its signing speed and key generation speed are faster than RSA. ECC 256 bits (SM2 uses a type of ECC 256 bits) has higher security strength than RSA 2048 bits, but the operation speed is faster than RSA. The SM2 encryption and decryption process is a logical calculation process. The embodiment of the present invention can use the hardware description language Verilog to implement the different calculation formulas involved therein, and call the registers of different modules according to the logical calculation process to realize the logical calculation. The encryption and decryption module is implemented by hardware, and the software finally realizes the encryption and decryption calculation by configuring the register (the register can be read and written). A random number generator, an elliptic point curve point calculation module, a point to character conversion module, a KDF key derivation function module, a bit string / character string concatenation function module, a bit string bitwise XOR operation module, a hash operation module and an output module can be realized, wherein the encryption process may include:

[0072] A1: Generate random numbers using a random number generator ; Implemented through a random number generator;

[0073] A2: Calculate elliptic curve points , convert the data type of C1 into a bit string; G is a base point of the elliptic curve, and its order is a prime number. It is implemented through the elliptic point curve point calculation module and the point to character conversion module;

[0074] A3: Calculate elliptic curve points , if S is a point at infinity, report an error and exit; is the public key, the cofactor of G of order n is h; it is implemented by the elliptic point curve point calculation module;

[0075] A4: Calculate elliptic curve points , convert the data types of coordinates x2 and y2 into bit strings; this is achieved through the ellipse point curve point calculation module and the point to character conversion module;

[0076] A5: Calculation , if t is a string of all 0 bits, then return to A1; KDF(): key derivation function; x∥y: concatenation of x and y; integer klen represents the bit length of the key data to be obtained; KDF key derivation function module, bit string / character string concatenation function module implementation;

[0077] A6: Calculation ; M: message to be encrypted; implemented through the bit-by-bit XOR operation module of the bit string.

[0078] A7: Calculation ;Hash():cryptographic hash function. Implemented by hash operation module.

[0079] A8: Output ciphertext . This is done through output modules.

[0080] According to the encryption algorithm flow and decryption algorithm flow, the specific modules that need to be implemented by hardware are determined. The calculation formula involved in the decryption algorithm is consistent with that in the encryption algorithm, so the hardware modules that need to be implemented are also consistent with those in the encryption algorithm.

[0081] For further information, see Figure 4 As shown, Figure 4 A schematic diagram of direct execution memory storage provided for an embodiment of the present invention. XIP flash is used in the XIP storage module. XIP, eXecute In Place, means in-chip execution, in-place execution, which means that the CPU reads the program code directly from the memory for execution without reading it into the memory. The application can run directly in the flash memory without having to read the code into the system RAM. The XIP flash is divided according to the 4KB sector size, and the voice signal received by the I2S controller is directly stored in the sector of the XIP flash (multiple fixed-size sectors can store one instruction). The voice recognition module obtains input from a fixed-size sector block, and after being processed by the network model, it obtains the BMC instruction information. These output instructions will overwrite the voice signal of the corresponding sector, and at the same time, the interrupt triggers the command execution module to process the BMC instructions (that is, the executable management controller control instructions).

[0082] The speech recognition and conversion module can use network models for speech recognition. The network models used include but are not limited to DNN-HMM, Attention-based RNN, Transformer, etc. The corpus for model training is built by yourself. Several different people in the company can record the specified BMC instructions. After recording, they are modulated by pitch, timbre, background sound, etc. to generate tens of thousands of corpora as input for the network model. The features of speech recognition are the specified BMC keywords. The output of the network model is the specified BMC instruction.

[0083] It should be pointed out that there is no data check in the I2S transmission protocol. If it is used directly to transmit BMC instructions, data loss and data interference may occur, resulting in errors in the final BMC instruction execution; at the same time, after receiving data from the I2S interface, DMA (Direct Memory Access) is often used to move data from FIFO to memory, and then process it. This process causes the processing of commands to be affected by the size of FIFO. The BMC voice command transmission system based on I2S proposed in the embodiment of the present invention. The correct transmission of voice signals is achieved by modifying the I2S signal transmission (adding CRC check); the correct transmission of voice commands is further guaranteed by the encryption and decryption hardware module, avoiding interference caused by transmission channels and electromagnetic radiation; the rapid recognition and processing of voice signals is achieved by the XIP storage module, and the BMC instructions overwrite the voice signals to avoid waste of storage space; the accuracy of BMC instruction recognition is ensured by constructing the input feature parameters, output targets and training corpus of the network model. The correct transmission of voice signals is achieved through customized I2S signal transmission and integrated SM2 encryption and decryption hardware modules. The conversion of voice commands to BMC executable commands is accelerated through the XIP module. The correct BMC execution commands are obtained by building network model training corpus and feature parameters.

[0084] The embodiment of the present invention provides a voice command transmission system integrating CODEC, I2S controller, XIP storage module, voice recognition network module, interrupt mechanism, and BMC command execution module. Although the implementation mode of the present invention is described as above, the above description and definition are only for the convenience of understanding the implementation mode of the present invention, and are not used to limit the present invention. Any modification and change made without departing from the spirit and scope of the present invention are within the protection scope of the present invention.

[0085] The embodiment of the present invention expands the application scenarios of BMC, and can integrate a voice receiving module and a voice recognition and conversion module in the BMC. The voice receiving module can include an integrated circuit built-in audio bus controller. The operator can transmit voice signals to the BMC through a microphone, and the audio instructions can be converted into instructions executable by the BMC through voice recognition technology.

[0086] For further information, see Figure 5 As shown, a method for transmitting voice commands of a management controller includes:

[0087] Step S11: using the audio bus controller built into the integrated circuit, obtaining voice command data through the audio bus interface built into the integrated circuit, extracting a verification code from the voice command data, and using the verification code to verify the voice command in the voice command data.

[0088] The embodiment of the present invention can implement the management controller voice instruction transmission method based on the management controller and the management controller voice instruction transmission system disclosed in the above embodiments.

[0089] In an optional implementation, the integrated circuit built-in audio bus controller also includes a decryption module. The embodiment of the present invention can use the decryption module to decrypt the voice command data to obtain decrypted voice command data; accordingly, a verification code is extracted from the voice command data, and the voice commands in the voice command data are verified using the verification code, including: extracting the verification code from the decrypted voice command data, and verifying the voice commands in the voice command data using the verification code.

[0090] Step S12: storing the verified voice command in the direct execution memory.

[0091] Step S13: reading the voice instruction from the direct execution memory, converting the voice instruction into an executable management controller control instruction, and writing the executable management controller control instruction into the direct execution memory so as to execute the executable management controller control instruction.

[0092] The embodiment of the present invention can utilize a speech recognition and conversion module to read speech instructions from the direct execution memory, convert the speech instructions into executable management controller control instructions, and write the executable management controller control instructions into the direct execution memory.

[0093] Furthermore, the storage address of the voice instruction in the direct execution memory can also be recorded; accordingly, writing the executable management controller control instruction into the direct execution memory includes: writing the executable management controller control instruction into the direct execution memory based on the storage address, overwriting the voice instruction.

[0094] Furthermore, the embodiment of the present invention may also read an executable management controller control instruction from the direct execution memory and execute the executable management controller control instruction.

[0095] Furthermore, the embodiment of the present invention can also utilize a two-wire serial bus connected to the audio codec to configure the audio codec so as to process the signal collected by the microphone through the audio codec and transmit the voice command data to the audio bus controller built into the integrated circuit.

[0096] In an optional implementation, the voice command may be converted into an executable management controller control command using a voice conversion network model, wherein the training process of the voice conversion network model includes:

[0097] Acquire an initial voice data set, wherein the initial voice data in the initial voice data set are voice instructions of a management controller recorded for different personnel; use a preset processing strategy to process the initial voice data in the initial voice data set to obtain processed voice data, wherein the preset processing strategy includes one or more processing strategies of pitch modulation, timbre modulation, and background sound modulation; add label information to the processed voice data to obtain a training sample set, wherein the label information is an executable management controller control instruction corresponding to the processed voice data; use the training sample set to train a deep learning network model, and during the training process, use the output result of the deep learning network model and the label information of the input processed voice data to calculate the training loss, and update the deep learning network model based on the training loss to obtain an updated deep learning network model, and when the training end condition is met, the current updated deep learning network model is determined as a voice conversion network model.

[0098] Figure 6 A structural diagram of an electronic device provided by an embodiment of the present invention, such as Figure 6 As shown, the electronic device includes: a memory 60 for storing a computer program;

[0099] The processor 61 is used to implement the steps of the method for transmitting voice instructions to the controller in the above embodiment when executing the computer program.

[0100] The electronic device provided in this embodiment may include a server and the like.

[0101] Among them, the processor 61 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 61 can be implemented in at least one hardware form of digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 61 may also include a main processor and a coprocessor. The main processor is a processor for processing data in the awake state, also known as a central processing unit (CPU); the coprocessor is a low-power processor for processing data in the standby state. In some embodiments, the processor 61 may be integrated with a graphics processing unit (GPU), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 61 may also include an artificial intelligence (AI) processor, which is used to process computing operations related to machine learning.

[0102] The memory 60 may include one or more computer-readable storage media, which may be non-transitory. The memory 60 may also include a high-speed random access memory, and a non-volatile memory, such as one or more disk storage devices, flash memory storage devices. In this embodiment, the memory 60 is at least used to store the following computer program 601, wherein, after the computer program is loaded and executed by the processor 61, it can implement the relevant steps of the management controller voice instruction transmission method disclosed in any of the aforementioned embodiments. In addition, the resources stored in the memory 60 may also include an operating system 602 and data 603, etc., and the storage method may be short-term storage or permanent storage. Among them, the operating system 602 may include Windows, Unix, Linux, etc. Data 603 may include but is not limited to training data, etc.

[0103] In some embodiments, the electronic device may further include a display screen 62 , an input / output interface 63 , a communication interface 64 , a power supply 65 , and a communication bus 66 .

[0104] Those skilled in the art will understand that Figure 6 The structure shown in the figure does not constitute a limitation on the electronic device, and may include more or fewer components than shown in the figure.

[0105] It is understandable that if the management controller voice command transmission method in the above embodiment is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or part of the contribution to the current technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium to execute all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), electrically erasable programmable ROM, register, hard disk, removable disk, CD-ROM, magnetic disk or optical disk and other media that can store program codes.

[0106] Based on this, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned management controller voice instruction transmission method are implemented.

[0107] An embodiment of the present invention discloses a computer program product, including a computer program / instruction. When the computer program / instruction is executed by a processor, the steps of the management controller voice instruction transmission method disclosed in the above embodiment are implemented.

[0108] The above is a detailed introduction to a management controller, a voice command transmission system, and a voice command transmission method provided by an embodiment of the present invention. The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same and similar parts between the various embodiments can be referred to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0109] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0110] The above is a detailed introduction to a management controller, a voice command transmission system and a voice command transmission method provided by the present invention. Specific examples are used in this article to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the scope of protection of the claims of the present invention.

Claims

1. A management controller, characterized in that: include: An integrated circuit built-in audio bus controller, used to obtain voice command data through an integrated circuit built-in audio bus interface, extract a check code from the voice command data, and use the check code to check the voice command in the voice command data; A direct execution memory for storing the voice commands that pass the verification; A speech recognition and conversion module connected to the direct execution memory is used to read speech instructions from the direct execution memory, convert the speech instructions into executable management controller control instructions, and write the executable management controller control instructions into the direct execution memory so as to execute the executable management controller control instructions.

2. The management controller according to claim 1, characterized in that: The speech recognition and conversion module is also used for: Recording the storage address of the voice command in the direct execution memory; Accordingly, writing the executable management controller control instruction into the direct execution memory includes: The executable management controller control instruction is written into the direct execution memory based on the storage address, covering the voice instruction.

3. The management controller according to claim 1, characterized in that: The integrated circuit built-in audio bus controller further includes a decryption module for: Decrypting the voice command data to obtain decrypted voice command data; Correspondingly, extracting a verification code from the voice command data, and using the verification code to verify the voice command in the voice command data includes: A verification code is extracted from the decrypted voice instruction data, and the voice instruction in the voice instruction data is verified using the verification code.

4. The management controller according to claim 1, characterized in that: It also includes a command execution module for: An executable management controller control instruction is read from the direct execution memory and executed.

5. The management controller according to claim 1, characterized in that: Also included is a two-wire serial bus that interfaces with the audio codec for: The audio codec is configured to process the signal collected by the microphone through the audio codec and transmit the voice command data to the audio bus controller built into the integrated circuit.

6. The management controller according to any one of claims 1 to 5, characterized in that: The speech recognition and conversion module is specifically used for: The voice command is converted into an executable management controller control command by using a voice conversion network model, wherein the training process of the voice conversion network model includes: Acquire an initial voice data set, wherein the initial voice data in the initial voice data set are voice instructions of a management controller recorded for different personnel; Using a preset processing strategy to process the initial speech data in the initial speech data set to obtain processed speech data, wherein the preset processing strategy includes one or more processing strategies of pitch modulation, timbre modulation, and background sound modulation; Adding label information to the processed voice data to obtain a training sample set, wherein the label information is an executable management controller control instruction corresponding to the processed voice data; The deep learning network model is trained using the training sample set. During the training process, the training loss is calculated using the output results of the deep learning network model and the label information of the input processed speech data, and the deep learning network model is updated based on the training loss to obtain an updated deep learning network model. When the training end condition is met, the current updated deep learning network model is determined as the speech conversion network model.

7. A management controller voice command transmission system, characterized in that: The method comprises a management controller as claimed in any one of claims 1 to 6, an audio codec connected to the integrated circuit built-in audio bus controller, and a microphone connected to the audio codec; wherein: The microphone is used to collect voice commands; The audio encoder is used to convert the voice instructions output by the microphone into a format corresponding to the integrated circuit built-in audio bus, so as to transmit it to the integrated circuit built-in audio bus controller through the integrated circuit built-in audio bus interface.

8. A method for transmitting voice commands of a management controller, characterized in that: include: Using the built-in audio bus controller of the integrated circuit, obtaining voice command data through the built-in audio bus interface of the integrated circuit, extracting a verification code from the voice command data, and verifying the voice command in the voice command data using the verification code; storing the verified voice command in the direct execution memory; The voice instructions are read from the direct execution memory and converted into executable management controller control instructions, and the executable management controller control instructions are written into the direct execution memory so as to execute the executable management controller control instructions.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the method for transmitting voice instructions to the management controller as claimed in claim 8 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the method for transmitting voice instructions to a management controller as claimed in claim 8 are implemented.

Images