Information processing method and device, equipment, medium and product

Abstract

The invention discloses an information processing method and device, equipment, a medium and a product. The method comprises the following steps: acquiring voice information; the voice information comprises a user instruction to be executed; determining a target slot position value of the voice information; expanding dictionary vocabularies based on the target slot position value; the extended dictionary vocabulary is used for executing the user instruction.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an information processing method, device, equipment, medium and product. Background Art

[0002] In the related art, the semantic recognition model has difficulty in recognizing speech information outside the vocabulary and cannot meet the user's usage needs. Summary of the invention

[0003] Embodiments of the present application provide an information processing method, apparatus, device, medium, and product.

[0004] The technical solution of the embodiment of the present application is implemented as follows:

[0005] An information processing method, the method comprising:

[0006] Acquiring voice information; the voice information includes a user instruction to be executed;

[0007] Determining a target slot value of the voice information;

[0008] The dictionary vocabulary is expanded based on the target slot value; the expanded dictionary vocabulary is used to execute the user instruction.

[0009] In the above scheme, determining the target slot value of the voice information includes:

[0010] Converting the voice information into text information;

[0011] Inputting the text information into a slot recognition model to obtain a classification slot value of the user instruction; the classification slot value includes a numerical slot value and / or a text slot value;

[0012] The target slot value is determined based on the classification slot value.

[0013] In the above solution, determining the target slot value based on the classification slot value includes:

[0014] Input the text class slot value into a first preset model to obtain a vector value and text segmentation corresponding to the text class slot value;

[0015] Determine a first slot value corresponding to the vector value based on the first mapping relationship;

[0016] Determine a second slot value corresponding to the text segmentation based on the second mapping relationship;

[0017] A first target slot value is determined based on the first slot value and the second slot value.

[0018] In the above solution, the expanding the dictionary vocabulary based on the target slot value includes:

[0019] Update the first text standard word in the first mapping relationship based on the first target slot value;

[0020] The second text standard word in the second mapping relationship is updated based on the first target slot value; the dictionary vocabulary includes the first text standard word and the second text standard word.

[0021] In the above solution, determining the target slot value based on the classification slot value includes:

[0022] Inputting the numerical slot value into a second preset model to obtain a numerical category corresponding to the numerical slot value;

[0023] Converting the numerical value category into numerical text based on a third mapping relationship;

[0024] A second target slot value is determined based on the numeric text.

[0025] In the above solution, the expanding the dictionary vocabulary based on the target slot value includes:

[0026] The numerical standard words in the third mapping relationship are updated based on the second target slot value; the dictionary vocabulary includes the numerical standard words.

[0027] An information processing device, comprising:

[0028] An acquisition unit, configured to acquire voice information; the voice information includes a user instruction to be executed;

[0029] A processing unit, configured to determine a target slot value of the voice information;

[0030] The processing unit is further used to expand the dictionary vocabulary based on the target slot value; the expanded dictionary vocabulary is used to execute the user instruction.

[0031] An electronic device comprising: a processor and a memory for storing a computer program that can be run on the processor,

[0032] Wherein, when the processor is used to run the computer program, it executes the steps of any one of the above methods.

[0033] A storage medium stores a computer program, wherein the computer program implements the steps of any of the above methods when executed by a processor.

[0034] A computer product comprises a computer program, wherein when the computer program is executed by a processor, the steps of any one of the above methods are implemented.

[0035] The information processing method, apparatus, device, medium and product provided by the embodiments of the present application obtain voice information; the voice information includes a user instruction to be executed; a target slot value of the voice information is determined; a dictionary vocabulary is expanded based on the target slot value; and the expanded dictionary vocabulary is used to execute the user instruction. In other words, the embodiments of the present application obtain voice information, determine the target slot value of the voice information, expand the dictionary vocabulary based on the target slot value, and use the dictionary vocabulary to execute the user instruction to be executed in the voice information, thereby solving the problem in the related art that the semantic recognition model is difficult to recognize voice information outside the vocabulary and cannot meet the user's usage needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 A flowchart of an information processing method provided in an embodiment of the present application;

[0037] Figure 2 A schematic diagram of a slot value identification process provided in an embodiment of the present application;

[0038] Figure 3 A schematic diagram of a process for normalizing slot values ​​provided in an embodiment of the present application;

[0039] Figure 4 A schematic diagram of a process for standardizing text slot values ​​provided in an embodiment of the present application;

[0040] Figure 5 A schematic diagram of another process for normalizing text slot values ​​provided in an embodiment of the present application;

[0041] Figure 6 A schematic diagram of a process for normalizing a numerical slot value provided in an embodiment of the present application;

[0042] Figure 7 A schematic diagram of another process for normalizing slot values ​​provided in an embodiment of the present application;

[0043] Figure 8 A schematic diagram of the structure of an information processing device provided in an embodiment of the present application;

[0044] Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] In order to make the purpose, technical solutions and advantages of the present application clearer, the technical solutions of the present application are further elaborated in detail below in conjunction with the drawings and embodiments. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of the present application.

[0046] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0047] The terms "first / second / third" involved in the present application are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first / second / third" can be interchanged with a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0048] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0049] In modern life, users can set warning conditions for the information they care about in order to provide reminders or automatic control. Usually, voice commands can be set using voice input. In related technologies, user needs are usually converted into standardized ones. The following takes numerical query and text query as examples:

[0050] 1. User query demand (query): The price rises to 18.8 yuan; slot value: 18.8 yuan; standardized value: 18.8.

[0051] 2. User query: The total market value exceeds 18.7 trillion; Slot value: 18.7 trillion; Normalized value: 187000000000000.

[0052] 3. User query: The price rises or falls by more than 4 points -> Slot value: 4 points -> Normalized value: 4%

[0053] 4. User query: Set reminder on Friday; slot value: Friday; standardized value: 20240809 (if the day is August 5).

[0054] 5. User query: Please enable intraday changes of individual stocks for me; Slot value: intraday changes of individual stocks; Normalized value: individual (there are many ways to say it in Chinese, such as: individual stock changes, intraday changes of individual stocks, stock changes, etc.).

[0055] 6. User query: Set collective bidding movement; slot value: collective bidding movement (collective bidding change [typo]); standardized value: call_auction.

[0056] In terms of numerical standardization, the relevant technology first mines out the slot value, then converts the slot value into a number, and multiplies the corresponding multiples for specific units, for example: 170,000 -> 170,000 -> 17*10000 (10,000) -> 170,000.

[0057] In terms of text standardization, the relevant technology collects relevant words into a map dictionary, such as:

[0058] {

[0059] "Individual stock intraday movement": "individual",

[0060] "Call Auction Mobile": "call_auction",

[0061] “Call auction changes”: “call_auction” ......

[0063] }

[0064] Once the corresponding map slot value is identified, it is converted into the corresponding map value. The dictionary can be used in semantic recognition models or other platforms to identify key information used in speech, but this method is difficult to identify speech information outside the vocabulary. For example: "17 yuan 89 cents", the word "mao" is not set in the rule, so it cannot be converted. It is also impossible to use prior knowledge to make judgments. For example, the word "ba" in "individual stock changes" is an irrelevant word, but it cannot be judged using rules. At the same time, this method requires manual maintenance of the dictionary and cannot be dynamically expanded. For example: "Individual stock intraday changes", "Individual stock market changes", "Individual stock information changes" actually belong to the "Individual stock changes" category, but because each user's statement is different, the map configuration can only be manually maintained in the map dictionary after discovery, and it is impossible to effectively convert words outside the map dictionary.

[0065] The embodiment of the present application provides an information processing method, referring to Figure 1 As shown, the method comprises the following steps:

[0066] Step S101: Acquire voice information.

[0067] The voice information includes user instructions to be executed.

[0068] It is understandable that the voice information may be transmitted through a user equipment (UE), where UE includes but is not limited to a mobile phone, a tablet, and a computer.

[0069] In actual applications, users can issue instructions for setting reminders or performing related operations through voice messages. User instructions may include but are not limited to numerical instructions and text instructions.

[0070] Step S102: Determine the target slot value of the voice information.

[0071] It is understandable that after acquiring the voice information, the user instructions can be classified by the slot recognition model. The slot recognition model can classify the user instructions into numerical slot values ​​and text slot values, and then perform a standardization process.

[0072] In practical applications, reference Figure 2 As shown, before inputting the voice information into the input slot recognition model, the voice information may be preprocessed, for example, the voice information may be converted into text information.

[0073] refer to Figure 3 As shown, for numerical slot values, the large language model (LLM) can be used for standardization, and for text slot values, the retrieval-augmented generation (RAG) model can be used for standardization. The target slot value can be understood as the standardized slot value.

[0074] Step S103: Expanding dictionary vocabulary based on the target slot value.

[0075] Among them, the extended dictionary vocabulary is used to execute user instructions.

[0076] In practical applications, after the large language model and retrieval enhancement generate the standardized slot value, the target slot value can be output as the result.

[0077] The dictionary vocabulary can be expanded according to the text information converted from the voice information and the standardized slot value, so as to realize the dynamic expansion of the dictionary vocabulary and quickly respond to the user's voice command.

[0078] From the above content, it can be seen that the embodiment of the present application obtains voice information, determines the target slot value of the voice information, expands the dictionary vocabulary based on the target slot value, and uses the dictionary vocabulary to execute the user instructions to be executed in the voice information, thereby solving the problem in the related technology that the semantic recognition model is difficult to recognize voice information outside the vocabulary and cannot meet the user's usage needs.

[0079] In some embodiments of the present application, determining a target slot value of voice information includes:

[0080] Convert voice information into text information;

[0081] Input the text information into the slot recognition model to obtain the classification slot value of the user instruction; the classification slot value includes a numerical slot value and / or a text slot value;

[0082] A target slot value is determined based on the classification slot value.

[0083] Understandable, reference Figure 3 As shown, after acquiring the voice information, the user instructions can be classified through the slot recognition model. The slot recognition model can classify the user instructions into numerical slot values ​​and text slot values, and then perform a standardization process.

[0084] For numerical slot values, the LLM model can be used for standardization, and for text slot values, the RAG model can be used for standardization. The target slot value can be understood as the standardized slot value.

[0085] In some embodiments of the present application, determining a target slot value based on a classification slot value includes:

[0086] Input the text class slot value into the first preset model to obtain the vector value and text segmentation corresponding to the text class slot value;

[0087] Determine a first slot value corresponding to the vector value based on the first mapping relationship;

[0088] Determine a second slot value corresponding to the text segmentation based on the second mapping relationship;

[0089] A first target slot value is determined based on the first slot value and the second slot value.

[0090] It can be understood that the first preset model can be understood as a RAG model. The RAG model can include a vector library and an inverted index library.

[0091] In practical applications, reference Figure 4 As shown, different text slot types can be pre-set, and then each text slot type is matched with a corresponding standardized Chinese word, and then each standardized Chinese word is vectorized using a text vector model (BAAI General Embedding, BGE). The processing results can be referred to in Table 1.

[0092] index key vector 1 Individual stock movement [0.12312,0.234234.....] 2 Stock intraday changes [0.12312,0.234234.....] 3 MACD Golden Cross [0.22312,0.134234.....]

[0093] Table 1

[0094] The vectorization results are used for subsequent standardized word recall. At the same time, in order to enhance the recall effect and realize multi-way parallel recall, after the standardized Chinese words are sorted out, the standardized Chinese words can be processed as inverted indexes. The text-like slot values ​​identified by the slot recognition model can enter two sub-processing processes. The first sub-process is the process of calculating the slot value vector, and the second process is the process of calculating the word segmentation. The first mapping relationship can be understood as a vector mapping relationship; the second mapping relationship can be understood as a word segmentation mapping relationship. After obtaining the first slot value and the second slot value, a union can be formed, and the union can be sorted to determine the first target slot value.

[0095] In a realistic scenario, refer to Figure 5 As shown, the standardization process of text slot values ​​can be achieved as follows:

[0096] Step S501: The text slot value identified by the slot recognition model can enter two sub-processing flows. Take the text slot value "individual stock movement" as an example.

[0097] Step S502: Calculate the embedding of the slot value through BGE. For example: individual stock movement: [0.123, 0.2312, 0.3232, ...].

[0098] Step S503: Recall a set of candidate similar words by means of vector retrieval, such as: ["individual stock movement", "individual stock intraday movement"].

[0099] Step S504: Segment the slot value. For example: individual stock movement: [individual stock movement]).

[0100] Step S505: Use the inverted index library and the Okapi BM25 ranking algorithm to recall the segmented query and calculate the top k values ​​["individual stock movement", "intraday stock movement"].

[0101] Step S506: Merge the two results and use the BGE-Ranker model to select the most relevant candidate result. For example: {"Individual stock movement": "individual"}.

[0102] In some embodiments of the present application, the dictionary vocabulary is expanded based on the target slot value, including:

[0103] Update the first text standard word in the first mapping relationship based on the first target slot value;

[0104] The second text standard word in the second mapping relationship is updated based on the first target slot value; the dictionary vocabulary includes the first text standard word and the second text standard word.

[0105] It can be understood that the text slot value identified by the slot recognition model can enter two sub-processes, the first sub-process is the slot value vector calculation process, and the second process is the word segmentation calculation process. The first mapping relationship can be understood as a vector mapping relationship; the second mapping relationship can be understood as a word segmentation mapping relationship.

[0106] In practical applications, the text slot value "individual stock movement" is used as an example. After obtaining the final result {"individual stock movement": "individual"}, the Chinese standard word corresponding to the vector in the vector mapping relationship and the Chinese standard word corresponding to the word segmentation in the word segmentation mapping relationship can be updated.

[0107] The "individual stock movement" contained in the voice information corresponds to {"individual stock movement": "individual"}, so as to expand the dictionary vocabulary and realize dynamic expansion of the dictionary vocabulary. The dictionary vocabulary is used to execute user instructions, which can improve the response efficiency.

[0108] In some embodiments of the present application, determining a target slot value based on a classification slot value includes:

[0109] Inputting the numerical slot value into the second preset model to obtain the numerical category corresponding to the numerical slot value;

[0110] Convert the numerical category into numerical text based on the third mapping relationship;

[0111] Determines the second target slot value based on the numeric text.

[0112] It can be understood that the second preset model can be understood as an LLM model. The discrimination of the numerical slots can be established on the LLM Agent link, for example: price_up, price_down belong to the price category, pp_ge belongs to the percentage category, tmc_ge belongs to the ordinary numerical category, etc. Different Agent processing sub-logics are established according to different numerical types. For example: the price module, by setting the prompt logic of calling the LLM module, the prompt specifies what the LLM needs to achieve.

[0113] The example of the set prompt is as follows: Convert the received voice into text, confirm whether the result price is correct, remove redundant characters, and convert the price unit to yuan. If the result does not contain the price, return -1. At the same time, convert the result into standard Arabic numerals, for example: 9,600 yuan, return 9,600.

[0114] In practical applications, the third mapping relationship can be understood as the mapping relationship between numerical categories and numerical results. For example, when the price rises to 18 yuan and 8 jiao, price_up corresponds to 18.8. The output result of the second target slot value can be {"18.8": "price_up"}.

[0115] In an implementable scenario, referring to Figure 6 as shown, the standardization process of the numerical category slot value can be implemented in the following way:

[0116] Step S601: Obtain the numerical category slot. For example: price_up (price rises to); pp_ge (daily increase or decrease rate exceeds); tmc_ge (total market value exceeds).

[0117] Step S602: Divide into price category, percentage category, and ordinary numerical category. For example: price category (price rises to, price drops to, total market value); percentage category (high price-earnings ratio, price-to-book ratio, turnover rate, increase or decrease rate); ordinary numerical category (number of consecutive rising days, number of consecutive falling days).

[0118] Step S603: Construct different agents according to different numerical categories. For example: In the price category model, construct a fixed prompt: Convert the input numerical value (price) into a standard Arabic numeral. Input: 18 yuan and 8 jiao and 8 fen; Output: 18.8.

[0119] Step S604: Different numerical category agents construct different pre- and post-processing flows. For example: Input the original slot value -> numerical preprocessing script -> LLM conversion -> post-conversion processing script.

[0120] In some embodiments of the present application, expanding the dictionary vocabulary based on the target slot value includes:

[0121] Updating the numerical standard words in the third mapping relationship based on the second target slot value; the dictionary vocabulary includes numerical standard words.

[0122] In practical applications, taking the example of the price rising to 18 yuan and 8 jiao, after obtaining the output result of the second target slot value 18.8, "jiao" can be updated into the numerical standard words.

[0123] As can be seen from the above content, the embodiments of the present application perform slot value standard value processing based on the LLM model and the RAG model, and different processing flows can be established according to the slot value type. It solves the problems of cumbersome rule writing for slot standardization and the inability to dynamically expand the thesaurus dictionary in the related art. It can quickly respond to the user's voice commands.

[0124] In an implementable scenario, referring to Figure 7 as shown, an information processing method can be implemented in the following way:

[0125] Step S701: Convert user voice input to text, for example, the converted text is: Set a reminder when the price rises to 18.88 yuan.

[0126] Step S702: Convert the result of the user's voice input into text, for example: set a reminder when the price rises to 18.88 yuan. The semantic understanding model recognizes the slot information of the text, and the result after recognition and sorting is {"price_up":"18.88 yuan"}

[0127] Step S704: Different processing flows will be entered according to different slot types. For example: {"price_up":"18.88"}, price_up belongs to the price type under the numerical type, and then the numerical slot value standardization flow will be entered. According to the set LLM Agent flow, the standardized slot result {"price_up":"18.88"} is finally returned. "price_up" is the slot value type, "18.88" is the original slot value, and "18.88" is the standardized slot value.

[0128] If the slot type is text type, the text slot value standardization process is used. For example: query: "turn on reminders for individual stock movement for me", the slot information proposed is {"Abnormal":"individual stock movement"}. The processing result is {"Abnormal":"individual"}. Note: "Abnormal" is the slot value type, "individual stock movement" is the original slot value, and "individual" is the standardized slot value.

[0129] Step S705: Return the final slot standard value.

[0130] From the above content, it can be seen that the embodiment of the present application obtains voice information, determines the target slot value of the voice information, expands the dictionary vocabulary based on the target slot value, and uses the dictionary vocabulary to execute the user instructions to be executed in the voice information, thereby solving the problem in the related technology that the semantic recognition model is difficult to recognize voice information outside the vocabulary and cannot meet the user's usage needs.

[0131] Based on the same inventive concept as above, Figure 8 A schematic diagram of the structure of an information processing device provided by an embodiment of the present invention, the information processing device 800 includes:

[0132] The acquisition unit 801 is used to acquire voice information; the voice information includes a user instruction to be executed;

[0133] The processing unit 802 is used to determine the target slot value of the voice information;

[0134] The processing unit 802 is further configured to expand the dictionary vocabulary based on the target slot value; the expanded dictionary vocabulary is used to execute user instructions.

[0135] In some embodiments of the present application, the processing unit 802 is further configured to convert the voice information into text information;

[0136] Input the text information into the slot recognition model to obtain the classification slot value of the user instruction; the classification slot value includes a numerical slot value and / or a text slot value;

[0137] A target slot value is determined based on the classification slot value.

[0138] In some embodiments of the present application, the processing unit 802 is further configured to input the text class slot value into the first preset model to obtain a vector value and a text segmentation corresponding to the text class slot value;

[0139] Determine a first slot value corresponding to the vector value based on the first mapping relationship;

[0140] Determine a second slot value corresponding to the text segmentation based on the second mapping relationship;

[0141] A first target slot value is determined based on the first slot value and the second slot value.

[0142] In some embodiments of the present application, the processing unit 802 is further configured to update the first text standard word in the first mapping relationship based on the first target slot value;

[0143] The second text standard word in the second mapping relationship is updated based on the first target slot value; the dictionary vocabulary includes the first text standard word and the second text standard word.

[0144] In some embodiments of the present application, the processing unit 802 is further configured to input the numerical slot value into a second preset model to obtain a numerical category corresponding to the numerical slot value;

[0145] Convert the numerical category into numerical text based on the third mapping relationship;

[0146] Determines the second target slot value based on the numeric text.

[0147] In some embodiments of the present application, the processing unit 802 is further used to update the numerical standard words in the third mapping relationship based on the second target slot value; the dictionary vocabulary includes the numerical standard words.

[0148] Based on the foregoing embodiments, an embodiment of the present application provides an electronic device, Fig. 9Schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. The electronic device 900 includes: at least one processor 901, a memory 902, and optionally, the electronic device 900 may further include at least one communication interface 903. The various components in the electronic device 900 are coupled together via a bus system 904. It can be understood that the bus system 904 is used to realize the connection and communication between these components. In addition to the data bus, the bus system 904 also includes a power bus, a control bus, and a status signal bus. However, for the sake of clarity, Figure 3 Various buses are labeled as bus system 904.

[0149] Based on the hardware implementation of the above program modules, the communication interface 903 can exchange information with other communication devices;

[0150] The processor 901 is connected to the communication interface 903 to implement information interaction with other communication devices, and is used to execute the method provided by one or more of the above technical solutions when running the computer program;

[0151] A memory 902 on which the computer program is stored.

[0152] Specifically, the processor 901 is used to obtain voice information; the voice information includes a user instruction to be executed;

[0153] Determine the target slot value of the voice information;

[0154] The dictionary vocabulary is expanded based on the target slot value; the expanded dictionary vocabulary is used to execute the user instruction.

[0155] In some embodiments of the present application, the processor 901 is configured to convert voice information into text information;

[0156] Input the text information into the slot recognition model to obtain the classification slot value of the user instruction; the classification slot value includes a numerical slot value and / or a text slot value;

[0157] A target slot value is determined based on the classification slot value.

[0158] In some embodiments of the present application, the processor 901 is used to input the text class slot value into the first preset model to obtain the vector value and text segmentation corresponding to the text class slot value;

[0159] Determine a first slot value corresponding to the vector value based on the first mapping relationship;

[0160] Determine a second slot value corresponding to the text segmentation based on the second mapping relationship;

[0161] A first target slot value is determined based on the first slot value and the second slot value.

[0162] In some embodiments of the present application, the processor 901 is configured to update the first text standard word in the first mapping relationship based on the first target slot value;

[0163] The second text standard word in the second mapping relationship is updated based on the first target slot value; the dictionary vocabulary includes the first text standard word and the second text standard word.

[0164] In some embodiments of the present application, the processor 901 is configured to input the numerical slot value into a second preset model to obtain a numerical category corresponding to the numerical slot value;

[0165] Convert the numerical category into numerical text based on the third mapping relationship;

[0166] Determines the second target slot value based on the numeric text.

[0167] In some embodiments of the present application, the processor 901 is used to update the numerical standard words in the third mapping relationship based on the second target slot value; the dictionary vocabulary includes the numerical standard words.

[0168] It can be understood that the memory 902 can be a volatile memory or a non-volatile memory, and can also include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a magnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disk, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as Static Random Access Memory (SRAM), Synchronous Static Random Access Memory (SSRAM), Dynamic Random Access Memory (DRAM), Synchronous Dynamic Random Access Memory (SDRAM), Double Data Rate Synchronous Dynamic Random Access Memory (DDRSDRAM), Enhanced Synchronous Dynamic Random Access Memory (ESDRAM), Sync Link Dynamic Random Access Memory (SLDRAM), and Direct Rambus Random Access Memory (DRRAM).The memory 902 described in the embodiments of the present invention is intended to include, but is not limited to, these and any other suitable types of memories.

[0169] The memory 902 in the embodiment of the present invention is used to store various types of data to support the operation of the electronic device 900. Examples of such data include: any computer program for operating on the electronic device 900. The program for implementing the method of the embodiment of the present invention may be included in the memory 902.

[0170] The method disclosed in the above embodiment of the present invention can be applied to the processor 901, or implemented by the processor 901. The processor may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by an integrated logic circuit of hardware in the processor or an instruction in the form of software. The above processor may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The processor can implement or execute the methods, steps and logic block diagrams disclosed in the embodiment of the present invention. The general-purpose processor may be a microprocessor or any conventional processor, etc. In combination with the steps of the method disclosed in the embodiment of the present invention, it can be directly embodied as a hardware decoding processor to execute, or it can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a storage medium, which is located in a memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0171] In an exemplary embodiment, the electronic device 900 can be implemented by one or more application specific integrated circuits (ASIC), DSP, programmable logic device (PLD), complex programmable logic device (CPLD), field programmable gate array (FPGA), general processor, controller, microcontroller (MCU), microprocessor, or other electronic components to execute the above method.

[0172] Based on the above embodiments, the embodiments of the present application further provide a computer product, including a computer program, which is executed by a processor to implement Figure 1 The corresponding embodiments provide steps in the information processing method.

[0173] Based on the above embodiments, the embodiments of the present application further provide a storage medium, wherein the storage medium stores computer executable instructions, and the computer executable instructions are configured to execute Figure 1 The corresponding embodiments provide information processing methods.

[0174] It should be noted that the above-mentioned computer storage medium can be a memory such as ROM, PROM, EPROM, EEPROM, FRAM, FlashMemory, magnetic surface storage, optical disk, or CD-ROM; it can also be various electronic devices including one or any combination of the above-mentioned memories, such as mobile phones, computers, tablet devices, personal digital assistants, etc.

[0175] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0176] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0177] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course, by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0178] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0179] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0180] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0181] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An information processing method, It is characterized in that The method comprises: Acquiring voice information; the voice information includes a user instruction to be executed; Determining a target slot value of the voice information; The dictionary vocabulary is expanded based on the target slot value; the expanded dictionary vocabulary is used to execute the user instruction.

2. The method according to claim 1, It is characterized in that Determining a target slot value of the voice information includes: Converting the voice information into text information; Inputting the text information into a slot recognition model to obtain a classification slot value of the user instruction; the classification slot value includes a numerical slot value and / or a text slot value; The target slot value is determined based on the classification slot value.

3. The method according to claim 2, It is characterized in that The determining the target slot value based on the classification slot value comprises: Input the text class slot value into a first preset model to obtain a vector value and text segmentation corresponding to the text class slot value; Determine a first slot value corresponding to the vector value based on the first mapping relationship; Determine a second slot value corresponding to the text segmentation based on the second mapping relationship; A first target slot value is determined based on the first slot value and the second slot value.

4. The method according to claim 3, It is characterized in that The expanding the dictionary vocabulary based on the target slot value includes: Update the first text standard word in the first mapping relationship based on the first target slot value; The second text standard word in the second mapping relationship is updated based on the first target slot value; the dictionary vocabulary includes the first text standard word and the second text standard word.

5. The method according to claim 2, It is characterized in that The determining the target slot value based on the classification slot value comprises: Inputting the numerical slot value into a second preset model to obtain a numerical category corresponding to the numerical slot value; Converting the numerical value category into numerical text based on a third mapping relationship; A second target slot value is determined based on the numeric text.

6. The method according to claim 5, It is characterized in that The expanding the dictionary vocabulary based on the target slot value includes: The numerical standard words in the third mapping relationship are updated based on the second target slot value; the dictionary vocabulary includes the numerical standard words.

7. A processing device, It is characterized in that The processing device comprises: An acquisition unit, configured to acquire voice information; the voice information includes a user instruction to be executed; A processing unit, configured to determine a target slot value of the voice information; The processing unit is further used to expand the dictionary vocabulary based on the target slot value; the expanded dictionary vocabulary is used to execute the user instruction.

8. An electronic device, It is characterized in that include: a processor and a memory for storing a computer program capable of being executed on the processor, Wherein, when the processor is used to run the computer program, it executes the steps of the method described in any one of claims 1 to 6.

9. A storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer product comprising a computer program, It is characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images