Intelligent number selection method, device, equipment and computer program product

Abstract

The invention provides an intelligent number selection method, device and equipment and a computer program product, and relates to the technical field of computers and Internet. The intelligent number selection method comprises the following steps: acquiring a number selection voice of a target object, and converting the number selection voice into a number selection text; obtaining a preset number selection prompt word template; calling a large language model according to the number selection prompt word template and the number selection text, and generating a number selection SQL statement by the large language model according to the number selection prompt word template and the number selection text; obtaining a number selection SQL statement generated by the large language model; and executing the SQL statement to perform number screening in the number pool database to obtain a number meeting the demand of the target object. According to the embodiment of the invention, personalized and accurate screening of massive number pools can be realized.

Description

Technical Field

[0001] This disclosure relates to the field of computer and Internet technology, and in particular to an intelligent number selection method, apparatus, electronic device, computer-readable storage medium, and computer program product. Background Technology

[0002] This section is intended to provide background or context for the embodiments of this disclosure as set forth in the claims. The description herein is not intended to be a prior art simply because it is included in this section.

[0003] Traditional customer number selection processes suffer from two major pain points: cumbersome operation and rigid conditions. On the one hand, users need to repeatedly perform manual operations, such as flipping through pages, entering filter conditions, and selecting specific numbers, which is lengthy and time-consuming. On the other hand, the system interface usually only provides limited preset filter conditions, such as number ranges, last digit rules, and whether a number contains specific digits, which severely restricts users' ability to express personalized or complex number selection needs.

[0004] In practice, users often find it difficult to accurately express their preferences using the system's existing fixed options, and are forced to "translate" their true needs into the limited conditions supported by the system. This conversion process is not only difficult to operate, but also prone to misunderstandings, resulting in the final selected numbers being significantly different from the user's actual expectations, leading to a poor overall experience. Summary of the Invention

[0005] The purpose of this disclosure is to provide an intelligent number selection method, device, electronic device, computer-readable storage medium, and computer program product that can achieve personalized and accurate screening of a massive number pool according to user needs.

[0006] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part by practice of this disclosure.

[0007] This disclosure provides an intelligent number selection method, comprising: acquiring a target object's voice prompt for number selection and converting the voice prompt into text; acquiring a preset number selection prompt word template; calling a large language model based on the number selection prompt word template and the number selection text, and having the large language model generate a number selection SQL statement based on the number selection prompt word template and the number selection text; acquiring the number selection SQL statement generated by the large language model; and executing the SQL statement to filter numbers in a number pool database to obtain numbers that meet the target object's requirements.

[0008] In some embodiments, the large language model generates a number selection SQL statement based on the number selection prompt word template and the number selection text, including: determining a target table name and / or target column name in the number pool database that has a similarity to the number selection voice higher than a first preset threshold, wherein numbers with the same category characteristics in the number pool database are grouped in the same table, and numbers with the same subcategory characteristics in a table are grouped in the same column; the large language model combines the number selection prompt word template to generate the SQL statement for filtering in the table corresponding to the target table name and / or in the column corresponding to the target column name.

[0009] In some embodiments, determining target table names and / or target column names in the number pool database that have a similarity higher than a first preset threshold with the number selection voice message includes: converting the number selection voice message into a number selection vector; obtaining the table name vector corresponding to each table name and the column name vector corresponding to each column name in the number pool database; and performing similarity matching between the number selection vector and the table name vector and the column name vector to determine target table names and / or target column names that have a similarity higher than the first preset threshold with the number selection voice message.

[0010] In some embodiments, the method further includes: recognizing at least one number selection intention of the target object based on the number selection voice, the at least one number selection intention including a first number selection intention and a second number selection intention; wherein the target table name and / or target list are related to the first number selection intention but not to the second number selection intention; wherein executing the SQL statement to filter numbers in the number pool database to obtain numbers that meet the requirements of the target object includes: executing the SQL statement to filter at least one candidate number in the table corresponding to the target table name and / or the column corresponding to the target column name; and filtering numbers that meet the requirements of the target object from the at least one candidate number according to the second number selection intention.

[0011] In some embodiments, before converting the voice selection into text selection, the method further includes: identifying at least one selection intention of the target object based on the voice selection, wherein the at least one selection intention includes a third selection intention; if the third selection intention does not match the preset selection filter items in the selection interface, then generating the selection SQL statement based on the voice selection.

[0012] In some embodiments, the number selection interface includes multiple filter items, the multiple filter items include a first filter item, and the at least one number selection intent includes a fourth number selection intent; wherein, the method further includes: determining that the fourth number selection intent matches a preset number selection filter item in the number selection interface; determining a first filter condition corresponding to the target object under the fourth number selection intent based on the number selection voice; locating the first filter item matching the fourth number selection intent among the multiple filter items in the number selection interface; filling the first filter condition into the input box corresponding to the first filter item; locating and triggering the control corresponding to the first filter item to filter out numbers that meet the needs of the target object in the number pool database through a preset SQL query statement corresponding to the first filter item.

[0013] In some embodiments, the at least one number selection intent includes a fifth number selection intent; wherein the method further includes: if the fifth number selection intent is determined to be an interface operation intent; determining the action requirement corresponding to the fifth number selection intent; generating an RPA execution action plan based on the fifth number selection intent; locating the target web page element in the number selection interface according to the action requirement; and performing an operation on the web page element according to the RPA execution action plan.

[0014] This disclosure provides an intelligent number selection device, including: a voice acquisition module, a template acquisition module, a large model calling module, a statement receiving module, and a number filtering module.

[0015] The system includes a voice acquisition module for acquiring the target object's voice prompt for selecting a phone number and converting it into text; a template acquisition module for acquiring a preset selection prompt word template; a large model invocation module for invoking a large language model based on the selection prompt word template and the selection text, with the large language model generating a selection SQL statement; a statement receiving module for acquiring the selection SQL statement generated by the large language model; and a number filtering module for executing the SQL statement to filter numbers in a number pool database to obtain numbers that meet the target object's requirements.

[0016] This disclosure provides an electronic device comprising: a memory and a processor; the memory for storing computer program instructions; and the processor for calling the computer program instructions stored in the memory to implement the intelligent number selection method described above.

[0017] This disclosure provides a computer-readable storage medium storing computer program instructions to implement the intelligent number selection method as described in any of the preceding embodiments.

[0018] This disclosure provides a computer program product or computer program that includes computer program instructions stored in a computer-readable storage medium. The computer program instructions are read from the computer-readable storage medium, and the processor executes the computer program instructions to implement the aforementioned intelligent number selection method.

[0019] The intelligent number selection method, apparatus, electronic device, computer-readable storage medium, and computer program product provided in this disclosure achieve an end-to-end intelligent number selection process from natural language input to accurate number filtering by integrating speech recognition and a large language model. Users only need to verbally state their number selection requirements, and the system can automatically convert them into executable SQL query statements and directly retrieve highly matching candidate numbers from the number pool. This method completely eliminates the reliance on preset filtering conditions in traditional number selection processes, enabling users to express personalized and complex number selection needs in the most natural way, significantly improving selection efficiency, accuracy, and user experience.

[0020] It should be understood that the above general description and the following detailed description are merely exemplary and do not limit this disclosure. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0022] Figure 1 A schematic diagram of a scenario that can be applied to the intelligent number selection method or intelligent number selection device in the embodiments of this disclosure is shown.

[0023] Figure 2 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0024] Figure 3 This is a flowchart illustrating a statement generation method according to an exemplary embodiment.

[0025] Figure 4 This is a flowchart illustrating a table and column filtering method according to an exemplary embodiment.

[0026] Figure 5 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0027] Figure 6 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0028] Figure 7 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0029] Figure 8 This is an overall design diagram of an intelligent number selection method according to an exemplary embodiment.

[0030] Figure 9 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0031] Figure 10 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0032] Figure 11 This is a block diagram illustrating an intelligent number selection device according to an exemplary embodiment.

[0033] Figure 12 A schematic diagram of the structure of an electronic device suitable for implementing embodiments of the present disclosure is shown. Detailed Implementation

[0034] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the embodiments set forth herein; rather, they are provided so that this disclosure will be thorough and complete, and will fully convey the concept of the exemplary embodiments to those skilled in the art. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted.

[0035] Those skilled in the art will recognize that embodiments of this disclosure can be a system, apparatus, device, method, or computer program product. Therefore, this disclosure can be implemented in the following forms: entirely hardware, entirely software (including firmware, resident software, microcode, etc.), or a combination of hardware and software.

[0036] The features, structures, or characteristics described in this disclosure can be combined in any suitable manner in one or more embodiments. Numerous specific details are provided in the following description to give a thorough understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more specific details omitted, or other methods, components, apparatuses, steps, etc., can be employed. In other instances, well-known methods, apparatuses, implementations, or operations are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0037] In this disclosure, the terms "module" or "unit" refer to a computer program or part of a computer program that has a predetermined function and works with other related parts to achieve a predetermined goal, and can be implemented wholly or partially using software, hardware (such as processing circuitry or memory), or a combination thereof. Similarly, a processor (or multiple processors or memory) can be used to implement one or more modules or units. Furthermore, each module or unit can be part of an overall module or unit that includes the functionality of that module or unit.

[0038] The accompanying drawings are merely illustrative of this disclosure, and the same reference numerals in the drawings denote the same or similar parts, thus omitting repeated descriptions of them. Some block diagrams shown in the drawings do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0039] The flowchart shown in the accompanying drawings is merely illustrative and does not necessarily include all content and steps, nor does it require execution in the described order. For example, some steps may be broken down, while others may be combined or partially combined; therefore, the actual execution order may change depending on the specific circumstances.

[0040] In the description of this disclosure, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, and B alone. Furthermore, "at least one" means one or more, and "multiple" means two or more. The terms "first," "second," etc., do not limit the quantity or order of execution, and "first," "second," etc., do not necessarily imply differences; the terms "contains," "includes," and "has" are used to indicate an open-ended meaning of inclusion and refer to the existence of additional elements / components / etc. besides those listed.

[0041] To better understand the above-mentioned objectives, features and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present disclosure can be combined with each other.

[0042] The following section will first explain some of the terms used in the embodiments of this disclosure so that those skilled in the art can understand them.

[0043] Schema (or database schema): A concept in databases, referring to the collection of data structures used to define and organize data within a database. In a database, a schema can be seen as a blueprint, describing the structure of various objects such as tables, fields, constraints, and indexes, as well as the relationships between them. A schema provides database designers and users with a way to describe and organize data, enabling data to be stored, queried, and managed effectively. Specifically, a schema can contain database objects such as tables, views, indexes, sequences, synonyms, procedures, functions, and packages.

[0044] Schema refers to the structured description of a database, used to define how data is organized, constraints, and relationships between data.

[0045] Schema vectorization: storing and processing metadata describing various tables, fields, constraints, indexes, etc. in the database in vector form.

[0046] Text-to-SQL, also known as NL2SQL, is a technique that converts natural language into SQL statements that can be executed on relational databases.

[0047] SQL (Structured Query Language) is a standard language specifically designed for communicating with relational databases.

[0048] Prompt engineering is a technique that guides generative AI models (such as the GPT series) to output specific content by carefully designing input prompts. Its core lies in optimizing the interaction with the model, so that it can be adapted to downstream tasks without additional fine-tuning.

[0049] Robotic Process Automation (RPA) is a business process automation technology based on software robots and artificial intelligence (AI). It provides an alternative way to automate manual processes by mimicking the way end users operate on computers.

[0050] The preceding text introduced some terms and concepts involved in the embodiments of this disclosure. The following text introduces the technical features involved in the embodiments of this disclosure.

[0051] The exemplary embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0052] Figure 1 A schematic diagram of a scenario that can be applied to the intelligent number selection method or intelligent number selection device in the embodiments of this disclosure is shown.

[0053] Please refer to Figure 1 The diagram illustrates an implementation environment provided by an exemplary embodiment of this disclosure.

[0054] like Figure 1 As shown, system architecture 100 may include terminal devices 101, 102, and 103, a network 104, and a server 105. Network 104 serves as the medium for providing communication links between terminal devices 101, 102, and 103 and server 105. Network 104 may include various connection types, such as wired or wireless communication links, or fiber optic cables, etc.

[0055] Users can use terminal devices 101, 102, and 103 to interact with server 105 via network 104 to receive or send messages, etc. Terminal devices 101, 102, and 103 can be various electronic devices with displays and web browsing capabilities, including but not limited to smartphones, tablets, laptops, desktop computers, wearable devices, virtual reality devices, smart home devices, etc.

[0056] Server 105 can be a server that provides various services, such as a backend management server that supports the devices operated by users using terminal devices 101, 102, and 103. The backend management server can analyze and process received requests and other data, and feed the processing results back to the terminal devices.

[0057] A server can be a standalone physical server, a server cluster or a distributed system consisting of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN (Content Delivery Network), and big data and artificial intelligence platforms. This disclosure does not impose any restrictions on this.

[0058] Server 105 may, for example, acquire the target object's voice prompt for selecting a number and convert the voice prompt into text for selecting a number; server 105 may, for example, acquire a preset number selection prompt word template; server 105 may, for example, call a large language model based on the number selection prompt word template and the number selection text, and the large language model generates a number selection SQL statement based on the number selection prompt word template and the number selection text; server 105 may, for example, acquire the number selection SQL statement generated by the large language model; server 105 may, for example, execute the SQL statement to filter numbers in the number pool database to obtain numbers that meet the requirements of the target object.

[0059] It should be understood that Figure 1The number of terminal devices, networks, and servers shown is merely illustrative. Server 105 can be a single physical server or a combination of multiple servers. Depending on actual needs, it can have any number of terminal devices, networks, and servers.

[0060] Figure 2 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment. The method provided in this disclosure can be executed by any electronic device with computing power, for example, the method can be executed by the above-described... Figure 1 The execution can be performed by a server or terminal device in the embodiments, or it can be performed by both a server and a terminal device. In the following embodiments, the server is used as the execution subject for illustration, but this disclosure is not limited to this.

[0061] Reference Figure 2 The intelligent number selection method provided in this embodiment may include the following steps.

[0062] Step S202: Obtain the target object's voice prompt for selecting a number and convert the voice prompt into text for selecting a number.

[0063] "Number selection voice" refers to a voice segment spoken by the target (usually the user) that contains their selection intention in a scenario where selection or confirmation is made through voice interaction.

[0064] Step S204: Obtain the preset number selection prompt template.

[0065] The "number selection prompt template" is a predefined structured text whose core function is to guide the large language model to generate standard SQL query statements that meet the system's expectations based on the user's input "number selection text". This template significantly improves the accuracy and consistency of the SQL generated by the large language model by standardizing the input format and clarifying the semantic intent.

[0066] Step S206: Based on the number selection prompt word template and the number selection text, call the large language model, and the large language model generates the number selection SQL statement based on the number selection prompt word template and the number selection text.

[0067] Step S208: Obtain the number selection SQL statement generated by the large language model.

[0068] Step S210: Execute an SQL statement to filter numbers in the number pool database and obtain numbers that meet the requirements of the target object.

[0069] The intelligent number selection method provided in this disclosure integrates speech recognition and large language model technology to construct an end-to-end intelligent processing flow from voice input to number selection. This method accurately converts the natural voice input of the target user into a structured query statement, significantly improving the naturalness and efficiency of human-computer interaction. Furthermore, it ensures the accuracy and standardization of SQL generation through preset templates, ultimately achieving rapid and accurate matching of user needs within the number pool, effectively balancing user experience and system processing performance.

[0070] Figure 3 This is a flowchart illustrating a statement generation method according to an exemplary embodiment.

[0071] refer to Figure 3 The above statement generation method may include the following steps.

[0072] Step S302: Determine the target table name and / or target column name in the number pool database that have a similarity to the selected voice that is higher than a first preset threshold. Numbers with the same category characteristics in the number pool database are grouped in the same table, and numbers with the same sub-category characteristics in one table are grouped in the same column.

[0073] In some embodiments, the massive number pool can be structured and categorized for storage to improve query efficiency and system performance.

[0074] Specifically, a multi-level classification strategy can be used to organize and manage numbers.

[0075] 1. Table-level classification (divided according to core features).

[0076] Numbers with the same category characteristics are stored in a unified data table, with each data table representing a number type. For example: Table 1: Store high-value numbers, including: premium numbers (e.g., 18888888888, 13999999999); serial numbers (e.g., 12345678, 98765432); and matching numbers (e.g., 138383838, 159595959). Table 2: Store special service numbers, including: birthday numbers (e.g., 19850214, 19991225); anniversary numbers (e.g., 20210808, 20191001). Table 3: Stored regional special numbers, including: area code associated numbers (e.g., 010xxxxxxx, 021xxxxxxx); local special numbers (e.g., numbers containing local postal codes).

[0077] 2. Column-level classification (subdivided by business attributes).

[0078] Within the same number table, numbers with the same subcategory characteristics can be grouped together in the same data column to support more refined query filtering.

[0079] For example, in Table 1, column 1 stores AAA-level premium numbers (e.g., 13888888888); column 2 stores AA-level premium numbers (e.g., 13999999999); and column 3 stores sequential numbers (e.g., 15812345678).

[0080] For example, in Table 2: Column 3 stores the birthday number; Column 4 stores the anniversary number; Column 5 stores the auspicious number (such as 168, 518, etc.).

[0081] For example, in Table 3, column 5 stores exclusive phone numbers for provincial capital cities; column 6 stores exclusive phone numbers for business purposes; and column 7 stores exclusive phone numbers for campuses.

[0082] Through the structured classification and storage mechanism described above, the target data table and columns can be quickly located when a query is initiated by the target object, significantly improving the efficiency of number retrieval. This classification method not only supports precise queries but also enables multi-dimensional combination filtering, providing a solid data foundation for generating accurate SQL statements for large language models, while ensuring the scalability and maintainability of the system.

[0083] Step S304: The large language model, combined with the selection prompt word template, generates SQL statements for filtering in the table corresponding to the target table name and / or in the column corresponding to the target column name.

[0084] This embodiment combines speech recognition with a large language model and innovatively employs a structured number classification and storage mechanism to achieve efficient and accurate SQL statement generation. The system first quickly locates the target data table and columns based on speech similarity matching, and then guides the large language model to generate accurate query statements through prompt word templates. This significantly improves the efficiency and accuracy of number retrieval while ensuring the system's scalability and multi-dimensional query capabilities.

[0085] Figure 4 This is a flowchart illustrating a table and column filtering method according to an exemplary embodiment.

[0086] refer to Figure 4 Determining target table names and / or target column names in the number pool database that have a similarity to the voice selection voice that is higher than a first preset threshold may include the following steps.

[0087] Step S402: Convert the voice selection of numbers into a number selection vector.

[0088] Step S404: Obtain the table name vector corresponding to each table name and the column name vector corresponding to each column name in the number pool database.

[0089] Step S406: Perform similarity matching between the number selection vector and the table name vector and column name vector to determine the target table name and / or target column name with a similarity to the number selection voice that is higher than a first preset threshold.

[0090] This disclosure achieves precise alignment between voice intent and database structure by introducing a vectorized matching mechanism. The method first converts the voice prompt for selecting a number into a vector representation, and simultaneously pre-constructs vectorized indexes for table names and column names. It then quickly locates the most relevant target table names and / or target column names by calculating semantic similarity. This retrieval method based on a vector space model effectively bridges the semantic gap between natural language expression and database structure, significantly improving the accuracy and efficiency of query location, and providing a reliable data foundation for subsequent intelligent generation of SQL statements.

[0091] In some embodiments, the target object's voice prompt for selecting a number may include at least one intention to select a number.

[0092] The user's intention in selecting a number refers to the core purpose or key condition they express in their voice message regarding their desire to select a specific number from the number pool. It is an abstraction and generalization of the user's needs and directly determines the logic for generating subsequent SQL query statements.

[0093] Suppose a user speaks a voice message about selecting a phone number, which may contain one or more combined intentions for selecting a number.

[0094] 1. Single intention.

[0095] User's voice message: "I want a premium phone number."

[0096] Selection intention: The filtering criteria is "number type = premium number".

[0097] 2. Complex intent (multiple conditions listed in parallel).

[0098] User's voice message: "Find a Beijing Mobile number that ends in 888."

[0099] Analysis of the intention behind number selection: Figure 1 (Number Characteristics): The last two digits of the number are "888"; meaning Figure 2 (Carrier): Carrier = "China Mobile"; meaning Figure 3 (Location): Location = "Beijing".

[0100] The user expressed three parallel filtering conditions, all of which need to be met simultaneously.

[0101] 3. Priority intentions (with preferences or weights).

[0102] User's voice: "Ideally, they should be consecutive numbers. If not, birthdates are fine too."

[0103] Number selection intention analysis: Primary intention: Prioritize filtering "Number type = consecutive numbers"; Alternative intention: If no results are found, filter "Number type = birthday number".

[0104] Users' intentions have a clear priority order, and the system needs to understand this logical relationship.

[0105] 4. Exclusionary intent (negation condition).

[0106] User's voice: "Give me a number that doesn't contain the number 4."

[0107] Intent for selecting numbers: to exclude numbers that contain the digit '4'.

[0108] The user's intention is to exclude numbers with certain characteristics, rather than to include them.

[0109] 5. Scope-based intentions.

[0110] User's voice: "The monthly fee is around one hundred yuan."

[0111] Intended purpose of number selection: Monthly package fee between 20 and 120.

[0112] Analysis: The user's intention is a price range, not a fixed value.

[0113] By accurately identifying these number selection intentions, the system can guide the large language model to generate the correct SQL statement (such as using operators like WHERE, AND, OR, NOT, BETWEEN, etc.), thereby accurately locating the user's desired number from a massive number of numbers.

[0114] In some embodiments, at least one selection intention of the target object can be recognized based on the voice selection process. The at least one selection intention includes a first selection intention and a second selection intention. The target table name and / or target list are related to the first selection intention but not to the second selection intention.

[0115] In some embodiments, executing an SQL statement to filter numbers in a number pool database to obtain numbers that meet the requirements of the target object may include: executing an SQL statement to filter at least one candidate number in the table corresponding to the target table name and / or the column corresponding to the target column name; and filtering numbers that meet the requirements of the target object from the at least one candidate number according to a second number selection intention.

[0116] This embodiment of the disclosure achieves accurate parsing and efficient processing of complex user number selection intentions by introducing a multi-level filtering mechanism. The system first identifies multiple number selection intentions from the user's voice, and at the database level, prioritizes queries related to the first intention using generated SQL statements to quickly narrow down the candidate range. Subsequently, at the application level, the system performs a second filtering of candidate numbers based on the second intention, ultimately accurately locating numbers that meet all of the user's needs. This layered processing approach leverages the efficiency of batch database queries while retaining the flexibility of business logic processing, effectively balancing query efficiency and result accuracy. It can well handle scenarios where users' voices contain multiple intentions such as parallel, priority, and exclusion.

[0117] Figure 5 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0118] refer to Figure 5 The above-mentioned intelligent number selection method may include the following steps.

[0119] Step S502: Based on the voice recognition of the number selection target object, at least one number selection intention is recognized, and the at least one number selection intention includes a third number selection intention.

[0120] The aforementioned third selection intention may be the same as or different from the second selection intention (or the third selection intention), and this application does not impose any restrictions on this.

[0121] Step S504: If the third selection intention does not match the preset selection filter items in the selection interface, the selection voice is converted into selection text.

[0122] The third selection intention does not match the preset selection filter items in the selection interface. This can refer to a specific need expressed by the user (target object) through voice (such as the third selection intention), which cannot be met by the existing fixed filter buttons or drop-down menus and other visual controls on the current selection interface.

[0123] Step S506: Obtain the preset number selection prompt template.

[0124] Step S508: Based on the number selection prompt word template and the number selection text, call the large language model, and the large language model generates the number selection SQL statement based on the number selection prompt word template and the number selection text.

[0125] Step S510: Obtain the number selection SQL statement generated by the large language model.

[0126] Step S512: Execute an SQL statement to filter numbers in the number pool database and obtain numbers that meet the requirements of the target object.

[0127] This embodiment of the disclosure effectively solves the pain point that traditional graphical interfaces cannot handle personalized number selection needs by using a collaborative mechanism of intelligent voice intent recognition and SQL generation from a large language model. When the system detects that the user's voice intent exceeds the preset filter range, it automatically starts an intelligent processing flow to convert natural language into precise database query statements. This achieves an interactive breakthrough from "fixed filtering" to "free expression," expanding the system's capabilities and providing users with a more natural and efficient number selection experience.

[0128] Figure 6 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0129] In some embodiments, the number selection interface may include multiple filter items, including a first filter item, and the aforementioned at least one number selection intent includes a fourth number selection intent.

[0130] refer to Figure 6 The above-mentioned intelligent number selection method may include the following steps.

[0131] Step S602: Determine that the fourth number selection intention matches the preset number selection filter items in the number selection interface.

[0132] Step S604: Determine the first screening condition corresponding to the target object under the fourth selection intention based on the selection voice.

[0133] In an intelligent number selection system, the first filtering condition refers to the structured query conditions parsed from the user's voice prompts, corresponding to their specific number selection intent. It is the fundamental semantic unit in the process of transforming the number selection needs expressed in natural language into a computer-executable database query statement, and is a core component in constructing complete SQL query conditions.

[0134] For example, suppose a user says, "A number with a monthly fee of around 50 yuan." The fourth intention for selecting a number could be: filtering by package price; the first filtering criterion could be a monthly package fee between 45 and 55 yuan.

[0135] Step S606: Locate the first filter that matches the fourth selection intention among the multiple filter options on the number selection interface.

[0136] In some embodiments, the specific filter control corresponding to the fourth core need (fourth number selection intent) expressed in the user's voice can be found on a graphical interface displayed to the user that includes multiple filtering conditions (such as "number type", "location", "package price" etc.).

[0137] Step S608: Fill the first filter condition into the input box corresponding to the first filter item.

[0138] After successfully locating the interface control (first filter) that matches the user's intent, the specific parameter value corresponding to the intent is automatically entered into the input area of ​​the control.

[0139] This actually simulates the process of a user manually operating the system: just like clicking an input box with a mouse and then typing content through the keyboard, it automatically completes this "input" action on behalf of the user.

[0140] Step S610: Locate and trigger the control corresponding to the first filter item, so as to filter out numbers that meet the requirements of the target object from the number pool database through the preset SQL query statement corresponding to the first filter item.

[0141] After the system identifies the first filtering condition by parsing the user's voice, it intelligently matches it with the preset filtering items in the graphical interface, locates the corresponding visual control (such as a button, drop-down menu, or checkbox), and automatically triggers the selection state of the control. Then, it calls the preset SQL query template bound to the control, fills the identified filtering parameters into the template to generate a complete query statement, and finally executes the filtering in the number pool database, returning the number results that accurately match the user's needs.

[0142] This embodiment utilizes the deep integration of speech recognition and interface automation technologies to construct a highly efficient intelligent number selection scheme. The system can accurately interpret the user's filtering intent in their speech, automatically match and operate the corresponding filter items in the graphical interface, intelligently fill the input box with structured query conditions, and finally complete the number selection by triggering a preset query statement. This method, while fully compatible with existing system architectures, achieves end-to-end automated processing from voice input to filtering results, preserving the stability of traditional interface operations while significantly improving operational efficiency and user experience through intelligent interaction.

[0143] Figure 7 This is a flowchart illustrating an intelligent number selection method according to an exemplary embodiment.

[0144] In some embodiments, at least one number selection intent may include a fifth number selection intent.

[0145] refer to Figure 7 The above-mentioned intelligent number selection method may include the following steps.

[0146] Step S702: If it is determined that the intention of the fifth selection number is an interface operation intention.

[0147] User interface intent refers to commands expressed by a user via voice, intended to directly control or interact with graphical user interface (GUI) elements, rather than directly expressing filtering criteria for number attributes. The core objective of this type of intent is to manipulate the interface itself to change its state, navigate, or trigger specific functions.

[0148] Step S704: Determine the action requirements corresponding to the fifth selection intention.

[0149] Action requirements refer to the specific interface interaction instructions that the system needs to execute after the user's intent is identified as "interface operation intent".

[0150] It is a refinement and specification of the "interface operation intent", clarifying which object to operate and what operation to perform.

[0151] Step S706: Generate RPA execution action plan based on the fifth selection intention.

[0152] The core idea of ​​RPA (Robotic Process Automation) is to create a software robot that can operate other software applications like a human to automate repetitive, tedious, rule-based tasks.

[0153] RPA (Execution Planning) is a precise and ordered sequence of computer operation instructions designed to automate specific business processes. It transforms high-level task objectives (such as "clicking the sort button") into standardized operational steps that can be executed by machines. This includes locating target interface elements, defining interactive actions (clicking, inputting, selecting, etc.), controlling execution logic, and handling exceptions, forming a complete automation solution.

[0154] Step S708: Locate the target webpage element in the number selection interface according to the action requirements.

[0155] Step S710: Perform operations on web page elements according to the RPA execution action plan.

[0156] This disclosed embodiment achieves a complete closed loop from interaction understanding to interface operation in the intelligent number selection system by integrating voice intent recognition and RPA automation technology. This solution can accurately analyze the user's interface operation intent in their voice, transform it into standardized RPA execution action planning, and ultimately complete the interface interaction by automating the operation of web page elements. This technical approach effectively breaks down the interaction barriers between voice commands and graphical interfaces, achieving a "hands-free" intelligent operation experience while maintaining the existing system architecture, significantly improving user operation efficiency and system automation level.

[0157] Figure 8 This is an overall design diagram of an intelligent number selection method according to an exemplary embodiment.

[0158] Combination Figure 8As shown in the design diagram, this application proposes the following intelligent number selection method, which may include the following steps.

[0159] (a) When the customer enters the number selection stage, monitor and recognize the customer's voice in real time. When the customer starts to speak, collect the customer's voice as input.

[0160] (b) Perform voice processing such as voice enhancement and speech-to-text on the collected customer voice to form text.

[0161] (c) Based on the above text output, perform intent recognition to determine the customer's intent from the content stated by the customer. Does the customer want to click to flip through the number set? Or does the customer want to click to confirm the selection of a number? Or does the customer want to select the number filtering conditions provided by the telecom provider, or does the customer's actual needs exceed the range of preset options on the number selection interface?

[0162] (d) If it is determined that the customer's description is related to the personalized number selection, then it is further determined whether the customer's described needs exceed the range of preset options on the number selection interface. If they do not exceed the range of preset options on the number selection interface, the agent schedules RPA to automatically operate the options on the number selection interface according to the customer's described needs, and directly executes preset fixed SQL statements to perform number filtering and push in the number pool database; if they exceed the range of preset options on the number selection interface, the customer's fuzzy needs are transformed into field conditions that can be recognized by the vector database, and the relevant table structure is recalled using similarity vectors. The recalled relevant table structure is then dynamically updated to the prompt word template, and the large model generates the corresponding SQL statement according to the prompt word template, and executes the SQL statement to perform number filtering and push in the number pool database.

[0163] Before retrieving relevant table structures based on vector similarity, the vector database needs to be built first. This involves converting the database table structures of the number pool into vectors, including table names, column names, column data types, primary keys, foreign key constraints, and other information, and storing these vectors in the vector database. Only after this step is completed can the system support retrieving fuzzy customer requirements as field conditions recognizable by the vector database.

[0164] (e) If it is determined that the content described by the customer is related to the customer's action, such as the customer wanting to flip through the number selection page, the agent performs semantic understanding and generates an RPA execution plan, locating the target webpage element on the current number selection page based on the customer's action requirements. The agent then invokes RPA to automatically execute the customer's action requirements.

[0165] refer to Figure 9 The above method can be specifically described in the following steps.

[0166] 1) When the monitoring detects that the "customer selects a number" process has been carried out, the monitoring starts to detect the customer's voice. If the monitoring detects that the customer is speaking, the voice stream is collected as input.

[0167] 2) The original customer voice streams are processed using voice enhancement techniques (such as deep noise suppression and echo cancellation) to improve the clarity and intelligibility of customer service and customer voices respectively.

[0168] 3) After the speech stream has been processed by voice enhancement technology, a high-precision speech-to-text engine is used to convert the speech stream into structured text in real time.

[0169] 4) Perform customer intent identification. Analyze text content to determine the customer's core intent.

[0170] 5) If the system identifies the customer's specific needs regarding the number, it needs to further determine whether the customer's described needs exceed the range of preset filtering options on the number selection interface.

[0171] (1) If it is within the range: the customer’s needs can be accurately expressed using the existing options on the interface (such as drop-down boxes, input boxes, and checkboxes for “last digits include”, “package type”, “price range”). For example: “last digits are 5678”.

[0172] The intelligent agent maps the customer's explicit requirements (e.g., "last digit = 5678") to the corresponding filter options on the number selection interface. It dispatches RPA to automatically locate and manipulate (e.g., click, input) the corresponding filter controls on the interface, setting the customer-specified conditions. It then triggers the system to execute preset, fixed SQL queries. These SQL statements are pre-written and correspond to standard filter operations on the interface. The SQL is executed in the number pool database, filtering out a set of numbers that meet the conditions. See the detailed process below. Figure 9 .

[0173] refer to Figure 9 The above-mentioned intelligent number selection method may include the following steps.

[0174] ① Real-time acquisition of customer voice conversations as input; ② Perform voice enhancement on the collected speech streams, such as speech enhancement based on hidden Markov models and speech enhancement based on deep neural networks, to improve the quality of the speech streams and enable them to be better converted into text. ③ Perform speech-to-text conversion on the processed speech stream to convert it into text; ④ When the intelligent agent receives text input, the large model, which acts as the central control, performs semantic understanding and judgment on the input to determine the customer's action requirements and generates an RPA execution plan. ⑤The agent locates the target web element on the current number selection page according to the customer's action requirements; ⑥The agent calls RPA to perform operations on the located web element according to the above customer action requirements, such as flipping pages or filling in the personalized requirement information of the number spoken by the customer.

[0175] (2) If it exceeds the range: The customer's requirements are vague, unstructured, or involve conditions not directly provided by the interface. For example: "Find a number that sounds smooth", "A number somewhat related to my license plate number Jing A12345".

[0176] Convert the customer's vague and unstructured requirement description text (such as "a number that sounds smooth") into a vector representation (usually using a text embedding model such as Sentence - BERT). This vector represents the semantics of the requirement. Based on this vector, perform similarity recall in the vector database to find the database table names and column names (fields) vectors that are most relevant to the customer's requirement semantic vector, so as to return these most similar database table names and column names, etc., and use them to update the prompt template, and dynamically insert the recalled relevant table structure information (table names, column names) into this template. The large model generates the corresponding SQL statement according to this prompt template, and executes the SQL statement generated by the LLM to filter in the number pool database. Specifically, reference can be made to Figure 10 the embodiments shown.

[0177] Reference Figure 10 , the above intelligent number selection method may include the following steps.

[0178] ①Vectorization of the number pool database Schema: Pre - convert the metadata (Schema) of the number pool database into vectors and store them in a dedicated vector database. The converted metadata includes: table name (Table Name), column name (ColumnName), column data type (Data Type), column meaning description (Description, if any), primary key (Primary Key) information, foreign key (Foreign Key) constraint information, etc.

[0179] ②Recall process: Use the customer requirement vector to perform similarity search (such as cosine similarity) in the vector database. The goal is to find the database table names and column names (fields) that are most relevant to the customer's requirement semantics. For example, the vector of "smooth" may recall the pronunciation_score (smoothness score) column or memorability_rating (memorability score) column in the numbers table (if the description or name of these columns is semantically similar to "smooth").

[0180] ③ Prompt Template Update: There is a preset prompt template to guide the LLM in generating SQL. The retrieved relevant table structure information (table name, column name) is dynamically inserted into this template.

[0181] ④ Generate SQL statements from the large model: Input the updated prompt word template (containing the original customer requirement text and relevant table / column information for recall) into the large language model (LLM). Based on its understanding of the requirements and the provided database structure context, the LLM dynamically generates the corresponding SQL query statement. For example, it generates a SELECT statement. FROM numbers WHEREpronunciation_score > 90 ORDER BY pronunciation_score DESC; statement.

[0182] Finally, the system executes the SQL statements generated by the large model and performs dynamic filtering in the number pool database.

[0183] 6) If the customer's description is found to be related to their actions, such as wanting to turn the page of the number set, the specific process is similar to that in (1) of 5), see [link to relevant documentation]. Figure 9 As shown.

[0184] To address the poor user experience and efficiency bottlenecks caused by the numerous repetitive manual operations (such as clicking to turn pages and entering information) required by traditional number selection schemes, this embodiment utilizes voice-driven technology (covering core modules such as speech recognition, voice enhancement, and speech-to-text) and agent-driven Robotic Process Automation (RPA) to achieve a zero-manual intervention process for the client. Customers issue commands through natural voice interaction, and the intelligent agent analyzes the voice intent in real time and drives the RPA engine to simulate and execute interface operations (such as automatic page turning, conditional input, and result filtering), completely freeing users from the burden of manual operation and significantly improving the smoothness of interaction and ease of use.

[0185] To overcome the limitations of existing number selection interfaces that rely on preset fixed filtering conditions and cannot meet customers' personalized and complex number selection needs, this embodiment proposes a method for dynamic filtering and number recommendation based on a large language model. The method transforms the customer's speech into textual requirements and integrates a number pool database schema vectorization representation, schema-aware dynamic injection of prompt words, and large-model-based Text-to-SQL generation technology. This enables the large model to deeply understand the user's semantic-level needs and dynamically generate precisely matching SQL query statements. This technology achieves real-time, dynamic retrieval of a massive number pool, automatically filtering and recommending highly relevant candidate number sets based on the customer's freely expressed needs, greatly improving the flexibility, personalization, and number resource utilization of the selection process.

[0186] This embodiment addresses two core issues in traditional number selection processes: cumbersome customer operations and rigid, fixed selection criteria. By integrating voice-driven technology (including speech recognition, voice enhancement, and speech-to-text) with agent-driven RPA (Robotic Process Automation) technology, it analyzes the customer's intentions in number selection in real time and triggers the agent to schedule RPA to execute interface operations (such as page turning, information input, and result filtering), completely eliminating the burden of manual operation and significantly improving interaction efficiency and user experience. Furthermore, it breaks the limitations of fixed selection criteria by using speech-to-text conversion of user requests. Through schema vectorization of the number pool database, dynamic injection of schema-aware prompts, and Text-to-SQL generation technology, the large model accurately analyzes complex semantic needs and automatically generates suitable SQL, enabling real-time dynamic retrieval of massive number pools. Based on any personalized needs, it intelligently pushes highly relevant candidate number sets, greatly enhancing selection flexibility, personalization capabilities, and number resource utilization.

[0187] It should be particularly noted that the steps in each embodiment of the above-described intelligent number selection method can be overlapped, substituted, added, or deleted. Therefore, these reasonable permutations and combinations of the intelligent number selection method should also fall within the protection scope of this disclosure, and the protection scope of this disclosure should not be limited to the described embodiments.

[0188] Based on the same inventive concept, this disclosure also provides an intelligent number selection device, as described in the following embodiments. Since the principle by which this device solves the problem is similar to that of the method embodiments described above, the implementation of this device embodiment can refer to the implementation of the method embodiments described above, and repeated details will not be repeated.

[0189] Figure 11 This is a block diagram illustrating an intelligent number selection device according to an exemplary embodiment. (Refer to...) Figure 11 The intelligent number selection device 1100 provided in this embodiment may include: a voice acquisition module 1101, a template acquisition module 1102, a large model calling module 1103, a statement receiving module 1104, and a number filtering module 1105.

[0190] The voice acquisition module 1101 can be used to acquire the target object's voice for selecting a number and convert the voice into text for selecting a number; the template acquisition module 1102 can be used to acquire a preset template for selecting a number; the large model invocation module 1103 can be used to invoke a large language model based on the template for selecting a number and the text for selecting a number, and the large language model generates a number selection SQL statement based on the template for selecting a number and the text for selecting a number; the statement receiving module 1104 can be used to acquire the number selection SQL statement generated by the large language model; and the number filtering module 1105 can be used to execute the SQL statement to filter numbers in the number pool database to obtain numbers that meet the needs of the target object.

[0191] It should be noted that the aforementioned voice acquisition module 1101, template acquisition module 1102, large model calling module 1103, statement receiving module 1104, and number filtering module 110 correspond to S202 to S210 in the method embodiment. The examples and application scenarios implemented by these modules and their corresponding steps are the same, but they are not limited to the content disclosed in the above method embodiment. It should also be noted that these modules, as part of the apparatus, can be executed in a computer system, such as a set of computer-executable instructions.

[0192] In some embodiments, the large model calling module 1103 may include a similarity filtering submodule and a statement generation submodule.

[0193] The similarity filtering submodule can be used to determine the target table name and / or target column name in the number pool database that has a similarity to the number selection voice that is higher than a first preset threshold. Numbers with the same category characteristics in the number pool database are grouped in the same table, and numbers with the same subcategory characteristics in one table are grouped in the same column. The statement generation submodule can be used by the large language model to combine the number selection prompt word template to generate the SQL statement for filtering in the table corresponding to the target table name and / or the column corresponding to the target column name.

[0194] In some embodiments, the similarity filtering submodule may include: a number selection vector determination unit, a column vector determination unit, and a matching unit.

[0195] The number selection vector determination unit can be used to convert the number selection voice into a number selection vector; the column vector determination unit can be used to obtain the table name vector corresponding to each table name and the column name vector corresponding to each column name in the number pool database; the matching unit can be used to perform similarity matching between the number selection vector and the table name vector and the column name vector to determine the target table name and / or target column name with a similarity higher than a first preset threshold to the number selection voice.

[0196] In some embodiments, the intelligent number selection device 1100 may further include an intent recognition module.

[0197] The intent recognition module can be used to recognize at least one number selection intent of the target object based on the number selection voice, wherein the at least one number selection intent includes a first number selection intent and a second number selection intent; wherein the target table name and / or target list is related to the first number selection intent but not to the second number selection intent.

[0198] The number filtering module 1105 may include a candidate number filtering submodule and a final filtering submodule.

[0199] The candidate number filtering submodule can be used to execute the SQL statement to filter at least one candidate number from the table corresponding to the target table name and / or the column corresponding to the target column name; the final filtering submodule can be used to filter out numbers that meet the requirements of the target object from the at least one candidate number according to the second number selection intention.

[0200] In some embodiments, the intelligent number selection device 1100 may further include: at least one number selection intention recognition module and a mismatch judgment module.

[0201] Among them, at least one number selection intent recognition module can be used to recognize at least one number selection intent of the target object based on the number selection voice before converting the number selection voice into number selection text, and the at least one number selection intent includes a third number selection intent; the mismatch judgment module can be used to generate the number selection SQL statement based on the number selection voice if the third number selection intent does not match the preset number selection filter items in the number selection interface.

[0202] In some embodiments, the number selection interface includes multiple filter items, the multiple filter items include a first filter item, and the at least one number selection intention includes a fourth number selection intention; wherein, the intelligent number selection device 1100 may further include: a filter item matching module, a first filter condition determination module, a first filter item determination module, a filling module, and a triggering module.

[0203] The filter matching module can be used to determine whether the fourth number selection intention matches the preset number selection filter items in the number selection interface; the first filter condition determination module can be used to determine the first filter condition corresponding to the target object under the fourth number selection intention based on the number selection voice; the first filter item determination module can be used to locate the first filter item that matches the fourth number selection intention among the multiple filter items in the number selection interface; the fill module can be used to fill the first filter condition into the input box corresponding to the first filter item; and the trigger module can be used to locate and trigger the control corresponding to the first filter item to filter out numbers that meet the needs of the target object in the number pool database through the preset SQL query statement corresponding to the first filter item.

[0204] In some embodiments, the at least one number selection intention includes a fifth number selection intention; wherein the intelligent number selection device 1100 may further include: an intention judgment module, an action requirement module, an action planning determination module, an element positioning module, and an execution module.

[0205] The intent determination module can be used to determine if the fifth selection intent is an interface operation intent; the action requirement module can be used to determine the action requirement corresponding to the fifth selection intent; the action planning determination module can be used to generate an RPA execution action plan based on the fifth selection intent; the element location module can be used to locate the target web page element in the selection interface based on the action requirement; and the execution module can be used to perform an operation on the web page element based on the RPA execution action plan.

[0206] Since the functions of the device 1100 have been described in detail in their respective method embodiments, they will not be repeated here.

[0207] The modules and / or sub-modules and / or units described in the embodiments of this disclosure can be implemented in software or hardware. The described modules and / or sub-modules and / or units can also be located in a processor. The names of these modules and / or sub-modules and / or units do not, in some cases, constitute a limitation on the module and / or sub-module and / or unit itself.

[0208] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in a flowchart or block diagram may represent a portion of a module or program segment containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer program instructions.

[0209] Furthermore, the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of this disclosure and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Additionally, it is readily understood that these processes may be executed synchronously or asynchronously, for example, in multiple modules.

[0210] Figure 12 A schematic diagram of an electronic device suitable for implementing embodiments of the present disclosure is shown. It should be noted that... Figure 12 The illustrated electronic device 1200 is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments disclosed herein.

[0211] like Figure 12 As shown, the electronic device 1200 includes a central processing unit (CPU) 1201, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1202 or a program loaded from a storage section 1208 into a random access memory (RAM) 1203. The RAM 1203 also stores various programs and data required for the operation of the electronic device 1200. The CPU 1201, ROM 1202, and RAM 1203 are interconnected via a bus 1204. An input / output (I / O) interface 1205 is also connected to the bus 1204.

[0212] The following components are connected to I / O interface 1205: an input section 1206 including a keyboard, mouse, etc.; an output section 1207 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 1208 including a hard disk, etc.; and a communication section 1209 including a network interface card such as a LAN card, modem, etc. The communication section 1209 performs communication processing via a network such as the Internet. A drive 1210 is also connected to I / O interface 1205 as needed. Removable media 1211, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 1210 as needed so that computer programs read from them can be installed into storage section 1208 as needed.

[0213] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing computer program instructions for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 1209, and / or installed from removable medium 1211. When the computer program is executed by central processing unit (CPU) 1201, it performs the functions defined above in the system of this disclosure.

[0214] It should be noted that the computer-readable storage medium disclosed herein may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable computer program instructions. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable storage medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. Computer program instructions contained on a computer-readable storage medium can be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0215] In another aspect, this disclosure also provides a computer-readable storage medium, which may be included in the device described in the above embodiments; or it may exist independently and not assembled into the device. The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the device, enable the device to perform the following functions: acquiring the target object's voice prompt for selecting a number and converting the voice prompt into text; acquiring a preset number selection prompt template; calling a large language model based on the number selection prompt template and the text, and having the large language model generate a number selection SQL statement based on the number selection prompt template and the text; acquiring the number selection SQL statement generated by the large language model; and executing the SQL statement to filter numbers in a number pool database to obtain numbers that meet the target object's requirements.

[0216] According to one aspect of this disclosure, a computer program product or computer program is provided, comprising computer program instructions stored in a computer-readable storage medium. The computer program instructions are read from the computer-readable storage medium, and a processor executes the computer program instructions to implement the methods provided in various optional implementations of the above embodiments.

[0217] From the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions of the embodiments of this disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, or portable hard drive) and includes several computer program instructions to cause an electronic device (such as a server or terminal device) to execute the method according to the embodiments of this disclosure.

[0218] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the claims.

[0219] It should be understood that this disclosure is not limited to the detailed structures, drawing arrangements or implementations shown herein; rather, this disclosure is intended to cover various modifications and equivalent arrangements contained within the spirit and scope of the appended claims.

Claims

1. A smart number selection method, characterized in that, include: Obtain the target object's voice prompt for selecting a phone number, and convert the voice prompt into text prompt for selecting a phone number; Get the preset number selection prompt template; The large language model is invoked based on the number selection prompt word template and the number selection text, and the large language model generates a number selection SQL statement based on the number selection prompt word template and the number selection text; Obtain the number selection SQL statement generated by the large language model; The SQL statement is executed to filter numbers in the number pool database to obtain numbers that meet the requirements of the target object.

2. The method according to claim 1, characterized in that, The large language model generates a number selection SQL statement based on the number selection prompt word template and the number selection text, including: In the number pool database, target table names and / or target column names with a similarity to the selected voice exceeding a first preset threshold are determined, wherein numbers with the same category characteristics in the number pool database are grouped in the same table, and numbers with the same sub-category characteristics in a table are grouped in the same column; The large language model, combined with the number selection prompt template, generates the SQL statement for filtering in the table corresponding to the target table name and / or in the column corresponding to the target column name.

3. The method according to claim 1, characterized in that, Determining target table names and / or target column names in the number pool database that have a similarity to the selected voice message exceeding a first preset threshold includes: Convert the voice prompt for selecting a number into a number selection vector; Obtain the table name vector corresponding to each table name and the column name vector corresponding to each column name in the number pool database; The number selection vector is matched with the table name vector and the column name vector for similarity to determine the target table name and / or target column name whose similarity to the number selection voice is higher than a first preset threshold.

4. The method according to claim 3, characterized in that, The method further includes: The system identifies at least one number selection intention of the target object based on the number selection voice recognition, the at least one number selection intention including a first number selection intention and a second number selection intention; wherein the target table name and / or target list are related to the first number selection intention but not to the second number selection intention; Executing the SQL statement to filter numbers in the number pool database and obtain numbers that meet the requirements of the target object includes: Execute the SQL statement to filter out at least one candidate number from the table corresponding to the target table name and / or the column corresponding to the target column name; According to the second number selection intention, numbers that meet the requirements of the target object are selected from the at least one candidate numbers.

5. The method according to claim 1, characterized in that, Before converting the voice selection into text selection, the method further includes: Based on the voice recognition of the number selection, at least one number selection intention of the target object is identified, and the at least one number selection intention includes a third number selection intention; If the third number selection intention does not match the preset number selection filter items in the number selection interface, then the number selection SQL statement is generated according to the number selection voice.

6. The method according to claim 5, characterized in that, The number selection interface includes multiple filter options, including a first filter option, and the at least one number selection intention includes a fourth number selection intention; wherein, the method further includes: Determine that the fourth number selection intention matches the preset number selection filter items in the number selection interface; The first filtering condition corresponding to the target object under the fourth number selection intention is determined based on the number selection voice. Locate the first filter item that matches the fourth selection intention among the plurality of filter items on the number selection interface; Fill the first filter criteria into the input box corresponding to the first filter item; Locate and trigger the control corresponding to the first filter item, so as to filter out numbers that meet the requirements of the target object in the number pool database through the preset SQL query statement corresponding to the first filter item.

7. The method according to claim 5, characterized in that, The at least one number selection intention includes a fifth number selection intention; The method further includes: If it is determined that the intention of the fifth selection is an interface operation intention; Determine the action requirement corresponding to the fifth selection intention; Generate an RPA execution action plan based on the fifth selection intention; According to the action requirements, locate the target webpage element in the number selection interface; The web page elements are operated according to the RPA execution action plan.

8. An intelligent number selection device, characterized in that, include: The voice acquisition module is used to acquire the voice of the target object selecting a number and convert the voice of the number selection into the text of the number selection. The template acquisition module is used to acquire preset number selection prompt word templates; The large model calling module is used to call the large language model according to the number selection prompt word template and the number selection text, and the large language model generates the number selection SQL statement according to the number selection prompt word template and the number selection text; The statement receiving module is used to obtain the number selection SQL statement generated by the large language model; The number filtering module is used to execute the SQL statement to filter numbers in the number pool database and obtain numbers that meet the requirements of the target object.

9. An electronic device, characterized in that, include: Memory and processor; The memory is used to store computer program instructions; the processor calls the computer program instructions stored in the memory to implement the intelligent number selection method as described in any one of claims 1-7.

10. A computer-readable storage medium storing computer program instructions thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the intelligent number selection method as described in any one of claims 1-7.

11. A computer program product comprising computer program instructions stored in a computer-readable storage medium, characterized in that, When the computer program instructions are executed by the processor, they implement the method of any one of claims 1-7.

Images