Intelligent voice call recognition method and device

Abstract

The invention relates to the technical field of intelligent calling, and particularly discloses an intelligent voice call recognition method and device, and the method comprises the steps: obtaining a voice call instruction of a user; determining call knowledge information associated with the voice call instruction based on a call knowledge base of the user; determining a call identification number based on the voice call instruction and the call knowledge information; and initiating a call based on the call identification number when the user confirms the call identification number. According to the technical scheme provided by the invention, the fuzzy call instruction of the user can be intelligently identified based on the call knowledge base of the user and the call knowledge information stored in the call knowledge base, so that the intelligent call experience of the user is improved.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent call technology, and in particular to an intelligent voice call recognition method and apparatus. Background Technology

[0002] In recent years, with the rapid development of artificial intelligence (AI) technology, especially breakthroughs in deep learning and natural language processing, speech recognition technology has seen unprecedented improvements. These advancements have significantly increased the accuracy and response speed of speech recognition, leading to its widespread application in various industries, particularly in call centers and customer service. The global AI call center market is experiencing strong growth, as the introduction of AI technology can greatly improve the efficiency and quality of customer service.

[0003] In enterprise call scenarios, due to the relatively large number of employees and densely distributed customer base in medium to large enterprises, users often cannot clearly know the specific caller ID of each party they need to contact. For example, user A needs to call B in Business Unit 1. In this case, B is the callee, but user A does not know B's specific caller ID and can only make the call using a vague method such as "Please contact employee B for me." If there is more than one employee named B in the company, it may lead to problems with accurately initiating the call. This situation greatly affects work efficiency and the user's intelligent experience. Summary of the Invention

[0004] The purpose of this disclosure is to provide an intelligent voice call recognition method and apparatus, which can intelligently recognize the user's fuzzy call commands based on the user's call knowledge base and the call knowledge information contained therein, thereby improving the user's intelligent call experience.

[0005] Specifically, the first aspect of this disclosure provides an intelligent voice call recognition method, which may include: acquiring a user's voice call command; determining call knowledge information associated with the voice call command based on the user's call knowledge base; determining a call identification number based on the voice call command and the call knowledge information; and initiating a call based on the call identification number if the user confirms the call identification number.

[0006] In one possible implementation of the first aspect above, the method further includes: generating call summary information based on the call recording of the call identification number; and updating the call knowledge information corresponding to the call identification number based on the call summary information and the call record corresponding to the call recording.

[0007] In one possible implementation of the first aspect above, the method further includes: updating the call address book in the call knowledge base based on the call identification number when the call identification number is the first call number; updating the call address book based on the called party information when the call recording of the call identification number contains the called party information corresponding to the call identification number; and prompting the user to supplement the called party information when the call recording does not contain the called party information, and updating the call address book based on the supplemented called party information.

[0008] In one possible implementation of the first aspect above, when the intelligent voice call recognition method is applied to an enterprise call scenario, when a call identification number is received or made, it is determined whether the call identification number exists in the call knowledge base: if the call identification number does not exist in the call knowledge base, the called party information corresponding to the call identification number is retrieved from the third-party business system associated with the enterprise call scenario, and the call address book is updated based on the retrieved called party information; if the call identification number exists in the call knowledge base, the called party information corresponding to the call identification number is retrieved from the third-party business system, and it is verified whether the retrieved called party information is consistent with the existing called party information in the call knowledge base: if they are inconsistent, the call address book and the third-party business system are updated synchronously based on the update time of the called party information, selecting the called party information with the most recent update time.

[0009] In one possible implementation of the first aspect above, the call knowledge base contains the user's call address book and call knowledge information corresponding to each call number in the call address book; the call knowledge information includes one or any combination of the following: the called party information corresponding to the call number, call records, and call summary information corresponding to each call record.

[0010] In one possible implementation of the first aspect above, when the intelligent voice call recognition method is applied to an enterprise call scenario, the call knowledge base contains enterprise identity information and / or historical call behavior information of the user and the called party; the call knowledge base also contains call association information in third-party business systems related to the enterprise call scenario.

[0011] In one possible implementation of the first aspect above, the process of determining call knowledge information associated with voice call commands based on the user's call knowledge base includes: obtaining call demand keywords based on the voice call commands; performing semantic reasoning retrieval in the call address book based on the call demand keywords to determine candidate call numbers associated with the voice call commands, and determining the call knowledge information corresponding to the candidate call numbers as call knowledge information associated with the voice call commands.

[0012] In one possible implementation of the first aspect above, the process of determining the call identification number based on the voice call command and call knowledge information includes: selecting the candidate call number with the highest matching degree with the voice call command as the call identification number based on a preset semantic parsing matching rule; wherein the semantic parsing matching rule determines the matching degree between the candidate call number and the voice call command based on the call knowledge information corresponding to the candidate call number.

[0013] In one possible implementation of the first aspect above, when the intelligent voice call recognition method is applied to an enterprise call scenario, the semantic parsing matching rule is customized based on the enterprise call scenario, and the degree of matching between the candidate call number and the voice call command is determined based on the user's enterprise identity information and / or historical call behavior information.

[0014] A second aspect of this disclosure provides an intelligent voice call recognition device, which may specifically include: an acquisition unit for acquiring a user's voice call command; a call knowledge determination unit for determining call knowledge information associated with the voice call command based on the user's call knowledge base; a call number recommendation unit for determining a call identification number based on the voice call command and the call knowledge information; and a call unit for initiating a call based on the call identification number when the user confirms the call identification number.

[0015] The technical solution provided in this disclosure can intelligently identify fuzzy call commands from users based on their call knowledge base and the call knowledge information contained therein, thereby improving the user's intelligent call experience. Furthermore, the technical solution provided in this disclosure does not require users to know the complete call number information; only the known partial information is needed to initiate a correct call. It is convenient to use and particularly suitable for enterprise call scenarios. It can dynamically and accurately provide call identification numbers based on the user's identity, call time, call location, and call history information, improving call efficiency. Simultaneously, it can synchronize call address book information from the call knowledge base to other external third-party enterprise systems, enhancing the user experience. Attached Figure Description

[0016] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0017] Figure 1 This is a flowchart illustrating an intelligent voice call recognition method provided according to an embodiment of the present disclosure.

[0018] Figure 2This is a schematic diagram of a process for determining call knowledge information associated with a voice call command, according to an embodiment of the present disclosure.

[0019] Figure 3 This is a schematic diagram of a process for updating call knowledge information according to an embodiment of the present disclosure.

[0020] Figure 4 This is a schematic diagram of a process for updating the call address book according to an embodiment of the present disclosure.

[0021] Figure 5 This is a schematic diagram of the structure of an intelligent voice call recognition device provided according to an embodiment of the present disclosure.

[0022] Figure 6 This is a schematic diagram of the structure of a computer device provided according to an embodiment of the present disclosure. Detailed Implementation

[0023] As can be understood from the background description, existing voice recognition calling methods still have certain limitations. For example, in the process of fuzzy call matching, they often rely solely on semantic analysis of the user's call commands and cannot perform targeted intelligent number matching based on the user's historical call records, calling habits, and calling scenarios. To solve the above technical problems, some embodiments of this disclosure provide an intelligent voice call recognition method and apparatus that can intelligently recognize the user's fuzzy call commands based on the user's call knowledge base and the call knowledge information contained therein, thereby improving the user's intelligent calling experience.

[0024] To make the objectives, technical solutions, and advantages of the embodiments of this disclosure clearer, the various embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details are provided in the embodiments of this disclosure to facilitate a better understanding of the disclosure. However, the technical solutions claimed in this disclosure can be implemented even without these technical details and various variations and modifications based on the following embodiments. The division of the following embodiments is for ease of description and should not constitute any limitation on the specific implementation of this disclosure. The various embodiments can be combined with and referenced by each other without contradiction.

[0025] In some embodiments of this disclosure, Figure 1 A flowchart illustrating an intelligent voice call recognition method is shown, such as... Figure 1 As shown, process 100 may specifically include the following steps: Step 110: Obtain the user's voice call command. In some embodiments, the user may input the voice call command through an audio input device such as a microphone or microphone, which is not limited here. In some embodiments, the intelligent voice call recognition method provided in this disclosure can be applied to terminal devices with call functions, such as telephone terminals, smartphones, etc., which is not limited here.

[0026] In some practical application scenarios of the above embodiments, terminal devices with calling functions can not only support intelligent voice calling, but also support targeted responses to other voice commands from users. That is, the voice commands provided by users may not only be used to make call requests. In the process of obtaining voice call commands, the voice commands provided by users need to be preprocessed. Specifically, based on semantic recognition, the obtained voice commands can be parsed. For example, the Automatic Speech Recognition Service (ASR) built into the terminal device can be used to convert the voice commands input by the user into text information. Then, functional modules that support natural language understanding, such as large language models, can be used to perform semantic analysis and understanding on the text information to obtain the true meaning of the voice command. Then, if the parsing result of the voice command contains call request information, the voice command input by the user can be used as a voice call command. For example, if the voice command input by the user contains information that can represent a call request, such as "dial," "call," or "contact," the voice command can be used as a voice call command. This is not limited here.

[0027] Step 120: Based on the user's call knowledge base, determine the call knowledge information associated with the voice call command. In some embodiments, the call knowledge base may be established and updated based on the user's call address book and historical call records, and can provide the necessary reference information for intelligent voice call recognition based on the user's historical call situation. In some embodiments, the call knowledge base includes the user's call address book and the call knowledge information corresponding to each call number in the call address book; specifically, the call knowledge information may include one or any combination of the following: the called party information corresponding to the call number, the call call record, and the call summary information corresponding to each call call record. The called party information may include basic information such as the name and age of the called party corresponding to the call number. The call call record may include the call time, call duration, call frequency, etc. between the user and the call number. The call summary information may be a summary of the content of the call between the user and the call number, which is not limited here. In some embodiments, further, when the intelligent voice call recognition method is applied to enterprise call scenarios, the call knowledge base contains enterprise identity information and / or historical call behavior information of the user and the called party. Specifically, the enterprise identity information may include the user's and the called party's corresponding internal positions, department, and location within the enterprise; the historical call behavior information may include the user's and the called party's historical call time, historical call location, historical call reason, and whether the historical call was answered, etc., without limitation. In some embodiments, the call knowledge base may also contain call association information from third-party business systems associated with the enterprise call scenario. For example, during an enterprise call through an IPPBX system, data associated with the enterprise call can be synchronized from third-party business systems associated with the IPPBX system, such as Customer Relationship Management (CRM) service systems, Property Management System (PMS) interface service systems, and Lightweight Directory Access Protocol (LDAP) service systems. Specifically, the call association information may include customer call numbers not entered into the IPPBX system, customer identity information, etc., without limitation. In some embodiments, the data content included in the call knowledge base may be partially stored locally in the call knowledge base, or it may be invoked and / or retrieved from other third-party business systems through a pre-configured communication interface, without limitation herein. The process for determining call knowledge information associated with voice call commands will be described in detail later and will not be repeated here.

[0028] Step 130: Determine the call identification number based on the voice call command and call knowledge information. In some embodiments, when the user's voice call command is accurate and clear, such as "Please dial number 5001" or "Please dial Zhang Sansi's office extension," a unique call identification number can be directly determined based on the voice call command and relevant call knowledge information in the call address book. In other embodiments, when the user's voice call command is not specific or is vague, such as "Please call Xiao Zhang" or "Call the most recent contact," call knowledge information can be used as a reference, and an artificial intelligence server can be used to determine the call identification number in the call address book that best matches the voice call command. Users can also choose to filter and determine the call identification number according to their actual needs, such as using a large language model or preset rule matching, which is not limited here. The specific process for determining the call identification number will be explained in detail later and will not be repeated here.

[0029] Step 140: If the user confirms the call identification number, initiate a call based on the call identification number. In some embodiments, for a determined call identification number, the call identification number can be converted into an interactive prompt tone for user confirmation through the text-to-speech service (TTS) built into the terminal device. For example, when the user inputs the voice call command "Please call Xiao Zhang", after obtaining the call identification number based on the voice call command and call knowledge information, the following interactive prompt tone can be generated: "The number information of customer Xiao Zhang has been found: Zhang San, 5011, confirm call?", and the user can confirm. In other embodiments, the user can also confirm the call identification number by presenting it in a visual interface on the terminal device, etc., which is not limited here. In some embodiments, if the user confirms the call identification number, a call can be initiated based on the call identification number, thereby realizing the entire intelligent voice call recognition process.

[0030] Based on the foregoing descriptions of the embodiments, it is understood that the intelligent voice call recognition method provided in this disclosure can perform multi-dimensional semantic analysis and reasoning based on the user's voice call command, combined with historical call records, the identity information of the called party, etc., and use artificial intelligence technology to filter out the call identification number that best matches the voice call command and initiate a call. It is particularly suitable for fuzzy calls in enterprise call scenarios, such as in IPPBX (Internet Protocol Private Branch Exchange) systems. In some embodiments, an IPPBX system is a telephone exchange system based on the IP protocol, typically used for internal enterprise communication. By combining traditional PBX (Private Branch Exchange) technology with modern IP technology, it enables enterprises to conduct voice communication via the Internet or local area network. In an IPPBX system, all users share the same call address book, making it even more necessary to utilize the solution provided in this disclosure to filter and determine the corresponding call number based on the actual call needs of different users. The implementation of the intelligent voice call recognition method provided in this disclosure will be further explained and illustrated below with reference to embodiments.

[0031] In some embodiments of this disclosure, further, Figure 2 A flowchart illustrating a process for determining call knowledge information associated with a voice call command is shown, such as... Figure 2 As shown, process 200 may specifically include the following steps: Step 210: Obtain call request keywords based on voice call commands. In some embodiments, considering that semantic analysis and understanding of the text information corresponding to the voice call command has been performed using functional modules supporting natural language understanding, such as large language models, in the aforementioned process of obtaining the voice call command, the semantic analysis and understanding results can be reused in the implementation of step 210 to extract call request keywords from the voice call command. The call request keywords can reflect the user's core call information. For example, taking the voice call command "Please call Xiao Zhang" as an example, the corresponding call request keyword can be "Xiao Zhang", and it can be further represented as "the called person's surname is 'Zhang'" based on semantic analysis; another example is taking the voice call command "Please contact Lao Liu from the R&D department" as an example, the corresponding call request keyword can be a combination of "R&D department" and "Lao Liu", and it can be further represented as "the called person's surname is 'Liu', and the department is the R&D department" based on semantic analysis. There is no limitation here.

[0032] Step 220: Based on the call demand keywords, perform semantic reasoning retrieval in the call address book to determine candidate call numbers associated with the voice call command, and determine the call knowledge information corresponding to the candidate call numbers as call knowledge information associated with the voice call command. It is understood that the call demand keywords contain the user's core call information. It is difficult to directly determine the user's required unique call identification number from the call address book using only this core call information. However, multiple associated call numbers can be filtered from the call address book as candidate call numbers, and the call knowledge information corresponding to the candidate call numbers can be determined as call knowledge information associated with the voice call command, laying the foundation for the intelligent determination of the subsequent call identification number. In some embodiments, taking the voice call command "Please call Xiao Zhang" as an example, the obtained call demand keywords can be represented as "the called person's surname is 'Zhang'". Based on this, a fuzzy matching retrieval can be performed in the call address book. According to the called person information corresponding to the call number, all call numbers whose names contain "Zhang" can be filtered as candidate call numbers, and the call knowledge information corresponding to these candidate call numbers can be extracted, which may include, for example, the following information: Candidate Call Number: 1000; Name: Zhang Er; Department: R&D Department 1; Location: Hangzhou; Recent Call Record: October 20, 2025; Call Summary: Communication regarding requirements for the XX project; Candidate Call Number: 1001; Name: Zhang San; Department: R&D Department 1; Location: Los Angeles; Candidate Call Number: 2001; Name: Zhang San; Department: R&D Department II; Location: Hangzhou; Candidate Call Number: 2002; Name: Zhang Si; Department: Administration; Location: Los Angeles; Candidate call number: 5001; Name: Jack Zhang; Partner client; Location: Shanghai.

[0033] In some embodiments, it is possible that the user's voice call command contains erroneous information that is difficult to match. For example, taking the voice call command "Please contact Lao Liu in the R&D department" as an example, the obtained call request keywords can be represented as "the called person's surname is 'Liu' and their department is the R&D department." Based on this, a matching search can be performed in the call address book. If no call number is found that simultaneously meets the conditions of having "Zhang" in the name and belonging to the "R&D department," it may be due to a memory bias, where the user has misremembered the department to which "Lao Liu" belongs. In this case, partial information from the call request keywords can be used to filter out all call numbers whose called person's name contains "Liu" and whose called person belongs to the "R&D department" as candidate call numbers. The corresponding call knowledge information for these candidate call numbers can then be extracted, without limitation. It is understood that the candidate call numbers and call knowledge information obtained from the above filtering are multiple alternatives highly correlated with the voice call command, providing a narrower and more accurate filtering data basis for subsequent call identification number confirmation.

[0034] In some embodiments of this disclosure, further, in the process of determining the call identification number, an artificial intelligence server with artificial intelligence analysis capabilities can be used to comprehensively obtain the most suitable call identification number by combining call knowledge information and voice call commands. Specifically, in some embodiments, candidate call numbers with the highest degree of matching with voice call commands can be selected as call identification numbers based on preset semantic parsing matching rules. The semantic parsing matching rules can determine the degree of matching between candidate call numbers and voice call commands based on call knowledge information corresponding to the candidate call numbers. Specifically, the degree of matching between candidate call numbers and voice call commands can be determined based on at least one or any combination of user identity information, called party information, voice call command initiation time, voice call command initiation location, and user's historical call records. In some embodiments, further, when the intelligent voice call recognition method is applied to enterprise call scenarios, the semantic parsing matching rules can be customized based on the enterprise call scenario and determine the degree of matching between candidate call numbers and voice call commands based on the user's enterprise identity information and / or historical call behavior information. For example, in enterprise call scenarios, for the same or similar voice call commands, different enterprise identity information of users will lead to differences in the matching call identification number. For example, taking the voice call command "Please call Xiao Zhang" as an example, if the user's identity is a staff member of R&D Department 2, then the matching degree of Zhang Si, who belongs to the same department, with call knowledge information "Call number: 2001; Name: Zhang San; Department: R&D Department 2; Location: Hangzhou" will be higher than the matching degree of other call numbers. If the user's identity is a staff member of the Administration Department, then the matching degree of Zhang Si, who belongs to the same department, with call knowledge information "Call number: 2002; Name: Zhang Si; Department: Administration Department; Location: Los Angeles" will be higher than the matching degree of other call numbers. That is, the higher the matching degree of the called party with the same department or related departments, the better the matching degree of the current user's voice call command. This will not be elaborated further here.For example, in enterprise call scenarios, the initiation time and / or location of different voice call commands can lead to differences in the matching call identification numbers. For instance, if a user in R&D Department 1 issues the voice call command "Please call Xiao Zhang" domestically, and the voice call command is issued at 10:00 AM Beijing time, the matching degree for Zhang Er, who belongs to the same department, with the call information "Call number: 1000; Name: Zhang Er; Department: R&D Department 1; Location: Hangzhou; Recent call record: October 20, 2025; Call summary: XX project requirements communication" will be higher than that of other call numbers. If the voice call command is issued at 10:00 PM Beijing time, the matching degree for Zhang San, who belongs to the same department but is located overseas, with the call information "Call number: 1001; Name: Zhang San; Department: R&D Department 1; Location: Los Angeles" will be higher than that of other call numbers. This is due to the different locations of the called parties and the different call times, and will not be limited here. In some embodiments, the aforementioned preset semantic parsing matching rules may be obtained by an artificial intelligence server through training based on the historical call behavior information of each user using machine learning or neural network learning methods, and continuously updated iteratively according to the continuous updates of changes in user call behavior. In other embodiments, at least some of the aforementioned preset semantic parsing matching rules may also be manually preset configured, such as "the matching degree of call numbers with recent call records is higher", "the matching degree of call numbers with the same location as the user is higher", "the matching degree of call numbers marked as 'cooperative customers' for sales department users is higher", etc. Users can choose the setting and update method of semantic parsing matching rules according to their actual needs, which is not limited here.

[0035] In some embodiments of this disclosure, the intelligent voice call recognition method provided by this disclosure, in addition to filtering out matching call identification numbers based on voice call commands during the dialing stage, also supports recording and summarizing the call during the call phase to achieve continuous updating of call knowledge information. Specifically, Figure 3 A schematic diagram of a process for updating call knowledge information is shown, such as... Figure 3 As shown, process 300 may include the following steps: Step 310: Generate call summary information based on the call recording of the call identification number. It is understood that call knowledge information includes the historical call records of the call number and the corresponding call summary information. After the call is completed using the call identification number, the call recording can be converted into text information using the automatic speech recognition service built into the terminal device. Then, semantic analysis and summarization of the text information can be performed using functional modules supporting natural language understanding, such as large language models, to obtain the corresponding call summary information. No specific limitations are imposed here.

[0036] Step 320: Update the call knowledge information corresponding to the call identification number based on the call summary information and the call record of the call identification number. It is understood that by binding the call summary information and the call record, the call knowledge information for the call identification number can be updated. This allows for an update of the call knowledge information for each call made by the user, ensuring that the call knowledge base can reflect the user's latest calling habits and recent calling activity in real time. No further limitations are imposed here.

[0037] In some embodiments of this disclosure, the intelligent voice call recognition method provided further supports timely and autonomous updating of the call knowledge base based on call details when the calling number is the first call number. Specifically, Figure 4 A flowchart illustrating a process for updating the caller's address book is shown, such as... Figure 4 As shown, process 400 may include the following steps: Step 410: If the call identification number is the first call number, update the call address book in the call knowledge base based on the call identification number. It is understood that the call identification number provided in the foregoing embodiments can be a unique call identification number directly determined based on the voice call command. For example, it can be that the number to be called is directly included in the voice call command, which is not limited here. If the number to be called is directly included in the voice call command, the call identification number may be the first call number not present in the call knowledge base. In this case, the call address book and call knowledge base can be updated accordingly to ensure that the call knowledge base reflects the user's latest call records.

[0038] In some embodiments, further, in enterprise call scenarios, during the process of making a call using an IPPBX system, users can selectively enable Customer Relationship Management (CRM) services, and automatically add the caller ID to the CRM system when the current caller ID does not exist in the CRM system, greatly saving users the potential work of caller ID synchronization and maintenance. In other embodiments, users can also selectively enable one or more combinations of Property Management System (PMS) integration services, Lightweight Directory Access Protocol (LDAP) services, and other third-party service functions, and automatically add the caller ID when the current caller ID does not exist in the management system corresponding to the above services, which will not be elaborated here.

[0039] Step 420: When the call recording of the call identification number contains the called party information corresponding to the call identification number, update the call address book based on the called party information. It is understood that for the first call number, the call knowledge base only contains isolated information about the call number and cannot directly obtain information such as the called party corresponding to the call number. In this case, the call recording corresponding to the current call can be used to proactively and intelligently identify the called party information that may be contained therein, saving the user the process of additionally inputting the called party information. The identification and acquisition of the called party information can be obtained in the same way as the call summary information in the aforementioned embodiments, and will not be elaborated here. In some embodiments, furthermore, the solution provided by this disclosure is not only applicable to point-to-point calls between users and called parties, but also applicable to multi-party call access conferences in enterprise call scenarios. For all parties in a multi-party call access conference, the identification and updating of the caller identity information corresponding to the call number can be performed through steps 410 to 420 above, and there are no limitations here.

[0040] Step 430: When the call recording does not contain the called party's information, prompt the user to supplement the called party's information, and update the call address book based on the supplemented called party's information. It is understandable that when the call recording does not contain the called party's information, the system can proactively prompt the user to supplement the called party's information, for example, through voice prompts, visual interface presentation, etc., and update the address book based on the user's supplemented called party information; this is not limited to this step.

[0041] In some embodiments of this disclosure, further, when the intelligent voice call recognition method is applied to enterprise call scenarios, when a call identification number is received or made, the content of the call knowledge base can be updated collaboratively and autonomously based on the current call status of the call identification number and relevant data in third-party business systems associated with the enterprise call scenario. Specifically, when a call identification number is received or made, it can first be determined whether the call identification number exists in the call knowledge base: if it does not exist, it means that the call identification number may be making its first incoming or outgoing call, and the call knowledge base needs to be updated in a timely manner. At this time, the called party information corresponding to the call identification number can be retrieved from third-party business systems associated with the enterprise call scenario. For example, during the process of making a call using an IPPBX system, the information can be retrieved from third-party business systems associated with the IPPBX system, such as Customer Relationship Management (CRM) service systems, Property Management System (PMS) interface service systems, and Lightweight Directory Access protocols. The system can automatically retrieve the called party information corresponding to the call identification number using Protocol (LDAP) service systems, etc. If the called party information corresponding to the call identification number has already been entered into the aforementioned third-party business system, the called party information can be automatically extracted directly and the call address book in the call knowledge base can be updated adaptively, providing users with a better automated information synchronization experience. If the called party information corresponding to the call identification number cannot be retrieved from the aforementioned third-party business system, the called party information can be obtained using the acquisition method related to process 400 provided in the aforementioned embodiment, which is not limited here.

[0042] In some embodiments, if the call identification number exists in the call knowledge base, the called party information corresponding to the call identification number can be retrieved from the third-party business system, and the retrieved called party information can be verified to be consistent with the existing called party information in the call knowledge base. If they are consistent, it means that the called party information in the call knowledge base and the called party information in the third-party business system have not been updated after historical synchronization. If they are inconsistent, the call party information with the most recent update time can be selected to synchronously update the call address book and the third-party business system based on the update time of the called party information. This enables the synchronous update of called party information between the third-party business system and the call knowledge base, allowing all systems associated with the enterprise call scenario to share the latest called party information in a timely manner, thus achieving automated information synchronization between systems.

[0043] In some embodiments of this disclosure, Figure 5 A schematic diagram of the structure of an intelligent voice call recognition device is shown, such as... Figure 5 As shown, this intelligent voice call recognition device may include an acquisition unit 510, a call knowledge determination unit 520, a call number recommendation unit 530, and a call unit 540.

[0044] In some embodiments, the acquisition unit 510 can be used to acquire the user's voice call command. Specifically, it can receive the user's voice command through an audio input device such as a microphone, and parse the voice command based on natural language recognition to determine whether the voice command contains a call request. If so, the received voice command is taken as the user's voice call command. In some embodiments, the call knowledge determination unit 520 can be used to determine call knowledge information associated with the voice call command based on the user's call knowledge base. The call knowledge base may contain the user's call address book and call knowledge information corresponding to each call number in the address book. The call knowledge information may be the identity information of the called party corresponding to the call number, historical call records, historical call content summaries, etc. The call knowledge information associated with the voice call command is filtered out by fuzzy matching. In some embodiments, the call number recommendation unit 530 can be used to determine the call identification number based on the voice call command and the call knowledge information. Specifically, it can filter out the call number with the highest matching degree with the voice call command from multiple call numbers according to semantic parsing matching rules. In some embodiments, the call unit 540 can be used to initiate a call based on the call identification number when the user confirms the call identification number, so as to realize the entire intelligent voice call recognition process. It is understood that the specific functional implementation of the above-mentioned acquisition unit 510 to call unit 540 can refer to the specific implementation process of each step in the intelligent voice call recognition method in the foregoing embodiment, and will not be repeated here.

[0045] Some embodiments of this disclosure also relate to a computer device, which can be configured as a telephone terminal device provided in the embodiments of this disclosure, or as an artificial intelligence server, call knowledge base external storage device, etc., provided in the embodiments of this disclosure, and is not limited thereto. In some embodiments, Figure 6 A schematic diagram of the structure of a computer device is shown, such as Figure 6 As shown, the computer device includes at least one processor 610 and a memory 620 communicatively connected to the at least one processor. The memory 620 stores instructions that can be executed by the at least one processor 610. The instructions are executed by the at least one processor 610 to enable the at least one processor 610 to execute and implement the intelligent voice call recognition method provided in the foregoing embodiments.

[0046] The memory 620 and processor 610 are connected via a bus, which may include any number of interconnecting buses and bridges, connecting various circuits of one or more processors 610 and memory 620 together. The bus may also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. A bus interface provides an interface between the bus and the transceiver. The transceiver may be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by processor 610 is transmitted over a wireless medium via an antenna, which further receives data and transmits it to the processor.

[0047] In some embodiments, the processor 610 may be responsible for managing the bus and general processing, and may also provide various functions, including reporting emergency stop button trigger events, receiving stop control commands, controlling devices such as safety relays to perform specific stop operations, and other control functions, while the memory 620 may be used to store data used by the processor when performing operations, without limitation.

[0048] Some embodiments of this disclosure also relate to a computer-readable storage medium storing computer instructions, which, when executed by a processor, implement the intelligent voice call recognition method provided in the foregoing embodiments. In some embodiments, the computer-readable storage medium may include flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the computer-readable storage medium may be an internal storage unit of a computer device, such as the hard disk or memory of the computer device. In other embodiments, the computer-readable storage medium may also be an external storage device of a computer device, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc., provided on the computer device. Of course, the computer-readable storage medium may also include both internal storage units and external storage devices of a computer device. In this embodiment, the computer-readable storage medium is typically used to store the operating system and various application software installed on the computer device, such as the program code of the intelligent voice call recognition method in this embodiment. Furthermore, the computer-readable storage medium can also be used to temporarily store various types of data that have been output or will be output.

[0049] The basic concepts have been described above. It is obvious that the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, various modifications, improvements, and corrections may be made to this specification by those skilled in the art. Such modifications, improvements, and corrections are taught in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

Claims

1. A method for intelligent voice call recognition, characterized in that, include: Obtain the user's voice call command; Based on the user's call knowledge base, determine the call knowledge information associated with the voice call command; Based on the voice call command and the call knowledge information, the call identification number is determined; If the user confirms the call identification number, a call is initiated based on the call identification number.

2. The intelligent voice call recognition method according to claim 1, characterized in that, Also includes: Based on the call recordings of the call identification number, call summary information is generated; Based on the call summary information and the call record corresponding to the call recording, update the call knowledge information corresponding to the call identification number.

3. The intelligent voice call recognition method according to claim 1, characterized in that, Also includes: If the call identification number is the first call number, the call address book is updated in the call knowledge base based on the call identification number; When the call recording of the call identification number contains the called party information corresponding to the call identification number, the call address book is updated based on the called party information; If the called party information is not included in the call recording, the user is prompted to supplement the called party information, and the call address book is updated based on the supplemented called party information.

4. The intelligent voice call recognition method according to claim 1, characterized in that, When the intelligent voice call recognition method is applied to an enterprise call scenario, if the call identification number is present when a call comes in or goes out, it is determined whether the call identification number exists in the call knowledge base: If the call identification number does not exist in the call knowledge base, the callee information corresponding to the call identification number is retrieved from the third-party business system associated with the enterprise call scenario, and the call address book is updated based on the retrieved callee information. If the call identification number exists in the call knowledge base, retrieve the called party information corresponding to the call identification number from the third-party business system, and verify whether the retrieved called party information is consistent with the called party information already existing in the call knowledge base. If they are inconsistent, select the called party information with the most recent update time according to the update time of the called party information to synchronously update the call address book and the third-party business system.

5. The intelligent voice call recognition method according to claim 1, characterized in that, The call knowledge base contains the user's call address book, and the call knowledge information corresponding to each call number in the call address book; The call knowledge information includes one or any combination of the following: the called party information corresponding to the call number, call records, and call summary information corresponding to each call record.

6. The intelligent voice call recognition method according to claim 5, characterized in that, When the intelligent voice call recognition method is applied to an enterprise call scenario, the call knowledge base contains the enterprise identity information and / or historical call behavior information of the user and the called party. The call knowledge base also contains call association information from third-party business systems related to the enterprise's call scenarios.

7. The intelligent voice call recognition method according to claim 5, characterized in that, The process of determining call knowledge information associated with the voice call command based on the user's call knowledge base includes: Based on the voice call command, obtain the call request keywords; Based on the call demand keywords, semantic reasoning retrieval is performed in the call address book to determine candidate call numbers associated with the voice call command, and the call knowledge information corresponding to the candidate call numbers is determined as call knowledge information associated with the voice call command.

8. The intelligent voice call recognition method according to claim 7, characterized in that, The process of determining the call identification number based on the voice call command and the call knowledge information includes: Based on preset semantic parsing and matching rules, the candidate call number with the highest degree of matching with the voice call command is selected as the call identification number; The semantic parsing matching rule determines the degree of matching between the candidate call number and the voice call command based on the call knowledge information corresponding to the candidate call number.

9. The intelligent voice call recognition method according to claim 8, characterized in that, When the intelligent voice call recognition method is applied to an enterprise call scenario, the semantic parsing matching rule is customized based on the enterprise call scenario, and the degree of matching between the candidate call number and the voice call command is determined based on the user's enterprise identity information and / or historical call behavior information.

10. An intelligent voice call recognition device, characterized in that, The device includes: The acquisition unit is used to acquire the user's voice call command; The call knowledge determination unit is used to determine call knowledge information associated with the voice call command based on the user's call knowledge base; A call number recommendation unit is used to determine a call identification number based on the voice call command and the call knowledge information; A calling unit is used to initiate a call based on the call identification number when the user confirms the call identification number.

Images