A method and system for handling intelligent messages in a voice assistant mailbox in an outbound calling system

Abstract

The application provides an outcalling system intelligent message method and system for coping with a voice assistant mailbox, belongs to the technical field of intelligent contact centers, and specifically comprises the following steps: a calling module realizes calling processing on a user; a message recognition module determines whether to enter a voice assistant mailbox according to a connection processing time length or a voice recognition result; a switching module automatically switches to a message processing module when it is determined that the voice assistant mailbox is entered, and continues to switch to the calling module to realize calling processing on other users; and the message processing module generates a voice message according to an analysis result of user information of the user, thereby improving the efficiency of touch processing.

Description

Technical Field

[0001] This invention belongs to the field of intelligent contact center technology, and in particular relates to a method and system for responding to intelligent messages in a voice assistant mailbox in an outbound calling system. Background Technology

[0002] Existing debt collection agencies or marketing organizations often use automated robots to handle calls with target users, as exemplified by patent applications CN202410075284.4 "A Debt Collection Device Based on a Smart Mobile Terminal" and CN202311577368.X "A Debt Collection Method and System Based on a Smart Voice Robot." However, whether using human agents or robots, voicemail messages are frequently encountered, which presents the following technical problems:

[0003] If voicemail mailboxes cannot be recognized, especially when using robots to process calls, the overall access efficiency may be slowed down.

[0004] When users access voicemail, existing technologies often process messages as if they were already connected, lacking a personalized message processing solution based on the caller's individual information. This can negatively impact a user's willingness to respond to messages received.

[0005] To address the aforementioned technical problems, this application provides a method and system for an outbound calling system to respond to intelligent messages from a voice assistant mailbox. Summary of the Invention

[0006] To achieve the objectives of this invention, the following technical solution is adopted:

[0007] Specifically, in the first aspect, this application provides a system for responding to intelligent messages in a voice assistant mailbox during an outbound call system, specifically including:

[0008] Call module, message recognition module, switching module, message processing module;

[0009] The call module described above handles user calls.

[0010] The message recognition module determines whether to enter the voice assistant mailbox based on the connection processing time or the voice recognition result.

[0011] The switching module automatically switches to the message processing module after confirming that the user has entered the voice assistant mailbox, and then switches to the call module to process calls to other users.

[0012] The message processing module generates a voice message based on the analysis results of the user's user information.

[0013] The beneficial effects of this invention are as follows:

[0014] Once the system confirms entry into the voice assistant mailbox, it automatically switches to the message processing module and then to the call module to process calls to other users. This avoids the impact of entering the voice assistant mailbox on the efficiency of collection processing, thus improving the efficiency of collection processing while ensuring its reliability.

[0015] Based on the analysis of user information, voice messages are generated, enabling the automatic generation of personalized collection voice messages for each user. This ensures that the collection voice messages accurately match the user's identity information and overdue status, further improving the efficiency of collection processing.

[0016] A further technical solution is that the call module uses either manual calling or voice robot calling.

[0017] A further technical solution is that when the connection processing time exceeds a preset processing time threshold or when it is determined from the voice recognition result that the call has entered the voice assistant mailbox, the user's call is confirmed to have entered the voice assistant mailbox.

[0018] A further technical solution involves generating a voice message based on the analysis results of the user's information, specifically including:

[0019] Based on the analysis results of the user's user information, a user profile of the user is determined;

[0020] A voice message is generated based on the preset voice message corresponding to the user's user profile.

[0021] A further technical solution involves using a voice robot to determine the user profile based on the user's user information as input.

[0022] A further technical solution involves building the voice robot using a voice synthesis model and a large model.

[0023] Secondly, this application provides a method for an outbound calling system to respond to intelligent messages in a voice assistant mailbox, applied to the aforementioned system for an outbound calling system to respond to intelligent messages in a voice assistant mailbox, specifically including:

[0024] S1 determines the user's call connection data in the current time period, and combines the voice recognition results of different connected calls to determine if the user's call connection probability meets the requirements, then proceeds to the next step;

[0025] S2 determines, based on the user's user information, that when the user does not have a matching user profile, that the user's historical communication data is stored on different servers based on the analysis and processing results of the user's historical communication data.

[0026] S3 determines the matching server in the server based on the storage situation, and determines the associated server of the user profile within the preset time period based on the storage situation of historical communication data in different matching servers at different time periods;

[0027] S4 determines the call delay data between the voice robot and different associated servers, and based on the call delay data and the historical communication data of the associated servers within a preset time period, determines the message generation processing method of the voice robot's voice assistant mailbox.

[0028] A further technical solution is that the time period is divided into one-hour or half-hour segments.

[0029] A further technical solution is that the connection data includes the number of connections and the connection rate during the time period.

[0030] A further technical solution is that the voice recognition result of the connected call includes the distribution data of keywords of the user in different connected calls during the time period.

[0031] A further technical solution involves determining the keywords using a rejection keyword database that includes phrases like "inconvenient during working hours," "in a meeting," and "don't call again."

[0032] A further technical solution involves determining that the user's connection probability meets the requirements, specifically including:

[0033] Based on the connected data in the current time period, determine the percentage of unconnected calls in the time period;

[0034] Based on the voice recognition results of connected calls in the current time period, determine the percentage of connected calls that match the rejection keyword database, and use this percentage as the matching call percentage.

[0035] The user's rejection probability is determined by the sum of the percentage of unanswered calls and the percentage of matched calls, and the rejection probability is used to determine whether the user's connection probability meets the requirements.

[0036] A further technical solution is that the method for determining the message generation and processing method of the voice assistant mailbox of the voice robot is as follows:

[0037] Using the call latency data between the voice robot and different associated servers, determine the average call latency duration of the voice robot in the call processing process with different associated servers, and determine the delayed call server among the associated servers based on the average call latency duration.

[0038] Based on the historical communication data of the delayed call server within a preset time period, determine the storage duration of the collection voice data of the delayed call server within the preset time period, and the proportion of the collection duration of the user's collection voice data within the preset time period, and use it as the collection duration proportion of the delayed call server.

[0039] The message generation and processing method of the voice robot's voice assistant mailbox is determined based on the sum of the collection time proportions of different delayed server calls.

[0040] Other features and advantages will be set forth in the following description, and the objects and other advantages of the invention are realized and obtained through the structures particularly pointed out in the description and the drawings.

[0041] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0042] The above and other features and advantages of the present invention will become more apparent from a detailed description of exemplary embodiments thereof with reference to the accompanying drawings.

[0043] Figure 1 This is a framework diagram of an outbound calling system that responds to intelligent messages in a voice assistant mailbox;

[0044] Figure 2 This is a flowchart illustrating a method for an outbound calling system to respond to intelligent messages in a voice assistant mailbox;

[0045] Figure 3 This is a flowchart for determining whether a user's connection probability meets the requirements;

[0046] Figure 4 This is a flowchart for determining if a user does not have a matching user profile;

[0047] Figure 5 This is a flowchart illustrating the method for determining the associated servers for a user profile within a preset time period. Detailed Implementation

[0048] To enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this specification, and not all embodiments. Based on the embodiments of this specification, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of this specification.

[0049] In this application, by recognizing and processing the user's voice assistant mailbox, a targeted voice message can be generated, and the call can be switched to another user in a timely manner, thereby improving the efficiency of call processing.

[0050] Example 1

[0051] like Figure 1 As shown, this application provides a system for responding to intelligent messages in a voice assistant mailbox during an outbound call system, specifically including:

[0052] Call module, message recognition module, switching module, message processing module;

[0053] The call module described above handles user calls.

[0054] The message recognition module determines whether to enter the voice assistant mailbox based on the connection processing time or the voice recognition result.

[0055] The switching module automatically switches to the message processing module after confirming that the user has entered the voice assistant mailbox, and then switches to the call module to process calls to other users.

[0056] The message processing module generates a voice message based on the analysis results of the user's user information.

[0057] Furthermore, the calling module can use either human call or voice robot call.

[0058] Specifically, when the connection processing time exceeds a preset processing time threshold or when it is determined from the voice recognition result that the call has entered the voice assistant mailbox, the user's call is confirmed to have entered the voice assistant mailbox.

[0059] It is understandable that, based on the analysis results of the user's information, a voice message is generated, specifically including:

[0060] Based on the analysis results of the user's user information, a user profile of the user is determined;

[0061] A voice message is generated based on the preset voice message corresponding to the user's user profile.

[0062] A further technical solution involves using a voice robot to determine the user profile based on the user's user information as input.

[0063] Specifically, the voice robot is built using a voice synthesis model and a large model.

[0064] Example 2

[0065] Secondly, such as Figure 2 As shown, this application provides a method for an outbound calling system to respond to intelligent messages in a voice assistant mailbox, applied to the aforementioned system for an outbound calling system to respond to intelligent messages in a voice assistant mailbox, specifically including:

[0066] S1 determines the user's call connection data in the current time period, and combines the voice recognition results of different connected calls to determine if the user's call connection probability meets the requirements, then proceeds to the next step;

[0067] If the sum of the percentage of unanswered calls and the percentage of answered calls matching the rejection keyword database during a given time period is greater than 0.6, then the user's call connection probability is determined to be unsatisfactory.

[0068] S2 determines, based on the user's user information, that when the user does not have a matching user profile, that the user's historical communication data is stored on different servers based on the analysis and processing results of the user's historical communication data.

[0069] If the percentage of identical information items with different preset user profiles is less than 0.4, then it is determined that there is no matching user profile.

[0070] S3 determines the matching server in the server based on the storage situation, and determines the associated server of the user profile within the preset time period based on the storage situation of historical communication data in different matching servers at different time periods;

[0071] The server that stores historical communication data within a preset time period is designated as the associated server.

[0072] S4 determines the call delay data between the voice robot and different associated servers, and based on the call delay data and the historical communication data of the associated servers within a preset time period, determines the message generation processing method of the voice robot's voice assistant mailbox.

[0073] Furthermore, the time period is divided into one-hour or half-hour segments.

[0074] Specifically, the connection data includes the number of connections and the connection rate during the specified time period.

[0075] It should be noted that the voice recognition results of the connected call include the distribution data of keywords in the user's connected calls at different times during the time period.

[0076] Understandably, the keywords were determined using a rejection keyword database that included phrases like "inconvenient during work hours," "in a meeting," and "don't call again."

[0077] Specifically, such as Figure 3 As shown, determining that the user's connection probability meets the requirements specifically includes:

[0078] Based on the connected data in the current time period, determine the percentage of unconnected calls in the time period;

[0079] Based on the voice recognition results of connected calls in the current time period, determine the percentage of connected calls that match the rejection keyword database, and use this percentage as the matching call percentage.

[0080] The user's rejection probability is determined by the sum of the percentage of unanswered calls and the percentage of matched calls, and the rejection probability is used to determine whether the user's connection probability meets the requirements.

[0081] It is understandable that when the rejection probability is greater than the preset rejection probability threshold, the user's connection probability is determined to be unsatisfactory.

[0082] It should be noted that when the user's connection probability does not meet the requirements, a voice robot is used to generate messages for the voice assistant's mailbox before the user is called.

[0083] Optionally, determining that the user's connection probability meets the requirements specifically includes:

[0084] The connection rate for the current time period is determined based on the connection data within that time period.

[0085] Based on the voice recognition results of connected calls in the current time period, determine the percentage of connected calls that match the rejection keyword database, and use this percentage as the matching call percentage.

[0086] Based on the ratio of the connection rate to the proportion of matched calls, the connection probability assessment value of the user is determined, and based on the connection probability assessment value, it is determined whether the connection probability of the user meets the requirements.

[0087] Furthermore, if the connection probability assessment value is less than a preset assessment value threshold, then it is determined that the user's connection probability does not meet the requirements.

[0088] It is understood that the user information includes the user's overdue amount and overdue time, the user's job, age, income, gender, marital status, the user's historical collection attempts, and collection voice data for different historical collection attempts.

[0089] It should be noted that, as Figure 4 As shown, determining that the user does not have a matching user profile specifically includes:

[0090] Based on the user information of the user, identify the similarities in user information across different dimensions with different preset user profiles;

[0091] Based on the aforementioned similarities, the number of identical information items is determined, and based on the proportion of identical information items, the profile matching coefficient between the user and different preset user profiles is determined.

[0092] Based on the profile matching coefficients with different preset user profiles, it is determined whether the user has a matching user profile.

[0093] Furthermore, if the matching coefficient of different preset user profiles is less than the preset user profile matching coefficient threshold, then it is determined that the user does not have a matching user profile.

[0094] Specifically, when there is a preset user profile with a profile matching coefficient not less than a preset profile matching coefficient threshold, the preset user profile with the largest profile matching coefficient is used as the matching user profile. When it is determined that the user has entered the voice assistant mailbox, the preset message information corresponding to the matching user profile is used to generate a message for the voice assistant.

[0095] Furthermore, the storage of the user's historical communication data on different servers includes collection voice data on different servers, historical collection calls corresponding to the collection voice data, and collection time of the historical collection calls.

[0096] It is understood that the preset time period is a time period within the most recent preset duration threshold.

[0097] Specifically, such as Figure 5 As shown, the method for determining the associated server of the user profile within a preset time period is as follows:

[0098] Based on the storage status of historical communication data in the matching server at different time periods, determine the number of historical collection calls stored by the matching server within a preset time period;

[0099] Based on the number of historical collection calls stored by the matching server within a preset time period, and the proportion of the number of historical collection calls of the user within the preset time period, the proportion of collection calls of the matching server is determined.

[0100] Based on the percentage of collection calls, determine whether the server is the associated server of the user profile within a preset time period.

[0101] Furthermore, when the proportion of collection calls exceeds a preset call proportion threshold, the server is determined to be the associated server of the user profile within a preset time period.

[0102] Optionally, the method for determining the associated server of the user profile within a preset time period is as follows:

[0103] Obtain the number of collection voice calls within a preset time period stored on the matching server and the storage duration of the collection voice data;

[0104] It should be noted that before proceeding to the next step, it is necessary to further determine whether the matching server has collection voice data within a preset time period. This involves checking whether the number of historical collection calls stored on the matching server within the preset time period, the percentage of such calls in the user's historical collection calls within the preset time period, and the storage duration of the collection voice data, as well as its percentage of the user's collection voice data collection duration within the preset time period, all meet the requirements. If the matching server does not have collection voice data within the preset time period, then it is determined that the matching server is not an associated server.

[0105] Furthermore, when the matching server stores collection voice data within a preset time period, if any one of the following conditions is met: the number of historical collection calls stored by the matching server within the preset time period, the proportion of the number of historical collection calls of the user within the preset time period, the storage duration of the collection voice data, and the proportion of the collection duration of the collection voice data of the user within the preset time period, then the matching server is determined to be an associated server.

[0106] The basic correlation coefficient between the matching server and the historical communication data within the preset time period is determined based on the number of historical collection calls stored in the matching server within the preset time period and the duration of different historical collection calls.

[0107] It should be noted that before proceeding to the next step, it is necessary to further determine whether the basic correlation coefficient is less than the preset correlation coefficient threshold. If so, it is determined that the matching server does not belong to the correlation server.

[0108] Based on the duration of different historical collection calls and the deviation between collection voice data and other historical collection calls, the data correlation coefficient of user profiles of different historical collection calls stored in the matching server is determined. Based on the data correlation coefficient of user profiles of different historical collection calls stored in the matching server, the comprehensive correlation coefficient of the matching server is determined. Based on the comprehensive correlation coefficient, it is determined whether the server is the associated server of the user profile within a preset time period.

[0109] It should be noted that the above steps include the following:

[0110] When the number of historical collection calls with a data correlation coefficient within the preset data correlation coefficient range is greater than the preset call number threshold, the matching server is determined to be an associated server. When the number of historical collection calls with a data correlation coefficient within the preset data correlation coefficient range is not greater than the preset call number threshold, the comprehensive correlation coefficient is then determined.

[0111] Furthermore, when the comprehensive correlation coefficient meets the requirements, the server is determined to be the associated server of the user profile within a preset time period.

[0112] Optionally, the method for determining the associated server of the user profile within a preset time period is as follows:

[0113] Based on the storage status of historical communication data in the matching server at different time periods, the storage duration of collection voice data in the matching server within a preset time period is determined.

[0114] Based on the storage duration of collection voice data of the matching server within a preset time period, the collection duration ratio of the user's collection voice data within the preset time period is determined.

[0115] Based on the collection time percentage, determine whether the server is an associated server of the user profile within a preset time period.

[0116] Furthermore, when the collection time percentage is greater than a preset collection time threshold, the server is determined to be the associated server of the user profile within the preset time period.

[0117] Specifically, the call delay data includes the average call delay duration of the voice robot during the call processing with different associated servers.

[0118] Specifically, the method for determining the message generation and processing method of the voice assistant mailbox of the voice robot is as follows:

[0119] Using the call latency data between the voice robot and different associated servers, determine the average call latency duration of the voice robot in the call processing process with different associated servers, and determine the delayed call server among the associated servers based on the average call latency duration.

[0120] Based on the historical communication data of the delayed call server within a preset time period, determine the storage duration of the collection voice data of the delayed call server within the preset time period, and the proportion of the collection duration of the user's collection voice data within the preset time period, and use it as the collection duration proportion of the delayed call server.

[0121] The message generation and processing method of the voice robot's voice assistant mailbox is determined based on the sum of the collection time proportions of different delayed server calls.

[0122] Furthermore, the delayed call server in the associated server is an associated server whose average call delay duration is greater than a preset duration threshold.

[0123] It is understandable that the message generation and processing method for the voice assistant mailbox of the voice robot is determined based on the sum of the collection time proportions of different delay call servers, specifically including:

[0124] When the sum of the collection time percentages of different delayed call servers is greater than the preset value of the time percentage, a voice robot is used to generate messages in the voice assistant mailbox before calling the user.

[0125] When the sum of the collection time percentages of different delayed call servers is not greater than the preset value of the time percentage, then when it is determined that the message will be entered into the voice assistant mailbox, the voice robot will be used to generate the message in the voice assistant mailbox.

[0126] It should be noted that the process of generating messages for the voice assistant's mailbox using a voice robot specifically includes:

[0127] Call the collection voice data from different associated servers within a preset time period as the collection information within the preset time period;

[0128] The collection information and the user's overdue information are used as input to the voice robot. Based on the output of the voice robot, the user's personalized message data is generated and processed.

[0129] The user's personalized message data is used to process messages in the user's voice assistant mailbox.

[0130] Optionally, the method for determining the message generation and processing method of the voice assistant mailbox of the voice robot is as follows:

[0131] S41 uses the call delay data between the voice robot and different associated servers to determine the average call delay duration of the voice robot in the call processing process with different associated servers, and determines the data call delay coefficient of different associated servers based on the average call delay duration.

[0132] S42 determines the storage duration of the collection voice data of the associated server within the preset time period based on the historical communication data of the associated server within the preset time period, and determines the user profile association value of the associated server by combining the duration of the historical collection calls stored by the associated server within the preset time period and the deviation of the collection voice data from other historical collection calls.

[0133] S43 determines the latency impact coefficient of the voice robot based on the data call latency coefficient of different associated servers and the user profile association value, and determines the message generation and processing method of the voice robot's voice assistant mailbox based on the latency impact coefficient.

[0134] Optionally, the message generation and processing method for the voice assistant mailbox of the voice robot is determined based on the delay impact coefficient, specifically including:

[0135] When the delay impact coefficient is greater than the preset delay impact coefficient threshold, the voice robot is used to generate messages in the voice assistant mailbox before calling the user.

[0136] When the delay impact coefficient is not greater than the preset delay impact coefficient threshold, when it is determined that the voice assistant mailbox is entered, the voice robot is used to process the message generation of the voice assistant mailbox.

[0137] Optionally, step S41 above includes the following:

[0138] S411 uses the call delay data between the voice robot and different associated servers to determine the average call delay duration of the voice robot in the call processing process with different associated servers. Based on the average call delay duration, the data call delay coefficient of different associated servers is determined. When there is no associated server with a data call delay coefficient greater than the preset delay coefficient threshold, it is determined that when entering the voice assistant mailbox, the voice robot is used to process the message generation of the voice assistant mailbox. When there is an associated server with a data call delay coefficient greater than the preset delay coefficient threshold, proceed to step S412.

[0139] S412 If there is an associated server whose data call delay coefficient is within the preset call delay coefficient range, proceed to step S413; if there is no associated server whose data call delay coefficient is within the preset call delay coefficient range, proceed to step S414.

[0140] S413 When the number of associated servers whose data call delay coefficient is within the preset call delay coefficient range does not meet the requirements, the voice robot is used to generate a message in the voice assistant mailbox before the user is called. When the number of associated servers whose data call delay coefficient is within the preset call delay coefficient range meets the requirements, proceed to step S414.

[0141] S414 determines the call delay evaluation value based on the data call delay coefficient of different associated servers. When the call delay evaluation value is greater than the preset call delay threshold, the voice robot is used to generate a message in the voice assistant mailbox before calling the user. When the call delay evaluation value is not greater than the preset call delay threshold, proceed to step S42.

[0142] Optionally, step S42 above includes the following:

[0143] S421 Based on the historical communication data of the associated server within a preset time period, determine the storage duration of the collection voice data of the associated server within the preset time period, and designate the associated server whose data call delay coefficient is within a preset call delay coefficient range as the delay server. When the sum of the storage duration of the collection voice data of the delay server within the preset time period does not meet the requirements, the voice robot is used to generate a message in the voice assistant mailbox before calling the user. When the sum of the storage duration of the collection voice data of the delay server within the preset time period meets the requirements, proceed to step S422.

[0144] S422 determines the user profile association value of the associated server based on the storage duration of the collection voice data of the associated server within a preset time period, combined with the duration of the historical collection calls stored by the associated server within the preset time period and the deviation between the collection voice data and other historical collection calls.

[0145] S423 determines the server latency impact coefficient of different associated servers based on the product of the user profile association value of different associated servers and the data call latency coefficient. When the server latency impact coefficient of different associated servers meets the requirements, it is determined that when entering the voice assistant mailbox, the voice robot is used to generate the message in the voice assistant mailbox. When there is an associated server whose server latency impact coefficient does not meet the requirements, the process proceeds to step S424.

[0146] S424 When the number of associated servers whose server latency impact coefficient does not meet the requirements is greater than the preset number of associated servers with latency, the voice robot is used to generate a message in the voice assistant mailbox before the user is called. When the number of associated servers whose server latency impact coefficient does not meet the requirements is not greater than the preset number of associated servers with latency, proceed to step S43.

[0147] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and non-volatile computer storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0148] The foregoing has described specific embodiments of this specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in a different order than that shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily require the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0149] The above description is merely one or more embodiments of this specification and is not intended to limit this specification. Various modifications and variations can be made to the one or more embodiments of this specification by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of one or more embodiments of this specification should be included within the scope of the claims of this specification.

Claims

1. A system for an outbound call system to handle intelligent messages for a voice assistant mailbox, the system comprising: Specifically comprising: The call module, the message identification module, the switching module, the message processing module; The call module realizes the call processing of the user; The message identification module determines whether to enter the voice assistant mailbox according to the connection processing time length or the voice recognition result; The switching module automatically switches to the message processing module when it is determined to enter the voice assistant mailbox, and continues to switch to the call module to realize the call processing of other users; The message processing module generates a voice message according to the analysis result of the user information of the user; The intelligent message method specifically comprises: Determine the connection data of the user in the current period, and determine the connection probability of the user to meet the requirements when the voice recognition result of the connection call is combined, and enter the next step; Based on the user information of the user, when it is determined that there is no matching user portrait, the storage situation of the historical communication data of the user in different servers is determined based on the analysis and processing result of the historical communication data of the user; Based on the storage situation, determine the matching server in the server, and determine the associated server of the user portrait in the preset period based on the storage situation of the historical communication data of the associated server in different periods; Determine the call delay data between the voice robot and different associated servers, and determine the message generation processing mode of the voice assistant mailbox of the voice robot based on the call delay data and the historical communication data of the associated server in the preset period; The determination method of the message generation processing mode of the voice assistant mailbox of the voice robot is: Determine the average value of the call delay time length of the voice robot in the call processing process with different associated servers based on the call delay data between the voice robot and different associated servers, and determine the delay call server in the associated server based on the average value of the call delay time length; According to the historical communication data of the delay call server in the preset period, determine the storage time length of the collection voice data of the delay call server in the preset period, and the proportion of the collection time length of the user in the preset period, and take it as the collection time length proportion of the delay call server; Determine the message generation processing mode of the voice assistant mailbox of the voice robot based on the sum of the collection time length proportions of different delay call servers; When the sum of the collection time length proportions of different delay call servers is greater than the time length proportion preset value, the voice robot is used to generate the message of the voice assistant mailbox before the call processing of the user is performed; When the sum of the collection time length proportions of different delay call servers is not greater than the time length proportion preset value, the voice robot is used to generate the message of the voice assistant mailbox when it is determined to enter the voice assistant mailbox.

2. The system of claim 1, wherein the intelligent message handling system for voice assistant voicemail of an outbound call system is characterized by, The call module adopts artificial call or voice robot call.

3. The system of claim 1, wherein the intelligent message handling system for voice assistant voicemail of an outbound call system is characterized by, When the connection processing time length is greater than the preset processing time length threshold or it is determined to enter the voice assistant mailbox according to the voice recognition result, it is determined that the call of the user enters the voice assistant mailbox.

4. The system of claim 1, wherein the intelligent message handling system for voice assistant voicemail of an outbound call system is characterized by, According to the analysis result of the user information of the user, a voice message is generated, specifically comprising: According to the analysis result of the user information of the user, the user portrait of the user is determined; According to the preset voice message corresponding to the user portrait of the user, a voice message is generated.

5. The system of claim 1, wherein the intelligent message handling system for voice assistant voicemail of an outbound call system is characterized by, The call data includes the number of calls and the call rate of the time period.

6. The system of claim 1, wherein the intelligent message handling system for voice assistant voicemail of an outbound call system is characterized by, The voice recognition result of the call includes the distribution data of the keywords in the different call of the user in the time period.

7. The system of claim 1, wherein the intelligent message handling system for voice assistant voicemail of an outbound call system is characterized by, The call probability of the user is determined to meet the requirements, specifically comprising: With the call data in the current time period, the number proportion of the missed call of the time period is determined; Based on the voice recognition result of the call in the current time period, the number proportion of the call matching the rejection keyword library is determined and used as the matching call proportion; According to the sum of the number proportion of the missed call and the matching call proportion, the rejection probability of the user is determined, and whether the call probability of the user meets the requirements is determined by using the rejection probability.

8. The system of claim 7, wherein the intelligent message of the voice assistant mailbox is handled by the outbound call system. When the call probability of the user does not meet the requirements, the voice message generation processing of the voice assistant mailbox is performed by using the voice robot before the call processing of the user.

Images