Voice interaction method, device, computer equipment and storage medium

Abstract

The present application discloses a voice interaction method, which includes: obtaining a call number according to an outbound call task; making an outbound call to the call number; if it is determined that the outbound call is answered, obtaining a task identifier of the outbound call task; initializing an intelligent agent according to configuration information corresponding to the task identifier to obtain a target intelligent agent; and performing voice interaction with a user who answers the outbound call through the target intelligent agent. The present application dynamically initializes an intelligent voice robot using configuration information, so that the intelligent voice robot can automatically configure service functions according to business needs, thereby improving the efficiency of voice interaction.

Description

Technical Field

[0001] The present application relates to the field of intelligent call technology, and more specifically, to a voice interaction method, apparatus, computer equipment and storage medium. Background Art

[0002] Intelligent voice robots can communicate with users in a very natural voice, so they can be seamlessly integrated into the enterprise customer service seat system, especially integrated with the existing customer service seat system, replacing or assisting human seats to complete data query, order processing, business handling and other operations. Intelligent voice robots can be widely used in various industries such as finance, telecommunications, e-commerce, and public utilities, guiding customers to use more self-service human-computer interaction to meet their needs, and improving the customer service quality and efficiency of enterprises.

[0003] In real application scenarios, as the types of business services provided by intelligent voice robots are constantly updated, the service functions of intelligent voice robots often need to be improved at the same time. However, due to the large differences in technical implementation of service functions of different business types, professional technicians currently only develop different versions of intelligent voice robots for different businesses. However, developing different versions of robots requires a lot of time and cost and cannot respond to business changes in a timely manner. Summary of the invention

[0004] The embodiments of the present application provide a voice interaction method, apparatus, computer equipment and storage medium, aiming to improve the efficiency and flexibility of intelligent voice robots in the voice interaction process.

[0005] In a first aspect, some embodiments of the present application provide a voice interaction method, which includes: obtaining a call number according to an outbound call task; making an outbound call to the call number; if it is determined that the outbound call is answered, obtaining a task identifier of the outbound call task; initializing an intelligent agent according to configuration information corresponding to the task identifier to obtain a target intelligent agent; and performing voice interaction with a user who answers the outbound call through the target intelligent agent.

[0006] On the second aspect, some embodiments of the present application also provide a voice interaction device, which includes: a number acquisition module, used to obtain a call number according to an outbound call task; a customer call module, used to make an outbound call to the call number; an identification acquisition module, used to obtain the task identification of the outbound call task if it is determined that the outbound call is answered; an initialization module, used to initialize the smart agent according to the configuration information corresponding to the task identification to obtain the target smart agent; a voice interaction module, used to perform voice interaction with the user who answers the outbound call through the target smart agent.

[0007] In a third aspect, some embodiments of the present application further provide a computer device, which includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are called by the processor, the video processing method provided in the first aspect is executed.

[0008] In a fourth aspect, some embodiments of the present application further provide a computer-readable storage medium, which stores a program code, wherein when the program code is executed by a processor, the voice interaction method provided in the first aspect is executed.

[0009] In a fifth aspect, some embodiments of the present application further provide a computer program product or a computer program, which includes computer instructions stored in a storage medium. A processor of a computer device reads the computer instructions from the storage medium, and the processor executes the computer instructions, so that the computer performs the steps in the above-mentioned voice interaction method.

[0010] It can be seen that in the embodiment of the present application, the call number can be obtained according to the outbound call task, and an outbound call can be made to the call number. If it is determined that the outbound call is answered, the task identifier of the outbound call task is obtained, and then the intelligent agent is initialized according to the configuration information corresponding to the task identifier to obtain the target intelligent agent. The intelligent voice robot is dynamically initialized through the configuration information, so that the intelligent voice robot can automatically switch to the corresponding service function according to business needs to interact with the customer by voice, and then interact with the user who answers the outbound call by voice through the target intelligent agent, thereby ensuring the continuity of the business service process and improving the efficiency and flexibility of voice interaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A schematic diagram of an application scenario of voice interaction provided in an embodiment of the present application is shown.

[0013] Figure 2 A flow chart of a voice interaction method provided in an embodiment of the present application is shown.

[0014] Figure 3 A schematic diagram showing a framework of a voice interaction system provided in an embodiment of the present application

[0015] Figure 4A flow chart of another voice interaction method provided in an embodiment of the present application is shown.

[0016] Figure 5 A module block diagram of a voice interaction device provided in an embodiment of the present application is shown.

[0017] Figure 6 It is a module block diagram of a computer device provided in an embodiment of the present application.

[0018] Figure 7 It is a module block diagram of a computer-readable storage medium provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] The embodiments of the present application are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present application, and cannot be understood as limiting the present application.

[0020] In order to enable those skilled in the art to better understand the solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present application.

[0021] With the development of intelligent voice robot technology, many companies currently use the voice interaction function of intelligent voice robots in call center systems to provide services to customers. Especially in the financial industry, especially in intelligent collection, intelligent return visits, intelligent telemarketing and other businesses, intelligent voice robots play a pivotal role. The intelligent voice robot (intelligent seat) of the call center system plays a role similar to that of a human agent. For example, it replaces the human agent to perform outbound call tasks and conduct business voice interaction with customers, so it can help the human agent share the work pressure to a great extent, thereby improving the operational efficiency of the call center and reducing operating costs.

[0022] In existing scenarios where intelligent voice robots are used, different types of businesses have different requirements for the functions of intelligent voice robots. Therefore, R&D personnel will conduct technical evaluations based on business needs, and then analyze the service functions that intelligent voice robots that serve a certain business need to have based on the results of the technical evaluation, and then complete a series of robot design, development, and testing processes, and finally release them online. This way of developing intelligent robots requires technical personnel to conduct relevant technical evaluations again and redevelop the changed technical points when business personnel propose changes to the functional requirements of intelligent voice robots. This also leads to high development costs for intelligent voice robots, extremely poor performance compatibility, and inability to cope with rapid changes in business.

[0023] In order to solve the above problems, the inventors have proposed a voice interaction method provided in an embodiment of the present application after long-term research. This method can obtain a call number according to an outbound call task, and make an outbound call request to the customer corresponding to the call number. If it is determined that the customer answers the outbound call request, the task identifier of the outbound call task is obtained, and then the smart agent is initialized according to the configuration information corresponding to the task identifier to obtain the target smart agent. Thus, the intelligent voice robot is dynamically initialized using specific configuration information, so that the intelligent voice robot can automatically configure service functions according to business needs, eliminating the need for redevelopment and reducing development costs. At the same time, the intelligent voice robot can be applied to the task execution of multiple businesses, thereby improving performance compatibility.

[0024] The following is an introduction to the application scenarios of the voice interaction method involved in this application.

[0025] like Figure 1 As shown, for example, the voice interaction method provided in the embodiment of the present application can be applied in Figure 1 In the application scenario of voice interaction shown, the scenario includes a service provider (system administrator) 410, a voice interaction system 420, a user 430, and a server 440. The voice interaction system 420 is in communication connection with the server 440. The voice interaction system 420 includes an intelligent voice robot for voice interaction, and the voice interaction system 420 can be integrated and installed in a computing device. The server 440 can be a file transfer protocol server (File Transfer Protocol Server, FTP) which can provide file storage and access services on the Internet.

[0026] In some embodiments, the service provider 410 can create a corresponding call task in the voice interaction system 420 according to business needs, so that the voice interaction system 420 can execute the call task to make an outbound call request to the user 430. When the user 430 answers the outbound call request, the voice interaction system 420 can initialize the intelligent voice robot that executes the call task, and use the initialized intelligent voice robot to communicate business with the user 430.

[0027] Specifically, when the voice interaction system 420 initializes the intelligent voice robot, the configuration information used includes at least text-to-speech configuration information and voice recognition configuration information. The text-to-speech configuration information is the configuration information of the text-to-speech technology (Text To Speech, TTS). TTS is a type of speech synthesis application that can convert files or web pages stored in a computer into natural speech output. TTS applications include voice-driven emails and voice-sensitive systems, and are often used together with voice recognition programs.

[0028] The speech recognition configuration information is the configuration information of the Automatic Speech Recognition (ASR) technology, which is a technology that converts human speech into text. ASR technology enables computers to "dictate" the continuous speech spoken by different people, which is commonly known as "speech dictation machine". It is a technology that realizes the conversion of "sound" to "text". The application of ASR technology in outbound call tasks can be keyword detection, such as identifying some professional terms in business, and can also be applied to automatic extension connection services.

[0029] The various embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0030] See also Figure 2 , Figure 2 A voice interaction method provided in an embodiment of the present application is shown, and the voice interaction method may include steps S110 to S150.

[0031] Step S110: Obtain a calling number according to the outbound call task.

[0032] Among them, outbound calling tasks refer to making calls to users for different businesses, for example, outbound calling tasks such as smart collection, smart return visits, and smart telemarketing related to consumer finance companies.

[0033] As an implementation mode, the step of obtaining a calling number according to an outbound call task may include:

[0034] (1) Obtain a call list according to an outbound call task, where the call list includes multiple call numbers.

[0035] (2) Obtaining a number of calling numbers corresponding to a preset number threshold from the calling list.

[0036] The call list refers to the phone list that matches the outbound call task. Each call list includes a large number of call numbers. In order to ensure the execution quality of the outbound call task, a certain number of call numbers corresponding to the preset number threshold can be obtained from the call list. The preset number threshold refers to the maximum number of call numbers for outbound calls.

[0037] For example, the outbound call task corresponding to the collection business has a collection phone list. When the outbound call task corresponding to the collection business is executed, the collection phone list can be obtained. Further, five to six call numbers can be obtained from the collection phone list.

[0038] Step S120: Make an outbound call to the calling number.

[0039] Among them, outbound calling refers to dialing a number of a user associated with the calling number. In some embodiments, the step of making an outbound call to the calling number may include:

[0040] (1) Determine whether the current time matches the preset outbound call time threshold.

[0041] (2) If the current time matches the preset outbound call time threshold, an outbound call is made to the user corresponding to the calling number.

[0042] The outbound call time threshold refers to the time specified by the voice interaction system for the user to make an outbound call request. Taking into account the user's daily routine and the service provider's business strategy requirements, in order to improve the efficiency of business services, an outbound call request can be made at a preset outbound call time. The preset outbound call time threshold can be a specified time period, for example, from 10 a.m. to 11 a.m., based on a day as a standard. Since the user's work tasks are usually handled during this time period and they are in a relatively idle state, when an outbound call request is made during this time period, the probability of the user answering the request is high.

[0043] Step S130: If it is determined that the outbound call is answered, the task identifier of the outbound call task is obtained.

[0044] In the embodiments of the present application, the task identifier refers to a mark used to identify the task, such as the task name, task ID (Identity Document), and other marks that can uniquely identify the outbound call task. The task identifier can be used to obtain the configuration information required for the execution of the task corresponding to the task identifier. As an implementation method, when making an outbound call to the user associated with the call number, if it is determined that the outbound call is answered by the user, the task identifier of the outbound call task can be obtained. As another implementation method, when making an outbound call to the customer corresponding to the call number, if it is determined that the user does not answer the outbound call, it is possible to wait for a period of time and then call the user again.

[0045] Step S140: Initialize the intelligent agent according to the configuration information corresponding to the task identifier to obtain a target intelligent agent.

[0046] Considering that each type of business has different business requirements, the intelligent agent (intelligent voice robot) providing services for the business needs to set specific service functions for different businesses. In the embodiment of the present application, by performing specific initialization on the intelligent agents under different outbound call tasks according to the configuration information, the intelligent agents can automatically switch service functions according to business requirements, thereby improving service efficiency.

[0047] The initialization refers to the program initialization of the intelligent agent used by the outbound call task by obtaining the configuration information corresponding to the different outbound call tasks (such as TTS configuration information, ASR configuration information, etc.). The target intelligent agent refers to the intelligent agent used to execute the outbound call task after the initialization.

[0048] In some embodiments, the step of initializing the intelligent agent according to the configuration information corresponding to the task identifier to obtain the target intelligent agent may include:

[0049] (1) Obtain configuration information based on the task identifier.

[0050] As an implementation mode, the step of acquiring configuration information according to the task identifier may include:

[0051] (1.1) The task identifier is used as an input parameter of the hypertext transfer protocol interface call.

[0052] (1.2) Call the input parameters through the hypertext transfer protocol interface to perform task query and obtain the query results.

[0053] (1.3) Obtain the configuration information corresponding to the task ID based on the query results.

[0054] Among them, the configuration information may include text-to-speech configuration information and speech recognition configuration information. The Hyper Text Transfer Protocol (HTTP) is a simple request-response protocol that specifies the request message that the voice interaction system sends to the server to obtain configuration information. After the voice interaction system obtains the task identifier, the task identifier can be used as an input parameter of the HTTP interface call. Further, the configuration information corresponding to the task identifier can be queried from the server that pre-stores the configuration information of each outbound call task through the HTTP interface call input parameter. Furthermore, after the required configuration information is queried, HTTP can return the query result, that is, the configuration information.

[0055] (2) Initialize the intelligent agent according to the configuration information to obtain the target intelligent agent.

[0056] The initialization may be to set the configuration of the service function of the intelligent agent. For example, in the outbound call task of intelligent telemarketing, the TTS configuration information and ASR configuration information of the intelligent agent may include:

[0057] TTS configuration information:

[0058] Synthesis method ();

[0059] protocol();

[0060] Pronouncer();

[0061] …

[0062] ASR configuration information:

[0063] timeout_duration();

[0064] SilenceDuration();

[0065] …

[0066] As an implementation method, when the voice interaction system obtains configuration parameters through HTTP, the TTS configuration information and ASR configuration information of the intelligent agent can be initialized:

[0067] TTS configuration information:

[0068] Synthesis method (splicing);

[0069] Protocol (MRCP);

[0070] Speaker (speaker_1);

[0071] …

[0072] ASR configuration information:

[0073] Timeout duration (5s);

[0074] Silence duration (3s);

[0075] …

[0076] At this point, the voice interaction system has completed the initialization of the smart agent, and can then use the initialized smart agent (target smart agent) to perform the corresponding outbound call task. It should be noted that the configuration information that needs to be initialized in the smart agent includes not only TTS configuration information and ASR configuration information, but also other configuration information that can realize the smart agent service function. The embodiment of this application only provides TTS and ASR examples. Other configuration information of the smart agent can be initialized through the same steps, which is not limited here.

[0077] In order to improve the service quality, for example, considering that the user portrait of the user in the consumer finance industry can often directly reflect the user's service demands, the user portrait can be used in the process of initializing the intelligent seat, so that the intelligent seat can provide targeted services to the user.

[0078] In other embodiments, the user portrait of the user corresponding to the call number can be obtained according to the call number. Optionally, in the application scenario of intelligent collection, the user portrait of the user can be extracted from the user's basic data and collection data, wherein the user portrait can be extracted using data mining technology. Furthermore, according to the configuration information corresponding to the task identifier and the user portrait, the intelligent agent is initialized to obtain the target intelligent agent.

[0079] Specifically, voice data from voice interactions between smart agents and users can be acquired, and text information of the voice interactions can be obtained using ASR. Then, sentiment analysis in natural language processing (NLP) technology can be used to analyze the user's emotions and obtain changes in the user's emotions. Furthermore, the smart agent can be initialized based on the changes in the user's emotions.

[0080] For example, after obtaining the user's negative emotions (Negative) from the text information of voice interaction through the Word2Vec word vector model, when the voice output module of the intelligent agent is initialized, the configuration information for controlling the voice output speed is adjusted to control the voice output speed of the intelligent agent to be moderate. In this way, the initialized target intelligent agent can provide the user with an adaptive voice interaction service to alleviate the user's negative emotions.

[0081] Step S150: Perform voice interaction with the user who answers the outbound call through the target intelligent agent.

[0082] As an implementation method, the initialized target intelligent agent can perform voice interaction with the user according to the dialogue process of the outbound call task. The dialogue process can be drawn by the process engine based on the needs of the outbound call task. During work, the target intelligent agent can continuously interact with the process engine through HTTP requests and return the results of each ASR recognition to the process engine, for example:

[0083] Smart Agent: Are you XXX?

[0084] User: Yes

[0085] When ASR recognizes that the user's answer is "yes", the intelligent agent returns the recognition result to the process engine. The process engine determines that the intelligent agent's response is an "affirmative answer" logic, and then returns the data required for the next conversation (such as: conversation type, broadcast text, etc.), and repeats the cycle to complete multiple conversations.

[0086] As can be seen from the above, the embodiment of the present application can construct different tasks according to different business needs by obtaining preset task parameters and determining the outbound call task according to the preset task parameters. Furthermore, the call number is obtained according to the outbound call task, and an outbound call is made to the user corresponding to the call number. If it is determined that the user answers the outbound call, the task identifier of the outbound call task is obtained, and then the intelligent agent is initialized according to the configuration information corresponding to the task identifier to obtain the target intelligent agent. Thus, the intelligent voice robot is dynamically initialized through the configuration information, so that the intelligent voice robot can automatically switch to the corresponding service function according to business needs to interact with the user by voice.

[0087] In combination with the method described in the above embodiment, the method will be further described in detail with examples below.

[0088] In this embodiment, the voice interaction system is specifically integrated and installed in a computer device as an example for description.

[0089] Please also read Figure 3 and Figure 4 , Figure 3 A schematic diagram of the framework of a voice interaction system provided in an embodiment of the present application. Figure 4 A flow chart of another voice interaction method provided in an embodiment of the present application. Figure 3 A voice interaction system 200 is shown, and the voice interaction system 200 includes a task management module 210, an outbound call module 230, and an intelligent agent (intelligent voice robot) 250. The voice interaction system 200 can be connected to an FTP server.

[0090] Step S310: The computer device obtains preset task parameters and determines an outbound call task according to the preset task parameters.

[0091] In actual voice interaction application scenarios, different tasks need to be constructed according to different business needs. And the execution of each task needs to be based on specific preset task parameters. In the embodiment of the present application, the preset task parameter refers to an adjustable parameter that needs to be constrained during the execution of a specific task by the voice interaction system.

[0092] As an implementation method, the voice interaction system can obtain the task parameters as preset task parameters when the service provider sets specific task parameters according to business needs. For example, when the call center of a consumer finance company uses the voice interaction system to make outbound calls, the preset task parameters required in the process of building the outbound call task may include dialing rounds, outbound call time, user level, whether it is a familiar customer, etc. The preset task parameters are specifically pre-set by the service provider according to different application scenarios and are not limited here.

[0093] The computer device is a device with an integrated voice interaction system, and has the functions of data storage and data calculation. The service provider needs to use the computer device to set preset task parameters to build different outbound call tasks according to different businesses. After the service provider sets the preset task parameters, the computer device can obtain the preset task parameters and then build the outbound call task according to the preset task parameters.

[0094] For example, the call center of a consumer finance company can set preset task parameters: outbound call time, dialing rounds, whether the familiar customer mode switch is turned on, whether the user level switch is turned on, etc., and then the computer equipment builds a specific outbound call task based on the above preset task parameters. Considering the reduction of the complexity of the voice interaction system, specific preset task parameters can be set to execute differentiated outbound call tasks.

[0095] As an implementation method, the computer device can determine the outbound call task for a familiar customer based on the familiar customer mode parameter. Specifically, the familiar customer mode parameter is used to determine whether the outbound call task has the familiar customer mode turned on. When the familiar customer mode parameter is "1", it can indicate that it is turned on, and when it is "0", it can indicate that it is not turned on. When creating an outbound call task, if the familiar customer mode parameter is "1", the voice interaction system can create an outbound call task for a familiar customer with the familiar customer mode, so that when transferring the user to manual service, the user can be matched with a seat that provides long-term service.

[0096] It should be noted that the service provider needs to configure and store the configuration information required by the intelligent agent corresponding to each outbound call task in advance when establishing an outbound call task. In order to enable the business party (user) to flexibly use various types of intelligent voice robots and dynamically switch intelligent voice robots. The service provider can maintain all TTS and ASR configuration information supported by the platform (voice interaction system) to the back-end database, so that when creating a task, the corresponding configuration information can be quickly selected to initialize the intelligent voice robot and ensure the correctness of the linkage relationship between related configuration items. At the same time, TTS configuration information, ASR configuration and grammar files used by ASR need to be placed in a specific directory of the computer device running the intelligent voice robot so as to realize the docking of TTS and ASR supported by all platforms in the intelligent voice robot.

[0097] For example, two outbound call tasks are created at the same time: intelligent telemarketing (ai_task1) and intelligent return call (ai_task2). Intelligent telemarketing is an intelligent outbound call task for telemarketing, and intelligent return call is an intelligent return call task for return call. When creating a task, the operator configures TTS and ASR as needed.

[0098] The configuration order can be:

[0099] TTS: Manufacturer → Protocol → Mode → Number of concurrent users → Speaker → Volume - Speed.

[0100] Specifically, after selecting a manufacturer, the protocols supported by the TTS of the manufacturer are listed; after selecting a protocol, the supported synthesis modes and the maximum number of concurrent connections based on the protocol are listed; after selecting a synthesis mode and setting the number of concurrent connections, the optional speakers based on the synthesis mode are listed; after selecting a speaker, the volume range and speech speed range can be set to automatic and automatic, and further, the TTS can be configured.

[0101] ASR: Manufacturer → Concurrency → Interruption → Timeout → Silence duration.

[0102] Specifically, after selecting a manufacturer, the modes and maximum concurrency numbers supported by ASR under the manufacturer are listed; after selecting a mode (interruption or no interruption allowed) and setting the concurrency number, set the timeout duration and the silence duration, and finally complete the ASR configuration.

[0103] The above TTS configuration information and ASR configuration information are stored in the task management module of the voice interaction system, which is uniformly managed by the business administrator and has the right to modify it. If the newly created outbound call task uses the same TTS and ASR technical requirements for the return call, you only need to modify the outbound call task configuration in the task management module. The intelligent voice robot can be developed using the Lua scripting language.

[0104] It should be noted that the number of concurrent calls of TTS and ASR is not equal to the number of numbers sent per unit time (number of outbound numbers), because a call often involves multiple rounds of conversations. The number of numbers sent is somewhat related to the number of concurrent calls. The number of numbers sent per unit time also depends on server resources and the average number of conversation rounds.

[0105] Step S320: the computer device obtains a call list according to the outbound call task and obtains a number of call numbers corresponding to a preset number threshold from the call list.

[0106] The computer device can retrieve the call list corresponding to the outbound call task from the database storing the call list through the application programming interface (API), and obtain a certain number of call numbers from the call list according to a preset number threshold, for example, 5 to 10 call numbers. The database can be deployed on an FTP server or locally, which is not limited here.

[0107] Step S330: The computer device makes an outbound call to the calling number.

[0108] In some embodiments, the step of the computer device making an outbound call to a user corresponding to the calling number may include:

[0109] (1) The computer device determines whether the current time matches the preset outbound call time threshold.

[0110] (2) If the computer device determines that the current time matches the preset outbound call time threshold, an outbound call is made to the user corresponding to the calling number.

[0111] Among them, before making an outbound call, the computer device can determine whether the current time matches the preset outbound call time threshold. If the current time matches the preset outbound call time threshold, the number is sent, that is, the call number is transmitted to the outbound call module of the voice interaction system, and then the outbound call module dials the call number, which is an outbound call.

[0112] Specifically, when the outbound call module receives an outbound call number and reaches the preset outbound call time threshold, the outbound call number can be dialed by calling the freeswit ESL originate command. The specific usage of the command is as follows:

[0113] originate:

[0114] -USAGE:<call url> <exten>|&<application_name>(<app_args>)[ <dialplan> ][ <context>][<cid_name> ][<cid_num> ][<timeout_sec> ]

[0115] When the outbound call module uses the originate command, it puts the task name into the call string (<call url> ) in the task name, which adds the task name to the call channel variable.

[0116] For example, in the intelligent telemarketing task, the outbound call command is as follows:

[0117] originate{...,task_name="ai_task1"}AI_robot XML default

[0118] Note: "..." is other unlisted call information in the form of key=value, such as:

[0119] sip_from_display = '66661234' is used to set the incoming call display number.

[0120] Step S340: If the computer device determines that the outbound call is answered, it obtains the task identifier of the outbound call task.

[0121] In the embodiment of the present application, the task identifier may be the task name of the outbound call task, for example, intelligent telemarketing: ai_task1, intelligent return call: ai_task2.

[0122] Step S350: The computer device initializes the intelligent agent according to the configuration information corresponding to the task identifier to obtain the target intelligent agent.

[0123] In some embodiments, after the user answers the outbound call and obtains the task identifier, the computer device can obtain configuration information according to the task identifier, and then initialize the smart agent based on the configuration information to obtain the target smart agent. Specifically, the computer device uses the task identifier as an input parameter of the hypertext transfer protocol interface call, and performs a task query through the hypertext transfer protocol interface call input parameter to obtain the query result, and then obtains the configuration information corresponding to the task identifier according to the query result.

[0124] For example, after the user answers the call, the computer device connects the user to the intelligent voice robot AI_robot. In the intelligent voice robot, the task name is taken from the call channel variable through the session:getVariable("task_name") method provided by lua, and used as the interface call input parameter to query the task management module through the HTTP interface. If the task management module queries the configuration information corresponding to task_name, the task management module can return the configuration information. For example:

[0125] If the input parameter is ai_task1, the task management module returns the following information:

[0126] 1. TTS: self-developed, concurrent (50), splicing, MRCP protocol, speaker (speaker_1), speaking speed (-200), volume (100);

[0127] 2. ASR: self-developed, concurrency (50), does not support interruption, timeout duration 5S, silent duration 3S;

[0128] If the input parameter is ai_task2, the task management module returns the following information:

[0129] 1. TTS: self-developed, concurrent (50), end-to-end, MRCP protocol, speaker (speaker_1), speaking speed (-200), volume (100);

[0130] 2. ASR: A certain three-party ASR, concurrent (50), does not support interruption, timeout duration 5S, silent duration 3S;

[0131] The main procedures are as follows:

[0132]

[0133]

[0134] Furthermore, the computer device can initialize the intelligent voice robot based on the acquired configuration information to obtain the target intelligent agent.

[0135] As another implementation method: obtaining a user portrait corresponding to the user according to the calling number, initializing the smart agent to obtain a target smart agent according to the configuration information and the user portrait corresponding to the task identifier. For example, in the application scenario of intelligent debt collection, the computer device can extract the user portrait of the user from the user's basic data and debt collection data using a deep learning (DeepLearning) / machine learning (Machine Learning) algorithm. Furthermore, initializing the smart agent to obtain a target smart agent according to the configuration information and the user portrait corresponding to the task identifier.

[0136] Step S360: The computer device performs voice interaction with the user who answers the outbound call through the target intelligent agent.

[0137] As an implementation method, the intelligent voice robot after the computer device is initialized can perform voice interaction with the user according to the dialogue process of the outbound call task. The dialogue process can be drawn by the process engine based on the needs of the outbound call task. The intelligent voice robot can continuously interact with the process engine through HTTP requests and return the results of each ASR recognition to the process engine.

[0138] In order to reduce the complexity of the voice interaction system, the voice interaction method provided in the embodiment of the present application can construct a specific outbound call task according to preset task parameters. Among them, the call history information refers to the call information stored by the service provider during the execution of the outbound call task, and the call history information may include the seat information, service type, service time, etc. that provide services to users.

[0139] As an implementation mode, if the familiar customer mode is turned on when the computer device creates an outbound call task, the familiar customer mode outbound call task can be executed. Specifically, it can include step S370 and step S380.

[0140] Step S370: The computer device obtains call history information according to the outbound call task for regular customers, and determines the optimal service agent for the user according to the call history information.

[0141] Step S380: If the user requests manual service, the computer device transfers the target intelligent agent to the optimal service agent.

[0142] In some embodiments, if the familiar customer mode is turned on when creating an outbound call task, the intelligent agent can query the call history information through the HTTP interface (input parameter: user mobile phone number or user ID) to filter out the manual agent who has served the user most recently, that is, the best service agent. For example, when a user initiates a transfer to manual service during a call with an intelligent agent or the intelligent agent automatically transfers to manual service, the target intelligent agent can transfer to the best service agent. Optionally, if the best service agent is in a call or is not available for call, the user is placed in the skill group queue to which the best service agent belongs. If the best service agent is not found, the user is placed in the corresponding skill group queue according to the transfer node business type or the user's transfer intention. Thus, by executing a specific outbound call task in the familiar customer mode, the steps in the entire process of interaction between the intelligent agent and the user have rapid continuity and accurate response.

[0143] As another implementation, the computer device can determine an outbound call task that performs outbound calls according to the customer level parameters. Specifically, if the computer device distinguishes user levels when constructing outbound call tasks, when a user transfers to a human agent through an intelligent voice robot, the intelligent voice robot can query the call history information before initiating a transfer instruction. Further, the user level is queried through the HTTP port method (input parameter: user mobile phone number or user ID). After the intelligent voice machine obtains the user level information, the user level information can be carried in the transfer instruction.

[0144] If the user is assigned to the skill group queue, the call distribution module will interrupt the queue according to the user level. The queue interruption rule can be: queue in order of priority. For high-level users (configuration determines the level value), if the phone is hung up during the queuing process (for example, the user actively hangs up during the waiting process), the user will be added to the reconnection list.

[0145] Considering that outbound call tasks are generally configured as multiple dialing rounds, after a round of dialing is completed, intelligent number washing will be performed before the next round of outbound calls, so that the numbers that have been connected, suspended or turned off will be washed out. Therefore, the list of numbers that have been washed out plus the reconnection list (the reconnection list belongs to the connected users and will be washed out during the number washing) are used as the new round of outbound call list.

[0146] As another implementation method, after the computer device makes a call request other than the first round, if it is determined that the user receives an outbound call and enters the main thread of the intelligent voice robot, the intelligent voice robot can determine whether the user belongs to the reconnection list. When the user belongs to the reconnection list, the intelligent voice robot can directly initiate a transfer to manual service, and the user does not need to have unnecessary "conversations" with the robot, thereby improving the execution efficiency of the outbound call task.

[0147] As can be seen from the above, the embodiment of the present application can obtain preset task parameters through a computer device, and determine an outbound call task according to the preset task parameters, so that different tasks can be constructed according to different business needs. Further, the call number is obtained according to the outbound call task, and an outbound call is made to the user corresponding to the call number. If it is determined that the user answers the outbound call, the task identifier of the outbound call task is obtained, and then the intelligent agent is initialized according to the configuration information corresponding to the task identifier to obtain the target intelligent agent. Thus, the intelligent voice robot is dynamically initialized through the configuration information, so that the intelligent voice robot can automatically switch to the corresponding service function according to business needs to interact with the user by voice.

[0148] See also Figure 5 , which shows a structural block diagram of a voice interaction device 500 provided in an embodiment of the present application. The voice interaction device 500 includes: a number acquisition module 510, which is used to acquire a call number according to an outbound call task; a customer call module 520, which is used to make an outbound call to the call number; an identification acquisition module 530, which is used to acquire the task identification of the outbound call task if it is determined that the outbound call is answered; an initialization module 540, which is used to initialize the smart agent according to the configuration information corresponding to the task identification to obtain the target smart agent; and a voice interaction module 550, which is used to perform voice interaction with the user who answers the outbound call through the target smart agent.

[0149] In some embodiments, the initialization module 540 may include: a configuration information acquisition unit, used to acquire configuration information according to the task identifier, the configuration information including text-to-speech configuration information and speech recognition configuration information; a first initialization unit, used to initialize the smart agent according to the configuration information to obtain a target smart agent.

[0150] In some embodiments, the configuration information acquisition unit can be specifically used to use the task identifier as an input parameter of a hypertext transfer protocol interface call; perform a task query through the hypertext transfer protocol interface call input parameter to obtain a query result; and obtain configuration information corresponding to the task identifier based on the query result.

[0151] In some embodiments, the initialization module 540 may further include: a user portrait acquisition unit, used to acquire a user portrait of the user answering the outbound call; a second initialization unit, used to initialize the smart agent to obtain a target smart agent according to the configuration information corresponding to the task identifier and the user portrait.

[0152] In some embodiments, the voice interaction device 500 may include: a parameter acquisition module, used to acquire preset task parameters; and a task determination module, used to determine the outbound call task according to the preset task parameters.

[0153] In some embodiments, the preset task parameters include a familiar customer mode parameter, and the task determination module may be specifically used to determine a familiar customer outbound call task according to the familiar customer mode parameter;

[0154] The voice interaction device 500 can also be specifically used to obtain call history information based on the outbound call tasks of regular customers; determine the best service seat for the user based on the call history information; if the user requests manual service, transfer the target intelligent seat to the best service seat.

[0155] In some embodiments, the customer call module 520 can be specifically used to determine whether the current time matches the preset outbound call time threshold; if the current time matches the preset outbound call time threshold, an outbound call request is made to the customer corresponding to the calling number.

[0156] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the above-described device and module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In several embodiments provided in this application, the coupling between modules can be electrical, mechanical or other forms of coupling.

[0157] In addition, each functional module in each embodiment of the present application can be integrated into a processing module, or each module can exist physically separately, or two or more modules can be integrated into one module. The above integrated modules can be implemented in the form of hardware or software functional modules.

[0158] The solution provided by the present application can construct different tasks according to different business needs by obtaining preset task parameters and determining the outbound call task according to the preset task parameters. Furthermore, the call number is obtained according to the outbound call task, and an outbound call is made to the call number. If it is determined that the customer answers the outbound call, the task identifier of the outbound call task is obtained, and then the intelligent agent is initialized according to the configuration information corresponding to the task identifier to obtain the target intelligent agent. Thus, the intelligent voice robot is dynamically initialized through the configuration information, so that the intelligent voice robot can automatically switch to the corresponding service function according to business needs to interact with the customer by voice.

[0159] like Figure 6 As shown, the embodiment of the present application also provides a computer device 600, which includes a processor 610, a memory 620, a power supply 630 and an input unit 640. The memory 620 stores computer program instructions. When the computer program instructions are called by the processor 610, the various method steps provided in the above embodiments can be implemented. Those skilled in the art can understand that the structure of the computer device shown in the figure does not constitute a limitation on the computer device, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently. Among them:

[0160] The processor 610 may include one or more processing cores. The processor 610 uses various interfaces and lines to connect various parts of the entire battery management system, and calls the data stored in the memory 620 by running or executing instructions, programs, code sets or instruction sets stored in the memory 620, and executes various functions and processes data of the battery management system, as well as various functions and processes data of the computer device, so as to monitor the computer device as a whole. Optionally, the processor 610 can be implemented in at least one hardware form of digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). The processor 610 can integrate one or a combination of a central processing unit 610 (CPU), a graphics processing unit 610 (GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 610, but may be implemented separately through a communication chip.

[0161] The memory 620 may include a random access memory 620 (Random Access Memory, RAM), and may also include a read-only memory 620 (Read-Only Memory). The memory 620 may be used to store instructions, programs, codes, code sets or instruction sets. The memory 620 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the various method embodiments described below, etc. The data storage area may also store data (such as a phone book, audio and video data, chat record data) created by the computer device during use. Accordingly, the memory 620 may also include a memory controller to provide the processor 610 with access to the memory 620.

[0162] The power supply 630 can be logically connected to the processor 610 through a power management system, so as to manage charging, discharging, power consumption and other functions through the power management system. The power supply 630 can also include one or more DC or AC power supplies, recharging systems, power failure detection circuits, power converters or inverters, power status indicators and other arbitrary components.

[0163] The input unit 640 can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal input related to user settings and function control.

[0164] Although not shown, the computer device 600 may also include a display unit, etc., which will not be described in detail herein. Specifically in this embodiment, the processor 610 in the computer device will load the executable files corresponding to the processes of one or more application programs into the memory 620 according to the following instructions, and the processor 610 will run the application programs stored in the memory 620, thereby implementing the various method steps provided in the aforementioned embodiments.

[0165] like Figure 7 As shown, an embodiment of the present application further provides a computer-readable storage medium 700, in which computer program instructions 710 are stored. The computer program instructions 710 can be called by a processor to execute the method described in the above embodiment.

[0166] The computer readable storage medium can be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk, or a ROM. Optionally, the computer readable storage medium includes a non-volatile computer readable storage medium (Non-Transitory Computer-Readable Storage Medium). The computer readable storage medium 700 has a storage space for program codes that execute any method steps in the above method. These program codes can be read from or written to one or more computer program products. The program code can be compressed, for example, in an appropriate form.

[0167] According to one aspect of the present application, a computer program product or a computer program is provided, the computer program product or the computer program includes computer instructions, the computer instructions are stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the method provided in various optional implementations provided in the above embodiments.

[0168] The above are only preferred embodiments of the present application, and are not intended to limit the present application in any form. Although the present application has been disclosed as above with preferred embodiments, it is not intended to limit the present application. Any technical personnel in the field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present application. However, any brief modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present application without departing from the content of the technical solution of the present application are still within the scope of the technical solution of the present application.< / context> < / dialplan> < / exten>

Claims

1. A voice interaction method, characterized in that: The method comprises: Obtaining preset task parameters, and determining the outbound call task according to the preset task parameters; wherein the preset task parameters include: dialing round, user level, and whether the user is a familiar customer; Get the calling number according to the outbound calling task; Making an outbound call to the calling number; If it is determined that the outbound call is answered, obtaining the task identifier of the outbound call task; The configuration of the service function of the intelligent agent is initialized according to the configuration information corresponding to the task identifier and the user portrait of the outbound user, so as to obtain the target intelligent agent; wherein the initialization is to set the configuration of the service function of the intelligent agent; and the configuration information is used to realize the service function of the intelligent agent; Voice interaction is performed through the target intelligent agent with the user who answers the outbound call.

2. The method according to claim 1, characterized in that Initializing the intelligent agent according to the configuration information corresponding to the task identifier to obtain the target intelligent agent includes: Acquire configuration information according to the task identifier, the configuration information including text-to-speech configuration information and speech recognition configuration information; The intelligent agent is initialized according to the configuration information to obtain a target intelligent agent.

3. The method according to claim 2, characterized in that The acquiring of configuration information according to the task identifier includes: Using the task identifier as an input parameter of a hypertext transfer protocol interface call; Perform task query by calling input parameters through the interface of the hypertext transfer protocol to obtain query results; The configuration information corresponding to the task identifier is acquired according to the query result.

4. The method according to claim 1, characterized in that: The determining of the outbound calling task according to the preset task parameters includes: determining the outbound calling task for a familiar customer according to a familiar customer mode parameter; After the target intelligent agent performs voice interaction with the user who answers the outbound call, the method further includes: Acquiring call history information according to the outbound calling task for regular customers; Determine the optimal service seat for the user according to the call history information; If the user requests manual service, the target intelligent agent is transferred to the optimal service agent.

5. The method according to claim 1, characterized in that The making an outbound call to the calling number includes: Determine whether the current time matches the preset outbound call time threshold; If the current time matches the preset outbound call time threshold, an outbound call is made to the calling number.

6. A voice interaction device, characterized in that: The device comprises: Parameter acquisition module, used to obtain preset task parameters; A task determination module is used to determine an outbound call task according to preset task parameters; wherein the preset task parameters include: dialing round, user level, whether the user is a familiar customer; The number acquisition module is used to obtain the calling number according to the outbound call task; A customer calling module, used for making an outbound call to the calling number; An identification acquisition module, used for acquiring a task identification of the outbound call task if it is determined that the outbound call is answered; An initialization module, used to initialize the configuration of the intelligent agent to realize the service function according to the configuration information corresponding to the task identifier and the user portrait of the outbound user, and obtain the target intelligent agent; wherein the initialization is to set the configuration of the intelligent agent to realize the service function; the configuration information is used to realize the intelligent agent service function; The voice interaction module is used to perform voice interaction with the user who answers the outbound call through the target intelligent agent.

7. A computer device, characterized in that: include: Memory; One or more processors coupled to the memory; One or more applications, wherein the one or more applications are stored in the memory and are configured to be executed by the one or more processors, and the one or more applications are configured to execute the voice interaction method as described in any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program code, and the program code can be called by a processor to execute the voice interaction method according to any one of claims 1 to 5.

9. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the voice interaction method according to any one of claims 1 to 5 is implemented.

Images