CN111063348B

CN111063348B - Information processing method, device and equipment and computer storage medium

Info

Publication number: CN111063348B
Application number: CN201911284989.2A
Authority: CN
Inventors: 罗佳晨
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2019-12-13
Filing date: 2019-12-13
Publication date: 2022-06-07
Anticipated expiration: 2039-12-13
Also published as: CN111063348A

Abstract

The application provides an information processing method, an information processing device, information processing equipment and a computer storage medium, relates to the technical field of artificial intelligence, and is used for optimizing an information processing process in a task type conversation scene. The method comprises the following steps: receiving dialog information sent by terminal equipment, wherein the dialog information comprises information indicating that the terminal equipment executes a target task by a first object; natural language understanding is carried out on the dialogue information to determine semantic parameters of the dialogue information, and a reply language template matched with the dialogue information is determined according to the semantic parameters; determining the target task and the execution result of the target task according to the semantic parameters, and generating a reply language corresponding to the dialog information according to the execution result and the reply language template; and sending the reply message to the terminal equipment. The method improves the flexibility of generating the reply language so that the first object can know the execution result of the target task according to the reply language, and further improves the user experience.

Description

Information processing method, device and equipment and computer storage medium

Technical Field

The present application relates to the field of artificial intelligence technologies, and in particular, to an information processing method, apparatus, device, and computer storage medium.

Background

With the rapid development of the Artificial Intelligence (AI) technology, various intelligent assistants are increasingly popularized, and the intelligent assistants can output corresponding reply words according to the dialog information of the user, so that the intelligent assistants are very convenient and fast. At present, the intelligent assistant can only generate the reply language by applying a fixed reply language template set by the intelligent assistant platform according to the dialog information indicated by the user, but in the scene of task-type dialog, the reply language output in the mode is single, fixed and lack of diversity, the actual execution condition of the task cannot be directly reflected, and the interaction experience of the user is poor.

Disclosure of Invention

The embodiment of the application provides an information processing method, an information processing device, information processing equipment and a computer storage medium, which are used for optimizing an information processing process in a task-type conversation scene.

In a first aspect of the present application, an information processing method is provided, including:

receiving dialog information sent by terminal equipment, wherein the dialog information comprises information indicating that the terminal equipment executes a target task by a first object;

natural language understanding is carried out on the dialogue information to determine semantic parameters of the dialogue information, and a reply language template matched with the dialogue information is determined according to the semantic parameters;

determining the target task and the execution result of the target task according to the semantic parameters, and generating a reply language corresponding to the dialog information according to the execution result and the reply language template;

and sending the reply language to the terminal equipment.

In a possible implementation manner, the reply language template further includes an authority reply language template, and before determining the execution result of the target task, the method further includes:

if the resource authority is determined to be needed for executing the target task, generating an authority acquisition reply according to the authority reply language template;

sending the permission acquisition reply to the terminal equipment so as to acquire the resource permission indicated by the first object according to the permission acquisition reply through the terminal equipment;

and receiving the resource authority sent by the terminal equipment, and executing the target task according to the resource authority.

In a second aspect of the present application, there is provided an information processing apparatus comprising:

the device comprises a conversation information receiving unit, a task processing unit and a task processing unit, wherein the conversation information receiving unit is used for receiving conversation information sent by terminal equipment, and the conversation information comprises information that a first object indicates the terminal equipment to execute a target task;

the reply language template determining unit is used for performing natural language understanding on the dialogue information to determine semantic parameters of the dialogue information and determining a reply language template matched with the dialogue information according to the semantic parameters;

a reply language determining unit, configured to determine the target task and an execution result of the target task according to the semantic parameters, and generate a reply language corresponding to the dialog information according to the execution result and the reply language template;

and the reply language sending unit is used for sending the reply language to the terminal equipment.

In a possible implementation manner, the reply language template includes a statement template configured by a second object for the terminal device, and the second object is a device accessor of the terminal device.

In a possible implementation manner, each reply language template includes at least one reply language sub-template of one task, wherein one reply language sub-template corresponds to one execution result of the one task, and the reply language determining unit is configured to:

determining a reply language sub-template corresponding to the execution result of the target task in the reply language template;

and generating a reply language of the dialog message according to the reply language sub-template.

In a possible implementation manner, the semantic parameter includes one of a domain, an intention, a slot, a receiving time of the dialog information, and device information of the terminal device, and the reply language determining unit is configured to:

determining a reply language template corresponding to the semantic parameter;

and selecting one reply language template matched with the dialogue information from the determined reply language templates.

In one possible implementation manner, the semantic parameters include at least two of a domain, an intention, a slot, a receiving time of the dialog information, and device information of the terminal device, and the reply language determination unit is configured to:

respectively determining the matching degree of each reply language template and the at least two semantic parameters;

and determining a reply language template with the matching degree in the designated ordinal as the reply language template of the dialogue information.

In a possible implementation manner, the reply phrase template further includes a conditional reply phrase template valid at the validation time, and the reply phrase determining unit is further configured to:

determining a receiving time of the dialogue information;

if the dialog information is determined to be the first dialog information sent after the terminal equipment is awakened and an effective conditional reply language template is determined to exist, generating a conditional reply language according to the effective conditional reply language template, wherein the effective conditional reply language template is determined according to the effective time of the template and the receiving time of the dialog information;

and sending the condition reply language to the terminal equipment so as to play or display the condition reply language to the first object through the terminal equipment.

In a possible implementation manner, the reply language template includes a reply language template that is not enabled and indicates that the terminal device does not support the target task, and the reply language determining unit is further configured to:

after the terminal device is determined not to support the target task, generating an un-enabled reply language according to the un-enabled reply language template;

and sending the non-enabled reply language to the terminal equipment so as to indicate that the terminal equipment does not support the target task to the first object through the terminal equipment.

In a possible implementation manner, the reply phrase template further includes an authority reply phrase template, and the reply phrase determining unit is further configured to:

before the execution result of the target task is determined, if the resource authority is determined to be needed for executing the target task, generating an authority acquisition reply according to the authority reply language template;

In a possible implementation manner, the reply phrase template determining unit is further configured to:

In a third aspect of the present application, there is provided a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of the first aspect and any one of the possible embodiments when executing the program.

In a fourth aspect of the present application, a computer-readable storage medium is provided, which stores computer instructions that, when executed on a computer, cause the computer to perform the method according to the first aspect and any one of the possible embodiments.

Due to the adoption of the technical scheme, the embodiment of the application at least has the following technical effects:

according to the method and the device, the reply language template is selected according to semantic parameters acquired by analyzing the dialogue information, the reply language is generated according to the execution result of the target task of the dialogue information, the reply language output in the mode is various, and after the terminal device displays the reply language to the object, the object can directly know the execution result of the target task from the reply language, the flexibility of generating the reply language is improved, so that the first object can know the execution result of the target task according to the reply language, and the user experience is further improved.

Drawings

Fig. 1 is a schematic diagram of an application scenario provided in an embodiment of the present application;

fig. 2 is a diagram illustrating an example of an information processing process provided in an embodiment of the present application;

fig. 3 is a schematic diagram illustrating a process of screening matched reply language templates according to an embodiment of the present application;

FIG. 4 is a diagram illustrating another process for screening matched reply phrase templates according to an embodiment of the present application;

fig. 5 is a schematic interface diagram of an apparatus dialog platform according to an embodiment of the present application;

FIG. 6 is a schematic diagram of an attribute editing interface provided in an embodiment of the present application;

FIG. 7 is a schematic diagram of a configuration domain interface according to an embodiment of the present application;

fig. 8 is a schematic diagram of a reply language configuration interface according to an embodiment of the present application;

FIG. 9 is an exemplary diagram of a configuration reply phrase template provided by an embodiment of the present application;

FIG. 10 is a diagram illustrating an exemplary configuration interface for an unenabled reply language according to the present application;

FIG. 11 is a schematic diagram of another exemplary configuration interface for an unenabled reply according to the present disclosure;

fig. 12 is a schematic interface diagram of a music operation reply according to an embodiment of the present application;

fig. 13 is a schematic interface diagram of another music operation reply provided in the embodiment of the present application;

FIG. 14 is a diagram illustrating a sensitive vocabulary detection process according to an embodiment of the present application;

fig. 15 is a process diagram of a specific example of processing information according to an embodiment of the present application;

fig. 16 is a schematic diagram illustrating a reply template that satisfies a device condition according to an embodiment of the present application;

FIG. 17 is a diagram illustrating a process of selecting reply language templates satisfying an intention according to an embodiment of the present application;

fig. 18 is a schematic diagram illustrating a process of screening reply language templates satisfying slots according to an embodiment of the present application;

fig. 19 is a schematic diagram illustrating a process of screening reply language templates satisfying a receiving time according to an embodiment of the present application;

fig. 20 is a schematic structural diagram of an information processing apparatus according to an embodiment of the present application;

fig. 21 is a schematic structural diagram of another information processing apparatus according to an embodiment of the present application.

Detailed Description

In order to better understand the technical solutions provided by the embodiments of the present application, the following detailed description is made with reference to the drawings and specific embodiments.

In order to facilitate those skilled in the art to better understand the technical solutions of the present application, the following description refers to the technical terms of the present application.

The terminal equipment: the system comprises various hardware, terminals, application software and the like which have the service capability of building a task conversation platform; the invention particularly relates to an intelligent terminal which can interact with a user through a human-computer interaction interface and can interact with a background server through a network, wherein the intelligent terminal comprises an intelligent mobile phone, a computer, an intelligent sound box, an intelligent wearable device, an application APP, a vehicle-mounted device, a television, a robot, a sound box and the like.

Task-based dialog: there are explicit tasks and desired conversations to return, such as "i want to listen to a song", "i want to listen to a phase", etc.

And (3) replying: that is, the reply sentence to the dialog information of the user refers to a sentence that answers the dialog information of the user in a natural, concise and friendly manner, and is usually presented in the form of a broadcast on the terminal device.

Natural Language understanding (NLP Processing): the branch disciplines in the fields of artificial intelligence and linguistics discuss how to process and use natural language, and natural language processing includes multiple aspects and steps, and basically comprises parts of cognition, understanding, generation and the like.

The field is as follows: NLP processing is carried out on the dialogue information of the user to obtain a preliminary intention category of the dialogue information expression semantics, for example, the field of the dialogue information 'i want to listen to songs' is music, and the field of the dialogue information 'how to listen to the songs' is weather.

Intention is: a concept in the technical field of NLP processing refers to the purpose that a user expresses in order to meet a certain requirement, in short, the user expresses "what is desired", such as the intention of playing a song, adding a collection, switching a previous song, and the like.

And (4) slot position: the key information carried in the dialog information, such as singer XX in the dialog information "song playing singer XX", if the key information is missing, the intention of this dialog information may not be clear, the key information is abstracted to be identified by character variables understood by the machine, i.e. slot, such as singer is identified as singer and song is identified as song.

The following explains the concept of the present application.

In a task type conversation scene, a terminal device uploads conversation information indicated by a user, such as ' please play a song of singer AA ', a task conversation platform executes a target task in the conversation information, namely plays the song of singer AA ', a reply language is determined by using a reply language template set by the task conversation platform, and the reply language is issued to the terminal device for playing, but the reply language is usually generated by using a reply language template fixed by the task conversation platform, such as ' play music for you ', and the reply language can only express the played music to the user, but what music is played specifically, and what personal music is played, and the user cannot know the reply language.

On the other hand, the device access side of the terminal device can only generate the reply language according to the reply language template set by the task dialogue platform, and cannot set the reply language according to the self requirement or individuation, so that the flexibility is poor.

In view of the above, the inventors devised an information processing method, apparatus, device, and computer storage medium, the method for optimizing an information processing process in a task-based dialog scenario, the method comprising: after receiving the dialogue information sent by the terminal equipment, the server comprises information indicating the terminal equipment to execute the target task by the first object; then natural language understanding is carried out on the dialogue information to determine semantic parameters of the expression semantics of the dialogue information, and a matched reply language template is selected according to the determined semantic parameters; and the server determines a target task and an execution result thereof according to the semantic parameters, combines the execution result with the selected reply language template to generate a reply language of the dialog information, and sends the reply language to the terminal equipment.

Further, in order to improve flexibility and personalized preference of generating the reply language, the reply language template may include a sentence template configured by the second object for the terminal device.

It should be noted that, in the present application, the first object is a device user of the terminal device, the second object is a device access person of the terminal device, and the first object and the second object may be the same object or different objects.

After introducing the design concept of the embodiment of the present application, some simple descriptions are provided below for application scenarios to which the technical solution of the embodiment of the present application can be applied, and it should be noted that the application scenarios described below are only used for describing the embodiment of the present application and are not limited. In a specific implementation process, the technical scheme provided by the embodiment of the application can be flexibly applied according to actual needs.

Referring to fig. 1, the application scenario includes at least one terminal device 110 and at least one server 120, where:

the terminal device 110 is configured to receive the dialog information indicated by the first object and send the dialog information to the server 120, and receive and play a reply language sent by the server 120.

The server 120 performs NLP processing on the received dialog information, selects a reply language template according to the understanding result, determines an execution result of the target task, generates a reply language according to the determined execution result and the selected reply language template, and sends the reply language to the terminal device 110.

And analyzing the music playing request, acquiring a secondary playlist of the target music unit according to the analyzed result, and sending the music unit information of the target music unit and the secondary playlist to the terminal.

The terminal device 110 may be installed with various types of application software, and the object may send the session information to the server 120 through the terminal device or various types of application software on the terminal device.

The terminal device 110 may be an intelligent speaker, and an intelligent interaction module, a media playing module, a communication module, and the like are configured in the intelligent speaker, where the intelligent interaction module is configured to obtain dialog information indicated by the first object, and the intelligent interaction module may further include a touch interaction screen, a microphone, and the like.

Based on the application scenario, an information processing method provided in the embodiment of the present application is described below, where the information processing process is described below only by taking the server 120 as an execution subject, please refer to fig. 2, and specifically includes the following steps:

step S201, receiving dialogue information sent by the terminal equipment, wherein the dialogue information comprises information that the first object indicates the terminal equipment to execute the target task.

As an example, the terminal device 110 may collect dialog information of the first object, such as "please play a certain song YYY of the singer" after being awakened by the first object, and send the dialog information to the server 120 through the communication network.

As an embodiment, the dialog information may be voice information or text information, and the terminal device may collect voice information of the first object through an intelligent interaction module, such as a microphone, or may collect text information input by the first object through a touch interaction screen.

Step S202, NLP processing is carried out on the dialogue information to determine semantic parameters of the dialogue information, and a reply language template matched with the dialogue information is determined according to the semantic parameters.

As an embodiment, if the dialog information is voice information, before performing NLP processing, ASR recognition is first performed on the dialog information to obtain text information corresponding to the dialog information;

as an example, the semantic parameter may include, but is not limited to, one or more parameters of a domain, an intention, a slot, etc. of the dialog information expression semantics; the domains may be, but are not limited to, classified by requesting resources according to the session information, such as classifying the domains into weather, music, news, games, stocks, etc.; when the field is music, the intention may be, but is not limited to, playing a song, adding a collection, switching a song of the previous song, querying a song, and the like, the slot may be a singer, a song, and the like, and a person skilled in the art may set the field, the intention, and the slot according to actual needs, for example, the dialog information "please play a song YY of a certain song by the singer" and the obtained field after NLP processing is "music", the intention is "playing a song", and the slot is "a certain song" and "YY".

As an embodiment, a plurality of reply language templates may be stored in the server 120, or may be stored in another server or device; after determining the semantic parameters of the dialog information, the server 120 may select a reply phrase template matching the dialog information according to the semantic parameters.

Step S203, determining the target task and the execution result of the target task according to the semantic parameters, and generating a reply language corresponding to the dialog information according to the execution result and the reply language template.

As an embodiment, the server 120 may further determine the target task of the dialog information according to the semantic parameters, such as "play a certain song YY" according to "music", "play a song", "zhao" and "YY"; further, the server 120 may determine an execution result of the target task, and further generate a reply language according to the execution result and the selected reply language template.

As an embodiment, there may be multiple execution results of some target tasks, and therefore, in this embodiment, one reply language template may include one or more reply language sub-templates of one task, and one reply language sub-template corresponds to one execution result of the task, and after determining the execution result of the target task, the server 120 may select, from the reply language templates corresponding to the dialog messages, a reply language sub-template corresponding to the determined execution result, and generate a reply language using the selected reply language sub-template.

Each reply language template and each reply language sub-template comprise key words and placeholders, and when reply languages are generated, the obtained slots and the like can be used for replacing the placeholders in the reply language sub-templates to obtain the reply languages.

And step S204, sending the generated reply language to the terminal equipment.

As an example, the server 120 may send the text message of the reply language to the terminal device 110, or may convert the text message of the reply language into a voice message and send the voice message to the terminal device 110; correspondingly, after receiving the text message of the reply language, the terminal device 110 may display the text message through the display screen, or after receiving the voice message of the reply language, the terminal device 110 may play the voice message through the speaker, and the first object may know the execution condition of the target task indicated by the dialog message through the reply language.

As an embodiment, the semantic parameters in the embodiment of the present application further include one or more of device information of the terminal device and receiving time of the session information; when the server 120 receives the session information sent by the terminal device, the receiving time of the received session information may be recorded, and the device information uploaded by the terminal device may also be received at the same time; the device information may include, but is not limited to, a device name, a device type, a brand, a model, etc. of the terminal device, and the reception time may include, but is not limited to, a specific time and date when the session information is received.

As an embodiment, the following describes the determination of a reply phrase template matching with the dialog information according to the semantic parameters in step S202:

the semantic parameter in the embodiment of the present application may include, but is not limited to, one or at least two of a field, an intention, a slot, a reception time of the session information, and device information of the terminal device, which will be described in the following cases.

Case 1: the semantic parameter includes only one

In this case, the reply language template corresponding to the semantic parameter may be determined first, and one reply language template determined to be a matching reply language template may be selected from the determined reply language templates, where if only one reply language template is determined, the reply language template is determined to be the matching reply language template, and if a plurality of reply language templates are determined, one reply language template determined to be a matching reply language template may be selected randomly from the plurality of reply language templates or according to other criteria.

Case 2: the semantic parameters include at least two

In this case, the matching degree between each reply language template and at least two semantic parameters can be respectively determined, and the reply language template with the matching degree in the designated sequence is selected as the reply language template matched with the dialogue information.

The designated ordinal is not limited too much, and those skilled in the art can set the order according to actual requirements, for example, the matching degrees of the reply language templates are arranged in the order from top to bottom, and the position of the most front matching degree is taken as the designated ordinal.

In this embodiment of the application, the set reply language template may be stored according to the semantic parameters, for example, reply language templates in the same field are stored in the same position, so that the server 120 may quickly search for a reply language template matching the dialog information.

As an embodiment, when the semantic parameters include the field, the intention, the slot, the receiving time of the session information, and the device information of the terminal device, the reply language template matching the session information may be selected according to the following process, please refer to fig. 3, which specifically includes:

and S301, screening the reply language template which accords with the field.

And when the terminal equipment executes the target task to call the resource, automatically screening a reply language template corresponding to the field of the dialogue information according to the field.

Step S302, determining the matching degree of the reply language template and a plurality of semantic parameters.

The matching degree of each reply language template and semantic parameters can be determined according to the intention, slot position and receiving time of each reply language template dialogue statement and the condition of matching with the equipment information of the terminal equipment, and each reply language template can be graded and sorted according to the matching degree of each reply language template and the equipment information, intention, slot position and receiving time, and the method is as follows:

A. respectively determining whether each reply language template is matched with the equipment information of the terminal equipment;

B. respectively determining whether each reply language template is matched with the intention;

C. respectively determining whether each reply language template is matched with the slot position;

D. and respectively determining whether each reply language template is matched with the receiving time of the dialogue information.

If the more semantic parameters are matched in the a-D, the higher the matching degree of the template with the dialog information is, further, the matching value of each semantic parameter in the a-D may be set to 1, when a certain reply language template is matched with the corresponding semantic parameter, the 1 score is given to the reply language template, when the certain reply language template is not matched with the corresponding semantic parameter, the 1 score is given to the reply language template, the sum of the total scores of the reply language templates a-D is taken as the matching degree of the reply language template with the dialog information, and if a certain reply language template is matched with the three semantic parameters in the a-D, the score of the reply language template is determined to be 3, that is, the matching degree of the reply language template with the dialog information is determined to be 3.

Here, in order to further increase the flexibility of generating the reply language, it may be considered that different weighted values of semantic parameters are given according to the importance degree of each semantic parameter, when the reply language template matches a certain semantic parameter, a corresponding score is given to the reply language template according to the corresponding weighted value, for example, the weighted values of the device information, the intention, the slot position, and the receiving time of the dialog information in the above a-D are respectively set to 3, 2, and 1, and further, the sum of the total score values of the reply language templates a-D is taken as the matching degree of the reply language template and the dialog information, when a certain reply language template matches two semantic parameters of the device information and the intention in the above a-D, the total score of the reply language template is determined to be 5, that is, the matching degree of the reply language template and the dialog information is 5.

Step S303, determining a reply language template matched with the dialogue information according to the matching degree of each reply language template.

After scoring the reply language templates, the reply language templates can be sorted according to the sequence of the matching degrees from large to small, and then the reply language template with the top matching degree can be used as the reply language template matched with the dialogue information.

As an embodiment, in the above case 2, the matched reply phrase template may be further selected according to the following process, please refer to fig. 4, which specifically includes:

step S401, screening reply language templates corresponding to the fields, and entering step S407 if only one reply language template corresponding to the field exists; if the reply language templates corresponding to the fields include at least two, the method goes to step S402; if there is no reply language template corresponding to the domain, the process proceeds to step S4011.

Step S4011, using the replying language template of the bibliographic bottom as the replying language template matching with the dialog information.

In order to further increase flexibility of generating a reply language, the embodiment of the present application is provided with the foregoing pocket-bottom reply language template, and when no reply language template corresponding to the dialog information can be found according to the semantic parameters, the pocket-bottom reply language template is used as a matched reply language template, and the content of the pocket-bottom reply language template may include, but is not limited to, "i don't hear clearly, it is troublesome for you to say again", { nickname } is vague, please say again ", and the like, where { nickname } may be a device name of the terminal device.

Step S402, screening reply language templates corresponding to the equipment information from the reply language templates corresponding to the fields, and entering step S407 if only one reply language template corresponding to the equipment information exists; if the reply language template corresponding to the device information includes at least two, go to step S403; if there is no reply language template corresponding to the device information, the process proceeds to step S4021.

Step S4021, randomly selecting one from the reply language templates corresponding to the fields as a reply language template matching the dialog information.

Step S403, screening reply language templates corresponding to the intentions from the reply language templates corresponding to the equipment information, and if only one reply language template corresponding to the intentions exists, entering step S407; if the reply language template corresponding to the intention includes at least two, go to step S404; if there is no reply language template corresponding to the intention, the process proceeds to step S4031.

Step S4031, randomly selects one from the reply language templates corresponding to the device information as a reply language template matching the session information.

Step S404, screening the reply language templates corresponding to the slot positions from the reply language templates corresponding to the intentions, and entering step S407 if only one reply language template corresponding to the slot position exists; if the reply language templates corresponding to the slot positions comprise at least two, the step S405 is executed; if there is no reply phrase template corresponding to the slot, the process proceeds to step S4041.

Step S4041, one of the reply phrase templates corresponding to the intention is randomly selected as a reply phrase template matching the dialogue information.

Step S405, screening reply language templates corresponding to the receiving time of the dialogue information from the reply language templates corresponding to the slot positions, and entering step S407 if only one reply language template corresponding to the receiving time of the dialogue information exists; if the reply language template corresponding to the receiving time of the dialog message includes at least two, step S406 is performed; if there is no reply language template corresponding to the reception time of the dialogue information, the process proceeds to step S4051.

Step S4051, randomly selecting one reply language template from the reply language templates corresponding to the slots as the reply language template matched with the dialogue information.

In step S406, one of the reply language templates corresponding to the reception time of the dialogue information is randomly selected as the reply language template matching the dialogue information.

Step S407, using the screened reply language template as a reply language template matched with the dialogue information.

As an embodiment, the reply phrase template in this embodiment further includes a conditional reply phrase template, and the conditional reply phrase template is valid in the valid time, in step S201, after the server 120 receives the dialog information sent by the terminal device, the receiving time of the dialog information may be determined, and then it is determined whether there is a valid conditional reply phrase template according to the receiving time, which includes the following specific processes:

the server 120 determines whether the dialog message is the first dialog message sent after the terminal device wakes up, if so, further determines whether an effective conditional reply language template exists, and if so, generates a conditional reply language according to the effective reply language template, and sends the conditional reply language to the terminal device, so that the terminal device plays or displays the conditional reply language to the first object.

In the above process, the condition reply may be determined according to the validation time of the condition reply word and the receiving time of the dialog message, and if the receiving time corresponds to the validation time of a certain condition reply word template, it is determined that the condition reply is a valid condition reply word.

The effective time may be a specific time point or time period, may be a specific date, may be a fixed time period of a specified date, may also be a fixed date of a specified year, and the like, if the effective time of a certain conditional reply phrase template is set to be "20: 00-20: 30", it means that the reply phrase template takes effect at 20:00-20:30 every day; if the effective time of a certain conditional reply language template is set to be 10 months and 1 days, 20:00-20:30, the reply language template is indicated to be effective at 20:00-20:30 days of 10 months and 1 days.

If a condition reply language template is set as 'meta-denier happy', the corresponding effective time is 1 month and 1 day per year, if the receiving time of the first dialogue information sent after the terminal equipment is awakened is 1 month and 1 day per year, the 'meta-denier happy' is determined as the effective reply language template, and the condition reply language such as 'wish you happy holiday' or 'meta-denier happy' or 'wish you happy' is generated according to the template, and the generated condition reply language is sent to the terminal equipment.

As an embodiment, a person skilled in the art may set the type of the execution result of the target task according to the actual situation, where the execution result may include, but is not limited to, success of execution, failure of execution, partial execution, and the like, for example, when a target task is "play song BB of singer AA", the success of execution may be successful play song BB of singer AA, the failure of execution may be no song found by singer AA or other songs played by other singers, and the like, and the partial execution may be songs played by singers other than BB of singer AA or songs CC played by other singers other than BB.

The reply language sub-template for "playing song BB of singer AA" corresponding to the above target task may be as in the case of table 1 below:

table 1:

in table 1, nickname is a placeholder for the device name of the terminal device; singer and e _ singer are singer placeholders, which may be the same, and are distinguished here only for convenience of representation, wherein singer represents a singer placeholder indicated by the target task, and e _ singer represents a singer placeholder other than the singer indicated by the target task; song and e _ song are placeholders for songs, both of which may be the same, and are distinguished here only for ease of representation, where song represents a placeholder for a song indicated by the target task and e _ song represents a placeholder for a song other than that indicated by the target task.

After determining the execution result of "playing the song BB of the singer AA", determining a reply language sub-template according to the determined execution result, if the execution result is determined to be successful, determining the determined reply language sub-template to be "{ nickname } already playing the song { song } of { singer } and" { song } asking you to enjoy the singer { singer } ", selecting one of the two reply language sub-templates, further replacing nickname with the device name, replacing singer in the reply language with AA, replacing song in the reply language with BB, namely generating a reply song" playing the BB of the singer AA ", where the reply language may be" BB song that a certain device has played AA for you "or" asking you to enjoy the BB of the singer AA ".

The method for selecting the reply language sub-template according to the other execution results and generating the reply language by using the reply language sub-template can refer to the above description, and the description is not repeated here.

As an embodiment, there are some target tasks that are not supported by the terminal device, and at this time, if a reply language prompt first object can be generated for the target task that is not supported by the terminal device, the user experience of the first object can be improved, so the reply language template provided in this embodiment further includes a reply language template that is not enabled, and is used to indicate that the terminal device does not support the target task; specifically, after the target task is determined according to the semantic parameters, if the terminal device is determined not to support the target task, the non-enabled reply is generated according to the non-enabled reply template, and then the non-enabled reply is sent to the terminal device, so that the terminal device is indicated to the first object through the terminal device that the terminal device does not support the target task.

As an embodiment, the reply phrase template further includes a permission reply phrase template, and after it is determined that the terminal device supports the target task, if it is determined that the execution of the target task requires the resource permission, the permission reply phrase may be generated according to the permission reply phrase template, and the permission acquisition reply phrase may be further sent to the terminal device, so that the terminal device acquires the resource permission indicated by the first object according to the permission acquisition reply phrase; and then after the resource authority is obtained, executing the target task according to the resource authority.

The specific form and content of the permission reply language model are not limited too much, and the technical personnel in the field can flexibly set the permission according to the resources required in the target task execution process.

To more clearly understand the rights reply template, an illustrative example is given here:

in this example, the execution of a certain target task needs to call the music resource of the music application, the resource permission is the account and password of the login music application YY, and the permission reply template may be "please input the account and password of the music application { music _ name }; at this time, if the music application needs to be called for execution of a certain target task, a permission acquisition reply is generated according to an account and a password of the music application { music _ name }, and the permission acquisition reply is sent to the terminal device, wherein the music _ name in the permission reply template is a placeholder of the name of the music application.

In order to improve the user experience, the reply language template may be set by the second object, and meanwhile, in order to facilitate management of the terminal device and setting of the reply language template, the embodiment further provides a device dialogue platform, which may provide all or part of the functions of the server 120, and the first object or the second object may access one or more terminal devices to the device dialogue platform, so as to complete a dialogue with the terminal device, or set the reply language template for the accessed terminal device.

Specifically, the second object may perform device version management, attribute setting, and reply language setting, etc. on the terminal device through the device dialog platform, which will be described in detail below.

Carrying out attribute setting on terminal equipment

Referring to fig. 5, as an embodiment, after the terminal device accesses the device dialog platform, and the second object logs in the account of the device dialog platform, an avatar picture of the second object and a user name "XX user" are displayed at 501, and after the user selects the terminal device in the application list, the user enters a management interface 502 of the terminal device.

The management interface 502 displays the device name of the terminal device, and in fig. 5, "XX _ device" is used as the name of the terminal device; the management interface 502 includes content related to an application overview, which includes application configuration, application access, quality test, release management, and the like of the terminal device.

The second object can set version management, attribute setting, reply language setting and the like of the terminal equipment through application configuration, wherein the second object can enter the attribute setting interface 504 by clicking an attribute setting key 503; further, the attribute of the terminal device is set in the attribute setting interface 504, for example, the second object can enter the attribute setting interface by clicking the add person button 505, but not limited to, adding a character attribute for the terminal device for representing the character attribute setting of the terminal device, and the character attribute can but not limited to include one or more of an age attribute and a nickname attribute.

Further, the second object may set an avatar attribute according to a feature of the first object facing the terminal device during use, for example, when the first object is a child, the age attribute of the terminal device may be set to 3 years old, and the nickname attribute may be set to "small fly".

As an embodiment, if the terminal device and other terminal devices form a smart home, the above-mentioned character attribute can also include a member attribute, for example, in fig. 5, the member attribute of the terminal device is set as dad, which indicates that the role of the terminal device in the smart home is dad.

As an embodiment, the second object may further click on the operation button 506 in each attribute to enter the corresponding attribute editing interface, please refer to fig. 6, which shows an attribute editing interface 601 of an age attribute.

The name, category, placeholder and value of the attribute are displayed in the attribute editing interface 601; the second object may input the age of the terminal device in the numeric value input box 602 of the attribute, and may add a reply template of the question example by clicking the add question and answer button 603, such as "i is three years old this year" or "tells you quietly," i [ age ] cheer ", where age refers to age; further, the second object may delete the corresponding reply phrase template through the deletion mark 604 on the right side of the reply phrase template.

Further, the second object may cancel the editing operation on the attribute this time by using a cancel button 605 on the lower right of the attribute editing interface 601, or confirm and save the editing operation on the attribute this time by using the determination button 606.

(II) configuring reply language template

As an embodiment, the reply language template may be configured in different domains according to a target task supported by the terminal device.

Referring to fig. 7, the second object enters the configuration field interface 702 after clicking the reply language configuration, in the configuration field interface 702, the second object may select and configure a reply language template of a music reply language, a radio reply language, a non-enabled reply language, and a linguet reply language, where the music reply language is a reply language in the music field, the radio reply language is a reply language in the radio field, and after determining that there is no reply language template matching with the dialog information in the reply language template configured for the second object, the linguet reply language template may be used to generate a linguet reply language and send the linguet reply language to the terminal device.

The second object can perform the setting of the detailed reply phrase template through the configuration key 703 in the reply phrase configuration of each field.

Referring to fig. 8, as an embodiment, after the second object clicks a configuration button 703 in a music reply in a configuration domain interface 702, the reply language configuration interface 801 of a customized music reply template interface is entered, one or at least two matching conditions may be added in the reply language configuration interface, where the matching conditions may include, but are not limited to, one or more of a song, a singer, an album, a slot tag, a date, and a time, and fig. 8 illustrates a case where one matching condition is added, that is, the adding matching condition is "song qilixiang".

Furthermore, through the reply language configuration interface 801, each reply language sub-template is configured according to the execution result of the target task, and when the dialog information includes the set matching condition, the following reply language templates are triggered, wherein the reply language templates include the reply language sub-templates corresponding to the three execution results of the target task.

Referring to fig. 9, an example of configuring a reply phrase template is shown, where a reply phrase sub-template 901 corresponds to a case where execution succeeds, "{ nickname } helps you find the song { tag }, where nickname is a placeholder for a device name of a terminal device and tag is a placeholder for a song name, and in the example shown in fig. 9, when a reply phrase is generated by using the reply phrase sub-template 901, the placeholder nickname may be replaced by text of the device name and the placeholder tag may be replaced by" qilix ".

The reply language sub-template 902 corresponds to a partially executed case, including two cases of "{ nickname } helping you find the song of { r _ singer } and" { nickname } playing the song of { r _ singer } for you, where r _ singer is the placeholder of the singer, in the example shown in fig. 9, when the reply language is generated by using the reply language sub-template 902, one of the reply language sub-templates 902 may be selected, the placeholder nickname may be replaced by the text of the device name, and the placeholder r _ singer may be replaced by the text of the singer of "qilixiang".

The reply language sub-template 903 may be set to "{ nickname } please enjoy the recommended song" or the like, corresponding to the case where the execution fails.

When the

reply language sub-templates

901, 902, and 903 are configured, if there are at least two reply language sub-templates, the different reply language sub-templates may be separated by a character "|".

After configuring the reply language template, the editing operation on the attribute at this time can be canceled through a cancel button 904 at the lower right of the reply language configuration interface 801, or the information of the reply language template at this time can be confirmed and saved through a confirm button 905.

As an embodiment, the second object may further configure a non-enabled reply, where the non-enabled reply may be set according to whether the terminal device supports the target task, and specifically, the non-enabled reply may be set as follows.

Referring to fig. 10 and fig. 11, enter the non-enabled reply configuration interface 1001 through the configuration key in the non-enabled reply in the configuration field interface 702; by means of an adding configuration key 1002 in the reply language configuration interface 1001 or entering the adding configuration interface 1101 through an editing identifier 1003 in the reply language configuration interface 1001, a skill which is not supported by the terminal device and an un-enabled reply template corresponding to the skill can be set in the adding configuration interface 1101, wherein the skill has the same meaning as a target task, that is, fig. 10 and 11 illustrate that the terminal device does not support a target task related to a menu (recipe).

After the non-enabled reply language template is configured, the non-enabled reply language template set this time may be cancelled by adding a cancel button 1102 at the right lower side of the configuration interface 1101, or the non-enabled reply language template configured this time may be confirmed and saved by a determination button 1103.

As an embodiment, a conditional reply template may also be set by the following method, where the conditional reply may be an operation reply set by the device access party for the operation terminal device, and fig. 12 shows a music operation reply set for the music field, where:

the reply word customization button 1201 is used for entering a setting interface 1202 of a music operation reply word template, the reply word adding button 1203 in the setting interface 1202 is used for entering a customization interface 1204 of a music operation reply word, and then the effective time of the music operation reply word can be set in the customization interface 1204, the effective time can be a specific time period or a date, fig. 12 shows that the music operation reply word ' night quiet, and a music bar!with quiet listening point ' is used for showing that the music operation reply word ' is quiet at night! "effective time is from 23:00 to 23: 30; the music operation reply may also be set to take effect according to date in customization interface 1204, with FIG. 12 showing the music operation reply "summer," let us use music to eliminate the Hot wave Bar! "effective time is 06 months and 06 days.

One skilled in the art can also set other forms of conditional reply language templates, which are not limited herein.

As an embodiment, in order to improve the security of the device dialog platform, a server in the device dialog platform may further perform vocabulary detection on a reply template after receiving the reply template configured by the second object, please refer to fig. 14, which shows a process of configuring the reply template and detecting the sensitive vocabulary, where information transmission among the second object, the device dialog platform, the sensitive vocabulary detection module, and a reply template database is described, where the device dialog platform may operate on a certain terminal such as a computer, and the sensitive vocabulary detection module and the reply template database may be implemented by the server, where the specific information transmission is as follows:

step 1401, the second object configures a reply language template, and calls a sensitive vocabulary detection interface through the device dialogue platform.

Step S1402, the sensitive vocabulary detection module performs sensitive vocabulary detection on the configured reply language template, and sends a sensitive vocabulary detection result to the device dialogue platform.

In this step, the detection can be performed according to the following sensitive vocabulary detection principle:

the above sensitive vocabulary detection principle is only an example, and those skilled in the art can flexibly set other sensitive vocabulary detection principles.

Step 1403, if the device dialogue platform determines that sensitive words are not detected, the configured reply language template is stored in a reply language template database; if it is determined that sensitive vocabulary is detected, a sensitive vocabulary alert message is indicated to the second object s1403, so that the second object deletes or modifies the detected sensitive vocabulary.

As an embodiment, after determining that the sensitive vocabulary is not detected, the configured reply language template may be stored in a reply language database according to a data protocol format agreed with skills invoked by the device dialog platform and the execution target task, where the reply language template may be stored in a local reply language template database, or the reply language template may be stored in a reply language template database of another server, so that a reply language template matching the dialog information may be quickly screened from the reply language template database during information processing.

The specific form of the data protocol format is not limited too much, and those skilled in the art can set the data protocol format according to actual requirements, for example, the data protocol format is set to be JSON format.

The present embodiment further provides a specific example of processing information according to the reply language template configured by the second object, and here, a task in which music is taken as a domain and a target task is taken as a music domain is described as an example.

The following description is made with respect to interaction and implementation effects among the first object, the terminal device, and the server:

referring to fig. 15, in this example, the functions of the server 120 are provided by a device dialogue system, which may include a plurality of servers, each of which provides a service of one function module, and in the following example, only the respective function modules are referred to as the respective servers, and the processing of the interaction information between the respective function modules and the terminal is described.

The function modules comprise a Voice Service (TVS) module, a management module AIproxy, an ASR module, an NLP module, a skill Service (TSKM) module, a music skill Service module and a content source module; the content source module may be, but is not limited to, a music application.

In step S1501, after the terminal device detects the voice information of "playing song BB of singer AA", the voice information is sent to the server as the session information.

The voice information is indicated to the terminal equipment by the first object.

Step S1502, after receiving the dialogue information 'playing singer AA song BB', the server sends the dialogue information to AIproxy through TVS module, AIproxy calls ASR module, and the received equipment information of the terminal equipment and the receiving time of the dialogue information are used as semantic parameters of the dialogue information;

step S1503, the ASR module carries out voice recognition on the received dialogue information to obtain text information 'playing song BB of singer AA', and returns to AIproxy;

step S1504, after the AIProxy receives the text information returned by the ASR module, invoking the NLP module to perform semantic understanding on the text information "play singer AA song BB", determining the field, intention, and slot position of the text information as semantic parameters, and returning the determined semantic parameters to the AIProxy, wherein it is determined in fig. 15 that the field in "play singer AA song BB" is music, intention is play, singer in slot position is AA, and song in slot position is song.

In step S1505, the AIProxy determines that the target task of the dialog information is "play song BB of singer AA" according to the semantic parameters, and invokes the TSKM module.

In step S1506, the TSKM module invokes the music skill service module according to the domain to obtain the music resource, i.e., here, the song resource of the song BB of the singer AA is obtained.

When the music skill service module requests music resources from the content source module, if resource permission is needed, a permission reply language is generated according to the permission reply language template and sent to the terminal device, so that the needed resource permission is obtained through the terminal device.

In step S1507, the content source module returns the song resource to the music skill service module.

In step S1508, the music skill service module may determine an execution result of the target task according to the returned song resource, further determine a reply language sub-template of the dialog message according to the reply language template screened according to the semantic parameters and the execution result of the target task, and generate a reply language by using the determined reply language sub-template.

Step S1509, the music skill service module returns the acquired song resources and the reply language to the TSKM module, the AIProxy, and the TVS module in sequence, and then sends the song resources and the reply language to the terminal device, and sends a play instruction to the terminal device.

In step S1510, after receiving the play instruction, the terminal device broadcasts the reply and plays the obtained song resource.

Taking a jingle smart screen as a terminal device as an example, the following description is given to a process of selecting a reply language template matched with dialogue information according to a plurality of semantic parameters in the process, and the process mainly comprises the following contents of steps 1) to 6):

step 1) screening reply language templates meeting equipment conditions according to the equipment information.

Referring to fig. 16, in the present embodiment, reply language templates of different terminal devices can be stored as a reply language template list, and here, a reply language template list corresponding to the jingle smart screen can be selected according to the device information.

And 2) screening reply language templates meeting the intention in the semantic parameters.

Referring to fig. 17, the semantic parameter is intended to be played, and then different reply templates are screened to obtain a reply template satisfying the intention.

And 3) screening reply language templates meeting the slot positions.

Specifically, referring to fig. 18, reply language templates satisfying the conditions that the slot position signer is AA and the song is BB are selected from different reply language templates.

And 4) screening reply language templates meeting the receiving time.

Referring to fig. 19, the semantic parameters entered at this time are the receiving time, which includes the time point "08: 00: 00" and the date "10 months and 1 day".

And 5) determining the matching degree of each reply language template and the dialogue information according to the matching condition of each reply language template and each semantic parameter and sequencing the matching degree and the dialogue information.

It should be noted that, the above steps 1) to 4) are not in strict sequence, and those skilled in the art can set them according to actual requirements.

Where the symbols "√" in fig. 16-19 and the solid line indicate that each reply phrase template matches the corresponding semantic parameter, and the dashed line in fig. 16-19 indicates that each reply phrase template does not match the corresponding semantic parameter.

And 6) selecting the reply language template with the highest matching degree as the reply language template matched with the dialogue information.

It should be understood that, in the above method, only the dialog information in the music domain is taken as an example, and the process of generating the reply phrase is described, and the scheme of generating the reply phrase in this embodiment can be flexibly applied to the dialog information in each domain, and examples of the dialog information in several different domains and the reply phrases thereof are given in table 2 below.

Table 2:

in the embodiment of the application, the reply language template matched with the dialogue information can be screened out by combining the matching condition of the reply language and a plurality of semantic parameters, so that the matching degree of the screened reply language template and the dialogue information is improved; in addition, the embodiment of the application supports that the corresponding reply language template is selected to generate the reply language according to one or more execution results of the target task, so that the flexibility of the configured reply language is improved, the reply language generated according to the execution results of the target task can accurately reflect the execution condition of the dialog information of the user, and the user experience is improved.

Moreover, the embodiment of the application also supports the equipment access party of the terminal equipment to configure each reply language template, so that the equipment access party can conveniently and individually configure the reply language template, the image and the product of the terminal equipment can be displayed, and the brand image of the equipment access party can be further conveniently embodied.

Referring to fig. 20, based on the same inventive concept, an embodiment of the present application provides an information processing apparatus 2000, including:

a session information receiving unit 2001 for receiving session information transmitted by a terminal device, the session information including information that a first object instructs the terminal device to execute a target task;

a reply language template determination unit 2002 for performing natural language understanding on the dialogue information to determine semantic parameters of the dialogue information, and determining a reply language template matched with the dialogue information according to the semantic parameters;

a reply language determining unit 2003 for determining the target task and the execution result of the target task according to the semantic parameters, and generating a reply language corresponding to the dialog information according to the execution result and the reply language template;

a reply language sending unit 2004, configured to send the reply language to the terminal device.

As an embodiment, the reply language template includes a statement template configured by a second object for the terminal device, and the second object is a device accessor of the terminal device.

As an embodiment, each reply phrase template includes at least one reply phrase sub-template of a task, wherein one reply phrase sub-template corresponds to an execution result of the task, and the reply phrase determining unit is configured to:

As an embodiment, the semantic parameter includes one of a field, an intention, a slot, a receiving time of the dialogue information, and device information of the terminal device, and the reply language determination unit is configured to:

determining a reply language template corresponding to the semantic parameter;

As an embodiment, the semantic parameter includes at least two of a field, an intention, a slot, a receiving time of the dialog information, and device information of the terminal device, and the reply determination unit is configured to:

and determining the reply language template with the matching degree in the designated ordinal position as the reply language template of the dialog message.

As an embodiment, the reply phrase template further includes a conditional reply phrase template valid at the validation time, and the reply phrase determining unit is further configured to:

determining the receiving time of the dialogue information;

As an embodiment, the reply language template includes an un-enabled reply language template indicating that the terminal device does not support the target task, and the reply language determining unit is further configured to:

after the terminal equipment is determined not to support the target task, generating an un-enabled reply language according to the un-enabled reply language template;

As an embodiment, the reply phrase template further includes an authority reply phrase template, and the reply phrase determining unit is further configured to:

before the execution result of the target task is determined, if the execution of the target task needs resource authority, generating an authority acquisition reply according to the authority reply language template;

As an embodiment, the reply phrase template determining unit is further configured to:

As an example, the apparatus in fig. 20 may be used to implement any of the information processing methods discussed above.

As an example of the hardware entities of the apparatus shown in fig. 21, the gateway includes a processor 2101, a storage medium 2102, and at least one external communication interface 2103; the processor 2101, the storage medium 2102, and the external communication interface 2103 are connected via a bus 2104.

Here, it should be noted that: the description related to the above device is similar to the description of the method of the present application, and the description of the beneficial effects of the method is not repeated. For technical details not disclosed in the gateway embodiment of the present application, refer to the description of the method embodiment of the present application.

Based on the same technical concept, the embodiment of the present application also provides a computer-readable storage medium, which stores computer instructions that, when executed on a computer, cause the computer to execute the objective function determination method as discussed above.

As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims

1. An information processing method characterized by comprising:

receiving dialog information sent by terminal equipment, wherein the dialog information comprises information of a first object indicating the terminal equipment to execute a target task, and the first object is an equipment user of the terminal equipment;

determining a receiving time of the dialogue information;

determining whether the dialog information is the first dialog information sent after the terminal equipment is awakened, and determining whether an effective conditional reply language template exists;

if so, generating a conditional reply language according to an effective conditional reply language template, and sending the conditional reply language to the terminal device so as to play or display the conditional reply language to the first object through the terminal device, wherein the effective reply language template is determined according to the effective time and the receiving time of the dialogue information;

otherwise, natural language understanding is carried out on the dialogue information to determine semantic parameters of the dialogue information, and a reply language template matched with the dialogue information is determined according to the semantic parameters; and determining the target task and the execution result of the target task according to the semantic parameters, generating a reply language corresponding to the conversation information according to the execution result and the reply language template, and sending the reply language to the terminal equipment so as to play or display the reply language to the first object through the terminal equipment.

2. The method of claim 1, wherein the reply-to template comprises a statement template configured for the terminal device by a second object, the second object being a device accessor of the terminal device.

3. The method of claim 1 or 2, wherein each reply language template comprises at least one reply language sub-template of a task, wherein one reply language sub-template corresponds to an execution result of the task, and the generating of the reply language corresponding to the dialog message according to the execution result and the reply language template comprises:

4. The method of claim 1 or 2, wherein the semantic parameter includes one of a domain of the dialog information, an intention, a reception time of the dialog information, and device information of the terminal device, and wherein determining a reply language template matching the dialog information according to the semantic parameter includes:

determining a reply language template corresponding to the semantic parameter;

5. The method of claim 1 or 2, wherein the semantic parameters include at least two of a domain, an intention, a slot, a reception time of the dialogue information, and device information of the terminal device, and wherein the determining a reply language template matching the dialogue information according to the semantic parameters includes:

6. The method according to claim 1 or 2, wherein the reply language template comprises a non-enabled reply language template indicating that the terminal device does not support the target task, and when determining the execution result of the target task, the method further comprises:

after the terminal device is determined not to support the target task, generating an un-started reply according to the un-started reply template;

and sending the un-started reply language to the terminal equipment so as to indicate that the terminal equipment does not support the target task to the first object through the terminal equipment.

7. An information processing apparatus characterized by comprising:

the terminal equipment comprises a conversation information receiving unit, a conversation information processing unit and a task processing unit, wherein the conversation information receiving unit is used for receiving conversation information sent by terminal equipment, the conversation information comprises information that a first object indicates the terminal equipment to execute a target task, and the first object is an equipment user of the terminal equipment;

a reply language template determining unit, configured to determine a receiving time of the dialog information, determine whether the dialog information is a first dialog information sent after the terminal device is awakened, determine whether a conditional reply language template in effect exists, if not, perform natural language understanding on the dialog information to determine semantic parameters of the dialog information, and determine a reply language template matched with the dialog information according to the semantic parameters;

the reply language determining unit is used for generating a conditional reply language according to an effective conditional reply language template, wherein the effective reply language template is determined according to the effective time and the receiving time of the dialogue information; or determining the target task and the execution result of the target task according to the semantic parameters, and generating a reply language corresponding to the dialog information according to the execution result and the reply language template;

and the reply language sending unit is used for sending the condition reply language or the reply language to the terminal equipment, and playing or displaying the condition reply language or the reply language to the first object through the terminal equipment.

8. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the method of any of claims 1-6 are implemented when the program is executed by the processor.

9. A computer-readable storage medium having stored thereon computer instructions which, when executed on a computer, cause the computer to perform the method of any one of claims 1-6.