A human-computer conversation output control method, a terminal and a human-computer conversation system

Abstract

The application provides a human-computer conversation output control method, a terminal and a human-computer conversation system. The method comprises the following steps: obtaining conversation information input by a user, and pushing the conversation information to a human-computer conversation large model; obtaining a reply message output by the human-computer conversation large model in a continuous push stream form after the human-computer conversation large model receives the conversation information; on this basis, a message object is established when the obtained reply message is the first stream data output by the human-computer conversation large model, and the reply message is added to the message object when the obtained reply message is the last stream data output by the human-computer conversation large model or the Nth stream data output by the human-computer conversation large model; the message object is pushed to the user, and the current conversation is ended; and a preset output condition is set to limit the message update frequency. The application can solve the problem of long reply time in the human-computer conversation system, thereby improving the user interaction experience.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence, and in particular to an output control method, terminal and human-computer dialogue system for human-computer dialogue. Background Technology

[0002] In existing human-computer dialogue systems, the dialogue information input by the user is transferred to a large AI (Artificial Intelligence) model. The AI ​​model then performs calculations and provides corresponding answers. However, when the amount of text is large, it takes a lot of time to process the dialogue information input by the user and provide corresponding answers, resulting in long human-computer interaction time and a low user experience.

[0003] Therefore, how to solve the problem of excessively long response times in human-computer dialogue systems in order to improve the user interaction experience is a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides an output control method for human-computer dialogue, which can solve the problem of excessively long response times in human-computer dialogue systems, thereby improving the user experience. This application also provides an output control terminal and a human-computer dialogue system, which have the same technical effects.

[0005] The first objective of this application is to provide an output control method for human-computer dialogue.

[0006] The aforementioned objective of this application is achieved through the following technical solution:

[0007] A method for output control in human-computer dialogue includes:

[0008] S11. Obtain the dialogue information input by the user and push the dialogue information to the human-computer dialogue model;

[0009] S12. Obtain the response message output by the human-computer dialogue model after receiving the dialogue information in the form of continuous push streaming;

[0010] S13. Determine whether the reply message is the first stream data output by the human-computer dialogue model. If yes, execute S14; otherwise, execute S15.

[0011] S14. Create a message object and determine whether the reply message is the last stream data output by the human-computer dialogue model. If yes, execute S19. If no, add the reply message to the message object, push the message object to the user, and then return to execute S12 again.

[0012] S15. Determine whether the reply message is the last stream data output by the human-computer dialogue model. If yes, execute S19; otherwise, execute S16.

[0013] S16. Determine whether the reply message is the Nth stream data output by the human-computer dialogue model. If yes, execute S19; otherwise, execute S17, where N is a positive integer greater than 1.

[0014] S17. Add the reply message to the message object, and then determine whether the reply message meets the preset output conditions. If yes, execute S18; otherwise, return to execute S12 again.

[0015] S18. Push the message object to the user, and then return to re-execute S12;

[0016] S19. Add the reply message to the message object, then push the message object to the user, and end the current round of dialogue.

[0017] Preferably, in the human-computer dialogue output control method, determining whether the response message meets preset output conditions includes:

[0018] Determine whether the time interval between the reply message and the previous stream of data output by the human-computer dialogue model meets a preset time interval, and whether the cumulative number of reply messages output by the human-computer dialogue model in this round of dialogue is less than a preset number. If yes, the reply message is deemed to meet the output conditions; otherwise, the reply message is deemed not to meet the output conditions.

[0019] Preferably, in the output control method for human-computer dialogue, determining whether the time interval between the reply message and the previous stream of data output by the large-scale human-computer dialogue model meets a preset time interval, and whether the cumulative number of reply messages output by the large-scale human-computer dialogue model in this round of dialogue is less than a preset number, if so, then the reply message is deemed to meet the output conditions, including:

[0020] S21. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the first preset time interval. If yes, execute S22; otherwise, execute S23.

[0021] S22. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the first preset number. If so, execute S29.

[0022] S23. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the second preset time interval. If yes, execute S24; otherwise, execute S25.

[0023] S24. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the second preset number. If so, execute S29.

[0024] S25. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the third preset time interval. If yes, execute S26; otherwise, execute S27.

[0025] S26. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the third preset number. If so, execute S29.

[0026] S27. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the fourth preset time interval. If so, execute S28.

[0027] S28. Determine whether the cumulative number of times the human-computer dialogue model outputs reply messages in this round of dialogue is less than the fourth preset number. If so, execute S29.

[0028] S29. Determine that the reply message meets the output conditions.

[0029] Preferably, in the human-computer dialogue output control method, the first preset time interval is (1, 1.5] seconds; the second preset time interval is (1.5, 3.5] seconds; the third preset time interval is (3.5, 4.5] seconds; and the fourth preset time interval is (4.5, ∞) seconds.

[0030] Preferably, in the human-computer dialogue output control method, the first preset number of times is 5 times; the second preset number of times is 8 times; the third preset number of times is 17 times; and the fourth preset number of times is 20 times.

[0031] Preferably, in the human-computer dialogue output control method, N is 20.

[0032] Preferably, in the human-computer dialogue output control method, the large human-computer dialogue model is an AI large model.

[0033] The second objective of this application is to provide an output control terminal for human-computer dialogue.

[0034] The second objective of this application is achieved through the following technical solution:

[0035] A human-computer dialogue output control terminal includes: a control terminal, which is used to execute instructions to implement any of the above-described human-computer dialogue output control methods.

[0036] The third objective of this application is to provide a human-computer dialogue system.

[0037] The aforementioned objective three of this application is achieved through the following technical solution:

[0038] A human-computer dialogue system includes a client, a processing terminal carrying a large human-computer dialogue model, and a control terminal as described in the above-mentioned human-computer dialogue output control terminal, wherein the client, the control terminal, and the processing terminal are connected in sequence.

[0039] Preferably, in the human-computer dialogue system, the client and the control terminal are connected through a message editing interface.

[0040] The above technical solution acquires user-input dialogue information and pushes it to a large-scale human-computer dialogue model. It then acquires the response messages continuously streamed from the model after receiving the dialogue information. Based on this, by setting judgment and control logic, a message object is created when the acquired response message is the first stream data output by the large-scale human-computer dialogue model. When the acquired response message is the last stream data output by the large-scale human-computer dialogue model or the Nth stream data output by the large-scale human-computer dialogue model, the response message is added to the message object, the message object is pushed to the user, and the current dialogue round ends. Compared to the traditional method of a large-scale human-computer dialogue model outputting all responses at once, this technical solution effectively avoids the problem of excessively long user waiting times caused by the large-scale human-computer dialogue model outputting all responses at once. Furthermore, by setting preset output conditions to limit the message update frequency, it avoids pushing the message object to the user with every stream, further improving the user experience. In summary, the above technical solution can solve the problem of excessively long response times in human-computer dialogue systems, thereby improving the user experience. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 This is a flowchart illustrating an output control method for human-computer dialogue provided in an embodiment of this application;

[0043] Figure 2 This is a schematic diagram of the structure of an output control terminal for human-computer dialogue provided in an embodiment of this application;

[0044] Figure 3 This is a schematic diagram of the structure of a human-computer dialogue system provided in the embodiments of this application. Detailed Implementation

[0045] To enable those skilled in the art to better understand the technical solutions in this application, the technical solutions in the embodiments of this application will be clearly and completely described below. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0046] In the embodiments provided in this application, it should be understood that the disclosed methods and systems can be implemented in other ways. The system embodiments described below are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple modules or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, and can be electrical, mechanical, or other forms.

[0047] It should be understood that the use of terms such as "system," "device," "unit," and / or "module" in this application is merely one method of distinguishing different components, elements, parts, sections, or assemblies at different levels. However, if other terms can achieve the same purpose, they may be replaced by other expressions.

[0048] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, "a plurality of" or "several" means two or more, unless otherwise explicitly specified.

[0049] If a flowchart is used in this application, it is used to illustrate the operations performed by the system according to embodiments of this application. It should be understood that the preceding or following operations are not necessarily performed in exact order. Instead, the steps can be processed in reverse order or simultaneously. Furthermore, other operations can be added to these processes, or one or more steps can be removed from them.

[0050] It should also be noted that, in this document, terms such as “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the article or device that includes the aforementioned element.

[0051] The embodiments in this application are written in a progressive manner.

[0052] like Figure 1 As shown, this application provides an output control method for human-computer dialogue, including:

[0053] S11. Obtain the dialogue information input by the user and push the dialogue information to the human-computer dialogue model;

[0054] In step S11, specifically, dialogue information input by the user can be obtained from the human-computer dialogue system. The input form of the dialogue information can be text, such as user-input queries, questions, casual conversations, etc. This application does not limit the specific input form. The large-scale human-computer dialogue model can adopt existing large-scale AI models, such as Large Language Models (LLMs), etc. This application does not limit this. Compared with traditional AI models, large-scale AI models can solve the problems of overly fragmented and diverse AI models, greatly improving the model's general applicability. Furthermore, large-scale AI models possess self-supervised learning capabilities, which can reduce training and development costs and improve model accuracy. In this step, after obtaining the dialogue information input by the user, the dialogue information can be input into the large-scale human-computer dialogue model.

[0055] S12. Obtain the response message output by the large human-computer dialogue model after receiving the dialogue information in the form of continuous push stream;

[0056] In S12, specifically, by configuring the response method of the large-scale human-computer dialogue model, the model can output response content in a continuous streaming manner. That is, after receiving dialogue information, the large-scale model sequentially outputs one stream of data at a time until the current streaming ends. In this step, the response message output by the large-scale model in a continuous streaming manner after receiving dialogue information can be obtained. Specifically, the response message refers to a single stream of data output by the large-scale model after receiving dialogue information, rather than multiple separate streams. Further, subsequent judgment and control logic is used to control the output of the response message, i.e., the streaming data.

[0057] S13. Determine if the reply message is the first stream data output by the human-computer dialogue model. If yes, execute S14; otherwise, execute S15.

[0058] In S13, specifically, for the obtained response message, that is, a single stream data output by the human-computer dialogue big model, it is determined whether it is the first stream data output by the human-computer dialogue big model in this round of dialogue.

[0059] S14. Create a message object and determine whether the reply message is the last stream data output by the human-computer dialogue model. If yes, execute S19. If no, add the reply message to the message object, push the message object to the user, and then return to execute S12 again.

[0060] In S14, specifically, when the reply message is the first stream data output by the human-computer dialogue model in this round of dialogue, a message object is created. The message object can add reply messages, and it can also update its display in the human-computer interaction interface based on the added reply messages, thus outputting the stream data generated by the human-computer dialogue model. In this step, it is also necessary to determine whether the reply message is the last stream data output by the human-computer dialogue model in this round of dialogue. If so, it can be considered that the current push of the human-computer dialogue model has ended, and S19 is executed, i.e., the reply message is added to the message object, the message object is pushed to the user, and the current round of dialogue ends. If not, it can be considered that the current push of the human-computer dialogue model has not ended, the reply message is added to the message object, the message object is pushed to the user, and then S12 is re-executed to obtain the next individual stream data output by the human-computer dialogue model.

[0061] S15. Determine if the reply message is the last stream data output by the human-computer dialogue model. If yes, execute S19; otherwise, execute S16.

[0062] In S15, specifically, if the reply message is not the first stream data output by the human-computer dialogue model in this round of dialogue, it is further determined whether the reply message is the last stream data output by the human-computer dialogue model in this round of dialogue. If so, it can be determined that the current push of the human-computer dialogue model has ended, and S19 is executed, that is, the reply message is added to the message object, and then the message object is pushed to the user, and the current round of dialogue is ended; if not, S16 is executed.

[0063] S16. Determine if the reply message is the Nth stream data output by the human-computer dialogue model. If yes, execute S19; otherwise, execute S17. Here, N is a positive integer greater than 1.

[0064] In S16, specifically, when the text volume is large, the large-scale human-computer dialogue model outputs the response content in a continuous push-streaming manner, which may still output a large amount of streaming data. If the full amount of streaming data is output, the user may still wait for a long time. In this step, if the response message is not the last piece of streaming data output by the large-scale human-computer dialogue model in this round of dialogue, it is further determined whether the response message is the Nth piece of streaming data output by the large-scale human-computer dialogue model in this round of dialogue. If yes, S19 is executed, that is, the response message is added to the message object, and then the message object is pushed to the user, ending the current round of dialogue. This controls the number of response messages (i.e., the push-streaming data of the large-scale human-computer dialogue model) output to the user to not exceed N, thereby effectively avoiding the problem of excessively long waiting time for the user to receive this streaming data. If not, S17 is executed. Here, N is a positive integer greater than 1, and the value of N can be set based on actual application needs; this application does not impose any restrictions on this. In a specific embodiment, based on the existing processing capacity of the large-scale human-computer dialogue model, N can be 20, thereby achieving a balance between the completeness of the push-streaming of the large-scale human-computer dialogue model and the user's waiting time as much as possible.

[0065] S17. Add the reply message to the message object, and then determine whether the reply message meets the preset output conditions. If yes, execute S18; otherwise, return to execute S12 again.

[0066] In S17, specifically, the large-scale human-computer dialogue model outputs response content through continuous push streaming. If each response is sent to the user, the frequent message updates could lead to a poor user experience. In this step, when the response message is not the Nth stream data output by the large-scale human-computer dialogue model in this round of dialogue, the response message is added to the message object, and then it is determined whether the response message meets the preset output conditions. If yes, S18 is executed, that is, the message object is pushed to the user, and then S12 is re-executed. If not, the message object is not pushed, and S12 is re-executed to obtain the next individual stream data output by the large-scale human-computer dialogue model. In this step, by setting preset output conditions, the message update frequency can be limited to avoid pushing the message object to the user with every push stream, thus further improving the user experience.

[0067] In some embodiments, one implementation of the step of determining whether a reply message meets preset output conditions includes: determining whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model meets a preset time interval, and whether the cumulative number of times the human-computer dialogue model outputs reply messages in this round of dialogue is less than a preset number. If yes, the reply message is deemed to meet the output conditions; otherwise, the reply message is deemed not to meet the output conditions. The preset time interval and preset number of outputs can be set based on actual application needs, and this application does not impose any restrictions on them. Specifically, a timer can be set to record the time interval between the reply message and the previous stream data output by the human-computer dialogue model, and a counter can be set to record the cumulative number of times the human-computer dialogue model outputs reply messages in this round of dialogue. The above implementation steps, based on the preset time interval and preset number of outputs, can effectively limit the message update frequency and ensure the user's interactive experience.

[0068] S18. Push the message object to the user, then return to re-execute S12;

[0069] In S18, specifically, when the reply message meets the preset output conditions, the update can be displayed in the human-computer interaction interface based on the reply message already added in the message object, so as to realize the push update of the human-computer dialogue model. Then, return to re-execute S12 to obtain the next individual stream data output by the human-computer dialogue model.

[0070] S19. Add the reply message to the message object, then push the message object to the user and end the current conversation.

[0071] In S19, specifically, when the obtained reply message is the last stream data output by the human-computer dialogue model in this round of dialogue or the Nth stream data output by the human-computer dialogue model in this round of dialogue, the reply message is added to the message object, and the message object is pushed to the user. Based on the reply message already added to the message object, the update can be displayed in the human-computer interaction interface, and finally the current round of dialogue ends.

[0072] In existing human-computer dialogue systems, the dialogue information input by the user is transferred to a large AI model, which then performs calculations and provides corresponding answers. However, when the amount of text is large, it takes a lot of time to process the dialogue information input by the user and provide corresponding answers, resulting in long human-computer interaction time and a low user experience.

[0073] The above embodiments acquire user-input dialogue information and push it to a large-scale human-computer dialogue model. The model then continuously streams response messages after receiving the dialogue information. Based on this, a judgment and control logic is set up: when the acquired response message is the first stream data output by the large-scale human-computer dialogue model, a message object is created; when the acquired response message is the last stream data output by the large-scale human-computer dialogue model or the Nth stream data output by the large-scale human-computer dialogue model, the response message is added to the message object, the message object is pushed to the user, and the current dialogue round ends. Compared to the traditional method of a large-scale human-computer dialogue model outputting all responses at once, this technical solution effectively avoids the problem of excessively long user waiting times caused by the large-scale human-computer dialogue model outputting all responses at once. Furthermore, by setting preset output conditions to limit the message update frequency, the message object is not pushed to the user with every stream, further improving the user experience. In summary, the above embodiments can solve the problem of excessively long response times in human-computer dialogue systems, thereby improving the user experience.

[0074] In other embodiments of this application, one implementation of the step of determining whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model meets a preset time interval, and whether the cumulative number of reply messages output by the human-computer dialogue model in this round of dialogue is less than a preset number, and if so, determining that the reply message meets the output conditions, includes:

[0075] S21. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the first preset time interval. If yes, execute S22; otherwise, execute S23.

[0076] S22. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the first preset number. If so, execute S29.

[0077] S23. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the second preset time interval. If yes, execute S24; otherwise, execute S25.

[0078] S24. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the second preset number. If so, execute S29.

[0079] S25. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the third preset time interval. If yes, execute S26; otherwise, execute S27.

[0080] S26. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the third preset number. If so, execute S29.

[0081] S27. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the fourth preset time interval. If so, execute S28.

[0082] S28. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the fourth preset number. If so, execute S29.

[0083] S29. Determine that the reply message meets the output conditions.

[0084] In this embodiment, to maintain a good interactive experience under different push times (fast, medium, and slow) of the large-scale human-computer dialogue model, output control logic in S21-S29 is set. The first preset time interval, second preset time interval, third preset time interval, and fourth preset actual interval, as well as the first preset number of times, second preset number of times, third preset number of times, and fourth preset number of times, can be set based on actual application needs, and this application does not impose any restrictions on them. In a specific embodiment, based on the processing capacity of the existing large-scale human-computer dialogue model, the first preset time interval can be set to (1, 1.5] seconds; the second preset time interval can be set to (1.5, 3.5] seconds; the third preset time interval can be set to (3.5, 4.5] seconds; the fourth preset time interval can be set to (4.5, ∞) seconds; the first preset number of times can be set to 5 times; the second preset number of times can be set to 8 times; the third preset number of times can be set to 17 times; and the fourth preset number of times can be set to 20 times, to maintain a good update frequency acceptable to the user.

[0085] like Figure 2 As shown, in another embodiment of this application, a human-computer dialogue output control terminal is also provided, including: a control terminal 10, which is used to execute instructions to implement any of the above-described human-computer dialogue output control methods. The control terminal 10 may be a processor, cloud server, etc., capable of performing various functions and data processing of this application; this application does not impose any limitations on this.

[0086] like Figure 3 As shown, in another embodiment of this application, a human-computer dialogue system is also provided, including a client 20, a processing terminal 21 carrying a large human-computer dialogue model, and a control terminal 22 as described in the above-mentioned human-computer dialogue output control terminal, wherein the client 20, the control terminal 22 and the processing terminal 21 are connected in sequence.

[0087] Specifically, the client 20 can be a smartphone, tablet, or other device used by the user. Through the client 20, the user can input dialogue information. The control terminal 22 can be a processor, cloud server, or other device. It can obtain the dialogue information input by the user from the client 20 and push the dialogue information to the large human-computer dialogue model carried in the processing terminal 21. The processing terminal 21 can be a processor or other device capable of executing various functions and data processing of the large human-computer dialogue model. The control terminal 22 can also obtain the response message output by the large human-computer dialogue model after receiving the dialogue information from the processing terminal 21 in the form of continuous push streaming, and execute instructions to implement any of the above-mentioned human-computer dialogue output control methods.

[0088] In other embodiments of this application, the client 20 and the control terminal 22 can be connected via a message editing interface. Specifically, the message editing interface can be an open API (Application Programming Interface), based on which the content of sent messages can be edited and updated according to set rules.

[0089] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A human-computer dialogue output control method, characterized in that, include: S11. Obtain the dialogue information input by the user and push the dialogue information to the human-computer dialogue model; S12. Obtain the response message output by the human-computer dialogue model after receiving the dialogue information in the form of continuous push streaming; S13. Determine whether the reply message is the first stream data output by the human-computer dialogue model. If yes, execute S14; otherwise, execute S15. S14. Create a message object and determine whether the reply message is the last stream data output by the human-computer dialogue model. If yes, execute S19. If no, add the reply message to the message object, push the message object to the user, and then return to execute S12 again. S15. Determine whether the reply message is the last stream data output by the human-computer dialogue model. If yes, execute S19; otherwise, execute S16. S16. Determine whether the reply message is the Nth stream data output by the human-computer dialogue model. If yes, execute S19; otherwise, execute S17, where N is a positive integer greater than 1. S17. Add the reply message to the message object, and then determine whether the reply message meets the preset output conditions. If yes, execute S18; otherwise, return to execute S12 again. S18. Push the message object to the user, and then return to re-execute S12; S19. Add the reply message to the message object, then push the message object to the user, and end the current round of dialogue.

2. The method as described in claim 1, characterized in that, The step of determining whether the reply message meets the preset output conditions includes: Determine whether the time interval between the reply message and the previous stream of data output by the human-computer dialogue model meets a preset time interval, and whether the cumulative number of reply messages output by the human-computer dialogue model in this round of dialogue is less than a preset number. If yes, the reply message is deemed to meet the output conditions; otherwise, the reply message is deemed not to meet the output conditions.

3. The method as described in claim 2, characterized in that, The determination of whether the time interval between the reply message and the previous stream of data output by the human-computer dialogue model meets a preset time interval, and whether the cumulative number of reply messages output by the human-computer dialogue model in this round of dialogue is less than a preset number, if so, then the reply message is deemed to meet the output conditions, including: S21. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the first preset time interval. If yes, execute S22; otherwise, execute S23. S22. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the first preset number. If so, execute S29. S23. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the second preset time interval. If yes, execute S24; otherwise, execute S25. S24. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the second preset number. If so, execute S29. S25. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the third preset time interval. If yes, execute S26; otherwise, execute S27. S26. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the third preset number. If so, execute S29. S27. Determine whether the time interval between the reply message and the previous stream data output by the human-computer dialogue model belongs to the fourth preset time interval. If so, execute S28. S28. Determine whether the cumulative number of response messages output by the human-computer dialogue model in this round of dialogue is less than the fourth preset number. If so, execute S29. S29. Determine that the reply message meets the output conditions.

4. The method as described in claim 3, characterized in that, The first preset time interval is (1, 1.5] seconds; The second preset time interval is (1.5, 3.5] seconds; The third preset time interval is (3.5, 4.5] seconds; The fourth preset time interval is (4.5, ∞) seconds.

5. The method as described in claim 4, characterized in that, The first preset number of times is 5 times; The second preset number of times is 8 times; The third preset number of times is 17 times; The fourth preset number of times is 20 times.

6. The method according to any one of claims 1-5, characterized in that, The value of N is 20.

7. The method according to any one of claims 1-5, characterized in that, The aforementioned human-computer dialogue model is an AI model.

8. A human-computer dialogue output control terminal, characterized in that, include: A control terminal, the control terminal being configured to execute instructions to implement the method as described in any one of claims 1 to 7.

9. A human-computer dialogue system, characterized in that, It includes a client, a processing terminal carrying a large human-computer dialogue model, and a control terminal as described in claim 8, wherein the client, the control terminal, and the processing terminal are connected in sequence.

10. The human-computer dialogue system as described in claim 9, characterized in that, The client and the control terminal are connected via a message editing interface.

Images