CN113722532A

CN113722532A - Audio discussion guiding method and device and computer equipment

Info

Publication number: CN113722532A
Application number: CN202110996634.7A
Authority: CN
Inventors: 刘博�; 汪沛毅; 李刚
Original assignee: Lenovo Beijing Ltd
Current assignee: Lenovo Beijing Ltd
Priority date: 2021-08-27
Filing date: 2021-08-27
Publication date: 2021-11-30

Abstract

The application provides an audio discussion guiding method, a device and a computer device, wherein in the process of speaking discussion of a plurality of members in any discussion group, the computer device can acquire audio information generated by speaking of each member and audio generation time thereof, so that the discussion participation degree of the corresponding member in a target discussion time length is acquired by counting the audio generation time of speaking of each member in the target discussion time length, thereby timely and accurately detecting the member to be guided with the discussion participation degree meeting the discussion guiding condition, automatically generating guidance statement audio aiming at the member to be guided according to the audio information acquired in the target discussion time length, sending the guidance statement audio to the member to be guided to terminals of each member of the discussion group for outputting, timely guiding the member without speaking/with few speaking to participate in the discussion, more friendly reminding the member to be guided to speak, and informing the member to be guided of the speaking direction, the discussion efficiency and effect are improved.

Description

Audio discussion guiding method and device and computer equipment

Technical Field

The present application relates to the field of communications technologies, and in particular, to an audio discussion guidance method, an audio discussion guidance device, and a computer device.

Background

At present, under a plurality of application scenes such as teaching, meeting, games and the like, a group discussion mode is often adopted for multi-person audio interaction, so that the application requirements are met. Taking a teaching scene as an example, teachers can group students and discuss the students according to the same or different subjects, so that the teaching mode is enriched, the learning interest of the students can be stimulated, and the students can participate in discussion, and the teaching quality and the enjoyment are improved.

However, in the actual discussion process, some members of a discussion group may speak less for some reasons, which goes against the original purpose of group discussion, in order to enable each student to participate in the discussion, a teacher usually observes the group discussion, and after discovering students not participating in the discussion, intervenes in time to designate the student to speak for discussion.

However, when a plurality of discussion groups discuss at the same time, it is difficult for teachers to track and understand the speaking situation of each member in each group, and students who speak less cannot be encouraged to make opinions in time, which affects the teaching effect of group discussion. And the way in which the teacher specifies the student to speak often puts stress on the student's mind, which is not friendly to the student.

Disclosure of Invention

In view of the above, the present application proposes an audio discussion guidance method, which includes:

acquiring audio information and audio generation time from members of the same discussion group;

according to the audio generation time, obtaining discussion participation of the corresponding member in the target discussion time length of the discussion group;

determining the member corresponding to the discussion participation degree meeting the discussion guidance condition as a member to be guided, and generating a guidance statement audio for the member to be guided according to the audio information acquired within the target discussion time length; the guidance statement audio is used for guiding the member to be guided to discuss and speak;

and sending the guide statement audio to the electronic equipment of each member in the discussion group for outputting.

Optionally, the obtaining, according to the audio generation time, discussion participation of the corresponding member in a target discussion duration of the discussion group includes:

according to the audio information and the audio generation time, counting the discussion participation time of the same member in the target discussion time of the discussion group; wherein, the time point corresponding to the target discussion time length is changed along with the change of the discussion content of the discussion group;

and carrying out ratio operation on the discussion participation duration and the target discussion duration to obtain the discussion participation degree of the corresponding member participating in the discussion in the target discussion duration.

Optionally, the counting discussion participation duration of the same member in the target discussion duration of the discussion group according to the audio information and the audio generation time includes:

performing keyword detection on the audio information acquired within the target discussion duration to obtain discussion keywords aiming at the discussion subjects of the discussion group;

and carrying out classification statistics on the audio generation time of the members participating in the discussion by utilizing the correlation between the discussion keywords contained in different audio information respectively to obtain the discussion participation time of the corresponding members in the target discussion time.

Optionally, the performing keyword detection on the audio information acquired within the target discussion time duration to obtain discussion keywords for the discussion topic of the discussion group includes:

obtaining a topic keyword configured aiming at the discussion topic of the discussion group;

and performing keyword detection on the audio information acquired within the target discussion duration by using the topic keywords to obtain discussion keywords aiming at the discussion topics of the discussion group.

Optionally, the performing, by using the topic keyword, keyword detection on the audio information acquired within the target discussion duration to obtain a discussion keyword for a discussion topic of the discussion group includes:

detecting whether each piece of audio information acquired within a target discussion time length contains any topic keyword, and determining the detected topic keyword as a discussion keyword aiming at a discussion topic of the discussion group; alternatively, the first and second electrodes may be,

according to the topic keywords, performing semantic detection on each piece of audio information acquired within the target discussion time length;

and determining the topic keywords contained in the audio information and the synonyms of the topic keywords as discussion keywords aiming at the discussion topics of the discussion group by utilizing semantic detection results.

Optionally, the generating a guidance statement audio for the member to be guided according to the audio information acquired within the target discussion time length includes:

obtaining discussion keywords contained in the audio information obtained in the target discussion time length;

screening guide keywords aiming at the member to be guided from the discussion keywords acquired within the target discussion duration;

calling a guide sentence template matched with the guide keyword;

and generating a guide sentence audio aiming at the member to be guided according to the guide sentence template by using the guide keyword.

Optionally, the screening of the guidance keywords for the member to be guided from the discussion keywords acquired within the target discussion time duration includes at least one of the following implementation manners:

randomly selecting one discussion keyword as a guide keyword of the member to be guided from the discussion keywords acquired in the target discussion time length;

screening the discussion keywords with the most occurrence times from the discussion keywords acquired within the target discussion duration as guide keywords of the member to be guided;

screening the discussion keywords with the highest attention degree of the members to be guided as guide keywords from the discussion keywords acquired within the target discussion duration; wherein the attention degree is determined based on a sight line tracking detection result of the member to be guided.

Optionally, the determining, as a member to be guided, a member corresponding to the discussion participation degree meeting the discussion guidance condition includes:

detecting whether the discussion participation of each of the members in the discussion group within the target discussion time period is less than a discussion threshold;

determining the member corresponding to the discussion participation degree smaller than the discussion threshold value as a member to be guided;

if the number of the members to be guided is multiple, the sending the guidance statement audio to the electronic equipment of each member in the discussion group for output comprises:

and sequentially sending the guide sentence audio of the corresponding member to be guided to the electronic equipment of each member in the discussion group for output according to the discussion participation degree of each member to be guided and a preset time interval.

The present application also proposes an audio discussion guidance apparatus, the apparatus including:

the information acquisition module is used for acquiring audio information and audio generation time of members from the same discussion group;

a discussion participation degree obtaining module, configured to obtain, according to the audio generation time, discussion participation degrees of the corresponding members within a target discussion duration of the discussion group;

the guide statement audio generation module is used for determining the member with the discussion participation degree meeting the discussion guide condition as a member to be guided, and generating a guide statement audio for the member to be guided according to the audio information acquired within the target discussion time length;

and the guide statement audio output module is used for sending the guide statement audio to the electronic equipment of each member in the discussion group for output.

The present application further proposes a computer device, the computer device comprising:

a communication module;

a memory for storing a program for implementing the audio discussion guidance method as described above;

a processor for loading and executing the program stored in the memory to implement the audio discussion guidance method as described above.

The present application also proposes a computer-readable storage medium, on which a computer program is stored, the computer program being invoked for execution by a processor, implementing the audio discussion guidance method as described above.

Therefore, the application provides an audio discussion guidance method, an audio discussion guidance device and a computer device, wherein in the process of speaking discussion of a plurality of members in any discussion group, the computer device can acquire audio information generated by speaking of each member and audio generation time thereof, so as to obtain discussion participation of the corresponding member in a target discussion time period by counting the audio generation time of speaking of each member in the target discussion time period, thereby timely and accurately detecting a member to be guided, the member not actively participating in speaking, of which the discussion participation satisfies discussion guidance conditions, and then, the application can also automatically generate guidance statement audio for the member to be guided according to the audio information acquired in the target discussion time period, send the guidance statement audio to terminals of the members in the discussion group for output, and timely guide members not speaking/having few speaking to participate in the discussion speaking, the method for generating the guidance statement audio output can remind the member to be guided to speak more friendly and inform the member to be guided of the speaking direction, and the discussion efficiency and effect are improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.

Fig. 1 is a schematic system architecture diagram of an optional application scenario applicable to the audio discussion guidance method proposed in the present application;

fig. 2 is a schematic diagram of a system architecture in a teaching scenario suitable for an audio discussion guidance method proposed in the present application;

FIG. 3 is a schematic diagram of a hardware configuration of an alternative example of a computer device suitable for use in the audio discussion guidance method proposed in the present application;

fig. 4 is a flowchart illustrating an alternative example of the audio discussion guidance method proposed in the present application;

fig. 5 is a flowchart illustrating yet another alternative example of the audio discussion guidance method proposed in the present application;

fig. 6 is a flowchart illustrating yet another alternative example of the audio discussion guidance method proposed in the present application;

fig. 7 is a flowchart illustrating yet another alternative example of the audio discussion guidance method proposed in the present application;

fig. 8 is a schematic structural diagram of an alternative example of the audio discussion guidance device proposed in the present application.

Detailed Description

Aiming at the technical problems described in the background technology section, the method and the device for automatically tracking and detecting the speaking condition of each member (such as a student) participating in the discussion are expected to determine few members participating in the discussion and speaking timely and accurately, solve the problem that a teacher cannot take care of each student of each discussion group, and reduce the workload of the teacher; later, the method and the device can automatically generate question audio, namely guidance statement audio, aiming at the member by combining discussion content, timely and correctly guide the member to speak, improve the teaching effect of group discussion, avoid the psychological pressure of the teacher calling the student to speak for the student, and are more friendly to the student who participates in discussion and speaks less.

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention. And only parts related to the related invention are shown in the drawings for convenience of description. In the present application, the embodiments and features of the embodiments may be combined with each other without conflict, and the present application does not describe in detail an example.

And it should be understood that "system", "apparatus", "unit" and/or "module" as used herein is a method for distinguishing different components, elements, parts, portions or assemblies at different levels. However, other words may be substituted by other expressions if they accomplish the same purpose.

It should be noted that, in the description of the present application, unless the context clearly dictates otherwise, words such as "a," "an," and/or "the" described herein do not denote the singular but may also include the plural. In general, the terms "comprises" and "comprising" merely indicate that steps and elements are included which are explicitly identified, that the steps and elements do not form an exclusive list, and that a method or apparatus may include other steps or elements. An element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in the process, method, article, or apparatus that comprises the element.

In the description of the embodiments herein, "/" means "or" unless otherwise specified, for example, a/B may mean a or B; "and/or" herein is merely an association describing an associated object, and means that there may be three relationships, e.g., a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, in the description of the embodiments of the present application, "a plurality" means two or more than two. The terms "first", "second" and the like are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating 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.

Additionally, flow charts are used herein to illustrate operations performed by systems according to embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in the exact order in which they are performed. Rather, the various steps may be processed in reverse order or simultaneously. Meanwhile, other operations may be added to the processes, or a certain step or several steps of operations may be removed from the processes.

Referring to fig. 1, a schematic diagram of a system architecture of an optional application scenario applicable to the audio discussion guidance method provided in the present application is shown, where the application scenario may be a multi-user audio discussion scenario, such as an online teaching scenario. As shown in fig. 1, the system may include: a plurality of terminals 11 and a server 12, wherein:

the terminal 11 may be an electronic device for a user to participate in audio discussion, and may include, but is not limited to, a smart phone, a tablet computer, a wearable device, a smart watch, an Augmented Reality (AR) device, a Virtual Reality (VR) device, an in-vehicle device, a smart home device, a robot, a desktop computer, and the like.

In order to facilitate users to participate in multi-user audio discussion, a communication client supporting an audio communication function, such as currently-used social communication software, or dedicated online teaching software, may be installed in the terminal 11 used by the user, or the user logs in an application platform through a browser, and uses an audio module of the terminal device to implement voice interaction.

It should be understood that, in the audio discussion scene, the types of the communication clients used by the members participating in the same discussion group are the same, so as to ensure that the members can successfully access the virtual room created for the discussion group, and implement multi-user voice interaction, and the present application does not describe in detail how the user uses the communication client to join the discussion group for audio discussion.

In addition, the speaking content of the members participating in the discussion, that is, the audio information output by the members, may be collected and sent to the terminals of other members by the audio collector configured in the terminal 11 itself, or may be sent to the terminals of other members participating in the discussion through the terminals of the members or the communication module provided in the audio collecting device after audio collection is performed by the independent audio collecting device.

The server 12 may be a service device that provides communication services for the communication clients described above, i.e., a communication server that is matched with the communication clients. In practical application, the server 12 may be an independent physical server, may also be a server cluster formed by a plurality of physical servers, and may also be a cloud server capable of implementing cloud computing, which may implement communication with a terminal through a wired network or a wireless network, so as to meet application requirements.

In the embodiment of the application, when a plurality of people need to perform online audio discussion, a corresponding virtual room is usually applied first, and users participating in the discussion are invited to enter the virtual room, so that each terminal entering the virtual room can receive audio information sent by any terminal, and audio interaction is realized. For example, in a teaching scenario, referring to the schematic diagram of the system architecture in the teaching scenario shown in fig. 2, a teacher using a terminal may apply to a server to create a virtual room for online teaching, and after determining relevant information of the virtual room, may notify a student to enter the virtual room to listen to a lecture of the teacher. If the teacher needs to discuss the students in groups, a virtual sub-room may be created for each discussion group, and the communication clients of the students participating in the discussion group are accessed to the virtual sub-room, so that the students entering the virtual sub-room can hear the speech between themselves, during which the teacher may choose to join one or more virtual sub-rooms to know the discussion situation of the students in the corresponding discussion group, and the implementation process of the present application is not described in detail.

The server can monitor the speaking conditions of students in the discussion group and timely discover the students who speak less or even do not participate in speaking by adopting the audio discussion guidance method provided by the application for each discussion group, so that corresponding guidance statement audio is automatically generated by combining the discussion content and is sent to the terminals of the students accessing into the corresponding virtual sub-rooms for output, the teacher supervises and guides the students who speak less to speak, the purpose of teaching quality is improved, the mental stress brought to the students by the teacher who speaks the students can be solved, the problem that the teacher cannot pay attention to the speaking conditions of each student under the condition that the discussion group is more can be solved, and the audio discussion guidance efficiency and reliability are improved.

It should be understood that the system architecture shown in fig. 1 does not constitute a limitation to the application scenario system architecture in the embodiment of the present application, and in practical applications, in different application scenarios, the system architecture may include more devices than those shown in fig. 1, or combine some devices, such as a separate storage device for storing data of each audio information of the discussion utterance, the uploaded related information, and the like, as the case may be, the present application is not listed here.

Based on the system architecture described above, referring to fig. 3, a hardware structure diagram of an alternative example of a computer device suitable for the audio discussion guidance method provided in the present application is shown, where the computer device may be a server or a terminal with certain data calculation capability, and may be determined according to the requirements of an application scenario, and the structure of the computer device is described in the present application by taking the computer device as a server as an example. As shown in fig. 3, the computer device may include at least one communication module 21, at least one memory 22, and at least one processor 23, wherein:

the communication module 21, the memory 22 and the processor 23 may all be connected to a communication bus, so as to implement data interaction therebetween through the communication bus, and a specific implementation process may be determined according to requirements of a specific application scenario, which is not described in detail herein.

The communication module 21 may include a communication module capable of implementing data interaction by using a wireless communication network, such as a WIFI module, a 5G/6G (fifth generation mobile communication network/sixth generation mobile communication network) module, a GPRS module, etc., so as to implement communication with other devices (such as terminals used by members participating in discussion), and meet audio interaction requirements between different devices; the communication module 21 may further include a communication interface, such as a USB interface, a serial/parallel interface, etc., for implementing data interaction between internal components of the computer device, and the present application does not limit the specific content included in the communication module 21.

The memory 22 may be used to store a program for implementing the audio discussion guidance method proposed in the present application; the processor 23 may be configured to load and execute the program stored in the memory 21 to implement the audio discussion guidance method proposed in the present application, and the implementation process may refer to, but is not limited to, the description of the corresponding parts of the following method embodiments.

In the present embodiment, the memory 22 may include a high speed random access memory, and may also include a non-volatile memory, such as at least one magnetic disk storage device or other volatile solid state storage device. The processor 23 may be a Central Processing Unit (CPU), an application-specific integrated circuit (ASIC), a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA), or other programmable logic devices. The present application is not limited as to the type of devices that the computer device contains, memory 22 and processor 23, as the case may be.

It should be noted that the structure of the computer device shown in fig. 3 does not constitute a limitation to the computer device in the embodiment of the present application, and in practical applications, the computer device may include more components than those shown in fig. 2, or some components may be combined, and the present application is not listed here.

With reference to fig. 4, a flowchart of an alternative example of the audio discussion guidance method proposed in the present application is shown in conjunction with the system architecture and the computer device structure described in the above embodiment, where the method can be applied to any multi-user audio discussion group in any discussion scenario, and the following embodiment will be described with reference to a student discussion group in a teaching scenario as an example. The audio discussion guiding method can be executed by a computer device, the application takes the computer device as a server, that is, a scene in which the audio discussion guiding method is executed by the server as an example for explanation, and the application is not exhaustive for similar implementation processes of other discussion scenes and other types of computer devices. As shown in fig. 4, the method may include:

step S11, obtaining audio information and audio generation time from members of the same discussion group;

in a scene of audio discussion by multiple persons, during the speaking process of any member in the discussion group, a local audio collecting device (such as an audio collector in a terminal used by the member, or an independent device different from the terminal, or the like) may collect audio information of the member, record the generation time of the audio information as audio generation time, and then send the audio information and the audio generation time of the member to a server. The data transmission method between the audio acquisition device and the server is not limited in the present application, and may be determined according to a corresponding communication method.

In an actual scenario, because the discussion time of the discussion group is often limited, members in the discussion group often speak intensively in a relatively short time (such as 5 minutes, 10 minutes, 15 minutes, and the like), and each member can collect audio information of the member and record audio generation time and send the audio information to the server, so that the server can record audio information of different members in the discussion group at different audio generation times.

In a possible implementation manner, in order to facilitate subsequent statistics of discussion participation of each member in a discussion group within a certain time period, audio information of each member may be sequentially recorded according to a sequence of audio generation time, but the present invention is not limited to this recording method, and may also perform classification recording according to the members, and the like, and detailed description thereof is not provided herein.

Step S12, according to the audio generation time, obtaining the discussion participation of the corresponding member in the target discussion time of the discussion group;

in the embodiment of the present application, the target discussion time duration may be a preset continuous time duration, and may be determined according to, but not limited to, a total expected discussion time duration of the discussion group, and in order to be able to find a member that is not actively involved in the discussion and speaking in time, the value of the preset target discussion time duration may be a relatively small value, such as 5 minutes, 10 minutes, and the like, and is generally a time duration from a certain historical time to a current time.

Because the discussion utterances of the members of the discussion group at different time points and for different contents are uncertain and have great randomness or subjectivity, the discussion participation situations of the same member may be different in different discussion time periods in the whole discussion process, that is, the discussion participation degrees of the same member in different discussion time periods may be different, and the discussion participation degree may represent the participation situation of the member in the discussion time period, and may be determined according to the utterance contents, utterance durations and other information of the member.

Based on this, in order to find out the members who do not actively participate in the discussion in time, the present application may count the audio generation time of the received audio information of each member within the target discussion time, determine the discussion participation time of each member within the target discussion time, and accordingly calculate the discussion participation degree of the corresponding member within the target discussion time. As described above, since the time point corresponding to the target discussion time length may change with the change of the discussion content of the discussion group, the calculated discussion participation time lengths of the same member may be different in different target discussion time lengths, that is, the discussion participation degree of the member a may be higher in a certain target discussion time length, but the discussion participation degree of the member a may be lower in another target discussion time length, which may be determined by the circumstances.

Step S13, determining the members with the discussion participation degree meeting the discussion guidance condition as the members to be guided, and generating guidance statement audios aiming at the members to be guided according to the audio information acquired within the target discussion time length;

as can be seen from the above description of the discussion participation degree, in the same discussion time length, a member with a higher discussion participation degree indicates that the discussion speaking aggressiveness of the member in the discussion time length is higher, for example, the total speaking time length is larger; conversely, the lower the discussion speaking aggressiveness of the member in the discussion duration, for example, the smaller the total speaking duration, the member may be considered not to participate in the discussion actively, and a reminder may be needed to guide such member to participate in speaking. Therefore, the discussion guidance condition for judging whether the member is the member to be guided or not can be configured according to the discussion participation of the member in the target discussion time length, the content of the discussion guidance condition is not limited by the application, and the discussion guidance condition can be determined by combining the requirement of the discussion scene.

Based on this, in the process of discussing each member of the discussion group, after the discussion participation of each member in the target discussion time length is obtained according to the above manner, whether the obtained discussion participation meets the discussion guidance condition or not is respectively detected, and if so, the member corresponding to the discussion participation can be determined as the member to be guided; if not, the corresponding member can be considered to speak relatively actively in the target discussion time length, no external intervention is needed to guide the member to speak, and no processing can be performed.

In some embodiments, to encourage a member who is actively speaking to continue active participation, discussion reward information, such as a verbal reward, may be generated for such member based on audio information and discussion engagement information of such member; the server can send the discussion reward information to the terminal of the corresponding member for output, and also can send the discussion reward information to the terminal of each member participating in the discussion group for output, and the like.

In the embodiment of the present application, to guide such members to discuss and speak, and increase the discussion participation degree thereof, in order to determine the members to be guided within the target discussion time length, that is, one or more members in the corresponding discussion group, in order to guide such members to discuss and speak, a corresponding guidance sentence audio, that is, audio information for guiding the members to discuss and speak, may be generated in combination with the audio information content of each member within the target discussion time length.

For a plurality of members to be guided determined within the same target discussion time length, guidance phrase audio generated by the server for each member to be guided may be the same or different, which may be determined according to a generation manner of the corresponding guidance phrase audio, which is not limited in the present application.

And step S14, sending the guide sentence audio to the electronic equipment of each member in the discussion group for output.

In conjunction with the above description of the system embodiment, for the terminals of each member of the same discussion group, the same virtual room created by the server may be accessed to implement data interaction between the terminals of each member, for example, implement synchronous output of audio information for each member participating in discussion speaking, and the like, and the implementation process is not described in detail in this application. Based on this, the server detects a member needing to be guided to speak within the target discussion time length, that is, a member to be guided, and after obtaining the guidance statement audio for the member to be guided, the server can send the guidance statement audio to each terminal accessing the virtual room corresponding to the discussion group, and the guidance statement audio is played by an audio player of the terminal or an independent audio player.

In some embodiments, if the members participating in the discussion group are located in the same space, for example, if the discussion is performed below the line, the audio information of each member may be collected by an audio collecting device disposed in the space, and sent to a server or an electronic device with certain data processing capability, and then after the guidance statement audio of the member to be guided is obtained in the above manner, the guidance statement audio may be directly or indirectly sent to an audio playing device in the space for playing. Therefore, in this scenario, each member participating in the discussion group may share an audio acquisition device and an audio playing device, and the server may directly perform communication connection with the audio acquisition device and the audio playing device, or may use a terminal with a communication function as a transit device located in the space, so as to achieve indirect communication connection between the server and the audio acquisition device and the audio playing device, which is not limited in this application.

In summary, in the course of speaking discussion by a plurality of members included in any discussion group, a computer device (e.g. a server) may obtain audio information generated by speaking of each member and audio generation time thereof, so as to obtain discussion participation of the corresponding member in a target discussion time period by counting the audio generation time of speaking of each member in the target discussion time period, thereby timely and accurately detecting a member to be guided whose discussion participation satisfies a discussion guidance condition, i.e. a member not actively participating in speaking, and then, the present application may also automatically generate guidance statement audio for the member to be guided according to the audio information obtained in the target discussion time period, send the guidance statement audio to a terminal of each member of the discussion group to output, and timely guide the member not speaking/having little speech to participate in the discussion, thereby not only solving the participation situation of each member in a manually supervised discussion group (e.g. a teacher supervises different student discussion groups in a teaching scene), the technical problem that members who do not speak or speak rarely can not be found in time, and compared with a processing mode that a person directly clicks a roll to call a member to be guided to speak, the mode for generating the guidance statement audio output can remind the member to be guided to speak more friendly, and meanwhile, the member to be guided can be informed of the speaking direction, so that the discussion efficiency and the discussion effect are improved.

Referring to fig. 5, a flow chart of another optional example of the audio discussion guidance method proposed in the present application is illustrated, and the present application embodiment may be an optional detailed implementation method of the audio discussion guidance method described in the above embodiments, but is not limited to such a detailed implementation method, and the method may be executed by a computer device, and still be described as an example executed by a server, as shown in fig. 5, and the method may include:

step S21, obtaining audio information and audio generation time from members of the same discussion group;

step S22, according to the audio information and the audio generation time, counting the discussion participation time of the same member in the target discussion time of the discussion group;

the server may accumulate the audio generation time of the audio information of the same member acquired within the target discussion time length according to the time sequence to obtain the discussion participation time length of the corresponding member within the target discussion time length.

Illustratively, for a discussion about a subject of a member A, a member B, a member C and a member D of a discussion group, if the member A speaks audio information 1 in a ratio of 8:32: 30-8: 33:48 and the member B speaks audio information 2 in a ratio of 8:37: 05-8: 39:49 and the member A speaks audio information 3 in a ratio of 8:40: 16-8: 43:25 and the member C speaks audio information 4 in a ratio of 8:43: 49-8: 45:09 and the member B speaks audio information 5, and the discussion participation duration of the member A is 3 minutes, the discussion participation duration of the member B is 4 minutes, the discussion participation duration of the member C is 3 minutes, and the discussion participation duration of the member D is 0. It will be appreciated that the accuracy of the time duration of participation in the discussion may also be consistent with the audio generation time, e.g., to seconds in the above example, or the time accuracy of the two may be different, and the present application is not limited thereto, as appropriate.

Step S23, carrying out ratio operation on the discussion participation duration and the target discussion duration to obtain the discussion participation degree of the corresponding member participating in the discussion in the target discussion duration;

still by way of example above, for the discussion participation degrees of the member a and the member C within the target discussion time period of 8: 30-8: 45 (i.e. the last 15 minutes), the discussion participation degree may be 20% in a percentage manner for the corresponding discussion participation time period/target discussion time period, i.e. 3 minutes/15 minutes ═ 0.2; similarly, the discussion participation degree for member B in the target discussion time period is 4 minutes/15 minutes is 0.267, i.e. 26.7%; and the discussion participation of the member D in the target discussion time length is 0, namely, the member D does not speak in the target discussion time length.

It should be noted that, the calculation method and the representation method of the discussion participation of the members of the discussion group in the target discussion time length include, but are not limited to, the above-described method.

Step S24, detecting whether each member in the discussion group has discussion participation degree smaller than the discussion threshold value in the discussion participation degrees in the target discussion time length, if yes, entering step S25; if not, returning to the step S22, and continuing to monitor the discussion participation duration of the members;

step S25, determining the member with the discussion participation degree smaller than the discussion threshold value as the member to be guided;

step S26, obtaining discussion keywords contained in the audio information obtained in the target discussion time length;

in combination with the above description of the technical concept of the present application, for a member who does not speak or has a small number of utterances in the target discussion time period, that is, a member to be guided, the present application is expected to not only remind the member of the category to speak actively, but also inform the member of the discussion content of other members in the target discussion time period before, and guide the member to speak continuously from which aspect, especially for the member to be guided who leaves for a short time or has no concentration, and therefore does not hear the discussion content of other members in the target discussion time period, the present application is an idea of guiding the utterance, and can guide the member to speak more friendly and effectively.

Based on this, in order to generate the guidance sentence audio for the member to be guided, the present application may extract the discussion keywords included in the audio information of each member received within the target discussion time duration, such as the keywords related to the purpose of establishing the discussion group this time.

Step S27, screening guide keywords aiming at the members to be guided from the obtained discussion keywords in the target discussion duration;

since the obtained content of the discussion keyword can represent the discussion content of the corresponding member, in order to more effectively guide the member to be guided to speak and inform the member to be guided of what content to speak, the present application proposes to filter out one or more guidance keywords from the discussion keywords obtained within the target discussion duration, that is, from the discussion keywords of the discussion group within the recent time period, so as to indicate the recent discussion content of the discussion group. It should be noted that, the number of the guide keywords and the screening method thereof are not limited in the present application, and may be determined according to the circumstances.

Step S28, a guide sentence template matched with the guide keyword is called;

step S29, generating a guide sentence audio for the member to be guided according to the guide sentence template by using the guide keyword;

and step S210, sending the guide statement audio to the electronic equipment of each member in the discussion group for output.

For different contents or different categories of guide keywords, the application can pre-configure a corresponding guide sentence template, for example, the guide keyword is "Mr. eight character set", the guide sentence template can include "Please say XXXXXX you know", and the guide sentence generated by using the guide sentence template for the guide keyword can be "Please say Mr. eight character set Bar you know! "to clarify the member to be guided, the audio of the guide sentence for the member to be guided in the present application may be" (identified by the name of the member to be guided, etc.) hello, say, mr. eight-mr.'s text, college bar you know! ".

It should be understood that, for different keywords, one or more guide sentence templates matched with the different keywords may be configured in advance, so that in practical application, the guide sentence template used at this time may be determined by combining a plurality of guide keywords of a discussion scene or a screening.

In addition, for the content of the guide sentence audio generated by the member to be guided, the guide keyword and the guide sentence template can be directly fused, in order to point out the member to be guided, the method and the device can add the identification such as the name of the member to be guided and generate the guide sentence audio, so that the member of the discussion group can definitely indicate which member needs to speak, and the output content of the guide sentence audio is enriched, and meanwhile, the discussion efficiency can be improved.

After the guidance keywords and the guidance sentence template matched with the guidance keywords are determined, the guidance sentence for guiding and indicating the member to be guided to speak is not limited to audio information, and a guidance sentence text for the member to be guided is generated according to the guidance sentence template by using the guidance keywords and sent to the electronic devices of the members in the discussion group for display.

In summary, in the embodiment of the present application, by monitoring the speeches of the members of the discussion group, counting the discussion participation duration of the speeches of the members, obtaining the discussion participation degree in the latest target discussion duration in time, and detecting the member to be guided whose discussion participation degree is smaller than the discussion threshold value, i.e., members who have not recently spoken, or have spoken very little, relative to manually monitoring the speaking situation of the members of the discussion group, the timeliness and the reliability of finding the member to be guided are improved, and then, the method screens the guiding key words from the key words recently spoken by the member, according to the matched guide statement template, a guide statement audio aiming at the member to be guided is generated and sent to each member terminal of the discussion group for output, so that the member to be guided can be guided to speak more friendly and effectively, the purpose of building the discussion group can be well achieved, and the teaching quality can be improved in a teaching scene.

Referring to fig. 6, a flowchart of a further optional example of the audio discussion guidance method proposed in the present application is shown, and the present application embodiment may be a further optional detailed implementation method of the audio discussion guidance method described in the above embodiments, and still take an implementation process of the audio discussion guidance method executed by the server as an example. As shown in fig. 6, the method may include:

step S31, obtaining audio information and audio generation time from members of the same discussion group;

step S32, carrying out keyword detection on the audio information acquired within the target discussion duration to obtain discussion keywords aiming at the discussion subjects of the discussion group;

in practical applications, before the discussion group is usually constructed, the discussion topic is determined, for example, for "what contributions are made to chinese literature by historical occurrences of eight nationalities of down and song? The topic is discussed, one or more discussion groups for the discussion topic may be constructed, and discussion group members may speak for the discussion topic, so that the content of the audio information of each member of the discussion group received by the server is about the discussion topic, and keyword detection is performed on the audio information, and the obtained discussion keywords are keywords about the discussion topic, such as one or more discussion keywords in a plurality of keywords configured for the discussion topic in advance, or discussion keywords that are synonyms/synonyms with pre-configured keywords, and the like.

Step S33, using the correlation between the discussion keywords contained in different audio information to classify and count the audio generation time of the members participating in discussion, and obtaining the discussion participation time of the corresponding members in the target discussion time;

as the discussion progress of the discussion group advances, the discussion content between the members of the discussion group may change, for example, discussion is performed at different angles with respect to the same discussion topic, so that the content included in the audio information of the members may be at different angles in different discussion time durations, and in order to accurately and timely detect the member to be guided in the current target discussion time duration, the present application may count the discussion participation of each member according to finer-grained sub-topics such as different angles of the discussion topic.

Based on this, for the discussion keywords detected from the audio information acquired within the target discussion time, the present application may further detect the correlation between these discussion keywords, classify the speaking members accordingly, for example, temporarily divide the members speaking for the same sub-topic into a class within the target discussion time, count the audio generation time of the speaking of each member in the class, and determine the total time of the speaking of each member for the sub-topic as the discussion participation time of the corresponding member within the target discussion time, but is not limited to this discussion participation time counting method.

Exemplarily, what contributions are made to chinese literature by the historical occurrences of eight down and sons families, still described above? "this discussion topic, keywords that may be involved may include down dynasty, calligraphy, painter, normal style drowning, prose, song dynasty, euonymus repair, susan, surut, literature, historical background, ancient sports, ancient innovation, mr. eight, liriod, wangsu, agalma, ancient great, etc., and different angles of topics may be discussed, such as dynasty, historical background, works, etc., and the discussion may be performed step by step, that is, the discussion may be performed for dynasty characters of eight during a certain discussion time period, the discussion may be performed for historical backgrounds of eight during a certain discussion time period, the discussion may be performed for works of eight during a certain discussion time period, etc., as the case may be, and the present application is not limited thereto.

Based on this, it can be understood that, for different angles of the above discussion topics, keywords that may be involved are often different, so that the present application may determine which one or more sub-topics are discussed for the target discussion time duration by the correlation of each detected discussion keyword, for example, the discussion keywords detected from the audio information of different members include korean willow, wangshi, great consolidation, song, eulogy, sudang, tang, and paradised, and the like, the discussion time duration may be considered to be discussed for dynasty and people, and then, the audio generation time of the audio information related to such keywords may be classified and counted according to the member categories to obtain the discussion participation time duration of each member in the target discussion time duration.

As the discussion progress advances, it is detected that the discussion keywords include calligraphy, painters, mr. eight, prose, and the like, and it may be considered that the discussion is performed on the aspects of works and speciality within the discussion time length, according to the above classification statistical manner, the audio generation time of the member speaking on the aspect may be counted, and the discussion participation time length of each member within the discussion time length and the aspects of works and speciality is determined.

Step S34, carrying out ratio operation on the discussion participation duration and the target discussion duration to obtain the discussion participation degree of the corresponding member participating in the discussion in the target discussion duration;

step S35, determining the members with the discussion participation degree meeting the discussion guidance condition as the members to be guided, and generating guidance statement audios aiming at the members to be guided according to the audio information acquired within the target discussion time length;

and step S36, sending the guide sentence audio to the electronic equipment of each member in the discussion group for output.

Regarding the implementation process of step S34 to step S36, reference may be made to the description of the corresponding parts in the above embodiments, which is not repeated in this embodiment.

In summary, in the embodiment of the present application, the obtained audio information of each member speaking is subjected to keyword detection, and the discussion keyword for the discussion topic in the discussion group is determined, so that by using the correlation between the discussion keywords, a finer-grained classification statistics on the audio generation time of each member participating in the discussion is achieved, the discussion participation time of the corresponding member for the current sub-topic within the target discussion time length is obtained, and further the discussion participation degree of each member for the current sub-topic is obtained by calculation, and accordingly, the member to be guided who speaks little or does not speak for the current sub-topic is timely and accurately identified, the guidance sentence audio for guiding the member to speak is generated in combination with the discussion keyword for the current sub-topic, and is output in the discussion group to effectively and friendly guide such member to speak for the current sub-topic, the problem that the member to be guided does not know what to speak is solved.

According to fig. 7, a schematic flow diagram of another optional example of the audio discussion guidance method provided in the present application is provided, where an embodiment of the present application may be another optional detailed implementation method of the audio discussion guidance method described in the above embodiment, for example, how to implement the method for detecting the keyword of the audio information in the above embodiment is detailed, but not limited to the keyword detection implementation method described in the present embodiment, and regarding other implementation steps of the audio discussion guidance method, reference may be made to descriptions of corresponding parts in the above embodiment, and this embodiment is not described again. As shown in fig. 7, the method may include:

step S41, obtaining the topic key words configured for the discussion topic of the discussion group;

in conjunction with the description in the corresponding part of the above embodiments, after determining the discussion topic, one or more topic keywords may be configured in advance for the discussion topic to guide or prompt each member participating in the discussion to speak.

For example, in a teaching scenario, after a teacher groups students for a specified discussion topic, several topic keywords about the discussion topic may be given for the convenience of the students to discuss. In practical application, a teacher may complete grouping of students, assignment of discussion topics, configuration of topic keywords of each discussion topic, and the like on a terminal used by the teacher, and send the information to a server, so that the server constructs a corresponding virtual room according to the information submitted by the teacher terminal, invites student terminals of corresponding discussion groups to access the virtual room, and sends related information about the discussion topics to the student terminals, so that the students determine whether to join the discussion groups according to the related information, or the students participating in discussion of the discussion groups speak with reference to the topic keywords, and the implementation process of the present application is not described in detail.

It should be noted that, the present application does not limit the implementation method for constructing different discussion groups, the discussion topics of each discussion group, and the implementation process for configuring the topic keywords thereof, and reference may be made to, but is not limited to, the teacher configuration method described above, and the configuration information of the corresponding discussion group, such as the discussion topics and the topic keywords thereof, may also be automatically generated according to the discussion requirements submitted by the teacher and according to algorithms such as machine learning/deep learning in artificial intelligence, and the implementation process is not described in detail in the present application.

Step S42, detecting whether each audio information obtained in the target discussion time length contains any topic keyword, and determining the detected topic keyword as the discussion keyword aiming at the discussion topic of the discussion group;

in practical application of the method, keyword detection can be performed on the audio information acquired within the target discussion time duration by using the topic keywords, so as to obtain discussion keywords for discussion topics of the discussion group, and based on the keyword detection concept, any topic keyword detected from a plurality of pieces of audio information acquired within the target discussion time duration can be determined as the discussion keyword for the discussion topic of the discussion group, namely, the discussion keyword of a member of the discussion group within the target discussion time duration.

In practical application, for the selection of the discussion keyword, the random selection manner described above may be adopted, or the frequency of each detected topic keyword may be counted, and the topic keyword with the highest frequency is determined as the discussion keyword; the determined nearest topic keyword can also be directly selected as a discussion keyword, and the like, and the application does not limit how to select the discussion keyword within the target discussion time length, and includes but is not limited to several implementation methods described herein.

Step S43, screening guide keywords aiming at the members to be guided from the obtained discussion keywords in the target discussion duration;

the number of the discussion keywords determined in the above manner is one, and the discussion keywords can be directly determined as the guide keywords for the member to be guided; if the number of the discussion keywords is multiple, the guidance keywords for the member to be guided can be screened from the discussion keywords according to a preset screening rule.

In some embodiments, one discussion keyword can be randomly selected from the discussion keywords acquired within the target discussion duration as a guide keyword of a member to be guided; the discussion keywords with the most occurrence times can be screened as the guide keywords of the members to be guided; the discussion keywords with the highest attention degree of the members to be guided can be screened as the guide keywords; the attention degree can be determined based on the sight tracking detection result of the member to be guided, and the sight tracking detection method of the member to be guided is not described in detail in the application.

Step S44, a guide sentence template matched with the guide keyword is called;

and step S45, generating a guide sentence audio aiming at the member to be guided according to the guide sentence template by using the guide keyword.

Regarding the implementation processes of step S44 and step S45, reference may be made to the description of the corresponding parts in the above embodiments, which is not repeated in this embodiment.

In summary, in the embodiment of the present application, the topic keywords configured for the discussion topic may be utilized to detect whether the speaking content of the discussion group member includes the topic keywords, determine the included topic keywords as the discussion keywords, then screen out one or more guidance keywords for the member to be guided, and generate, according to the guidance sentence template, a guidance sentence audio for guiding the member to be guided to speak, so as to ensure the validity of the guidance sentence, thereby encouraging each member of the discussion group to speak actively and speeding up the discussion progress.

In still other embodiments provided by the present application, in the process of determining the discussion keyword, the present application may further perform semantic detection on each audio information acquired within the target discussion time duration according to each acquired topic keyword preconfigured for the discussion topic of the discussion group, so as to determine, by using a semantic detection result, a topic keyword included in the audio information and a synonym of the topic keyword as the discussion keyword, and then, may continue to screen the guide keyword according to the above-described method, and generate the required guide sentence audio.

Therefore, the semantic detection method implemented by the artificial intelligence technology provided by the embodiment of the application improves the reliability and accuracy of the detection of the discussion keyword, and solves the technical problem that in the subject keyword comparison method, the member cannot be detected because the member says the synonym or the synonym of the subject keyword, so that the reliability and accuracy of the generated guide sentence audio are reduced. It should be noted that the implementation process of the semantic detection method is not described in detail in this application.

In still other embodiments provided by the present application, based on the description of the foregoing embodiments, if the number of the determined members to be guided is multiple, the present application may sequentially send the guidance sentence audio of the corresponding member to be guided to the electronic devices of the members in the discussion group for output according to the respective discussion participation degrees of the multiple members to be guided and according to the preset time interval. Optionally, the guidance sentence audio is constructed and sent to the electronic device of each member for output only for the member to be guided with the minimum discussion participation degree, which is not limited in the present application and may be determined according to the situation.

Based on the audio discussion guidance method described in the above embodiments, the following scenario will be described by taking a teaching scene as an example. In the process of online teaching of teachers to a plurality of students in a certain class, in order to improve teaching quality, the method proposes to make students discuss in groups according to a certain discussion topic, and still makes a history of which contributions are brought to Chinese literature by the appearances of eight families, down and song? The main topic is described as an example, assuming that students are divided into two discussion groups, referring to the system architecture diagram in the teaching scenario shown in fig. 2, a terminal of a teacher may request a server to construct a corresponding virtual room for different discussion groups, and access each student terminal of a corresponding discussion group member to the virtual room. In order to facilitate the teacher to understand the discussion situation of each discussion group, the teacher terminal can be accessed into each virtual room, and the implementation process is not described in detail in this application.

The server can send the discussion subjects of each discussion group distributed by the teacher and the information such as the subject keywords configured for the discussion subjects to the terminals of the corresponding virtual rooms for displaying, so that students of the discussion group can start to discuss accordingly. For any discussion group, in the speaking process of any participating student, an audio collector of the student terminal or an independent audio collecting device can collect audio information of the student, determine audio generation time, and send the audio information and the audio generation time thereof, and speaking student identification (such as student name, unique identification of the student using the terminal, and the like) to the server.

Then, the data analyzer in the server may analyze and record the speech content of each student in the discussion group, for example, according to the method described in the above embodiment, determine which students participate in the discussion speech in different discussion durations, and what the discussion keyword of the speech is, so as to calculate the discussion participation degree of the corresponding student according to the duration of the speech. For example, taking the example that the discussion group includes four members, i.e., student a, student B, student C, and student D, the statistical discussion record table of each student may be:

TABLE 1

Participating members	Time period of participation in discussion	Keywords participating in discussion content
			Student A	0-5 min	Tang dynasty, calligraphy and painter
Student B	0-5 min	Is free of
			Student C	0-5 min	Tang dynasty, the greater energizer being in the middle of the body and the abdomen being in the middle of the body
Student D	0-5 min	Is free of
			Student A	5-10 minutes	Song dynasty, Ouyang and Su
Student B	5-10 minutes	Innovation of Tang dynasty, ancient sports and ancient renovation
			Student C	5-10 minutes	Song dynasty, Mr. eight character corpus
Student D	5-10 minutes	Is free of
			Student A	10-15 minutes	Susan, surut, literature and history background
Student B	10-15 minutes	Eight Mr. Wenjin, Korean willow, Wang' an stone, Zeng Gu and Ouyang Xian
			Student C	10-15 minutes	Is free of
Student D	10-15 minutes	Is free of

As shown in table 1 above, the present embodiment may count the discussion participation time of the student every 5 minutes, and the actual discussion participation time of each member is not shown in table 1. Based on the above statistics, the discussion participation of each student in different target discussion time periods may be calculated according to, but not limited to, the calculation manner of discussion participation duration/target discussion time period, for example, student B participates in 10 minutes, and within the target discussion time period of the last 15 minutes, the discussion participation of student B may be 10/15 ═ 0.667, that is, 66.7%, so that the discussion participation of each member within the target discussion time period of the last 15 minutes shown in table 2 may be calculated:

TABLE 2

Participating members	Duration of participation in discussion	Discussion engagement
			Student A	15 minutes	100％
Student B	10 minutes	66.7％
			Student C	10 minutes	66.7％
Student D	0 minute	0％

The data analyzer of the server obtains the discussion participation of each student in the same discussion group within the target discussion duration, and then sends the discussion participation to the task trigger, and by detecting whether the discussion participation smaller than the discussion threshold, such as smaller than 5%, exists, the corresponding student can be determined as a member to be guided, such as student D, and trigger a guidance question for the student D, and specifically, a guidance statement template for the student D can be generated in the manner described above, such as "please say that you know about eight mr. newsletters corpus bar |)! ", the question wizard may combine it with student D to get" D classmate your good, please say one says you know about the eight Mr. Buck! The guidance sentence audio of the content is sent to each student terminal and teacher terminal of the discussion group to be played, so that the student D is encouraged and guided to speak for the eight-Mr. character set.

Therefore, compared with the method that the teacher directly calls the roll to speak, the audio discussion guiding method is more friendly to the student D; under the scene that a plurality of discussion groups go on simultaneously, solved the teacher and hardly tracked and know every student's speech, can't guarantee that every student who participates in the discussion group can be from the problem of having a certain amount of results, help improving the teaching quality.

Referring to fig. 8, a schematic structural diagram of an alternative example of an audio discussion guidance device proposed in the present application may include:

an information acquisition module 31 for acquiring audio information and audio generation time from members of the same discussion group;

a discussion participation degree obtaining module 32, configured to obtain, according to the audio generation time, a discussion participation degree of the corresponding member within a target discussion duration of the discussion group;

a guidance statement audio generating module 33, configured to determine a member whose discussion participation meets a discussion guidance condition as a member to be guided, and generate a guidance statement audio for the member to be guided according to the audio information acquired within the target discussion time length;

and the guide statement audio output module 34 is used for sending the guide statement audio to the electronic equipment of each member in the discussion group for output.

In some embodiments, the discussion engagement obtaining module 32 may include:

the discussion participation duration counting unit is used for counting the discussion participation duration of the same member in the target discussion duration of the discussion group according to the audio information and the audio generation time; wherein, the time point corresponding to the target discussion time length is changed along with the change of the discussion content of the discussion group;

and the discussion participation degree calculating unit is used for carrying out ratio operation on the discussion participation duration and the target discussion duration to obtain the discussion participation degree of the corresponding member participating in the discussion in the target discussion duration.

Optionally, the above discussion participation duration statistic unit may include:

a discussion keyword obtaining unit, configured to perform keyword detection on the audio information obtained within the target discussion duration to obtain a discussion keyword for a discussion topic of the discussion group;

and the classification statistical unit is used for performing classification statistics on the audio generation time of the members participating in the discussion by utilizing the correlation between the discussion keywords contained in different audio information respectively to obtain the discussion participation time of the corresponding members in the target discussion time.

In practical applications, the above discussion keyword obtaining unit may include:

a topic keyword obtaining unit, configured to obtain a topic keyword configured for a discussion topic of the discussion group;

and the keyword detection unit is used for carrying out keyword detection on the audio information acquired within the target discussion time length by utilizing the topic keywords to obtain discussion keywords aiming at the discussion topics of the discussion group.

In a possible implementation manner, the keyword detection unit may include:

a first detecting unit, configured to detect whether each piece of audio information acquired within a target discussion duration includes any one of the topic keywords, and determine the detected topic keyword as a discussion keyword for a discussion topic of the discussion group

In another possible implementation manner, the keyword detection unit may further include:

the semantic detection unit is used for carrying out semantic detection on each piece of audio information acquired within the target discussion time length according to the topic keywords;

a first determining unit, configured to determine, by using a semantic detection result, the topic keyword included in the audio information and a synonym of the topic keyword as a discussion keyword for a discussion topic of the discussion group.

In still other embodiments of the present application, the guide sentence audio generating module 33 may include:

the discussion keyword acquisition unit is used for acquiring discussion keywords contained in the audio information acquired within the target discussion time length;

the keyword screening unit is used for screening the guide keywords aiming at the member to be guided from the discussion keywords acquired within the target discussion time length;

a guide sentence template calling unit for calling a guide sentence template matched with the guide keyword;

and the guide sentence audio generating unit is used for generating a guide sentence audio aiming at the member to be guided according to the guide sentence template by utilizing the guide keyword.

Based on the above description of the embodiments, the keyword filtering unit may include at least one filtering subunit as follows:

the first screening subunit is used for randomly selecting one discussion keyword as a guide keyword of the member to be guided from the discussion keywords acquired within the target discussion time length;

the second screening subunit is used for screening the discussion keyword with the largest occurrence frequency from the discussion keywords acquired in the target discussion time length as the guide keyword of the member to be guided;

a third screening subunit, configured to screen the discussion keyword with the highest attention of the member to be guided as a guide keyword from the discussion keywords acquired within the target discussion time; wherein the attention degree is determined based on a sight line tracking detection result of the member to be guided.

Based on the above description of the embodiments, the above guidance sentence audio generating module 33 may include:

a detecting unit, configured to detect whether the discussion participation degree of each member in the discussion group in the target discussion time period is smaller than a discussion threshold;

a member to be guided determining unit, configured to determine, as a member to be guided, a member corresponding to the discussion participation degree smaller than the discussion threshold;

based on this, the above guidance sentence audio output module 34 may include:

and the transmission control unit is used for sequentially sending the guide statement audio of the corresponding member to be guided to the electronic equipment of each member in the discussion group for output according to the discussion participation degree of each member to be guided and a preset time interval under the condition that the number of the members to be guided is multiple.

It should be noted that, various modules, units, and the like in the embodiments of the foregoing apparatuses may be stored in the memory as program modules, and the processor executes the program modules stored in the memory to implement corresponding functions, and for the functions implemented by the program modules and their combinations and the achieved technical effects, reference may be made to the description of corresponding parts in the embodiments of the foregoing methods, which is not described in detail in this embodiment.

The present application further provides a computer-readable storage medium, on which a computer program may be stored, where the computer program may be called and loaded by a processor to implement the steps of the audio discussion guidance method described in the foregoing embodiments, and the specific implementation process may refer to, but is not limited to, the descriptions of the corresponding parts of the foregoing embodiments.

The embodiments in the present description are described in a progressive or parallel manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device, the system and the computer equipment disclosed by the embodiment correspond to the method disclosed by the embodiment, so that the description is relatively simple, and the relevant points can be referred to the method part for description.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended 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. An audio discussion guidance method, the method comprising:

2. The method of claim 1, said obtaining discussion engagement of a respective said member within a target discussion time duration of said discussion group in dependence on said audio production time, comprising:

3. The method of claim 2, wherein said counting discussion participation durations of the same member within a target discussion duration of the discussion group according to the audio information and the audio generation time comprises:

4. The method according to claim 3, wherein the performing keyword detection on the audio information obtained within the target discussion duration to obtain discussion keywords for the discussion topic of the discussion group includes:

5. The method according to claim 4, wherein the performing keyword detection on the audio information obtained within the target discussion duration by using the topic keyword to obtain a discussion keyword for the discussion topic of the discussion group includes:

6. The method according to any one of claims 1 to 5, wherein generating a guidance statement audio for the member to be guided according to the audio information acquired within the target discussion time period includes:

calling a guide sentence template matched with the guide keyword;

7. The method according to claim 6, wherein the step of filtering the guide keywords for the member to be guided from the discussion keywords acquired within the target discussion duration includes at least one of the following implementation manners:

8. The method according to any one of claims 1 to 5, wherein the determining, as the member to be guided, the member corresponding to the discussion participation degree meeting the discussion guidance condition includes:

9. An audio discussion guidance apparatus, the apparatus comprising:

10. A computer device, the computer device comprising:

a communication module;

a memory for storing a program for implementing the audio discussion guidance method according to any one of claims 1 to 8;

a processor for loading and executing the program stored in the memory to implement the audio discussion guidance method according to any one of claims 1 to 8.