CN113241070A

CN113241070A - Hot word recall and updating method, device, storage medium and hot word system

Info

Publication number: CN113241070A
Application number: CN202110488840.7A
Authority: CN
Inventors: 徐文铭; 赵立; 杨晶生; 韩晓; 杜春赛
Original assignee: Beijing Zitiao Network Technology Co Ltd
Current assignee: Beijing Zitiao Network Technology Co Ltd
Priority date: 2021-04-28
Filing date: 2021-04-28
Publication date: 2021-08-10
Anticipated expiration: 2041-04-28
Also published as: CN113241070B

Abstract

The invention provides a hotword recall and updating method, a hotword recall and updating device, a storage medium and a hotword system.

Description

Hot word recall and updating method, device, storage medium and hot word system

Technical Field

The embodiment of the disclosure relates to the technical field of audio and video conferences, in particular to a hotword recall and updating method, a hotword recall and updating device, a storage medium and a hotword system.

Background

Currently, an ASR (Automatic Speech Recognition) technology is commonly used in an audio/video conference to convert real-time voice information during the conference process into text information to form a conference record, and in order to improve the voice Recognition accuracy, the same Recognition method is adopted for different users in most voice Recognition processes, and a Recognition strategy is not customized for different users, so that the voice Recognition accuracy is low.

Disclosure of Invention

The embodiment of the disclosure provides a hotword recall and presenting method, a hotword recall and presenting device, a storage medium and a hotword system.

In a first aspect, an embodiment of the present disclosure provides a hotword recall method, including:

searching a user-defined hot word set corresponding to the target user identification in a user-defined hot word database;

and performing hot word extraction by using a preset hot word extraction algorithm and the searched user-defined hot word set to obtain a customized hot word information set corresponding to the target user identifier, wherein the customized hot word information comprises a customized hot word and a corresponding heat value.

In some optional embodiments, the customized hotword information set corresponding to the target user identifier includes a customized hotword information subset.

In some optional embodiments, the customized hotword information set corresponding to the target user identifier further includes a non-customized hotword information subset, and each hotword value in the customized hotword information subset is greater than or equal to each hotword value in the non-customized hotword information subset.

In some optional embodiments, the preset hotword extraction algorithm is any one of: bayesian averaging, Newton's cooling law-fixing method, theme model method.

In some optional embodiments, the target user identifier is a general user identifier or a user group identifier, and the user group identifier is associated with at least one general user identifier.

In some optional embodiments, the set of customized hotwords corresponding to the target user identifier is sent by the client corresponding to the target user identifier in real time and updated into the customized hotword database.

In a second aspect, an embodiment of the present disclosure provides a hotword updating method, including: responding to the detection of a hot word updating operation for triggering the updating of a user-defined hot word set corresponding to a target user identifier into a target hot word set, sending a hot word updating request to a user-defined hot word database so as to update the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database into the target hot word set, enabling a hot word server to query the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database, and performing hot word extraction by using a preset hot word extraction algorithm and the searched user-defined hot word set to obtain a customized hot word information set corresponding to the target user identifier, wherein the customized hot word information comprises customized hot words and corresponding heat values.

In some optional embodiments, the target user identifier is a common user identifier or a user group identifier, and the user group identifier is associated with at least one common user identifier.

In some optional embodiments, the responding to the detection of the hotword updating operation for triggering the updating of the customized hotword set corresponding to the target user identification to the target hotword set includes:

responding to the detected conversation operation with the robot contact person identification, and determining the conversation intention according to the conversation content of the conversation operation to update the user-defined hot word set corresponding to the target user identification into a target hot word set, wherein the robot contact person identification is used for indicating a virtual conversation robot.

In some optional embodiments, the dialogue operation with the robot contact identification includes:

inputting characters, pictures or voice conversation messages by taking the target user identifier as a sender identifier and the robot contact identifier as a receiver identifier; or

Receiving and presenting a preset candidate dialog intention information set with the robot contact person identifier as a sender identifier and the target user identifier as a receiver identifier, and detecting selection operation aiming at candidate dialog intention information in the preset candidate dialog intention information set.

In some optional embodiments, the sending a hotword update request to the custom hotword database includes:

and sending the hot word updating request to the user-defined hot word database by taking the robot contact person identifier as a sender identifier.

In some optional embodiments, the method further comprises:

and in response to the detection of a user-defined hot word checking operation for triggering the user-defined hot word set corresponding to the target user identification, acquiring the user-defined hot word set corresponding to the target user identification from the user-defined hot word database, and presenting the acquired user-defined hot word set.

In some optional embodiments, the responding to the detection of the custom hotword viewing operation for triggering viewing of the custom hotword set corresponding to the target user identification includes:

responding to the detected conversation operation with the robot contact person identification, and determining the conversation intention to view the user-defined hot word set corresponding to the target user identification according to the conversation content of the conversation operation, wherein the robot contact person identification is used for indicating a virtual conversation robot.

In some optional embodiments, the updating the customized hotword set corresponding to the target user identifier in the customized hotword database to the target hotword set includes:

and updating the user-defined hot word set corresponding to the target user identification in the user-defined hot word database into data obtained by encrypting the target hot word set.

In a third aspect, an embodiment of the present disclosure provides a hotword recall apparatus, which is applied to a hotword server, and includes: the query unit is configured to query a user-defined hot word set corresponding to the target user identification in a user-defined hot word database; and the recall unit is configured to extract the hotwords by utilizing a preset hotword extraction algorithm and the searched user-defined hotword set to obtain a customized hotword information set corresponding to the target user identification, wherein the customized hotword information comprises the customized hotwords and corresponding heat values.

In a fourth aspect, an embodiment of the present disclosure provides a hotword updating apparatus, applied to a client, where the apparatus includes: the updating unit is configured to respond to the detection of a hot word updating operation for triggering the updating of a user-defined hot word set corresponding to a target user identifier into a target hot word set, send a hot word updating request to a user-defined hot word database so as to update the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database into the target hot word set, enable a hot word server to query the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database, and extract hot words by using a preset hot word extraction algorithm and the searched user-defined hot word set to obtain a customized hot word information set corresponding to the target user identifier, wherein the customized hot word information comprises customized hot words and corresponding heat values.

In some optional embodiments, the apparatus further comprises:

and the hot word checking unit is configured to respond to the detection of a user-defined hot word checking operation for triggering the user-defined hot word set corresponding to the target user identification to be checked, acquire the user-defined hot word set corresponding to the target user identification from the user-defined hot word database, and present the acquired user-defined hot word set.

In a fifth aspect, an embodiment of the present disclosure provides a hotword server, including: one or more processors; a storage device, on which one or more programs are stored, which, when executed by the one or more processors, cause the one or more processors to implement the method as described in any implementation manner of the first aspect.

In a sixth aspect, an embodiment of the present disclosure provides a client, including: one or more processors; a storage device, on which one or more programs are stored, which, when executed by the one or more processors, cause the one or more processors to implement the method as described in any implementation manner of the second aspect.

In a seventh aspect, embodiments of the present disclosure provide a computer-readable storage medium on which a computer program is stored, wherein the computer program, when executed by one or more processors, implements the method as described in any of the implementations of the first aspect and/or the method as described in any of the implementations of the second aspect.

In an eighth aspect, an embodiment of the present disclosure provides a hotword system, including a hotword server as described in any implementation manner of the fifth aspect and a client as described in any implementation manner of the sixth aspect.

In the application of audio and video conferences, most of the existing hotword services are general hotwords, that is, the hotwords for all users are the same, or a background worker can manually and independently set corresponding hotwords for a background of the user and then restart a hotword server, instead of the user who actively sets the hotwords and takes effect in real time. If the user wants to set the hotword by himself, the hotword can be achieved only by restarting the hotword server after the background staff replaces the hotword server for operation. Or, the existing hot word service can be obtained by adopting an artificial intelligence technology based on a large amount of corpus information analysis, and cannot embody the user-defined hot words and take effect in real time.

According to the hotword recall and updating method, the hotword recall and updating device, the storage medium and the hotword system, the user configuration custom hotwords are provided at the client side and are updated into the custom hotword database in real time, when the hotword is recalled for the target user in the hotword server, the custom hotword database is read to obtain the custom hotword set of the target user, then the hotword recall is carried out, and the custom hotword and the corresponding hotword value of the target user can be embodied in the hotword information recalled for the target user. The user-defined hotword of the target user can be understood as a problem and a transaction which are relatively concerned by the target user, or can be understood as a topic, a transaction, a problem, an entity and the like which are relatively frequently mentioned by the target user. Due to the above property of the hotword, in the future speech recognition process, the speech recognition server may invoke the hotword information service provided by the hotword server, and then may perform recognition based on the speech data to be recognized and the customized hotword information set of the target user in the speech recognition process related to the target user (e.g., real-time recognition of conference speech of an audio-video conference in which the target user participates), for example, a relatively higher recognition weight is set in the speech recognition process for a customized hotword with a higher heat value in the customized hotword information set of the target user, and a relatively lower recognition weight is set in the speech recognition process for a customized hotword with a lower heat value, since the customized hotword information set of the target user is customized by the target user in real time using the client and updated into the hotword database, it is possible to avoid speech recognition errors, the speech recognition accuracy is improved. Compared with the prior art that the hot word server is restarted after manual setting by background staff, the user can set the hot word server in real time without restarting the hot word server. Compared with the method only adopting artificial intelligence technology corpus information analysis, the method can reflect the user-defined hot words in real time.

Drawings

Other features, objects, and advantages of the disclosure will become apparent from a reading of the following detailed description of non-limiting embodiments which proceeds with reference to the accompanying drawings. The drawings are only for purposes of illustrating the particular embodiments and are not to be construed as limiting the invention. In the drawings:

FIG. 1 is a system architecture diagram of one embodiment of a hotword recall system according to the present disclosure;

FIG. 2 is a timing diagram for one embodiment of a hotword recall system according to the present disclosure;

FIG. 3 is a flow diagram for one embodiment of a hotword recall method according to the present disclosure;

FIG. 4 is a flow diagram for one embodiment of a hotword update method according to the present disclosure;

FIG. 5 is a schematic diagram illustrating one embodiment of a hotword recall device according to the present disclosure;

FIG. 6 is a schematic block diagram illustrating one embodiment of a hotword update apparatus according to the present disclosure;

FIG. 7 is a schematic block diagram of a computer system suitable for use as a client or server for implementing embodiments of the present disclosure.

Detailed Description

The present disclosure is described in further detail below with reference to the accompanying drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the related invention are shown in the drawings.

It should be noted that, in the present disclosure, the embodiments and features of the embodiments may be combined with each other without conflict. The present disclosure will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.

FIG. 1 illustrates an exemplary hotword system architecture 100 to which one embodiment of the hotword recall method and apparatus and the hotword update method and apparatus of the present disclosure may be applied.

As shown in FIG. 1, the system 100 may include

clients

101, 102, 103, a network 104, a hotword server 105, and a custom hotword database 106. The network 104 is a medium used to provide communication links between the

clients

101, 102, 103, the hotword server 105 and the custom hotword database 106. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.

The user may use the

clients

101, 102, 103 to interact with the hotword server 105 and the custom hotword database 106 over the network 104 to receive or send messages, etc. The

clients

101, 102, 103 may have various communication client applications installed thereon, such as an audio video conference application, an Instant Message (IM) application, a voice recognition application, a web browser application, a shopping application, a search application, a mailbox client, social platform software, and the like.

The

clients

101, 102, 103 may be hardware or software. When the

clients

101, 102, 103 are hardware, they may be various electronic devices having a display screen and supporting sound collection and/or video collection, including but not limited to smart phones, tablet computers, e-book readers, MP3 players (Moving Picture Experts Group Audio Layer III, mpeg compression standard Audio Layer 3), MP4 players (Moving Picture Experts Group Audio Layer IV, mpeg compression standard Audio Layer 4), laptop and desktop computers, etc. When the

clients

101, 102, 103 are software, they can be installed in the electronic devices listed above. It may be implemented as multiple pieces of software or software modules (e.g., to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.

The hotword server 105 may be a server that provides various services, such as a server that provides support for hotword services issued by an automatic speech recognition server (not shown in FIG. 1). The hotword server 105 may process the hotword service request issued from the automatic speech recognition server and feed back the processing results (e.g., the customized hotword information set corresponding to the target user) to the automatic speech recognition server. Here, the hotword server 105 and the automatic speech recognition server may be deployed in one server, or they may be different servers.

The hotword server 105 may be hardware or software. When the hotword server 105 is hardware, it may be implemented as a distributed server cluster composed of a plurality of servers, or may be implemented as a single server. When the hotword server 105 is software, it may be implemented as multiple pieces of software or software modules (e.g., to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.

It should be noted that the custom hotword database 106 may be deployed in the hotword server 105, and the custom hotword database 106 may also be deployed in other electronic devices different from the hotword server 105.

It should be noted that the hotword updating method provided by the present disclosure is generally executed by the

clients

101, 102, 103, and accordingly, the hotword updating apparatus is generally disposed in the

clients

101, 102, 103.

It should be noted that the hotword recall method provided by the present disclosure is generally executed by the hotword server 105, and accordingly, the hotword recall apparatus is generally disposed in the hotword server 105.

It should be understood that the number of clients, networks, hotword servers, and custom hotword databases in FIG. 1 are merely illustrative. There may be any number of clients, networks, hotword servers, and custom hotword databases, as desired for implementation.

With continued reference to FIG. 2, a timing sequence 200 for one embodiment of a hotword system according to the present disclosure is shown. The hotword system in the embodiment of the disclosure can comprise a client and a hotword server. The sequence 200 includes the following steps:

step 201, in response to detecting a hotword update operation for triggering the update of the user-defined hotword set corresponding to the target user identifier into the target hotword set, the client sends a hotword update request to the user-defined hotword database so as to update the user-defined hotword set corresponding to the target user identifier in the user-defined hotword database into the target hotword set.

In this embodiment, the client may be installed with an audio and video conference application or an instant message application, or an operation object that can trigger the audio and video conference application may be set in the instant message application, so that the client may jump from the instant message application to the audio and video conference application. The audio-video conference application can also provide an instant message function. The client may also have a web browser-like application installed therein.

The target user identification may be a user identification that the user successfully logs in an application or a website page providing the hotword updating operation by using the client device and is authorized to pass verification.

Accordingly, hotword update operations may be presented in various forms. For example, after a user successfully logs in an audio and video conference application by using a client through a target user identifier, a corresponding hot word setting interface is provided in the audio and video conference application, after the user enters the hot word setting interface, a hot word input text box or a hot word input pull-down list for inputting hot words is provided in the hot word setting interface, and then the user inputs a target hot word set therein, clicks a determination or submission button in the hot word setting interface, and further can trigger a hot word updating operation.

It is understood that the hotword update request sent by the client to the custom hotword database may include the target user identification and the target hotword set. The custom hotword database may be a database in various forms, for example, a relational or non-relational database, in which a user identifier and a corresponding set of custom hotwords may be stored correspondingly. The client side can send a hot word updating request to the custom hot word database by calling a data updating operation service corresponding to the custom hot word database (for example, a corresponding application program interface can be called), so that a custom hot word set corresponding to a target user identifier in the custom hot word database is updated to a target hot word set.

In some optional embodiments, the customized hot word set corresponding to the target user identifier in the customized hot word database is updated to the target hot word set, which may be that the customized hot word set corresponding to the target user identifier in the customized hot word database is updated to data obtained by encrypting the target hot word set, that is, the customized hot word set stored in the customized hot word database is encrypted data. The target hot word set can be encrypted at the client and then sent to the user-defined hot word database, so that the leakage of secret in the data sending process can be prevented, and the safety is improved. The target hot word set can be encrypted and updated by the electronic equipment where the user-defined hot word database is located, so that the encryption calculation amount of the client can be reduced. Due to the encryption, even a service provider of the customized hotword database can hardly know the hotword updated by the target user, namely the privacy of the target user can be protected, and the method is more suitable for enterprise users to customize the hotword.

In some alternative embodiments, the target user identifier may be a general user identifier or a user group identifier, and the user group identifier may be associated with at least one general user identifier. Here, the user group identification is used to uniquely indicate the user group, and the user group may be understood as an organization, an enterprise, a group, and the like, which are composed of different general users, and accordingly, the general users may be members of the organization, employees of the enterprise, members of the group, and the like. The general user identifier associated with the user group identifier may be understood as the user identifier of each specific member, employee, etc. in the user group indicated by the user group identifier. Furthermore, the user group and the common user can update the corresponding user-defined hot word set to the user-defined hot word database in real time through step 201.

In some optional embodiments, in response to detecting a hotword update operation for triggering to update the customized hotword set corresponding to the target user identifier to the target hotword set, the method may include: and responding to the detected conversation operation with the robot contact person identifier, and determining the conversation intention to update the user-defined hot word set corresponding to the target user identifier into the target hot word set according to the conversation content of the conversation operation. Wherein the robot contact identification is used to indicate the virtual dialogue robot. Here, the dialogue operation with the robot contact identifier may be regarded as a dialogue operation with the virtual dialogue robot. Specifically, the dialog operation with the virtual dialog robot may be realized through an instant messaging service provided in an audio-video application installed in the client, an instant messaging application installed in the client, or an instant messaging service provided in a preset web page accessed by the client. The determination of the dialog intention according to the dialog contents of the dialog operation may be performed by analyzing the dialog contents by a client or a server serving an instant message using a Natural Language Processing (NLP) technique.

In some optional embodiments, the dialogue operation with the robot contact identification may include:

and the user inputs characters, pictures or voice conversation messages by using the client and taking the target user identifier as a sender identifier and the robot contact identifier as a receiver identifier. Alternatively, it may be:

the client receives and presents a preset candidate dialogue intention information set which takes the robot contact person identifier as a sender identifier and takes the target user identifier as a receiver identifier, and detects selection operation aiming at candidate dialogue intention information in the preset candidate dialogue intention information set.

Here, the candidate dialog intention information in the preset candidate dialog intention information set may be used to characterize different dialog intents. For example, the dialog intent may be: reserving an audio and video conference, starting the audio and video conference, ending the audio and video conference, recording audio and video during the audio and video conference, customizing a custom hot word set of a target user and the like.

In some optional embodiments, the hotword update request is sent to the custom hotword database, or the hotword update request may be sent to the custom hotword database by using the identifier of the robot contact as the sender identifier, which may also be implemented to facilitate the user to implement hotword update operation through instant message application, and facilitate the user to use.

Step 202, the hot word server queries a user-defined hot word set corresponding to the target user identifier in a user-defined hot word database.

Here, for the hotword server, the target user identifier may be any specified user identifier in a preset user identifier set. And the hotword server may perform step 202 and step 203 at the specified times.

Alternatively, the target user identifier may also be a user identifier specified by a caller who calls the hotword recall service when the hotword recall service is provided by the hotword server.

For example, after or at the same time of sending the hotword update request to the custom hotword database in step 201, the client may send a message to the hotword server to indicate that the set of custom hotwords corresponding to the target user identifier is updated, and then trigger the hotword server to perform step 202 and step 203.

For another example, after the device in which the custom hotword database is located updates the custom hotword set corresponding to the target user identifier in the custom hotword database to the target hotword set, a message may be sent to the hotword server to indicate that the custom hotword set corresponding to the target user identifier is updated, and then the hotword server is triggered to execute step 202 and step 203.

As another example, the automatic speech recognition server may invoke a hotword recall service and designate the user identification as the target user identification. The target user id during the execution of step 202 and step 203 by the hotword server is then the user id specified in the invocation of the hotword recall service.

As can be seen from the above description, the user identifier and the corresponding user-defined hot word set are correspondingly stored in the user-defined hot word database, so that the hot word server can query the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database, and then go to step 203.

And 203, the hot word server extracts hot words by using a preset hot word extraction algorithm and the searched user-defined hot word set to obtain a customized hot word information set corresponding to the target user identification.

Here, the customized hotword information may include a customized hotword and a corresponding hotword value. The hot value corresponding to the customized hot word in the customized hot word information set corresponding to the target user identifier may be a numerical value obtained by quantizing the usage frequency, attention degree, and the like of the customized hot word for the target user.

Here, the preset hotword extraction algorithm may be any currently known or future developed hotword/keyword extraction algorithm. For example, a bayesian averaging method (for example, TF-IDF (Term Frequency-Inverse text Frequency index)), a newton's cooling law method, a topic model method, or the like may be used.

The hot word server can extract hot words by using a preset hot word extraction algorithm based on the general corpus information or the related corpus information corresponding to the target user identification in the process of extracting the hot words to obtain the customized hot word information set corresponding to the target user identification, and relatively higher hot degree values can be given to the searched hot words in the customized hot word set in the process of extracting the hot words. And then the target user can endow the hot words in the customized user-defined hot word set with a relatively high hot value.

The related corpus information corresponding to the target user id may be, for example:

in the scene of the audio and video conference, a user can use a client to log in an audio and video conference application by using a target user identifier to participate in the audio and video conference, and in the process of the conference, an audio and video conference server performs voice recognition and speaker identity recognition on audio data of participants in real time, so that conference records comprising conference content texts and speaker identities can be obtained. And the corpus information related to the target user identifier may be text of each conference content in which the speaker identity is the target user identifier in the conference record recognized by the audio/video conference server before. The conference content texts reflect words spoken by the target user in the conference participating process, and hot words can be extracted based on the conference content texts, so that hot words more specific to the target user can be obtained.

In some optional embodiments, the customized hotword information set corresponding to the target user identification obtained in step 203 may include a customized hotword information subset. Here, the hot words respectively defined in the custom hot word information subset may include the hot words in the custom hot word set corresponding to the target user identifier, which are queried in the custom hot word database by the hot word server in step 202. Based on this, optionally, the customized hotword information set corresponding to the target user identification may further include a non-customized hotword information subset. Here, the hotword in the non-custom hotword information subset may not include the hotword in the custom hotword set corresponding to the target user identifier, and each hotness value in the custom hotword information subset is greater than or equal to each hotness value in the non-custom hotword information subset. That is, for the target user, the hot word hot value customized by the target user is not lower than the hot word hot value customized by other non-target users, and the purpose that the target user wants to customize the hot word by himself/herself is reflected.

In some optional embodiments, if the target user identifier is a common user identifier, step 203 may be performed as follows:

and performing hot word extraction by using a preset hot word extraction algorithm, the global hot word set and the personal hot word set to obtain a customized hot word information set corresponding to the target user identification. Here, the global hot word set is a user-defined hot word set corresponding to a user group identifier associated with the target user identifier, and the personal hot word set is a user-defined hot word set corresponding to the target user identifier. That is, the user-defined hot words of the user group associated with the target user are comprehensively considered in addition to the user-defined hot word set of the target user, and when the target user participates in the audio/video conference and also includes other common users associated with the same user group as the target user, the user-defined hot words of the user group associated with the target user are considered, so that the accuracy of voice recognition for the audio/video conference can be improved.

Optionally, based on the above optional embodiment, considering that the priority of the target user's own customized hotword is often higher than the priority of the customized hotword of the user group associated with the target user, the heat values corresponding to the individual hotwords, the heat values corresponding to the individual global hotwords, and the heat values corresponding to the individual human hotwords and other hotwords except the global hotwords in the obtained customized hotword information set corresponding to the target user identifier are arranged from large to small. According to the alternative mode, the recognition accuracy of the subsequent speech recognition can be further improved.

Through the steps 202 and 203, the targeted customized hotword and the corresponding hotness value for the target user can be realized, and further, in the subsequent process of performing voice recognition for the target user or the design target user, recognition can be performed based on the voice data to be recognized and the customized hotword information set of the target user. For example, for a customized hotword with a high corresponding hotword value in the customized hotword information set of the target user, a relatively high recognition weight is set in the speech recognition process, and for a customized hotword with a low hotword value, a relatively low recognition weight is set in the speech recognition process, so that speech recognition errors can be avoided, and the speech recognition accuracy is improved.

And the speech recognition scenario for the target user or the design target user may be, for example: audio and video conferences participated by the target users, audio and video calls participated by the target users, audio and video conferences hosted by the target users and the like.

In some optional embodiments, the timing sequence 200 may further include the following step 204:

and 204, in response to the detection of the user-defined hot word checking operation for triggering the user-defined hot word set corresponding to the target user identifier, the client acquires the user-defined hot word set corresponding to the target user identifier from the user-defined hot word database, and presents the acquired user-defined hot word set.

Here, the custom hotword view operation may be presented in various forms. For example, after a user successfully logs in the audio and video conference application with a target user identifier by using the client, a corresponding custom hotword viewing interface can be provided in the audio and video conference application, when the user opens the custom hotword viewing interface, a custom hotword viewing operation can be triggered, and then the client can obtain a custom hotword set corresponding to the target user identifier from a custom hotword database and present the obtained custom hotword set on the opened custom hotword viewing interface. Of course, it is understood that the custom hotword viewing interface and the hotword setting interface described above can also be combined into one interface.

In some optional embodiments, a dialog operation with the robot contact identifier may also be detected, and in a case that it is determined from the dialog content of the dialog operation that the dialog is intended to view the custom hotword set corresponding to the target user identifier, it is determined that a custom hotword viewing operation is detected. For details, reference may be made to the above related descriptions, which are not repeated herein.

According to the hotword system provided by the embodiment of the disclosure, the user configuration custom hotwords are provided at the client and are updated into the custom hotword database in real time, when the hotword is recalled for the target user in the hotword server, the custom hotword database is read to obtain the custom hotword set of the target user, and then the hotword recall is performed, so that the custom hotword and the corresponding hotword value of the target user can be embodied in the hotword information recalled for the target user. Then, when the hot word information service in the hot word server is called by the future voice recognition service, the target user can be identified based on the voice data to be recognized and the customized hot word information set of the target user instead of the general hot word information set in the voice recognition process related to the target user. And the target user updates the corresponding user-defined hot word set to the user-defined hot word database in real time by using the client, and the hot word server can read the user-defined hot word set of the target user from the hot word database in real time on line when determining the user-defined hot word information set of the target user, without restarting the hot word server, thereby realizing the real-time property of the user-defined hot words.

With continued reference to FIG. 3, a flow 300 of one embodiment of a hotword recall method according to the present disclosure is shown. The hot word recall method can be applied to a hot word server, and comprises the following steps:

step 301, searching a user-defined hot word set corresponding to the target user identifier in a user-defined hot word database.

And 302, performing hotword extraction by using a preset hotword extraction algorithm and the searched user-defined hotword set to obtain a customized hotword information set corresponding to the target user identifier.

In this embodiment, the specific operations of step 301 and step 302 and the technical effects thereof are substantially the same as the operations and effects of step 202 and step 203 in the embodiment shown in fig. 2, and are not repeated herein.

In the method provided by the above embodiment of the present disclosure, when a hotword is recalled for a target user in the hotword server, the custom hotword database is read first to obtain a custom hotword set of the target user, and then a hotword recall is performed, so that the custom hotword and the corresponding hotword value of the target user can be embodied in the hotword information recalled for the target user. Then, when the hot word information service in the hot word server is called by the future voice recognition service, the voice recognition can be carried out based on the voice data to be recognized and the customized hot word information set of the target user in the voice recognition process related to the target user, so that the voice recognition error can be avoided, the voice recognition accuracy is improved, and the hot word server does not need to be restarted in the whole process.

With continued reference to FIG. 4, a flow 400 of one embodiment of a hotword update method according to the present disclosure is shown. The hotword updating method can be applied to a client, for example, and comprises the following steps:

step 401, in response to detecting a hotword update operation for triggering the update of the user-defined hotword set corresponding to the target user identifier into the target hotword set, sending a hotword update request to the user-defined hotword database, so as to update the user-defined hotword set corresponding to the target user identifier in the user-defined hotword database into the target hotword set.

In the present embodiment, the specific operation of step 401 and the technical effect thereof are substantially the same as the operation and effect of step 201 in the embodiment shown in fig. 2, and are not repeated herein.

In some optional embodiments, the flow 400 may further include the following step 402:

step 402, in response to detecting a custom hotword check operation for triggering checking of a custom hotword set corresponding to a target user identifier, obtaining the custom hotword set corresponding to the target user identifier from a custom hotword database, and presenting the obtained custom hotword set.

In the present embodiment, the detailed operation of step 402 and the technical effects thereof are substantially the same as the operation and effects of step 204 in the embodiment shown in fig. 2, and are not repeated herein.

The hotword updating method provided by the above embodiment of the present disclosure provides a user configuration custom hotword at the client and updates the custom hotword database in real time, and then provides a data base for subsequently reading the custom hotword database when a hotword is recalled for a target user in the hotword server, and then the hotword server can recall the hotword based on the custom hotword set of the target user read from the custom hotword database, so that the custom hotword and the corresponding hotword value of the target user can be embodied in the hotword information recalled for the target user. Then, when the hot word information service in the hot word server is called by the future voice recognition service, the voice recognition can be carried out based on the voice data to be recognized and the customized hot word information set of the target user in the voice recognition process related to the target user, and the voice recognition accuracy can be improved.

With further reference to fig. 5, as an implementation of the methods shown in the above-mentioned figures, the present disclosure provides an embodiment of a hotword recall apparatus, which corresponds to the method embodiment shown in fig. 3, and which is specifically applicable to various servers.

As shown in fig. 5, the hotword recall device 500 of the present embodiment includes: a query unit 501 and a recall unit 502. The query unit 501 is configured to query a user-defined hot word set corresponding to a target user identifier in a user-defined hot word database; a recall unit 502 configured to perform hotword extraction by using a preset hotword extraction algorithm and the searched user-defined hotword set to obtain a customized hotword information set corresponding to the target user identifier, where the customized hotword information includes a customized hotword and a corresponding hotword value.

In this embodiment, the detailed processing of the query unit 501 and the recall unit 502 of the hotword recall device 500 and the technical effects thereof can refer to the related descriptions of step 301 and step 302 in the corresponding embodiment of fig. 3, which are not repeated herein.

In some optional embodiments, the preset hotword extraction algorithm may be any one of the following: bayesian averaging, Newton's cooling law-fixing method, theme model method.

In some optional embodiments, the target user identifier may be a general user identifier or a user group identifier, and the user group identifier is associated with at least one general user identifier.

In some optional embodiments, the user-defined hot word set corresponding to the target user identifier is sent by the client corresponding to the target user identifier in real time and updated into the user-defined hot word database.

It should be noted that, for details of implementation and technical effects of each unit in the hotword recall device provided in the embodiments of the present disclosure, reference may be made to descriptions of other embodiments in the present disclosure, and details are not described herein again.

With further reference to fig. 6, as an implementation of the methods shown in the above-mentioned figures, the present disclosure provides an embodiment of a hotword updating apparatus, which corresponds to the method embodiment shown in fig. 4, and which may be specifically applied to various client devices.

As shown in fig. 6, the hotword updating apparatus 600 of the present embodiment includes: a hotword update unit 601. The hotword updating unit 601 is configured to, in response to detecting a hotword updating operation for triggering updating of a custom hotword set corresponding to a target user identifier into a target hotword set, send a hotword updating request to a custom hotword database to update the custom hotword set corresponding to the target user identifier in the custom hotword database into the target hotword set, enable a hotword server to query the custom hotword set corresponding to the target user identifier in the custom hotword database, and perform hotword extraction by using a preset hotword extraction algorithm and the searched custom hotword set to obtain a customized hotword information set corresponding to the target user identifier, where the customized hotword information includes a customized hotword and a corresponding heat value.

In this embodiment, the detailed processing of the hotword updating unit 601 of the hotword updating device 600 and the technical effects thereof can refer to the related description of step 401 in the corresponding embodiment of fig. 4, and are not repeated herein.

In some optional embodiments, the above, in response to detecting a hotword update operation for triggering the update of the customized hotword set corresponding to the target user identifier to the target hotword set, may include:

responding to the detected conversation operation with the robot contact person identifier, and determining the conversation intention to update the user-defined hot word set corresponding to the target user identifier into a target hot word set according to the conversation content of the conversation operation, wherein the robot contact person identifier is used for indicating the virtual conversation robot.

In some optional embodiments, the above dialogue operation with the robot contact identifier may include:

inputting characters, pictures or voice conversation messages by taking the target user identifier as a sender identifier and taking the robot contact identifier as a receiver identifier; or

Receiving and presenting a preset candidate dialog intention information set which takes the robot contact person identifier as a sender identifier and takes the target user identifier as a receiver identifier, and detecting selection operation aiming at the candidate dialog intention information in the preset candidate dialog intention information set.

In some optional embodiments, the sending the hotword update request to the custom hotword database may include:

In some optional embodiments, the apparatus 600 may further include:

and the hot word checking unit 602 is configured to, in response to detecting a custom hot word checking operation for triggering checking of a custom hot word set corresponding to the target user identifier, acquire the custom hot word set corresponding to the target user identifier from the custom hot word database, and present the acquired custom hot word set.

In some optional embodiments, the responding to the detection of the custom hotword viewing operation for triggering viewing of the custom hotword set corresponding to the target user identifier may include:

and responding to the detected conversation operation with the robot contact person identifier, and determining the conversation intention to view the user-defined hot word set corresponding to the target user identifier according to the conversation content of the conversation operation, wherein the robot contact person identifier is used for indicating the virtual conversation robot.

In some optional embodiments, the updating the customized hotword set corresponding to the target user identifier in the customized hotword database to the target hotword set may include:

Referring now to FIG. 7, shown is a block diagram of a computer system 700 suitable for use as a client or server for implementing embodiments of the present disclosure. The computer system 700 shown in fig. 7 is only an example and should not bring any limitations to the functionality or scope of use of the embodiments of the present disclosure.

As shown in fig. 7, computer system 700 may include a processing device (e.g., central processing unit, graphics processor, etc.) 701 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)702 or a program loaded from a storage device 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data necessary for the operation of the electronic apparatus 700 are also stored. The processing device 701, the ROM 702, and the RAM 703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.

Generally, the following devices may be connected to the I/O interface 705: input devices 706 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, etc.; an output device 707 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage 708 including, for example, magnetic tape, hard disk, etc.; and a communication device 709. The communications device 709 may allow the computer system 700 to communicate wirelessly or by wire with other devices to exchange data. While fig. 7 illustrates a computer system 700 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.

In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via the communication means 709, or may be installed from the storage means 708, or may be installed from the ROM 702. The computer program, when executed by the processing device 701, performs the above-described functions defined in the methods of embodiments of the present disclosure.

It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.

The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.

The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to implement the hotword recall method shown in the embodiment shown in fig. 3 and its alternative embodiments, and/or the hotword update method shown in the embodiment shown in fig. 4 and its alternative embodiments.

Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The units described in the embodiments of the present disclosure may be implemented by software or hardware. Where the name of a cell does not in some cases constitute a limitation on the cell itself, for example, a query cell may also be described as a "cell that queries a custom hotword set corresponding to a target user identification in a custom hotword database".

The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents does not depart from the spirit of the disclosure. For example, the above features and (but not limited to) the features disclosed in this disclosure having similar functions are replaced with each other to form the technical solution.

Claims

1. A hotword recall method applied to a hotword server comprises the following steps:

2. The method of claim 1, wherein the set of customized hotword information corresponding to the target user identification comprises a subset of customized hotword information.

3. The method of claim 2, wherein the customized hotword information set corresponding to the target user identifier further comprises a non-custom hotword information subset, and each hotword value in the custom hotword information subset is greater than or equal to each hotword value in the non-custom hotword information subset.

4. The method of claim 1, wherein the preset hotword extraction algorithm is any one of: bayesian averaging, Newton's cooling law-fixing method, theme model method.

5. The method of claim 1, wherein the target user identity is a general user identity or a user group identity, the user group identity being associated with at least one general user identity.

6. The method of claim 1, wherein the set of custom hotwords corresponding to the target user identifier is sent by the client corresponding to the target user identifier in real time and updated into the custom hotword database.

7. A hotword updating method is applied to a client and comprises the following steps:

responding to the detection of a hot word updating operation for triggering the updating of a user-defined hot word set corresponding to a target user identifier into a target hot word set, sending a hot word updating request to a user-defined hot word database so as to update the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database into the target hot word set, enabling a hot word server to query the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database, and performing hot word extraction by using a preset hot word extraction algorithm and the searched user-defined hot word set to obtain a customized hot word information set corresponding to the target user identifier, wherein the customized hot word information comprises customized hot words and corresponding heat values.

8. The method of claim 7, wherein the target user identity is a common user identity or a user group identity, the user group identity being associated with at least one common user identity.

9. The method of claim 7, wherein the responding to the detection of the hotword update operation for triggering the update of the customized hotword set corresponding to the target user identification to the target hotword set comprises:

10. The method of claim 9, wherein the dialogue operation with the robot contact identification comprises:

11. The method of claim 7, wherein the sending a hotword update request to a custom hotword database comprises:

12. The method of claim 7, wherein the method further comprises:

13. The method of claim 7, wherein the responding to the detection of the custom hotword viewing operation for triggering viewing of the set of custom hotwords corresponding to the target user identification comprises:

14. The method of claim 7, wherein the updating the set of custom hotwords corresponding to the target user identifier in the custom hotword database to the target hotword set comprises:

15. A hotword recall device applied to a hotword server, the device comprising:

the query unit is configured to query a user-defined hot word set corresponding to the target user identification in a user-defined hot word database;

and the recall unit is configured to extract the hotwords by utilizing a preset hotword extraction algorithm and the searched user-defined hotword set to obtain a customized hotword information set corresponding to the target user identification, wherein the customized hotword information comprises the customized hotwords and corresponding heat values.

16. A hotword updating device is applied to a client and comprises:

the updating unit is configured to respond to the detection of a hot word updating operation for triggering the updating of a user-defined hot word set corresponding to a target user identifier into a target hot word set, send a hot word updating request to a user-defined hot word database so as to update the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database into the target hot word set, enable a hot word server to query the user-defined hot word set corresponding to the target user identifier in the user-defined hot word database, and extract hot words by using a preset hot word extraction algorithm and the searched user-defined hot word set to obtain a customized hot word information set corresponding to the target user identifier, wherein the customized hot word information comprises customized hot words and corresponding heat values.

17. A hotword server, comprising:

one or more processors;

a storage device having one or more programs stored thereon,

the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method recited in any of claims 1-6.

18. A client, comprising:

one or more processors;

a storage device having one or more programs stored thereon,

the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method of any of claims 7-14.

19. A computer readable storage medium having a computer program stored thereon, wherein the computer program, when executed by one or more processors, implements the method of any of claims 1-6 and/or the method of any of claims 7-14.

20. A hotword system comprising the hotword server of claim 17 and the client of claim 18.