CN112908311A - Training and sharing method of voice assistant - Google Patents
Training and sharing method of voice assistant Download PDFInfo
- Publication number
- CN112908311A CN112908311A CN202110094701.6A CN202110094701A CN112908311A CN 112908311 A CN112908311 A CN 112908311A CN 202110094701 A CN202110094701 A CN 202110094701A CN 112908311 A CN112908311 A CN 112908311A
- Authority
- CN
- China
- Prior art keywords
- voice assistant
- user
- task model
- task
- voice
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012549 training Methods 0.000 title claims abstract description 70
- 238000000034 method Methods 0.000 title claims abstract description 65
- 230000003993 interaction Effects 0.000 claims abstract description 85
- 230000006870 function Effects 0.000 claims description 22
- 238000007726 management method Methods 0.000 claims description 9
- 238000013475 authorization Methods 0.000 claims description 4
- 230000001360 synchronised effect Effects 0.000 claims description 4
- 238000013500 data storage Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 abstract description 28
- 239000000284 extract Substances 0.000 abstract description 5
- 238000012795 verification Methods 0.000 description 20
- 230000002452 interceptive effect Effects 0.000 description 13
- 235000012054 meals Nutrition 0.000 description 9
- 238000010586 diagram Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 238000004891 communication Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 230000001413 cellular effect Effects 0.000 description 2
- 230000000977 initiatory effect Effects 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000010267 cellular communication Effects 0.000 description 1
- 238000012790 confirmation Methods 0.000 description 1
- 238000005429 filling process Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 229910052709 silver Inorganic materials 0.000 description 1
- 239000004332 silver Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/28—Constructional details of speech recognition systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
- G10L2015/0631—Creating reference templates; Clustering
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
Abstract
The invention discloses a training and sharing method of a voice assistant, which comprises the following steps: 101. the user initiates a command to the voice assistant; 102. the voice assistant recognizes a user command and extracts key knowledge data; 103. the voice assistant judges whether the voice assistant contains a corresponding task model according to the key knowledge data, and if so, the voice assistant calls the task model to execute the task; 104. if not, the voice assistant initiates a search operation to the intelligent interaction platform; 105. if the corresponding task model is found, loading the task model to a voice assistant and executing the task; 106. if the corresponding task model is not found, the voice assistant initiates a process of creating the task model; 107. and the voice assistant saves the created task model and uploads the task model to the intelligent interaction platform. The method of the invention allows the user to train and share the task model, and improves the training efficiency of the voice assistant.
Description
The present patent application is a divisional application of the parent patent application No. 2019107066572.
Technical Field
The embodiment of the invention relates to the technical field of information processing, in particular to a training and sharing method of a voice assistant.
Background
The voice assistant is the most common interactive interface with the user in artificial intelligence, most of the existing voice assistants are created and trained by developers, the developers need to conduct requirement research and function collection, the process is complex, the requirements of the users cannot be comprehensively reflected and responded, and the functions are relatively limited. Moreover, most of the existing voice assistants have no pertinence, cannot meet individual requirements of a single user, do not have a sharing function, cannot quickly meet the self requirements by acquiring a task model shared by others, and shorten the training time. Therefore, how to improve the training efficiency of the voice assistant and provide the voice assistant closer to the user's needs becomes a problem to be solved urgently.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a training and sharing method of a voice assistant, wherein the voice assistant can be in communication connection with an intelligent interaction platform, and download data from the intelligent interaction platform or upload data to the intelligent interaction platform; the method is characterized by comprising the following steps:
104, if not, the voice assistant initiates a search operation to the intelligent interaction platform;
and step 107, the voice assistant saves the created task model and uploads the task model to the intelligent interaction platform.
The voice assistant training and sharing method further comprises the following steps: in step 106, the process of initiating the creation of the task model by the voice assistant specifically includes the following steps:
step 106a, calling a basic slot position to generate a problem;
106b, asking questions of the user through multiple rounds of conversations, acquiring answers of the user, extracting key knowledge data, and filling basic slot positions;
106c, acquiring a specific slot position according to the key knowledge data of the filled one or more basic slot positions;
step 106d, generating a question according to the specific slot position, asking a question to the user through a plurality of rounds of conversations, obtaining an answer of the user, extracting key knowledge data, and filling the specific slot position;
and 106e, generating a task model.
The step 106a further comprises: the key knowledge data extracted from the user's command generates a base slot or calls a base slot stored in the voice assistant that is relevant to creating a new task model.
Said step 106c further comprises: and acquiring specific slot position information according to information related to the interactive object of the voice assistant in the key knowledge data filled by the user.
The voice assistant training and sharing method further comprises the following steps:
106f, establishing an expansion slot position, and generating a problem according to the expansion slot position;
step 106g, filling the expansion slot position;
and step 106h, updating the task model and uploading the task model to the intelligent interaction platform.
Preferably, in step 106g, key knowledge data is acquired by using a man-machine conversation or multiple rounds of conversations to fill the expansion slot; or, the stored user information is obtained according to authorization to fill the expansion slot position; or calling a nesting submodel to fill the extended slot.
Alternatively, step 107 of the voice assistant training and sharing method may be: step 107a, the voice assistant saves the created task model, and the task model is directly shared with other voice assistants through the user equipment.
Preferably, the task is a self-help book borrowing task of an online library.
Furthermore, other voice assistants download the task model uploaded by the voice assistant through the intelligent interaction platform, further modify the task model according to the self-demand, and upload the modified task model and the description information to the intelligent interaction platform.
The embodiment of the invention also provides a training and sharing system of the voice assistant, which comprises: the intelligent interaction platform is located at the cloud end, and the voice assistants are loaded in local user equipment. The intelligent interaction platform comprises: the task model training module is used for training a corresponding task model according to a user instruction;
the task model is used for executing the instructions of the user and completing the task;
the functional component is used for realizing the functions of the voice assistant;
the data storage module is used for storing data information related to the voice assistant;
the extended function module is used for realizing extended functions;
a network interface for implementing network connection;
the voice assistant comprises:
the human-computer interaction interface is used for realizing voice interaction with a user;
the task model training module is used for training a task model according to a user instruction;
the task model is used for executing the instructions of the user and completing the task;
the voice assistant further comprises: the device comprises a voice recognition module, a semantic understanding module and a transceiving interface.
Furthermore, the intelligent interaction platform realizes the management of the voice assistants through user IDs, and one user ID corresponds to one or more voice assistants; when one user ID corresponds to a plurality of voice assistants, if one voice assistant is updated, the intelligent interaction platform initiates synchronous updating of other voice assistants of the same user ID.
Further, after the voice assistant trains the task model, the task model is shared to the intelligent interaction platform for being downloaded by other voice assistants; or after the voice assistant trains the task model, the task model is directly shared to other voice assistants through the user equipment.
The embodiment of the invention also provides a voice assistant training method, which is executed by the voice assistant or an intelligent interaction platform, and comprises the following steps:
step 106a, calling a basic slot position to generate a problem;
106b, asking questions of the user through multiple rounds of conversations, acquiring answers of the user, extracting key knowledge data, and filling basic slot positions;
106c, acquiring specific slot position information according to the key knowledge data of the filled one or more basic slot positions;
step 106d, generating a question according to the specific slot position, asking a question to the user through a plurality of rounds of conversations, obtaining an answer of the user, extracting key knowledge data, and filling the specific slot position;
and 106e, generating a task model.
Further, the method comprises the following steps:
106f, establishing an expansion slot position, and generating a problem according to the expansion slot position;
step 106g, filling the expansion slot position;
and step 106h, updating the task model and uploading the task model to the intelligent interaction platform.
In a preferred step 106g, key knowledge data are acquired by using a man-machine conversation or multiple rounds of conversations to fill the expansion slot; or, the stored user information is obtained according to authorization to fill the expansion slot position; or calling a nesting submodel to fill the extended slot.
By the method and the system, the user can download the task model from the intelligent interaction platform according to the requirement, the efficiency is improved, the task model can be trained automatically, the intelligent interaction platform shares the task model with the voice assistant of other users, the personalized requirement of the user is met, and the training efficiency of the voice assistant in the whole system is improved greatly.
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, and it is obvious that the drawings in the following description are some embodiments of the present invention, and those skilled in the art can also obtain other drawings according to the drawings without creative efforts.
FIG. 1 is a block diagram of a voice assistant training and sharing system in one embodiment of the invention.
FIG. 2 is a block diagram of a voice assistant in one embodiment of the invention.
FIG. 3 is a block diagram of an intelligent interaction platform in one embodiment of the invention.
FIG. 4 is a flow chart of a method for a voice assistant training and sharing method in another embodiment of the present invention.
FIG. 5 is a flow chart of a voice assistant training method in another embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings. The embodiments and specific features of the embodiments of the present invention are detailed descriptions of technical solutions of the embodiments of the present invention, and are not limited to technical solutions of the present invention, and the technical features of the embodiments and the embodiments of the present invention may be combined with each other without conflict.
Example one
Fig. 1 is a block diagram of a voice assistant training and sharing system according to a first embodiment of the present invention, where the voice assistant training and sharing system 1 mainly includes a plurality of voice assistants 2 and an intelligent interactive platform 3. The voice assistant 2 is located in local user equipment, the intelligent interaction platform 3 is located at the cloud end, communication connection is established between the voice assistant and the intelligent interaction platform through a public network or a private network, and the voice assistant and the intelligent interaction platform are further connected with the public network through network interfaces respectively. Public networks of the present invention include, but are not limited to, all networks accessed and used by users, such as the internet, cellular networks, PSTN networks, WiFi, WLAN, etc. The private network of the present invention includes, but is not limited to, a communication network with a certain encryption level, such as a private channel and a virtual channel in a public network, a specially constructed network channel, and the like.
The voice assistant 2 may be loaded in a user device, and the user device of the present invention includes, but is not limited to, a car machine, a mobile phone, a portable terminal, an intelligent wearable device, an intelligent television, an intelligent home device, an intelligent wearable device, a personal assistant, and the like. The voice assistant can communicate with the intelligent interaction platform through the user equipment in the intelligent interaction platform through a public network or a private network, access the intelligent interaction platform, and upload and download data. The voice assistant can realize man-machine voice interaction, recognize the intention of the user and execute the instruction of the user. The voice assistant can access an external network through the user equipment in the voice assistant via a public network, for example, access an external website, platform or application, etc., so as to execute the instructions of the user and complete the task. For example: accessing a meal ordering website according to the instruction of the user to complete meal ordering service; accessing the navigation application according to the instruction of the user to complete the navigation task; and accessing the search website according to the instruction of the user to complete the query task and the like.
The intelligent interaction platform 3 stores massive task models, functional components, data packets and the like required by the voice assistant. The task model comprises but is not limited to a task engine, a program module, an installation package, an installation-free package and the like for executing a user task instruction, such as a meal ordering task model, a navigation task model and the like; the functional components are the components for realizing the functions of the voice assistant, and the collection of a plurality of functional components is used for realizing the whole functions of the voice assistant, such as: a speech recognition component, a semantic understanding component, etc.; the data includes, but is not limited to, all data information related to the voice assistant, such as user ID, authentication information of the user, history data, slot information, and the like. The intelligent interaction platform provides an access interface for a user to access, receives task models, functional components, data and the like trained and uploaded by the user, allows the user to download the task models, the functional components, the data and the like, provides information management, sharing and interaction functions, provides an account management function, and allows the user to directly train a voice assistant or train the task models, the functional components and the like required by the voice assistant.
Generally, the interaction between the voice assistants is completed through the intelligent interaction platform, for example, the voice assistant a needs to share the self-trained book borrowing task model with the voice assistant B, the voice assistant a first uploads the book borrowing task model to the intelligent interaction platform, and the voice assistant B downloads the book borrowing task model by accessing the intelligent interaction platform. Further, direct communication can be established between the voice assistants, and the voice assistant A is allowed to directly share the borrowing task model with the voice assistant B. The direct communication may be established directly by the user equipment in which it is located, including but not limited to bluetooth, WiFi, cellular communications, etc.
Further, the voice assistant corresponds to a user ID, one voice assistant corresponds to a user ID, or multiple voice assistants correspond to a user ID. The intelligent interaction platform realizes the management of a plurality of voice assistants by a mode of managing user IDs. When one voice assistant of one user ID updates the task model, the intelligent interaction platform automatically completes the synchronous operation of the task models of other voice assistants corresponding to the user ID. For example: the USER USER123 is provided with three voice assistants which are respectively loaded in a mobile phone, a car machine and a notebook computer of the USER, and when the voice assistant in the mobile phone trains a new book borrowing task model, the book borrowing task model can be automatically loaded into the voice assistants in the car machine and the notebook computer through the intelligent interaction platform. Similarly, when a new meal ordering website is added to the meal ordering task model in the car machine, the user can update the meal ordering task model of the voice assistant in the mobile phone and the notebook computer through the intelligent interaction platform. Through rapid synchronous updating operation, the synchronization among the voice assistants is realized, and the use requirements of users in different occasions are met.
FIG. 2 is a block diagram of a voice assistant according to a first embodiment of the invention. The voice assistant 2 includes at least a human-machine interface 21, a speech recognition module 22, a semantic understanding module 23, a task model training module 24, a transceiving interface 25, and a plurality of task models 26. The man-machine interaction interface 21 is used for interaction between the voice assistant and the user, receiving voice input of the user and outputting a voice instruction to the user; the speech recognition module 22 is configured to recognize speech input of a user, generate a speech recognition result, and input the speech recognition result to the semantic understanding module 23, and the semantic understanding module 23 is configured to perform semantic understanding on the recognition result of the speech recognition module, generate a control instruction, call a task model execution instruction, or trigger the task model training module 24 to start a task model training process; the task model training module 24 performs training, modification, or updating of the task model. The plurality of task models 26 refers to task engines capable of performing different tasks, such as: the order model 261, namely the order task engine, completes the order task, and the navigation model 262, namely the navigation task engine, completes the navigation task. The task model 26 comprises a plurality of task slot positions, when the task model 26 is called, the voice assistant generates a problem by calling the task slot positions, acquires key knowledge data through man-machine conversation or multi-round conversation with a user to fill the task slot positions, generates a task instruction, and interacts with a target website to execute a task. The user can modify, delete, add, update and the like the task module through the voice assistant or through other data connection, so that the task module is closer to the actual requirement of the user.
The voice assistant can be connected with a public network through the user equipment where the voice assistant is located, so that functions of data uploading and downloading, information interaction and the like are achieved, for example, a meal ordering task model is connected with a multi-takeaway website through the public network, and the like. The voice assistant has an extensible interface that can be compatibly connected with the open interface of different websites that need to be accessed. Preferably, the voice assistant corresponds to a user ID, and the user uses the user ID to register on the intelligent interaction platform, so that the intelligent interaction platform can realize the unified management of the voice assistant.
Fig. 3 is a block diagram of an intelligent interaction platform according to a first embodiment of the present invention. The intelligent interaction platform 3 is located at a cloud end or a remote server end, and has a large storage space, and can store and manage the task model 31, namely the task engine. The intelligent interaction platform provides classified storage and management of the task models, for example, classified storage is performed according to the types, application scenes, applicable objects and the like of the task models, sequencing display is performed according to the downloading times, updating time, the sizes of the task models and the like of users, and functions of searching, browsing, demonstrating and the like of the task models can be provided. The task model may be a complete or closed task model, or may be an incomplete task model that provides partial functionality or an extensible task model. The preferred intelligent interaction platform can be configured as a cloud server.
The intelligent interaction platform also stores and manages functional components 32, the functional components can be further divided into general components and special components, the general components refer to non-personalized functional components which can be needed by the voice assistant, and can be used by all users or a certain class of users, such as semantic recognition components, voice recognition components and the like, and the components can be developed and trained by developers and used for direct downloading, installation and use of the users; for some users, the general components can provide basic functions but cannot meet personalized requirements, the intelligent interaction platform can also provide special components, the special components can be trained by the intelligent interaction platform, and the special components can also be further trained into personalized components by the users according to the basic general components for uploading and sharing. The intelligent interactive platform also stores data information 33 required for implementing voice assistant functions, such as user profiles, historical data, slot information, etc.
The intelligent interactive platform is also provided with a task model training module 34, and a user directly trains a task model in the intelligent interactive platform through the task model training module 34 and shares and uses the task model. The method does not need to be loaded to the local, and is suitable for the situation that the voice assistant of the user is not nearby.
The intelligent interactive platform also has a network interface 35 enabling access to and access to a public network. The intelligent interaction platform can provide online voice assistant service, a user does not need to download a voice assistant to the local, and the voice assistant of the intelligent interaction platform can access other websites, platforms or applications through a public network to execute instructions of the user to complete tasks. For example: accessing a meal ordering website according to the instruction of the user to complete meal ordering service; accessing the navigation application according to the instruction of the user to complete the navigation task; and accessing the search website according to the instruction of the user to complete the query task and the like.
Preferably, the intelligent interactive platform has a UI interface 36 for facilitating browsing and access by the user. For example, the intelligent interactive platform is configured as a website.
Furthermore, the intelligent interactive platform also has an open extended function interface 37, which can implement function extension to be compatible with more applications and functions.
Preferably, the intelligent interactive platform realizes remote management of the local voice assistant of the user equipment through the user ID, such as notification, updating, synchronization, upgrading and the like. The account number of the user can also be managed.
The intelligent interaction platform 3 is capable of interconnecting with the user equipment where the voice assistant 2 is located over a network including, but not limited to: WiFi, cellular network, internet, etc.; the intelligent interactive platform 3 can also be connected with a public network such as the internet and the like, and can send information to the public network or download information from the public network. Common users and developers can access the intelligent interaction platform through account login, visitor browsing and other modes. Authorized users can download or upload data from the intelligent interactive platform, such as: download/upload task models, download/upload voice assistant generic components, etc.
Example two
Fig. 4 is a flowchart of a method for training and sharing voice assistants according to a second embodiment of the present invention. The voice assistant training and sharing method of the embodiment comprises the following steps:
104, if not, the voice assistant initiates a search operation to the intelligent interaction platform;
and step 107, the voice assistant saves the created task model and uploads the task model to the intelligent interaction platform.
To facilitate an understanding of the steps of the method, the following is described in detail in the context of a borrowing task, it being understood that the following examples are intended to facilitate an intuitive understanding of the method by those skilled in the art and should not be considered as limiting the method, which may be applied to various types of stand-alone, online or network tasks including borrowing, for example: ordering, navigation, web search, instant messaging, and the like.
when a user needs to borrow a book through a network library, a command can be sent to a voice assistant loaded in the mobile phone: hello, small e, borrow volume 1 from the national library.
after the voice assistant receives the command, key knowledge data of national library, borrow, the emperor of the Galaxy, book 1 are obtained through the voice recognition module and the semantic understanding module.
the step can be that the voice assistant searches the key knowledge data in a task model library, finds a book friend of a book borrowing task model through fuzzy matching, calls the task model, and fills corresponding slot positions in the model by using the key knowledge data, for example, a destination address slot position fills a national library, a title slot position fills a galaxy empire state, a book information slot position fills a book 1, generates a book borrowing command to access a national library website, and uses the existing registration information to complete book borrowing operation. If when filling the slot position, when finding the vacant slot position, can guide the user to accomplish the input through many rounds of dialogues, for example when the destination address slot position leads to the vacancy because can't match, the voice assistant generates the problem: asking where to borrow a book? The user answers the national library to fill the empty slot.
104, if not, the voice assistant initiates a search operation to the intelligent interaction platform;
and if the voice assistant searches the key knowledge data in the task model library and does not find a book borrowing task model, generating a search command word according to the key knowledge data, automatically logging in the intelligent interaction platform and initiating a search operation. And searching in a task model of the intelligent interaction platform.
and if the corresponding task model is found in the task model library of the intelligent interaction platform, loading the task model which is matched most to the voice assistant. In this step, the search result may not be unique, and the intelligent interaction platform may push the optimal task model through calculation, or may send names and function descriptions of a plurality of task models with higher matching degrees to the voice assistant, and select a task model that the user desires to load in a man-machine conversation manner. After the task model is loaded, the voice assistant calls the task model to execute a book borrowing task.
if the intelligent interaction platform does not return the search result, or the returned search result has low matching degree, or the returned result does not meet the requirement of the user, the voice assistant triggers the task model training module 24 to initiate the task model creating process. In the process, basic slot position information associated with a task to be created in the task model training module 24 is called, basic slot positions are filled in a multi-round conversation mode, key information in one or more basic slot positions is obtained, a specific slot position is obtained according to the key information, multi-round conversations are initiated again, and a user is guided to fill the specific slot position to generate a task model. The training process of the task model is specifically referred to in the third embodiment.
And step 107, the voice assistant saves the created task model and uploads the task model to the intelligent interaction platform.
After the book borrowing task model is created, the created book borrowing task model is stored in a task model of the voice assistant, and the user can also upload the book borrowing task model to the intelligent interaction platform to share the book borrowing task model to other authorized users. Preferably, the user who creates the borrowing task model can obtain higher authority, such as management authority for updating, upgrading, expanding, downloading and the like of the task model. And the voice assistant executes the book borrowing task according to the created book borrowing task model.
EXAMPLE III
In the training and sharing method of the voice assistant, when neither the voice assistant nor the intelligent interaction platform meets the task model required by the user, the voice assistant initiates a task model creating process, the creating process is a process for training the task model, and a task model training module of the voice assistant completes the training of the task model. The embodiment discloses a training method of a voice assistant, which is applied to the voice assistant or an intelligent interaction platform to train a task model for executing a specific task. The training method of the voice assistant comprises the following steps:
step 106a, calling a basic slot position to generate a problem;
106b, asking questions of the user through multiple rounds of conversations, acquiring answers of the user, extracting key knowledge data, and filling basic slot positions;
106c, acquiring specific slot position information according to the key knowledge data of the filled one or more basic slot positions;
step 106d, generating a question according to the specific slot position, asking a question to the user through a plurality of rounds of conversations, obtaining an answer of the user, extracting key knowledge data, and filling the specific slot position;
and 106e, generating a task model.
Still taking the book borrowing task scenario as an example, when a user wants to borrow book 1 of the 'silver river empire state' through a digital library, a command is sent to the voice assistant, the voice assistant does not find a task model meeting the requirements in the local and intelligent interaction platforms, and then the voice assistant triggers the task model training module to initiate the process of creating the book borrowing task model. The voice assistant starts a training process of the task model by calling basic slot position information in a task model training module; the task model training module stores logically related basic slot positions, wherein the basic slot positions refer to slot positions corresponding to the most basic information required by a common task model creating process. For example: task name, destination address, task object, etc. The task model training module stores the most basic information for training the task model, and not only includes the basic slot position information, but also includes the logic relationship or incidence relationship between the basic slot position information, the preferable basic slot position and incidence relationship are created, maintained and updated in the intelligent interaction platform by developers, and when the basic slot position information in the intelligent interaction platform is updated, the intelligent interaction platform notifies the voice assistant or directly pushes the update.
And the task model training module sequentially calls the basic slot position information according to the incidence relation to generate a question of man-machine conversation, asks the user in a multi-round conversation mode, and acquires key knowledge data according to the answer of the user to fill the basic slot position. The multiple rounds of dialog may include the following:
the voice assistant: what is the task name?
The user: book borrowing device
The voice assistant: what is the destination address of the task?
The user: national library website/APP
The target address of the task is to trigger to obtain key information of the specific slot, the voice assistant automatically accesses the national library website or APP according to the key information, and reads the specific slot information with logic association required by book borrowing, for example, the read specific slot information includes a user name, a password, a book name and identity information. And the voice assistant sequentially generates questions according to the specific slot position information, continuously asks questions to the user, and extracts key knowledge data from the obtained answers to fill the specific slot position.
The voice assistant: what is the user name?
The user: USER123
The voice assistant: what is the password?
The user: 12ab
The voice assistant: what is the name of the book?
The user: galaxy empire 1
The voice assistant: what is the identity information?
The user: *************
Through multiple rounds of conversations, the voice assistant fills corresponding specific slots according to the acquired key knowledge data to generate a book borrowing task model. The book borrowing task model can only contain basic slot position information, and can also contain key knowledge data of partial or complete filling slot positions.
The voice assistant further inquires whether the user stores the read slot position information into a task model training module, and if so, the read slot position information is stored into the task model training module for use when a new task model is trained again.
The slot position in the task model training module can be further divided into a general slot position and a special slot position according to the user intention, the general slot position corresponds to the general intention, and the general intention refers to some general instructions during slot position establishment or slot position filling, for example: the confirmation, cancellation and the like are general intents, and the corresponding slot positions are general slot positions. And specific instructions exist for a specific task model and are used for realizing the special intention of a user, the corresponding slot position is the special slot position, in the book borrowing task model, book borrowing is the special intention, and the corresponding slot position is the special slot position.
Preferably, when storing the slot position information, the task model training module performs classified storage according to the basic slot position, the general slot position and the special slot position, so that management and calling of the slot position information are facilitated.
Preferably, in the task model training process, after the user inputs an instruction through voice, the voice assistant firstly identifies whether the intention of the user belongs to a general intention or a special intention, then establishes or calls corresponding slot position information, and guides the user to fill in the slot position according to multiple rounds of conversations to complete establishment of the task model.
The training of the task model may not be completed at one time, for example, when the flow required for executing the task changes, or the task model cannot complete the task, the training of the task needs to be continued, and the speech assistant training method of the present invention further includes the following steps: 106f, establishing an expansion slot position, and generating a problem according to the expansion slot position;
step 106g, filling the expansion slot position;
and step 106h, updating the task model and uploading the task model to the intelligent interaction platform.
Preferably, in step 106g, key knowledge data is acquired by using a man-machine conversation or multiple rounds of conversations to fill the expansion slot; or, the stored user information is obtained according to authorization to fill the expansion slot position; or calling a nesting submodel to fill the extended slot.
Still take the book borrowing scenario as an example: the voice assistant finds that a mobile phone verification link is added in a book borrowing process of the library website, judges that slot position information needs to be expanded according to the found voice assistant, reads a mobile phone number, and sends a verification code and three expansion slot positions of the verification code, then initiates a new multi-turn conversation to the user, guides the user to fill in the expansion slot positions, and perfects a task model. The voice assistant constructs corresponding questions according to the mobile phone number, whether the verification code is sent or not, the three expansion slot positions of the verification code and the association relationship of the three expansion slot positions, and sequentially asks questions to the user. Question one is what is the cell phone number? Filling in mobile phone number information according to the answer of the user; the second problem is that: is the authentication code transmitted? Selecting to send the verification code according to the user response; the third problem is that: what is the authentication code? And filling the verification code slot position according to the answer of the user, or automatically reading the verification code filling verification code slot position. And generating further instruction information according to the extended slot position, and finishing the book borrowing process by interacting with the library website or the APP. And after the book borrowing process is completed, the three expansion slot positions are supplemented into the task model, the task model is updated, and the task model is uploaded to the intelligent interaction platform.
Nesting of task models may exist in the slot expansion process, that is, other task models need to be called to obtain key knowledge data for filling the expansion slot. For example: the voice assistant finds that a fingerprint verification link is added in the book borrowing process of the library website, firstly, the voice assistant creates an expansion slot position corresponding to a 'fingerprint', generates a problem and asks a question to a user: what is the fingerprint? The user answers: and calling a camera, or calling a camera task model when the voice assistant senses that the user places a finger at the position of the camera, acquiring fingerprint information of the user, and filling the user slot position. And generating a corresponding instruction after the slot position is filled, and completing the book borrowing task. At the moment, the camera task model is nested in the book borrowing task model, and the camera task model is a nested sub-model of the book borrowing task model. When the task model which can be used as a nesting submodel is stored, a general task model label can be added, and a logic interface is established so as to be more conveniently nested in other task models.
Example four
The embodiment discloses a using method of a voice assistant. After the training of the task model of the voice assistant is finished, the task model is stored in the voice assistant, and the task model is called to execute a command when a user gives a task instruction. For example: after the book borrowing task model is successfully created, a book borrowing task model exists in a task model library of the voice assistant. When the user needs to borrow the book, the voice assistant is directly given an instruction: hi, i want to borrow a book. The voice assistant recognizes the special intention of book borrowing, searches in the task model library to obtain a book friend task model, calls the model, starts the following multi-turn conversation process, guides the user to fill a plurality of slot position information corresponding to the task model,
the voice assistant: what book you want to borrow?
The user: first volume of the Galaxy empire
The voice assistant: is a borrowing from a national library?
The user: is
The voice assistant: is the username the default?
The user: is
The voice assistant: is the password default?
The user: is
After the slot position information is filled, a book borrowing instruction is generated, and the voice assistant realizes a book borrowing command through the butt joint with a website or APP of a national library.
When executing a task, it may be found that there is a change in information required to complete the task, at which time a slot update or slot expansion needs to be performed. The voice assistant finds that a mobile phone verification link is added in a book borrowing process of the library website, judges that slot position information needs to be expanded according to the found voice assistant, reads a mobile phone number, and sends a verification code and three expansion slot positions of the verification code, then initiates a new multi-turn conversation to the user, guides the user to fill in the expansion slot positions, and perfects a task model. The voice assistant constructs corresponding questions according to the mobile phone number, whether the verification code is sent or not, the three expansion slot positions of the verification code and the association relationship of the three expansion slot positions, and sequentially asks questions to the user. Question one is what is the cell phone number? Filling in mobile phone number information according to the answer of the user; the second problem is that: is the authentication code transmitted? Selecting to send the verification code according to the user response; the third problem is that: what is the authentication code? And filling the verification code slot position according to the answer of the user, or automatically reading the verification code filling verification code slot position. And generating further instruction information according to the extended slot position, and finishing the book borrowing process by interacting with the library website or the APP. And after the book borrowing process is completed, the three expansion slot positions are supplemented into the task model, the book friend task model is updated, and the updated book friend task model is uploaded to the intelligent interaction platform.
When executing a task, it may be found that there is a change in information required to complete the task, at which time a slot update or slot expansion needs to be performed. Nesting of task models may exist in the slot expansion process, that is, other task models need to be called to obtain key knowledge data for filling the expansion slot. For example: the voice assistant finds that a fingerprint verification link is added in the book borrowing process of the library website, firstly, the voice assistant creates an expansion slot position corresponding to a 'fingerprint', generates a problem and asks a question to a user: what is the fingerprint? The user answers: and calling a camera, or calling a camera task model when the voice assistant senses that the user places a finger at the position of the camera, acquiring fingerprint information of the user, and filling the user slot position. And generating a corresponding instruction after the slot position is filled, and completing the book borrowing task. At the moment, the camera task model is nested in the book borrowing task model, and the camera task model is a nested sub-model of the book borrowing task model. When the task model which can be used as a nesting submodel is stored, a general task model label can be added, and a logic interface is established so as to be more conveniently nested in other task models.
The task model related in the embodiment of the present invention at least includes the minimum slot position information required for completing the task, including the basic slot position, the specific slot position, and in some cases, the extended slot position. Some task models may also include key knowledge data for slot filling, so that the voice assistant simplifies the slot filling process when invoking the task model. For the task model containing the key knowledge data, the sharing level can be correspondingly set according to which key knowledge data are contained. For example, a task model containing user name and password information is only shared with family members.
An embodiment of the present invention further provides a user equipment, where the user equipment includes a processor and a memory, where the memory stores a computer program that is executable on the processor, and the computer program, when executed by the processor, implements the voice assistant training and sharing method, the voice assistant training method, and the voice assistant using method described above.
An embodiment of the present invention further provides a computer-readable storage medium, where a computer program operable on a processor is stored in the computer-readable storage medium, and when the computer program is executed, the method for training and sharing a voice assistant, the method for training a voice assistant, and the method for using a voice assistant are implemented as described above.
Any combination of one or more computer-readable media may be employed. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. 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. The computer-readable storage medium may include: 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), a flash memory, an erasable programmable read-only memory (EPROM), 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 context of this document, 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.
Computer program code for carrying out operations of the present invention may be written in one or more programming languages, or a combination thereof.
The above description is only an example for the convenience of understanding the present invention, and is not intended to limit the scope of the present invention. In the specific implementation, a person skilled in the art may change, add, or reduce the components of the apparatus according to the actual situation, and may change, add, reduce, or change the order of the steps of the method according to the actual situation without affecting the functions implemented by the method.
While embodiments of the invention have been shown and described, it will be understood by those skilled in the art that: various changes, modifications, substitutions and alterations can be made to the embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents, and all changes that come within the meaning and range of equivalency of the claims are to be embraced within their scope.
Claims (5)
1. A voice assistant training and sharing system, the system comprising: the intelligent interaction platform is located at the cloud end, the voice assistants are loaded in local user equipment, and a user shares a task model to the voice assistants of other users through the intelligent interaction platform, and the intelligent interaction platform is characterized in that:
the intelligent interaction platform comprises: the task model training module is used for training a corresponding task model according to a user instruction;
the task model is used for executing the instructions of the user and completing the task;
the functional component is used for realizing the functions of the voice assistant;
the data storage module is used for storing data information related to the voice assistant;
the extended function module is used for realizing extended functions;
a network interface for implementing network connection;
the voice assistant comprises:
the human-computer interaction interface is used for realizing voice interaction with a user;
the task model training module is used for training a task model according to a user instruction;
the task model is used for executing the instructions of the user and completing the task;
the voice assistant further comprises: the device comprises a voice recognition module, a semantic understanding module and a transceiving interface.
2. The voice assistant training and sharing system of claim 1 wherein: the intelligent interaction platform realizes the management of the voice assistants through user IDs, and one user ID corresponds to one or more voice assistants; when one user ID corresponds to a plurality of voice assistants, if one voice assistant is updated, the intelligent interaction platform initiates synchronous updating of other voice assistants of the same user ID.
3. The voice assistant training and sharing system of claim 1 wherein: after the voice assistant trains the task model, the task model is shared to an intelligent interaction platform for other voice assistants to download; or after the voice assistant trains the task model, the task model is directly shared to other voice assistants through the user equipment.
4. A voice assistant training method is executed by a voice assistant, and is characterized by comprising the following steps:
step 106a, calling a basic slot position to generate a problem;
106b, asking questions of the user through multiple rounds of conversations, acquiring answers of the user, extracting key knowledge data, and filling basic slot positions;
106c, acquiring specific slot position information according to the key knowledge data of the filled one or more basic slot positions;
step 106d, generating a question according to the specific slot position, asking a question to the user through a plurality of rounds of conversations, obtaining an answer of the user, extracting key knowledge data, and filling the specific slot position;
step 106e, generating a task model;
106f, establishing an expansion slot position, and generating a problem according to the expansion slot position;
step 106g, filling the expansion slot position;
and step 106h, updating the task model and uploading the task model to the intelligent interaction platform.
5. The voice assistant training method of claim 4, wherein: step 106g, filling the extended slot, including:
obtaining key knowledge data by using man-machine conversation or multi-round conversation to fill the expansion slot position;
or, the stored user information is obtained according to authorization to fill the expansion slot position;
or calling a nesting submodel to fill the extended slot.
Applications Claiming Priority (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910141268 | 2019-02-26 | ||
| CN201910141268X | 2019-02-26 | ||
| CN201910706657.2A CN110310630B (en) | 2019-02-26 | 2019-08-01 | Training and sharing method of voice assistant |
Related Parent Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910706657.2A Division CN110310630B (en) | 2019-02-26 | 2019-08-01 | Training and sharing method of voice assistant |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112908311A true CN112908311A (en) | 2021-06-04 |
Family
ID=68082871
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110094701.6A Pending CN112908311A (en) | 2019-02-26 | 2019-08-01 | Training and sharing method of voice assistant |
| CN201910706657.2A Active CN110310630B (en) | 2019-02-26 | 2019-08-01 | Training and sharing method of voice assistant |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910706657.2A Active CN110310630B (en) | 2019-02-26 | 2019-08-01 | Training and sharing method of voice assistant |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN112908311A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110310641A (en) * | 2019-02-26 | 2019-10-08 | 北京蓦然认知科技有限公司 | A kind of method and device for voice assistant |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110941693A (en) * | 2019-10-09 | 2020-03-31 | 深圳软通动力信息技术有限公司 | Task-based man-machine conversation method, system, electronic equipment and storage medium |
| CN110826481A (en) * | 2019-11-01 | 2020-02-21 | 北京云迹科技有限公司 | Data processing method, commodity identification method, server and storage medium |
| CN111124649B (en) * | 2019-12-26 | 2023-04-18 | 杭州蓦然认知科技有限公司 | Method and device for generating APP ecosystem |
| CN111026538B (en) * | 2019-12-26 | 2023-04-14 | 杭州蓦然认知科技有限公司 | APP ecosystem establishing and using method and device |
| CN114267356B (en) * | 2021-12-30 | 2024-04-02 | 重庆特斯联智慧科技股份有限公司 | Man-machine interaction logistics robot and control method thereof |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070130186A1 (en) * | 2005-12-05 | 2007-06-07 | Microsoft Corporation | Automatic task creation and execution using browser helper objects |
| CN106598538A (en) * | 2016-11-29 | 2017-04-26 | 努比亚技术有限公司 | Method and system for updating instruction set |
| CN107665704A (en) * | 2016-07-29 | 2018-02-06 | 科大讯飞股份有限公司 | Phonetic order detection model construction method, detection method and system, man-machine interaction method and equipment |
| CN108847229A (en) * | 2018-05-23 | 2018-11-20 | 上海爱优威软件开发有限公司 | A kind of information interacting method and terminal based on voice assistant |
| CN108962244A (en) * | 2018-06-29 | 2018-12-07 | 百度在线网络技术（北京）有限公司 | Method and apparatus for sending information |
| CN109086282A (en) * | 2017-06-14 | 2018-12-25 | 杭州方得智能科技有限公司 | A kind of method and system for the more wheels dialogue having multitask driving capability |
| CN109241250A (en) * | 2018-07-25 | 2019-01-18 | 南京瓦尔基里网络科技有限公司 | A kind of dialogue of policing rule promotes and intention method of discrimination and system |
| CN109256122A (en) * | 2018-09-05 | 2019-01-22 | 深圳追科技有限公司 | machine learning method, device, equipment and storage medium |
Family Cites Families (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7664644B1 (en) * | 2006-06-09 | 2010-02-16 | At&T Intellectual Property Ii, L.P. | Multitask learning for spoken language understanding |
| US8086443B2 (en) * | 2007-08-24 | 2011-12-27 | Siemens Medical Solutions Usa, Inc. | System and method for text tagging and segmentation using a generative/discriminative hybrid hidden markov model |
| CN103813003B (en) * | 2012-11-15 | 2016-06-08 | 三星电子（中国）研发中心 | Data sharing method and the mobile terminal of mobile terminal in call |
| AU2014233517B2 (en) * | 2013-03-15 | 2017-05-25 | Apple Inc. | Training an at least partial voice command system |
| CN103646646B (en) * | 2013-11-27 | 2018-08-31 | 联想(北京)有限公司 | A kind of sound control method and electronic equipment |
| US10659851B2 (en) * | 2014-06-30 | 2020-05-19 | Apple Inc. | Real-time digital assistant knowledge updates |
| US10446141B2 (en) * | 2014-08-28 | 2019-10-15 | Apple Inc. | Automatic speech recognition based on user feedback |
| US9646609B2 (en) * | 2014-09-30 | 2017-05-09 | Apple Inc. | Caching apparatus for serving phonetic pronunciations |
| WO2016054230A1 (en) * | 2014-10-01 | 2016-04-07 | XBrain, Inc. | Voice and connection platform |
| CN104881678A (en) * | 2015-05-11 | 2015-09-02 | 中国科学技术大学 | Multitask learning method of model and characteristic united learning |
| US10884503B2 (en) * | 2015-12-07 | 2021-01-05 | Sri International | VPA with integrated object recognition and facial expression recognition |
| US10417346B2 (en) * | 2016-01-23 | 2019-09-17 | Microsoft Technology Licensing, Llc | Tool for facilitating the development of new language understanding scenarios |
| US10635281B2 (en) * | 2016-02-12 | 2020-04-28 | Microsoft Technology Licensing, Llc | Natural language task completion platform authoring for third party experiences |
| US10847135B2 (en) * | 2017-05-18 | 2020-11-24 | Aiqudo, Inc. | Sharing commands and command groups across digital assistant operations |
| US20180366108A1 (en) * | 2017-05-18 | 2018-12-20 | Aiqudo, Inc. | Crowdsourced training for commands matching |
| CN107733722B (en) * | 2017-11-16 | 2021-07-20 | 百度在线网络技术（北京）有限公司 | Method and apparatus for configuring voice service |
| CN108564946B (en) * | 2018-03-16 | 2019-09-20 | 苏州思必驰信息科技有限公司 | Technical ability, the method and system of voice dialogue product are created in voice dialogue platform |
| CN109120774A (en) * | 2018-06-29 | 2019-01-01 | 深圳市九洲电器有限公司 | Terminal applies voice control method and system |
| CN108984157B (en) * | 2018-07-27 | 2022-01-11 | 思必驰科技股份有限公司 | Skill configuration and calling method and system for voice conversation platform |
| CN109246467A (en) * | 2018-08-15 | 2019-01-18 | 上海蔚来汽车有限公司 | Label is to the method, apparatus of sharing video frequency, video camera and smart phone |
-
2019
- 2019-08-01 CN CN202110094701.6A patent/CN112908311A/en active Pending
- 2019-08-01 CN CN201910706657.2A patent/CN110310630B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070130186A1 (en) * | 2005-12-05 | 2007-06-07 | Microsoft Corporation | Automatic task creation and execution using browser helper objects |
| CN107665704A (en) * | 2016-07-29 | 2018-02-06 | 科大讯飞股份有限公司 | Phonetic order detection model construction method, detection method and system, man-machine interaction method and equipment |
| CN106598538A (en) * | 2016-11-29 | 2017-04-26 | 努比亚技术有限公司 | Method and system for updating instruction set |
| CN109086282A (en) * | 2017-06-14 | 2018-12-25 | 杭州方得智能科技有限公司 | A kind of method and system for the more wheels dialogue having multitask driving capability |
| CN108847229A (en) * | 2018-05-23 | 2018-11-20 | 上海爱优威软件开发有限公司 | A kind of information interacting method and terminal based on voice assistant |
| CN108962244A (en) * | 2018-06-29 | 2018-12-07 | 百度在线网络技术（北京）有限公司 | Method and apparatus for sending information |
| CN109241250A (en) * | 2018-07-25 | 2019-01-18 | 南京瓦尔基里网络科技有限公司 | A kind of dialogue of policing rule promotes and intention method of discrimination and system |
| CN109256122A (en) * | 2018-09-05 | 2019-01-22 | 深圳追科技有限公司 | machine learning method, device, equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 易定: "基于Agent的语音交互界面模型与应用", 《微型电脑应用》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110310641A (en) * | 2019-02-26 | 2019-10-08 | 北京蓦然认知科技有限公司 | A kind of method and device for voice assistant |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110310630A (en) | 2019-10-08 |
| CN110310630B (en) | 2021-02-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110310630B (en) | Training and sharing method of voice assistant | |
| US9172747B2 (en) | System and methods for virtual assistant networks | |
| US8380047B2 (en) | Video editing system, video editing server and communication terminal | |
| WO2020114368A1 (en) | Man-machine interaction system and multi-task processing method in man-machine interaction system | |
| CN109240670A (en) | Modular software development methodology, system, equipment and medium | |
| CN106445951B (en) | File transmission method and device | |
| CN106164909A (en) | The task of natural language input completes | |
| US11570253B1 (en) | Method of adapting a user interface on a mobile communication device based on different environments | |
| CN112699257A (en) | Method, device, terminal, server and system for generating and editing works | |
| US10078692B2 (en) | Method and system for providing a social service based on music information | |
| US20180190292A1 (en) | Voice recognition system and construction method thereof | |
| CN108776592B (en) | Page construction method, device, equipment and storage medium | |
| CN106201610A (en) | Web application accesses the method and device of the primary function of terminal | |
| WO2008121533A1 (en) | Configuration management of an electronic device | |
| JP6619488B2 (en) | Continuous conversation function in artificial intelligence equipment | |
| CN111598428B (en) | Management method and device of flow node, storage medium and server | |
| US20180336050A1 (en) | Action recipes for a crowdsourced digital assistant system | |
| KR101943430B1 (en) | User Device, Driving Method of User Device, Apparatus for Providing Service and Driving Method of Apparatus for Providing Service | |
| CN112243016B (en) | Middleware platform, terminal equipment, 5G artificial intelligence cloud processing system and processing method | |
| CN110209392A (en) | Technical ability shares development approach and device | |
| CN111367561B (en) | Remote development method and device for software program | |
| KR20130023490A (en) | System and method for synchronizing applications | |
| US11757976B2 (en) | Unified application management for heterogeneous application delivery | |
| CN117435215A (en) | Development environment deployment method, system, computer device and storage medium | |
| CN110855750B (en) | Downloading method of software development kit and related equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220113 Address after: 310024 floor 5, zone 2, building 3, Hangzhou cloud computing Industrial Park, Zhuantang street, Xihu District, Hangzhou City, Zhejiang Province Applicant after: Hangzhou suddenly Cognitive Technology Co.,Ltd. Address before: Room 401, gate 2, block a, Zhongguancun 768 Creative Industry Park, 5 Xueyuan Road, Haidian District, Beijing 100083 Applicant before: BEIJING MORAN COGNITIVE TECHNOLOGY Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210604 |