WO2020143381A1 - 对话处理方法及设备 - Google Patents

对话处理方法及设备 Download PDF

Info

Publication number
WO2020143381A1
WO2020143381A1 PCT/CN2019/124441 CN2019124441W WO2020143381A1 WO 2020143381 A1 WO2020143381 A1 WO 2020143381A1 CN 2019124441 W CN2019124441 W CN 2019124441W WO 2020143381 A1 WO2020143381 A1 WO 2020143381A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
information
slot
slot type
dialog
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.)
Ceased
Application number
PCT/CN2019/124441
Other languages
English (en)
French (fr)
Inventor
罗红枫
张轶博
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huawei Technologies Co Ltd
Original Assignee
Huawei Technologies Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Huawei Technologies Co Ltd filed Critical Huawei Technologies Co Ltd
Priority to EP19909437.6A priority Critical patent/EP3889799A4/en
Publication of WO2020143381A1 publication Critical patent/WO2020143381A1/zh
Priority to US17/371,377 priority patent/US11977815B2/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/01—Protocols
    • H04L67/12—Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16—Sound input; Sound output
    • G06F3/167—Audio in a user interface, e.g. using voice commands for navigating, audio feedback
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/02—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/903—Querying
    • G06F16/9032—Query formulation
    • G06F16/90332—Natural language query formulation or dialogue systems
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/20—Natural language analysis
    • G06F40/205—Parsing
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/04—Real-time or near real-time messaging, e.g. instant messaging [IM]

Definitions

  • This application relates to the field of computers, and in particular to a dialog processing method and device.
  • the human-machine dialogue system (hereinafter referred to as the dialogue system) can respond to the user's input voice feedback information to the user, the dialogue system may include task-oriented dialogue systems (task-oriented systems) and non-task-oriented dialogue systems (non-task-oriented systems) .
  • the non-task dialogue system may be a chat dialogue system.
  • the task-based dialogue system can trigger the determined service according to the user's voice and feed back the result of the execution of the corresponding service to the user.
  • the service dialogue system may correspond to a target server, and the target server is used to provide and execute one or more services. Based on this, after receiving the voice input by the user, the dialogue system can obtain one or more required slot information from the voice.
  • the required slot information is information necessary to indicate a service. For example, the required slot information for booking a ticket includes the origin, departure time and destination.
  • the dialogue system can detect whether the one or more required slot information is included in the preset slot information of the dialogue system, if the one or more required slot information is included in the dialogue system In the set slot information, the dialogue system requests the target server to execute the target service according to the one or more required slot information.
  • the target server detects whether the target service is included in the one or more services in response to the request.
  • the target server executes the target service. Then, the dialogue system can feed back information about the successful execution of the target service to the user. In addition, if any one of the one or more required slot information is not included in the preset slot information of the dialog system, or if the target service is not included in the one or more services , The dialogue system may feed back information about the failed execution of the target service to the user. Then, the user may continue to input voices of other contents so that a certain service can be successfully executed.
  • the present application provides a dialogue processing method and device, which solves the problem of using the existing dialogue processing method and having many dialogue rounds between the user and the dialogue system.
  • the present application provides a dialog processing method, the method includes:
  • the dialogue processing device receives dialogue information from the user equipment
  • the dialog processing device obtains a service identifier set corresponding to the first slot type from the server, the first slot The type indicates the slot type to which the slot information of the service identifier in the required slot information belongs, and the service identifier set includes multiple service identifiers corresponding to the first slot type;
  • the conversation processing device sends the service identification set to the user equipment
  • the dialog processing device If a target service identifier is received from the user equipment, the dialog processing device requests the server for a service corresponding to the target service identifier, and the target service identifier belongs to the service identifier set;
  • the dialog processing device sends the execution success information to the user equipment.
  • the dialog processing device feeds back the multiple service identifiers maintained by the server to the user, so that the user enters the corresponding slot
  • the information must correspond to a service item in the server. Not only can the service items be displayed to the user in time, but also the user can be prevented from initiating multiple rounds of dialogue with the dialogue processing device, and the efficiency of executing the service can be improved.
  • the method includes:
  • the dialogue processing device parses the dialogue information
  • the dialog processing device determines that the dialog information does not include the slot information corresponding to the first slot type that can determine the service .
  • the method includes:
  • the dialog processing device parses the dialog information to obtain slot information corresponding to the first slot type
  • the dialog processing device requests the server for a service corresponding to the slot information
  • the dialog processing device determines that the dialog information does not include slot information corresponding to the first slot type that can determine the service.
  • the slot information preset in the dialog processing device does not correspond to the required slot information input by the user, and/or, the slot information preset in the dialog processing device and the services included in the server
  • the dialogue processing device can provide effective information and/or processing strategies in a timely manner, which not only can reduce unnecessary dialogue rounds, but also has flexible processing methods, which can improve the user experience.
  • the method further includes:
  • the dialog processing device parses the dialog information to obtain slot information corresponding to the first slot type
  • the dialog processing device requests the server for a service corresponding to the slot information
  • the dialog processing device adds the slot information to the first slot type Corresponding information library.
  • the conversation processing device not only can the conversation processing device and the user be reduced in the number of conversation rounds, and to provide services to the user in time, but also, the conversation processing device can timely update the slot information maintained by the conversation processing device.
  • the dialog processing device before the dialog processing device receives the dialog information from the user equipment, it includes:
  • the dialog processing device receives a tagging instruction, and the tagging instruction corresponds to the first slot type;
  • the dialog processing device adds a label to the first slot type.
  • the required slot information includes the slot information indicating the service identifier, for example, the "ride” in the call service, and the conventional slot information used to determine the service, For example, the destination is "Tsinghua West Gate".
  • the service provided by the server to the user is usually customized by the relevant service personnel, so the service provided by the server may be changed, for example, the "Shunfeng" service item is off the shelf, and the "carpooling" service item is on the shelf, resulting in no corresponding response.
  • the service corresponding to the service ID is usually customized by the relevant service personnel, so the service provided by the server may be changed, for example, the "Shunfeng" service item is off the shelf, and the "carpooling" service item is on the shelf, resulting in no corresponding response.
  • the service corresponding to the service ID is usually customized by the relevant service personnel, so the service provided by the server may be changed, for example, the "Shunfeng" service item is off the shelf, and the "carpooling" service item is on the
  • the application may pre-label the first slot type by adding a label according to the attributes of the above-mentioned mandatory slot type, so as Provide a basis for the realization of the technical solution of this application.
  • the dialogue information is text or audio.
  • the present application provides a dialog processing method, which includes:
  • the user equipment sends dialogue information to the dialogue processing device
  • the user equipment receives and prompts a service identifier set corresponding to a first slot type from the dialog processing device, where the first slot type indicates a slot to which the slot information used to indicate the service identifier in the required slot information belongs Bit type, the service identifier set includes multiple service identifiers corresponding to the first slot type;
  • the user equipment sends a target service identifier to the dialog processing device, where the target service identifier belongs to the service identifier set;
  • the user equipment receives information of successful service execution from the dialog processing device, and the service is a service corresponding to the target service identifier.
  • the user can obtain the desired service. This not only prevents the user from initiating multiple rounds of conversations, but also enables the user to be informed of multiple services provided by the server in time, thereby improving the user experience.
  • the user equipment prompting the service identifier set corresponding to the first slot type includes:
  • the user equipment prompts all service identifiers in the service identifier set.
  • the user equipment prompting the service identifier set corresponding to the first slot type includes:
  • the user equipment prompts to link an entrance identifier of the service identifier set
  • the user equipment receives an opening instruction input by a user
  • the user equipment prompts the page linked to the entrance identifier, and the page includes all service identifiers in the service identifier set.
  • the user equipment prompting the service identifier set corresponding to the first slot type includes:
  • the user equipment prompts a menu control, and the menu control prompts a part of the service identifier in the service identifier set;
  • the user equipment receives a prompt instruction input by a user
  • the user equipment switches the service identifier prompted in the menu control.
  • the service identifier corresponding to the first slot type may be two or three. In other embodiments, there may be more than a dozen service identifiers corresponding to the first slot type. With this implementation, it is possible to flexibly set the service identification prompting method, thereby improving the user experience.
  • the present application provides a dialog processing method.
  • the method includes: a dialog processing device receives dialog information from a user device, and then determines keywords corresponding to all required slot types in the dialog information. If any keyword corresponding to all the required slot types corresponds to at least two slot information in the corresponding slot type, the dialogue processing device determines that the at least two slot information corresponds to the optimal service plan Specific slot information. Furthermore, the dialog processing device may request the server for the service corresponding to the specific slot information, and after receiving the information of the successful execution of the corresponding service from the server, send the information of the successful execution of the corresponding service to the user device.
  • the dialogue processing device can automatically provide users with relatively superior services when the meaning of the keywords entered by the user is ambiguous. Not only can they avoid multiple rounds of dialogue with the user, but also the operation is flexible, which can further improve the user’s Use experience.
  • the present application provides a dialog processing method, which includes:
  • the dialogue processing device receives dialogue information from the user equipment
  • the conversation processing device requests the service corresponding to the slot information from the first server;
  • the dialog processing device If the service execution failure information received from the first server, the dialog processing device requests the second server for the service corresponding to the slot information;
  • the dialog processing device sends the user equipment success execution information to the user equipment.
  • the dialog processing device can trigger the second server to provide the corresponding service. This can reduce the number of dialogue rounds between the dialogue processing device and the user, and provide users with services in a timely manner.
  • the present application provides a dialog processing method, which includes:
  • the dialog information does not include slot information that can determine the first slot type of the service, obtain a service identifier set corresponding to the first slot type from a local application or server, and the first slot type indicates In the required slot information used to indicate the slot type to which the slot information of the service identifier belongs, the service identifier set includes multiple service identifiers corresponding to the first slot type;
  • the method after receiving the dialog information input by the user, before acquiring the service identification set corresponding to the first slot type from a local application or a server, the method includes:
  • the dialog information does not include slot information corresponding to the first slot type that can determine the service.
  • the method after receiving the dialog information input by the user, before acquiring the service identification set corresponding to the first slot type from a local application or a server, the method includes:
  • the dialog information does not include slot information corresponding to the first slot type that can determine the service.
  • the acquiring the service identifier set corresponding to the first slot type from a local application or a server includes:
  • prompting the service identifier in the service identifier set includes:
  • the page includes all service identifications in the service identification set; or,
  • the menu control prompts a part of service identifiers in the service identifier set;
  • the service identifier prompted in the menu control is switched.
  • the present application provides a dialog processing device that has a function to implement the behavior of a radio access network device in the above method.
  • the functions can be realized by hardware, or can also be realized by hardware executing corresponding software.
  • the hardware or software includes one or more modules corresponding to the above functions.
  • the structure of the radio access network device includes a processor and a transceiver, and the processor is configured to process the dialog processing device to perform the corresponding function in the above method.
  • the transceiver is used to realize the information interaction between the dialog processing device, the user equipment and the server.
  • the dialog processing device may further include a memory for coupling with a processor, which stores necessary program instructions and data of the dialog processing device.
  • the present application provides a user equipment, the user equipment having a function to implement the behavior of the user equipment in the above method.
  • the functions can be realized by hardware, or can also be realized by hardware executing corresponding software.
  • the hardware or software includes one or more modules corresponding to the above functions.
  • the structure of the user equipment includes a processor and a transceiver, and the processor is configured to process the user equipment to perform the corresponding function in the above method.
  • the transceiver is used to implement the communication between the user equipment and the conversation processing device.
  • the user equipment may further include a memory for coupling with the processor, which stores necessary program instructions and data of the user equipment.
  • the present application provides a human-computer interaction system.
  • the human-computer interaction system includes a user equipment, a conversation processing device, and a server.
  • the conversation processing device communicates with the user equipment and the server, respectively, to complete the first
  • the man-machine dialogue operation described in any one of the possible implementation manners of the aspect, the second aspect, the third aspect, the first aspect, and the various possible implementation manners of the second aspect.
  • the dialog processing device is the dialog processing device according to the fifth aspect
  • the user equipment is the user device according to the sixth aspect.
  • the present application provides a human-computer interaction system.
  • the human-computer interaction system includes a user equipment, a dialog processing device, a first server, and a second server.
  • the dialog processing device is in communication with the user equipment and the first server, respectively. Communicate with the second server to complete the man-machine dialogue operation described in the fourth aspect above.
  • the dialog processing device is the dialog processing device according to the fifth aspect
  • the user equipment is the user device according to the sixth aspect.
  • the present application provides a computer-readable storage medium having instructions stored therein, which when run on a computer, causes the computer to perform any possible method described above.
  • the dialogue processing device feeds back the service identifier maintained by the server to The user, so that the corresponding slot information input by the user must correspond to a certain service item in the server.
  • FIG. 1 is a user interface of an embodiment of a multi-round dialogue provided by this application
  • 2A is a schematic diagram of the hardware structure of the dialog processing device 10 provided by the present application.
  • 2B is a schematic diagram of the software architecture of the dialog processing device 10 provided by the present application.
  • 3A is a schematic diagram of a hardware structure of user equipment 20 provided by the present application.
  • 3B is a schematic diagram of the software architecture of the user equipment 20 provided by this application.
  • 5A is a schematic diagram of a first implementation manner of a user interface provided by this application.
  • 5B is a schematic diagram of a second implementation manner of the user interface provided by this application.
  • 5C is a schematic diagram of a third implementation manner of the user interface provided by this application.
  • FIG. 6 is a schematic diagram of a user interface of a first implementation scenario provided by this application.
  • FIG. 7 is a schematic diagram of a user interface of a second implementation scenario provided by this application.
  • FIG. 8 is a schematic diagram of a user interface of a third implementation scenario provided by this application.
  • FIG. 9 is a schematic diagram of a user interface of a fourth implementation scenario provided by this application.
  • FIG. 10 is a schematic diagram of a user interface of a fifth implementation scenario provided by this application.
  • FIG. 11 is a schematic structural diagram of a first implementation manner of a human-computer interaction system provided by this application.
  • FIG. 14 is a schematic structural diagram of a second implementation manner of a human-computer interaction system provided by this application.
  • 15 is a method flowchart of a fourth implementation manner of a dialog processing method provided by this application.
  • 16 is a method flowchart of a fifth implementation manner of a dialog processing method provided by the present application.
  • This application relates to information interaction between a dialog processing device and a user equipment (UE), and between the dialog processing device and a server.
  • the user equipment is a medium for the user to have a dialogue with the dialogue processing device.
  • the user equipment transmits the information input by the user to the dialog processing device, and prompts the information fed back by the dialog processing device.
  • the dialog processing device may request the server to perform the service indicated by the corresponding information.
  • the dialog processing device After receiving the execution result of the service from the server, the dialog processing device sends the execution result to the user device.
  • the dialog management device can maintain intent, slot type and slot information.
  • Service information is maintained in the server.
  • intent is maintained in the server.
  • slot type is maintained in the server.
  • service information is maintained in the server.
  • the dialogue processing device can recognize the user's intention from the information input by the user device.
  • the input information of the user device is "help me book a flight ticket from city A to city B”
  • the dialog processing device can recognize from the input information that the user's intention is "book a ticket”.
  • the recognition model of intent can be obtained by training a large amount of corpus, which is a corpus expressing the intent in different expressions.
  • the dialog processing device can maintain one or more intents.
  • Slot information refers to specific information of the intended content, and is also key information for triggering a specific service.
  • the slot information is, for example, a keyword in input information of the user equipment.
  • the "A city” and “B city” in the input information "help me book a flight from city A to city B” are the slot information of the input information.
  • the slot information may include required slot information and optional slot information.
  • the required slot information refers to the slot information necessary to trigger the service.
  • Optional slot information refers to slot information that is not necessary to trigger the service.
  • the required slot information that triggers the service "book a ticket” includes the starting place "A city", the departure time "XX year XX month XX day XX hours XX minutes" and the destination "B city”.
  • the optional slot information is, for example, "Country to which City B belongs”.
  • the slot type (slot_type) is used to identify the category to which the slot information belongs.
  • the slot information "city A” and the slot information "city B” belong to the slot type "address”
  • the slot information "XX year XX month XX day XX hour XX minutes” belong to the slot type "time”.
  • a slot type corresponds to a specific knowledge base.
  • all slot information belonging to a slot type can be stored in the dialog processing device in different structural forms.
  • the slot information corresponding to the slot type "address” may be stored in the dialog processing device in the form of a list of city names.
  • the slot information corresponding to the slot type "time” may be all expressions of time.
  • an intent can correspond to one or more slot types.
  • the intention of "booking a ticket” corresponds to the slot type "address” and the slot type "time”.
  • Service refers to the purposeful execution of the server to complete a specific service item.
  • the service can be marked by the service identifier.
  • one or more required slot information can determine a service.
  • the service identifier of the service corresponds to one of the one or more required slot information, or the service identifier of the service is one of the one or more required slot information information.
  • the slot information "A city”, the slot information "B city” and the slot information "XX year XX month XX day XX hour XX minutes” determine the service "book XX year Air ticket to fly from city A to city B at XX, XX, XX and XX.”
  • the ticket booking server establishes a connection between the ticket of "Flying from City A to City B at XX XX, XX, XX, XX, XX, XX, XX, XX, XX, and marking the ticket as being booked, that is, performing the booking "XX "A flight from city A to city B" on XX month XX day XX XX.
  • the form of service can be instructions, procedures or processes.
  • the dialogue processing device is a device that can process the voice or text sent by the user and feed back the voice or text to the user.
  • the dialog processing device is carried in a robot (Bot), where the Bot exists independently of the user equipment used by the user, and performs information interaction with the user equipment through wireless communication.
  • the skills or programs that support the dialog processing are carried on the user equipment.
  • the user equipment and dialog processing are an electronic device entity.
  • the server may be a computer that hosts the program corresponding to the service item, an operating system, or a cloud server (elastic compute (ECS).
  • ECS elastic compute
  • the program carried by the server can provide the above service items when running, and each service item corresponds to a corresponding service identifier.
  • the program carried by the server may be source code written in a specific computer programming language, such as hypertext markup language (hypertextmarkuplanguage, HTML), hypertext preprocessor (PHP), java script (JavaScript , JS) etc.
  • the source code can be loaded and displayed as user-recognizable content by a browser or a webpage display component similar to the browser function. Based on this, related technical personnel can add new service items to the server or delete existing service items by modifying the source code.
  • the user equipment may be a portable electronic device that also contains other functions such as a personal digital assistant and/or music player functions, such as a mobile phone, a tablet computer, a wearable electronic device (such as a smart watch) with wireless communication function, etc. .
  • portable electronic devices include, but are not limited to Or portable electronic devices of other operating systems.
  • the above portable electronic device may also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface or a touch panel. It should also be understood that, in some other embodiments, the above electronic device may not be a portable electronic device, but a desktop computer with a touch-sensitive surface or a touch panel.
  • the user equipment can interact with the dialogue processing equipment through a user interface (UI).
  • UI user interface
  • the user interface is a media interface for interaction and information exchange between the application program or the operating system and the user. It implements the conversion between the internal form of information and the form that the user can accept.
  • the user interface of the application is source code written in a specific computer language such as java, extensible markup language (XML), etc.
  • the interface source code is parsed and rendered on the terminal device, and finally presented as user-recognizable content, Such as pictures, text, buttons and other controls.
  • Controls also called widgets
  • Typical controls include toolbars, menu bars, text boxes, buttons, and scroll bars. (scrollbar), pictures and text.
  • the attributes and contents of the controls in the interface are defined by tags or nodes.
  • tags or nodes For example, XML uses ⁇ Textview>, ⁇ ImgView>, ⁇ VideoView> and other nodes to specify the controls contained in the interface.
  • a node corresponds to a control or attribute in the interface, and the node is rendered as user-visible content after being parsed and rendered.
  • the dialogue mentioned in this application refers to "human-machine dialogue", that is, the information interaction between the user and the dialogue processing device.
  • the user may send audio to the conversation processing device through the user device.
  • the user may send text to the conversation processing device through the user device.
  • the dialog processing device may convert the execution result into audio, and then send the audio to the user device.
  • the dialog processing device may convert the execution result into text, and then send the text to the user device.
  • the user's audio interaction with the dialog processing device, the user's text interaction with the dialog processing device, and the user's interaction with one of the dialog processing device to send audio and the other to send text belong to the category of "human-machine dialogue" in this application.
  • the dialog processing device In conjunction with the existing dialog processing method, if the slot information preset in the dialog processing device does not contain the required slot information input by the user device, the dialog processing device will not be able to determine the first service to be requested, resulting in the first service Execution failed. If the service included in the server does not include the first service requested by the dialog processing device, the server cannot execute the first service requested by the dialog processing device, which may also cause the execution of the first service to fail. The service requested by the user can be provided only when the first service is successfully executed. Therefore, after the dialog processing device sends the user device the information that the first service fails, the dialog processing device may receive the dialog sent by the user device that triggers the second service. The attributes of the second service and the first service are the same.
  • Figure 1 illustrates an exemplary user interface for multiple rounds of conversation.
  • the user requests vehicle service through a human-machine dialogue.
  • the user inputs "help me call a ride to Tsinghua West Gate” to the user device.
  • the dialogue processing device After receiving the message "Help me call a ride to Tsinghua Ximen", the dialogue processing device generates a request message and sends the request message to the target call server.
  • the request information includes the current address of the user, the destination "Tsinghua West Gate” and the required slot information "Shunfeng".
  • the target ride-hailing server does not provide a "ride” service, so the user-triggered service cannot be executed.
  • the conversation processing device sends a message "Sorry, the service you requested cannot be performed" to the user device. Furthermore, the user can continue to input “Now help me get a fast train” to the user device, thereby initiating a new round of dialogue with the dialogue processing device.
  • the dialogue processing device requests the "fast train” service from the target call server in accordance with the above-mentioned process of requesting the "ride” service. If the target taxi server also does not provide the "Express" service, the user and the dialogue processing device may need to conduct more rounds of dialogue. This not only causes the dialog processing device to continuously execute the dialog processing operation, occupying resources, but also makes the user's use experience poor.
  • the present application provides a dialog processing method and device, the slot information preset in the dialog processing device does not correspond to the required slot information input by the user, and/or, the dialog processing device
  • the dialogue processing device can provide effective information and/or processing strategies in time, which not only reduces unnecessary dialogue rounds, but also provides flexible processing methods and can improve User experience.
  • FIG. 2A shows a schematic diagram of the structure of the dialog processing device 10.
  • the conversation processing device 10 includes a processor 11, a memory 12, and a transceiver 13. Among them, the processor 11, the memory 12 and the transceiver 13 are connected through a bus interface.
  • the structure illustrated in FIG. 2A does not constitute a specific limitation on the dialog processing device 10.
  • the dialog processing device 10 may include more or fewer components than shown, or combine certain components, or split certain components, or arrange different components.
  • the illustrated components can be implemented in hardware, software, or a combination of software and hardware.
  • the processor 11 may be a central processing unit (CPU), a network processor (NP), or a combination of CPU and NP.
  • the processor may further include a hardware chip.
  • the hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD) or a combination thereof.
  • the PLD may be a complex programmable logic device (complex programmable logic device, CPLD), a field programmable logic gate array (field-programmable gate array, FPGA), a general array logic (generic array logic, GAL), or any combination thereof.
  • the memory 12 may include volatile memory (volatile memory), such as random-access memory (RAM); the memory may also include non-volatile memory (non-volatile memory), such as read-only memory (read-memory) only memory (ROM), flash memory (flash memory), hard disk (hard disk drive) or solid-state drive (SSD); the memory may also include a combination of the above types of memory.
  • volatile memory volatile memory
  • non-volatile memory such as read-only memory (read-memory) only memory (ROM), flash memory (flash memory), hard disk (hard disk drive) or solid-state drive (SSD)
  • ROM read-only memory
  • flash memory flash memory
  • hard disk drive hard disk drive
  • SSD solid-state drive
  • the transceiver 13 may be a wired transceiver, a wireless transceiver, or a combination thereof.
  • the wired transceiver may be an Ethernet interface, for example.
  • the Ethernet interface may be an optical interface, an electrical interface, or a combination thereof.
  • the wireless transceiver may be, for example, a wireless local area network transceiver, a cellular network transceiver, or a combination thereof.
  • the bus interface may include any number of interconnected buses and bridges. Specifically, one or more processors represented by the processor 11 and various circuits of the memory represented by the memory 12 are linked together. The bus interface can also link various other circuits such as peripheral devices, voltage regulators, and power management circuits.
  • the transceiver 13 provides a unit for communicating with various other devices (such as user equipment, servers) on the transmission medium.
  • the processor 11 is responsible for managing the bus architecture and general processing, and the memory 12 can store data used by the processor in performing operations.
  • the hardware architecture shown in FIG. 2A involves hardware components that perform dialog processing.
  • the transceiver 13 may be used to perform transmission and reception of information between the conversation processing device and the user equipment, for example, receiving voice information sent by the user equipment, sending a service identifier and an execution result to the user equipment, and so on.
  • the transceiver 13 can also be used for sending and receiving information between the conversation processing device and the server, for example, sending information requesting service to the server, and receiving service identification and execution results from the server.
  • the memory 12 is used to store instructions and data. In some embodiments, the memory 12 may store instructions or data that the processor 11 has just used or recycled. If the processor 11 needs to use the instruction or data again, it can be directly called from the memory 12 to avoid repeated access.
  • the memory 12 is further used to store one or more intent information, one or more slot types corresponding to each intent information in the one or more intent information, and each slot type in the one or more slot types Contains slot information.
  • the processor 11 invokes the instructions and data in the memory 12 to perform other operations besides sending and receiving information, for example, recognizing speech, parsing semantics, and determining user intention and slot information according to the parsed semantics.
  • 2B is a block diagram of the software structure of the dialog processing device 10 of the present application.
  • the software system of the dialog processing device 10 may include functional modules for performing various functions of the dialog processing. For example, operating system, transceiver module, automatic speech recognition (ASR) module, semantic understanding (natural language understanding, NLU) module, dialogue generation (natural language generation, NLG) module, slot information recognition module, request information generation Modules etc.
  • ASR automatic speech recognition
  • NLU natural language understanding
  • NLG natural language generation
  • slot information recognition module request information generation Modules etc.
  • the dialog processing device 10 further includes an editing interface.
  • an editing operation interface can be presented.
  • the software system shown in FIG. 2B involves functional modules that perform dialog processing, where the operating system drives each module to perform operations.
  • the ASR module recognizes the voice
  • the NLU module analyzes the semantics of the recognized voice.
  • the slot information identification module can identify the slot information from the analyzed semantics, and the request information generation module generates the information requesting the service.
  • the transceiving module receives the execution result of the service from the server
  • the NLG module generates a voice to generate the execution result, and then the transceiving module sends the generated voice to the user equipment.
  • FIG. 3A shows a schematic structural diagram of user equipment 20.
  • the user equipment 20 may include a processor 110, an external memory interface 120, an internal memory 121, a universal serial bus (USB) interface 130, a charging management module 140, a power management module 141, a battery 142, an antenna 1, an antenna 2 , Mobile communication module 150, wireless communication module 160, audio module 170, speaker 170A, receiver 170B, microphone 170C, headphone jack 170D, sensor module 180, key 190, motor 191, indicator 192, camera 193, display 194, and Subscriber identification module (SIM) card interface 195, etc.
  • SIM Subscriber identification module
  • the sensor module 180 may include a pressure sensor 180A, a gyro sensor 180B, an air pressure sensor 180C, a magnetic sensor 180D, an acceleration sensor 180E, a distance sensor 180F, a proximity light sensor 180G, a fingerprint sensor 180H, a temperature sensor 180J, a touch sensor 180K, and ambient light Sensor 180L, bone conduction sensor 180M, etc.
  • the structure illustrated in this application does not constitute a specific limitation on the user equipment 20.
  • the user equipment 20 may include more or fewer components than shown, or combine some components, or split some components, or arrange different components.
  • the illustrated components can be implemented in hardware, software, or a combination of software and hardware.
  • the processor 110 may include one or more processing units, for example, the processor 110 may include an application processor (application processor, AP), a controller, a memory, and the like. Among them, different processing units may be independent devices, or may be integrated in one or more processors. In some embodiments, the user equipment 20 may also include one or more processors 110.
  • the processor 110 may include an application processor (application processor, AP), a controller, a memory, and the like. Among them, different processing units may be independent devices, or may be integrated in one or more processors.
  • the user equipment 20 may also include one or more processors 110.
  • the controller may be the nerve center and command center of the user equipment 20.
  • the controller can generate the operation control signal according to the instruction operation code and the timing signal to complete the control of fetching instructions and executing instructions.
  • the processor 110 may also be provided with a memory for storing instructions and data.
  • the memory in the processor 110 is a cache memory.
  • the memory may store instructions or data that the processor 110 has just used or recycled. If the processor 110 needs to use the instruction or data again, it can be directly called from the memory. The repeated access is avoided, the waiting time of the processor 110 is reduced, and thus the efficiency of the user equipment 20 is improved.
  • the processor 110 may include one or more interfaces.
  • the one or more interfaces include an interface that implements communication between the processor 110 and the audio module 170, and an interface that connects the processor 110 and the display screen 194. No more details here.
  • the wireless communication function of the user equipment 20 can be realized by the antenna 1, the antenna 2, the mobile communication module 150, the wireless communication module 160, the modem processor, and the baseband processor.
  • the wireless communication module 160 can provide wireless communication solutions including wireless local area networks (wireless local area networks, WLAN) (such as wireless fidelity (Wi-Fi) networks) applied to the user equipment 20, and the like.
  • WLAN wireless local area networks
  • Wi-Fi wireless fidelity
  • the antenna 2 of the user equipment 20 and the wireless communication module 160 are coupled so that the user equipment 20 can communicate with the network and other devices through wireless communication technology.
  • the display screen 194 of the user equipment 20 and the application processor can implement the prompt function.
  • the display screen 194 is used for prompting text, menus, audio controls, etc.
  • the user equipment 20 may implement audio functions through an audio module 170, a speaker 170A, a receiver 170B, a microphone 170C, a headphone interface 170D, and an application processor. For example, audio playback, recording, etc.
  • the audio module 170 is used to convert digital audio information into analog audio signal output, and also used to convert analog audio input into digital audio signal.
  • the audio module 170 can also be used to encode and decode audio signals.
  • the speaker 170A also known as a “speaker”, is used to convert audio electrical signals into sound signals.
  • the user device 20 can listen to the audio conversation sent by the conversation processing device through the speaker 170A.
  • the receiver 170B also known as "handset" is used to convert audio electrical signals into sound signals.
  • the user device 20 receives the audio information, the user can listen to the sound of the audio information by holding the receiver 170B to the ear.
  • the microphone 170C also known as “microphone”, “microphone”, is used to convert sound signals into electrical signals.
  • the user can make a sound by approaching the microphone 170C through the human mouth, and input the sound signal to the microphone 170C.
  • the user equipment 20 may set at least one microphone 170C.
  • the user equipment 20 exemplarily shown in FIG. 3A may display and prompt various user interfaces described in the following embodiments through the display screen 194.
  • the user device 20 can input voice information to each user interface through the microphone 170C, and can also play audio prompted by the user interface through the speaker 170A or the receiver 170B.
  • the software system of the user equipment 20 may adopt a layered architecture.
  • This application takes an Android system with a layered architecture as an example to illustrate the software structure of the user equipment 20 by way of example.
  • 3B is a block diagram of the software structure of the user equipment 20 of the present application.
  • the layered architecture divides the software into several layers, and each layer has a clear role and division of labor.
  • the layers communicate with each other through a software interface.
  • the Android system is divided into four layers, from top to bottom are the application layer, the application framework layer, the Android runtime and the system library, and the kernel layer.
  • the application layer may include a series of application packages.
  • the application package may include applications such as camera, gallery, WLAN, Bluetooth, music, video, and short message.
  • the application framework layer provides application programming interfaces (application programming interfaces) and programming frameworks for applications at the application layer.
  • the application framework layer includes some predefined functions.
  • the application framework layer may include a window manager, a content provider, a view system, a phone manager, a resource manager, a notification manager, and so on.
  • the window manager is used to manage window programs.
  • the window manager can obtain the size of the display screen, determine whether there is a status bar, lock the screen, intercept the screen, etc.
  • Content providers are used to store and retrieve data and make it accessible to applications.
  • the data may include video, images, audio, text, etc.
  • the view system includes visual controls, such as controls that display text.
  • the view system can be used to build applications.
  • the display interface can be composed of one or more views.
  • the media library supports a variety of commonly used audio playback and recording.
  • the media library can support multiple audio and video encoding formats, such as: MPEG4, H.264, MP3, AAC, AMR, JPG, PNG, etc.
  • the kernel layer is the layer between hardware and software.
  • the kernel layer contains at least the display driver, audio driver, and sensor driver.
  • the software system shown in FIG. 3B is related to the presentation of applications that perform human-machine dialogue (eg window manager, view system), the provision of audio codec services, and the kernel and basic communication protocols.
  • human-machine dialogue eg window manager, view system
  • audio codec services e.g., voice codec services
  • kernel and basic communication protocols e.g., voice codec services
  • the following describes the workflow of the software and hardware of the user device 20 in combination with the scene of human-machine dialogue.
  • the microphone 170C receives the user's voice audio.
  • the audio driver of the kernel layer is triggered to convert the user's voice and audio into electrical signals.
  • the wireless communication module 160 sends the electric signal to the conversation processing device.
  • the display driver of the kernel layer calls the window manager to display the audio.
  • the pressure sensor 180A receives the instruction to play audio, the corresponding hardware interrupt is sent to the kernel layer.
  • the audio driver of the kernel layer triggers the audio module 170 to decode the audio, and the decoded audio is played through the speaker 170A or the receiver 170B.
  • the required slot information includes slot information indicating the service identifier, for example, the "ride" in the call service, and the conventional slot information used to determine the service, For example, the destination is "Tsinghua West Gate".
  • the service provided by the server to the user is usually customized by the relevant service personnel, so the service provided by the server may be changed, for example, the "Shunfeng" service item is off the shelf, and the "carpooling" service item is on the shelf, resulting in no corresponding response.
  • the service corresponding to the service ID is usually customized by the relevant service personnel, so the service provided by the server may be changed, for example, the "Shunfeng" service item is off the shelf, and the "carpooling" service item is on the shelf, resulting in no corresponding response.
  • the service corresponding to the service ID is usually customized by the relevant service personnel, so the service provided by the server may be changed, for example, the "Shunfeng" service item is off the shelf, and the "carpooling" service item is on the shelf
  • the user may not be able to know in time what services are provided by the server, on the other hand, the relevant slot information maintained in the dialog processing device may also not be updated in time, thereby causing some of the problems mentioned above in this application.
  • the slot type to which the slot information used to indicate the service identifier belongs in the required slot information is referred to as the "first slot type", and the slot of the conventional slot information in the required slot information The type is called "second slot type".
  • first slot type refers to the slot type to which the slot information used to indicate the service identifier belongs in the required slot information
  • second slot type refers to The slot type of the conventional slot information in the required slot information will not be repeated below.
  • the developer can design the dialog processing device according to the attributes of the above-mentioned mandatory slot type, by labeling the first Slot type. For example, the developer can add the label "M" to the first slot type in the required slot type, and not add any label to the first slot type in the required slot type.
  • the conversation processing device reads the slot type carrying the label “M”, it determines the slot type as the first slot type.
  • the conversation processing device reads the required slot type that does not carry the tag "M”, it determines the required slot type as the second slot type.
  • the developer may directly add a label to the first slot type during the design of the program related to the slot type.
  • the application may provide an editing interface on which developers can classify slot types by checking.
  • FIG. 4 illustrates an exemplary editing interface of a slot type.
  • the list of slots indicated on the editing interface may include the slot type name, the slot information database corresponding to the slot type, the required slot options, prompt options, and five questions.
  • the slot type name is used to name the slot type to indicate the corresponding slot type.
  • the slot information database corresponding to the slot type may be a link for linking to the slot information database corresponding to the slot type of the same row.
  • the question indicates various expressions corresponding to the slot information of the corresponding row slot type. If the required slot option is selected, the slot type of the corresponding row is considered to be marked as the required slot type. Correspondingly, the slot type is added with a "required" label.
  • the prompt option is an option for marking whether it is the first slot type in this implementation. Based on this, in the same way, if the prompt option is selected, it is considered that the slot type of the corresponding row is marked as the first slot type. Correspondingly, the slot type is added with a "prompt" label. It should be noted that in this embodiment, only the required slot option of the corresponding row is selected, the prompt option of the row can be operated. In addition, the developer can trigger the "+Add Slot Type" to add the slot type according to the requirements, and can click the delete symbol to delete the slot type of the row corresponding to the deleted symbol.
  • the slot type “vehicle type” and the slot type “destination” are required slot types, and the slot type “starting point” is an optional slot type.
  • the required slot types “vehicle type” is the first slot type, and "destination” is the second slot type.
  • the dialog processing device when the user does not input the slot information corresponding to the first slot type, or the slot information corresponding to the first slot type entered by the user does not have a matching service, the dialog processing device Multiple service identifiers corresponding to the first slot type are fed back to the user, so that the user selects one service identifier from the multiple service identifiers as corresponding slot information, so that a service can be triggered quickly and accurately.
  • the multiple service identifiers may be all service identifiers corresponding to the first slot type in the server. In other embodiments, the multiple service identifiers may be part of the service identifiers corresponding to the first slot type in the server.
  • the plurality of service identifiers may be service identifiers in which the user praise is higher than a certain value, for example, a service identifier corresponding to a service item that the user praises is higher than 95%.
  • the multiple service identifiers may be service identifiers of service items updated within a certain time from the current time, for example, the multiple service identifiers are service identifiers of service items updated within the last 3 months.
  • the selection conditions of the above “multiple service identifiers” are only selection conditions corresponding to some exemplary implementation scenarios of this application, and do not constitute a specific limitation on the selection of “multiple service identifiers”. In other embodiments, the selection of "multiple service identifiers" can be flexibly set according to other implementation scenarios. I won't repeat them here.
  • First implementation scenario The user inputs dialogue information to the user equipment, and the dialogue information does not include the slot information corresponding to the first slot type. Then, the user equipment may receive all service identifiers corresponding to the first slot type from the conversation processing device. The user may select a service identifier from all the service identifiers as the first slot type slot information and input it to the user equipment.
  • all the service identifiers corresponding to the first slot type may be two or three.
  • the first slot type is the “vehicle type” in the call-hailing service, and the “vehicle type” corresponds to the “downwind car” ", "special car” and "express” service marks.
  • the first slot type is "drink" in the meal order service, and the "drink” corresponds to "Coke” and “Sprite”. , "Soybean Milk”, “Coffee”, “Juice”, “Milk”, “Black Tea”, “Green Tea”, “Milk Tea” and “Beer” and so on.
  • the size of the user interface is limited. Therefore, in some embodiments, when the number of all service identifiers is less than or equal to a preset threshold, the user equipment may prompt the all service identifiers on the user interface, as shown in the user interface shown in FIG. 5A .
  • the preset threshold is 5, for example.
  • the user equipment may prompt the all service identifiers on the user interface in a link manner, as shown in the user interface shown in FIG. 5B.
  • the user equipment may set a menu bar (menu bar) for the scroll bar, as shown in the user interface shown in FIG. 5C.
  • the user interface illustrated in FIGS. 5A to 5C does not constitute a specific limitation on the present application.
  • the user interface may prompt the service identifier in a manner different from that described above. No more details here.
  • the user equipment may receive information that the service execution is successful from the dialog processing device. If the dialogue information does not include the slot information corresponding to the second slot type, the user equipment may receive a dialogue requesting the user to input the corresponding slot information from the dialogue processing device, and the user may input the corresponding slot information in response to the corresponding dialogue. Finally, the user The device may receive the information that the service execution is successful from the conversation processing device.
  • the user inputs "help me get a car” to the user interface, and then the user equipment prompts "the currently available vehicle types include: special cars, express cars, and carpools. Which one do you want?" . Furthermore, the user selects "Express”. Then, the user interface may prompt "Where are you going?". After the user enters "Tsinghua West Gate", the user interface prompts "The train has been successfully reserved for you from your location to Tsinghua West Gate".
  • Second implementation scenario The user inputs dialogue information to the user equipment, the dialogue information includes the slot information corresponding to the first slot type, but the server does not include the service corresponding to the slot information. Furthermore, the user equipment may receive multiple service identifiers corresponding to the first slot type from the conversation processing device. The user may select one service identifier from the plurality of service identifiers as the first slot type slot information and input it to the user equipment. The user interface corresponding to the service identifier is shown in FIGS. 5A to 5C. In addition, the subsequent operations of the user and the dialog content subsequently prompted by the user equipment are similar to the first implementation scenario, and will not be repeated here.
  • the user inputs "help me order a coffee” to the user interface.
  • coffee is sold out.
  • the user device prompts "The drinks currently available include: milk, cola, sprite, fruit juice, milk tea, soy milk, black tea, which one do you want?".
  • the user equipment prompts the above marks through the menu bar with a scroll bar.
  • the user can view the logo corresponding to "drink” by sliding the scroll bar on the user interface of FIG. 6 up and down.
  • the user selects "black tea", for example.
  • the user interface prompts the message "The black tea you ordered will be delivered in 15 minutes".
  • the types of drinks listed in this embodiment may be some of the drinks in "Currently Available Beverages”.
  • the user inputs dialogue information to the user equipment, and the dialogue information includes slot information corresponding to the first slot type.
  • the dialog processing device does not contain the slot information, but the server contains the service corresponding to the slot information.
  • the user equipment prompts the service execution success message.
  • the user inputs “help me book a double-decker Burger A package” to the user interface.
  • the dialogue processing device does not include the slot information corresponding to the "double-layer burger A package", but the user device may receive the message "the double-layer burger A package you booked will be delivered within half an hour”.
  • the user inputs dialogue information to the user equipment, and the dialogue information includes slot information corresponding to the first slot type.
  • the first server does not include the service corresponding to the slot information.
  • the user equipment prompts that the service is successfully performed by the second server.
  • the user inputs “check the flight from Hong Kong to New York tomorrow morning from 10 to 11 am” to the user interface.
  • the first server is a server that maintains domestic flight data, for example, a server of a domestic travel platform.
  • the user device can receive the message "The official data of international flights prompts that there will be AA8925 flights from Hong Kong to New York tomorrow morning from 10 to 11 am, and will depart from Hong Kong at 10:35 am.”
  • the user inputs dialogue information to the user equipment, the dialogue information includes slot information corresponding to the first slot type, and/or, slot information corresponding to the first slot type, but at least one of the above slots
  • the user equipment prompts the information that the corresponding service is executed according to the best solution.
  • the meaning of the slot information indication is unclear means that the slot information may indicate a plurality of different specific information, thereby causing the dialog processing device to fail to trigger the determined service according to the slot information.
  • the best solution refers to the most, fast, good, and/or provincial solution for the user, for example, the route with the shortest itinerary for the user, or the package that requires the user to pay the least cost.
  • the user inputs "help me book a hotel near Beijing Railway Station” to the user interface.
  • the user equipment receives and prompts "a standard room for a five-star hotel X hotel is 400 meters away from Beijing Railway Station".
  • the "X Hotel” is, for example, the five-star hotel closest to Beijing Railway Station.
  • FIGS. 6 to 10 are only a part of the exemplary implementation scenarios of the present application, and do not constitute a limitation on the applicable scenarios of the present application.
  • the present application can also be applied to any other scenario where any user triggers a service through a human-machine dialogue. I won't repeat them here.
  • the technical solution of the present application is adopted, regardless of whether the dialog information input by the user includes the slot information corresponding to the first slot type, regardless of the slot information input by the user and What is the correspondence between the slot information preset in the dialogue processing device, and regardless of the correspondence between the slot information input by the user and the service provided by the server, the user can obtain the desired service. This not only prevents users from initiating multiple rounds of conversations, but also enables users to be informed of the services provided by the server in time, thereby improving the user's experience.
  • FIG. 11 shows an architecture diagram of the first embodiment of the applicant’s machine dialogue system.
  • the intention, slot type, and slot information maintained in the dialog processing device correspond to the services supported by the server.
  • the user can talk to the dialogue device through the user device.
  • the dialog processing device may be as described in FIGS. 2A and 2B, and the user equipment may be as described in FIGS. 3A and 3B.
  • FIG. 12 shows a method flowchart of an implementation manner of the dialogue processing method of the present application.
  • the dialog processing method 100 shown in FIG. 12 includes the following steps:
  • step S101 the dialog processing device receives dialog information from the user equipment.
  • the dialogue information may be audio or text. If the dialogue information is audio, the dialogue processing device may perform voice recognition on the audio, and then, semantically understand the content after the voice recognition to determine the intention corresponding to the dialogue information. If the dialogue information is text, the dialogue processing device can semantically understand the text to determine the intent corresponding to the dialogue information. Further, the dialog processing device determines one or more mandatory slot types corresponding to the intention, and determines whether each mandatory slot type in the one or more mandatory slot types is the first slot type or the second Slot type. Then, the dialog processing device may recognize whether the dialog information includes the slot information corresponding to the first slot type.
  • the first slot type described in this embodiment refers to the slot type to which the slot information used to indicate the service identifier belongs in the required slot information.
  • the second slot type described in this embodiment refers to the slot type of the conventional slot information in the mandatory slot information.
  • the dialog processing device may determine whether the required slot type is by identifying whether each required slot type includes a preset label Type of the first slot. If the required slot type includes a preset label, the dialog processing device considers the required slot type to be the first slot type. If the required slot type does not include the preset label, the dialog processing device considers the required slot type to be the second slot type.
  • the setting process of the preset label and the content of the preset label are described in the embodiment corresponding to FIG. 4 and will not be described in detail here.
  • the dialogue processing device may recognize one or more keywords in the dialogue information. Furthermore, for each keyword in one or more keywords, in some embodiments, the dialog processing device determines the slot type to which the keyword belongs by using a word vector model. In other embodiments, the dialog processing device determines the slot type to which the keyword belongs according to the expression format of the keyword, for example: according to the format of the keyword "12:30", it can be determined that the keyword "12:30" belongs to the slot Type "Time”.
  • Step S102 if the dialog information does not contain slot information corresponding to the first slot type that can determine the service, the dialog processing device obtains the service identifier set corresponding to the first slot type from the server.
  • the service identifier set includes all service identifiers corresponding to the first slot type.
  • step S103 the conversation processing device sends the service identification set to the user device.
  • the dialog information that does not include the slot information corresponding to the first slot type that can determine the service may include two scenarios: Scenario 1: The dialog information does not include the slot information corresponding to the first slot type.
  • the dialog processing device may send an acquisition request to the server, and the acquisition request may include the first slot type. logo.
  • the server sends the service identifier set formed by all the service identifiers corresponding to the first slot type to the dialog processing device.
  • the dialog processing device may send the service identification set to the user equipment, so that the user can select any service identification as the slot information corresponding to the first slot type.
  • the dialog processing device determines that the dialog information does not include the slot information corresponding to the first slot type.
  • the dialog processing device obtains the slot information corresponding to the first slot type from the dialog information, and further, the dialog processing device requests the server for all required options based on all required slot information contained in the dialog information The service corresponding to the slot information. If the service execution failure information is received from the server, the dialog processing device determines that the dialog information does not include slot information that can determine the corresponding first slot type.
  • the dialogue processing device can determine the number from the dialogue information “Help me order a coffee”.
  • the slot information "coffee” corresponding to a slot type.
  • the dialogue processing device requests the service of "order coffee” from the server. Since the coffee has been sold out, the server fails to execute the service "order coffee” and can send a message to the dialogue processing device that the execution failed.
  • the server after the server fails to perform the service "order coffee", the server can also directly send the in-stock beverage illustrated in FIG. 7 to the dialogue processing device.
  • step S104 the dialog processing device receives the target service identifier sent by the user equipment, and requests the server for the service corresponding to the target service identifier.
  • the target service identifier belongs to the above service identifier set.
  • the target service identifier is used as slot information corresponding to the first slot type.
  • the dialog processing device has learned the slot information corresponding to the second slot type.
  • the dialog processing device learns the slot information corresponding to the second slot type.
  • step S105 the dialog processing device receives the service execution success information from the server, and sends the service execution success information to the user equipment.
  • the dialog processing device feeds back all the service identifiers maintained by the server to the user, so that the user enters the corresponding slot Bit information must correspond to a service item in the server.
  • FIG. 13 shows a method flowchart of another implementation manner of the dialogue processing method of the present application.
  • the dialog processing method 200 shown in FIG. 13 includes the following steps:
  • step S201 the dialog processing device receives dialog information from the user equipment.
  • the dialog information includes slot information corresponding to the first slot type.
  • the operation after the dialog processing device receives the dialog information is similar to the description of step S101 in the method 100, and is not repeated here.
  • step S202 the dialog processing device requests the server for the service corresponding to the slot information.
  • the dialog processing device even if the slot information is not included in the slot information preset in the dialog processing device, the dialog processing device still requests the server to provide corresponding services according to the slot information.
  • Step S203 if the slot processing device does not include the slot information, and the service execution success message is received from the server, the dialogue processing device adds the slot information to the information library corresponding to the first slot type.
  • the slot information is not included in the slot information preset in the dialog processing device, and the service is successfully executed by the server, it means that the service item maintained in the server is updated, but the first slot type in the dialog processing device The corresponding slot information is not updated in time.
  • a new product "dual hamburger A package” is added on the server side, but the dialog processing device does not update the information library corresponding to the slot type "package", then the dialog processing device may change " The slot information corresponding to the double-decker Burger A package is added to the information library corresponding to the slot type "package".
  • the slot information corresponding to the "double-layer burger A course” may be "double-layer burger A course".
  • step S204 the dialog processing device sends to the user equipment the information that the service execution is successful.
  • the conversation processing device not only can the conversation processing device and the user be reduced in the number of conversation rounds, and to provide services to the user in time, but also, the conversation processing device can timely update the slot information maintained by the conversation processing device.
  • the user interface corresponding to the method 200 is as shown in the above third implementation scenario and shown in FIG. 8 and will not be described in detail here.
  • the dialog processing method 300 includes:
  • the dialogue processing device receives the dialogue information from the user equipment, and then determines the keywords corresponding to all the required slot types in the dialogue information. If any keyword corresponding to all the required slot types corresponds to at least two slot information in the corresponding slot type, the dialogue processing device determines that the at least two slot information corresponds to the optimal service plan Specific slot information. Furthermore, the dialog processing device may request the server for the service corresponding to the specific slot information, and after receiving the information that the corresponding service is successfully executed from the server, send the information that the corresponding service is successfully executed to the user device.
  • the keyword corresponding to at least two slot information in a slot type may be a general vocabulary containing multiple specific slot information.
  • the slot information that may be included in the "hotel near Beijing Station” includes, for example, the names of all hotels within 1000 meters from Beijing Station.
  • the slot information that "Tsinghua" may include is, for example, addresses of all doors of Tsinghua University.
  • the "optimal” standard in the “optimal service plan” is different according to the service content and user needs.
  • the best service solution for "Book me a hotel near Beijing Railway Station” may be, for example, the nearest hotel to Beijing Railway Station, or the hotel with the highest star rating within 1,000 meters from Beijing Railway Station.
  • the dialogue message "Help me call a car to Tsinghua” the best service plan can be, for example, the plan with the shortest travel route, or the vehicle with the lowest cost.
  • the dialog processing device can be determined according to the calculation model of the corresponding service scenario.
  • the calculation model can be trained based on the big data of the corresponding service scenario and the user's historical order data. No more details here.
  • the dialog processing device can separately determine whether the service corresponding to each piece of slot information is the above-mentioned "optimal service plan", and if so, the The slot information is determined to be specific slot information.
  • the dialogue processing device can automatically provide users with relatively superior services when the meaning of the keywords entered by the user is ambiguous. Not only can they avoid multiple rounds of dialogue with the user, but also the operation is flexible, which can further improve User experience.
  • FIG. 14 shows an architectural diagram of a second embodiment of the applicant’s machine dialogue system.
  • the conversation processing device establishes a connection with the first server and the second server.
  • the user can talk to the dialogue device through the user device.
  • the dialog processing device may be as described in FIGS. 2A and 2B, and the user equipment may be as described in FIGS. 3A and 3B.
  • FIG. 15 shows a method flowchart of a fourth implementation manner of the dialogue processing method of the present application.
  • the dialog processing method 400 shown in FIG. 15 includes the following steps:
  • step S401 the dialogue processing device receives dialogue information from the user equipment.
  • the dialog information includes slot information corresponding to the first slot type.
  • the operation after the dialog processing device receives the dialog information is similar to the description of step S101 in the method 100, and is not repeated here.
  • step S402 the dialog processing device requests the first server for the service corresponding to the slot information.
  • Step S403 if the service execution failure information received from the first server, the dialog processing device requests the second server for the service corresponding to the slot information.
  • Step S404 if the service execution success information received from the second server, the dialog processing device sends the user equipment success execution information to the user equipment.
  • the user interface corresponding to the method 400 is as shown in the above fourth implementation scenario and shown in FIG. 9, and will not be described in detail here.
  • the first server may be a server that provides customized services
  • the second server may be an official server related to the services provided by the first server.
  • the first server is, for example, a server that provides domestic travel services
  • the second server is, for example, an official server that provides domestic and international flight data.
  • the dialog processing device can trigger the second server to provide the corresponding service. This can reduce the number of dialogue rounds between the dialogue processing device and the user, and provide users with services in a timely manner.
  • the dialog processing device may be a cloud solution, that is, the dialog processing device is a server, which is a relatively independent entity from the user device.
  • the dialog processing device may be a solution combining end clouds, that is, the dialog processing device includes a function module on the server and the user device.
  • the dialog processing device may include the server's Bot platform and applications on the user's device.
  • the dialog processing device may be an end-side solution, that is, the dialog processing device is a functional module integrated on the user device.
  • the conversation processing device may be an application on the user device.
  • the hardware of the dialog processing device is the hardware corresponding to the user equipment.
  • the following embodiments provide a user equipment method.
  • the user equipment may be as described in FIGS. 3A and 3B.
  • FIG. 16 shows a method flowchart of a fifth implementation manner of a dialog processing method for user equipment according to an embodiment of the present application.
  • the dialog processing method 500 shown in FIG. 16 includes the following steps:
  • Step S501 Receive dialogue information input by the user.
  • the user equipment may receive the user's dialogue information through the man-machine dialogue application, where the dialogue information indicates the service that the user wants to reserve.
  • the man-machine dialogue application can be a voice assistant or other installed applications that can implement dialogue functions.
  • the dialogue information may be voice information, and there may be text, gestures, and other information.
  • Step S502 If the dialog information does not include slot information that can determine the first slot type of the service, obtain a service identifier set corresponding to the first slot type from a local application or server.
  • the man-machine conversation application can query the service identifier set corresponding to the first slot type from the local application for the user to choose.
  • the local application is an application that can provide a service that the user wants to book.
  • the local application may be a third-party application, such as the Didi taxi application, which is different from the man-machine dialogue application.
  • the man-machine dialog application can also be accessed from the server Query the service identifier set corresponding to the first slot type for users to choose.
  • the server may be a server corresponding to the local application or a server corresponding to the man-machine conversation application. This application is not limited.
  • the man-machine dialogue application may query the service identification set corresponding to the first slot type from a local application or from a server for the user to select.
  • Step S503 prompting the service identifier in the service identifier set.
  • the user equipment may prompt the user by voice, for example, by voice announcement of a voice assistant; the user equipment may also prompt the user by text, or by a combination of methods, and so on.
  • voice for example, by voice announcement of a voice assistant
  • the user equipment may also prompt the user by text, or by a combination of methods, and so on.
  • Step S504 If the user determines the target service identifier from the service identifier set, the user requests a service corresponding to the target service identifier from the local application or the server.
  • the user equipment may request a service from a local application, such as the Didi taxi application, or may directly request a service from the server.
  • the user equipment requests the service from the server, which may be a request to the server corresponding to the man-machine dialogue application, or a request directly to the server corresponding to the local application. This application is not limited here.
  • Step S505 prompting the information that the service execution is successful.
  • the user may be prompted to make a successful vehicle reservation, etc.
  • the dialog processing device feeds back the service identifier maintained by the server to the user, so that the user inputs the corresponding The slot information must correspond to a service item in the server.
  • the solutions of the dialog processing method provided by the embodiments of the present application are introduced from the perspectives of the devices themselves and the interaction between the devices.
  • the above dialog processing device, user equipment, and server all include hardware structures and/or software modules corresponding to performing various functions.
  • the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is performed by hardware or computer software driven hardware depends on the specific application of the technical solution and design constraints. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
  • embodiments of the present application also provide a computer storage medium, wherein the computer storage medium provided in any device may store a program, and when the program is executed, it may be implemented as shown in FIGS. 4 to 15. Part or all of the steps in the various embodiments.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (read-only memory, ROM for short) or a random access memory (random access memory, RAM for short), etc.
  • the size of the sequence number of each process does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not be implemented in this application.
  • the implementation process of the examples constitutes no limitation.
  • the computer program product includes one or more computer instructions.
  • the computer may be a general-purpose computer, a dedicated computer, a computer network, or other programmable devices.
  • the computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions may be from a website site, computer, server or data center Transmission to another website, computer, server or data center via wired (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (such as infrared, wireless, microwave, etc.).
  • the computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device including one or more available medium integrated servers, data centers, and the like.
  • the usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, Solid State Disk (SSD)), or the like.
  • a magnetic medium for example, a floppy disk, a hard disk, a magnetic tape
  • an optical medium for example, a DVD
  • a semiconductor medium for example, Solid State Disk (SSD)

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Databases & Information Systems (AREA)
  • Human Computer Interaction (AREA)
  • Multimedia (AREA)
  • Data Mining & Analysis (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Telephonic Communication Services (AREA)

Abstract

一种对话处理方法及设备，该方法包括：对话处理设备从用户设备接收对话信息；若所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息，从服务器获取第一槽位类型对应的服务标识集，并将服务标识集发送到用户设备；从用户设备接收到目标服务标识后，向服务器请求目标服务标识对应的服务，并将执行成功的信息发送到用户设备。采用该方法，不但能够及时向用户展示服务项目，而且能够避免用户与对话处理设备发起多轮对话，提高执行服务的效率，进而提高用户的使用体验。

Description

对话处理方法及设备
本申请要求在2019年1月10日提交中国国家知识产权局、申请号为201910025807.3、发明名称为“对话处理方法及设备”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及计算机领域，尤其涉及一种对话处理方法及设备。
背景技术
人机对话系统(以下简称对话系统)可以响应用户输入的语音向用户反馈信息，该对话系统可以包括任务型对话系统(task-oriented systems)和非任务型对话系统(non-task-oriented systems)。非任务型对话系统可以是聊天式对话系统。任务型对话系统可以根据用户的语音触发确定的服务，并向该用户反馈相应服务被执行的结果。
其中，服务型对话系统可以对应目标服务器，该目标服务器用于提供和执行一项或多项服务。基于此，该对话系统接收用户输入的语音之后，可以从该语音中获取一个或者多个必选槽位信息。必选槽位信息是指示一项服务必不可少的信息，例如，预订机票这项服务的必选槽位信息包括起始地，起飞时间和目的地。进而，该对话系统可以检测该一个或者多个必选槽位信息是否均包含在该对话系统预置的槽位信息中，若该一个或者多个必选槽位信息均包含在该对话系统预置的槽位信息中，该对话系统根据该一个或者多个必选槽位信息请求该目标服务器执行目标服务。该目标服务器响应该请求检测该目标服务是否包含于该一项或多项服务中，若该目标服务包含于该一项或多项服务中，该目标服务器执行该目标服务。然后，该对话系统可以向用户反馈该目标服务执行成功的信息。此外，若该一个或者多个必选槽位信息中任一槽位信息不包含在该对话系统预置的槽位信息中，或者，若该目标服务不包含在该一项或多项服务中，该对话系统可能向用户反馈该目标服务执行失败的信息，那么，用户可能继续输入其他内容的语音，以使某项服务能够被执行成功。
可见，采用上述对话处理方法，用户与对话系统的对话轮次较多，使得用户的使用体验较差。
发明内容
本申请提供了一种对话处理方法及设备，解决了采用现有对话处理方法，用户与对话系统的对话轮次多的问题。
第一方面，本申请提供了一种对话处理方法，该方法包括：
对话处理设备从用户设备接收对话信息；
若所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息，所述对话处理设备从服务器获取所述第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述 第一槽位类型对应的多个服务标识；
所述对话处理设备将所述服务标识集发送到所述用户设备；
若从所述用户设备接收到目标服务标识，所述对话处理设备向所述服务器请求所述目标服务标识对应的服务，所述目标服务标识属于所述服务标识集；
若从所述服务器接收服务执行成功的信息，所述对话处理设备将所述执行成功的信息发送到所述用户设备。
采用本实现方式，在服务器自定义的服务项目与相应槽位类型的槽位信息不对应的情况下，对话处理设备将服务器维护的多个服务标识反馈给用户，从而使得用户输入的相应槽位信息一定对应服务器中某项服务项目。从而不但能够及时向用户展示服务项目，而且能够避免用户与对话处理设备发起多轮对话，提高执行服务的效率。
一种可能的实现方式中，对话处理设备从用户设备接收对话信息之后，从服务器获取所述第一槽位类型对应的服务标识集之前，包括：
所述对话处理设备解析所述对话信息；
若从所述对话信息中未解析出所述第一槽位类型对应的槽位信息，所述对话处理设备确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。一种可能的实现方式中，对话处理设备从用户设备接收对话信息之后，从服务器获取所述第一槽位类型对应的服务标识集之前，包括：
所述对话处理设备解析所述对话信息得到所述第一槽位类型对应的槽位信息；
所述对话处理设备向所述服务器请求所述槽位信息对应的服务；
若从所述服务器接收到服务执行失败的信息，所述对话处理设备确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
采用本实现方式，在该对话处理设备中预置的槽位信息与用户输入的必选槽位信息不对应，和/或，该对话处理设备中预置的槽位信息与服务器所包含的服务不对应的情况下，该对话处理设备能够及时提供有效的信息和/或处理策略，不仅能够减少不必要的对话轮次，而且处理方式灵活，能够提高用户的使用体验。
一种可能的实现方式中，对话处理设备从用户设备接收对话信息之后，将所述执行成功的信息发送到所述用户设备之前，还包括：
所述对话处理设备解析所述对话信息得到所述第一槽位类型对应的槽位信息；
所述对话处理设备向所述服务器请求所述槽位信息对应的服务；
若从所述服务器接收到所述服务执行成功的信息，且所述对话处理设备中不包含所述槽位信息，所述对话处理设备将所述槽位信息添加到所述第一槽位类型对应的信息库中。
可见，采用本实现方式，不但能够减少对话处理设备与用户的对话轮次，及时为用户提供服务，而且，对话处理设备能够及时更新该对话处理设备所维护的槽位信息。
一种可能的实现方式中，对话处理设备从用户设备接收对话信息之前，包括：
所述对话处理设备接收添加标签指令，所述添加标签指令与所述第一槽位类型对应；
所述对话处理设备为所述第一槽位类型添加标签。
其中，根据槽位信息与服务的关系的描述，必选槽位信息包括指示服务标识的槽位信息，例如，叫车服务中的“顺风车”，以及用于确定服务的常规槽位信息，例如，目的地“清华西门”。其中，服务器为用户提供的服务通常由相关服务人员自定义，所以，服 务器所提供的服务可能会更改，例如，“顺风车”服务项目下架，“拼车”服务项目上架，从而导致不存在相应服务标识对应的服务。
基于此，为了便于对话处理设备识别第一槽位类型和第二槽位类型，本申请可以预先按照上述必选槽位类型的属性，通过添加标签的方式标示出第一槽位类型，从而为本申请技术方案的实现提供依据。
一种可能的实现方式中，所述对话信息是文本或者音频。
第二方面，本申请提供了一种对话处理方法，该方法包括：
用户设备向对话处理设备发送对话信息；
所述用户设备从所述对话处理设备接收并提示第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述第一槽位类型对应的多个服务标识；
所述用户设备向所述对话处理设备发送目标服务标识，所述目标服务标识属于所述服务标识集；
所述用户设备从所述对话处理设备接收服务执行成功的信息，所述服务是所述目标服务标识对应的服务。
采用本实现方式，无论用户输入的对话信息是否包含第一槽位类型对应的槽位信息，无论用户输入的槽位信息与对话处理设备中预置的槽位信息的对应关系怎样，并且，无论用户输入的槽位信息与服务器提供的服务的对应关系怎样，该用户均能够获得想要的服务。这样不仅避免用户发起多轮次对话，而且，使得用户能够及时获知服务器所提供的多个服务，从而提高用户的使用体验。
一种可能的实现方式中，所述用户设备提示第一槽位类型对应的服务标识集，包括：
所述用户设备提示所述服务标识集中的全部服务标识。
一种可能的实现方式中，所述用户设备提示第一槽位类型对应的服务标识集，包括：
所述用户设备提示链接所述服务标识集的入口标识；
所述用户设备接收用户输入的打开指令；
所述用户设备提示所述入口标识链接的页面，所述页面中包括所述服务标识集中的全部服务标识。
一种可能的实现方式中，所述用户设备提示第一槽位类型对应的服务标识集，包括：
所述用户设备提示菜单控件，所述菜单控件中提示所述服务标识集中的部分服务标识；
所述用户设备接收用户输入的提示指令；
所述用户设备切换所述菜单控件中所提示的服务标识。
其中，一些实施例中，第一槽位类型对应的服务标识可能是两个或者三个。另一些实施例中，第一槽位类型对应的服务标识可能是十几个。采用本实现方式，能够灵活设置服务标识提示的方式，从而提高用户的使用体验。
第三方面，本申请提供了一种对话处理方法，该方法包括：对话处理设备从用户设备接收对话信息，然后，确定该对话信息中全部必选槽位类型所对应的关键词。若该全部必选槽位类型对应的关键词中，任一关键词对应相应槽位类型中的至少两条槽位信息，对话处理设备确定该至少两条槽位信息中对应最优服务方案的特定槽位信息。进而，对话处理设备可以向服务器请求该特定槽位信息对应的服务，在从服务器接收到相应服务 执行成功的信息之后，将相应服务执行成功的信息发送到用户设备。
采用本实现方式，对话处理设备可以在用户输入的关键词含义模糊的情况下，自动为用户提供相对较优的服务，不仅能够避免与用户多轮次对话，而且操作灵活，能够进一步提高用户的使用体验。
第四方面，本申请提供了一种对话处理方法，该方法包括：
对话处理设备从用户设备接收对话信息；
对话处理设备向第一服务器请求该槽位信息对应的服务；
若从第一服务器接收到的该服务执行失败的信息，对话处理设备向第二服务器请求该槽位信息对应的服务；
若从第二服务器接收到的该服务执行成功的信息，对话处理设备向用户设备发送该服务执行成功的信息。
采用本实现方式，在第一服务器无法提供服务时，对话处理设备可以触发第二服务器提供相应服务。这样能够减少对话处理设备与用户的对话轮次，及时为用户提供服务。
第五方面，本申请提供了一种对话处理方法，该方法包括：
接收用户输入的对话信息；
若所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息，从本地应用或者服务器获取所述第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述第一槽位类型对应的多个服务标识；
提示所述服务标识集中的服务标识；
若所述用户从所述服务标识集中确定的目标服务标识，向所述本地应用或者所述服务器请求所述目标服务标识对应的服务；
提示所述服务执行成功的信息。
一种可能的实现方式中，在接收用户输入的对话信息之后，在从本地应用或者服务器获取所述第一槽位类型对应的服务标识集之前，包括：
若从所述对话信息中未解析出所述第一槽位类型对应的槽位信息，确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
一种可能的实现方式中，在接收用户输入的对话信息之后，在从本地应用或者服务器获取所述第一槽位类型对应的服务标识集之前，包括：
解析所述对话信息得到所述第一槽位类型对应的槽位信息；
向所述本地应用或者所述服务器请求所述槽位信息对应的服务；
若从所述本地应用和所述服务器均请求服务失败，确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
一种可能的实现方式中，所述从本地应用或者服务器获取所述第一槽位类型对应的服务标识集，包括：
从所述本地应用获取所述第一槽位类型对应的服务标识集；或者，
从所述服务器获取所述第一槽位类型对应的服务标识集；或者，
调用所述本地应用从所述服务器获取所述第一槽位类型对应的服务标识集。
一种可能的实现方式中，提示所述服务标识集中的服务标识包括：
提示所述服务标识集中的全部服务标识；或者，
提示链接所述服务标识集的入口标识；
若用户输入的打开指令，提示所述入口标识链接的页面，所述页面中包括所述服务标识集中的全部服务标识；或者，
提示菜单控件，所述菜单控件中提示所述服务标识集中的部分服务标识；
若用户输入的提示指令，切换所述菜单控件中所提示的服务标识。
第六方面，本申请提供了一种对话处理设备，该对话处理设备具有实现上述方法中无线接入网设备行为的功能。所述功能可以通过硬件实现，也可以通过硬件执行相应的软件实现。所述硬件或软件包括一个或多个与上述功能相对应的模块。在一个可能的设计中，上述无线接入网设备的结构中包括处理器和收发器，所述处理器被配置为处理该对话处理设备执行上述方法中相应的功能。所述收发器用于实现上述对话处理设备与用户设备和服务器之间的信息交互。所述对话处理设备还可以包括存储器，所述存储器用于与处理器耦合，其保存该对话处理设备必要的程序指令和数据。
第七方面，本申请提供了一种用户设备，该用户设备具有实现上述方法中用户设备行为的功能。所述功能可以通过硬件实现，也可以通过硬件执行相应的软件实现。所述硬件或软件包括一个或多个与上述功能相对应的模块。在一个可能的设计中，上述用户设备的结构中包括处理器和收发器，所述处理器被配置为处理该用户设备备执行上述方法中相应的功能。所述收发器用于实现上述用户设备与对话处理设备之间的通信。所述用户设备还可以包括存储器，所述存储器用于与处理器耦合，其保存该用户设备必要的程序指令和数据。
第八方面，本申请提供了一种人机交互系统，该人机交互系统包括用户设备、对话处理设备和服务器，该对话处理设备分别与该用户设备和该服务器进行通信，以完成上述第一方面、第二方面、第三方面、第一方面各种可能的实现方式、第二方面各种可能的实现方式中的任意一种实现方式所述的人机对话操作。其中，该对话处理设备是如第五方面所述的对话处理设备，该用户设备是如第六方面所述的用户设备。
第九方面，本申请提供了一种人机交互系统，该人机交互系统包括用户设备、对话处理设备、第一服务器和第二服务器，该对话处理设备分别与该用户设备、该第一服务器和该第二服务器进行通信，以完成上述第四方面所述的人机对话操作。其中，该对话处理设备是如第五方面所述的对话处理设备，该用户设备是如第六方面所述的用户设备。
第十方面，本申请提供了一种计算机可读存储介质，所述计算机可读存储介质中存储有指令，当其在计算机上运行时，使得计算机执行上述任意可能的方法。
为解决现有人机对话对话轮次多的问题，本申请中，在服务器自定义的服务项目与相应槽位类型的槽位信息不对应的情况下，对话处理设备将服务器维护的服务标识反馈给用户，从而使得用户输入的相应槽位信息一定对应服务器中某项服务项目。采用本实现方式，不但能够及时向用户展示服务项目，而且能够避免用户与对话处理设备发起多轮对话，提高执行服务的效率，进而，提高用户的使用体验。
附图说明
为了更清楚地说明本申请的技术方案，下面将对实施例中所需要使用的附图作简单地介绍，显而易见地，对于本领域普通技术人员而言，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请提供的多轮次对话的一种实施方式的用户界面；
图2A是本申请提供的对话处理设备10的硬件结构的示意图；
图2B是本申请提供的对话处理设备10的软件架构的示意图；
图3A是本申请提供的用户设备20的硬件结构的示意图；
图3B是本申请提供的用户设备20的软件架构的示意图；
图4是本申请提供的一种操作界面的示意图；
图5A是本申请提供的用户界面的第一种实施方式的示意图；
图5B是本申请提供的用户界面的第二种实施方式的示意图；
图5C是本申请提供的用户界面的第三种实施方式的示意图；
图6是本申请提供的第一实施场景的用户界面示意图；
图7是本申请提供的第二实施场景的用户界面示意图；
图8是本申请提供的第三实施场景的用户界面示意图；
图9是本申请提供的第四实施场景的用户界面示意图；
图10是本申请提供的第五实施场景的用户界面示意图；
图11是本申请提供的人机交互系统的第一种实施方式的结构示意图；
图12是本申请提供的对话处理方法的第一种实施方式的方法流程图；
图13是本申请提供的对话处理方法的第二种实施方式的方法流程图；
图14是本申请提供的人机交互系统的第二种实施方式的结构示意图；
图15是本申请提供的对话处理方法的第四种实施方式的方法流程图；
图16是本申请提供的对话处理方法的第五种实施方式的方法流程图。
具体实施方式
下面将结合本申请中的附图，对本申请的技术方案进行清楚地描述。
本申请以下实施例中所使用的术语只是为了描述特定实施例的目的，而并非旨在作为对本申请的限制。如在本申请的说明书和所附权利要求书中所使用的那样，单数表达形式“一个”、“一种”、“所述”、“上述”、“该”和“这一”旨在也包括复数表达形式，除非其上下文中明确地有相反指示。还应当理解，尽管在以下实施例中可能采用术语第一、第二等来描述某一类对象，但所述对象不应限于这些术语。这些术语仅用来将该类对象的具体对象进行区分。例如，以下实施例中可能采用术语第一、第二等来描述服务器，但服务器不应限于这些术语。这些术语仅用来将所述不同的服务器进行区分。以下实施例中可能采用术语第一、第二等来描述的其他类对象同理，此处不再赘述。另外，本申请中使用的术语“和/或”是指并包含一个或多个所列出项目的任何或所有可能组合。
在对本申请的技术方案进行说明之前，首先对本申请的技术场景和本申请涉及的技术术语进行介绍。
本申请涉及对话处理设备与用户设备(user equipment，UE)之间，以及该对话处理设备与服务器之间的信息交互。其中，该用户设备是用户与对话处理设备进行对话的媒介。该用户设备将用户输入的信息传输至对话处理设备，并提示对话处理设备反馈的信息。该对话处理设备接收到用户设备发送的信息之后，可以请求服务器执行相应信息指示的服务。在从该服务器接收到该服务的执行结果后，该对话处理设备将该执行结果发送到用户设备。
其中，对话管理设备中可以维护意图、槽位类型和槽位信息。服务器中维护服务信息。以下结合示例对技术术语“意图”、“槽位类型”、“槽位信息”和“服务信息”的含义以及各技术术语之间的关系进行介绍。
意图(intent)是指用户的目的，用于指示用户的需求。对话处理设备可以从用户设备输入的信息中识别用户的意图。示例性的，用户设备的输入信息是“帮我预订A市飞到B市的机票”，对话处理设备可以从该输入信息中识别出用户的意图是“预订机票”。其中，意图的识别模型可以根据大量语料训练得到，该大量语料是采用不同表达方式表达该意图的语料。另外，对话处理设备可以维护一个或多个意图。
槽位信息(slot)是指意图内容的具体信息，也是用于触发特定服务的关键信息。槽位信息例如是用户设备输入信息中的关键词。示例性的，输入信息“帮我预订A市飞到B市的机票”中的“A市”和“B市”即为该输入信息的槽位信息。槽位信息可以包括必选槽位信息和可选槽位信息。必选槽位信息是指触发服务必不可少的槽位信息。可选槽位信息是指触发服务非必要的槽位信息。示例性的，触发服务“预订机票”的必选槽位信息包括起始地“A市”，起飞时间“XX年XX月XX日XX时XX分”和目的地“B市”。而该示例中，可选槽位信息例如是“B市所属国家”。
槽位类型(slot_type)用于标识槽位信息所属的分类。示例性的，槽位信息“A市”和槽位信息“B市”属于槽位类型“地址”，槽位信息“XX年XX月XX日XX时XX分”属于槽位类型“时间”。可以理解为，一个槽位类型对应标识一个特定的知识库。其中，根据槽位信息的属性，属于一种槽位类型的所有槽位信息可以以不同的结构形式存储在对话处理设备中。例如，槽位类型“地址”对应的槽位信息可以以城市名列表的形式存储在对话处理设备中。再如，槽位类型“时间”对应的槽位信息可以是时间的全部表达方式。通常，一个意图可以对应一个或者多个槽位类型。例如，意图“预订机票”对应槽位类型“地址”和槽位类型“时间”。
服务是指服务器有目的的执行的动作，以完成特定的服务项目。在服务器中，服务可以通过服务标识标示。本申请中，一个或者多个必选槽位信息可以确定一项服务。其中，该服务的服务标识对应该一个或者多个必选槽位信息中的一项槽位信息，或者，该服务的服务标识是该一个或者多个必选槽位信息中的一项槽位信息。如上述对必选槽位信息的介绍所述，槽位信息“A市”，槽位信息“B市”和槽位信息“XX年XX月XX日XX时XX分”确定服务“预订XX年XX月XX日XX时XX分从A市飞到B市的机票”。而机票预订服务器将“XX年XX月XX日XX时XX分从A市飞到B市”的一张机票与用户信息建立联系，且将该机票标记为被预订状态，即为执行预订“XX年XX月XX日XX时XX分从A市飞到B市”的机票的服务。其中，服务的表现形式可以是指令、程序或者进程。
对话处理设备是能够处理用户发送的语音或者文本，并向用户反馈语音或者文本的设备。一些实施例中，该对话处理设备承载在人机对话的机器人(robot，Bot)中，其中，该Bot独立于用户所使用的用户设备存在，并与该用户设备通过无线通信进行信息交互。在另一些实施例中，支持该对话处理的技能(skill)或者程序搭载在用户设备中。此种场景下，用户设备与对话处理是一台电子设备实体。
服务器可以是承载服务项目对应的程序的计算机，操作系统或者云服务器(elastic compute service,ECS)。该服务器所承载的程序在运行时，能够提供上述的服务项目，并且每项服务项目对应相应的服务标识。其中，该服务器所承载的程序可以是通过特定 计算机程序语言编写的源代码，例如超文本标记语言(hyper text markup language，HTML)，超文本预处理器(hypertext preprocessor，PHP)，java脚本(JavaScript，JS)等。该源代码可以由浏览器或与浏览器功能类似的网页显示组件加载和显示为用户可识别的内容。基于此，相关技术人员可以通过修改该源代码的方式在该服务器中增加新的服务项目，或者删除已有的服务项目。
一些实施例中，用户设备可以是还包含其它功能诸如个人数字助理和/或音乐播放器功能的便携式电子设备，诸如手机、平板电脑、具备无线通讯功能的可穿戴电子设备(如智能手表)等。便携式电子设备的示例性实施例包括但不限于搭载
Figure PCTCN2019124441-appb-000001
Figure PCTCN2019124441-appb-000002
或者其它操作系统的便携式电子设备。上述便携式电子设备也可以是其它便携式电子设备，诸如具有触敏表面或触控面板的膝上型计算机(Laptop)等。还应当理解的是，在其他一些实施例中，上述电子设备也可以不是便携式电子设备，而是具有触敏表面或触控面板的台式计算机。
用户设备可以通过用户界面(user interface，UI)与对话处理设备进行信息交互。用户界面是应用程序或操作系统与用户之间进行交互和信息交换的介质接口，它实现信息的内部形式与用户可以接受形式之间的转换。应用程序的用户界面是通过java、可扩展标记语言(extensible markup language，XML)等特定计算机语言编写的源代码，界面源代码在终端设备上经过解析，渲染，最终呈现为用户可以识别的内容，比如图片、文字、按钮等控件。控件(control)也称为部件(widget)，是用户界面的基本元素，典型的控件有工具栏(toolbar)、菜单栏(menu bar)、文本框(text box)、按钮(button)、滚动条(scrollbar)、图片和文本。界面中的控件的属性和内容是通过标签或者节点来定义的，比如XML通过<Textview>、<ImgView>、<VideoView>等节点来规定界面所包含的控件。一个节点对应界面中一个控件或属性，节点经过解析和渲染之后呈现为用户可视的内容。
本申请所述的对话是指“人机对话”，即用户与对话处理设备的信息交互。一些实施例中，用户可以通过用户设备向对话处理设备发送音频。另一些实施例中，用户可以通过用户设备向对话处理设备发送文本。相应的，一些实施例中，对话处理设备可以将执行结果转换成音频，然后，将该音频发送到用户设备。另一些实施例中，对话处理设备可以将执行结果转换成文本，然后，将该文本发送到用户设备。基于此，用户与对话处理设备进行音频交互，用户与对话处理设备进行文本交互，以及用户与对话处理设备中一方发送音频另一方发送文本的交互，均属于本申请“人机对话”的范畴。
结合现有的对话处理方法，若对话处理设备中预置的槽位信息不包含用户设备输入的必选槽位信息，该对话处理设备将无法确定待请求的第一服务，从而导致第一服务执行失败。若服务器所包含的服务中不包含对话处理设备请求的第一服务，该服务器也无法执行对话处理设备请求的第一服务，从而也会导致第一服务执行失败。而在第一服务执行成功的情况下，用户请求的服务才能够被提供。所以，对话处理设备向用户设备发送第一服务执行失败的信息后，该对话处理设备可能会接收到用户设备发送的触发第二服务的对话。其中，第二服务与第一服务的属性相同。
图1示意了多轮次对话的示例性用户界面。本实施例中，用户通过人机对话的方式请求车辆服务。如图1所示，用户向用户设备输入“帮我叫个顺风车去清华西门”。对话处理设备接收到用户设备发送的信息“帮我叫个顺风车去清华西门”之后，生成请求信 息，并将该请求信息发送到目标叫车服务器。其中，该请求信息包括用户当前的地址，目的地“清华西门”和必选槽位信息“顺风车”。示例性的，该目标叫车服务器不提供“顺风车”服务，所以无法执行用户触发的服务。相应的，对话处理设备向用户设备发送“对不起，您所请求的服务无法执行”的信息。进而，用户可以继续向用户设备输入“那帮我叫个快车吧”，从而发起与对话处理设备的新一轮的对话。对话处理设备按照上述请求“顺风车”服务的流程，向目标叫车服务器请求“快车”服务。若目标叫车服务器也不提供“快车”服务，则用户与对话处理设备可能需要进行更多轮次的对话。这样不仅使得对话处理设备不断的执行对话处理操作，占用资源，而且使得用户的使用体验较差。
有鉴于此，本申请提供了一种对话处理方法及设备，在该对话处理设备中预置的槽位信息与用户输入的必选槽位信息不对应，和/或，该对话处理设备中预置的槽位信息与服务器所包含的服务不对应的情况下，该对话处理设备能够及时提供有效的信息和/或处理策略，不仅能够减少不必要的对话轮次，而且处理方式灵活，能够提高用户的使用体验。
以下介绍本申请中提供的示例性对话处理设备10。
图2A示出了对话处理设备10的结构示意图。
对话处理设备10包括处理器11，存储器12和收发器13。其中，处理器11，存储器12和收发器13通过总线接口连接。
可以理解的是，图2A示意的结构并不构成对对话处理设备10的具体限定。在本申请另一些实施例中，对话处理设备10可以包括比图示更多或更少的部件，或者组合某些部件，或者拆分某些部件，或者不同的部件布置。图示的部件可以以硬件，软件或软件和硬件的组合实现。
处理器11可以是中央处理器(central processing unit，CPU)，网络处理器(network processor，NP)或者CPU和NP的组合。处理器还可以进一步包括硬件芯片。上述硬件芯片可以是专用集成电路(application-specific integrated circuit，ASIC)，可编程逻辑器件(programmable logic device，PLD)或其组合。上述PLD可以是复杂可编程逻辑器件(complex programmable logic device，CPLD)，现场可编程逻辑门阵列(field-programmable gate array，FPGA)，通用阵列逻辑(generic array logic，GAL)或其任意组合。
存储器12可以包括易失性存储器(volatile memory)，例如随机存取存储器(random-access memory，RAM)；存储器也可以包括非易失性存储器(non-volatile memory)，例如只读存储器(read-only memory，ROM)，快闪存储器(flash memory)，硬盘(hard disk drive，HDD)或固态硬盘(solid-state drive，SSD)；存储器还可以包括上述种类的存储器的组合。
收发器13可以是有线收发器，无线收发器或其组合。有线收发器例如可以为以太网接口。以太网接口可以是光接口，电接口或其组合。无线收发器例如可以为无线局域网收发器，蜂窝网络收发器或其组合。
总线接口可以包括任意数量的互联的总线和桥，具体由处理器11代表的一个或多个处理器和存储器12代表的存储器的各种电路链接在一起。总线接口还可以将诸如外围设备、稳压器和功率管理电路等之类的各种其他电路链接在一起。收发器13提供用于在传输介质上与各种其他设备(如用户设备，服务器)通信的单元。处理器11负责管理总线 架构和通常的处理，存储器12可以存储处理器在执行操作时所使用的数据。
本领域技术任何还可以了解到本申请实施例列出的各种说明性逻辑块(illustrative logical block)和步骤(step)可以通过电子硬件、电脑软件，或两者的结合进行实现。这样的功能是通过硬件还是软件来实现取决于特定的应用和整个系统的设计要求。本领域技术人员可以对于每种特定的应用，可以使用各种方法实现所述的功能，但这种实现不应被理解为超出本申请实施例保护的范围。
图2A所示的硬件架构涉及到执行对话处理的硬件部件。示例性的，收发器13可以用于执行对话处理设备与用户设备之间信息的收发，例如，接收用户设备发送的语音信息，向用户设备发送服务标识和执行结果等。该收发器13还可以用于对话处理设备与服务器之间信息的收发，例如，向服务器发送请求服务的信息，以及从服务器接收服务标识和执行结果等。存储器12用于存储指令和数据。在一些实施例中，存储器12可以保存处理器11刚用过或循环使用的指令或数据。如果处理器11需要再次使用该指令或数据，可从该存储器12中直接调用，从而避免重复存取。存储器12还用于存储一个或者多个意图信息，该一个或者多个意图信息中每个意图信息对应的一个或多个槽位类型，以及该一个或多个槽位类型中每个槽位类型所包含的槽位信息。处理器11调用存储器12中的指令和数据执行除信息收发之外的其他操作，例如，识别语音，解析语义，并根据解析后的语义确定用户意图和槽位信息等。
图2B是本申请的对话处理设备10的软件结构框图。
对话处理设备10的软件系统可以包括用于执行对话处理的各种功能的功能模块。例如，操作系统，收发模块，语音识别(automatic speech recognition，ASR)模块，语义理解(natural language understanding，NLU)模块，对话生成(natural language generation，NLG)模块，槽位信息识别模块，请求信息生成模块等。
一些实施例中，对话处理设备10还包括编辑接口。该编辑接口被触发时，可以呈现编辑操作界面。
图2B所示的软件系统涉及到执行对话处理的功能模块，其中，操作系统驱动各模块执行操作。示例性的，收发模块接收到用户的对话语音后，ASR模块对语音进行识别，然后，NLU模块解析识别后语音的语义。进而，槽位信息识别模块可以从解析后的语义中识别槽位信息，请求信息生成模块生成请求服务的信息。在收发模块从服务器接收到服务的执行结果之后，NLG模块生成将执行结果生成语音，进而收发模块将所生成的语音发送到用户设备。
以下介绍本申请以下实施例中提供的示例性用户设备20。
图3A示出了用户设备20的结构示意图。
用户设备20可以包括处理器110，外部存储器接口120，内部存储器121，通用串行总线(universal serial bus，USB)接口130，充电管理模块140，电源管理模块141，电池142，天线1，天线2，移动通信模块150，无线通信模块160，音频模块170，扬声器170A，受话器170B，麦克风170C，耳机接口170D，传感器模块180，按键190，马达191，指示器192，摄像头193，显示屏194，以及用户标识模块(subscriber identification module，SIM)卡接口195等。其中传感器模块180可以包括压力传感器180A，陀螺仪传感器180B，气压传感器180C，磁传感器180D，加速度传感器180E，距离传感器180F，接近光传感器180G，指纹传感器180H，温度传感器180J，触摸传感器180K，环境光传 感器180L，骨传导传感器180M等。
可以理解的是，本申请示意的结构并不构成对用户设备20的具体限定。在本申请另一些实施例中，用户设备20可以包括比图示更多或更少的部件，或者组合某些部件，或者拆分某些部件，或者不同的部件布置。图示的部件可以以硬件，软件或软件和硬件的组合实现。
处理器110可以包括一个或多个处理单元，例如：处理器110可以包括应用处理器(application processor，AP)，控制器，存储器等。其中，不同的处理单元可以是独立的器件，也可以集成在一个或多个处理器中。在一些实施例中，用户设备20也可以包括一个或多个处理器110。
其中，控制器可以是用户设备20的神经中枢和指挥中心。控制器可以根据指令操作码和时序信号，产生操作控制信号，完成取指令和执行指令的控制。
处理器110中还可以设置存储器，用于存储指令和数据。在一些实施例中，处理器110中的存储器为高速缓冲存储器。该存储器可以保存处理器110刚用过或循环使用的指令或数据。如果处理器110需要再次使用该指令或数据，可从所述存储器中直接调用。避免了重复存取，减少了处理器110的等待时间，因而提高了用户设备20的效率。
在一些实施例中，处理器110可以包括一个或多个接口。该一个或多个接口中包括实现处理器110与音频模块170之间的通信的接口，以及连接处理器110与显示屏194的接口等。此处不再详述。
用户设备20的无线通信功能可以通过天线1，天线2，移动通信模块150，无线通信模块160，调制解调处理器以及基带处理器等实现。
无线通信模块160可以提供应用在用户设备20上的包括无线局域网(wireless local area networks，WLAN)(如无线保真(wireless fidelity，Wi-Fi)网络)，等无线通信的解决方案。在一些实施例中，用户设备20的天线2和无线通信模块160耦合，使得用户设备20可以通过无线通信技术与网络以及其他设备通信。
用户设备20的显示屏194，以及应用处理器等可以实现提示功能。显示屏194用于提示文本，菜单，音频控件等。
用户设备20可以通过音频模块170，扬声器170A，受话器170B，麦克风170C，耳机接口170D，以及应用处理器等实现音频功能。例如音频播放，录音等。
音频模块170用于将数字音频信息转换成模拟音频信号输出，也用于将模拟音频输入转换为数字音频信号。音频模块170还可以用于对音频信号编码和解码。
扬声器170A，也称“喇叭”，用于将音频电信号转换为声音信号。用户设备20可以通过扬声器170A收听对话处理设备发送的音频对话。
受话器170B，也称“听筒”，用于将音频电信号转换成声音信号。当用户设备20接收到音频信息时，用户可以通过将受话器170B靠近耳朵接听该音频信息的声音。
麦克风170C，也称“话筒”，“传声器”，用于将声音信号转换为电信号。当发送语音信息时，用户可以通过人嘴靠近麦克风170C发声，将声音信号输入到麦克风170C。用户设备20可以设置至少一个麦克风170C。
图3A示例性所示的用户设备20可以通过显示屏194显示提示以下各个实施例中所描述的各个用户界面。用户设备20可以通过麦克风170C向各个用户界面输入语音信息，还可以通过扬声器170A或者受话器170B播放用户界面所提示的音频。
用户设备20的软件系统可以采用分层架构。本申请以分层架构的Android系统为例，示例性说明用户设备20的软件结构。
图3B是本申请的用户设备20的软件结构框图。
分层架构将软件分成若干个层，每一层都有清晰的角色和分工。层与层之间通过软件接口通信。在一些实施例中，将Android系统分为四层，从上至下分别为应用程序层，应用程序框架层，安卓运行时(Android runtime)和系统库，以及内核层。
应用程序层可以包括一系列应用程序包。
如图3B所示，应用程序包可以包括相机，图库，WLAN，蓝牙，音乐，视频，短信息等应用程序。
应用程序框架层为应用程序层的应用程序提供应用编程接口(application programming interface，API)和编程框架。应用程序框架层包括一些预先定义的函数。
如图3B所示，应用程序框架层可以包括窗口管理器，内容提供器，视图系统，电话管理器，资源管理器，通知管理器等。
窗口管理器用于管理窗口程序。窗口管理器可以获取显示屏大小，判断是否有状态栏，锁定屏幕，截取屏幕等。
内容提供器用来存放和获取数据，并使这些数据可以被应用程序访问。所述数据可以包括视频，图像，音频，文本等。
视图系统包括可视控件，例如显示文字的控件等。视图系统可用于构建应用程序。显示界面可以由一个或多个视图组成的。
媒体库支持多种常用的音频回放和录制等。媒体库可以支持多种音视频编码格式，例如:MPEG4，H.264，MP3，AAC，AMR，JPG，PNG等。
内核层是硬件和软件之间的层。内核层至少包含显示驱动，音频驱动，传感器驱动。
图3B所示的软件系统涉及到执行人机对话的应用呈现(如窗口管理器，视图系统)，提供音频编解码的服务，以及内核和基本通信协议。
下面结合人机对话的场景，示例性说明用户设备20软件以及硬件的工作流程。
显示器121呈现人机对话的应用界面之后，麦克风170C接收用户的语音音频。相应的，内核层的音频驱动被触发，将用户的语音音频转换为电信号。无线通信模块160将该电信号发送到对话处理设备。无线通信模块160从对话处理设备接收音频之后，内核层的显示驱动调用窗口管理器显示该音频。压力传感器180A接收到播放音频的指令后，相应硬件中断被发给内核层。内核层的音频驱动触发音频模块170将音频解码，解码后的音频通过扬声器170A或者受话器170B播放。
结合上述对槽位信息与服务的关系的描述，必选槽位信息包括指示服务标识的槽位信息，例如，叫车服务中的“顺风车”，以及用于确定服务的常规槽位信息，例如，目的地“清华西门”。其中，服务器为用户提供的服务通常由相关服务人员自定义，所以，服务器所提供的服务可能会更改，例如，“顺风车”服务项目下架，“拼车”服务项目上架，从而导致不存在相应服务标识对应的服务。然而，一方面，用户可能无法及时获知服务器所提供的服务有哪些，另一方面，对话处理设备中维护的相关槽位信息可能也无法被及时更新，从而产生本申请上述的一些列问题。本申请以下将必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，称为“第一槽位类型”，将必选槽位信息中的常规槽位信息的槽位类型称为“第二槽位类型”。相应的，下文中引用的“第一槽位类型”均 指必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，下文中引用的“第二槽位类型”均指必选槽位信息中的常规槽位信息的槽位类型，以下不再赘述。
其中，为了便于对话处理设备识别第一槽位类型和第二槽位类型，开发人员在设计对话处理设备过程中，可以按照上述必选槽位类型的属性，通过添加标签的方式标示出第一槽位类型。例如，开发人员可以为必选槽位类型中第一槽位类型添加标签“M”，而不为必选槽位类型中第一槽位类型添加任何标签。相应的，对话处理设备读取到携带标签“M”的槽位类型，则将该槽位类型确定为第一槽位类型。对话处理设备读取到未携带标签“M”的必选槽位类型，则将该必选槽位类型确定为第二槽位类型。
一些实施例中，开发人员在设计槽位类型相关的程序过程中，可以直接为第一槽位类型添加上标签。另一些实施例中，本申请可以提供编辑界面，开发人员可以在该编辑界面上通过勾选的方式为槽位类型分类。
图4示意了一种槽位类型的示例性编辑界面。该编辑界面示意的槽位列表中可以包括槽位类型名称，槽位类型对应的槽位信息库，必选槽位选项，提示选项和提问五个项目。其中，槽位类型名称用于为槽位类型命名，以标示相应槽位类型。槽位类型对应的槽位信息库可以是一个链接，用于链接同一行槽位类型对应的槽位信息库。提问指示相应行槽位类型的槽位信息对应的各种表达方式。必选槽位选项若被选中，则认为相应行的槽位类型被标注为必选槽位类型。相应的，该槽位类型被添加“必选”的标签。提示选项是本实施中用于标注是否是第一槽位类型的选项，基于此，同理，提示选项若被选中，则认为相应行的槽位类型被标注为第一槽位类型。相应的，该槽位类型被添加“提示”的标签。需要指出的，本实施例中，只有相应行的必选槽位选项被选中，该行的提示选项才能够被操作。另外，开发者还可以按照需求触发“+新增槽位类型”添加槽位类型，并且可以点击删除符号删除相应删除符号对应行的槽位类型。
根据上述描述，图4示意的实施例中，槽位类型“车辆类型”和槽位类型“目的地”是必选槽位类型，槽位类型“出发地”是可选槽位类型。必选槽位类型中“车辆类型”是第一槽位类型，“目的地”是第二槽位类型。
基于此，本申请的技术方案，在用户未输入第一槽位类型对应的槽位信息，或者，用户输入的第一槽位类型对应的槽位信息没有匹配的服务的情况下，对话处理设备将第一槽位类型对应的多个服务标识反馈给用户，以使用户从该多个服务标识中选择一个服务标识作为相应槽位信息，从而能够快速准确的触发一项服务。
一些实施例中，该多个服务标识可以是服务器中第一槽位类型对应的全部服务标识。另一些实施例中，该多个服务标识可以是服务器中第一槽位类型对应的部分服务标识。具体的，该多个服务标识可以是其中用户好评高于一定值的服务标识，例如，用户好评高于95％的服务项目对应的服务标识。或者，该多个服务标识可以是距离当前时间一定时间内更新的服务项目的服务标识，例如，该多个服务标识是最近3个月内更新的服务项目的服务标识。
可以理解的是，上述“多个服务标识”的选取条件，仅是本申请的部分示意性实施场景对应的选取条件，不构成对“多个服务标识”选取的具体限定。在另一些实施例中，“多个服务标识”的选取可以根据其他实施场景灵活设定。此处不再赘述。
以下结合示例性用户界面对本申请用户端的实施场景进行介绍。
第一实施场景：用户向用户设备输入对话信息，该对话信息中不包含第一槽位类型 对应的槽位信息。然后，该用户设备可以从对话处理设备接收该第一槽位类型对应的全部服务标识。用户可以从该全部服务标识中选择一个服务标识作为第一槽位类型的槽位信息输入用户设备。
一些实施例中，第一槽位类型对应的全部服务标识可能是两个或者三个，例如，第一槽位类型是叫车服务中的“车辆类型”，该“车辆类型”对应“顺风车”，“专车”和“快车”三个服务标识。另一些实施例中，第一槽位类型对应的全部服务标识可能是十几个，例如，第一槽位类型是订餐服务中的“饮品”，该“饮品”对应“可乐”，“雪碧”，“豆浆”，“咖啡”，“果汁”，“牛奶”，“红茶”，“绿茶”，“奶茶”和“啤酒”等十个。而用户界面的尺寸有限，所以，一些实施例中，该全部服务标识的数量小于或者等于预设阈值时，用户设备可以在用户界面上提示该全部服务标识，如图5A示意的用户界面所示。其中，该预设阈值例如是5。另一些实施例中，该全部服务标识的数量大于该预设阈值时，用户设备可以以链接(link)的方式在用户界面上提示该全部服务标识，如图5B示意的用户界面所示。或者，用户设备可以以设置滚动条的菜单栏(menu bar)，如图5C示意的用户界面所示。
可以理解的是，图5A至图5C示意的用户界面并不构成对本申请的具体限定。在本申请另一些实施例中，用户界面可以采用不同于上述描述的方式提示服务标识。此处不再详述。
进一步的，若对话信息中包括第二槽位类型对应的槽位信息，用户设备可以从对话处理设备接收到服务执行成功的信息。若对话信息中不包括第二槽位类型对应的槽位信息，用户设备可以从对话处理设备接收到要求用户输入相应槽位信息的对话，用户可以响应相应对话输入相应槽位信息，最终，用户设备可以从对话处理设备接收到服务执行成功的信息。
示例性的，如图6所示，用户向用户界面输入“帮我叫个车”，然后，用户设备提示“当前可用车辆类型包括：专车、快车和拼车，请问您想要哪一种？”。进而，用户选择“快车”。然后，用户界面可以提示“请问您要去哪里？”。在用户输入“清华西门”之后，用户界面提示信息“已经成功为您预约了从您的位置到清华西门的快车”。
第二实施场景：用户向用户设备输入对话信息，该对话信息中包含第一槽位类型对应的槽位信息，但是服务器中不包含该槽位信息对应的服务。进而，该用户设备可以从对话处理设备接收该第一槽位类型对应的多个服务标识。用户可以从该多个服务标识中选择一个服务标识作为第一槽位类型的槽位信息输入用户设备。其中，服务标识对应的用户界面如图5A至图5C所示。另外，用户后续的操作，以及用户设备后续提示的对话内容，与第一实施场景相似，此处不再赘述。
可以理解的是，该多个服务标识在用户界面上提示的方式，如图5A至图5C所述，此处不再赘述。
示例性的，如图7所示，用户向用户界面输入“帮我点一份咖啡”。本实施例中，例如咖啡已售完。那么，用户设备提示“当前有货的饮品包括：牛奶，可乐，雪碧，果汁，奶茶，豆浆，红茶，请问您想要哪一种？”。如图6所示，“饮品”对应的服务标识较多，所以，用户设备通过设置滚动条的菜单栏提示上述标识。用户可以通过上下滑动图6用户界面上的滚动条，查看“饮品”对应的标识。进而，用户例如选择了“红茶”。然后，用户界面提示信息“您所点的红茶15分钟后送到”。其中，本实施例所罗列的饮品种类 可以是“当前有货的饮品”中的部分饮品。
第三实施场景：用户向用户设备输入对话信息，该对话信息中包含第一槽位类型对应的槽位信息。其中，对话处理设备中不包含该槽位信息，但是服务器中包含该槽位信息对应的服务。在用户输入第二槽位类型对应的槽位信息后，用户设备提示服务执行成功的信息。
示例性的，如图8所示，用户向用户界面输入“帮我订一下双层汉堡A套餐”。本实施例中，对话处理设备中不包含“双层汉堡A套餐”对应的槽位信息，但是，用户设备可以接收信息“您预订的双层汉堡A套餐半个小时候送到”。
第四实施场景：用户向用户设备输入对话信息，该对话信息中包含第一槽位类型对应的槽位信息。但是第一服务器中不包含该槽位信息对应的服务。在用户输入第二槽位类型对应的槽位信息后，用户设备提示服务由第二服务器执行成功的信息。
示例性的，如图9所示，用户向用户界面输入“查一下明天上午10点到11点香港飞纽约的航班”。本实施例中，第一服务器是维护国内航班数据的服务器，例如是某国内旅游平台的服务器。但是，用户设备可以接收信息“国际航班官方数据提示，明天上午10点到11点香港飞纽约的航班有AA8925，上午10：35分从香港起飞”。
第五实施场景：用户向用户设备输入对话信息，该对话信息中包含第一槽位类型对应的槽位信息，和/或，第一槽位类型对应的槽位信息，但是至少一条上述槽位信息指示的含义不清楚。进而，用户设备提示相应服务按照最佳方案被执行的信息。
其中，槽位信息指示的含义不清楚是指，该槽位信息可以指示多个不同的具体信息，从而导致对话处理设备根据该槽位信息无法触发确定的服务。例如，信息“帮我预订北京站附近的酒店”，其中，“北京站附近的酒店”至少包括“距离北京站500米以内的酒店”，“北京站附近的五星级酒店”以及“距离北京站500米以内的五星级酒店”等更具体的信息。所以，本实施例中“北京站附近”和“酒店”指示的含义均不清楚。另外，最佳方案是指对于用户来讲，最多、快、好和/或省的方案，例如，对于用户来讲行程最短的路线，或者，需要用户支付最少费用的套餐。
示例性的，如图10所示，用户向用户界面输入“帮我预订北京站附近的酒店”。用户设备接收并提示“为您预定了距离北京站400米的五星级酒店X酒店的标准间”。其中，该“X酒店”例如是距离北京站最近的五星级酒店。
可以理解的是，图6至图10示意的实施场景，仅是本申请的部分示意性实施场景，并不构成对本申请适用场景的限定。在另一些实施例中，本申请还可以适用于其他任何用户通过人机对话触发服务的其他任何场景。此处不再赘述。
根据上述对第一实施场景至第五实施场景的描述可知，采用本申请的技术方案，无论用户输入的对话信息是否包含第一槽位类型对应的槽位信息，无论用户输入的槽位信息与对话处理设备中预置的槽位信息的对应关系怎样，并且，无论用户输入的槽位信息与服务器提供的服务的对应关系怎样，该用户均能够获得想要的服务。这样不仅避免用户发起多轮次对话，而且，使得用户能够及时获知服务器所提供的服务，从而提高用户的使用体验。
上述第一实施场景至第五实施场景均是从用户端进行描述，而本申请实现上述各场景效果的执行过程，详见下文对对话处理设备端的描述。以下结合附图和上述各实施场景对本申请对话处理设备端的执行过程进行介绍。
图11示出了本申请人机对话系统第一种实施方式的架构图。本实施例中，对话处理设备中维护的意图、槽位类型和槽位信息等，与服务器支持的服务对应。用户可以通过用户设备与对话设备对话。其中，对话处理设备可以如图2A和图2B所述，用户设备可以如图3A和图3B所述。
结合图11示意的人机对话系统，图12示出了本申请对话处理方法的一种实施方式的方法流程图。图12示出的对话处理方法100包括如下步骤：
步骤S101，对话处理设备从用户设备接收对话信息。
其中，对话信息可以是音频或者文本。若该对话信息是音频，对话处理设备可以对该音频进行语音识别，然后，对语音识别后的内容进行语义理解，从而确定该对话信息对应的意图。若该对话信息是文本，对话处理设备可以对该文本进行语义理解，从而确定该对话信息对应的意图。进一步的，对话处理设备确定该意图对应的一个或者多个必选槽位类型，并确定该一个或多个必选槽位类型中每个必选槽位类型是第一槽位类型还是第二槽位类型。然后，对话处理设备可以识别对话信息中是否包括第一槽位类型对应的槽位信息。其中，本实施例所述的第一槽位类型是指，必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型。本实施例所述的第二槽位类型是指，必选槽位信息中的常规槽位信息的槽位类型。
可选的，结合上述对第一槽位类型和第二槽位类型的描述，对话处理设备可以通过识别每个必选槽位类型是否包含预设标签，来确定该必选槽位类型是否是第一槽位类型。若该必选槽位类型包含预设标签，对话处理设备认为该必选槽位类型是第一槽位类型。若该必选槽位类型不包含预设标签，对话处理设备认为该必选槽位类型是第二槽位类型。其中，预设标签的设置过程以及该预设标签的内容，详见图4对应的实施例所述，此处不再详述。
可选的，对话处理设备可以识别对话信息中的一个或者多个关键词。进而，对应该一个或者多个关键词中每个关键词，一些实施例中，对话处理设备通过词向量模型确定该关键词所属的槽位类型。另一些实施例中，对话处理设备通过该关键词的表达格式确定该关键词所属的槽位类型，例如：根据关键词“12:30”的格式可以确定关键词“12:30”属于槽位类型“时间”。
步骤S102，若对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息，对话处理设备从服务器获取第一槽位类型对应的服务标识集。
其中，服务标识集包括第一槽位类型对应的全部服务标识。
步骤S103，对话处理设备将服务标识集发送到用户设备。
其中，对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息可以包括两种场景：场景一：对话信息中不包含第一槽位类型对应的槽位信息。场景二，对话信息中包含第一槽位类型对应的槽位信息，但是服务器中不包含该槽位信息对应的服务。基于此，为了避免用户多次输入第一槽位类型对应的槽位信息，在上述任一实施方式出现时，对话处理设备可以向服务器发送获取请求，该获取请求可以包括第一槽位类型的标识。然后，服务器响应该获取请求，将第一槽位类型对应的全部服务标识形成的服务标识集发送到对话处理设备。对话处理设备可以将该服务标识集发送到用户设备，以供用户选择任意服务标识作为该第一槽位类型对应的槽位信息。
对应情况一，对话处理设备若从对话信息中未获取到第一槽位类型对应的槽位信息， 则对话处理设备确定对话信息中不包含第一槽位类型对应的槽位信息。
示例性的，如上述第一实施场景以及图6示意的实施例。此处不再详述。
对应情况二，对话处理设备从对话信息中获取到第一槽位类型对应的槽位信息，进而，对话处理设备根据该对话信息中包含的全部必选槽位信息，向服务器请求该全部必选槽位信息对应的服务。若从服务器接收到服务执行失败的信息，对话处理设备确定对话信息中不包含能够确定第一槽位类型对应的槽位信息。
示例性的，如上述第二实施场景以及图7示例的实施例，对话处理设备接收到对话信息“帮我点一份咖啡”之后，能够从对话信息“帮我点一份咖啡”中确定第一槽位类型对应的槽位信息“咖啡”。然后，对话处理设备向服务器请求“订咖啡”的服务。由于咖啡已售完，所以服务器执行服务“订咖啡”失败，并可以向对话处理设备发送执行失败的信息。在另一些实施例中，服务器执行服务“订咖啡”失败后，还可以直接将图7示意的有货的饮品发送到对话处理设备。
其中，用户设备提示所述全部服务标识的方式，如图5A至图5C所示，此处不再赘述。
步骤S104，对话处理设备接收用户设备发送的目标服务标识，并向服务器请求该目标服务标识对应的服务。
其中，该目标服务标识属于上述服务标识集。本申请中，该目标服务标识被用作第一槽位类型对应的槽位信息。
可以理解的是，在执行本步骤之前，对话处理设备已经获知第二槽位类型对应的槽位信息。对于对话处理设备获知第二槽位类型对应的槽位信息的执行过程，本实施例不再详述。
步骤S105，对话处理设备从服务器接收服务执行成功的信息，并将该服务执行成功的信息发送到用户设备。
由此可见，本申请的技术方案，若服务器自定义的服务项目与相应槽位类型的槽位信息不对应，对话处理设备将服务器维护的全部服务标识反馈给用户，从而使得用户输入的相应槽位信息一定对应服务器中某项服务项目。采用本实现方式，不但能够及时向用户展示服务项目，而且能够避免用户与对话处理设备发起多轮对话，提高执行服务的效率，进而，提高用户的使用体验。
结合图11示意的人机对话系统，图13示出了本申请对话处理方法的另一种实施方式的方法流程图。图13示意的对话处理方法200包括如下步骤：
步骤S201，对话处理设备从用户设备接收对话信息。
其中，对话信息中包含第一槽位类型对应的槽位信息。此外，本实施例中，对话处理设备接收到对话信息之后的操作与方法100中步骤S101的描述相似，此处不再赘述。
步骤S202，对话处理设备向服务器请求该槽位信息对应的服务。
其中，本实施例中，即使该槽位信息未包含在对话处理设备中预置的槽位信息中，该对话处理设备依然根据该槽位信息请求服务器提供相应服务。
步骤S203，若对话处理设备中不包含该槽位信息，且从服务器接收到该服务执行成功的信息，对话处理设备将该槽位信息添加到第一槽位类型对应的信息库中。
其中，若该槽位信息未包含在对话处理设备中预置的槽位信息中，且该服务被服务器执行成功，说明服务器中维护的服务项目被更新，但是对话处理设备中第一槽位类型 对应的槽位信息未被及时更新。示例性的，结合图8示意的实施例，服务器端添加了新产品“双层汉堡A套餐”，但是对话处理设备未更新槽位类型“套餐”对应的信息库，则对话处理设备可以将“双层汉堡A套餐”对应的槽位信息添加到槽位类型“套餐”对应的信息库中。另外，本实施例中，“双层汉堡A套餐”对应的槽位信息可以是“双层汉堡A套餐”。
步骤S204，对话处理设备向用户设备发送该服务执行成功的信息。
可见，采用本实现方式，不但能够减少对话处理设备与用户的对话轮次，及时为用户提供服务，而且，对话处理设备能够及时更新该对话处理设备所维护的槽位信息。
其中，与方法200相对应的用户界面如上述第三实施场景以及图8所示，此处不再详述。
结合图11示意的人机对话系统，本申请还提供了对话处理方法的第三种实施方式。其中，对话处理方法300包括：
对话处理设备从用户设备接收对话信息，然后，确定该对话信息中全部必选槽位类型所对应的关键词。若该全部必选槽位类型对应的关键词中，任一关键词对应相应槽位类型中的至少两条槽位信息，对话处理设备确定该至少两条槽位信息中对应最优服务方案的特定槽位信息。进而，对话处理设备可以向服务器请求该特定槽位信息对应的服务，在从服务器接收到相应服务执行成功的信息之后，将相应服务执行成功的信息发送到用户设备。
其中，对应一个槽位类型中至少两条槽位信息的关键词，可以是包含多个具体槽位信息的概括性词汇。例如，“北京站附近的酒店”可以包括的槽位信息例如有，距离北京站1000米以内的全部的酒店的名称。再如，“清华”可以包括的槽位信息例如有，清华大学的所有门的地址。
本申请中，“最优服务方案”中“最优”的标准，根据服务内容以及用户的需求不同。例如，结合图10示意的实施例，“帮我预订北京站附近的酒店”最优的服务方案例如可以是距离北京站最近的酒店，也可以是距离北京站1000米以内星级最高的酒店。再如，对话信息“帮我叫个车去清华”，最优的服务方案例如可以是行驶路线最短的方案，也可以是费用最低的车辆。基于此，对话处理设备可以根据相应服务场景的计算模型确定。计算模型可以根据相应服务场景的大数据以及用户历史订单数据训练得到。此处不再详述。
进一步的，本实施例中，对话处理设备确定关键词对应的至少两条槽位信息后，可以分别判断每条槽位信息对应的服务是否是上述“最优服务方案”，若是，将该条槽位信息确定为特定槽位信息。
可见，采用本实现方式，对话处理设备可以在用户输入的关键词含义模糊的情况下，自动为用户提供相对较优的服务，不仅能够避免与用户多轮次对话，而且操作灵活，能够进一步提高用户的使用体验。
图14示出了本申请人机对话系统第二种实施方式的架构图。本实施例中，对话处理设备与第一服务器和第二服务器建立连接。用户可以通过用户设备与对话设备对话。其中，对话处理设备可以如图2A和图2B所述，用户设备可以如图3A和图3B所述。
结合图14示意的人机对话系统，图15示出了本申请对话处理方法的第四种实施方式的方法流程图。图15示出的对话处理方法400包括如下步骤：
步骤S401，对话处理设备从用户设备接收对话信息。
其中，对话信息中包含第一槽位类型对应的槽位信息。此外，本实施例中，对话处理设备接收到对话信息之后的操作与方法100中步骤S101的描述相似，此处不再赘述。
步骤S402，对话处理设备向第一服务器请求该槽位信息对应的服务。
步骤S403，若从第一服务器接收到的该服务执行失败的信息，对话处理设备向第二服务器请求该槽位信息对应的服务。
步骤S404，若从第二服务器接收到的该服务执行成功的信息，对话处理设备向用户设备发送该服务执行成功的信息。
其中，与方法400相对应的用户界面如上述第四实施场景以及图9所示，此处不再详述。
示例性的，本实施例中，第一服务器可以是提供定制化服务的服务器，第二服务器可以是与第一服务器提供的服务相关的官方服务器。结合图9示意的实施例，第一服务器例如是提供国内旅游服务的服务器，第二服务器例如是提供国内和国际航班数据的官方服务器。
可见，采用本实现方式，在第一服务器无法提供服务时，对话处理设备可以触发第二服务器提供相应服务。这样能够减少对话处理设备与用户的对话轮次，及时为用户提供服务。
在一种可选的实施方式中，上述图11至图15示意的实施例，对话处理设备可以是云端的方案，即对话处理设备为服务器，是与用户设备相对独立的实体。
在另一种可选的实施方式中，上述图11至图15示意的实施例，对话处理设备可以是端云结合的方案，即对话处理设备包括服务器和用户设备上的功能模块。例如对话处理设备可以包括服务器的Bot平台和用户设备上的应用。
在另一种可选的实施方式中，上述图11至图15示意的实施例，对话处理设备可以是端侧的方案，即对话处理设备是集成在于用户设备上的功能模块。例如，对话处理设备可以是用户设备上的应用。可以理解的是，对话处理设备可以是端侧的方案时，对话处理设备的硬件即为用户设备对应的硬件。
以下实施例提供一种用户设备的方法。该用户设备可以如图3A和图3B所述。
图16示出了本申请实施例的用户设备的对话处理方法的第五种实施方式的方法流程图。图16示出的对话处理方法500包括如下步骤：
步骤S501，接收用户输入的对话信息。
在一种可选的实施方式中，用户设备可以通过人机对话应用接收用户的对话信息，该对话信息中指示了用户想要预定的服务。人机对话应用可以是语音助手，也可以是其他安装的可以实现对话功能的应用。对话信息可以是语音信息，有可以是文字、手势等其他信息。
步骤S502，若所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息，从本地应用或者服务器获取第一槽位类型对应的服务标识集。
在一种可实现方式中，若用户输入的对话信息中包含的第一槽位类型对应的槽位信息与当前能提供服务的应用的服务标识集不匹配，例如用户想要预定滴滴打车应用的顺风车服务，但是当前滴滴打车应用不提供顺风车服务了，那么人机对话应用可以从本地应用中查询第一槽位类型对应的服务标识集，供用户选择。该本地应用是能提供用户想 预订服务的应用。该本地应用可以是第三方应用，如滴滴打车应用，不同于人机对话应用。
在另一种可实现方式中，若用户输入的对话信息中包含的第一槽位类型对应的槽位信息与当前能提供服务的应用的服务标识集不匹配，人机对话应用也可以从服务器查询第一槽位类型对应的服务标识集，供用户选择。该服务器可以是本地应用对应的服务器，也可以是人机对话应用对应的服务器。本申请不做限定。
类似的，若用户输入的对话信息中不包含的第一槽位类型对应的槽位信息，例如用户想要预定车出行，但是没有车辆类型的信息，而车辆类型为预定车的必要信息，那么人机对话应用可以从本地应用中或从服务器查询第一槽位类型对应的服务标识集，供用户选择。
步骤S503，提示服务标识集中的服务标识。
可选的，用户设备可以通过语音的方式提示用户，例如通过语音助手的语音播报；用户设备也可以通过文字提示用户，或通过多种方式结合提示用户，等。本申请不做限定
步骤S504，若用户从服务标识集中确定的目标服务标识，向本地应用或者服务器请求目标服务标识对应的服务。
与步骤S502的描述类似的，本步骤中，用户设备可以向本地应用，例如滴滴打车应用，请求服务，也可以直接向服务器请求服务。用户设备向服务器请求服务，可以是向人机对话应用对应的服务器发送请求，也可以是直接向本地应用对应的服务器请求。本申请此处不做限定。
步骤S505，提示服务执行成功的信息。
例如，用户如果是想预订车，此时可以提示用户车辆预订成功，等。
需要说明的是，上述各示例性的对话处理方法均基于一个总的技术构思，基于此，各示例性对话处理方法相同或者相似的部分，本申请不再赘述，各示例性对话处理方法的执行过程以及相应用户界面呈现的提示效果，可相互借鉴。
综上，本申请的技术方案，在服务器自定义的服务项目与相应槽位类型的槽位信息不对应的情况下，对话处理设备将服务器维护的服务标识反馈给用户，从而使得用户输入的相应槽位信息一定对应服务器中某项服务项目。采用本实现方式，不但能够及时向用户展示服务项目，而且能够避免用户与对话处理设备发起多轮对话，提高执行服务的效率，进而，提高用户的使用体验。
上述本申请提供的实施例中，分别从各个设备本身、以及从各个设备之间交互的角度对本申请实施例提供的对话处理方法的各方案进行了介绍。例如上述对话处理设备、用户设备和服务器，均包含了执行各个功能相应的硬件结构和/或软件模块。本领域技术人员应该很容易意识到，结合本文中所公开的实施例描述的各示例的模块及算法步骤，本申请能够以硬件或硬件和计算机软件的结合形式来实现。某个功能究竟以硬件还是计算机软件驱动硬件的方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范围。
具体实现中，对应上述各设备，本申请实施例还提供一种计算机存储介质，其中，设置在任意设备中计算机存储介质可存储有程序，该程序执行时，可实施包括图4至图 15示意的各实施例中的部分或全部步骤。所述存储介质可为磁碟、光盘、只读存储记忆体(read-only memory，简称：ROM)或随机存储记忆体(random access memory，简称：RAM)等。
本领域技术任何还可以了解到本申请实施例列出的各种说明性逻辑块(illustrative logical block)和步骤(step)可以通过电子硬件、电脑软件，或两者的结合进行实现。这样的功能是通过硬件还是软件来实现取决于特定的应用和整个系统的设计要求。本领域技术人员可以对于每种特定的应用，可以使用各种方法实现所述的功能，但这种实现不应被理解为超出本申请实施例保护的范围。
应理解，在本申请实施例的各种实施例中，各过程的序号的大小并不意味着执行顺序的先后，各过程的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
在上述实施例中，可以全部或部分地通过软件、硬件、固件或者其任意组合来实现。当使用软件实现时，可以全部或部分地以计算机程序产品的形式实现。所述计算机程序产品包括一个或多个计算机指令。在计算机上加载和执行所述计算机程序指令时，全部或部分地产生按照本申请实施例所述的流程或功能。所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。所述计算机指令可以存储在计算机可读存储介质中，或者从一个计算机可读存储介质向另一个计算机可读存储介质传输，例如，所述计算机指令可以从一个网站站点、计算机、服务器或数据中心通过有线(例如同轴电缆、光纤、数字用户线(DSL))或无线(例如红外、无线、微波等)方式向另一个网站站点、计算机、服务器或数据中心进行传输。所述计算机可读存储介质可以是计算机能够存取的任何可用介质或者是包含一个或多个可用介质集成的服务器、数据中心等数据存储设备。所述可用介质可以是磁性介质，(例如，软盘、硬盘、磁带)、光介质(例如，DVD)、或者半导体介质(例如固态硬盘Solid State Disk(SSD))等。
尽管已描述了本申请实施例的优选实施例，但本领域内的技术人员一旦得知了基本创造性概念，则可对这些实施例作出另外的变更和修改。所以，所附权利要求意欲解释为包括优选实施例以及落入本申请实施例范围的所有变更和修改。
显然，本领域的技术人员可以对本申请实施例进行各种改动和变型而不脱离本申请实施例的精神和范围。这样，倘若本申请实施例的这些修改和变型属于本申请实施例权利要求及其等同技术的范围之内，则本申请实施例也意图包含这些改动和变型在内。

Claims (30)

  1. 一种对话处理方法，其特征在于，所述方法包括：
    对话处理设备从用户设备接收对话信息；
    若所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息，所述对话处理设备从服务器获取所述第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述第一槽位类型对应的多个服务标识；
    所述对话处理设备将所述服务标识集发送到所述用户设备；
    若从所述用户设备接收到目标服务标识，所述对话处理设备向所述服务器请求所述目标服务标识对应的服务，所述目标服务标识属于所述服务标识集；
    若从所述服务器接收服务执行成功的信息，所述对话处理设备将所述执行成功的信息发送到所述用户设备。
  2. 如权利要求1所述的对话处理方法，其特征在于，对话处理设备从用户设备接收对话信息之后，从服务器获取所述第一槽位类型对应的服务标识集之前，包括：
    若从所述对话信息中未解析出所述第一槽位类型对应的槽位信息，所述对话处理设备确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
  3. 如权利要求1所述的对话处理方法，其特征在于，对话处理设备从用户设备接收对话信息之后，从服务器获取所述第一槽位类型对应的服务标识集之前，包括：
    所述对话处理设备解析所述对话信息得到所述第一槽位类型对应的槽位信息；
    所述对话处理设备向所述服务器请求所述槽位信息对应的服务；
    若从所述服务器接收到服务执行失败的信息，所述对话处理设备确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
  4. 如权利要求1所述的对话处理方法，其特征在于，对话处理设备从用户设备接收对话信息之后，将所述执行成功的信息发送到所述用户设备之前，还包括：
    所述对话处理设备解析所述对话信息得到所述第一槽位类型对应的槽位信息；
    所述对话处理设备向所述服务器请求所述槽位信息对应的服务；
    若从所述服务器接收到所述服务执行成功的信息，且所述对话处理设备中不包含所述槽位信息，所述对话处理设备将所述槽位信息添加到所述第一槽位类型对应的信息库中。
  5. 如权利要求1至4中任一项所述的对话处理方法，其特征在于，对话处理设备从用户设备接收对话信息之前，包括：
    所述对话处理设备接收添加标签指令，所述添加标签指令与所述第一槽位类型对应；
    所述对话处理设备为所述第一槽位类型添加标签。
  6. 如权利要求1至4中任一项所述的对话处理方法，其特征在于，
    所述对话信息是文本或者音频。
  7. 一种对话处理方法，其特征在于，所述方法包括：
    用户设备向对话处理设备发送对话信息；
    所述用户设备从所述对话处理设备接收并提示第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述第一槽位类型对应的多个服务标识；
    所述用户设备向所述对话处理设备发送目标服务标识，所述目标服务标识属于所述服务标识集；
    所述用户设备从所述对话处理设备接收服务执行成功的信息，所述服务是所述目标服务标识对应的服务。
  8. 如权利要求7所述的对话处理方法，其特征在于，所述用户设备提示第一槽位类型对应的服务标识集，包括：
    所述用户设备提示所述服务标识集中的全部服务标识。
  9. 如权利要求7所述的对话处理方法，其特征在于，所述用户设备提示第一槽位类型对应的服务标识集，包括：
    所述用户设备提示链接所述服务标识集的入口标识；
    所述用户设备接收用户输入的打开指令；
    所述用户设备提示所述入口标识链接的页面，所述页面中包括所述服务标识集中的全部服务标识。
  10. 如权利要求7所述的对话处理方法，其特征在于，所述用户设备提示第一槽位类型对应的服务标识集，包括：
    所述用户设备提示菜单控件，所述菜单控件中提示所述服务标识集中的部分服务标识；
    所述用户设备接收用户输入的提示指令；
    所述用户设备切换所述菜单控件中所提示的服务标识。
  11. 一种对话处理设备，其特征在于，包括处理器，收发器，所述处理器用于与存储器耦合，读取存储器中的指令并根据所述指令执行操作，其中，
    所述收发器，用于从用户设备接收对话信息；
    所述处理器，用于在所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息时，从服务器获取所述第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述第一槽位类型对应的多个服务标识；
    所述收发器，还用于将所述服务标识集发送到所述用户设备；
    所述处理器，还用于若从所述用户设备接收到目标服务标识，向所述服务器请求所述目标服务标识对应的服务，所述目标服务标识属于所述服务标识集；
    所述收发器，还用于从所述服务器接收服务执行成功的信息，并将所述执行成功的信息发送到所述用户设备。
  12. 如权利要求11所述的对话处理设备，其特征在于，
    所述处理器，还用于解析所述对话信息，并响应从所述对话信息中未解析出所述第一槽位类型对应的槽位信息，确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
  13. 如权利要求11所述的对话处理设备，其特征在于，
    所述处理器，还用于解析所述对话信息得到所述第一槽位类型对应的槽位信息，并向所述服务器请求所述槽位信息对应的服务；还用于响应从所述服务器接收到服务执行失败的信息，确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
  14. 如权利要求11所述的对话处理设备，其特征在于，
    所述处理器，还用于响应从所述服务器接收到所述服务执行成功的信息，且所述对话处理设备中不包含所述槽位信息，将所述槽位信息添加到所述第一槽位类型对应的信息库中。
  15. 如权利要求11所述的对话处理设备，其特征在于，
    所述收发器，还用于接收添加标签指令，所述添加标签指令与所述第一槽位类型对应；
    所述处理器，还用于为所述第一槽位类型添加标签。
  16. 一种用户设备，其特征在于，包括一个或多个处理器；
    存储器；
    多个应用程序；
    以及一个或多个计算机程序，其中所述一个或多个计算机程序被存储在所述存储器中，所述一个或多个计算机程序包括指令，当所述指令被所述用户设备执行时，使得所述用户设备执行以下步骤：
    向对话处理设备发送对话信息；
    从所述对话处理设备接收并提示第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述第一槽位类型对应的多个服务标识；
    向所述对话处理设备发送目标服务标识，所述目标服务标识属于所述服务标识集；
    从所述对话处理设备接收服务执行成功的信息，所述服务是所述目标服务标识对应的服务。
  17. 如权利要求16所述的用户设备，其特征在于，
    所述用户设备，还用于提示所述服务标识集中的全部服务标识。
  18. 如权利要求16所述的用户设备，其特征在于，
    所述用户设备，还用于提示链接所述服务标识集的入口标识；接收用户输入的打开指令；并提示所述入口标识链接的页面，所述页面中包括所述服务标识集中的全部服务标识。
  19. 如权利要求16所述的用户设备，其特征在于，
    所述用户设备，还用于提示菜单控件，所述菜单控件中提示所述服务标识集中的部分服务标识；接收用户输入的提示指令；并切换所述菜单控件中所提示的服务标识。
  20. 一种人机对话系统，其特征在于，包括服务器，如权利要求11至15中任一项所述的对话处理设备，以及如权利要求16至19中任一项所述的用户设备，其中，所述对话处理设备分别与所述用户设备和所述服务器进行信息交互。
  21. 一种对话处理方法，其特征在于，所述方法包括：
    接收用户输入的对话信息；
    若所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息，从本地应 用或者服务器获取所述第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述第一槽位类型对应的多个服务标识；
    提示所述服务标识集中的服务标识；
    若所述用户从所述服务标识集中确定的目标服务标识，向所述本地应用或者所述服务器请求所述目标服务标识对应的服务；
    提示所述服务执行成功的信息。
  22. 如权利要求21所述的对话处理方法，其特征在于，在接收用户输入的对话信息之后，在从本地应用或者服务器获取所述第一槽位类型对应的服务标识集之前，包括：
    若从所述对话信息中未解析出所述第一槽位类型对应的槽位信息，确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
  23. 如权利要求21所述的对话处理方法，其特征在于，在接收用户输入的对话信息之后，在从本地应用或者服务器获取所述第一槽位类型对应的服务标识集之前，包括：
    解析所述对话信息得到所述第一槽位类型对应的槽位信息；
    向所述本地应用或者所述服务器请求所述槽位信息对应的服务；
    若从所述本地应用和所述服务器均请求服务失败，确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
  24. 如权利要求21所述的对话处理方法，其特征在于，所述从本地应用或者服务器获取所述第一槽位类型对应的服务标识集，包括：
    从所述本地应用获取所述第一槽位类型对应的服务标识集；或者，
    从所述服务器获取所述第一槽位类型对应的服务标识集；或者，
    调用所述本地应用从所述服务器获取所述第一槽位类型对应的服务标识集。
  25. 如权利要求21所述的对话处理方法，其特征在于，提示所述服务标识集中的服务标识包括：
    提示所述服务标识集中的全部服务标识；或者，
    提示链接所述服务标识集的入口标识；
    若用户输入的打开指令，提示所述入口标识链接的页面，所述页面中包括所述服务标识集中的全部服务标识；或者，
    提示菜单控件，所述菜单控件中提示所述服务标识集中的部分服务标识；
    若用户输入的提示指令，切换所述菜单控件中所提示的服务标识。
  26. 一种用户设备，其特征在于，包括一个或多个处理器；
    存储器；
    多个应用程序；
    以及一个或多个计算机程序，其中所述一个或多个计算机程序被存储在所述存储器中，所述一个或多个计算机程序包括指令，当所述指令被所述用户设备执行时，使得所述用户设备执行以下步骤：
    接收用户输入的对话信息；
    若所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息，从本地应用或者服务器获取所述第一槽位类型对应的服务标识集，所述第一槽位类型指示必选槽 位信息中用于指示服务标识的槽位信息所属的槽位类型，所述服务标识集包括所述第一槽位类型对应的多个服务标识；
    提示所述服务标识集中的服务标识；
    若所述用户从所述服务标识集中确定的目标服务标识，向所述本地应用或者所述服务器请求所述目标服务标识对应的服务；
    提示所述服务执行成功的信息。
  27. 如权利要求26所述的用户设备，其特征在于，
    所述用户设备，还用于执行若从所述对话信息中未解析出所述第一槽位类型对应的槽位信息，确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
  28. 如权利要求26所述的用户设备，其特征在于，
    所述用户设备，还用于执行解析所述对话信息得到所述第一槽位类型对应的槽位信息；向所述本地应用或者所述服务器请求所述槽位信息对应的服务；若从所述本地应用和所述服务器均请求服务失败，确定所述对话信息中不包含能够确定服务的第一槽位类型对应的槽位信息。
  29. 如权利要求26所述的用户设备，其特征在于，
    所述用户设备，还用于执行从所述本地应用获取所述第一槽位类型对应的服务标识集；或者，
    从所述服务器获取所述第一槽位类型对应的服务标识集；或者，
    调用所述本地应用从所述服务器获取所述第一槽位类型对应的服务标识集。
  30. 如权利要求26所述的用户设备，其特征在于，
    所述用户设备，还用于执行提示所述服务标识集中的全部服务标识；还用于执行提示链接所述服务标识集的入口标识；若用户输入的打开指令，提示所述入口标识链接的页面，所述页面中包括所述服务标识集中的全部服务标识；还用于执行提示菜单控件，所述菜单控件中提示所述服务标识集中的部分服务标识；若用户输入的提示指令，切换所述菜单控件中所提示的服务标识。
PCT/CN2019/124441 2019-01-10 2019-12-11 对话处理方法及设备 Ceased WO2020143381A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP19909437.6A EP3889799A4 (en) 2019-01-10 2019-12-11 DIALOG PROCESSING METHOD AND DEVICE
US17/371,377 US11977815B2 (en) 2019-01-10 2021-07-09 Dialogue processing method and device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910025807.3 2019-01-10
CN201910025807.3A CN111427992B (zh) 2019-01-10 2019-01-10 对话处理方法及设备

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/371,377 Continuation US11977815B2 (en) 2019-01-10 2021-07-09 Dialogue processing method and device

Publications (1)

Publication Number Publication Date
WO2020143381A1 true WO2020143381A1 (zh) 2020-07-16

Family

ID=71520916

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/124441 Ceased WO2020143381A1 (zh) 2019-01-10 2019-12-11 对话处理方法及设备

Country Status (4)

Country Link
US (1) US11977815B2 (zh)
EP (1) EP3889799A4 (zh)
CN (1) CN111427992B (zh)
WO (1) WO2020143381A1 (zh)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112201250B (zh) * 2020-09-30 2024-03-19 中移（杭州）信息技术有限公司 一种语义分析方法、装置、电子设备及存储介质
CN113590750B (zh) 2021-07-30 2024-09-13 北京小米移动软件有限公司 人机对话方法、装置、电子设备及存储介质
CN114117009B (zh) * 2021-11-30 2025-05-30 深圳壹账通智能科技有限公司 基于对话机器人的子流程配置方法、装置、设备及介质
US12231377B1 (en) * 2022-06-22 2025-02-18 Amazon Technologies, Inc. Chatbot conversation flow designer

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170177715A1 (en) * 2015-12-21 2017-06-22 Adobe Systems Incorporated Natural Language System Question Classifier, Semantic Representations, and Logical Form Templates
CN107665704A (zh) * 2016-07-29 2018-02-06 科大讯飞股份有限公司 语音指令检测模型构建方法、检测方法及系统、人机交互方法及设备
CN108920497A (zh) * 2018-05-23 2018-11-30 北京奇艺世纪科技有限公司 一种人机交互方法及装置
CN109033223A (zh) * 2018-06-29 2018-12-18 北京百度网讯科技有限公司 用于跨类型对话的方法、装置、设备以及计算机可读存储介质
CN109063035A (zh) * 2018-07-16 2018-12-21 哈尔滨工业大学 一种面向出行领域的人机多轮对话方法
CN109101545A (zh) * 2018-06-29 2018-12-28 北京百度网讯科技有限公司 基于人机交互的自然语言处理方法、装置、设备和介质

Family Cites Families (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8073681B2 (en) * 2006-10-16 2011-12-06 Voicebox Technologies, Inc. System and method for a cooperative conversational voice user interface
US9466294B1 (en) * 2013-05-21 2016-10-11 Amazon Technologies, Inc. Dialog management system
US9508339B2 (en) * 2015-01-30 2016-11-29 Microsoft Technology Licensing, Llc Updating language understanding classifier models for a digital personal assistant based on crowd-sourcing
US10460227B2 (en) * 2015-05-15 2019-10-29 Apple Inc. Virtual assistant in a communication session
US9996531B1 (en) * 2016-03-29 2018-06-12 Facebook, Inc. Conversational understanding
CN107347109B (zh) * 2016-05-06 2020-06-30 腾讯科技（深圳）有限公司 信息交互系统、方法和装置
US20180114528A1 (en) * 2016-10-26 2018-04-26 IPsoft Incorporated Systems and methods for generic flexible dialogue management
DK201770432A1 (en) * 2017-05-15 2018-12-21 Apple Inc. Hierarchical belief states for digital assistants
US11172063B2 (en) * 2017-05-22 2021-11-09 Genesys Telecommunications Laboratories, Inc. System and method for extracting domain model for dynamic dialog control
CN107886948A (zh) * 2017-11-16 2018-04-06 百度在线网络技术（北京）有限公司 语音交互方法及装置，终端，服务器及可读存储介质
CN108874774B (zh) * 2018-06-05 2022-02-11 浪潮软件股份有限公司 一种基于意图理解的服务调用方法和系统
CN108932946B (zh) * 2018-06-29 2020-03-13 百度在线网络技术（北京）有限公司 客需服务的语音交互方法和装置
CN109002501A (zh) * 2018-06-29 2018-12-14 北京百度网讯科技有限公司 用于处理自然语言对话的方法、装置、电子设备以及计算机可读存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20170177715A1 (en) * 2015-12-21 2017-06-22 Adobe Systems Incorporated Natural Language System Question Classifier, Semantic Representations, and Logical Form Templates
CN107665704A (zh) * 2016-07-29 2018-02-06 科大讯飞股份有限公司 语音指令检测模型构建方法、检测方法及系统、人机交互方法及设备
CN108920497A (zh) * 2018-05-23 2018-11-30 北京奇艺世纪科技有限公司 一种人机交互方法及装置
CN109033223A (zh) * 2018-06-29 2018-12-18 北京百度网讯科技有限公司 用于跨类型对话的方法、装置、设备以及计算机可读存储介质
CN109101545A (zh) * 2018-06-29 2018-12-28 北京百度网讯科技有限公司 基于人机交互的自然语言处理方法、装置、设备和介质
CN109063035A (zh) * 2018-07-16 2018-12-21 哈尔滨工业大学 一种面向出行领域的人机多轮对话方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3889799A4

Also Published As

Publication number Publication date
US11977815B2 (en) 2024-05-07
CN111427992B (zh) 2024-10-15
EP3889799A4 (en) 2022-02-09
CN111427992A (zh) 2020-07-17
US20210334071A1 (en) 2021-10-28
EP3889799A1 (en) 2021-10-06

Similar Documents

Publication Publication Date Title
US12094208B2 (en) Video classification method, electronic device and storage medium
US11977815B2 (en) Dialogue processing method and device
US20220214894A1 (en) Command execution method, apparatus, and device
US11934394B2 (en) Data query method supporting natural language, open platform, and user terminal
WO2022134110A1 (zh) 一种语音理解方法及装置
US12367352B2 (en) Deep learning techniques for extraction of embedded data from documents
US20160170710A1 (en) Method and apparatus for processing voice input
CN108701128A (zh) 解释和解析条件自然语言查询
US10749986B2 (en) Platform for interaction via commands and entities
US20170249934A1 (en) Electronic device and method for operating the same
CN110018858A (zh) 一种基于语音控制的应用管理方法、装置
WO2023064575A1 (en) Systems and methods to generate contextual based actions
US20240265211A1 (en) Programatically invokable conversational chatbot
WO2023165225A9 (zh) 多任务场景下的悬浮球处理方法和电子设备
WO2022161132A1 (zh) 语音播报方法及装置
CN114465975B (zh) 一种内容推送方法、装置、存储介质和芯片系统
US20210109960A1 (en) Electronic apparatus and controlling method thereof
WO2026081707A1 (zh) 一种基于民航大语言模型的智能系统
WO2018076616A1 (zh) 信息交互方法、服务器、客户端、存储介质及设备
CN118251668A (zh) 用于从数据中提取问题答案对的基于规则的技术
WO2024078419A1 (zh) 语音交互方法、语音交互装置和电子设备
US20160048498A1 (en) Method for providing alternative service and electronic device thereof
TWI915279B (zh) 需求資訊處理方法及系統與電腦程式產品
US20250118307A1 (en) Speech control method, apparatus, and device
KR102004968B1 (ko) 메신저 서비스의 첨부파일 관리 방법, 이를 위한 시스템, 이를 위한 단말기 및 이를 위한 서비스장치

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19909437

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2019909437

Country of ref document: EP

Effective date: 20210630

NENP Non-entry into the national phase

Ref country code: DE