CN111539219A

CN111539219A - Method, equipment and system for disambiguating natural language content title

Info

Publication number: CN111539219A
Application number: CN202010325483.8A
Authority: CN
Inventors: 袁志伟; 戴帅湘
Original assignee: Beijing Moran Cognitive Technology Co Ltd
Current assignee: Hangzhou suddenly Cognitive Technology Co.,Ltd.
Priority date: 2017-05-19
Filing date: 2017-05-19
Publication date: 2020-08-14
Also published as: CN111539204A; CN107193810A; CN111539217B; CN111539218A; CN111539215A; CN111538812A; CN107193810B; CN111539217A; CN111539202A; CN111539216A; CN111538811A; CN111539203A; CN111539214A; CN111539201A

Abstract

The invention aims to provide a method, equipment and a system for disambiguating natural language content titles. Specifically, the actual semantic item information corresponding to the content title is obtained by determining the natural language command of the user from the multiple semantic item information after conversion according to the equipment type information of the output equipment, and the content corresponding to the content title is output according to the actual semantic item information, so that when the meaning expressed by the natural language of the user is converted into a clear expression but the content indicated by the meaning is not determined, the true meaning of the user can be understood, the accuracy of user graph identification is improved, and the user voice interaction experience is also improved.

Description

Method, equipment and system for disambiguating natural language content title

The scheme is a divisional application with a parent application number of 201710357079.7 and application dates of 2017-05-19, and is named as 'a method, equipment and system for disambiguating natural language content titles'.

Technical Field

The present invention relates to the field of speech recognition technology, and more particularly, to a method, apparatus, and system for natural speech content title disambiguation.

Background

The interaction between the user and the device comprises key control, remote control, somatosensory control, touch control and the like. With the development of NLP (natural language processing) technology, an interaction mode, namely voice interaction control, is added between a user and a device.

The user uses natural language to interact with the device, and it is expected that the device can maximally understand the user's intention, but when a human being expresses through natural language, words with uncertain meanings such as omitted words and pronouns often appear, and the existing speech recognition technology generally utilizes context to determine the specific meanings of the omitted words and the pronouns in the user's language expression. However, when the user interacts with the device using the natural language, there is also a case that the meaning of the user's natural language expression is determined after being converted into a character, and the character expression of the user's natural language after being recognized by the voice is unique, and does not include an omitted word or a representative word. There are situations where the word expression is unambiguous and the content to which it refers is uncertain. At the moment, the equipment for receiving the natural language input of the user cannot understand the real meaning of the user, so that the accuracy of the intention identification of the user is reduced, and the voice interaction experience of the user is influenced.

Disclosure of Invention

The invention provides a method applied to a natural language content title disambiguation system, wherein the system comprises a first device, other devices at least comprising a second device, a cloud and a network, and the method comprises the following steps:

a user sends a natural language command;

the first equipment or the second equipment determines whether an object of a user for sending a natural language command is the first equipment or not;

after the first device or the second device determines that an object of a user for sending a natural language command is self, the natural language command is sent to the cloud;

the cloud receives the natural language command and converts the natural language command into a content title, wherein the content title is uniquely determined in literal representation, but the content represented by the content title is not uniquely determined;

the cloud determines the content specifically represented by the content title, wherein the cloud combines the device type of the first device or the second device when determining the content specifically represented by the content title;

the cloud returns the content specifically represented by the content title to the first device or the second device after determining the content specifically represented by the content title;

and the first device or the second device outputs the content specifically represented by the determined content title after receiving the content specifically represented by the determined content title returned by the cloud.

Optionally, the cloud receives the device type tag from the first device or the second device.

Optionally, the cloud obtains the device type tag of the first device or the second device locally.

Optionally, the first device or the second device determines whether the object of the natural language command sent by the user is itself by one or more of the following ways:

according to the distance between the user and the equipment;

detecting a user natural language command direction according to the audio;

the front face orientation of the user is detected from the image.

The present invention provides a method for natural language content title disambiguation, the method comprising performing at a device the steps of:

determining whether an object of a user sending a natural language command is the object;

the method comprises the steps of determining that an object of a natural language command sent by a user is the natural language command, and then receiving the natural language command input by the user;

the natural language command is sent to the cloud,

enabling the cloud to convert the natural language command into a content title after receiving the natural language command, wherein the content title is uniquely determined in the literal expression, but the content represented by the content title is not uniquely determined; the cloud determines the content specifically represented by the content title, and the cloud combines the device type of the device when determining the content specifically represented by the content title;

receiving content specifically represented by the determined content title returned by the cloud;

and outputting the content specifically represented by the determined content title.

The present invention provides a computer medium having stored thereon a computer program, the program being executable by an apparatus to cause the apparatus to:

the natural language command is sent to the cloud,

The invention provides a device for natural language content title disambiguation, the device comprising a processor, a memory and a program stored on the memory and executable by the processor, the processor executing the program to implement the steps of:

the natural language command is sent to the cloud,

The present invention provides an output device for natural language content title disambiguation, the output device comprising:

a module for determining whether an object from which a user sends a natural language command is itself;

a module for receiving a natural language command input by a user after determining that an object of the natural language command sent by the user is the user;

means for sending a natural language command to the cloud,

means for receiving content returned by the cloud that is specifically represented by the determined content title;

means for outputting content specifically represented by the determined content title.

The present invention provides an apparatus for natural language content title disambiguation, the apparatus comprising an information processing device comprising:

means for sending a natural language command to the cloud,

The present invention provides an apparatus for natural language content title disambiguation, the apparatus comprising:

a detection unit configured to determine whether an object to which a user transmits a natural language command is itself;

a sound pickup unit configured to receive a natural language command input by a user after determining that an object to which the user transmits the natural language command is itself;

a transmitting unit transmitting the natural language command to the cloud,

a receiving unit configured to receive content specifically represented by the determined content title returned by the cloud;

an output unit configured to output the content specifically represented by the determined content title.

The present invention provides a method for natural language content title disambiguation, the method performing the following steps at the cloud:

receiving a natural language command sent by equipment;

converting the natural language command into a content title, wherein the content title is uniquely determined in literal representation but the content represented by the content title is not uniquely determined;

determining content specifically represented by a content title, wherein the device type of the device is combined when determining the content specifically represented by the content title;

the content specifically represented by the content title is returned to the device after the content specifically represented by the content title is determined.

The present invention provides a computer medium having stored thereon a computer program for execution by a cloud to cause the cloud to:

receiving a natural language command sent by equipment;

The invention provides a cloud for natural language content title disambiguation, the cloud comprising a processor, a memory and a program stored on the memory and executable by the processor, the processor executing the program to implement the steps of:

receiving a natural language command sent by equipment;

The present invention provides a cloud for natural language content title disambiguation, the cloud comprising:

a module for receiving a natural language command sent by a device;

a module for converting a natural language command into a content title, wherein the content title is uniquely identified in textual representation, but does not represent a content that is uniquely identified;

means for determining content specifically represented by a content title, wherein a device type of the device is incorporated in determining the content specifically represented by the content title;

means for returning the content specifically represented by the content title to the device after determining the content specifically represented by the content title.

The present invention provides a cloud for natural language content title disambiguation, the cloud comprising an information processing apparatus comprising:

a module for receiving a natural language command sent by a device;

the receiving unit is used for receiving the natural language command sent by the equipment;

a first processing unit coupled to the receiving unit, the first processing unit configured to convert the natural language command into a content title, wherein the content title is uniquely determined in textual representation but does not represent content that is uniquely determined;

a second processing unit coupled to the first processing unit, the second processing unit configured to determine content specifically represented by a content title, wherein a device type of the device is incorporated in determining the content specifically represented by the content title;

a sending unit configured to return the content specifically represented by the content title to the device after determining the content specifically represented by the content title.

Alternatively, the first processing unit and the second processing unit may be combined into one processing unit.

The invention provides a natural language content title disambiguation system comprising a cloud as described, a device as described and a network as described.

Compared with the prior art, according to the embodiment of the invention, the actual semantic item information corresponding to the content title is obtained by determining the natural language command of the user from the multiple semantic item information after conversion according to the equipment type information of the output equipment, and the content corresponding to the content title is output according to the actual semantic item information, so that when the meaning expressed by the natural language of the user is converted into the content which is expressed clearly but is not determined, the true meaning of the user can be understood, the accuracy of the intention identification of the user is improved, and the voice interaction experience of the user is also improved.

Drawings

Other features, objects and advantages of the invention will become more apparent upon reading of the detailed description of non-limiting embodiments made with reference to the following drawings:

FIG. 1 shows a block diagram of a system for natural language content title disambiguation according to an embodiment of the invention;

FIG. 2 illustrates a method flow for natural language content title disambiguation according to an embodiment of the invention;

FIG. 3 illustrates a first embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 4 shows a second embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 5 illustrates a third embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 6 shows a fourth embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 7 shows a fifth embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 8 shows a sixth embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 9 shows a seventh embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 10 shows an eighth embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 11 shows a ninth embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 12 shows a tenth embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 13 shows an eleventh embodiment for natural language content title disambiguation according to embodiments of the invention;

FIG. 14 shows a twelfth embodiment for natural language content title disambiguation according to embodiments of the invention.

The same or similar reference numbers in the drawings identify the same or similar elements.

Detailed Description

The present invention is described in further detail below with reference to the attached drawing figures.

In the following description, since the description is provided to describe many technical details to assist understanding of the present invention, the present invention can be implemented without these technical details, and the technical solution of the present invention is not limited to the scenarios and environments described in the description.

Description of the problem

When a user sends a control command to a device using natural language, it is desirable that the natural language command more closely resemble the natural language expression, while also expecting that the device can really understand the intent of its control command. For example, when a user wishes to play the music of zhou jilun "secret not to say", the conventional practice is to not only need to explicitly play the content "secret not to say", but also to explicitly play the category "music", that is, at least to establish a template "i want to listen to … …" or "please play music … …", and to express the template in the natural language input, that is, the natural language input "secret that i want to listen to cannot say", or "please play the secret that music cannot say", so as to enable the receiving device to understand the true intention of the user. If the user only inputs 'please play secret which can not be said', the device can not immediately determine whether the user requests music 'secret which can not be said' or the movie 'secret which can not be said', in the prior art, some users can determine the intention of the user request according to the context information, and some users can select to determine the true intention of the request by prompting. However, these existing schemes have a possibility of erroneously recognizing the user's intention, or have problems in that the recognition response time is excessively long and the user's use burden is increased.

The above problem scenario may exist in a residential or vehicular environment. Furthermore, in an office environment, there may be similar problem scenarios as well. For example, when a user inputs a "first AI congress" to the device in a meeting room through a natural language, and a file corresponding to the "first AI congress" includes video, audio and text or a web page, the device cannot determine whether the user's real intention is to play the video "first AI congress", play the audio "first AI congress" or play text or a web page related to the "first AI congress". The device may determine the user's intent through contextual information or may prompt the user to choose to determine the true intent of their request. However, also, these existing solutions have a possibility of either misidentifying the user's intention or have problems in that the recognition response time is too long and the user's use burden is increased.

The above user natural language input contents, scenes, environments, etc. are only for illustrative purposes and are not limited thereto. For better understanding of the present invention, the following embodiments will refer to some input contents, scenarios and environments in the description process, however, the understanding of the technical solution of the present invention should not be limited thereto, and all possible input contents, scenarios and environments are within the scope of the present invention.

General overview

When a user sends a command by using a natural language, the natural language command is converted into a content title through a voice recognition technology, the content title has a plurality of actual meaning item information, namely, the content title is uniquely determined in literal representation, but the content represented by the content title is not uniquely determined, in particular different contents with the same title are referred, the different contents are particularly different contents with different output forms, and the device type can be combined when the content specifically pointed by the content title is determined. The device is an output device, and is configured to output content corresponding to a content title, where the content corresponding to the content title is adapted to the actual meaning item information corresponding to the content title, that is, the output device is configured to output content to which the content title specifically points. In some embodiments, the output device is also a device that receives user natural language commands; in some embodiments, the output device is a different device than the device that receives the user natural language command. In some embodiments, the output device may be a device having only output capabilities. In some embodiments, the output device may be a device that, in addition to having an output capability, may also have a capability of acquiring content corresponding to the content title according to the actual semantic item information corresponding to the content title.

In some embodiments, when the output device has a sound pickup function, the output device may simultaneously be the device that receives the user's natural language commands; when the output device does not have the sound pickup function, the output device is different from the device for receiving the natural language command of the user, and at the moment, the output device can be controlled by other control devices with the sound pickup function; in some embodiments, when the output device has a function of picking up sound, or a part of the output device has a function of picking up sound, the output device may be a different device from the device receiving the user's natural language command, and in this case, the output device may be controlled by another control device having a function of picking up sound; in some embodiments, when part of the output devices have a sound pickup function, the output device having a sound pickup function is the same device as the device receiving the user's natural language command, the output device not having a sound pickup function is a different device from the device receiving the user's natural language command, and the output device not having a sound pickup function is controllable by the other control device having a sound pickup function.

In some embodiments, if the device type of the output device is an audio playing device, the content pointed by the content title is preferentially determined to be audio content related to the content title; and if the device type of the output device is the video playing device, preferentially determining the content pointed by the content title as the video content related to the content title.

In some embodiments, if the output device includes an audio playing device and a video playing device, and the device type of the output device is the video playing device, the content pointed by the content title is preferentially determined to be video content related to the content title; if only the video playing device exists, prompting the user to select to play the audio content or the video content pointed by the content title, or preferentially determining the content pointed by the content title as the video content related to the content title, prompting the user to determine the content as the video content related to the content title, and optionally simultaneously prompting the user whether to switch to the audio content related to the content title.

In some embodiments, the control device stores the output device and its device type, and the user can set the output device and its device type information through the input device of the control device. In some embodiments, a user may activate the output devices via the input device of the control device, may activate all of the output devices, or may activate some of the output devices. The setting and starting process can be input through characters such as a keyboard, a touch screen and the like, and can also be input through natural language commands received by the sound pickup device. The sound pick-up device of the control equipment receives the natural voice input command of the user and completes the process of the content specifically pointed by the content title.

For example, the user's natural language input "secret cannot be said", which is converted into the text "secret cannot be said" after the speech conversion word processing, is uniquely determined here for the speech conversion word processing, but the "secret cannot be said" may represent music "secret cannot be said", and may represent movie "secret cannot be said", at which time, if the output device is a sound, the true intention of preferentially determining the "secret cannot be said" natural language input "of the user is music" secret cannot be said "; if the output device is a television, then it is prioritized that the true intent of the user's natural language input "secret cannot be said" is the movie "secret cannot be said"; alternatively, if the user is in an environment where both stereo and television are present and the output device is television, then it is prioritized that the true intent of the user's natural language input "secret cannot be said" is the movie "secret cannot be said"; if the user is in the environment where only television is present and the output device is television, then the true intent of the user's natural language input "secret not to speak" is preferentially to determine that the movie "secret not to speak" and prompt the user for the determination of the movie "secret not to speak" and optionally prompt the user to switch to music "secret not to speak".

In the interaction process, the sound box and the television have sound pickup functions, so that the sound box and the television are output devices and can also be used as devices for receiving natural language commands of users; alternatively, the user natural language command may be received through a control device storing a tv in which the user is in an environment including a sound for outputting an audio file and an output video file, the user inputs a natural language through the control device and the control device performs the previous process of recognizing the user's true intention, and transmits the determined contents to the output device. In this process, the user may enter the devices and device types included in the user environment in text or natural language, and may activate all or a portion of the devices included in the user environment in text or natural language input.

The above examples may be used in a residential environment, as well as in an in-vehicle environment, or any environment where similar devices are present.

For example, the user natural language input "the first AI congress" is converted into the text "the first AI congress" after the speech conversion word processing, where the processing of the speech conversion word is uniquely determined, but the "first AI congress" may represent the video "the first AI congress" and may also represent the text or web page related to the audio "the first AI congress" or "the first AI congress", and at this time, if the output device is a sound, the real intention of the user natural language input "the first AI congress" is preferentially determined to be the audio "the first AI congress"; if the output device is a stereo-connected projector or television, then it is preferably determined that the true intent of the user natural language input "first-to-end AI convention" is the video "first-to-end AI convention"; if the output equipment is a projector which is not connected with a sound box, the real intention of inputting the first terminal AI meeting by the natural language of the user is preferentially determined to be the relevant characters or the webpage of the first terminal AI meeting; optionally, if the sound and the projector exist in the environment where the user is located at the same time, and the output device is the projector, firstly determining whether the projector is connected with the sound, and if not, preferentially determining that the real intention of the user natural language input "the first AI congress" is the relevant text or webpage of the "first AI congress"; if the connection is made, the real intention of the user to input the 'first terminal AI congress' by natural language is preferentially determined to be the video 'first terminal AI congress', optionally, the user is prompted to determine the video 'first terminal AI congress', optionally, whether the user is prompted to switch to the audio 'first terminal AI congress', or the relevant words or webpage of the 'first terminal AI congress'.

In the above interaction process, both the sound box and the television have a sound pickup function, but the projector does not have a sound pickup function. Therefore, the sound and the television are output devices and can also be used as devices for receiving the natural language commands of the user, the projector is output device but not used as a device for receiving the natural language commands of the user, the projector can receive the natural language commands of the user through the control device, the control device completes the process of identifying the real intentions of the user in the front and controls and sends output contents to the projector; optionally, since the environment where the user is located includes the projector without the sound pickup function, all the natural language commands are received by the control device, the control device completes the process of identifying the real intention of the user in the front, and controls to send the output content to the output device; in some embodiments, the control device is further configured to determine an output capability that the device has.

Device type information

The device type information may be represented by any identifier that can distinguish between different device types, and in particular embodiments, the identifier may be a type of the output file or may be a device type tag. Here, the distinguishing identification of the device type may be automatically performed or may be manually set by the user. The information of the device self attribute, the user attribute, the context and the like can be combined in the distinguishing and identifying process of the device types. In the process of distinguishing and identifying the device types, the disambiguation of the device type can be completed through interaction, and meanwhile, the disambiguation effect can be continuously improved through enhancing a learning model by user behaviors in the interaction process.

Content title

The content corresponding to the content title may be content stored locally in an internal memory, may be content stored in an external memory, or may be content obtained through internet search.

System architecture

As shown in fig. 1, system 100 includes at least a first device 110, and in some embodiments, a network device 130 and a network 140; in some embodiments, the first device 110 may perform the present invention independently; in some embodiments, the first device 110 interfaces with the network device 130 via the network 140 to accomplish the present invention; in some embodiments, the system 100 further comprises at least one second device 120, and the first device 110 is connected to the second device 120 via the network 140 to implement the present invention; in some embodiments, the first device 110 and the second device 120 are coupled to the network device 130 via the network 140 to implement the present invention; in some embodiments, the system 100 further includes an intelligent device 150, the intelligent device 150 controlling the system 100 through the network 140 to accomplish the present invention.

Here, the first device 110 may be any electronic product capable of outputting media contents of one or more combinations of video, audio, pictures, text, and the like. In some embodiments, the first device 110 includes, by way of non-limiting example, a television, a stereo, a projector, a car stereo, a car display, a smart rearview mirror, and the like. In some embodiments, outputting content includes, by way of non-limiting example, speakers outputting audio and video in cooperation with a display, displays outputting text and pictures, and the like. In some embodiments, the first device 110 has sound pickup capabilities. In some embodiments, the first device 110 optionally has voice-to-text capability. In some embodiments, the first device 110 optionally includes a processor, a memory, and a program stored in the memory and executable on the processor, the processor executing the program to implement the respective functions and/or methods of the present invention; in some embodiments, the first device 110 optionally includes or is coupled to a computer-readable medium, such as a Random Access Memory (RAM) and/or a cache memory, wherein the computer-readable medium stores a program that is executed by a processor to perform corresponding functions and/or methods of the present invention. The first device 110 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. In some embodiments, the first device 110 optionally includes or is coupled to a computer-readable medium for storing other information, optionally including user-requested content; in some embodiments, the first device 110 optionally includes several modules, which may be program modules configured to perform the respective functions of the embodiments of the present invention; in some embodiments, the first device 110 comprises a controller comprising a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, is capable of implementing the respective functions and/or methods of the present invention; in some embodiments, the first device 110 optionally includes a wireless or wired network connection unit; in some embodiments, the first device 110 may implement NLP functionality, recognizing a user's natural language input; in some embodiments, the first device 110 may implement a web search function; in some embodiments, the first device 110 may include an operating system; the first device 110 may install an Application (APP) for performing a corresponding function of the present invention; the first device 110 may be activated by means of a button, a remote control, a wake-up word, or the smart device 150.

The second device 120 may be any electronic product capable of outputting media content in one or more combinations of video, audio, graphics, text, etc., but having a different media output type than the first device 110. In some embodiments, the second device 120 illustratively includes, without limitation, a television, a stereo, a projector, a car stereo, a car display, a smart rearview mirror, and the like. In some embodiments, outputting content includes, by way of non-limiting example, speakers outputting audio and video in cooperation with a display, displays outputting text and pictures, and the like. In some embodiments, the second device 120 has sound pickup capability. In some embodiments, the second device 120 optionally has voice-to-text capability. In some embodiments, the second device 120 optionally includes a processor, a memory, and a program stored on the memory and executable on the processor, the processor executing the program to implement the respective functions and/or methods of the present invention; in some embodiments, the second device 120 optionally includes or is coupled to a computer readable medium, such as a Random Access Memory (RAM) and/or a cache memory, wherein the computer readable medium stores a program that is executed by a processor to perform the respective functions and/or methods of the present invention. The second device 120 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. In some embodiments, the second device 120 optionally includes or is coupled to a computer-readable medium for storing other information, optionally including user-requested content. In some embodiments, the second device 120 optionally includes several modules, which may be program modules configured to perform the functions and/or methods corresponding to the embodiments of the present invention. In some embodiments, the second device 120 comprises a controller comprising a memory and a processor, wherein the memory stores a computer program that when executed by the processor is capable of performing the respective functions of the present invention. In some embodiments, the second device 120 optionally comprises a wireless or wired network connection unit; in some embodiments, the second device 120 may implement NLP functionality, recognizing a user's natural language input; the second device 120 may implement a network search function; in some embodiments, the second device 120 may include an operating system; the second device 120 may install an application APP for performing the corresponding functions of the present invention; the second device 120 may be activated by means of a button, a remote control, a wake-up word, or the smart device 150.

The network device 130 is connected to the first device 110 and/or the second device 120. Herein, the network device 130 includes, but is not limited to, implementations such as a network host, a single network server, a plurality of network server sets, or a cloud computing-based computer collection. Here, the Cloud is made up of a large number of hosts or web servers based on Cloud Computing (Cloud Computing), which is a type of distributed Computing, a super virtual computer consisting of a collection of loosely coupled computers. In particular embodiments, network device 130 may host multiple servers. In some embodiments, a network device may be referred to as a network device node; in some embodiments, a network device node may be a node that performs the corresponding function of the present invention, or a plurality of nodes that perform the corresponding function of the present invention together, and the plurality of nodes may be arranged in a centralized manner or in a distributed manner. In some embodiments, the network device 130 includes a processor, a memory, and a program stored on the memory and executable on the processor, the processor executing the program to implement the corresponding functions of the present invention. In some embodiments, the network device 130 optionally includes or is coupled to a computer-readable medium, such as a Random Access Memory (RAM) and/or a cache memory, wherein the computer-readable medium stores a program that is executed by a processor to perform corresponding functions and/or methods of the present invention. The network device 130 may further include other removable/non-removable, volatile/nonvolatile computer system storage media. In some embodiments, network device 130 optionally includes or is coupled to a computer-readable medium for storing other information. In some embodiments, network device 130 optionally includes several modules, which may be program modules, configured to perform the functions and/or methods corresponding to the various embodiments of the present invention. In some embodiments, the network device 130 comprises a controller comprising a memory and a processor, wherein the memory stores a computer program that when executed by the processor is capable of performing the corresponding functions of the present invention. In some embodiments, network device 130 may be implemented by one or more computing devices. In some embodiments, network device 130 is implemented by one or more computing units. In some embodiments, network device 130 may be implemented by a server; in some embodiments, the network device 130 may be implemented by a distributed server; in some embodiments, the network device 130 is disposed within the environment in which the user is located; in some embodiments, the network device 130 is provided separately; in some embodiments, the network device 130 may be disposed in the first device 110 or the second device 120; in some embodiments, the network device 130 may also be located outside of the user's environment; in some embodiments, the network device 130 may implement storage functionality, storing other information, optionally storing user-requested content; in some embodiments, network device 130 may implement NLP functionality, recognizing a user's natural language input; in some embodiments, network device 130 may implement a network search function; in some embodiments, network device 130 may implement network connection functionality; in some embodiments, network device 130 may include an operating system; the network device 130 may install an application APP for performing the corresponding functions of the present invention.

The network 140 may connect the first device 110 with the second device 120, and may also connect the network device 130 with the first device 110 and/or the second device 120. Network 140 illustratively includes, without limitation, a local area network LAN, a wide area network WAN, an Ethernet network, the Internet, any mobile communications network, a satellite network, or any other wired/wireless network; in some embodiments, network 140 may include a combination of networks; in some embodiments, the wireless connection may also be implemented by near field communication.

The smart device 150 has a sound pickup function, and may be any electronic product that can perform human-computer interaction with a user through a keyboard, a touch pad, a touch screen, or a handwriting device, voice, and the like, such as a PC, a mobile phone, a smart phone, a PDA, a wearable device, a palm PC PPC, a smart remote controller, or a tablet computer. Particularly for environments where the first device 110, the second device 120, etc. do not have a sound pickup function. In some embodiments, the smart device 150 comprises a processor, a memory, and a program stored on the memory and executable on the processor, the processor executing the program to implement the corresponding functions of the present invention; in some embodiments, the smart device 150 includes a computer readable medium storing a program for execution by a processor to perform the corresponding functions of the present invention; in some embodiments, the smart device 150 comprises a controller comprising a memory and a processor, wherein the memory stores a computer program that, when executed by the processor, enables the corresponding functions of the present invention; in some embodiments, the smart device 150 optionally includes several modules, which may be program modules, for performing the corresponding functions of the present invention; in some embodiments, the smart device 150 optionally includes an operating system; in some embodiments, the smart device 150 optionally includes a touch-sensitive display screen or keyboard; in some embodiments, the smart device 150 optionally includes a display device; in some embodiments, the smart device 150 optionally includes a keyboard, mouse; in some embodiments, the smart device 150 has installed an application APP for performing the corresponding functions of the present invention.

It should be understood by those skilled in the art that the first device 110, the second device 120, the network device 130 and the intelligent device 150 are only examples, and other existing or future first devices, second devices, network devices or intelligent devices may be suitable for the present invention, and are included in the scope of the present invention and are incorporated herein by reference. Here, the first device, the second device, the network device, and the intelligent apparatus each include an electronic device capable of automatically performing numerical calculation and information processing according to instructions set or stored in advance, and hardware thereof includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a programmable gate array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like.

In some embodiments, the processor may include, by way of non-limiting example, a central processing unit CPU, a graphics processing unit GPU, a CPU and a GPU, a microprocessor, a digital signal processor, or any other processing unit or component known in the art; in some embodiments, the functions of the processor may be performed alternatively by hardware logic components, which illustratively include, without limitation, field programmable gate arrays FPGAs, application specific integrated circuits ASICs, application specific standard products ASSPs, system on a chip SOCs, complex programmable logic devices CPLDs, and the like.

In some embodiments, a computer readable medium is used to store information. Computer-readable media can include, by way of example only, volatile and nonvolatile memory, removable and non-removable media for storing information by any technique; computer-readable media include, by way of example only, RAM, ROM, EEPROM, flash memory, CD-ROM, Digital Versatile Disks (DVD), optical storage, magnetic disks, magnetic tape, any other magnetic storage device, RAID storage systems, USB connected to a device by hot-plug, or any other medium capable of storing information.

In some embodiments, the program modules are stored in a computer readable medium and executable by a processor; in some embodiments, the program modules may be application programs stored on computer readable media and executed on processors.

Summary of the flow

As shown in FIG. 2, a method for natural language content title disambiguation includes the steps of:

step 201, receiving a natural language command input by a user;

when a user performs voice interaction with the device, the user is more inclined to use a more natural and simpler expression mode, for example, the user may directly send a natural language command "play secret which cannot be spoken by zhou jilun", or even directly send a natural language command "secret which cannot be spoken", "first AI meeting", or the like, so that the natural language command input by the user can be collected through a sound collection device such as a microphone or the like.

Step 203, converting the natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information.

Specifically, the natural voice command is converted into words through a voice recognition technology such as a method based on a vocal tract model and voice knowledge, a method of template matching, and a method using an artificial neural network, so that a corresponding content title can be obtained, wherein the text corresponding to the content title has a plurality of semantic information.

Here, the fact that the text corresponding to the content title has a plurality of items of meaning information means that the text corresponding to the content title is unique in literal representation, but the content represented by the text is not uniquely determined, that is, has a plurality of items of meaning. For example, "secret cannot be said" may represent music "secret cannot be said" or may represent movie "secret cannot be said"; the "first AI congress" may represent the "first AI congress" video, may also represent the "first AI congress" audio, or the "first AI congress" associated text or web page.

Step 205, according to the device type information of the corresponding output device, determining the actual meaning item information corresponding to the content title from the plurality of meaning item information.

For example, assuming that the natural language command input by the user is "secret cannot be said", and the device type information of the output device is sound, it can be determined that the actual meaning item information is music "secret cannot be said"; if the device type information of the output device is television, the actual meaning item information can be determined to be ' secret which cannot be said ' of the movie '; if the output device includes both television and stereo, then when the output device is television, the actual semantic item information may be determined to be "secret cannot be said" of the movie. In another embodiment, if the output device includes only a television, the actual semantic item information can be determined to be the movie "secret cannot be said" and the user can be prompted to determine that the content is the movie "secret cannot be said" and optionally whether to switch to music "secret cannot be said".

For another example, assuming that the natural language command input by the user is "the first AI congress" and the device type information of the output device is a sound-connected projector or television, the actual meaning item information may be determined to be "the first AI congress" video; if the device type information of the output device is sound, the actual meaning item information can be determined to be the audio of the first AI conventions; if the device type information of the output device is a projector which is not connected with a sound, the actual meaning item information can be determined to be the relevant characters, pictures or web pages of the first AI congress.

And step 207, outputting the content corresponding to the content title by an output device, wherein the content corresponding to the content title is adapted to the actual meaning item information.

Specifically, the output device firstly obtains the content corresponding to the content title according to the actual meaning item information, and then outputs the content; or the output device directly outputs the content corresponding to the content title, wherein the content corresponding to the content title is matched with the actual meaning item information. For example, the output device may obtain and output the content corresponding to the content title by locally searching the internal memory or the external memory, may obtain and output the content corresponding to the content title by searching the network, or may receive and output the content corresponding to the content title transmitted by another device.

For example, an output device such as a stereo may locally search a built-in memory or connected CD, network search, or receive music from other devices to obtain "secret not to say" and play; the output device such as a television can locally search the built-in memory or the connected VCD, DVD, USB memory, network search, or receive and obtain the 'secret which cannot be said' of the movie from other devices and play the movie; an output device such as a projector may obtain the "first AI congress" associated text or web page, etc. from other devices.

The following will describe possible embodiments by way of non-limiting example, and the present invention is not limited to the following list, in the following examples, the text corresponding to the content title has multiple items of meaning information, which means that the text corresponding to the content title is unique in terms of word expression, but the content represented by the text is not unique.

First embodiment

As shown in fig. 3, in a first embodiment, the system 100 includes a first device 110. A method for natural language content title disambiguation comprising the steps of:

in step 301, the first device 110 receives a natural language command input by a user.

Users tend to use more natural and simpler expressions when interacting with the device, for example, the user may directly send the natural language command "play secret that cannot be spoken by zhou jeghen", or even directly send the natural language command "secret that cannot be spoken", "first AI party", etc.

Step 303, after receiving the natural language command, the first device 110 converts the natural language command into a content title, where a text corresponding to the content title has a plurality of semantic item information.

Specifically, the first device 100 may collect a natural language command input by the user through a sound collection device such as a microphone or the like; then, the natural language command is converted into a content title by converting the natural language command into words through a speech recognition technology such as a method based on a vocal tract model and speech knowledge, a method of template matching, and a method using an artificial neural network, so that the corresponding content title can be obtained, and the conversion of the natural language command into the content title is realized, wherein the text corresponding to the content title has a plurality of semantic item information.

Here, the fact that the text corresponding to the content title has a plurality of items of semantic information means that the text corresponding to the content title is unique in terms of word expression, but the content represented by the text is not uniquely determined. For example, "secret cannot be said" may represent music "secret cannot be said" or may represent movie "secret cannot be said"; the "first AI congress" may represent the "first AI congress" video, may also represent the "first AI congress" audio, or the "first AI congress" associated text or web page.

In step 305, the first device 110 determines actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the corresponding output device.

For example, assuming that the natural language command input by the user is "secret cannot be said" and the device type information of the output device is sound, first device 110 may determine that the actual semantic item information is music "secret cannot be said"; if the device type information of the output device is television, then first device 110 may determine that the actual semantic item information is movie "secret cannot be said"; if the output device includes both television and stereo, then when the output device is television, first device 110 may determine that the virtual item information is a "secret cannot be said" of the movie. In another embodiment, if the output device includes only a television, first device 110 may determine that the actual semantic item information is a movie "secret cannot be spoken" and prompt the user for the determined content to be a movie "secret cannot be spoken", optionally prompting the user whether to switch to music "secret cannot be spoken".

For another example, assuming that the natural language command input by the user is "the first current AI meeting" and the device type information of the output device is a sound-connected projector or television, the first device 110 may determine that the actual meaning item information is "the first current AI meeting" video; if the device type information of the output device is a sound, the first device 110 may determine that the real meaning item information is a "first-kind AI congress" audio; if the device type information of the output device is a projector to which no audio is connected, the first device 110 may determine that the real meaning item information is "first-due AI conventions" related characters, pictures, or web pages.

Step 307, the first device 110 outputs the content corresponding to the content title according to the actual meaning item information.

Specifically, the first device 110 obtains and outputs the content corresponding to the content title through local search of the internal memory or the external memory according to the actual meaning item information, or obtains and outputs the content corresponding to the content title through network search, or receives and outputs the content corresponding to the content title sent by another device. For example, the first device 110 may be a stereo, which may locally search a built-in memory or connected CD, network search, or receive and play music "secret not to speak" from another device; the first device 110 is, for example, a television, which can locally search the built-in memory or connected VCD, DVD, USB memory, network search, or receive and obtain the movie "secret cannot be said" from other devices and play it; the first device 110 is, for example, a projector, which can obtain the relevant text or web page of the "first AI congress" from other devices and present the corresponding text or web page.

Optionally, a method for natural language content title disambiguation comprises the steps of:

receiving, at the first device 110, a natural language command input by a user; converting a natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information; incorporating device type information of the first device 110 in determining the actual item of interest information; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, a computer medium storing one or more programs for execution by the first device 110 to cause the first device 110 to: receiving a natural language command input by a user; converting the natural language command into a content title; incorporating device type information of the first device 110 in determining the actual item of interest information; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a natural language command input by a user; converting the natural language command into a content title; incorporating device type information of the first device 110 in determining the actual item of interest information; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; means for incorporating device type information of the first device 110 in determining the actual sense information; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises an information processing apparatus, wherein the information processing apparatus comprises: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; means for incorporating device type information of the first device 110 in determining the actual sense information; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises: a sound pickup unit configured to receive a natural language command input by a user; a first processing unit configured to convert a natural language command into a content title; a second processing unit configured to determine the actual sense information, which is combined with the device type information of the first device 110 when determining the actual sense information; and the output unit is configured to output the content corresponding to the content title according to the actual meaning item information.

Second embodiment

As shown in fig. 4, in a second embodiment, system 100 includes a first device 110, a network device 130, and a network 140. A method for natural language content title disambiguation comprising the steps of:

in step 401, the first device 110 receives a natural language command input by a user.

In step 409, the first device 110 receives the natural language command and then sends the natural language command to the network device 130.

Optionally, the first device 110 further sends the self device type information to the network device 130.

In step 403, after receiving the natural language command, the network device 130 converts the natural language command into a content title, where a text corresponding to the content title has a plurality of semantic item information.

For example, "secret cannot be said" may represent music "secret cannot be said" or may represent movie "secret cannot be said"; the "first AI congress" may represent the "first AI congress" video, may also represent the "first AI congress" audio, or the "first AI congress" associated text or web page.

Optionally, the network device 130 receives its device type information from the first device 110;

optionally, the network device 130 obtains device type information of the first device 110 locally.

In step 405, the network device 130 determines actual semantic item information corresponding to the content title from the plurality of semantic item information according to the device type information of the first device 110.

For example, assuming that the natural language command input by the user is "secret cannot be said" and the device type information of first device 110 is sound, network device 130 may determine that the actual semantic item information is music "secret cannot be said"; if the device type information of first device 110 is television, then network device 130 may determine that the real-world item information is movie "secret cannot be said"; if first device 110 includes both television and audio, then network device 130 may determine that the actual item of semantic information is a "secret cannot be said" of the movie when it is television. In another embodiment, if first device 110 includes only a television, network device 130 may determine that the actual semantic item information is a movie "secret cannot be spoken" and prompt the user for the determined content to be a movie "secret cannot be spoken", optionally prompting the user whether to switch to music "secret cannot be spoken".

For another example, assuming that the natural language command input by the user is "first current AI meeting" and the device type information of the first device 110 is a sound-connected projector or television, the network device 130 may determine that the actual meaning item information is "first current AI meeting" video; if the device type information of the first device 110 is stereo, the network device 130 may determine that the real meaning item information is "first-then AI congress" audio; if the device type information of the first device 110 is a projector to which no sound is connected, the network device 130 may determine that the real meaning item information is "first-due AI congress" related text, picture, or web page.

In step 411, after determining the actual sense item information, the network device 130 returns the actual sense item information to the first device 110.

Step 407, after receiving the actual item information returned by the network device 130, the first device 110 outputs the content corresponding to the content title according to the actual item information.

Specifically, the first device 110 may obtain and output the content corresponding to the content title by locally searching the internal memory or the external memory, may obtain and output the content corresponding to the content title by searching the network, and may receive and output the content corresponding to the content title sent by another device. For example, the first device 110 may be a stereo, which may locally search a built-in memory or connected CD, network search, or receive and play music "secret not to speak" from another device; the first device 110 is, for example, a television, which can locally search the built-in memory or connected VCD, DVD, USB memory, network search, or receive and obtain the movie "secret cannot be said" from other devices and play it; the first device 110 is, for example, a projector, which can obtain the text or web page related to the "first AI congress" from other devices and present the text or web page.

receiving, at the first device 110, a natural language command input by a user; sending the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 when determining the actual meaning item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, a computer medium storing one or more programs for execution by the first device 110 to cause the first device 110 to: receiving a natural language command input by a user; sending the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 when determining the actual meaning item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a natural language command input by a user; sending the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 when determining the actual meaning item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises: means for receiving a natural language command input by a user; a unit for transmitting a natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 when determining the actual meaning item information; means for receiving the real meaning item information returned by the network device 130; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises an information processing apparatus, wherein the information processing apparatus comprises: means for receiving a natural language command input by a user; a unit for transmitting a natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 when determining the actual meaning item information; means for receiving the real meaning item information returned by the network device 130; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises: a sound pickup unit configured to receive a natural language command input by a user; a transmitting unit configured to transmit a natural language command to the network device 130, to convert the natural language command into a content title after the network device 130 receives the natural language command, and to combine the device type information of the first device 110 when determining the actual significand information; a receiving unit configured to receive the real meaning item information returned by the network device 130; and the output unit is configured to output the content corresponding to the content title according to the actual meaning item information.

receiving the natural language command transmitted by the first device 110 and converting the natural language command into a content title at the network device 130; incorporating device type information of the first device 110 in determining the actual item of interest information; the actual sense information is returned to the first device 110.

Optionally, a computer medium storing one or more programs, which when executed by the network device 130, cause the network device 130 to: receiving a natural language command transmitted by the first device 110 and converting the natural language command into a content title; incorporating device type information of the first device 110 in determining the actual item of interest information; the actual sense information is returned to the first device 110.

Optionally, the network device 130 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a natural language command transmitted by the first device 110 and converting the natural language command into a content title; incorporating device type information of the first device 110 in determining the actual item of interest information; the actual sense information is returned to the first device 110.

Optionally, the network device 130 includes: means for receiving a natural language command sent by the first device 110; means for converting the natural language command into a content title; means for incorporating device type information of the first device 110 in determining the actual sense information; means for returning the actual sense information to the first device 110.

Optionally, the network device 130 includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving a natural language command sent by the first device 110; means for converting the natural language command into a content title; means for incorporating device type information of the first device 110 in determining the actual sense information; means for returning the actual sense information to the first device 110.

Optionally, the network device 130 includes: a receiving unit, configured to receive a natural language command sent by the first device 110; a first processing unit for converting a natural language command into a content title; a second processing unit for determining the actual sense information, which is combined with the device type information of the first device 110 when determining the actual sense information; a sending unit, configured to return the actual sense item information to the first device 110.

Third embodiment

As shown in fig. 5, in a third embodiment, the system 100 includes a first device 110, a network device 130, and a network 140. A method for natural language content title disambiguation comprising the steps of:

step 501, the first device 110 receives a natural language command input by a user;

users tend to use more natural, simpler expressions when interacting with the device, for example, the user may send a natural language command "play the secret that zhou jilun cannot say" directly, or even send a natural language command such as "secret cannot say" directly, the first AI party ".

Step 503, the first device 110 converts the natural language command into a content title and sends the content title to the network device 130; the text corresponding to the content title has a plurality of semantic item information;

Optionally, the first device 110 further sends the device type information of itself to the network device 130;

step 505, the network device 130 receives the content title, and determines actual item information corresponding to the content title from the plurality of item information according to the device type information of the first device 110;

Step 511, the network device 130 returns the actual item information to the first device 110;

in step 507, after receiving the actual item information returned by the network device 130, the first device 110 outputs the content corresponding to the content title according to the actual item information.

receiving, at the first device 110, a natural language command input by a user; converting the natural language command into a content title and sending the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 when determining the actual meaning item information after receiving the content title; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, a computer medium storing one or more programs for execution by the first device 110 to cause the first device 110 to: receiving a natural language command input by a user; converting the natural language command into a content title and sending the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 when determining the actual meaning item information after receiving the content title; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a natural language command input by a user; converting the natural language command into a content title and sending the content title to the network device 130, so that the network device 130 combines the device type of the first device 110 when determining the actual item information after receiving the content title; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; means for transmitting the content title to the network device 130, so that the network device 130, after receiving the content title, incorporates the device type information of the first device 110 when determining the real item information; means for receiving the real meaning item information returned by the network device 130; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises an information processing apparatus, wherein the information processing apparatus comprises: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; means for transmitting the content title to the network device 130, so that the network device 130, after receiving the content title, incorporates the device type information of the first device 110 when determining the real item information; means for receiving the real meaning item information returned by the network device 130; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 comprises: a sound pickup unit configured to receive a natural language command input by a user; a first processing unit configured to convert a natural language command into a content title; a transmitting unit configured to transmit the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 when determining the real item information after receiving the content title; a receiving unit configured to receive the real meaning item information returned by the network device 130; and the output unit is configured to output the content corresponding to the content title according to the actual meaning item information.

at the network device 130: receiving a content title transmitted by the first device 110; incorporating device type information of the first device 110 in determining the actual item of interest information; the actual sense information is returned to the first device 110.

Optionally, a computer medium storing one or more programs, which when executed by the network device 130, cause the network device 130 to: receiving a content title transmitted by the first device 110; incorporating device type information of the first device 110 in determining the actual item of interest information; the actual sense information is returned to the first device 110.

Optionally, the network device 130 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: and receiving the content title sent by the first device 110, and returning the actual item information to the first device 110 in combination with the device type of the first device 110 when determining the actual item information.

Optionally, the network device 130 includes: means for receiving a content title transmitted by the first device 110; means for incorporating device type information of the first device 110 in determining the actual sense information; means for returning the actual sense information to the first device 110.

Optionally, the network device 130 includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving a content title transmitted by the first device 110; means for incorporating device type information of the first device 110 in determining the actual sense information; means for returning the actual sense information to the first device 110.

Optionally, the network device 130 includes: a receiving unit configured to receive a content title transmitted by the first device 110; a first processing unit for determining the actual sense information, which is combined with the device type information of the first device 110 when determining the actual sense information; a sending unit, configured to return the actual sense item information to the first device 110.

Fourth embodiment

As shown in fig. 6, in a fourth embodiment, the system 100 includes a first device 110, at least one second device 120, and a network 140. A method for natural language content title disambiguation comprising the steps of:

601, a user sends a natural language command;

Step 603, after the first device 110 or the second device 120 determines that the object of the natural language command sent by the user is itself, converting the natural language command into a content title, wherein a text corresponding to the content title has a plurality of items of semantic information.

Optionally, the first device 110 or the second device 120 determines whether the object of the natural language command sent by the user is itself by one or a combination of the following ways:

-determining the output device from a distance between the user and the first device and a distance between the user and the second device;

-determining the output device according to the direction from which the user issued the natural language command;

-determining the output device in dependence of the current frontal orientation of the user.

For example, if the output device is determined according to the distance between the user and the first device and the distance between the user and the second device, assuming that the distance between the user and the first device 110 is d1, the distance between the user and the second device 120 is d2, and d1> d2, it may be determined that the device closest to the user, i.e., the second device 120, is the output device.

For another example, if the output device is determined according to the direction from which the user issues the natural language command. Specifically, the direction of the natural language command of the user can be detected through voice, and the device existing in the direction can be detected through images, so that the device in the direction of the natural language command is determined as the output device, for example, the device closest to the user in the direction of the natural language command is used as the output device.

If, for example, the output device is determined based on the current frontal orientation of the user. Specifically, it is detected through the image that a device existing in the direction in which the user is currently facing is the output device, such as a device closest to the user in the direction in which the user is currently facing is the output device.

Step 605, the first device 110 or the second device 120 determines actual item information corresponding to the content title from the plurality of item information according to its own device type information;

for example, assuming that the natural language command input by the user is "secret cannot be said", and the device type information of the first device 110 or the second device 120 is sound, it may be determined that the actual semantic item information is music "secret cannot be said"; if the device type information of first device 110 or second device 120 is a television, determining that the actual meaning item information is a "secret that cannot be said" of the movie; if first device 110 or second device 120 includes both television and stereo, then when it is television, the actual semantic item information may be determined to be a movie "secret cannot be said". In another embodiment, if first device 110 or second device 120 only includes a television, the actual semantic item information may be determined to be a movie "secret cannot be spoken" and the user may be prompted to determine that the content is a movie "secret cannot be spoken", optionally prompting the user to switch to music "secret cannot be spoken".

For another example, assuming that the natural language command input by the user is "first current AI meeting", and the device type information of the first device 110 or the second device 120 is a sound-connected projector or a television, the virtual item information may be determined to be "first current AI meeting" video; if the device type information of the first device 110 or the second device 120 is a stereo, it may be determined that the real meaning item information is a "first-kind AI congress" audio; if the device type information of the first device 110 or the second device 120 is the projector to which the audio is not connected, it may be determined that the real meaning item information is the "first AI congress" related text, picture, or web page.

Step 607, the first device 110 or the second device 120 outputs the content corresponding to the content title according to the actual sense item information.

Specifically, the first device 110 or the second device 120 may obtain and output the content corresponding to the content title by locally searching the internal memory or the external memory, may also obtain and output the content corresponding to the content title by searching through the network, and may also receive and output the content corresponding to the content title sent by another device. For example, first device 110 or second device 120 may be a stereo, for example, which may locally search a built-in memory or connected CD, network search, or receive and play music "secret not to speak" from another device; first device 110 or second device 120 is, for example, a television, which may locally search built-in memory or a connected VCD, DVD, USB memory, web search, or receive and obtain a movie "secret cannot be said" from another device and play it; the first device 110 or the second device 120 is, for example, a projector, which can obtain the text or web page related to the "first AI congress" from other devices and present the text or web page.

determining, at the first device 110 or the second device 120, that an object of the user to send the natural language command is itself; receiving a natural language command input by a user; converting the natural language command into a content title; combining the device type information of the user when determining the actual meaning item information; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, a computer medium storing one or more programs for execution by the first device 110 or the second device 120 to cause the first device 110 or the second device 120 to: determining that an object of a user sending a natural language command is the user; receiving a natural language command input by a user; converting the natural language command into a content title; combining the device type information of the user when determining the actual meaning item information; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises at least one processor and a memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: determining that an object of a user sending a natural language command is the user; receiving a natural language command input by a user; converting the natural language command into a content title; combining the device type information of the user when determining the actual meaning item information; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; means for incorporating its own device type information in determining the actual item information; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 includes an information processing apparatus, wherein the information processing apparatus includes: means for determining that an object from which a user sends a natural language command is itself; means for receiving a natural language command input by a user; means for converting the natural language command into a content title; means for incorporating its own device type information in determining the actual item information; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises: a determination unit configured to determine that an object of a user to send a natural language command is itself; a sound pickup unit configured to receive a natural language command input by a user; a first processing unit configured to convert a natural language command into a content title; a second processing unit configured to determine the actual sense information in combination with a device type of itself when determining the actual sense information; and the output unit is configured to output the content corresponding to the content title according to the actual meaning item information.

Fifth embodiment

As shown in fig. 7, in a fifth embodiment, the system 100 includes a first device 110, at least one second device 120, a network device 130, and a network 140. A method for natural language content title disambiguation comprising the steps of:

step 701, a user sends a natural language command;

users tend to use more natural and simpler expressions when interacting with the device, for example, the user may directly issue a natural language command "play the secret that zhou jilun cannot say", or even directly issue a natural language command such as "secret cannot say", "first AI party", etc.

Step 709, after the first device 110 or the second device 120 determines that the object of the user sending the natural language command is itself, sending the natural language command to the network device 130;

optionally, the first device 110 or the second device 120 further sends the device type information of itself to the network device 130;

In step 703, after receiving the natural language command, the network device 130 converts the natural language command into a content title, where a text corresponding to the content title has a plurality of semantic item information.

Step 705, the network device 130 determines actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the first device 110 or the device type information of the second device 120;

for example, assuming that the natural language command input by the user is "secret cannot be said" and the device type information of first device 110 or second device 120 is sound, network device 130 may determine that the actual semantic item information is music "secret cannot be said"; if the device type information of first device 110 or second device 120 is a television, then network device 130 determines that the actual semantic item information is a "secret that cannot be said" of the movie; if first device 110 or second device 120 includes both television and audio, then network device 130 may determine that the actual semantic item information is "secret cannot be said" of the movie when it is television. In another embodiment, if first device 110 or second device 120 only includes a television, network device 130 may determine that the real-meaning information is a movie "secret cannot be said" and prompt the user for the determined content to be a movie "secret cannot be said," optionally prompting the user to switch to music "secret cannot be said.

For another example, assuming that the natural language command input by the user is "the first AI congress", and the device type information of the first device 110 or the second device 120 is a sound-connected projector or a television, the network device 130 may determine that the actual meaning item information is "the first AI congress" video; if the device type information of the first device 110 or the second device 120 is a stereo, the network device 130 may determine that the real meaning item information is "first-due AI conventions" audio; if the device type information of the first device 110 or the second device 120 is a projector to which a sound is not connected, the network device 130 may determine that the actual meaning item information is "first-kind AI congress" related text, picture, or web page.

Optionally, the network device 130 receives the corresponding device type information from the first device 110 or the second device 120;

optionally, the network device 130 obtains the device type information of the first device 110 or the device type information of the second device 120 locally.

Step 711, the network device 130 returns the actual sense item information to the first device 110 or the second device 120;

in step 707, after receiving the actual item information returned by the network device 130, the first device 110 or the second device 120 outputs the content corresponding to the content title according to the actual item information.

determining, at the first device 110 or the second device 120, that an object of the user to send the natural language command is itself; receiving a natural language command input by a user; sending the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 or the device type information of the second device 120 when determining the actual meaning item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, a computer medium storing one or more programs for execution by the first device 110 to cause the first device 110 or the second device 120 to: determining that an object of a user sending a natural language command is the user; receiving a natural language command input by a user; sending the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 or the device type information of the second device 120 when determining the actual meaning item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises at least one processor and a memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: determining that an object of a user sending a natural language command is the user; receiving a natural language command input by a user; sending the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 or the device type information of the second device 120 when determining the actual meaning item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises: means for receiving a natural language command input by a user; means for determining that an object from which a user sends a natural language command is itself; a unit for transmitting a natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 or the device type information of the second device 120 when determining the actual semantic item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 includes an information processing apparatus, wherein the information processing apparatus includes: means for determining that an object from which a user sends a natural language command is itself; means for receiving a natural language command input by a user; a unit for transmitting a natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type information of the first device 110 or the device type information of the second device 120 when determining the actual semantic item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises: a determination unit configured to determine that an object of a user to send a natural language command is itself; a sound pickup unit configured to receive a natural language command input by a user; a transmitting unit configured to transmit a natural language command to the network device 130, convert the natural language command into a content title after the network device 130 receives the natural language command, and combine the device type information of the first device 110 or the device type information of the second device 120 when determining the actual significand information; receiving the actual meaning item information returned by the network equipment 130; and the output unit is configured to output the content corresponding to the content title according to the actual meaning item information.

at the network device 130, receiving the natural language command sent by the first device 110 or the second device 120, converting the natural language command into a content title, and combining the device type information of the first device 110 or the device type information of the second device 120 when determining the actual semantic item information; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, a computer medium storing one or more programs, which when executed by the network device 130, cause the network device 130 to: and converting the natural language command into a content title after receiving the natural language command sent by the first device 110 or the second device 120, and returning the actual item information to the first device 110 or the second device 120 in combination with the device type information of the first device 110 or the device type information of the second device 120 when determining the actual item information.

Optionally, the network device 130 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a natural language command sent by the first device 110 or the second device 120, converting the natural language command into a content title, and combining the device type information of the first device 110 or the device type information of the second device 120 when determining the actual semantic item information; the actual sense item information is returned to the first device 110 or the second device 120.

Optionally, the network device 130 includes: means for receiving a natural language command sent by first device 110 or second device 120; means for converting the natural language command into a content title; means for incorporating the device type information of the first device 110 or the device type information of the second device 120 in determining the real item information; means for returning the actual sense information to the first device 110 or the second device 120.

Optionally, the network device 130 includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving a natural language command sent by first device 110 or second device 120; means for converting the natural language command into a content title; means for incorporating the device type information of the first device 110 or the device type information of the second device 120 in determining the real item information; means for returning the actual sense information to the first device 110 or the second device 120.

Optionally, the network device 130 includes: a receiving unit, configured to receive a natural language command sent by the first device 110 or the second device 120; a first processing unit for converting a natural language command into a content title; a second processing unit, configured to determine the actual item information, which is combined with the device type information of the first device 110 or the device type information of the second device 120 when determining the actual item information; a sending unit, configured to return the actual sense item information to the first device 110 or the second device 120.

Sixth embodiment

As shown in fig. 8, in a sixth embodiment, the system 100 includes a first device 110, at least one second device 120, a network device 130, and a network 140. A method for natural language content title disambiguation comprising the steps of:

step 801, a user sends a natural language command;

Step 803, after the first device 110 or the second device 120 determines that the object of the user sending the natural language command is itself, the natural language command is converted into a content title and sent to the network device 130;

the text corresponding to the content title has a plurality of semantic item information, for example, "secret cannot be said" can represent music "secret cannot be said" and can also represent movie "secret cannot be said"; the "first AI congress" may represent the "first AI congress" video, may also represent the "first AI congress" audio, or the "first AI congress" associated text or web page.

Optionally, the first device 110 or the second device 120 further sends device type information of itself to the network device 130;

Step 805, after receiving the content title, the network device 130 determines, according to the device type information of the first device 110 or the device type information of the second device 120, actual item information corresponding to the content title from the plurality of item information;

optionally, the network device 130 obtains device type information of the first device 110 or the second device 120 locally.

Step 811, the network device 130 returns the actual item information to the first device 110 or the second device 120;

in step 807, after receiving the real item information returned by the network device 130, the first device 110 or the second device 120 outputs the content corresponding to the content title according to the real item information.

receiving a natural language command input by a user after determining that an object of the natural language command sent by the user is self, at the first device 110 or the second device 120; converting the natural language command into a content title and sending the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 or the device type information of the second device 120 when determining the actual meaning item information after receiving the content title; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, a computer medium storing one or more programs for execution by the first device 110 or the second device 120 to cause the first device 110 or the second device 120 to: determining that an object of a natural language command sent by a user is a natural language command input by the user after the object is self; converting the natural language command into a content title and sending the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 or the device type information of the second device 120 when determining the actual meaning item information after receiving the content title; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises at least one processor and a memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: determining that an object of a natural language command sent by a user is a natural language command input by the user after the object is self; converting the natural language command into a content title and sending the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 or the device type information of the second device 120 when determining the actual meaning item information after receiving the content title; receiving the actual meaning item information returned by the network equipment 130; and outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises: means for determining that an object from which a user sends a natural language command is itself; means for receiving a natural language command input by a user; means for converting the natural language command into a content title; a unit for transmitting the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 or the device type information of the second device 120 when determining the real meaning item information after receiving the content title; means for receiving the real meaning item information returned by the network device 130; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 includes an information processing apparatus, wherein the information processing apparatus includes: means for determining that an object from which a user sends a natural language command is itself; means for receiving a natural language command input by a user; means for converting the natural language command into a content title; a unit for transmitting the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 or the device type information of the second device 120 when determining the real meaning item information after receiving the content title; means for receiving the real meaning item information returned by the network device 130; and the unit is used for outputting the content corresponding to the content title according to the actual meaning item information.

Optionally, the first device 110 or the second device 120 comprises: a determination unit configured to determine that an object of a user to send a natural language command is itself; a sound pickup unit configured to receive a natural language command input by a user; a first processing unit configured to convert a natural language command into a content title; a transmitting unit configured to transmit the content title to the network device 130, so that the network device 130 combines the device type information of the first device 110 or the device type information of the second device 120 when determining the real meaning item information after receiving the content title; a receiving unit configured to receive the real meaning item information returned by the network device 130; and the output unit is configured to output the content corresponding to the content title according to the actual meaning item information.

at the network device 130, the content title sent by the first device 110 or the second device 120 is received, and the actual item information is returned to the first device 110 or the second device 120 in combination with the device type information of the first device 110 or the device type information of the second device 120 when determining the actual item information.

Optionally, a computer medium storing one or more programs, which when executed by the network device 130, cause the network device 130 to: and receiving a content title sent by the first device 110 or the second device 120, and returning the actual item information to the first device 110 or the second device 120 in combination with the device type information of the first device 110 or the device type information of the second device 120 when determining the actual item information.

Optionally, the network device 130 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: and receiving a content title sent by the first device 110 or the second device 120, and returning the actual item information to the first device 110 or the second device 120 in combination with the device type information of the first device 110 or the device type information of the second device 120 when determining the actual item information.

Optionally, the network device 130 includes: means for receiving a content title transmitted by the first device 110 or the second device 120; means for incorporating the device type information of the first device 110 or the device type information of the second device 120 in determining the real item information; means for returning the actual sense information to the first device 110 or the second device 120.

Optionally, the network device 130 includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving a content title transmitted by the first device 110 or the second device 120; means for incorporating the device type information of the first device 110 or the device type information of the second device 120 in determining the real item information; means for returning the actual sense information to the first device 110 or the second device 120.

Optionally, the network device 130 includes: a receiving unit, configured to receive a content title sent by the first device 110 or the second device 120; a first processing unit, configured to determine the actual item information, which is combined with the device type information of the first device 110 or the device type information of the second device 120 when determining the actual item information; a sending unit, configured to return the actual item information to the first device 110 or the second device 120.

Seventh embodiment

As shown in fig. 9, in the seventh embodiment, the system 100 includes a first device 110, a smart appliance 150, and a network 140. A method for natural language content title disambiguation comprising the steps of:

step 901, a user sends a natural language command;

In step 903, after receiving the natural language command, the smart device 150 converts the natural language command into a content title, where a text corresponding to the content title has a plurality of semantic item information.

In step 905, the intelligent device 150 determines, according to the device type information of the first device 110, actual meaning item information corresponding to the content title from the plurality of meaning item information.

For example, assuming that the natural language command input by the user is "secret cannot be said" and the device type information of the first device 110 is sound, the smart device 150 may determine that the actual semantic item information is music "secret cannot be said"; if the device type information of the first device 110 is a television, the intelligent device 150 determines that the actual semantic item information is a "secret that cannot be said" of the movie; if the first device 110 includes both television and audio, then when it is television, the smart device 150 may determine that the actual semantic item information is a movie "secret cannot be said". In another embodiment, if the first device 110 includes only a television, the smart device 150 may determine that the actual semantic item information is a movie "secret cannot be said" and prompt the user for the determined content to be a movie "secret cannot be said," optionally prompting the user whether to switch to music "secret cannot be said.

For another example, assuming that the natural language command input by the user is "the first current AI meeting" and the device type information of the first device 110 is a sound-connected projector or a television, the intelligent device 150 may determine that the actual meaning item information is "the first current AI meeting" video; if the device type information of the first device 110 is a stereo, the intelligent device 150 may determine that the real meaning item information is a "first-kind AI congress" audio; if the device type information of the first device 110 is a projector to which no sound is connected, the intelligent device 150 may determine that the real meaning item information is "first AI congress" related text, picture or web page.

Step 911, the intelligent device 150 sends the actual item information to the first device 110;

in step 907, after receiving the actual meaning item information sent by the intelligent device 150, the first device 110 outputs the content corresponding to the content title according to the actual meaning item information.

Optionally, step 913 (not shown) is further included before step 905: the smart device 150 determines that the first device 110 is an output device.

Optionally, the output device is determined according to an operation submitted by the user based on the smart device with respect to the output device. Specifically, the smart device 150 may provide an interface for the user to select, and the user selects the first device 110 as an output device through a touch screen or a keyboard; the smart device 150 may provide a voice interactive interface, and the user may determine that the first device 110 is an output device through natural language instructions.

Optionally, the smart device 150 determines the output device by detecting a distance between the user and the device, for example, determining the first device 110 as the output device when the user is closest to the first device 110;

alternatively, the smart device 150 determines the output device by detecting the direction of the user's natural language command, for example, if the smart device 150 detects the direction of the user's natural language command by voice and detects the device existing in the direction as the first device 110 by an image, the first device 110 is determined as the output device;

alternatively, the smart device 150 determines the output device by detecting the front facing direction of the user, for example, the smart device 150 detects the device existing in the direction facing the front facing direction of the user as the first device 110 by an image, and then determines the first device 110 as the output device.

receiving, at the smart device 150, a natural language command input by a user; converting a natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information; according to the device type information of the first device 110, determining actual meaning item information corresponding to the content title from the plurality of meaning item information; the actual sense item information is sent to the first device 110.

Optionally, a computer medium storing one or more programs for execution by the smart device 150 to cause the smart device 150 to: receiving a natural language command input by a user; converting the natural language command into a content title; according to the device type information of the first device 110, determining actual meaning item information corresponding to the content title from the plurality of meaning item information; the actual sense item information is sent to the first device 110.

Optionally, the smart device 150 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a natural language command input by a user; converting the natural language command into a content title; according to the device type information of the first device 110, determining actual meaning item information corresponding to the content title from the plurality of meaning item information; the actual sense item information is sent to the first device 110.

Optionally, the smart device 150 comprises: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; a unit configured to determine, from the plurality of item information, actual item information corresponding to the content title according to device type information of the first device 110; means for sending the actual sense item information to the first device 110.

Optionally, the smart device 150 comprises an information processing device, the information processing device comprising: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; a unit configured to determine, from the plurality of item information, actual item information corresponding to the content title according to device type information of the first device 110; means for sending the actual sense item information to the first device 110.

Optionally, the smart device 150 comprises: a sound pickup unit configured to receive a natural language command input by a user; a first processing unit configured to convert a natural language command into a content title; a second processing unit configured to determine actual semantic item information corresponding to the content title from the plurality of semantic item information according to the device type information of the first device 110; a transmitting unit configured to transmit the real item information to the first device 110.

after receiving the actual meaning item information sent by the intelligent device 150, the first device 110 outputs the content corresponding to the content title according to the actual meaning item information,

the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command input by the user is received and converted into the content title by the intelligent device 150.

Optionally, a computer medium storing one or more programs for execution by the first device 110 to cause the first device 110 to: receiving actual meaning item information sent by the intelligent device 150; outputting the content corresponding to the content title according to the actual meaning item information;

Optionally, the first device 110 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving actual meaning item information sent by the intelligent device 150; outputting the content corresponding to the content title according to the actual meaning item information;

Optionally, the first device 110 comprises: means for receiving the real item information transmitted by the smart device 150; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command input by the user is received and converted into the content title by the intelligent device 150.

Optionally, the first device 110 comprises an information processing apparatus, wherein the information processing apparatus comprises: means for receiving the real item information transmitted by the smart device 150; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command input by the user is received and converted into the content title by the intelligent device 150.

Optionally, the first device 110 comprises: a receiving unit configured to receive the actual meaning item information transmitted by the smart device 150; an output unit configured to output the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command input by the user is received and converted into the content title by the intelligent device 150.

Eighth embodiment

As shown in fig. 10, in the eighth embodiment, the system 100 includes a device 110, at least one second device 120, a smart appliance 150, and a network 140. A disambiguation method for natural language content titles comprising the steps of:

step 1013, the smart device 150 determines an output device from the first device 110 and the second device 120;

alternatively, the output device is determined according to an operation submitted by the user based on the smart device 150 with respect to the output device. Specifically, the smart device 150 may provide an interface for a user to select, and the user may select the first device 110 as an output device through a touch screen or a keyboard; the smart device 150 may provide a voice interactive interface through which the user may determine that the first determination device 110 is an output device through natural language instructions.

Optionally, the smart device 150 determines the output device by detecting the distance between the user and the device, for example, when the distance between the user and the first device 110 is the closest, the first device 110 is determined to be the output device;

alternatively, the smart device determines the output device by detecting the front facing direction of the user, for example, the smart device 150 detects the device existing in the direction facing the front facing direction of the user as the first device 110 by an image, and then determines the first device 110 as the output device.

1001, a user sends a natural language command;

In step 1003, after receiving the natural language command, the intelligent device 150 converts the natural language command into a content title, where a text corresponding to the content title has a plurality of items of information.

Step 1005, the intelligent device 150 determines actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the output device;

for example, assuming that the natural language command input by the user is "secret cannot be said" and the device type information of the output device is sound, the smart device 150 may determine that the actual meaning item information is music "secret cannot be said"; if the device type information of the output device is a television, the intelligent device 150 determines that the actual meaning item information is a "secret that cannot be said" of the movie "; if the output device includes both television and stereo, then when it is television, the smart device 150 may determine that the actual semantic item information is a "secret cannot be said" of the movie. In another embodiment, if the output device includes only a television, the smart device 150 may determine that the actual semantic item information is a movie "secret cannot be said" and prompt the user for the determined content to be a movie "secret cannot be said", optionally prompting the user whether to switch to music "secret cannot be said".

For another example, assuming that the natural language command input by the user is "the first AI congress" and the device type information of the output device is a sound-connected projector or a television, the intelligent device 150 may determine that the actual meaning item information is "the first AI congress" video; if the device type information of the output device is a sound, the intelligent device 150 may determine that the actual meaning item information is "first-minute AI conventions" audio; if the device type information of the output device is a projector to which no sound is connected, the smart device 150 may determine that the actual meaning item information is "first AI congress" related text, picture, or web page.

Step 1011, the intelligent device 150 sends the actual meaning item information to an output device;

step 1007, the output device outputs the content corresponding to the content title according to the actual meaning item information.

Specifically, the output device obtains and outputs the content corresponding to the content title by locally searching the internal memory or the external memory, or obtains and outputs the content corresponding to the content title by searching the network, or receives and outputs the content corresponding to the content title sent by another device. For example, the output device may be a sound device, which may locally search a built-in memory or a connected CD, network search, or receive and play music "secret not to say" from another device; the output device is a television, for example, which can locally search the built-in memory or connected VCD, DVD, USB memory, network search, or receive and obtain the movie 'secret unable to say' from other devices and play; the output device is, for example, a projector, which can obtain the relevant text or web page of the "first AI congress" from other devices and present the text or web page.

Optionally, the method further comprises step 1015 (not shown): the intelligent apparatus 150 determines the device type information of the output device, for example, the output device may actively send the device type information to the intelligent apparatus, or the intelligent apparatus 150 performs a matching query in the device description information base based on the identifier of the output device.

at the smart device 150, determining a corresponding output device; receiving a natural language command input by a user; converting a natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information; according to the device type information of the output device, determining actual meaning item information corresponding to the content title from the plurality of meaning item information; and sending the actual meaning item information to the output equipment.

Optionally, a computer medium storing one or more programs for execution by the smart device 150 to cause the smart device 150 to: determining corresponding output equipment; receiving a natural language command input by a user; converting a natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information; according to the device type information of the output device, determining actual meaning item information corresponding to the content title from the plurality of meaning item information; and sending the actual meaning item information to the output equipment.

Optionally, the smart device 150 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: determining corresponding output equipment; receiving a natural language command input by a user; converting a natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information; according to the device type information of the output device, determining actual meaning item information corresponding to the content title from the plurality of meaning item information; and sending the actual meaning item information to the output equipment.

Optionally, the smart device 150 comprises: means for determining a corresponding output device; means for receiving a natural language command input by a user; the unit is used for converting the natural language command into a content title, wherein the text corresponding to the content title has a plurality of semantic item information; a unit configured to determine, from the plurality of items of semantic information, actual semantic item information corresponding to the content title according to device type information of the output device; means for sending the virtual item information to an output device.

Optionally, the smart device 150 includes an information processing device, wherein the information processing device includes: means for receiving a natural language command input by a user; the unit is used for converting the natural language command into a content title, wherein the text corresponding to the content title has a plurality of semantic item information; a unit configured to determine, from the plurality of items of semantic information, actual semantic item information corresponding to the content title according to device type information of the output device; means for sending the virtual item information to an output device.

Optionally, the smart device 150 comprises: a determination unit for determining a corresponding output device; the pickup unit is used for receiving natural language commands input by a user; the first processing unit is used for converting the natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information; the second processing unit is used for determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to the equipment type information of the output equipment; and the sending unit is used for sending the actual meaning item information to the output equipment.

receiving, at the output device, the actual item of interest information sent by the corresponding smart device 150; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information; the actual semantic item information is determined from the plurality of semantic item information by the smart device 150 after receiving a natural language command input by a user and converting it into the content title, according to the device type information of the output device.

Optionally, a computer medium storing one or more programs, the execution of which by an output device causes the output device to: receiving actual meaning item information sent by the intelligent device 150; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the natural language command input by the user is received by the intelligent device 150 and converted into the content title.

Optionally, the output device comprises at least one processor and a memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving actual meaning item information sent by the intelligent device 150; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the natural language command input by the user is received by the intelligent device 150 and converted into the content title.

Optionally, the output device comprises: means for receiving the real item information transmitted by the smart device 150; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the natural language command input by the user is received by the intelligent device 150 and converted into the content title.

Optionally, the output device includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving the real item information transmitted by the smart device 150; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the natural language command input by the user is received by the intelligent device 150 and converted into the content title.

Optionally, the output device comprises: a receiving unit configured to receive the actual meaning item information transmitted by the smart device 150; the output unit is configured to output the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the natural language command input by the user is received by the intelligent device 150 and converted into the content title.

Ninth embodiment

As shown in fig. 11, in the ninth embodiment, the system 100 includes a first device 110, an intelligent apparatus 150, a network device 130, and a network 140. A method for natural language content title disambiguation comprising the steps of:

step 1101, a user sends a natural language command;

Step 1109, after receiving the natural language command, the intelligent device 150 sends the natural language command to the network device 130;

optionally, the intelligent device 150 further sends device type information to the network device 130;

step 1103, after receiving the natural language command, the network device 130 converts the natural language command into a content title, where a text corresponding to the content title has a plurality of semantic item information;

Step 1105, the network device 130 determines the actual item information corresponding to the content title from the multiple item information according to the device type information of the first device 110;

for example, assuming that the natural language command input by the user is "secret cannot be said" and the device type information of first device 110 is sound, network device 130 may determine that the actual semantic item information is music "secret cannot be said"; if the device type information of first device 110 is a television, then network device 130 determines that the actual semantic item information is a "secret that cannot be said" of the movie "; if first device 110 includes both television and audio, then network device 130 may determine that the actual item of semantic information is a "secret cannot be said" of the movie when it is television. In another embodiment, if first device 110 includes only a television, network device 130 may determine that the actual semantic item information is a movie "secret cannot be spoken" and prompt the user for the determined content to be a movie "secret cannot be spoken", optionally prompting the user whether to switch to music "secret cannot be spoken".

Optionally, the network device 130 receives the device type information of the first device 110 from the smart appliance 150;

Step 1111, the network device 130 sends the actual item information to the first device 110;

in step 1107, after receiving the actual item information sent by the network device 130, the first device 110 outputs the content corresponding to the content title according to the actual item information.

Optionally, step 1113 (not shown) is further included before step 1103: the smart device 150 determines that the first device 110 is an output device.

Optionally, the output device is determined according to an operation submitted by the user based on the smart device 150 with respect to the output device. Specifically, the smart device 150 may provide an interface for the user to select, and the user selects the first device 110 as an output device through a touch screen or a keyboard; the smart device 150 may provide a voice interactive interface, and the user may determine that the first device 110 is an output device through natural language instructions.

receiving, at the smart device 150, a natural language command input by a user; the natural language command is sent to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and determines actual content information corresponding to the content title from the plurality of content information according to the device type information of the first device 110.

Optionally, a computer medium storing one or more programs for execution by the smart device 150 to cause the smart device 150 to: receiving a natural language command input by a user; the natural language command is sent to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and determines actual content information corresponding to the content title from the plurality of content information according to the device type information of the first device 110.

Optionally, the smart device 150 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a natural language command input by a user; the natural language command is sent to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and determines actual content information corresponding to the content title from the plurality of content information according to the device type information of the first device 110.

Optionally, the smart device 150 comprises: means for receiving a natural language command input by a user; a unit configured to send the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and determines, according to the device type information of the first device 110, actual meaning item information corresponding to the content title from the plurality of meaning item information.

Optionally, the smart device 150 comprises an information processing device, the information processing device comprising: means for receiving a natural language command input by a user; a unit configured to send the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and determines, according to the device type information of the first device 110, actual meaning item information corresponding to the content title from the plurality of meaning item information.

Optionally, the smart device 150 comprises: a sound pickup unit configured to receive a natural language command input by a user; a sending unit configured to send the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and determines actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the first device 110.

at the network device 130, after receiving the natural language command sent by the intelligent apparatus 150, the natural language command is converted into a content title, actual item information corresponding to the content title is determined from the plurality of item information according to the device type information of the first device 110, and the actual item information is sent to the first device 110.

Optionally, a computer medium storing one or more programs, which when executed by the network device 130, cause the network device 130 to: the natural language command is converted into a content title after receiving the natural language command sent by the intelligent device 150, actual item information corresponding to the content title is determined from the plurality of item information according to the device type information of the first device 110, and the actual item information is sent to the first device 110.

Optionally, the network device 130 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: the natural language command is converted into a content title after receiving the natural language command sent by the intelligent device 150, actual item information corresponding to the content title is determined from the plurality of item information according to the device type information of the first device 110, and the actual item information is sent to the first device 110.

Optionally, the network device 130 includes: means for natural language commands sent by the smart device 150; means for converting the natural language command into a content title; a unit configured to determine, from the plurality of item information, actual item information corresponding to the content title according to device type information of the first device 110; means for sending the actual sense item information to the first device 110.

Alternatively, the network device 130 includes an information processing apparatus including: means for receiving a natural language command sent by the smart device 150; means for converting the natural language command into a content title; a unit configured to determine, from the plurality of item information, actual item information corresponding to the content title according to device type information of the first device 110; means for sending the actual sense item information to the first device 110.

Optionally, the network device 130 includes: a receiving unit, configured to receive a natural language command sent by the smart device 150; a first processing unit for converting a natural language command into a content title; the second processing unit is used for determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to the equipment type information of the first equipment 110; a sending unit, configured to send the actual item information to the first device 110.

receiving, at the first device 110, the real item information transmitted by the network device 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command received by the network device 130 from the intelligent apparatus 150 is converted into the content title.

Optionally, a computer medium storing one or more programs for execution by the first device 110 to cause the first device 110 to: receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command received by the network device 130 from the intelligent apparatus 150 is converted into the content title.

Optionally, the first device 110 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command received by the network device 130 from the intelligent apparatus 150 is converted into the content title.

Optionally, the first device 110 comprises: means for receiving the real meaning item information transmitted by the network device 130; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command received by the network device 130 from the intelligent apparatus 150 is converted into the content title.

Optionally, the first device 110 comprises an information processing apparatus, wherein the information processing apparatus comprises: means for receiving the real meaning item information transmitted by the network device 130; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command received by the network device 130 from the intelligent apparatus 150 is converted into the content title.

Optionally, the first device 110 comprises: a receiving unit configured to receive the real meaning item information transmitted by the network device 130; the output unit is configured to output the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the first device 110 after the natural language command received by the network device 130 from the intelligent apparatus 150 is converted into the content title.

Tenth embodiment

As shown in fig. 12, in the tenth embodiment, the system 100 includes a first device 110, an intelligent apparatus 150, a network device 130, and a network 140. The embodiment illustrates a method for natural language content title disambiguation with an output device as the first device 110, wherein the method comprises the steps of:

step 1201, the user sends a natural language command;

Step 1203, after receiving the natural language command, the intelligent device 150 converts the natural language command into a content title, and sends the content title to the network device 130, where a text corresponding to the content title has a plurality of items of information.

Optionally, the intelligent device 150 further sends device type information of the corresponding output device to the network device 130;

step 1205, after receiving the content title, the network device 130 determines, according to the device type information of the output device, actual item information corresponding to the content title from the multiple item information;

for example, assuming that the natural language command input by the user is "secret cannot be said" and the device type information of the output device is sound, the network device 130 may determine that the actual semantic item information is music "secret cannot be said"; if the device type information of the output device is a television, the network device 130 determines that the actual meaning item information is a "secret that cannot be said" of the movie "; if the output device includes both television and audio, then the network device 130 may determine that the real meaning item information is "secret cannot be said" of the movie when it is television. In another embodiment, if the output device includes only a television, the network device 130 may determine that the actual semantic item information is a movie "secret cannot be said" and prompt the user for the determined content to be a movie "secret cannot be said", optionally prompting the user whether to switch to music "secret cannot be said".

For another example, assuming that the natural language command input by the user is "the first-ending AI congress" and the device type information of the output device is a sound-connected projector or television, the network device 130 may determine that the actual meaning item information is "the first-ending AI congress" video; if the device type information of the output device is a stereo, the network device 130 may determine that the real meaning item information is a "first-minute AI congress" audio; if the device type information of the output device is a projector to which no audio is connected, the network device 130 may determine that the real meaning item information is "first-due AI congress" related text, picture, or web page.

Optionally, the network device 130 receives device type information of the output device from the smart appliance 150;

optionally, the network device 130 locally outputs device type information of the device;

step 1211, the network device 130 sends the actual sense item information to an output device such as the first device 110;

step 1207, after receiving the actual item information sent by the network device 130, the first device 110 outputs the content corresponding to the content title according to the actual item information.

Specifically, the output device, such as the first device 110, obtains and outputs the content corresponding to the content title by locally searching the internal memory or the external memory, or obtains and outputs the content corresponding to the content title by searching the network, or receives and outputs the content corresponding to the content title sent by another device. For example, the first device 110 may be a stereo, which may locally search a built-in memory or connected CD, network search, or receive and play music "secret not to speak" from another device; the first device 110 is, for example, a television, which can locally search the built-in memory or connected VCD, DVD, USB memory, network search, or receive and obtain the movie "secret cannot be said" from other devices and play it; the first device 110 is, for example, a projector, which can obtain the text or web page related to the "first AI congress" from other devices and present the text or web page.

Optionally, step 1213 (not shown) is also included before step 1203: the smart device 150 determines that the first device 110 is an output device.

Alternatively, the output device is determined according to an operation submitted by the user based on the smart device 150 with respect to the output device. Specifically, the smart device 150 may provide an interface for the user to select, and the user selects the first device 110 as an output device through a touch screen or a keyboard; the smart device 150 may provide a voice interactive interface, and the user may determine that the first device 110 is an output device through natural language instructions.

receiving, at the smart device 150, a natural language command input by a user; and converting the natural language command into a content title and sending the content title to the network equipment 130, so that after the network equipment 130 receives the content title, the actual meaning item information corresponding to the content title is determined from the plurality of meaning item information according to the equipment type information of the corresponding output equipment.

Optionally, a computer medium storing one or more programs for execution by the smart device 150 to cause the smart device 150 to: receiving a natural language command input by a user; and converting the natural language command into a content title and sending the content title to the network equipment 130, so that after the network equipment 130 receives the content title, the actual meaning item information corresponding to the content title is determined from the plurality of meaning item information according to the equipment type information of the corresponding output equipment.

Optionally, the smart device 150 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a natural language command input by a user; and converting the natural language command into a content title and sending the content title to the network equipment 130, so that after the network equipment 130 receives the content title, the actual meaning item information corresponding to the content title is determined from the plurality of meaning item information according to the equipment type information of the corresponding output equipment.

Optionally, the smart device 150 comprises: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; and a unit for transmitting the content title to the network device 130, so that after the network device 130 receives the content title, the network device determines the actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the corresponding output device.

Optionally, the smart device 150 comprises an information processing device, the information processing device comprising: means for receiving a natural language command input by a user; means for converting the natural language command into a content title; and a unit for transmitting the content title to the network device 130, so that after the network device 130 receives the content title, the network device determines the actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the corresponding output device.

Optionally, the smart device 150 comprises: a sound pickup unit configured to receive a natural language command input by a user; a first processing unit configured to convert a natural language command into a content title; and a transmitting unit configured to transmit the content title to the network device 130, so that after the network device 130 receives the content title, the network device determines the actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the corresponding output device.

Optionally, a method for natural language content title disambiguation comprises the steps of: at the network device 130: receiving a content title sent by a corresponding output device (such as the first device 110), determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the output device, and sending the actual meaning item information to the output device.

Optionally, a computer medium storing one or more programs, which when executed by the network device 130, cause the network device 130 to: receiving a content title sent by a corresponding output device (such as the first device 110), determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the output device, and sending the actual meaning item information to the output device.

Optionally, the network device 130 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a content title sent by a corresponding output device (such as the first device 110), determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the output device, and sending the actual meaning item information to the output device.

Optionally, the network device 130 includes: means for receiving a content title transmitted by a corresponding output device (e.g., the first device 110); the device type information unit is used for determining actual meaning item information units corresponding to the content titles from the meaning item information according to the device type information of the output device; means for sending the virtual item information to the output device.

Optionally, the network device 130 includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving a content title transmitted by a corresponding output device (e.g., the first device 110); a unit configured to determine, from the plurality of items of semantic information, actual semantic item information corresponding to the content title according to device type information of the output device; means for sending the virtual item information to the output device.

Optionally, the network device 130 includes: a receiving unit, configured to receive a content title sent by a corresponding output device (e.g., the first device 110); the first processing unit is used for determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to the equipment type information of the output equipment; and the sending unit is used for sending the actual meaning item information to the output equipment.

Optionally, a method for natural language content title disambiguation comprises the steps of: at an output device (e.g., first device 110): receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the output device after the network device 130 receives the content title.

Optionally, a computer medium storing one or more programs for execution by the first device 110 to cause the first device 110 to: receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the first device 110 after the network device 130 receives the content title.

Optionally, the first device 110 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the first device 110 after the network device 130 receives the content title.

Optionally, the first device 110 comprises: means for receiving the real meaning item information transmitted by the network device 130; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the first device 110 after the network device 130 receives the content title.

Optionally, the first device 110 comprises an information processing apparatus, wherein the information processing apparatus comprises: means for receiving the real meaning item information transmitted by the network device 130; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the first device 110 after the network device 130 receives the content title.

Optionally, the first device 110 comprises: a receiving unit configured to receive the real meaning item information transmitted by the network device 130; the output unit is configured to output the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the first device 110 after the network device 130 receives the content title.

Eleventh embodiment

As shown in fig. 13, in the eleventh embodiment, the system 100 includes a first device 110, at least one second device 120, a smart appliance 150, a network device 130, and a network 140. A method for natural language content title disambiguation comprising the steps of:

step 1313, the smart device 150 determines an output device from the first device 110 and the second device 120;

Optionally, the smart device 150 determines the output device by detecting the distance between the user and the first device 110 and the distance between the user and the second device 120, for example, determining the first device 110 as the output device when the user is closest to the first device 110;

Step 1301, a user sends a natural language command;

Step 1309, after receiving the natural language command, the intelligent device 150 forwards the natural language command to the network device 130;

optionally, the intelligent device 150 further sends the device type information of the output device to the network device 130;

step 1303, after receiving the natural language command, the network device 130 converts the natural language command into a content title;

Step 1305, the network device 130 determines, according to the device type information of the output device, actual meaning item information corresponding to the content title from the plurality of meaning item information;

Step 1311, the network device 130 sends the information of the real meaning item to an output device;

step 1307, after the output device receives the actual meaning item information sent by the network device 130, the output device outputs the content corresponding to the content title according to the actual meaning item information.

Optionally, a method for natural language content title disambiguation comprises the steps of: at the smart appliance 150, determining the output device as either device 110 or device 120; receiving a natural language command input by a user; the natural language command is transmitted to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type of the output device when determining the actual meaning item information.

Optionally, a computer medium storing one or more programs for execution by the smart device 150 to cause the smart device 150 to: determining that the output device is device 110 or device 120; receiving a natural language command input by a user; the natural language command is transmitted to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type of the output device when determining the actual meaning item information.

Optionally, the smart device 150 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: determining that the output device is device 110 or device 120; receiving a natural language command input by a user; the natural language command is transmitted to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type of the output device when determining the actual meaning item information.

Optionally, the smart device 150 comprises: means for determining that the output device is device 110 or device 120; means for receiving a natural language command input by a user; a unit for transmitting the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and combines the device type of the output device when determining the actual meaning item information.

Optionally, the smart device 150 comprises an information processing device, the information processing device comprising: means for determining an output device from the first device 110 and the second device 120; means for receiving a natural language command input by a user; a unit for sending the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and determines actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the output device.

Optionally, the smart device 150 comprises: a determination unit configured to determine an output device from the first device 110 and the second device 120; a sound pickup unit configured to receive a natural language command input by a user; a sending unit configured to send the natural language command to the network device 130, so that the network device 130 converts the natural language command into a content title after receiving the natural language command, and determines actual meaning item information corresponding to the content title from the plurality of meaning item information according to the device type information of the output device.

Optionally, a method for natural language content title disambiguation comprises the steps of: at the network device 130, after receiving the natural language command sent by the intelligent apparatus 150, converting the natural language command into a content title, where a text corresponding to the content title has a plurality of items of information; determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to equipment type information corresponding to output equipment; and sending the actual meaning item information to an output device.

Optionally, a computer medium storing one or more programs, which when executed by the network device 130, cause the network device 130 to: the method comprises the steps of receiving a natural language command sent by an intelligent device 150, and converting the natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information; determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to equipment type information corresponding to output equipment; and sending the actual meaning item information to an output device.

Optionally, the network device 130 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: the method comprises the steps of receiving a natural language command sent by an intelligent device 150, and converting the natural language command into a content title, wherein a text corresponding to the content title has a plurality of semantic item information; determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to equipment type information corresponding to output equipment; and sending the actual meaning item information to an output device.

Optionally, the network device 130 includes: means for receiving a natural language command sent by the smart device 150; the unit is used for converting the natural language command into a content title, wherein the text corresponding to the content title has a plurality of semantic item information; a unit configured to determine actual semantic item information corresponding to the content title from the plurality of semantic item information according to device type information corresponding to an output device; means for sending the virtual item information to an output device.

Optionally, the network device 130 includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving a natural language command sent by the smart device 150; the unit is used for converting the natural language command into a content title, wherein the text corresponding to the content title has a plurality of semantic item information; a unit configured to determine actual semantic item information corresponding to the content title from the plurality of semantic item information according to device type information corresponding to an output device; means for sending the virtual item information to an output device.

Optionally, the network device 130 includes: a receiving unit, configured to receive a natural language command sent by the smart device 150; a first processing unit for converting a natural language command into a content title; the second processing unit is used for determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to the equipment type information of the corresponding output equipment; a transmitting unit configured to transmit the real item information to an output device.

Optionally, a method for natural language content title disambiguation comprises the steps of: receiving, at the output device, the real item information transmitted by the network device 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the network device 130 receives the natural language command sent by the intelligent device 150 and converts the natural language command into the content title.

Optionally, a computer medium storing one or more programs, the execution of which by an output device causes the output device to: receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the network device 130 receives the natural language command sent by the intelligent device 150 and converts the natural language command into the content title.

Optionally, the output device comprises at least one processor and a memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the network device 130 receives the natural language command sent by the intelligent device 150 and converts the natural language command into the content title.

Optionally, the output device comprises: means for receiving the real meaning item information sent by the network device 130; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the network device 130 receives the natural language command sent by the intelligent device 150 and converts the natural language command into the content title.

Optionally, the output device includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving the real meaning item information transmitted by the network device 130; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the network device 130 receives the natural language command sent by the intelligent device 150 and converts the natural language command into the content title.

Optionally, the output device comprises: a receiving unit configured to receive the real meaning item information transmitted by the network device 130; the output unit is configured to output the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of semantic item information, and the actual semantic item information is determined from the plurality of semantic item information according to the device type information of the output device after the network device 130 receives the natural language command sent by the intelligent device 150 and converts the natural language command into the content title.

Twelfth embodiment

As shown in fig. 14, in the twelfth embodiment, the system 100 includes a first device 110, at least one second device 120, a smart appliance 150, a network device 130, and a network 140. A method for natural language content title disambiguation comprising the steps of:

step 1413, the smart device 150 determines an output device from the first device 110 and the second device 120;

1401, a user sends a natural language command;

Step 1403, after receiving the natural language command, the intelligent device 150 converts the natural language command into a content title and sends the content title to the network device 130, where a text corresponding to the content title has a plurality of items of information;

step 1405, the network device 130 receives the content title, and determines actual item information corresponding to the content title from the plurality of item information according to the device type information of the output device;

Step 1411, the network device 130 sends the actual item information to an output device;

step 1407, after the output device receives the actual meaning item information sent by the network device 130, the output device outputs the content corresponding to the content title according to the actual meaning item information.

Optionally, a method for natural language content title disambiguation comprises the steps of: determining, at the smart device 150, an output device from the first device 110 and the second device 120; and after receiving the natural language command, converting the natural language command into a content title and sending the content title to the network equipment 130, so that after receiving the content title, the network equipment 130 determines the actual meaning item information corresponding to the content title from the plurality of meaning item information according to the equipment type information of the output equipment.

Optionally, a computer medium storing one or more programs for execution by the smart device 150 to cause the smart device 150 to: determining an output device from the first device 110 and the second device 120; and after receiving the natural language command, converting the natural language command into a content title and sending the content title to the network equipment 130, so that after receiving the content title, the network equipment 130 determines the actual meaning item information corresponding to the content title from the plurality of meaning item information according to the equipment type information of the output equipment.

Optionally, the smart device 150 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: determining an output device from the first device 110 and the second device 120; and after receiving the natural language command, converting the natural language command into a content title and sending the content title to the network equipment 130, so that after receiving the content title, the network equipment 130 determines the actual meaning item information corresponding to the content title from the plurality of meaning item information according to the equipment type information of the output equipment.

Optionally, the smart device 150 comprises: means for determining that the output device is device 110 or device 120; means for receiving a natural language command input by a user; means for determining an output device from the first device 110 and the second device 120; and a unit for sending the content title to the network device 130, so that after the network device 130 receives the content title, the actual meaning item information corresponding to the content title is determined from the plurality of meaning item information according to the device type information of the output device.

Optionally, the smart device 150 includes an information processing device, wherein the information processing device includes: means for determining an output device from the first device 110 and the second device 120; means for receiving a natural language command input by a user; means for converting the natural language command into a content title; and a unit for sending the content title to the network device 130, so that after the network device 130 receives the content title, the actual meaning item information corresponding to the content title is determined from the plurality of meaning item information according to the device type information of the output device.

Optionally, the smart device 150 comprises: a determination unit configured to determine an output device from the first device 110 and the second device 120; a sound pickup unit configured to receive a natural language command input by a user; a first processing unit configured to convert a natural language command into a content title; and a sending unit configured to send the content title to the network device 130, so that after the network device 130 receives the content title, the actual meaning item information corresponding to the content title is determined from the plurality of meaning item information according to the device type information of the output device.

Optionally, a method for natural language content title disambiguation comprises the steps of: receiving, at the network apparatus 130, the content title transmitted by the smart device 150; determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to equipment type information corresponding to output equipment; and sending the actual meaning item information to the output equipment.

Optionally, a computer medium storing one or more programs, which when executed by the network device 130, cause the network device 130 to: receiving a content title transmitted by the smart device 150; determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to equipment type information corresponding to output equipment; and sending the actual meaning item information to the output equipment.

Optionally, the network device 130 comprises at least one processor and memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving a content title transmitted by the smart device 150; determining actual meaning item information corresponding to the content title from the plurality of meaning item information according to equipment type information corresponding to output equipment; and sending the actual meaning item information to the output equipment.

Optionally, the network device 130 includes: means for receiving a content title transmitted by the smart device 150; a unit configured to determine actual semantic item information corresponding to the content title from the plurality of semantic item information according to device type information corresponding to an output device; means for sending the virtual item information to the output device.

Optionally, the network device 130 includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving a content title transmitted by the smart device 150; the device type information unit is used for determining actual meaning item information unit corresponding to the content title from the meaning item information according to the device type information of the corresponding output device; means for sending the virtual item information to the output device.

Optionally, the network device 130 includes: a receiving unit, configured to receive a content title sent by the smart device 150; a first processing unit configured to determine actual semantic item information corresponding to the content title from the plurality of semantic item information according to device type information of a corresponding output device; a transmission unit configured to transmit the realistic meaning item information to the output device.

Optionally, a method for natural language content title disambiguation comprises the steps of: receiving, at the output device, the real item information transmitted by the network device 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the output device after the network device receives the content title from the intelligent device.

Optionally, a computer medium storing one or more programs, the execution of which by an output device causes the output device to: receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the output device after the network device receives the content title from the intelligent device.

Optionally, the output device comprises at least one processor and a memory storing one or more programs for execution by the at least one processor, the programs comprising instructions for: receiving the actual meaning item information sent by the network equipment 130; outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the output device after the network device receives the content title from the intelligent device.

Optionally, the output device comprises: means for receiving the real meaning item information transmitted by the network device 130; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the output device after the network device receives the content title from the intelligent device.

Optionally, the output device includes an information processing apparatus, wherein the information processing apparatus includes: means for receiving the real meaning item information transmitted by the network device 130; a unit for outputting the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the output device after the network device receives the content title from the intelligent device.

Optionally, the output device comprises: a receiving unit configured to receive the real meaning item information transmitted by the network device 130; the output unit is configured to output the content corresponding to the content title according to the actual meaning item information; the text corresponding to the content title has a plurality of items of information, and the actual items of information are determined from the plurality of items of information according to the device type information of the output device after the network device receives the content title from the intelligent device.

It should be noted that the present invention may be implemented in software and/or in a combination of software and hardware, for example, as an Application Specific Integrated Circuit (ASIC), a general purpose computer or any other similar hardware device. In one embodiment, the software program of the present invention may be executed by a processor to implement the steps or functions described above. Also, the software programs (including associated data structures) of the present invention can be stored in a computer readable recording medium, such as RAM memory, magnetic or optical drive or diskette and the like. Further, some of the steps or functions of the present invention may be implemented in hardware, for example, as circuitry that cooperates with the processor to perform various steps or functions.

In addition, some of the present invention can be applied as a computer program product, such as computer program instructions, which when executed by a computer, can invoke or provide the method and/or technical solution according to the present invention through the operation of the computer. Program instructions which invoke the methods of the present invention may be stored on a fixed or removable recording medium and/or transmitted via a data stream on a broadcast or other signal-bearing medium and/or stored within a working memory of a computer device operating in accordance with the program instructions. An embodiment according to the invention herein comprises an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein the computer program instructions, when executed by the processor, trigger the apparatus to perform a method and/or solution according to embodiments of the invention as described above.

It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned. Furthermore, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means recited in the apparatus claims may also be implemented by one unit or means in software or hardware. The terms first, second, etc. are used to denote names, but not any particular order.

Claims

1. A method applied to a natural language content title disambiguation system comprising a first device, other devices including at least a second device, a cloud, and a network, the method comprising the steps of:

a user sends a natural language command;

2. The method of claim 1, the cloud receiving the device type tag from the first device or the second device.

3. The method of claim 1, the cloud obtaining the device type tag of the first device or the second device locally.

4. A method as claimed in any one of claims 1 to 3, wherein the first device or the second device determines whether the object from which the user sends the natural language command is itself by one or more of:

according to the distance between the user and the equipment;

detecting a user natural language command direction according to the audio;

the front face orientation of the user is detected from the image.

5. A method for natural language content title disambiguation, the method comprising performing, at a device, the steps of:

the natural language command is sent to the cloud,

6. A computer medium having stored thereon a computer program, the program being executable by an apparatus to cause the apparatus to:

the natural language command is sent to the cloud,

7. An apparatus for natural language content title disambiguation, the apparatus comprising a processor, a memory, and a program stored on the memory and executable by the processor, the processor executing the program to perform the steps of:

the natural language command is sent to the cloud,

8. An output device for natural language content title disambiguation, the output device comprising:

means for sending a natural language command to the cloud,

9. An apparatus for natural language content title disambiguation, the apparatus comprising an information processing device comprising:

means for sending a natural language command to the cloud,

10. An apparatus for natural language content title disambiguation, the apparatus comprising:

a transmitting unit transmitting the natural language command to the cloud,

11. A method for natural language content title disambiguation, the method performing the following steps at a cloud:

receiving a natural language command sent by equipment;

12. A computer medium having stored thereon a computer program for execution by a cloud to cause the cloud to:

receiving a natural language command sent by equipment;

13. A cloud for natural language content title disambiguation, the cloud comprising a processor, a memory, and a program stored on the memory and executable by the processor, the processor executing the program to perform the steps of:

receiving a natural language command sent by equipment;

14. A cloud for natural language content title disambiguation, the cloud comprising:

a module for receiving a natural language command sent by a device;

15. A cloud for natural language content title disambiguation, the cloud comprising an information processing apparatus comprising:

a module for receiving a natural language command sent by a device;

16. A cloud for natural language content title disambiguation, the cloud comprising:

17. The cloud of claim 16, said first and second processing units being combinable into one processing unit.

18. A natural language content title disambiguation system comprising a cloud according to any of claims 7-10, a device according to any of claims 13-17 and a network.