Display device and voice identifier recommendation method

Abstract

The invention provides a display device and a voice identifier recommendation method. The method comprises the following steps: receiving voice interaction data of a user; determining a target keyword of the voice interaction data; searching the target keyword in a pre-constructed voice identifier library, and determining a target subject term corresponding to the target keyword; determining a target voice identifier corresponding to the target subject term in the voice identifier library in response to the fact that the number of times that the target subject term is searched in the voice identifier library exceeds a target search threshold value corresponding to the target subject term; and displaying the target voice identifier. According to the process, the subject term can be determined based on the voice interaction data of the user, and the voice identifier corresponding to the subject term is displayed to the user by using the voice identifier library, so that the user can replace the virtual image of the voice assistant, the voice interaction frequency of the user can be increased, and the voice service experience of the user is improved.

Description

Technical Field

[0001] The present application relates to the field of voice processing technology, and in particular to a display device and a voice identification recommendation method. Background Art

[0002] Voice assistants for smart TVs have become an important part of modern smart homes. They not only provide basic voice control functions, such as searching for programs and adjusting the volume, but also enable more advanced operations through voice interaction, such as controlling other smart devices in the home and querying information. The design of the virtual images of voice assistants on the market tends to be simple and intuitive, making them easy for users to identify and interact with.

[0003] However, the existing virtual images of voice assistants are usually simple icons or have no specific image. They are only used as a display of functional output, but not as part of the user's personalized experience. Their images lack personalized characteristics, causing users to feel aesthetic fatigue after using them for a period of time. As users' demand for personalized experience increases, the single image of voice assistants can no longer meet users' usage needs. Summary of the invention

[0004] The present application provides a display device and a voice identification recommendation method, which can identify keywords and subject terms that the user is interested in based on the user's voice interaction data, and then use a voice identification library to recommend voice identifications that match the user's interests based on the subject terms, thereby improving the display effect of the voice assistant's virtual image and improving the user's usage experience.

[0005] In a first aspect, an embodiment of the present application provides a display device, including:

[0006] monitor;

[0007] an audio receiver, coupled to the display, for receiving audio data;

[0008] A controller is coupled to the display and the audio receiver and is configured to:

[0009] Controlling the audio receiver to receive the user's voice interaction data;

[0010] Determine the target keywords for voice interaction data;

[0011] Search for target keywords in the pre-built speech markup library and determine the target subject words corresponding to the target keywords;

[0012] In response to the number of times the target subject word is searched in the voice identification library exceeding the target search threshold corresponding to the target subject word, determining a target voice identification corresponding to the target subject word in the voice identification library;

[0013] The control display shows the target voice mark.

[0014] In the technical solution provided by the present application, after receiving the user's voice interaction data, the target keyword of the voice interaction data can be determined first, and then the target subject word can be determined in the pre-built voice identification library based on the target keyword. Afterwards, when the number of searches for the target subject word exceeds the corresponding target search threshold, the voice identification corresponding to the target subject word is determined as the target voice identification, and the target voice identification is displayed. This technical solution enables the display device to identify the subject words that the user is interested in based on the acquired user's voice interaction data when the user performs voice interaction, and recommend the voice identification corresponding to the subject word as the virtual image of the voice assistant to the user for use, thereby enhancing the user's sense of interaction, gain and participation in using voice, and allowing the user to have a better experience.

[0015] In some feasible embodiments, the controller determines the target keyword of the voice interaction data and is configured to:

[0016] Convert voice interaction data into text data;

[0017] Preprocess the text data, including word segmentation and removal of banned words;

[0018] Based on the preset word extraction model, target keywords are extracted from the preprocessed text data.

[0019] By adopting this embodiment, the user's voice interaction data can be converted into text data, the target keywords in the text data can be determined, and the important information in the voice interaction data can be quickly identified to obtain the words that the user is interested in. This provides strong data support for the subsequent method flow of determining the target voice identifier.

[0020] In some feasible embodiments, before the controller searches for a target keyword in a pre-built voice identification library and determines a target subject word corresponding to the target keyword, the controller is configured to:

[0021] Get historical voice interaction data;

[0022] Determine a plurality of historical subject words and a plurality of historical keywords corresponding to each historical subject word according to the historical voice interaction data, wherein all the historical subject words include the target subject word;

[0023] Set a preset search threshold for each historical subject word, wherein the preset search threshold for the target subject word is the target search threshold;

[0024] Generate a first image related to each historical subject word by using an image generation model, and use the first image as a first voice identifier corresponding to the historical subject word, wherein the first voice identifier corresponding to the target subject word is a target voice identifier;

[0025] A voice identification library is constructed based on all historical keywords, all historical subject words, all preset search thresholds, and all first voice identifications.

[0026] By adopting this embodiment, a voice identification library that meets the personalized recommendations of users can be established based on the voice interaction data of different users, thereby improving the user experience and speeding up the speed of determining the corresponding voice identification of keywords.

[0027] In some feasible embodiments, the controller determines a plurality of historical subject words and a plurality of historical keywords corresponding to each historical subject word according to the historical voice interaction data, and is specifically configured as follows:

[0028] Convert historical voice interaction data into historical text data;

[0029] Inputting the historical text data into a preset subject word distribution model to obtain a subject word distribution result, wherein the subject word distribution result includes a plurality of historical keywords classified according to the subject;

[0030] The subject identification is performed on the subject word distribution results to obtain historical subject words corresponding to the multiple historical keywords that have been classified.

[0031] By adopting this embodiment, historical keywords and corresponding historical subject words can be determined based on historical voice interaction data, topics and key points of user concern can be identified, and data support can be provided for the subsequent design of voice identification that is more in line with user interests, thereby improving user satisfaction.

[0032] In some feasible embodiments, the controller executes setting a preset search threshold for each historical subject word and is configured as follows:

[0033] Recording the number of occurrences of each historical subject word in the process of determining multiple historical subject words and multiple historical keywords corresponding to each historical subject word;

[0034] Determine the popularity level of historical keywords according to the number of occurrences of each historical keyword;

[0035] Set a preset search threshold for each historical keyword based on its popularity level.

[0036] By adopting this embodiment, the popularity level of the historical keyword can be determined according to the number of occurrences, the popularity of the keyword can be determined according to the popularity level, and then the preset search threshold can be determined according to the popularity, so that influential keywords can be quickly located and processed.

[0037] In some feasible embodiments, the controller searches for a target keyword in a pre-built voice identification library and determines a target subject word corresponding to the target keyword, and is specifically configured as follows:

[0038] In response to the target keyword existing in the voice identification library, a target subject word corresponding to the target keyword is determined in the voice identification library.

[0039] By adopting this embodiment, when there is a target keyword in the voice identification library, the corresponding subject word (ie, target subject word) can be successfully determined according to the target keyword, providing data support for the subsequent determination of the target voice identification.

[0040] In some feasible embodiments, the controller determines the target subject word corresponding to the target keyword in the voice identification library, and is specifically configured as follows:

[0041] In response to the target keyword corresponding to at least one first subject word in the voice identification library, the target keyword is determined based on the matching degree of each first subject word, and the first subject word with the highest matching degree is selected as the target subject word.

[0042] With this embodiment, when the target keyword corresponds to at least one first subject word, selecting the first subject word with the highest matching degree can improve the accuracy of subject word confirmation.

[0043] In some feasible embodiments, the controller searches for a target keyword in a pre-built voice identification library and determines a target subject word corresponding to the target keyword, and is further configured to:

[0044] In response to the target keyword not existing in the voice identification library, searching for a target synonymous keyword related to the target keyword, wherein the target synonymous keyword includes words that are semantically similar and / or identical to the target keyword;

[0045] In response to at least one target synonymous keyword existing in the voice identification library, a second subject word corresponding to the at least one target synonymous keyword is determined, and a target subject word corresponding to the target keyword is determined according to the second subject word.

[0046] With this embodiment, when the target keyword does not exist in the voice identification library, but the target synonymous keyword of the target keyword exists, the second subject word with the highest matching degree with the target synonymous keyword can be selected as the target subject word. In this way, when the keywords in the voice identification library are not completely the same as the keywords in the voice interaction, the appropriate target subject word can be successfully determined, and the speed and accuracy of the target subject word confirmation are guaranteed, so that the subsequent target voice identification can be smoothly obtained.

[0047] In some feasible embodiments, the controller is configured to:

[0048] In response to the fact that the target keyword and the target synonymous keyword do not exist in the voice identification library, a third subject word corresponding to the target keyword is determined by using a preset subject determination model;

[0049] Generate a second image related to the third keyword by using the image generation model, and use the second image as a second voice identifier corresponding to the third keyword;

[0050] Based on the target keyword, the third subject word, and the second voice identification, the voice identification library is updated, and a preset search threshold corresponding to the third subject word is set.

[0051] By adopting this embodiment, when there is no target keyword in the voice identification library and no target synonymous keyword of the target keyword, a corresponding third subject word and a second voice identification can be generated according to the target keyword. This can ensure the integrity of the voice identification library and make the voice identification library more in line with user interests.

[0052] In some feasible embodiments, the controller controls the display to display the target voice mark, and is configured to:

[0053] Controlling the audio receiver to receive a voice identification switching instruction;

[0054] In response to the voice identification switching instruction, the target voice identification is used as the virtual image of the voice assistant, and the display is controlled to display the target voice identification.

[0055] With this embodiment, when a voice identification switching instruction is received, the virtual image of the voice assistant is switched and displayed on the display, which can improve the user experience.

[0056] In some feasible embodiments, the controller controls the display to display the target voice identifier, and is further configured to:

[0057] Controlling the audio receiver to receive voice interaction instructions;

[0058] In response to the voice interaction instruction, the display is controlled to display a voice interaction interface, and a target voice identifier is displayed on the voice interaction interface.

[0059] By adopting this embodiment, the virtual image of the voice assistant is switched in the voice interaction interface to meet the user's demand for the virtual image of the voice assistant.

[0060] In a second aspect, an embodiment of the present application provides a method for recommending a voice identification, which is applied to a display device and includes:

[0061] Receive user's voice interaction data;

[0062] Determine the target keywords for voice interaction data;

[0063] Search for target keywords in the pre-built speech markup library and determine the target subject words corresponding to the target keywords;

[0064] In response to the number of times the target subject word is searched in the voice identification library exceeding the target search threshold corresponding to the target subject word, determining a target voice identification corresponding to the target subject word in the voice identification library;

[0065] Displays the target voice identifier.

[0066] It can be understood that the beneficial effects that can be achieved by the technical solution provided by the second aspect provided above can refer to the beneficial effects in the first aspect and any optional implementation method thereof, and will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0068] Figure 1 A schematic diagram of an operation scenario between a display device and a control device provided in an embodiment of the present application;

[0069] Figure 2 A schematic diagram of the hardware configuration of a display device provided in an embodiment of the present application;

[0070] Figure 3 A schematic diagram of software configuration of a display device provided in an embodiment of the present application;

[0071] Figure 4 A schematic diagram of a process for a display device according to an embodiment of the present application to determine a corresponding target voice identifier according to voice interaction data;

[0072] Figure 5 A schematic diagram of a process for a display device provided in an embodiment of the present application to construct a voice identification library based on historical voice interaction data;

[0073] Figure 6 A schematic diagram of a keyword table provided for an embodiment of the present application;

[0074] Figure 7 A schematic diagram of the perplexity of different numbers of topics provided in the embodiment of the present application;

[0075] Figure 8 A schematic diagram of the structure of the LDA model provided in the embodiment of the present application;

[0076] Fig. 9 A schematic diagram of the subject word distribution results provided in the embodiment of the present application;

[0077] Fig.10 A schematic diagram of the subject identification result provided in the embodiment of the present application;

[0078] Fig.11 A schematic diagram of a process for a display device according to an embodiment of the present application to determine a target voice identifier based on voice interaction data;

[0079] Fig.12 A schematic diagram of a flow chart of a display device communicating with a server to determine a target voice identifier based on voice interaction data provided in an embodiment of the present application. DETAILED DESCRIPTION

[0080] The following embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following embodiments do not represent all implementations consistent with the present application. They are only examples of systems and methods consistent with some aspects of the present application as detailed in the claims.

[0081] It should be noted that the brief description of terms in this application is only for the convenience of understanding the embodiments described below, and is not intended to limit the embodiments of this application. Unless otherwise specified, these terms should be understood according to their common and usual meanings.

[0082] The terms "first", "second", "third", etc. in the specification and claims of this application and the above drawings are used to distinguish similar or similar objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise noted. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances.

[0083] The terms "including" and "having" and any variations thereof are intended to cover but not be exclusive, for example, a product or device comprising a series of components is not necessarily limited to all the components explicitly listed, but may include other components not explicitly listed or inherent to these products or devices. The term "module" refers to any known or later developed hardware, software, firmware, artificial intelligence, fuzzy logic or combination of hardware and / or software code that is capable of performing the functions associated with the element.

[0084] The display device 200 provided in the embodiment of the present application generally refers to a device with image display and data processing capabilities, and can have various implementation forms, for example, it can be a television, a smart TV, a computer, a laser projection device, a display (monitor), an electronic whiteboard (electronic bulletin board), an electronic desktop (electronic table), a monitor, an advertising screen, a wearable device, a virtual reality device, an augmented reality device, etc.

[0085] Figure 1 Schematic diagram of an operation scenario between a display device 200 and a control device 100 provided in an embodiment of the present application. Figure 1 As shown, the user can operate the display device 200 through the terminal device 300 or the control device 100 .

[0086] In some embodiments, the control device 100 is used to receive operation instructions input by the user and convert the operation instructions into control instructions that the display device 200 can recognize and respond to. The control device 100 can be a remote control, a stylus, a handle, etc., wherein the communication between the remote control and the display device 200 includes infrared protocol communication or Bluetooth protocol communication, and other short-range communication methods, and the display device 200 is controlled wirelessly or wired. The user can input user instructions through buttons on the remote control, voice input, control panel input, etc. to control the display device 200.

[0087] In some embodiments, a terminal device 300 (such as a mobile terminal, a tablet computer, a computer, a laptop computer, etc.) can also be used as a control device to perform human-computer interaction between a user and the display device 200. The mobile terminal 300 can also be used as a communication device to establish a communication connection with the display device 200 and perform data interaction. The mobile terminal 300 can install the same software application as the display device 200, and then connect and communicate through a network communication protocol to achieve the purpose of one-to-one control operation and data communication. The audio and video content displayed on the mobile terminal 300 can also be transmitted to the display device 200 to achieve a synchronous display function.

[0088] In some embodiments, the mobile terminal 300 or other electronic devices may also simulate the functions of the control device 100 by running an application program for controlling the display device 200 .

[0089] In some embodiments, the display device 200 may also be controlled in a manner other than the control device 100 and the terminal device 300, for example, the user's voice command control may be directly received through a module for acquiring voice commands configured inside the display device 200, or the user's voice command control may be received through a voice control device provided outside the display device 200. The display device 200 may also receive user control through touch or gestures.

[0090] In some embodiments, the display device 200 also communicates data with the server 400 through various communication methods. The display device 200 may be allowed to communicate via a local area network (LAN), a wireless local area network (WLAN), and other networks. The server 400 may provide various content and interactions to the display device 200. The server 400 may be one cluster or multiple clusters, and may include one or more types of servers.

[0091] The display device 200 may provide a broadcast receiving television function, and may also additionally provide an intelligent network television function with a computer support function, including but not limited to network television, smart television, Internet Protocol television (IPTV), and the like.

[0092] Figure 2 FIG. 1 shows a hardware configuration block diagram of a control device 100 according to some embodiments. Figure 2 As shown, the control device 100 includes a controller 250, a communication interface 130, a user input / output interface 140, a memory, and a power supply. The control device 100 can receive input operation instructions from the user, and convert the operation instructions into instructions that the display device 200 can recognize and respond to, playing the role of an interactive intermediary between the user and the display device 200.

[0093] Figure 2 Some embodiments of the present application provide Figure 1 2 is a block diagram of the hardware configuration of the display device 200.

[0094] like Figure 2 As shown, the display device 200 may include at least one of a tuner 210, a communication device 220, a detector 230, an external device interface 240, a controller 250, a display 260, an audio output interface 270, a memory, a power supply, and a user interface.

[0095] In some embodiments, the tuner-demodulator 210 receives broadcast television signals via wired or wireless reception, and demodulates audio and video signals, such as EPG data signals, from a plurality of wireless or wired broadcast television signals.

[0096] In some embodiments, the communication device 220 is a component for communicating with an external device or server according to various communication protocol types. The display device 200 can be provided with a plurality of communication devices 220 according to the different communication modes supported. For example, when the display device 200 supports wireless network communication, the display device 200 can be provided with a communication device 220 including a WiFi function. When the display device 200 supports Bluetooth connection communication, the display device 200 needs to be provided with a communication device 220 including a Bluetooth function. The communication device 220 may include at least one of a Wifi module, a Bluetooth module, a wired Ethernet module, and other network communication protocol chips or near field communication protocol chips, and an infrared receiver. The display device 200 can establish the sending and receiving of control signals and data signals with the external control device 100 or the server 400 through the communication device 220.

[0097] The communication device 220 can enable the display device 200 to communicate with the external device or server 400 by wireless or wired connection. Among them, the wired connection can connect the display device 200 with the external device through components such as data cables and interfaces. The wireless connection can connect the display device 200 with the external device through wireless signals or wireless networks. The display device 200 can establish a connection relationship with the external device directly, or indirectly establish a connection relationship through a gateway, a router, a connection device, etc.

[0098] In some embodiments, the detector 230 is used to collect signals of the external environment or external interaction. For example, the detector 230 includes a light receiver, a sensor for collecting the intensity of ambient light; or, the detector 230 includes an image collector, such as a camera, which can be used to collect external environment scenes, user attributes or user interaction gestures; or, the detector 230 includes a sound collector, such as a microphone, etc., for receiving external sounds.

[0099] In some embodiments, the external device interface 240 may include, but is not limited to, any one or more of the following interfaces: a high-definition multimedia interface (HDMI), an analog or digital high-definition component input interface (component), a composite video input interface (CVBS), a USB input interface (USB), an RGB port, etc. It may also be a composite input / output interface formed by the above multiple interfaces.

[0100] In some embodiments, the controller 250 controls the operation of the display device 200 and responds to user operations through various software control programs stored in the memory. The controller 250 controls the overall operation of the display device 200. For example, in response to receiving a user command for selecting a UI object to be displayed on the display 260, the controller 250 can perform operations related to the object selected by the user command.

[0101] In some embodiments, the controller 250 includes at least one of a central processing unit (CPU), a video processor, an audio processor, a graphics processing unit (GPU), a random access memory (RAM), a read-only memory (ROM), a first interface to an nth interface for input / output, a communication bus, etc. The controller 250 controls the operation of the display device and responds to the user's operation through various software control programs stored in the memory. The controller 250 controls the overall operation of the display device 200.

[0102] In some embodiments, the controller 250 and the tuner-demodulator 210 may be located in different separate devices, that is, the tuner-demodulator 210 may also be located in an external device of the main device where the controller 250 is located, such as an external set-top box.

[0103] In some embodiments, the user may input a user command in a graphical user interface (GUI) displayed on the display 260, and the user input interface receives the user input command through the graphical user interface (GUI). Alternatively, the user may input a user command by inputting a specific sound or gesture, and the user input interface receives the user input command by recognizing the sound or gesture through a sensor.

[0104] In some embodiments, the display 260 includes a display screen component for presenting a picture, and a driving component for driving an image display. The display 260 is used to receive an image signal output from the controller 250 for display. For example, the display 260 can be used to display video content, image content, and components of a menu control interface and a user control UI interface.

[0105] In some embodiments, the "user interface" is a medium interface for interaction and information exchange between an application or operating system and a user, which realizes the conversion between the internal form of information and the form acceptable to the user. The commonly used form of user interface is the graphical user interface (GUI), which refers to a user interface related to computer operation displayed in a graphical manner. It can be an interface element such as an icon, window, control, etc. displayed on the display screen of an electronic device, where the control can include icons, buttons, menus, tabs, text boxes, dialog boxes, status bars, navigation bars, widgets, and other visual interface elements.

[0106] In some embodiments, the display 260 may be a liquid crystal display, an OLED display, or a projection display, and may also be a projection device and a projection screen.

[0107] In some embodiments, the audio output device 270 may be a local speaker of the display device 200, or may be an external audio output device of the display device 200. In particular, for the external audio output device of the display device 200, the display device 200 may also be provided with an external audio output terminal, and the audio output device may be connected to the display device 200 through the external audio output terminal to output the sound of the display device 200.

[0108] In some embodiments, the user interface may be used to receive a control signal from the control device 100 (eg, an infrared remote controller, etc.).

[0109] In order to perform user interaction, in some embodiments, the display device 200 may run an operating system. The operating system is a computer program for managing and controlling hardware resources and software resources in the display device 200. The operating system can control the display device to provide a user interface. For example, the operating system can directly control the display device to provide a user interface, or provide a user interface by running an application program. The operating system also allows the user to interact with the display device 200.

[0110] It should be noted that the operating system can be a native operating system based on a specific operating platform, a third-party operating system deeply customized based on a specific operating platform, or an independent operating system specially developed for the display device.

[0111] The operating system can be divided into different modules or layers according to the functions implemented, such as Figure 3 As shown, in some embodiments, the system is divided into four layers, from top to bottom, namely, the application layer (Applications) layer (referred to as "application layer"), the application framework layer (Application Framework) layer (referred to as "framework layer"), the system library layer and the kernel layer.

[0112] In some embodiments, the application layer is used to provide services and interfaces for applications so that the display device 200 can run applications and interact with users based on the applications. At least one application can be run in the application layer, and these applications can be window programs, system settings programs, clock programs, etc. that come with the operating system; they can also be applications developed by third-party developers. In specific implementations, the application packages in the application layer are not limited to the above examples.

[0113] The framework layer provides application programming interfaces (APIs) and programming frameworks for applications. The application framework layer includes some predefined functions. The application framework layer is equivalent to a processing center that determines the actions that applications in the application layer take. Through the API interface, applications can access system resources and obtain system services during execution.

[0114] Figure 3 A schematic diagram of the software configuration of the display device 200 provided in an embodiment of the present application.

[0115] like Figure 3 As shown, in the embodiment of the present application, the application framework layer includes a view system, managers, content providers, etc., wherein the view system can design and implement the interface and interaction of the application, and the view system includes lists, grids, text boxes, buttons, etc. The manager includes at least one of the following modules: an activity manager for interacting with all activities running in the system; a location manager for providing system services or applications with access to system location services; a package manager for retrieving various information related to the application package currently installed on the device; a notification manager for controlling the display and clearing of notification messages; and a window manager for managing icons, windows, toolbars, wallpapers, and desktop widgets on the user interface.

[0116] In some embodiments, the activity manager is used to manage the life cycle of each application and the usual navigation back function, such as controlling the exit, opening, and back of the application. The window manager is used to manage all window programs, such as obtaining the display screen size, determining whether there is a status bar, locking the screen, capturing the screen, and controlling the display window changes, for example, reducing the display window, shaking the display, distorting the display, etc.

[0117] In some embodiments, the system runtime layer can provide support for the framework layer. When the framework layer is used, the operating system will run the instruction library contained in the system runtime layer, such as the C / C++ instruction library, to implement the functions to be implemented by the framework layer.

[0118] In some embodiments, the kernel layer is a functional layer between the hardware and software of the display device 200. The kernel layer can implement functions such as hardware abstraction, multitasking, and memory management. Figure 3 As shown, the kernel layer may be configured with hardware drivers, and the drivers included in the kernel layer may be at least one of the following drivers: audio driver, display driver, Bluetooth driver, camera driver, WIFI driver, USB driver, HDMI driver, sensor driver (such as fingerprint sensor, temperature sensor, pressure sensor, etc.) and power driver, etc.

[0119] It should be noted that the above example is only a simple division of the operating system functions and does not constitute a limitation on the specific operating system form of the display device 200 in the embodiment of the present application. Depending on factors such as the function of the display device and the type of operating system, the number of levels and specific level types contained in the operating system may be expressed in other forms.

[0120] In some embodiments, the display device 200 may also have a voice interaction function. To this end, a voice application may be integrated in the display device 200. The voice application is an artificial intelligence application integrated on the display device. The user can wake up the voice assistant by shouting a wake-up word, and then send a voice command to the voice application. The voice application can recognize the voice command and execute the interactive effect corresponding to the voice command. The voice application can respond to a variety of voice commands, such as a voice start command, a voice play command, a voice pause command, a voice switch command, or a voice wake-up command.

[0121] Based on the above hardware framework and software framework, in some embodiments, the user can input voice interaction data through an audio receiver coupled to the display, so that the display device 200 displays the corresponding voice identification. For example, the user sends voice interaction data with the content of "braised pork tutorial" to the display device, and the display device 200 receives the voice interaction data based on the audio receiver, and displays the voice identification related to the voice interaction data "braised pork tutorial", so that the voice identification serves as the virtual image of the voice assistant.

[0122] Voice assistants for smart TVs have become an important part of modern smart homes. They not only provide basic voice control functions, such as searching for programs and adjusting the volume, but also enable more advanced operations through voice interaction, such as controlling other smart devices in the home and querying information. With the continuous advancement of technology, the application scenarios of voice assistants are becoming increasingly diverse, ranging from simple audio responses to complex task management and information processing, greatly improving the user experience.

[0123] However, the existing virtual images of voice assistants have certain limitations in image design. In most cases, the virtual image of a voice assistant is just a simple icon or has no specific image at all. This makes the voice assistant more of a tool rather than a part of the user's personalized experience, lacking appeal and affinity to users. In addition, a single design can also cause users to develop aesthetic fatigue after using it for a period of time, reducing their interest in and frequency of use of voice assistants.

[0124] In view of the above problems, the embodiment of the present application provides a display device that can determine the subject words that the user is interested in based on the user's voice interaction data, and then determine the relevant voice identifier based on the subject words, and use the voice identifier as the virtual image of the voice assistant. Here, the voice identifier can be an image, a virtual character image or an animation related to the subject word, and the present application does not limit it. Among them, the image can be a real or abstract illustration related to the subject, the virtual character image can be a virtual character dressed up in a related to the subject word, and can be a cartoon style, a realistic style or a 3D model image, and the animation can be a dynamic icon or image related to the subject word, as well as an interactive animation, etc.

[0125] In some embodiments, Figure 4 A flow chart of a display device according to an embodiment of the present application determining a corresponding target voice identifier according to voice interaction data. Figure 4 As shown, some embodiments of the present application provide a display device 200, and the display device 200 includes: a display 260, an audio receiver and a controller 250. Among them, the audio receiver is coupled to the display and is used to receive audio data. Here, the audio data can be audio data played by the display device 200, and can also be audio data input by the user. The audio data played by the display device 200 may include audio of TV programs, music from music streaming media, audio signals transmitted by an external audio transmission device, and prompt sounds and menu sound effects during TV operation. The audio data input by the user may include voice commands and feedback audio for the display device to interact with the user. In addition, the controller 250, coupled to the display 260 and the audio receiver, can be configured to perform the following program steps:

[0126] S100: Controlling an audio receiver to receive user's voice interaction data.

[0127] In some embodiments, the user can wake up the voice assistant by pressing the voice button on the remote control, saying a specific wake-up word (such as "Hello, TV"), and using the voice control icon of the user interface, thereby activating the audio receiver to receive the user's voice interaction data. For example, after waking up the voice assistant, the user can send voice interaction data, such as "play XX program", "open video software", etc., and the audio receiver can receive the user's voice interaction data.

[0128] S200: Determine target keywords for voice interaction data.

[0129] In some embodiments, after receiving the user's voice interaction data, the controller may preprocess the voice interaction data to improve the accuracy of subsequent voice recognition. The preprocessing may include denoising, echo cancellation, gain control, etc. Then, the preprocessed voice interaction data may be used to extract target keywords, and then determine the target subject words and target voice identifiers corresponding to the voice interaction data. By extracting the target keywords of the voice interaction data, the subject words corresponding to the voice interaction data can be more accurately understood and determined, and then a voice identifier that meets the user's interests can be provided.

[0130] In some embodiments, the controller executes step S200, which is specifically configured to: convert the voice interaction data into text data; preprocess the text data; and extract target keywords from the preprocessed text data based on a preset word extraction model.

[0131] Here, the controller may convert the voice interaction data into text data. Specifically, the voice features of the voice interaction data may be extracted using voice recognition technology, and the voice features may include Mel-frequency cepstral coefficients (MFCCs), linear predictive coding (LPC) coefficients, spectrum centroid, etc.

[0132] Furthermore, deep learning models such as acoustic models and language models can be used to identify speech features and understand language patterns, and convert speech interaction data into text data. Specifically, acoustic models such as recurrent neural networks (RNN), long short-term memory networks (LSTM) or convolutional neural networks (CNN) can be used to identify the relationship between speech features and phonemes (the smallest unit of speech), and then use language models to determine the word sequence probability of phonemes. Afterwards, based on the word sequence probability of phonemes, the decoder is used to determine the most likely word sequence. Finally, the word sequence output by the decoder is converted into text data.

[0133] Next, the controller can preprocess the text data. The preprocessing can include tokenizing the text data and removing stop words. Specifically, tokenizing the text data means splitting the continuous text data into words, phrases, or other meaningful elements. Then, stop words in the tokenized text data are removed. Stop words can include words that frequently appear in the text but contribute little to the meaning of the text, such as "de", "he", "shi", etc. By preprocessing the text data, a cleaner text form can be obtained, which helps with subsequent natural language processing tasks, such as sentiment analysis, topic modeling, keyword extraction, etc.

[0134] Finally, the controller can extract the target keywords of the preprocessed text data based on a preset keyword extraction model. Here, the preset keyword extraction model can use keyword extraction models in existing natural language processing libraries, such as the Term Frequency-Inverse Document Frequency (TF-IDF) model, text ranking, Bidirectional Encoder Representations from Transformers (BERT), etc., or can also be obtained by training other large models according to a specific text training set. This application places no restrictions on this. Further, the target keywords of the text data are extracted using the preset keyword extraction model, and post-processing is performed on the extracted target keywords, such as removing duplicate words and further screening according to requirements.

[0135] According to the above technical solution, by extracting the target keywords from the preprocessed text data, data support can be provided for subsequent natural language processing tasks (such as determining the target topic words corresponding to the voice interaction data).

[0136] S300: Search for the target keywords in the pre-constructed voice identification library and determine the target topic words corresponding to the target keywords.

[0137] In some embodiments, a voice identification library can be pre-constructed. The voice identification library is a database or data structure that includes keywords, the topic words corresponding to the keywords, and the voice identifications corresponding to the topic words. Each keyword has one or more associated topic words. The voice identification library can be constructed based on historical keywords, historical topic words, and the first voice identification. All historical topic words include the target topic words. The controller can search in the voice identification library according to the target keywords and, based on the found target keywords, find the corresponding target topic words.

[0138] In some embodiments, Figure 5 is a schematic flowchart of a display device provided by an embodiment of the present application for constructing a voice identification library according to historical voice interaction data. As Figure 5As shown, the controller executes the construction of the voice identification library and is configured to execute the following program steps:

[0139] S501: Acquire historical voice interaction data.

[0140] In some embodiments, the controller can obtain historical voice interaction data of multiple other display devices. Specifically, the current display device can establish a data sharing agreement with multiple other display devices, and the data sharing agreement can determine the type of data, access rights and sharing conditions, etc. Based on the data sharing agreement, the controller can discover other display devices on the network, identify the identities of other display devices, and establish a trust relationship with other display devices. After that, the controller can send a data acquisition request to other display devices to obtain the historical voice interaction data of other display devices. In addition, the current display device can also be connected to the server and send a data acquisition request to the server. After receiving the data acquisition request, the server forwards the data acquisition request to the corresponding display device according to the request content to obtain the historical voice interaction data of other display devices. Furthermore, the historical voice interaction data of each display device is taken as a document, and the display device can obtain multiple documents.

[0141] S502: Determine a plurality of historical subject words and a plurality of historical keywords corresponding to each historical subject word according to the historical voice interaction data.

[0142] In some embodiments, the controller executes step S502, which is specifically configured to: convert historical voice interaction data into historical text data; input the historical text data into a preset subject word distribution model to obtain a subject word distribution result, wherein the subject word distribution result includes multiple historical keywords completed according to subject classification; perform subject identification on the subject word distribution result to obtain historical subject words corresponding to the multiple historical keywords completed by classification.

[0143] In some embodiments, the historical voice interaction data in each document is first converted into historical text data, and each historical text data is preprocessed, including but not limited to segmenting the historical text data using word segmentation, removing stop words, such as removing characters that cannot express the subject content, such as a large number of numbers, English, punctuation marks, or meaningless Chinese words. After that, a keyword table is set to retain the preprocessed special nouns related to the subject. Figure 6 Schematic diagram of a keyword table provided in an embodiment of the present application. Figure 6 As shown, the first column is the historical text data obtained based on the user's historical voice interaction data, and the second column is the vocabulary list after preprocessing the historical text data, which may include: play, food, volume, football, World Cup, etc.

[0144] In some embodiments, after obtaining the historical text data, the preprocessed historical text can be analyzed for hot topic words using the LDA model to obtain the topic word distribution results. Before using the LDA model for topic word analysis, the optimal number of topics for the LDA model can be determined. Specifically, the LDA model can be used to calculate the perplexity under different numbers of topics, and then determine the optimal number of topics. Perplexity is an indicator to measure the quality of the sample predicted by the probability model, which can be used to evaluate the prediction ability of the LDA model for new texts. The calculation formula of perplexity is as follows:

[0145]

[0146] in, is the perplexity value, M is the number of documents, N m is the number of words in document m, is the probability value of the word in the mth document. Here, the smaller the perplexity value, the better the LDA model classifies text data and the stronger its generalization ability. Figure 7 A schematic diagram of the perplexity of different numbers of topics provided in the embodiment of the present application, the perplexity of different numbers of topics changes as shown in FIG. Figure 7 As shown in the figure, the perplexity decreases with the increase of the number of topics. When the number of topics of the LDA model is 12, the perplexity value is the lowest, and the perplexity value when the number of topics is greater than 12 has no obvious change compared with the perplexity value when the number of topics is equal to 12. Therefore, the optimal number of topics for the LDA model can be set to 12.

[0147] In some embodiments, Figure 8 This is a schematic diagram of the structure of the LDA model provided in the embodiment of the present application. Figure 8 As shown, the input variable α represents the feature vector, the input variable β represents the category label, θ is the multinomial distribution of the topic, Z is the extracted topic word, and W is the predicted value finally obtained based on the topic word Z. In the process of training the LDA model, the training sample data is first obtained. The training sample data may include multiple sample speech text data, and the topic word distribution results of the sample speech text data. Then, the LDA model is trained using the training sample data, and the number of topics K of the LDA model is set to 12, the initial parameters α=0.1, β=0.01, and the number of iterations is 20 times. The LDA model is trained based on the above parameters to obtain the trained LDA model. The pre-processed historical text data is processed based on the trained LAD model to obtain the topic word distribution result. Fig. 9 This is a schematic diagram of the subject word distribution results provided in the embodiment of this application. Fig. 9As shown, the results of the subject word distribution can be represented by a word list including the subject number and the keyword distribution. The first column of the word list is the subject number, which can be 0 to 11. The number of subject numbers is 12, which is the same as the optimal number of subjects mentioned above. The second column of the word list is the keyword distribution, and each subject number corresponds to a series of keywords. Based on the subject word distribution results, the top 50 words with the highest probability under each subject can be extracted, and each word can be sorted from large to small according to the probability distribution as a representative of the meaning of the subject, and the subject can be identified in combination with these words.

[0148] In some embodiments, after determining the subject word distribution results, the subject word distribution results can be subject identified and the subject can be marked. Specifically, the subject words corresponding to each subject can be determined based on the keywords corresponding to each subject. Here, the keywords ranked in the front can be used as the basis for determining the subject words, and then the subject words can be determined based on the keywords. Furthermore, the subject can be marked based on the results of subject identification. Fig.10 This is a schematic diagram of the results of the subject recognition provided by the embodiment of the present application. Fig.10 As shown in FIG. 1 , the result of topic identification can be composed of a word list including topics, topic numbers, and keyword distribution. It is also necessary to ignore topics with repeated expressions and topics with vague content, and then determine the final topic words. In addition, custom topics and related keywords can be added as needed.

[0149] S503: Setting a preset search threshold for each historical keyword.

[0150] In some embodiments, the controller executes step S503 and is configured to: record the number of occurrences of each historical subject word in the process of determining multiple historical subject words and multiple historical keywords corresponding to each historical subject word; determine the popularity level of the historical subject word according to the number of occurrences of each historical subject word; and set a preset search threshold for each historical subject word based on the popularity level.

[0151] Here, for each historical keyword, the number of times it appears in all historical text data is recorded, and then its popularity level is calculated based on the number of times the historical keyword appears. The popularity level can be defined based on the percentage of occurrences, ranking, or other evaluation methods, and this application does not impose any restrictions. Next, according to the popularity level of each historical keyword, a corresponding preset search threshold is set. Here, the higher the popularity level of the historical keyword, the more the historical keyword is liked by users, and the corresponding preset search threshold can be set lower to make it easier for users to reach the preset search threshold. In addition, the preset search threshold of the target keyword is the target search threshold.

[0152] According to the above technical solution, the number of occurrences of each historical keyword can be recorded, and a preset search threshold can be set according to its popularity level, so as to effectively filter and analyze when querying historical data.

[0153] S504: Generate a first image related to each historical keyword using an image generation model, and use the first image as a first voice identifier corresponding to the historical keyword.

[0154] In some embodiments, the image generation model can be trained based on the text or keyword to generate the corresponding image, and the image generation model can include deep learning models such as generative adversarial networks and variational autoencoders. Here, the trained image generation model can be used to generate a corresponding first image according to each historical keyword. Afterwards, the generated first image is associated with the historical keyword as the first voice identifier corresponding to the historical keyword, and stored in a database or data structure for subsequent query and use. Furthermore, the first voice identifier corresponding to the target keyword is the target voice identifier.

[0155] In some embodiments, the voice identification corresponding to the keyword can also be determined according to user needs, that is, the content subject needs of the user can be confirmed based on the user data obtained by the display device, and the corresponding voice identification can be selected and saved in the voice identification library based on the identified subject of the user's interest content.

[0156] S505: Construct a voice identification library according to all historical keywords, all historical subject words, all preset search thresholds, and all first voice identifications.

[0157] In some embodiments, the voice identification library is shown in Table 1. As can be seen from Table 1, the voice identification library may include a voice identification column, a subject word column, a keyword column, and a preset search threshold column. Specifically, all historical keywords, historical subject words, preset search thresholds, and the first voice identification may be collected. Then, these data are cleaned and preprocessed to ensure the quality and consistency of the data. Afterwards, the voice identification library is stored based on a preset data structure, and the data structure may be a database, a file system, etc. Finally, the preprocessed data is stored in a designed data structure to obtain a voice identification library, and an index is established for the voice identification library for quick retrieval and query.

[0158]

[0159]

[0160] Table 1

[0161] In some embodiments, the controller performs step S300, and is specifically configured to: in response to the existence of the target keyword in the voice identification library, determine the target subject word corresponding to the target keyword in the voice identification library. Here, the target keyword can be used as a query condition to search for records related to the target keyword in the voice identification library. In the search results of the target keyword, find the record matching the target keyword, and then extract the target subject word from the matching record.

[0162] According to the above technical solution, by determining the target subject word corresponding to the target keyword in the voice identification library, information related to the target keyword can be quickly retrieved, reducing the time consumption of fuzzy matching.

[0163] In some embodiments, the controller executes step S300 and is also configured to: in response to the target keyword not existing in the voice identification library, search for target synonymous keywords related to the target keyword, wherein the target synonymous keywords include words that are semantically similar and / or identical to the target keyword; in response to the presence of at least one target synonymous keyword in the voice identification library, determine a second subject word corresponding to the at least one target synonymous keyword, and determine a target subject word corresponding to the target keyword based on the second subject word.

[0164] Here, when the target keyword does not exist in the voice identification library, the search scope can be expanded by finding target synonymous keywords with similar or identical meanings to the target keyword to improve the matching degree and user experience. Specifically, the target synonymous keywords of the target keyword can be determined by a synonym dictionary, semantic analysis, or machine learning model. For example, the target keyword determined according to the voice interaction data is "landline", and "landline" does not exist in the keyword column of the voice identification library. Then, the target synonymous keywords of "landline" can be determined as "landline", "desk phone", "extension", etc. according to methods such as a synonym dictionary, semantic analysis, or a machine learning model.

[0165] Furthermore, the target synonymous keyword is used as a query condition, and the records related to the target synonymous keyword are searched in the voice identification library. If there is at least one target synonymous keyword, the corresponding second subject word is determined according to the target synonymous keyword, and the target subject word corresponding to the target keyword is determined according to the second subject word. For example, when there is only one target synonymous keyword such as "landline" with a similar meaning to "landline" in the voice identification library, and there is only one second subject word corresponding to the target synonymous keyword, the second subject word "communication equipment" corresponding to the target synonymous keyword "landline" is used as the target subject word.

[0166] When there is a target synonymous keyword in the voice identification library, and there are multiple second subject words corresponding to the target synonymous keyword, the matching degree of each second subject word with the target synonymous keyword is calculated, and the second subject word with the highest matching degree is used as the target subject word. For example, there is a target synonymous keyword "landline" in the voice identification library, and "landline" corresponds to multiple second subject words, "communication equipment", "household appliances" and "office equipment". At this time, the matching degree between "landline" and "communication equipment", "household appliances" and "office equipment" needs to be calculated respectively, and the calculation method of the matching degree may include word frequency, correlation between keywords and subject words, context matching, etc., which is not limited in this application. Here, the corresponding matching degrees can be calculated according to the correlation between "landline" and "communication equipment", "household appliances" and "office equipment", respectively: 100, 70, 90, then the second subject word "communication equipment" with the highest matching degree with "landline" can be used as the target subject word.

[0167] When there are multiple target synonymous keywords in the voice identification library, and the second subject word corresponding to each target synonymous keyword is the same, the second subject word can be used as the target subject word. For example, there are target synonymous keywords "landline phone", "desk phone", and "extension" in the voice identification library. "Landline phone", "desk phone", and "extension" all correspond to the same second subject word "communication equipment", and "communication equipment" can be used as the target subject word.

[0168] When there are multiple target synonymous keywords in the voice identification library, and the second subject words corresponding to each target synonymous keyword are different, the matching degree between each target synonym and the corresponding second subject word is calculated, and the second subject word with the highest matching degree is used as the target subject word. For example, there are target synonymous keywords "landline phone" and "desk phone" in the voice identification library. The second subject word corresponding to "landline phone" is "communication equipment", and the second subject word corresponding to "desk phone" is "household appliances". The matching degree can be determined as 100 based on the correlation between "landline phone" and "communication equipment", and the matching degree is calculated as 70 based on the correlation between "desk phone" and "household appliances", then "communication equipment" with a higher matching degree is used as the target subject word.

[0169] It is understandable that when there are multiple target synonymous keywords in the speech identification library and each target synonymous keyword corresponds to multiple second subject words, the matching degree between each target synonymous keyword and the corresponding multiple second subject words can be calculated, and the second subject word with the highest matching degree among all target synonymous keywords and the corresponding second subject words can be taken as the target subject word.

[0170] In some embodiments, the controller executes determination of a target subject word corresponding to a target keyword, and is also configured to: in response to the absence of a target keyword and a target synonymous keyword in a voice identification library, determine a third subject word corresponding to the target keyword using a preset subject determination model; generate a second image related to the third subject word using an image generation model, and use the second image as a second voice identification corresponding to the third subject word; based on the target keyword, the third subject word, and the second voice identification, update the voice identification library, and set a preset search threshold corresponding to the third subject word.

[0171] Here, when neither the target keyword nor the target synonym keyword exists in the voice identification library, a new third subject word can be generated using the preset topic determination model, and the image generation model can be used to generate a second image related to the third subject word, and the second image is used as the target voice identification corresponding to the target keyword. Specifically, a new subject word is generated according to the context information of the target keyword using the preset topic determination model, and a related second image is generated according to the generated third subject word using the image generation model, and the generated second image is used as the second voice identification corresponding to the third subject word; then, the target keyword, the third subject word and the second voice identification are added to the voice identification library. A preset search threshold is also set for the third subject word according to the generation method and context information of the third subject word.

[0172] Exemplarily, when the target keyword "landline" does not exist in the voice identification library, nor does it exist target synonymous keywords such as "landline", "desk phone", "extension", etc., the preset topic determination model can be used to generate a corresponding new third subject word "communication equipment" based on the target keyword. By inputting "communication equipment" into the image generation model, a third image related to the communication equipment can be obtained, such as an image of a phone or an image of a landline. Furthermore, a preset search threshold for "communication equipment" can be set. Afterwards, "landline", "communication equipment", the third image (i.e., the image of the phone or the image of the landline) and the preset search threshold for "communication equipment" are added to the voice identification library.

[0173] According to the above technical solution, in response to the absence of target keywords and target synonymous keywords in the voice identification library, a preset topic determination model can be used to determine the third subject word corresponding to the target keyword, and the voice identification library can be updated in real time to ensure the integrity of the voice identification library and make the data in the voice identification library more in line with the user's interests, providing a better user experience.

[0174] S400: In response to the number of times the target subject word is searched in the voice identification library exceeding the target search threshold corresponding to the target subject word, determining a target voice identification corresponding to the target subject word in the voice identification library.

[0175] In some embodiments, it is first determined whether the number of searches for the target subject word in the voice identification library exceeds its corresponding target search threshold. If the number of searches exceeds the target search threshold, the target voice identification corresponding to the target subject word is determined.

[0176] In some embodiments, the controller executes determining a target subject word corresponding to a target keyword in a voice identification library, and is specifically configured as follows: in response to the target keyword corresponding to at least one first subject word in the voice identification library, determining the target keyword based on a matching degree of each first subject word, and selecting a first subject word with the highest matching degree as the target subject word.

[0177] Here, when the target keyword corresponds to only one first subject word, the first subject word can be used as the target subject word. When the target keyword corresponds to multiple first subject words, the matching degree of each matching first subject word and the target keyword can be calculated. Among all the matching first subject words, the first subject word with the highest matching degree is selected as the target subject word. For example, when the target keyword is "landline", there is only one first subject word "communication equipment" corresponding to "landline" in the voice identification library, then "communication equipment" is used as the target subject word. When the target keyword is "landline", there are multiple first subject words (such as "communication equipment" and "office equipment") corresponding to "landline" in the voice identification library, then the matching degree of "landline" is calculated according to the correlation between "communication equipment" and "office equipment". Specifically, according to the correlation between "landline" and "communication equipment", the matching degree is calculated to be 100, and according to the correlation between "landline" and "office equipment", the matching degree is calculated to be 80, then "communication equipment" with a higher matching degree is used as the target subject word.

[0178] According to the above technical solution, a first keyword with a higher matching degree can be selected as a target keyword to accurately understand and respond to the user's input, thereby improving user experience and satisfaction.

[0179] S500: Control the display to display the target voice mark.

[0180] In some embodiments, the controller executes step S500 and is configured to: control the audio receiver to receive a voice identification switching instruction; in response to the voice identification switching instruction, use the target voice identification as the virtual image of the voice assistant, and control the display to display the target voice identification.

[0181] Here, the audio receiver can be controlled to receive the voice identification switching instruction input by the user, parse the voice identification switching instruction, and determine the target voice identification. Then, in response to the voice identification switching instruction, the target voice identification is used as the virtual image of the voice assistant, and the display is controlled to display the target voice identification.

[0182] In some embodiments, the controller executes step S500 and is further configured to: control the audio receiver to receive the voice interaction instruction; in response to the voice interaction instruction, control the display to display the voice interaction interface, and display the target voice identifier on the voice interaction interface. Here, after receiving and identifying the voice interaction instruction, the controller controls the display to display the voice interaction interface, and displays the target voice identifier, that is, the virtual image of the voice assistant, on the voice interaction interface. The voice interaction interface can be a graphical user interface for providing visual feedback to let the user know that the device is responding to its voice instruction.

[0183] According to the embodiments of the present application, the present application provides a display device, including: a display; an audio receiver coupled to the display for receiving audio data; a controller coupled to the display and the audio receiver, and configured to: control the audio receiver to receive the user's voice interaction data; determine the target keyword of the voice interaction data; search for the target keyword in the pre-built voice identification library, and determine the target subject word corresponding to the target keyword; in response to the number of times the target subject word is searched in the voice identification library exceeding the target search threshold corresponding to the target subject word, determine the target voice identification corresponding to the target subject word in the voice identification library; control the display to display the target voice identification. The above-mentioned display device allows the user to identify the topic of interest to the user according to the voice interaction data when using voice interaction, and recommend the voice identification corresponding to the topic as the virtual image of the voice assistant to the user, thereby enhancing the user's sense of interaction, sense of gain and sense of participation in using voice, so that the user can have a better experience.

[0184] In some embodiments, Fig.11 A flow chart of a display device according to an embodiment of the present application determining a target voice identifier based on voice interaction data. Fig.11 As shown, the display device includes an audio receiver, a page display module, a speech-to-text module, a keyword determination module, a subject word determination module, and a speech identification library. Specifically, the audio receiver receives the user's voice interaction data and sends the speech interaction data to the speech-to-text module. The speech-to-text module converts the speech interaction data into text data and sends the text data to the keyword determination module. The keyword determination module can use a preset model to determine the keywords in the text data and send the keywords to the subject word determination module. The subject word determination module can determine the corresponding subject words based on the keywords and send the subject words to the speech identification library. The speech identification library searches for the corresponding target speech identification based on the subject words and sends the target speech identification to the page display module. The page display module can display the target speech identification for the user to select and switch.

[0185] In some embodiments, Fig.12A flow chart of the communication between the display device and the server provided in the embodiment of the present application to determine the target voice identification according to the voice interaction data. Fig.12 As shown, the present application also provides a reality device 200, including: a display 260; an audio receiver 280, coupled to the display, for receiving audio data; a controller 250, configured to:

[0186] Controlling the audio receiver to receive the user's voice interaction data;

[0187] In response to receiving the voice interaction data, forwarding the voice interaction data to the server, so that the server determines the target keyword of the voice interaction data and determines the target voice identifier corresponding to the target subject word according to the voice interaction data;

[0188] The target voice identifier is received from the server, and the display is controlled to display the target voice identifier.

[0189] The display device 200 may be a smart TV with high computing power, and may obtain voice interaction data and determine the target voice identifier through its own computing chip. However, in some application scenarios, the display device 200 may be a low computing power device such as an advertising machine, and the low computing power terminal device may determine the target voice identifier slowly or may not be equipped with a computing model.

[0190] Based on the above problems, in some embodiments, the display device 200 can also interact with the server in real time to determine the target voice identifier through the server, receive the target voice identifier sent by the server, and then display the target voice identifier through the display 260 connected to the display device 200. The server can be a server arranged in the cloud, a server arranged in the local network, or a server searched based on edge computing. The method of communicating with the server to obtain the target voice identifier alleviates the problem that the display device 200 cannot determine the target voice identifier in time according to the voice interaction data input by the user due to insufficient computing power.

[0191] In some embodiments, the present application further provides a server, including: a communicator, configured to communicate with a terminal device; a processor, configured to:

[0192] Receiving voice interaction data forwarded by a terminal device; determining target keywords of the voice interaction data;

[0193] Search for target keywords in the pre-built speech markup library and determine the target subject words corresponding to the target keywords;

[0194] In response to the number of times the target subject word is searched in the voice identification library exceeding the target search threshold corresponding to the target subject word, determining a target voice identification corresponding to the target subject word in the voice identification library;

[0195] The target voice identifier is sent to the display device so that the display device controls the display to display the target voice identifier.

[0196] The server may receive the voice interaction data forwarded by the display device 200, and determine the target voice identifier according to the voice interaction data. And send the target voice identifier to the display device 200, so that the display device 200 displays the target voice identifier for the user to select. This alleviates the problem that it is difficult for the low-computing display device 200 to determine the target voice identifier.

[0197] In some embodiments, the present application further provides a voice identification recommendation method, which is applied to a display device with high computing power, including:

[0198] Receive user's voice interaction data;

[0199] Determine the target keywords for voice interaction data;

[0200] Search for target keywords in the pre-built speech markup library and determine the target subject words corresponding to the target keywords;

[0201] In response to the number of times the target subject word is searched in the voice identification library exceeding the target search threshold corresponding to the target subject word, determining a target voice identification corresponding to the target subject word in the voice identification library;

[0202] Displays the target voice identifier.

[0203] The above method is applied to a display device 200 equipped with a computing chip with high computing power, such as a smart TV, a computer, etc. The voice interaction data input by the user can be received through the audio receiver user interface, and the controller of the display device 200 determines the target voice identifier according to the voice interaction data.

[0204] It is understandable that, in some embodiments, a voice identification recommendation method for a display device with low computing power is also provided. The controller 250 of the display device 200 forwards the voice interaction data input by the user to the server, and the server performs a process of determining the target voice identification and sends the target voice identification to the display device 200. The display device 200 displays the target voice identification through the display 260.

[0205] According to the embodiments of the present application, the present application provides a method for recommending a voice identification, which is applied to a display device, including: receiving a user's voice interaction data; determining a target keyword of the voice interaction data; searching for the target keyword in a pre-built voice identification library, and determining a target subject word corresponding to the target keyword; in response to the number of searches for the target subject word in the voice identification library exceeding the target search threshold corresponding to the target subject word, determining a target voice identification corresponding to the target subject word in the voice identification library; and displaying the target voice identification. The method allows the user to identify the content topic that the user is interested in based on the voice interaction data when using voice interaction, and recommend the voice identification corresponding to the topic as a virtual image of the voice assistant to the user, thereby enhancing the user's sense of interaction, gain, and participation in using voice, and allowing the user to have a better experience.

[0206] Similar parts between the embodiments provided in this application can be referenced to each other. The specific implementation methods provided above are only a few examples under the general concept of this application and do not constitute a limitation on the protection scope of this application. For those skilled in the art, any other implementation methods expanded based on the scheme of this application without creative work belong to the protection scope of this application.

Claims

1. A display device, characterized in that: include: monitor; an audio receiver, coupled to the display, for receiving audio data; A controller is coupled to the display and the audio receiver and is configured to: Controlling the audio receiver to receive the user's voice interaction data; Determining target keywords of the voice interaction data; Searching for the target keyword in a pre-built voice identification library and determining a target subject word corresponding to the target keyword; In response to the number of times the target subject word is searched in the voice identification library exceeding the target search threshold corresponding to the target subject word, determining a target voice identification corresponding to the target subject word in the voice identification library; The display is controlled to display the target voice identifier.

2. The display device according to claim 1, characterized in that The controller determines the target keyword of the voice interaction data and is configured to: Converting the voice interaction data into text data; Preprocessing the text data, wherein the preprocessing includes segmenting the text data and removing banned words; Based on a preset word extraction model, the target keywords are extracted from the preprocessed text data.

3. The display device according to claim 1, characterized in that Before the controller searches for the target keyword in the pre-built voice identification library and determines the target subject word corresponding to the target keyword, the controller is configured as follows: Get historical voice interaction data; Determining a plurality of historical subject words and a plurality of historical keywords corresponding to each of the historical subject words according to the historical voice interaction data, wherein all the historical subject words include the target subject word; Setting a preset search threshold for each of the historical subject words, wherein the preset search threshold for the target subject word is the target search threshold; Generate a first image related to each of the historical subject words by using an image generation model, and use the first image as a first voice identifier corresponding to the historical subject word, wherein the first voice identifier corresponding to the target subject word is the target voice identifier; The voice identification library is constructed according to all the historical keywords, all the historical subject words, all the preset search thresholds and all the first voice identifications.

4. The display device according to claim 3, characterized in that The controller determines a plurality of historical subject words and a plurality of historical keywords corresponding to each of the historical subject words according to the historical voice interaction data, and is specifically configured as follows: Converting the historical voice interaction data into historical text data; Inputting the historical text data into a preset subject word distribution model to obtain a subject word distribution result, wherein the subject word distribution result includes a plurality of historical keywords classified according to the subject; The subject identification is performed on the subject word distribution result to obtain the historical subject words corresponding to the multiple historical keywords that have been classified.

5. The display device according to claim 4, characterized in that The controller executes setting of a preset search threshold for each of the historical subject words, and is configured as follows: Recording the number of occurrences of each of the historical subject words in the process of determining the plurality of historical subject words and the plurality of historical keywords corresponding to each of the historical subject words; Determining the popularity level of the historical subject words according to the number of occurrences of each of the historical subject words; The preset search threshold for each of the historical keywords is set based on the popularity level.

6. The display device according to claim 1, characterized in that The controller searches for the target keyword in a pre-built voice identification library and determines a target subject word corresponding to the target keyword, and is specifically configured as follows: In response to the target keyword existing in the voice identification library, the target subject word corresponding to the target keyword is determined in the voice identification library.

7. The display device according to claim 6, characterized in that The controller determines the target subject word corresponding to the target keyword in the voice identification library, and is specifically configured as follows: In response to the target keyword corresponding to at least one first subject word in the voice identification library, the matching degree of the target keyword based on each of the first subject words is determined, and the first subject word with the highest matching degree is selected as the target subject word.

8. The display device according to claim 6, characterized in that The controller searches for the target keyword in a pre-built voice identification library and determines a target subject word corresponding to the target keyword, and is further configured to: In response to the target keyword not existing in the voice identification library, searching for a target synonymous keyword related to the target keyword, wherein the target synonymous keyword includes words with semantic similarity and / or the same as the target keyword; In response to at least one target synonymous keyword existing in the voice identification library, a second subject word corresponding to at least one target synonymous keyword is determined, and the target subject word corresponding to the target keyword is determined according to the second subject word.

9. The display device according to claim 8, characterized in that The controller determines the target subject word corresponding to the target keyword, and is further configured to: In response to the target keyword and the target synonymous keyword not existing in the voice identification library, determining a third subject word corresponding to the target keyword by using a preset subject determination model; Generate a second image related to the third keyword by using an image generation model, and use the second image as a second voice identifier corresponding to the third keyword; Based on the target keyword, the third subject word, and the second voice identifier, the voice identifier library is updated, and a preset search threshold corresponding to the third subject word is set.

10. The display device according to claim 1, characterized in that The controller controls the display to display the target voice identifier, and is configured to: Controlling the audio receiver to receive a voice identification switching instruction; In response to the voice identification switching instruction, the target voice identification is used as a virtual image of the voice assistant, and the display is controlled to display the target voice identification.

11. The display device according to claim 1, characterized in that The controller controls the display to display the target voice identifier, and is further configured to: Controlling the audio receiver to receive voice interaction instructions; In response to the voice interaction instruction, the display is controlled to display a voice interaction interface, and the target voice identifier is displayed on the voice interaction interface.

12. A method for recommending a voice tag, characterized in that: The voice identification recommendation method is applied to a display device, comprising: Receive user's voice interaction data; Determining target keywords of the voice interaction data; Searching for the target keyword in a pre-built voice identification library and determining a target subject word corresponding to the target keyword; In response to the number of times the target subject word is searched in the voice identification library exceeding the target search threshold corresponding to the target subject word, determining a target voice identification corresponding to the target subject word in the voice identification library; The target voice identifier is displayed.

Images