CN115438625A - Text error correction server, terminal device and text error correction method - Google Patents

Abstract

The embodiment of the application provides a text error correction server, terminal equipment and a text error correction method, and relates to the technical field of voice recognition. The text correction server includes: a communicator configured to receive a text to be corrected transmitted by a first terminal device; the controller is configured to judge whether a trigger condition for marking the text to be corrected is met or not under the condition that first error correction data for correcting the error of the text to be corrected is not included in an error correction database of the text error correction server; the communicator is further configured to send a labeling request for requesting labeling of the text to be corrected to at least one terminal device and receive labeling data sent by the at least one terminal device under the condition that the triggering condition is met; and the controller is also configured to generate error correction data corresponding to the text to be corrected according to the number of the labels sent by the at least one terminal device, and store the error correction data corresponding to the text to be corrected in an error correction database. The embodiment of the application is used for text error correction.

Description

Text error correction server, terminal device and text error correction method

Technical Field

The embodiment of the application relates to the technical field of voice recognition. And more particularly, to a text correction server, a terminal apparatus, and a text correction method.

Background

Text error correction refers to correcting errors that exist in text. Errors often occur in the process of converting speech into text due to dialects, accents, noise, etc., and the errors in the text will greatly affect the effect of tasks such as intent recognition, text retrieval, etc. Therefore, text error correction is one of the research hotspots in the field of speech recognition.

Currently, a text error correction scheme commonly used in the industry is a Natural Language Processing (NLP) based text error correction scheme. For example: and training through sample data to obtain a text error correction model, inputting the voice text to be error corrected into the text error correction model, and obtaining the output of the text error correction model as an error correction result of the text to be error corrected. However, the development cycle of the NLP-based text correction scheme is generally very long, and the time limit requirement of the related product cannot be met frequently.

Disclosure of Invention

The exemplary embodiment of the application provides a text error correction server, a terminal device and a text error correction method, which are used for solving the problem that the development period of a text error correction scheme is long.

The embodiment of the application provides the following technical scheme:

in a first aspect, an embodiment of the present application provides a text correction server, including:

the communicator is configured to receive the text to be corrected sent by the first terminal equipment;

the controller is configured to judge whether a trigger condition for marking the text to be corrected is met or not under the condition that first error correction data for correcting the text to be corrected is not included in an error correction database of the text correction server;

the communicator is further configured to send an annotation request for requesting to annotate the text to be corrected to at least one terminal device and receive annotation data sent by the at least one terminal device when the trigger condition is met;

the controller is further configured to generate error correction data corresponding to the text to be corrected according to the number of labels sent by the at least one terminal device, and store the error correction data corresponding to the text to be corrected in the error correction database.

In a second aspect, an embodiment of the present application provides a terminal device, including:

the communicator is configured to receive a labeling request which is sent by the text correction server and used for requesting to label the text to be corrected;

a display configured to display an annotation interface according to the annotation request;

the user interface is configured to receive a marking operation input in the marking interface by a user and generate marking data according to the marking operation;

the communicator is further configured to transmit the annotation data to the text correction server.

In a third aspect, an embodiment of the present application provides a text correction method, which is applied to a text correction server, and the method includes:

receiving a text to be corrected sent by first terminal equipment;

under the condition that an error correction database of the text error correction server does not comprise first error correction data used for correcting the text to be corrected, judging whether a triggering condition for marking the text to be corrected is met;

under the condition that the triggering condition is met, sending a labeling request for requesting to label the text to be corrected to at least one terminal device, and receiving labeling data sent by the at least one terminal device;

and generating error correction data corresponding to the text to be corrected according to the number of the labels sent by the at least one terminal device, and storing the error correction data corresponding to the text to be corrected in the error correction database.

In a fourth aspect, an embodiment of the present application provides a text error correction method, which is applied to a terminal device, and the method includes:

receiving a labeling request which is sent by a text error correction server and used for requesting to label a text to be corrected;

displaying a labeling interface according to the labeling request;

receiving a marking operation input by a user in the marking interface, and generating marking data according to the marking operation;

and sending the annotation data to the text error correction server.

In a fifth aspect, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a computing device, the computing device is caused to implement the text error correction method according to the third aspect or the fourth aspect.

In a sixth aspect, the present application provides a computer program product, which when run on a computer, causes the computer to implement the text error correction method of the third or fourth aspect.

According to the technical scheme, when a text to be corrected sent by a first terminal device is received, the text correction server and the text correction method provided by the embodiment of the application judge whether a first correction data for correcting the text to be corrected is included in an error correction database, and judge whether a trigger condition for labeling the text to be corrected is met under the condition that the first correction data for correcting the text to be corrected is not included in the error correction database; under the condition that the triggering condition is met, sending a labeling request for requesting to label the text to be corrected to at least one terminal device, and receiving labeling data sent by the at least one terminal device; and finally, generating error correction data corresponding to the text to be corrected according to the number of the labels sent by the at least one terminal device, and storing the error correction data corresponding to the text to be corrected in the error correction database. In the embodiment of the invention, the error correction data is used for correcting the error of the text to be corrected, and the error correction data is generated according to the number of the labels sent by the terminal equipment, so that the embodiment of the invention can quickly acquire the error correction data of the text to be corrected, which meets the triggering condition, in a user labeling mode, and further correct the error of the text to be corrected, and therefore, the embodiment of the invention can solve the problem of long development period of the text error correction scheme.

Drawings

In order to more clearly illustrate the embodiments of the present application or the implementation manner in the related art, a brief description will be given below of the drawings required for the description of the embodiments or the related art, and it is obvious that the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art according to the drawings.

FIG. 1 is a diagram illustrating a scene architecture of a text error correction method in some embodiments;

FIG. 2 illustrates a scene architecture diagram of a terminal device in some embodiments;

fig. 3 shows a block diagram of a hardware configuration of a terminal device in some embodiments;

fig. 4 shows a block diagram of a software configuration of a terminal device in some embodiments;

FIG. 5 illustrates a voice interaction network architecture diagram in some embodiments;

FIG. 6 shows a block diagram of a configuration of a text correction server in some embodiments;

FIG. 7 is a flowchart illustrating an interaction of a text correction method according to an embodiment of the present application;

FIG. 8 is a schematic diagram illustrating one of the annotation interfaces provided by embodiments of the present application;

FIG. 9 is a second interaction flowchart of a text correction method provided by an embodiment of the present application;

FIG. 10 is a second schematic diagram of a tagging interface provided by an embodiment of the present application;

FIG. 11 is a diagram illustrating a third interaction flow of a text error correction method according to an embodiment of the present application;

fig. 12 shows a flowchart of a method for determining a user emotion according to an embodiment of the present application.

Detailed Description

To make the purpose and embodiments of the present application clearer, the following will clearly and completely describe the exemplary embodiments of the present application with reference to the attached drawings in the exemplary embodiments of the present application, and it is obvious that the described exemplary embodiments are only a part of the embodiments of the present application, and not all of the embodiments.

It should be noted that the brief descriptions of the terms in the present application are only for convenience of understanding of the embodiments described below, and are not intended to limit the embodiments of the present application. These terms should be understood in their ordinary and customary meaning unless otherwise indicated.

The terms "comprises" and "comprising," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a product or apparatus that comprises a list of elements is not necessarily limited to all elements expressly listed, but may include other elements not expressly listed or inherent to such product or apparatus.

Fig. 1 is a schematic view of a scene architecture of a text error correction method according to an embodiment of the present application. As shown in fig. 1, a scenario architecture provided in the embodiment of the present application includes: the system comprises a terminal device 100 with a voice receiving function and configured to receive voice input of a user, a semantic recognition server 200, a text correction server 300, and at least one terminal device 400 (illustrated in fig. 1 by including three terminal devices 400) configured to label a text to be corrected.

The terminal device 100 may receive a voice input of a user and transmit the received voice input to the semantic recognition server 200. The semantic recognition server 200 is configured to convert the voice input sent by the terminal device 100 into a voice text, and obtain a user intention according to the converted voice text. In the process of acquiring the user intention according to the converted voice text, if there is no error in the converted voice text, the semantic recognition server 200 may acquire the user intention, and then the semantic recognition server 200 directly calls the corresponding service system according to the acquired user intention, and returns the service data returned by the service system to the terminal device 100. However, if there is an error in the converted speech text and the semantic recognition server 200 cannot acquire the user's intention, the semantic recognition server 200 returns the converted speech text to the terminal device 100. After receiving the voice text sent by the semantic recognition server 200, the terminal device 100 sends the voice text as a text to be corrected to the error correction server 300, so that the error correction server 300 corrects the text to be corrected. After receiving the text to be corrected sent by the terminal device 100, the error correction server 300 first determines whether the error correction database includes error correction data for correcting the text to be corrected; if the error correction database of the text error correction server 300 includes error correction data for correcting the text to be corrected, correcting the text to be corrected based on the error correction data, and returning the error correction result to the terminal device 100; the terminal device 100 makes a service request to the semantic recognition server 200 again according to the error correction result. If the error correction database of the text error correction server 300 does not include error correction data for correcting the text to be corrected, then returning indication information that the text error correction cannot be performed to the terminal device 100, and determining whether a trigger condition for labeling the text to be corrected is satisfied, if the trigger condition for labeling the text to be corrected is not satisfied, then only caching the text to be corrected without performing other processing, and if the trigger condition for labeling the text to be corrected is satisfied, then sending a labeling request for requesting labeling of the text to be corrected to each terminal device 400. After receiving the annotation request sent by the text correction server 300, the terminal device 400 displays an annotation interface according to the annotation request, receives an annotation operation for inputting on the annotation interface, generates annotation data according to the annotation operation, and returns the annotation data to the text correction server 300. After receiving the label data returned by each terminal device 400, the text error correction server generates error correction data corresponding to the text to be error corrected according to the label data returned by each terminal device 400, and stores the error correction data corresponding to the text to be error corrected in an error correction database, so as to correct the error of the text to be error corrected in the following process.

Fig. 2 is a schematic structural diagram of a terminal device according to an embodiment of the present application. As shown in fig. 1, an architecture of a terminal device provided in an embodiment of the present application includes: terminal device 100, control apparatus 500, and smart device 600.

The user can control and operate the terminal device 100 through the smart device 1002 or the control apparatus 500, and can receive a voice input of the user through the smart device 1002 or the control apparatus 500. The display device provided by the embodiment of the present application may have various implementation forms, for example, a television, a smart speaker, a refrigerator, a curtain with a display function, a Personal Computer (PC), a laser projection device, a display (monitor), an electronic whiteboard (electronic whiteboard), a wearable device, an in-vehicle device, an electronic desktop (electronic table), and the like.

In some embodiments, the control device 500 may be a remote controller, and the communication between the remote controller and the terminal device 100 includes infrared protocol communication or bluetooth protocol communication, and other short-distance communication methods, and controls the terminal device 100 through wireless or wired methods. The user can input a user instruction through a key on a remote controller, voice input, control panel input, or the like to control the terminal device 100.

In some embodiments, the terminal device 100 may also be controlled using a smart device 600 (e.g., a mobile phone, a tablet computer, a notebook computer, etc.). For example, the terminal device 100 is controlled using an application program running on the smart device 600.

In some embodiments, the terminal device 100 may also be controlled in a manner other than the control apparatus 500 and the smart device 600, for example, the voice command control of the user may be directly received by a module configured inside the terminal device 100 for acquiring the voice command.

Fig. 3 shows a hardware configuration block diagram of a terminal device in an exemplary embodiment. The terminal apparatus 100 shown in fig. 3 includes at least one of a communicator 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 280.

The controller 250 includes a central processor, an audio processor, a RAM, a ROM, first to nth interfaces for input/output.

The communicator 220 is a component for communicating with an external device or a server according to various communication protocol types. For example: the communicator 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 terminal device may establish transmission and reception of control signals and data signals with the semantic recognition server 200 through the communicator 220.

The user interface 280 may be used to receive external control signals.

The detector 230 is used to collect signals of the external environment or interaction with the outside. For example, detector 230 includes a light receiver, a sensor for collecting ambient light intensity; alternatively, the detector 230 includes an image collector, such as a camera, which can be used to collect external environment scenes, attributes of the user, or user interaction gestures, or the detector 230 includes a sound collector, such as a microphone, which is used to receive external sounds.

The sound collector can be a microphone, also called a microphone or a microphone, and can be used for receiving the sound of a user and converting a sound signal into an electric signal. The terminal device may be provided with at least one microphone. In other embodiments, the terminal device may be provided with two microphones, so as to achieve a noise reduction function in addition to collecting sound signals. In other embodiments, the terminal device may further include three, four, or more microphones to collect and reduce noise of the sound signal, and may further identify a sound source to realize a directional recording function.

In addition, the microphone may be built in the terminal device, or the microphone may be connected to the terminal device by wire or wirelessly. Of course, the position of the microphone on the terminal device is not limited in the embodiments of the present application. Alternatively, the terminal device may not include a microphone, i.e., the microphone is not provided in the terminal device. The terminal device may be externally connected to a microphone (also referred to as a microphone) via an interface (e.g., USB interface 130). The external microphone can be fixed on the terminal equipment through an external fixing piece (such as a camera support with a clip).

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 terminal device.

In some embodiments the controller comprises at least one of a Central Processing Unit (CPU), a video processor, an audio processor, a RAM Random Access Memory, a ROM (Read-Only Memory), a first to nth interface for input/output, a communication Bus (Bus), and the like.

In some examples, the operating system of the terminal device 100 is an Android system as an example, and as shown in fig. 4, the terminal device 100 may be logically divided into an application (Applications) layer 21, a kernel layer 22 and a hardware layer 23.

As shown in fig. 4, the hardware layer may include the communicator 220, the controller 250, the detector 230, the display 260, and the like shown in fig. 2. The application layer 21 includes one or more applications. The application may be a system application or a third party application. For example, the application layer 21 includes a voice recognition application, which may provide a voice interactive interface and service for the connection of the terminal device 100 with the semantic recognition server 200.

The kernel layer 22 acts as software middleware between the hardware layer and the application layer 21 for managing and controlling hardware and software resources.

In some examples, the kernel layer 22 includes a detector driver to send voice data collected by the detector 230 to a voice recognition application. Illustratively, when the speech recognition application in the terminal device 100 is started and the terminal device 100 establishes a communication connection with the semantic recognition server 200, the detector driver is configured to send the speech data input by the user and collected by the detector 230 to the speech recognition application. After that, the voice recognition application transmits query information containing the voice data to the communication control module 101 in the semantic recognition server 200. The communication control module 101 is used to input voice data transmitted by the terminal device 100 to the intention recognition module 102 and the data storage module 103.

For clarity of explanation of the embodiments of the present application, a speech recognition network architecture provided by the embodiments of the present application is described below with reference to fig. 5.

Referring to fig. 5, fig. 5 is a schematic diagram of a voice interaction network architecture according to an embodiment of the present application. In fig. 5, the voice interactive apparatus is configured to receive input information and output a processing result of the information. The voice recognition module is deployed with voice recognition service and used for recognizing the audio frequency as a text; the semantic understanding module is deployed with semantic understanding service and used for performing semantic analysis on the text; the business management module is provided with a business instruction management service for providing business instructions; the language generation module is deployed with a language generation service (NLG) and used for converting an instruction indicating the execution of the voice interaction equipment into a text language; and the voice synthesis module is deployed with a voice synthesis (TTS) service and used for processing a text language corresponding to the instruction and then sending the processed text language to a loudspeaker for broadcasting. In one embodiment, in the architecture shown in fig. 5, there may be multiple entity service devices deployed with different business services, and one or more function services may also be aggregated in one or more entity service devices.

In some embodiments, the following describes, by way of example, a process for processing information input to the voice interaction device based on the architecture shown in fig. 5, where the information input to the voice interaction device is a voice instruction input by voice:

[ Speech recognition ]

The voice interaction device can perform noise reduction processing and feature extraction on the audio frequency of the voice instruction after receiving the voice instruction input by voice, wherein the noise reduction processing can comprise the steps of removing echo and ambient noise and the like.

[ semantic understanding ]

And performing natural language understanding on the identified candidate texts and associated context information by using the acoustic model and the language model, and analyzing the texts into structured and machine-readable information, information such as business fields, intentions, word slots and the like to express semantics and the like. Deriving an actionable intent determination intent confidence score, a semantic understanding module selects one or more candidate actionable intents based on the determined intent confidence score,

[ Business management ]

The semantic understanding module issues an execution instruction to the corresponding service management module according to a semantic analysis result of the text of the voice instruction so as to execute the operation corresponding to the voice instruction, complete the operation requested by the user, and feed back the execution result of the operation corresponding to the voice instruction.

In some examples, as shown in fig. 6, fig. 6 is a block diagram of a configuration of the text correction server 300. When the semantic recognition server 200 converts the audio into the text, an error exists in the text, and the semantic recognition server 200 returns the converted text to the terminal device 100. Then, the terminal device 100 sends the converted text as a text to be corrected to the communication control module 301 in the text correction server 300, and the communication control module 301 outputs the received text to be corrected to the correction module 302; the error correction module 302 is configured to query whether the error correction database 303 contains error correction data of the text to be error corrected, correct the text to be error corrected according to the error correction data if the error correction database 303 contains the error correction data of the text to be error corrected, and return the error correction result to the terminal device 100 through the communication control module 301. If the error correction database 303 does not contain error correction data of the text to be corrected, the decision module 304 is invoked to determine whether a trigger condition for labeling the text to be corrected is satisfied. In the case that the trigger condition is met, the voting module 305 sends a labeling request for requesting to label the text to be corrected to the terminal device 400, and receives labeling data sent by the terminal device 400. The decision module 304 is further configured to generate error correction data corresponding to the text to be error corrected according to the number of labels sent by the at least one terminal device, and store the error correction data corresponding to the text to be error corrected in the error correction database 303.

In some embodiments, as shown in fig. 5, a text error correction method provided in an embodiment of the present application includes the following steps:

s701, the first terminal device sends a text to be corrected to a text correction server.

Correspondingly, the text error correction server receives the text to be error corrected sent by the first terminal device.

In some embodiments, the implementation manner of the first terminal device sending the text to be corrected to the text correction server may include: the method comprises the steps that a first terminal device sends a voice to be recognized to a semantic recognition server and receives a semantic recognition result returned by the semantic recognition server; and if the semantic recognition result only comprises the text corresponding to the voice to be recognized and does not comprise the user intention corresponding to the voice to be recognized, sending the text corresponding to the voice to be recognized as the text to be corrected to the text correction server.

For example: the user inputs a section of voice input corresponding to the text content of 'how to do preserved eggs' to the first terminal device, after the first terminal device sends the voice input to the semantic recognition server, the semantic recognition server converts the voice input into a text 'how to do send flowers', the user intention corresponding to the voice input cannot be obtained according to the text 'how to do send flowers', the text 'how to do send flowers' is returned to the first terminal device, and then the text 'how to do send flowers' is sent to the text error correction server by the first terminal device as the text to be corrected.

In some embodiments, the first terminal device may transmit the text to be corrected to the text correction server through the communicator, and the text correction server may receive the text to be corrected transmitted by the first terminal device through the communicator.

S702, the text error correction server judges whether the error correction database of the text error correction server comprises first error correction data for correcting the text to be corrected.

In the above step S702, if the error correction database of the text correction server does not include the first error correction data for correcting the text to be corrected, the following step S703 is executed.

S703, the text error correction server judges whether the triggering condition for marking the text to be error corrected is met.

That is, in the embodiment of the present application, all texts that do not obtain the intention of the user are not labeled, but only texts that satisfy the trigger condition are labeled.

For example: the triggering conditions are that the accumulated unrecognized times exceed 5 times, the text to be corrected is 'what do with flower send' and then when first correction data for correcting the 'what do with flower send' are not included in the correction database, whether the times of sending the 'what do with flower send' of the text to be corrected to the text correction server by the terminal device exceeds 5 times or not is judged, and if the times exceed 5 times, the triggering conditions for marking the 'what do with flower send' are determined to be met; and if the number of times does not exceed 5, determining that the triggering condition for marking 'sending flower is not met'.

In some embodiments, the text correction server may execute the above steps S702 and S703 (determine whether the first error correction data for correcting the text to be corrected is included in the error correction database of the text correction server, and determine whether the trigger condition for labeling the text to be corrected is satisfied in a case that the first error correction data for correcting the text to be corrected is not included in the error correction database of the text correction server).

In the above step S703, if the text correction server determines that the trigger condition for labeling the text to be corrected is satisfied, the following step S704 is executed.

S704, the text correction server sends a labeling request for requesting to label the text to be corrected to at least one terminal device.

Correspondingly, the at least one terminal device receives a labeling request which is sent by the text correction server and used for requesting to label the text to be corrected.

In some embodiments, the at least one terminal device may comprise the first terminal device. Namely, the text error correction server sends a labeling request for requesting to label the text to be corrected to the terminal equipment sending the data to be corrected. For example: the terminal device a sends the text to be corrected "what do it spend" to the text correction server, and the text correction server sends a labeling request for requesting the labeling of the text to be corrected "what do spend" to a plurality of terminal devices including the terminal device a.

In some embodiments, the text correction server may perform the above step S704 (sending an annotation request for requesting to annotate the text to be corrected to at least one terminal device) through the communicator.

S705, the at least one terminal device displays a labeling interface according to the labeling request.

For example, referring to fig. 8, the text to be corrected is "what do with flower send' in fig. 8 as an example of a labeling interface for explanation, as shown in fig. 8, a prompt text 81 and a text 82 to be corrected may be displayed on the labeling interface, the prompt text 81 may be" please input the following text meaning, "and the text 82 to be corrected is" what do with flower send ".

In some embodiments, the at least one terminal device may perform the above step S705 (displaying an annotation interface according to the annotation request) through a display respectively.

S706, the at least one terminal device receives the labeling operation input by the user in the labeling interface respectively, and generates labeling data according to the labeling operation.

In some embodiments, the annotation operation entered by the user in the annotation interface may be a text entry operation. For example: and the user inputs the text 'doing how preserved eggs' in the labeling interface through text input operation, and generates labeling data according to the text 'doing how preserved eggs'.

In some embodiments, the at least one terminal device may perform the above step S706 (receiving a labeling operation input in the labeling interface by a user, and generating labeling data according to the labeling operation) through a user interface respectively.

S707, the at least one terminal device sends the annotation data to the text correction server respectively.

Correspondingly, the text error correction server receives the annotation data sent by the at least one terminal device.

In some embodiments, the at least one terminal device may transmit the annotation data to the text correction server via a communicator, respectively, and the text correction server may receive the annotation data transmitted by the at least one terminal device via the communicator.

S708, the text error correction server generates error correction data corresponding to the text to be error corrected according to the label number sent by the at least one terminal device.

For example: the text to be corrected is "how do to send a preserved egg", the annotation data sent by the at least one terminal device to the text correction server is "how do to make a preserved egg", and the error correction data that can be generated can be "how to make a preserved egg".

S709, the text error correction server stores the error correction data corresponding to the text to be error corrected to the error correction database.

In some embodiments, the text error correction server may execute the above steps S708 and S709 (generate error correction data corresponding to the text to be error corrected according to the number of labels sent by the at least one terminal device, and store the error correction data corresponding to the text to be error corrected in the error correction database) through the controller.

The text error correction method provided by the embodiment of the application judges whether an error correction database comprises first error correction data for correcting the text to be corrected when a text error correction server receives the text to be corrected sent by a first terminal device, and judges whether a trigger condition for labeling the text to be corrected is met under the condition that the error correction database does not comprise the first error correction data for correcting the text to be corrected; under the condition that the trigger condition is met, sending a labeling request for requesting to label the text to be corrected to at least one terminal device, and receiving labeling data sent by the at least one terminal device; and finally, generating error correction data corresponding to the text to be corrected according to the number of the labels sent by the at least one terminal device, and storing the error correction data corresponding to the text to be corrected in the error correction database. In the embodiment of the invention, the error correction data is used for correcting the error of the text to be corrected, and the error correction data is generated according to the number of the labels sent by the terminal equipment, so that the embodiment of the invention can quickly acquire the error correction data of the text to be corrected, which meets the triggering condition, in a user labeling mode, and further correct the error of the text to be corrected, and therefore, the embodiment of the invention can solve the problem of long development period of the text error correction scheme.

As an extension and refinement of the above embodiment, an embodiment of the present invention provides another text error correction method, which is shown in fig. 9 and includes the following steps:

s901, the first terminal equipment receives voice input of a user.

In some embodiments, after receiving the voice input of the user, the first terminal device may further perform noise reduction processing on the voice input.

S902, the first terminal device sends the voice input to a semantic recognition server.

Correspondingly, the semantic recognition server receives the voice input sent by the first terminal device.

And S903, converting the voice input into a text by the semantic recognition server, and acquiring the user intention according to the text.

In step S903, if the semantic recognition server successfully obtains the user intention according to the text, the semantic recognition server directly calls the relevant service system to obtain the service data, and returns the obtained service data to the first terminal device.

For example: and the semantic recognition server converts the text obtained by voice input into 'play static night thinking', calls a service system to acquire audio resources or video resources related to 'static night thinking', and sends the acquired audio resources or video resources related to 'static night thinking' to the first terminal equipment, so that the first terminal equipment plays the audio resources or video resources related to 'static night thinking'.

In the step S903, if the semantic recognition server fails to obtain the user intention according to the text, the semantic recognition server executes the following step S904.

S904, the semantic recognition server sends the text obtained by converting the voice input to the first terminal device.

Correspondingly, the first terminal equipment receives the text which is sent by the semantic recognition server and obtained by converting the voice input.

For example: the semantic recognition server converts the text from the speech input to "where the province of the netherlands is? "as can not be followed by the text" where the dutch province will be? "get the user's intent, and thus the text" where the province of the netherlands "that will convert the speech input? "return to the first terminal device.

S905, the first terminal device determines the text obtained by converting the voice input as the text to be corrected, and sends the text to be corrected to the text correction server.

Correspondingly, the text error correction server receives the text to be error corrected sent by the first terminal device.

S906, the text error correction server judges whether the error correction database of the text error correction server comprises first error correction data for correcting the text to be corrected.

In some embodiments, the first error correction data is error correction data for correcting errors of the text to be corrected sent by all terminal devices.

For example: the text to be corrected is "how big is a build? "the error correction database includes a text" how big to build? "how large is error correction data for error correction" created? "the error correction data is determined to include first error correction data for correcting the error of the text to be corrected, and the first error correction data is error correction data for correcting the error of the text to be corrected sent by all the terminal devices.

In some embodiments, the first error correction data is error correction data configured to be used only for error correction of the text to be error corrected sent by the first terminal device.

For example: is the text to be corrected "unknown", and includes error correction data "unknown" for correcting errors in the text "unknown" transmitted by the first terminal device in the error correction database? "the first error correction data is only used for correcting the text to be corrected sent by the first terminal device.

For another example: the text to be corrected is "two yellowbird cuilius", the error correction database includes error correction data "two yellowbird cuilius" for correcting the error of the text "two yellowbird cuilius" sent by the second terminal device, and it can be determined that the error correction database does not include first error correction data for correcting the error of the text to be corrected.

In the above step S906, if the error correction database of the text error correction server includes the first error correction data for performing error correction on the text to be error corrected, the following steps S907 to S910 are performed.

S907, the text correction server corrects the error of the first text according to the first correction data to obtain a second text.

Illustratively, the text to be corrected is "how big is a build? "is the error correction database included for" how big is build? "how large is the first error correction data to be error corrected" created? "the first text may be error corrected based on the first error correction data to obtain a second text" how big is a good creation? ".

In some embodiments, the text correction server may execute the above step S907 by the controller (the text correction server corrects the first text according to the first correction data to obtain the second text).

S908, the text error correction server sends the second text to the first terminal device.

Correspondingly, the first terminal equipment receives the second text sent by the text correction server.

And S909, the first terminal device sends the second text to the semantic recognition server.

Correspondingly, the semantic recognition server receives the second text sent by the first terminal equipment.

S910, the semantic recognition server obtains the user intention according to the second text.

For example: the second text is "how big is the Fujian? If the second text is a text of the second user, the semantic recognition server can query the area of the Fujian province for the purpose of obtaining the user intention according to the second text, then call the related service system to obtain the area of the Fujian province, and return the queried area of the Fujian province to the first terminal device.

In the above step S906, if the error correction database of the text correction server does not include the first error correction data for correcting the text to be corrected, the following steps S911 to S918 are executed.

And S911, the text error correction server judges whether the triggering condition for marking the text to be corrected is met.

In the step S911, if the text correction server determines that the trigger condition for labeling the text to be corrected is satisfied, the following step S912 is executed.

S912, the text error correction server sends a labeling request to at least one terminal device.

Correspondingly, the at least one terminal device receives the marking request sent by the text correction server respectively.

The marking request is used for requesting to mark the text to be corrected, and the marking request carries at least one marking option.

And S913, the at least one terminal device displays a labeling interface according to the labeling request.

Illustratively, referring to fig. 10, in fig. 10, a labeling interface is illustrated by taking an example that the text to be corrected is "this am unknown", as shown in fig. 10, a prompt text 1001, a text 1002 to be corrected, a labeling option 1003 and a labeling option may be displayed on the labeling interface, the prompt text 1001 may be "please select the correct meaning of the text", the text 1002 to be corrected is "this am unknown", and the labeling option 1003 includes "this am aware" and the labeling option 1004 is "this am unknown".

S914, the at least one terminal device respectively receives the selection operation of the user to the at least one annotation option, and generates annotation data according to the annotation option selected by the user.

For example: and if the at least one terminal device 1 receives the selection operation of the user on the labeling option a in the at least one labeling option, the terminal device 1 generates labeling data according to the labeling option a, and if the at least one terminal device 2 receives the selection operation of the user on the labeling option B in the at least one labeling option, the terminal device 2 generates labeling data according to the labeling option B.

S915, the at least one terminal device respectively sends the marking data to the text correction server.

Correspondingly, the text error correction server receives the annotation data sent by the at least one terminal device.

S916, the text correction server judges whether the label data with the ratio exceeding the threshold ratio exists in the label data respectively sent by the at least one terminal device to the text correction server.

Illustratively, the threshold is 60%, and the terminal device includes: the labeling data returned by the terminal devices 1 to 10 are as shown in the following table 1:

TABLE 1

Terminal device	Annotating data
		Terminal device 1	Annotation data A
Terminal device 2	Annotation data A
		Terminal device 3	Annotation data B
Terminal equipment 4	Annotation data A
		Terminal device 5	Annotation data A
Terminal device 6	Annotation data C
		Terminal device 7	Annotation data C
Terminal device 8	Annotation data A
		Terminal device 9	Annotation data A
Terminal device 10	Annotation data A

From table 1 above, it can be seen that: the percentage of the annotation data A is 70%, the percentage of the annotation data B is 10%, and the percentage of the annotation data C is 20%, wherein the percentage (70%) of the annotation data A is greater than a threshold percentage (60%), so that the text correction server determines that the at least one terminal device can respectively send annotation data with the percentage exceeding the threshold percentage to the text correction server, and the annotation data with the percentage exceeding the threshold percentage is the annotation data A.

Illustratively, the threshold is 75%, and the terminal device includes: the labeling data returned by the terminal devices 1 to 10 are as shown in the following table 2:

TABLE 2

Terminal device	Annotating data
		Terminal device 1	Annotation data A
Terminal device 2	Annotation data E
		Terminal device 3	Annotation data A
Terminal device 4	Annotation data A
		Terminal device 5	Annotation data C
Terminal device 6	Annotation data E
		Terminal device 7	Labeling data B
Terminal device 8	Annotation data C
		Terminal device 9	Annotation data B
Terminal device 10	Labeling data D

From table 2 above, it can be seen that: the percentage of the annotation data A is 30%, the percentage of the annotation data B is 20%, the percentage of the annotation data C is 10%, the percentage of the annotation data E is 20%, and the percentages of the annotation data A, the annotation data B, the annotation data C, the annotation data D and the annotation data E are all less than a threshold percentage (75%), so that the text error correction server determines that the at least one terminal device can respectively send the annotation data to the text error correction server, wherein the percentage of the annotation data does not exceed the threshold percentage.

In the above step S916, if the ratio of the target annotation data in the annotation data sent by the at least one terminal device exceeds the threshold ratio, the following step S917 is executed:

s917, the text error correction server generates error correction data for correcting the text to be error-corrected sent by all the terminal devices according to the target marking data.

Bearing the example shown in table 1, the ratio of the annotation data a in the annotation data sent by the terminal device 1 to the terminal device 10 is greater than the threshold ratio, so that error correction data for correcting the error of the text to be corrected sent by all the terminal devices is generated according to the annotation data a.

In the step S916, if the ratio of each piece of annotation data in the annotation data sent by the at least one terminal device is smaller than the threshold ratio, the following step S918 is executed:

s918, the text error correction server generates error correction data only used for correcting the error of the text to be corrected sent by the corresponding terminal device according to the marking data of each terminal device in the at least one terminal device.

In the example shown in table 1, the ratio of each piece of annotation data in the annotation data sent by the terminal device 1 to the terminal device 10 is smaller than the threshold ratio, so that error correction data only used for correcting the text to be corrected sent by the terminal device 1 is generated according to the annotation data a sent by the terminal device 1, error correction data only used for correcting the text to be corrected sent by the terminal device 2 is generated according to the annotation data E sent by the terminal device 2, 8230 \ 8230, and error correction data only used for correcting the text to be corrected sent by the terminal device 10 is generated according to the annotation data D sent by the terminal device 10.

S919, the text error correction server stores the error correction data corresponding to the text to be error corrected to the error correction database.

In some embodiments, referring to fig. 11, on the basis of the embodiment shown in fig. 9, in a case that the error correction database of the text correction server does not include the first error correction data for correcting the error of the text to be corrected, the text correction method provided in the embodiment of the present invention further includes the following steps:

s111, the text error correction server sends indication information used for indicating that error correction of the text to be error corrected fails to the first terminal device.

Correspondingly, the first terminal device receives the message error correction server-sent message for indicating that the message error correction to be corrected fails.

And S112, the first terminal equipment outputs indication information for indicating that the voice input recognition fails.

In some embodiments, the implementation manner of the first terminal device outputting the indication information for indicating the voice input recognition failure may be: an audio file indicating a failure of speech input recognition is played. For example: and playing an audio file with the corresponding text of 'I does not understand your meaning'.

In some embodiments, the implementation manner of the first terminal device outputting the indication information for indicating the voice input recognition failure may be: text information indicating a failure of speech input recognition is displayed. For example: the text information "you say i have not learned" is displayed.

In some embodiments, the implementation manner of determining whether the trigger condition for labeling the text to be corrected is satisfied may include:

if the number of times that the first terminal device continuously sends the text to be corrected to the text correction server exceeds a first threshold number of times; alternatively, the first and second liquid crystal display panels may be,

if the number of times that terminal equipment in a target geographic area sends the text to be corrected to the text correction server in a first preset time exceeds a second threshold number of times, wherein the target geographic area is the geographic area to which the first terminal equipment belongs; alternatively, the first and second electrodes may be,

if the number of times that the first terminal device sends the text to the text error correction server within a second preset time length exceeds a third threshold number of times; alternatively, the first and second liquid crystal display panels may be,

if the user emotion corresponding to the text to be corrected is a negative emotion;

then, determining that the triggering condition for marking the text to be corrected is met.

That is, whether the following conditions 1 to 4 are satisfied is determined, and when at least one of the following conditions 1 to 4 is satisfied, it is determined that a trigger condition for labeling the text to be corrected is satisfied:

and in the condition 1, the number of times that the first terminal device continuously sends the text to be corrected to the text correction server exceeds a first threshold number of times.

The first threshold number of times may be issued by a background configuration of the text correction server.

For example: and if the first threshold number of times is configured to be 3 times, judging whether a trigger condition for labeling the text to be corrected is met by judging whether the number of times that the first terminal device continuously sends the text to be corrected to the text correction server exceeds 3 times.

And 2, sending the times of the text to be corrected to the text correction server by the terminal equipment in the target geographic area within a first preset time length to exceed a second threshold value. And the target geographic area is the geographic area to which the first terminal equipment belongs.

The text error correction method comprises the steps that a text error correction server divides geographical areas in advance, then when judging whether a triggering condition for labeling a text to be error corrected is met, the geographical area to which a first terminal device belongs is firstly obtained, then text error correction historical data of all terminal devices in the geographical area to which the first terminal device belongs within a first preset time length are obtained, and whether the triggering condition for labeling the text to be error corrected is met is judged by judging whether the sending times of the text to be error corrected in the text error correction historical data of all devices in the geographical area to which the first terminal device belongs within the first preset time length exceed a second threshold value. The geographical area to which the first terminal device belongs may be obtained according to an Internet Protocol (IP) address of the first terminal device.

For example: and if the first preset duration is 24 hours, the geographic area to which the first terminal device belongs is XX town, and the second threshold frequency is 10, acquiring whether the frequency of sending the text to be corrected to the text correction server by all the terminal devices in XX town exceeds 10 times in the past 24 hours to judge whether the triggering condition for labeling the text to be corrected is met.

And 3, the number of times that the first terminal device sends the text to the text correction server within a second preset time exceeds a third threshold number of times.

For example: and if the second preset time is 7 days and the third threshold time is 5, judging whether the triggering condition for labeling the text to be corrected is met by judging whether the times of sending the text to the text correction server by the first terminal device in the last 7 days exceeds 5 times.

And 4, the user emotion corresponding to the text to be corrected is a negative emotion.

And judging whether the triggering condition for labeling the text to be corrected is met or not by judging whether the user emotion corresponding to the text to be corrected is a negative emotion or not. In some embodiments, referring to fig. 12, an implementation manner of obtaining a user emotion corresponding to the text to be corrected may include the following steps a to b:

s121, performing word segmentation on the text to be corrected, and acquiring word segmentation results of the text to be corrected.

For example: if the text to be corrected is "i like playing a game", the word segmentation result of the text to be corrected may include: "I", "like", "play", "game".

And S122, loading a dirty word database.

S123, judging whether the word segmentation result of the text to be corrected contains the dirty words in the dirty word database.

In the step c, if the word segmentation result of the text to be corrected includes the dirty word in the dirty word database, executing the following step d:

and S124, determining the user emotion corresponding to the text to be corrected as a negative emotion.

In the step c, if the word segmentation result of the text to be corrected does not include the dirty word in the dirty word database, executing the following step e:

s125: and determining whether the user emotion corresponding to the text to be corrected is negative emotion or not based on an emotion analysis model.

In some embodiments, the emotion analysis model may be a model obtained by training a machine learning model based on sample data, where the sample data includes a sample text and emotion labels of the sample text, the emotion labels are negative emotions or non-negative emotions, and the machine learning model may be a deep neural network model, a convolutional neural network model, or the like.

In some embodiments, the present application provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a computing device, the computing device is caused to implement the text error correction method according to any one of the above embodiments.

In some embodiments, the present application provides a computer program product, which when run on a computer, causes the computer to implement the text error correction method described in any of the above embodiments.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present application.

The foregoing description, for purposes of explanation, has been presented in conjunction with specific embodiments. However, the illustrative discussions above are not intended to be exhaustive or to limit the embodiments to the precise forms disclosed above. Many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles and the practical application, to thereby enable others skilled in the art to best utilize the embodiments and various embodiments with various modifications as are suited to the particular use contemplated.

Claims (10)

1. A text correction server, comprising:

the communicator is configured to receive the text to be corrected sent by the first terminal equipment;

the controller is configured to judge whether a trigger condition for marking the text to be corrected is met or not under the condition that first error correction data for correcting the text to be corrected is not included in an error correction database of the text correction server;

the communicator is further configured to send an annotation request for requesting to annotate the text to be corrected to at least one terminal device and receive annotation data sent by the at least one terminal device when the trigger condition is met;

the controller is further configured to generate error correction data corresponding to the text to be error corrected according to the number of labels sent by the at least one terminal device, and store the error correction data corresponding to the text to be error corrected in the error correction database.

2. The text correction server of claim 1,

the controller is further configured to, in a case where the first error correction data is included in the error correction database, perform error correction on the first text according to the first error correction data to obtain a second text;

the communicator is further configured to transmit the second text to the first terminal device.

3. The text correction server of claim 1,

the first error correction data is error correction data used for correcting errors of the texts to be corrected sent by all the terminal devices;

or;

the first error correction data is error correction data configured to be used only for correcting the text to be corrected sent by the first terminal device.

4. The text correction server of claim 1, wherein the controller is specifically configured to:

if the number of times that the first terminal device continuously sends the text to be corrected to the text correction server exceeds a first threshold number of times; alternatively, the first and second electrodes may be,

if the number of times that terminal equipment in a target geographical area sends the text to be corrected to the text correction server in a first preset time exceeds a second threshold number of times, wherein the target geographical area is a geographical area to which the first terminal equipment belongs; alternatively, the first and second electrodes may be,

if the number of times that the first terminal device sends the text to the text correction server within a second preset time length exceeds a third threshold number of times; alternatively, the first and second electrodes may be,

if the user emotion corresponding to the text to be corrected is a negative emotion;

and determining that the triggering condition for marking the text to be corrected is met.

5. The text correction server of claim 1, wherein the controller is specifically configured to:

under the condition that the proportion of target marking data in marking data sent by at least one terminal device exceeds a threshold proportion, generating error correction data for correcting the text to be corrected sent by all terminal devices according to the target marking data;

and under the condition that the proportion of each marking data in the marking data sent by the at least one terminal device is smaller than the threshold proportion, generating error correction data only used for correcting the text to be corrected sent by the corresponding terminal device according to the marking data of each terminal device in the at least one terminal device.

6. The text correction server according to claim 1, wherein the communicator is further configured to transmit indication information indicating that the error correction of the text to be corrected fails to the first terminal device in a case that first error correction data for error correction of the text to be corrected is not included in an error correction database of the text correction server.

7. A terminal device, comprising:

the communicator is configured to receive a labeling request which is sent by the text correction server and used for requesting to label the text to be corrected;

a display configured to display an annotation interface according to the annotation request;

the user interface is configured to receive a marking operation input in the marking interface by a user and generate marking data according to the marking operation;

the communicator is further configured to transmit the annotation data to the text correction server.

8. The terminal device according to claim 7, wherein the annotation request carries at least one annotation option;

the display is specifically configured to display the text to be corrected and the at least one labeling option;

the user interface is specifically configured to receive a selection operation of the user on the at least one annotation option, and generate the annotation data according to the annotation option selected by the user.

9. A text correction method is applied to a text correction server, and the method comprises the following steps:

receiving a text to be corrected sent by first terminal equipment;

under the condition that an error correction database of the text error correction server does not comprise first error correction data used for correcting the text to be corrected, judging whether a triggering condition for marking the text to be corrected is met;

under the condition that the triggering condition is met, sending a labeling request for requesting to label the text to be corrected to at least one terminal device, and receiving labeling data sent by the at least one terminal device;

and generating error correction data corresponding to the text to be corrected according to the number of the labels sent by the at least one terminal device, and storing the error correction data corresponding to the text to be corrected in the error correction database.

10. A text error correction method is applied to a terminal device, and comprises the following steps:

receiving a labeling request which is sent by a text correction server and used for requesting to label a text to be corrected;

displaying a labeling interface according to the labeling request;

receiving a marking operation input by a user in the marking interface, and generating marking data according to the marking operation;

and sending the annotation data to the text correction server.

Images