CN112069805A - Text labeling method, device, equipment and storage medium combining RPA and AI - Google Patents
Text labeling method, device, equipment and storage medium combining RPA and AI Download PDFInfo
- Publication number
- CN112069805A CN112069805A CN202011062863.3A CN202011062863A CN112069805A CN 112069805 A CN112069805 A CN 112069805A CN 202011062863 A CN202011062863 A CN 202011062863A CN 112069805 A CN112069805 A CN 112069805A
- Authority
- CN
- China
- Prior art keywords
- text
- pinyin
- chinese
- marking
- pause information
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000002372 labelling Methods 0.000 title claims abstract description 38
- 238000012545 processing Methods 0.000 claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 47
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 17
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 17
- 238000005516 engineering process Methods 0.000 claims abstract description 16
- 238000012937 correction Methods 0.000 claims description 28
- 238000003058 natural language processing Methods 0.000 claims description 17
- 230000008859 change Effects 0.000 claims description 14
- 230000008569 process Effects 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 9
- 238000012549 training Methods 0.000 claims description 7
- 238000013473 artificial intelligence Methods 0.000 description 43
- 101100048435 Caenorhabditis elegans unc-18 gene Proteins 0.000 description 7
- 238000004891 communication Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 5
- 230000011218 segmentation Effects 0.000 description 5
- 238000013528 artificial neural network Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 238000004801 process automation Methods 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000003780 insertion Methods 0.000 description 1
- 230000037431 insertion Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L13/00—Speech synthesis; Text to speech systems
- G10L13/02—Methods for producing synthetic speech; Speech synthesisers
Abstract
The application provides a text labeling method and device combining RPA and AI, relating to the RPA and AI speech synthesis technology, wherein the method comprises the following steps: acquiring a Chinese text to be marked, and generating a pinyin text corresponding to the Chinese text; determining pause information corresponding to the Chinese text by adopting an NLP technology, and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information; marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking; the Chinese text to be marked, the Chinese text with pause information and the pinyin text after marking processing are provided for a user to be corrected, and the user can obtain the pinyin marking result corresponding to the Chinese text to be marked with high accuracy rate only by correcting the pinyin text after marking processing, so that the user can be assisted in marking the text, and the marking efficiency and the accuracy of the text are improved.
Description
Technical Field
The present application relates to the field of computer technologies, and in particular, to a RPA (robot Process Automation) and AI (Artificial Intelligence) speech synthesis technology, and more particularly, to a text labeling method, apparatus, device, and storage medium combining RPA and AI.
Background
Robot Process Automation (RPA) simulates the operation of a human on a computer through specific robot software and automatically executes Process tasks according to rules.
Artificial Intelligence (AI) is a new technology science for researching and developing theories, methods, techniques and application systems for simulating, extending and expanding human Intelligence. Research in the field of artificial intelligence includes robotics, language recognition, image recognition, natural language processing, and expert systems, among others.
Speech synthesis is a technique that can generate speech from text. The construction of the speech synthesis system requires a large amount of audio and pinyin annotation results corresponding to the audio. The quality of the pinyin annotation result determines the tone quality of the speech synthesis system. At present, people mainly perform pinyin annotation on a Chinese text corresponding to audio, which is long in time consumption and high in error rate.
Disclosure of Invention
The object of the present application is to solve at least to some extent one of the above mentioned technical problems.
Therefore, a first objective of the present application is to provide a text labeling method combining RPA and AI, in which a chinese text to be labeled is preprocessed to generate a pinyin text to be corrected, and pinyins meeting conditions to be corrected in the pinyin text are marked to be provided for a user for correction, and the user only needs to correct the pinyin text after marking, so that a pinyin labeling result corresponding to the chinese text to be labeled with high accuracy can be obtained, thereby assisting the user in labeling the text and improving the labeling efficiency and accuracy of the text.
A second objective of the present application is to provide a text annotation device that combines RPA and AI.
A third object of the present application is to provide a text annotation device that combines RPA and AI.
A fourth object of the present application is to propose a non-transitory computer-readable storage medium.
To achieve the above object, an embodiment of a first aspect of the present application provides a text annotation method combining an RPA and an AI, including: acquiring a Chinese text to be marked, and generating a pinyin text corresponding to the Chinese text; determining pause information corresponding to the Chinese text by adopting a Natural Language Processing (NLP) technology, and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information; marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking; and providing the Chinese text to be marked, the Chinese text with the pause information and the pinyin text after marking processing for a user for correction.
According to the text marking method combining the RPA and the AI, the Chinese text to be marked is obtained, and the pinyin text corresponding to the Chinese text is generated; determining pause information corresponding to the Chinese text by adopting an NLP technology, and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information; marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking; and providing the Chinese text to be marked, the Chinese text with the pause information and the pinyin text after marking processing for a user for correction. The method preprocesses the Chinese text to be marked to generate the pinyin text to be corrected, marks the pinyin meeting the condition to be corrected in the pinyin text and provides the pinyin text to a user for correction, and the user only needs to correct the pinyin text after marking processing to obtain the pinyin marking result corresponding to the Chinese text to be marked with high accuracy, so that the user can be assisted in marking the text, and the marking efficiency and the accuracy of the text are improved.
In order to achieve the above object, a second aspect of the present application provides a text annotation device combining RPA and AI. The device includes: the system comprises an acquisition module, a marking module and a marking module, wherein the acquisition module is used for acquiring a Chinese text to be marked and generating a pinyin text corresponding to the Chinese text; the determining module is used for determining the pause information corresponding to the Chinese text by adopting an NLP technology, inserting the pause information into the Chinese text and obtaining the Chinese text with the pause information; the marking module is used for marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking; and the providing module is used for providing the Chinese text to be marked, the Chinese text with the pause information and the pinyin text after marking processing for a user for correction.
The text marking device combining the RPA and the AI in the embodiment of the application generates a pinyin text corresponding to a Chinese text by acquiring the Chinese text to be marked; determining pause information corresponding to the Chinese text by adopting an NLP technology, and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information; marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking; and providing the Chinese text to be marked, the Chinese text with the pause information and the pinyin text after marking processing for a user for correction. The device can realize preprocessing the Chinese text to be labeled, generate the pinyin text to be corrected, mark the pinyin meeting the conditions to be corrected in the pinyin text, provide the pinyin text to the user for correction, and the user only needs to correct the pinyin text after marking processing, so that the pinyin labeling result corresponding to the Chinese text to be labeled with high accuracy can be obtained, the user can be assisted in labeling the text, and the labeling efficiency and the accuracy of the text are improved.
In order to achieve the above object, a third aspect of the present application provides a text annotation device combining RPA and AI, including: memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor implements the text annotation method combining RPA and AI as described above when executing the program.
In order to achieve the above object, a fourth aspect of the present application provides a non-transitory computer-readable storage medium, on which a computer program is stored, where the computer program, when executed by a processor, implements the text annotation method combining RPA and AI as described above.
Additional aspects and advantages of the present application will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the present application.
Drawings
The foregoing and/or additional aspects and advantages of the present application will become apparent and readily appreciated from the following description of the embodiments, taken in conjunction with the accompanying drawings of which:
FIG. 1 is a schematic flow chart of a text annotation method combining RPA and AI according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating a text annotation method according to another embodiment of the present application;
FIG. 3 is a flowchart illustrating a text annotation method in combination with RPA and AI according to another embodiment of the present application;
FIG. 4 is a schematic structural diagram of a text annotation device combining an RPA and an AI according to an embodiment of the present application;
FIG. 5 is a schematic structural diagram of a text annotation device combining an RPA and an AI according to another embodiment of the present application;
fig. 6 is a schematic structural diagram of a text annotation device combining an RPA and an AI according to an embodiment of the present application.
Detailed Description
Reference will now be made in detail to embodiments of the present application, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to the same or similar elements or elements having the same or similar function throughout. The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining the present application and should not be construed as limiting the present application.
The following describes a text labeling method and apparatus combining RPA and AI according to an embodiment of the present application with reference to the drawings. The main execution body of the text labeling method combining the RPA and the AI is a text labeling device combining the RPA and the AI.
Fig. 1 is a schematic flowchart of a text annotation method combining an RPA and an AI according to an embodiment of the present disclosure. As shown in fig. 1, the text annotation method combining RPA and AI includes the following steps:
In the embodiment of the present application, the chinese text to be labeled may be a chinese text corresponding to a related audio, a partial chinese text in a document, a chinese text on a network, or the like. Correspondingly, the method for acquiring the to-be-labeled Chinese text can be, but is not limited to, converting the audio into the corresponding Chinese text by converting the related audio, intercepting the Chinese text in the document, downloading the Chinese text on the network, and the like.
And then, generating a pinyin text which is composed of pinyin and corresponds to the Chinese text according to the acquired Chinese text to be labeled.
As an example, the pinyin corresponding to each word in the chinese text may be obtained by querying a pinyin library, and the pinyins corresponding to each word are combined to obtain a pinyin text corresponding to the chinese text. For example, the Chinese text to be labeled is "speech synthesis", and the corresponding pinyin text is "yu 3 yin1 he2 cheng 2", wherein the numbers represent pinyin tones.
And 102, determining pause information corresponding to the Chinese text by adopting an NLP technology, and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information.
In the embodiment of the application, the pause information corresponding to the Chinese text can be determined through the Chinese text to be labeled or the audio corresponding to the Chinese text to be labeled. For example, the semantic recognition can be performed on the Chinese to be labeled by the NLP technology, the word segmentation information is determined according to the semantic corresponding to the text to be labeled, and the word segmentation information is used as the pause information of the Chinese text to be labeled. For another example, the audio corresponding to the to-be-labeled chinese text may be analyzed by the NLP technique to determine the pause information corresponding to the audio, and the pause information is used as the pause information of the to-be-labeled chinese text corresponding to the audio. It should be noted that the pause information may be, but is not limited to, word segmentation information, and in general, the pause information is word segmentation information based on "prosodic words" or "prosodic phrases".
As an example, for example, a chinese text to be labeled is input into a pre-trained text pause recognition model, and the text pause recognition model can output pause information corresponding to the chinese text.
As another example, acquiring an audio corresponding to a to-be-labeled Chinese text; and identifying the audio and determining pause information in the audio. For example, the audio corresponding to the chinese text may be obtained, the audio is input to a pre-trained audio pause recognition model, the audio pause recognition model may output pause information corresponding to the audio, and the pause information corresponding to the audio is determined as the pause information corresponding to the chinese text. It should be noted that the audio pause recognition model may analyze the audio based on the NLP technology, and determine pause information corresponding to the audio.
And then, inserting the determined pause information into the Chinese text to obtain the Chinese text with the pause information. For example, the Chinese text to be labeled is: after inserting corresponding pause information, for example, word segmentation information, the exhibition hall obtains a Chinese text with pause information as follows: exhibition | museum.
And 103, marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking.
In the embodiment of the present application, the pinyin meeting the condition to be corrected may include, but is not limited to, any one or more of the following: the pinyin corresponding to polyphone, the pinyin corresponding to inflexion character and the pinyin corresponding to the word composed of continuous three-tone Chinese characters. Preferably, the word is obtained by segmenting the Chinese text according to the pause information.
The pinyin corresponding to the polyphone, the pinyin corresponding to the inflexion character, and the pinyin corresponding to the word composed of the continuous three-tone Chinese characters may be determined based on the pinyin text determined in step 101 and a preset determination mechanism. As an example, the pinyin texts determined in step 101 are matched based on a preset polyphone recognition library and an inflexion character recognition library, and the pinyins corresponding to the polyphone characters and the inflexion characters are determined. As an example, successive three-tone pinyins in the pinyin text determined in step 101 may be detected, and pinyins corresponding to words formed by successive three-tone chinese characters may be determined.
For example, for "a language", where "a" is a polyphone (zhong3, zhong4), and the pinyin corresponding to "a" is labeled, the labeled pinyin text after the labeling process is: yi1{ zhong3} yu3 yan 1. For another example, "a language", "a" is not a polyphone, but needs to be inflected in this context, and is a inflected character, where yi4 is read, and after inflectioning the pinyin corresponding to "a" and performing the marking process, the pinyin after the marking process is expressed as: { yi4} { zhong3} yu3 yan 1.
In addition, in the embodiment of the present application, as shown in fig. 2, when the condition to be corrected is that there is pinyin corresponding to a word composed of consecutive three-tone chinese characters, the pinyin corresponding to the word is subjected to sound change processing to obtain a pinyin text after sound change processing, and the pinyin after sound change processing is marked to obtain a pinyin text after marking processing, which is specifically implemented as follows:
In the embodiment of the application, the pinyin text corresponding to the Chinese text to be labeled can be intercepted to obtain the pinyin corresponding to the words consisting of the continuous three-tone Chinese characters in the pinyin text.
Optionally, acquiring a word sequence corresponding to the continuous three tones in the words consisting of the continuous three-tone Chinese characters; determining the pinyin of all characters before the last character in the character sequence as the pinyin of the sound to be changed; and performing two-tone processing on the pinyin to be varied in the pinyin corresponding to the word consisting of the continuous three-tone Chinese characters.
For example, a word composed of three consecutive tones of chinese characters is a word composed of more than 2 consecutive adjacent 3 tones of chinese characters, i.e. when the word "exhibition" is read by, for example, 2 consecutive 3 tones of chinese characters, the first 3 tones become 2 tones, for example: "exhibition (zhan3) overview (lan 3)", and the pinyin text after the sound change processing is as follows: "zhan 2 lan 3"; for another example, for more than 3 consecutive 3-pronouncing Chinese characters, all 3 pronunciations before the last 2 pronunciations become 2 pronunciations. For example: "exhibition (zhan3) visit (lan3) library (guan 3)", the pinyin text after the sound change processing is: "zhan 2 lan2 guan 3".
And 203, marking the pinyin subjected to the sound changing treatment in the pinyin text to obtain the pinyin text subjected to the marking treatment.
For example, for example: zhan2 lan2 guan3, wherein zhan2 lan2 is the pinyin subjected to the sound change processing, the pinyin subjected to the sound change processing is subjected to marking processing, and the text subjected to the marking processing is as follows: < zhan2> < lan2> guan 3.
And 104, providing the Chinese text to be marked, the Chinese text with the pause information and the pinyin text after marking processing for a user for correction.
Optionally, as shown in fig. 3, the chinese text to be labeled, the chinese text with pause information, and the pinyin text after marking processing are provided to the user for correction, and a pinyin labeling result corresponding to the chinese text to be labeled is obtained, which specifically implements the following process:
In the embodiment of the application, the Chinese text to be marked, the Chinese text with pause information and the pinyin text after marking processing can be sent to a user (such as a marking person), and when the user finds that the pinyin text after marking processing has errors, a pinyin correction request can be sent to a text marking device. And then, the text marking device corrects the pinyin text after the marking process according to the pinyin correction request to obtain the corrected pinyin text. The correction request may include, but is not limited to, a pinyin to be corrected and a corresponding corrected pinyin.
For example, take "an exhibition hall" as an example, wherein the Chinese text to be labeled is: an exhibition hall, the Chinese text with pause information is: a seat (1) exhibition hall; the pinyin text after marking processing is as follows: { yi2} zuo 4< zhan2> < lan2> guan 3; when text labeling is carried out, only a text needing to be labeled is displayed in a graphical mode, and parts in brackets (), angle brackets < >, or braces { } in the pinyin text after label processing, namely the parts needing manual inspection in the pinyin text, are displayed in a highlight mode. The annotator only needs to check whether the highlighted portion is correct.
And step 304, deleting the mark in the corrected pinyin text to obtain a pinyin marking result corresponding to the Chinese text to be marked.
And finally, deleting the mark in the corrected pinyin text to obtain a pinyin marking result corresponding to the marked text. For example, the Chinese text to be labeled is: an exhibition hall, the corresponding corrected pinyin is: { yi2} zuo 4< zhan2> < lan2> guan3, deleting the mark in the pinyin text, wherein the result of the pinyin mark corresponding to an exhibition hall is as follows: yi2 zuo 4zhan2 lan2 guan 3.
Optionally, the Chinese text to be marked, the Chinese text with pause information and the pinyin text marked and processed are provided for a user for correction, and after the pinyin marking result corresponding to the Chinese text to be marked is obtained, the pause information in the Chinese text with pause information can be inserted into the pinyin marking result to obtain the pinyin marking result with pause information; and training a preset voice synthesis system according to the audio corresponding to the Chinese text and the pinyin marking result with pause information to obtain the trained voice synthesis system.
That is, the pinyin labeling result with pause information can be used for building a speech synthesis system by matching with corresponding audio data. For example, a speech synthesis system based on neural networks has the ability to "learn". The pinyin marking result with pause information is used as the input of the system, the audio corresponding to the text is used as the output, and a large number of pinyin text-audio pairs are used as training data to train the system. Through training, the speech synthesis system based on the neural network can 'learn' pronunciation, tone and the like corresponding to the Chinese sentences. After training, only need input arbitrary Chinese in the front end of the speech synthesis system, the speech synthesis system can output the speech similar to real person's speech directly.
According to the text marking method combining the RPA and the AI, the Chinese text to be marked is obtained, and the pinyin text corresponding to the Chinese text is generated; determining pause information corresponding to the Chinese text by adopting an NLP technology, and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information; marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking; and providing the Chinese text to be marked, the Chinese text with pause information and the pinyin text marked and processed for a user to correct. The method preprocesses the Chinese text to be marked to generate the pinyin text to be corrected, marks the pinyin meeting the condition to be corrected in the pinyin text and provides the pinyin text to a user for correction, and the user only needs to correct the pinyin text after marking processing to obtain the pinyin marking result corresponding to the Chinese text to be marked with high accuracy, so that the user can be assisted in marking the text, and the marking efficiency and the accuracy of the text are improved.
Corresponding to the text labeling methods provided in the foregoing embodiments, an embodiment of the present application further provides a text labeling apparatus combining an RPA and an AI, and since the text labeling apparatus combining an RPA and an AI provided in the embodiment of the present application corresponds to the text labeling methods combining an RPA and an AI provided in the foregoing embodiments, the foregoing embodiments of the text labeling method combining an RPA and an AI are also applicable to the text labeling apparatus combining an RPA and an AI provided in the embodiment, and are not described in detail in the embodiment. Fig. 4 is a schematic structural diagram of a text annotation device combining an RPA and an AI according to an embodiment of the present application. As shown in fig. 4, the text labeling apparatus combining RPA and AI includes: an acquisition module 410, a determination module 420, a labeling module 430, and a providing module 440.
The obtaining module 410 is configured to obtain a chinese text to be labeled, and generate a pinyin text corresponding to the chinese text; the determining module 420 is configured to determine pause information corresponding to the chinese text by using an NLP technique, and insert the pause information into the chinese text to obtain the chinese text with the pause information; the marking module 430 is configured to mark pinyin meeting the condition to be corrected in the pinyin text to obtain a pinyin text after marking; a providing module 440, configured to provide the chinese text to be labeled, the chinese text with pause information, and the pinyin text after label processing to the user for correction.
As a possible implementation manner of the embodiment of the present application, the determining module 420 is specifically configured to obtain an audio corresponding to a to-be-labeled chinese text; identifying the audio, determining pause information in the audio, and determining the pause information in the audio as pause information corresponding to the Chinese text; and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information.
As a possible implementation manner of the embodiment of the present application, the condition to be corrected includes any one or more of the following conditions: the pinyin corresponding to polyphone, the pinyin corresponding to inflexion character and the pinyin corresponding to the word composed of continuous three-tone Chinese characters; the words are obtained by segmenting the Chinese text according to the pause information.
As a possible implementation manner of the embodiment of the present application, when the condition to be corrected is that pinyin corresponding to a term composed of consecutive three-tone chinese characters exists, the marking module 430 is specifically configured to obtain pinyin corresponding to a term composed of consecutive three-tone chinese characters in a pinyin text; performing voice change processing on pinyin corresponding to words consisting of continuous three-tone Chinese characters to obtain a pinyin text after the voice change processing; and marking the pinyin subjected to the sound changing treatment in the pinyin text to obtain the pinyin text subjected to marking treatment.
As a possible implementation manner of the embodiment of the application, the sound change processing of the pinyin corresponding to the words consisting of the continuous three-sound Chinese characters is to obtain word sequences corresponding to the continuous three-sound Chinese characters in the words consisting of the continuous three-sound Chinese characters; determining the pinyin of all characters before the last character in the character sequence as the pinyin of the sound to be changed; and performing two-tone processing on the pinyin to be varied in the pinyin corresponding to the word consisting of the continuous three-tone Chinese characters.
As a possible implementation manner of the embodiment of the present application, the providing module 440 is specifically configured to provide the chinese text to be labeled, the chinese text with pause information, and the pinyin text after the marking process to the user; receiving a pinyin correction request of a user, wherein the correction request comprises: the pinyin to be corrected and the corresponding corrected pinyin; correcting the pinyin text subjected to marking processing according to the pinyin correction request to obtain a corrected pinyin text; and deleting the mark in the corrected pinyin text to obtain a pinyin marking result corresponding to the Chinese text to be marked.
As a possible implementation manner of the embodiment of the present application, as shown in fig. 5, on the basis of fig. 4, the text annotation device combining the RPA and the AI further includes: an insertion module 450 and a training module 460.
The inserting module 450 is configured to insert the pause information in the Chinese text with pause information into the pinyin annotation result to obtain the pinyin annotation result with pause information; the training module 460 is configured to train a preset speech synthesis system according to the audio corresponding to the chinese text and the pinyin annotation result with the pause information, so as to obtain a trained speech synthesis system.
The text marking device combining the RPA and the AI in the embodiment of the application generates a pinyin text corresponding to a Chinese text by acquiring the Chinese text to be marked; determining pause information corresponding to the Chinese text by adopting an NLP technology, and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information; marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking; and providing the Chinese text to be marked, the Chinese text with pause information and the pinyin text marked and processed for a user to correct. The device can realize preprocessing the Chinese text to be labeled, generate the pinyin text to be corrected, mark the pinyin meeting the conditions to be corrected in the pinyin text, provide the pinyin text to the user for correction, and the user only needs to correct the pinyin text after marking processing, so that the pinyin labeling result corresponding to the Chinese text to be labeled with high accuracy can be obtained, the user can be assisted in labeling the text, and the labeling efficiency and the accuracy of the text are improved.
In order to implement the foregoing embodiments, the present application further provides a text annotation device combining an RPA and an AI, and fig. 6 is a schematic structural diagram of a text annotation device combining an RPA and an AI according to an embodiment of the present application. The text labeling device combining the RPA and the AI comprises:
The processor 1002, when executing the program, implements the text annotation method combining the RPA and the AI provided in the above embodiments.
Further, the text labeling device combining the RPA and the AI further comprises:
a communication interface 1003 for communicating between the memory 1001 and the processor 1002.
A memory 1001 for storing computer programs that may be run on the processor 1002.
The processor 1002 is configured to implement the text labeling method combining RPA and AI according to the foregoing embodiment when executing the program.
If the memory 1001, the processor 1002, and the communication interface 1003 are implemented independently, the communication interface 1003, the memory 1001, and the processor 1002 may be connected to each other through a bus and perform communication with each other. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is shown in FIG. 6, but this is not intended to represent only one bus or type of bus.
Optionally, in a specific implementation, if the memory 1001, the processor 1002, and the communication interface 1003 are integrated on one chip, the memory 1001, the processor 1002, and the communication interface 1003 may complete communication with each other through an internal interface.
The processor 1002 may be a Central Processing Unit (CPU), an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits configured to implement embodiments of the present Application.
In order to implement the foregoing embodiments, the present application further proposes a non-transitory computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the text annotation method combining RPA and AI as described in the foregoing embodiments.
In the description herein, reference to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the application. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present application, "plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing steps of a custom logic function or process, and alternate implementations are included within the scope of the preferred embodiment of the present application in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present application.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. If implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present application may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application, and that variations, modifications, substitutions and alterations may be made to the above embodiments by those of ordinary skill in the art within the scope of the present application.
Claims (10)
1. A text labeling method combining RPA and AI is characterized by comprising the following steps:
s1, acquiring a Chinese text to be labeled, and generating a pinyin text corresponding to the Chinese text;
s2, determining pause information corresponding to the Chinese text by adopting a Natural Language Processing (NLP) technology, and inserting the pause information into the Chinese text to obtain the Chinese text with the pause information;
s3, marking the pinyin meeting the condition to be corrected in the pinyin text to obtain a pinyin text after marking;
s4, providing the Chinese text to be marked, the Chinese text with pause information and the pinyin text after marking processing for a user to correct.
2. The method according to claim 1, wherein the determining the pause information corresponding to the chinese text by the NLP technique, and inserting the pause information into the chinese text to obtain the chinese text with pause information comprises:
s21, acquiring the audio corresponding to the Chinese text to be labeled;
s22, identifying the audio, determining pause information in the audio, and determining the pause information in the audio as pause information corresponding to the Chinese text;
and S23, inserting the pause information into the Chinese text to obtain the Chinese text with the pause information.
3. The method according to claim 1, wherein the condition to be corrected comprises any one or more of the following conditions: the pinyin corresponding to polyphone, the pinyin corresponding to inflexion character and the pinyin corresponding to the word composed of continuous three-tone Chinese characters;
the words are obtained by segmenting the Chinese text according to the pause information.
4. The method as claimed in claim 3, wherein when the condition to be corrected is existence of pinyin corresponding to a word consisting of consecutive three-tone Chinese characters, the marking of the pinyin meeting the condition to be corrected in the pinyin text to obtain a pinyin text after marking, includes:
s31, obtaining pinyin corresponding to words formed by continuous three-tone Chinese characters in the pinyin text;
s32, performing voice change processing on the pinyin corresponding to the words consisting of the continuous three-tone Chinese characters to obtain a pinyin text after the voice change processing;
s33, marking the pinyin subjected to the sound changing processing in the pinyin text to obtain the pinyin text subjected to marking processing.
5. The method of claim 4, wherein the performing of the voicing process on the pinyin corresponding to the word consisting of the three consecutive Chinese characters comprises:
s321, acquiring a word sequence corresponding to continuous three tones in the words consisting of the continuous three-tone Chinese characters;
s322, determining the pinyin of all characters before the last character in the character sequence as the pinyin of the sound to be changed;
s323, the pinyin of the sound to be changed in the pinyin corresponding to the words consisting of the continuous three-sound Chinese characters is processed by two sound.
6. The method according to claim 1, wherein the providing the chinese text to be labeled, the chinese text with pause information, and the pinyin text after labeling processing to a user for correction to obtain a pinyin labeling result corresponding to the chinese text to be labeled comprises:
s41, providing the Chinese text to be labeled, the Chinese text with pause information and the pinyin text after marking processing for a user;
s42, receiving a pinyin correction request of the user, wherein the correction request comprises: the pinyin to be corrected and the corresponding corrected pinyin;
s43, correcting the pinyin text after the marking processing according to the pinyin correction request to obtain a corrected pinyin text;
s44, deleting the mark in the corrected Pinyin text to obtain the Pinyin marking result corresponding to the Chinese text to be marked.
7. The method according to claim 1, wherein after providing the chinese text to be labeled, the chinese text with pause information, and the pinyin text after labeling processing for user correction and obtaining a pinyin labeling result corresponding to the chinese text to be labeled, the method further comprises:
s5, inserting pause information in the Chinese text with the pause information into the pinyin annotation result to obtain the pinyin annotation result with the pause information;
s6, training a preset voice synthesis system according to the audio corresponding to the Chinese text and the pinyin marking result with the pause information to obtain the trained voice synthesis system.
8. A text labeling device combining RPA and AI, comprising:
the system comprises an acquisition module, a marking module and a marking module, wherein the acquisition module is used for acquiring a Chinese text to be marked and generating a pinyin text corresponding to the Chinese text;
the determining module is used for determining the pause information corresponding to the Chinese text by adopting an NLP technology, inserting the pause information into the Chinese text and obtaining the Chinese text with the pause information;
the marking module is used for marking the pinyin meeting the condition to be corrected in the pinyin text to obtain the pinyin text after marking;
and the providing module is used for providing the Chinese text to be marked, the Chinese text with the pause information and the pinyin text after marking processing for a user for correction.
9. A text annotation device that combines RPA and AI, comprising:
memory, processor and computer program stored on the memory and executable on the processor, characterized in that the processor, when executing the program, implements the text annotation method according to any one of claims 1 to 7 in combination with RPA and AI.
10. A non-transitory computer-readable storage medium, on which a computer program is stored, which, when executed by a processor, implements the method for text annotation combining RPA and AI according to any one of claims 1 to 7.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911329215 | 2019-12-20 | ||
| CN2019113292157 | 2019-12-20 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112069805A true CN112069805A (en) | 2020-12-11 |
Family
ID=73683790
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011062863.3A Pending CN112069805A (en) | 2019-12-20 | 2020-09-30 | Text labeling method, device, equipment and storage medium combining RPA and AI |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112069805A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112632967A (en) * | 2020-12-30 | 2021-04-09 | 广东德诚科教有限公司 | Chinese pinyin automatic generation method and device oriented to set strategy |
| CN112669814A (en) * | 2020-12-17 | 2021-04-16 | 北京猎户星空科技有限公司 | Data processing method, device, equipment and medium |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1116336A (en) * | 1994-11-03 | 1996-02-07 | 王昭宁 | Substitution type Chinese phonetic character, word input coding method and keyboard thereof |
| CN1282072A (en) * | 1999-07-27 | 2001-01-31 | 国际商业机器公司 | Error correcting method for voice identification result and voice identification system |
| CN1945657A (en) * | 2006-11-01 | 2007-04-11 | 邓云泽 | Method for really marking phonetic notation in English quick oral sentence |
| CN101141666A (en) * | 2006-09-05 | 2008-03-12 | 中兴通讯股份有限公司 | Method of converting text note to voice broadcast in mobile phone |
| CN101819469A (en) * | 2009-11-06 | 2010-09-01 | 无敌科技(西安)有限公司 | Method for correcting Chinese content spelling |
| CN102478968A (en) * | 2010-11-23 | 2012-05-30 | 腾讯科技（深圳）有限公司 | Chinese pinyin input method and chinese pinyin input system |
| CN105632484A (en) * | 2016-02-19 | 2016-06-01 | 上海语知义信息技术有限公司 | Voice synthesis database pause information automatic marking method and system |
| CN107203508A (en) * | 2016-03-17 | 2017-09-26 | 富士施乐实业发展（中国）有限公司 | Braille document generating method and system |
| CN107678561A (en) * | 2017-09-29 | 2018-02-09 | 百度在线网络技术（北京）有限公司 | Phonetic entry error correction method and device based on artificial intelligence |
| CN109065031A (en) * | 2018-08-02 | 2018-12-21 | 阿里巴巴集团控股有限公司 | Voice annotation method, device and equipment |
| CN109977361A (en) * | 2019-03-01 | 2019-07-05 | 广州多益网络股份有限公司 | A kind of Chinese phonetic alphabet mask method, device and storage medium based on similar word |
| CN110083711A (en) * | 2019-05-13 | 2019-08-02 | 成都启英泰伦科技有限公司 | A kind of phonetic transcriptions of Chinese characters conversion method and converting system |
| CN110390000A (en) * | 2019-07-30 | 2019-10-29 | 同方赛威讯信息技术有限公司 | A kind of legal documents automatic identification generates system and method |
| CN110534089A (en) * | 2019-07-10 | 2019-12-03 | 西安交通大学 | A kind of Chinese speech synthesis method based on phoneme and rhythm structure |
| CN110556093A (en) * | 2019-09-17 | 2019-12-10 | 浙江核新同花顺网络信息股份有限公司 | Voice marking method and system |
-
2020
- 2020-09-30 CN CN202011062863.3A patent/CN112069805A/en active Pending
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1116336A (en) * | 1994-11-03 | 1996-02-07 | 王昭宁 | Substitution type Chinese phonetic character, word input coding method and keyboard thereof |
| CN1282072A (en) * | 1999-07-27 | 2001-01-31 | 国际商业机器公司 | Error correcting method for voice identification result and voice identification system |
| CN101141666A (en) * | 2006-09-05 | 2008-03-12 | 中兴通讯股份有限公司 | Method of converting text note to voice broadcast in mobile phone |
| CN1945657A (en) * | 2006-11-01 | 2007-04-11 | 邓云泽 | Method for really marking phonetic notation in English quick oral sentence |
| CN101819469A (en) * | 2009-11-06 | 2010-09-01 | 无敌科技(西安)有限公司 | Method for correcting Chinese content spelling |
| CN102478968A (en) * | 2010-11-23 | 2012-05-30 | 腾讯科技（深圳）有限公司 | Chinese pinyin input method and chinese pinyin input system |
| CN105632484A (en) * | 2016-02-19 | 2016-06-01 | 上海语知义信息技术有限公司 | Voice synthesis database pause information automatic marking method and system |
| CN107203508A (en) * | 2016-03-17 | 2017-09-26 | 富士施乐实业发展（中国）有限公司 | Braille document generating method and system |
| CN107678561A (en) * | 2017-09-29 | 2018-02-09 | 百度在线网络技术（北京）有限公司 | Phonetic entry error correction method and device based on artificial intelligence |
| CN109065031A (en) * | 2018-08-02 | 2018-12-21 | 阿里巴巴集团控股有限公司 | Voice annotation method, device and equipment |
| CN109977361A (en) * | 2019-03-01 | 2019-07-05 | 广州多益网络股份有限公司 | A kind of Chinese phonetic alphabet mask method, device and storage medium based on similar word |
| CN110083711A (en) * | 2019-05-13 | 2019-08-02 | 成都启英泰伦科技有限公司 | A kind of phonetic transcriptions of Chinese characters conversion method and converting system |
| CN110534089A (en) * | 2019-07-10 | 2019-12-03 | 西安交通大学 | A kind of Chinese speech synthesis method based on phoneme and rhythm structure |
| CN110390000A (en) * | 2019-07-30 | 2019-10-29 | 同方赛威讯信息技术有限公司 | A kind of legal documents automatic identification generates system and method |
| CN110556093A (en) * | 2019-09-17 | 2019-12-10 | 浙江核新同花顺网络信息股份有限公司 | Voice marking method and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112669814A (en) * | 2020-12-17 | 2021-04-16 | 北京猎户星空科技有限公司 | Data processing method, device, equipment and medium |
| CN112632967A (en) * | 2020-12-30 | 2021-04-09 | 广东德诚科教有限公司 | Chinese pinyin automatic generation method and device oriented to set strategy |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107731228B (en) | Text conversion method and device for English voice information | |
| CN103714048B (en) | Method and system for correcting text | |
| JP6581356B2 (en) | Speech synthesis method and apparatus based on large-scale corpus | |
| JP5413622B2 (en) | Language model creation device, language model creation method, and program | |
| WO2019198386A1 (en) | Request rephrasing system, method for training of request rephrasing model and of request determination model, and conversation system | |
| CN108984679B (en) | Training method and device for dialogue generation model | |
| CN109410913B (en) | Voice synthesis method, device, equipment and storage medium | |
| CN108804526B (en) | Interest determination system, interest determination method, and storage medium | |
| CN108846124B (en) | Training method, training device, computer equipment and readable storage medium | |
| JP2006031228A (en) | Morphemic analysis device, method, and program | |
| CN112069805A (en) | Text labeling method, device, equipment and storage medium combining RPA and AI | |
| CN109656554B (en) | User interface generation method and device | |
| CN112016303B (en) | Text error correction method, device, equipment and storage medium based on graphic neural network | |
| CN112242185A (en) | Medical image report automatic generation method and system based on deep learning | |
| CN110335608B (en) | Voiceprint verification method, voiceprint verification device, voiceprint verification equipment and storage medium | |
| CN109166569B (en) | Detection method and device for phoneme mislabeling | |
| CN112669845B (en) | Speech recognition result correction method and device, electronic equipment and storage medium | |
| CN112241629A (en) | Pinyin annotation text generation method and device combining RPA and AI | |
| JP2016224483A (en) | Model learning device, method and program | |
| CN110188327B (en) | Method and device for removing spoken language of text | |
| CN110135583B (en) | Method and device for generating label information and electronic equipment | |
| CN108829896B (en) | Reply information feedback method and device | |
| CN108897872B (en) | Dialogue processing method, device, computer equipment and storage medium | |
| CN111833847A (en) | Speech processing model training method and device | |
| CN115908775A (en) | Chemical structural formula identification method and device, storage medium and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |