Children literature material voice library management method and system based on big data, and medium

Abstract

The embodiment of the invention provides a children literature data voice library management method and system based on big data and a medium, and the method comprises the steps: obtaining literature data, inputting the literature data into a voice generation model, and outputting voice information; analyzing the voice information to obtain analysis information, and analyzing the matching degree of the voice information and the literature data according to the analysis information; judging whether the matching degree is greater than a set matching degree threshold value or not; if yes, classifying the voice information to obtain a voice category, and storing the voice information according to the voice category; if yes, adjusting model parameters of the voice generation model; the literature data is subjected to voice conversion through the voice generation model, so that early education voice prompt of children is realized, after voice information conversion, the voice information is subjected to reverse analysis, and the matching of the voice information and the literature data is analyzed, so that accurate conversion is completed, and the conversion precision is improved.

Description

Technical Field

[0001] This application relates to the field of voice conversion storage management of literary materials, and more specifically, to a method, system, and medium for managing a children's literary material voice library based on big data. Background Art

[0002] In children's early education course learning, children generally cannot read text, and it is necessary to convert text materials into voice so as to carry out children's education in the way of playing voice. In the existing literary material voice management methods, voice output cannot be performed through a voice generation model, resulting in large errors in the process of converting literary materials, affecting the output of the semantics of literary materials, and having an adverse impact on children's education; in view of the above problems, effective technical solutions are urgently needed. Summary of the Invention

[0003] The purpose of the embodiments of this application is to provide a method, system, and medium for managing a children's literary material voice library based on big data. The literary materials are converted into voice through a voice generation model, so as to realize voice prompts for children's early education. After the voice information is converted, the voice information is reversely parsed, and the matching degree between the voice information and the literary materials is analyzed, so as to complete accurate conversion and improve the conversion accuracy.

[0004] The embodiments of this application also provide a method for managing a children's literary material voice library based on big data, including:

[0005] Obtain literary materials, input the literary materials into a voice generation model, and output voice information;

[0006] Parse and process the voice information to obtain parsed information, and analyze the matching degree between the voice information and the literary materials according to the parsed information;

[0007] Judge whether the matching degree is greater than a set matching degree threshold;

[0008] If it is greater, classify the voice information to obtain a voice category, and store the voice information according to the voice category;

[0009] If it is less, adjust the model parameters of the voice generation model.

[0010] Optionally, in the method for managing a children's literary material voice library based on big data described in the embodiments of this application, obtaining literary materials, inputting the literary materials into a voice generation model, and outputting voice information specifically includes:

[0011] Establish a training set through big data, and iteratively train the voice generation model through the training set to obtain a training result;

[0012] Judge whether the training result converges;

[0013] If it converges, input the literary materials into the trained speech generation model to obtain speech information;

[0014] If it does not converge, generate extended information, expand the data in the training set according to the extended information, and perform secondary training on the speech generation model.

[0015] Optionally, in the method for managing a speech library of children's literary materials based on big data described in the embodiments of the present application, the speech information is parsed to obtain parsing information, which specifically includes:

[0016] Obtain the speech information, perform slicing processing on the speech information to obtain a number of slices;

[0017] Perform semantic analysis on the text according to the sentences, grammar, and voices in the literary materials, extract the semantic information in the slices respectively, and convert it into text;

[0018] Connect the text and determine whether the text description is smooth;

[0019] If it is smooth, generate parsing information;

[0020] If it is not smooth, adjust the text connection relationship.

[0021] Optionally, in the method for managing a speech library of children's literary materials based on big data described in the embodiments of the present application, analyze the matching degree between the speech information and the literary materials according to the parsing information, which specifically includes:

[0022] According to the parsing information, reverse parse the speech information, perform semantic analysis on the parsing information to obtain semantic information;

[0023] Obtain the literary materials, analyze according to the literary materials to obtain text information;

[0024] Perform semantic matching between the text information and the semantic information to determine the semantic difference degree;

[0025] Generate the matching degree between the speech information and the literary materials according to the speech difference degree.

[0026] Optionally, in the method for managing a speech library of children's literary materials based on big data described in the embodiments of the present application, if it is greater, classify the speech information to obtain a speech category, and store the speech information according to the speech category, which specifically includes:

[0027] Obtain the speech information, perform semantic parsing on the speech information to obtain semantic information;

[0028] Match the corresponding literary materials according to the semantic information, classify according to the content of the literary materials to obtain category information;

[0029] Store the voice information of literary materials of the same category within the same category range;

[0030] Generate a title header according to the category of the literary material, match the title header to the corresponding category range, and generate a storage area with the title header.

[0031] Optionally, in the method for managing a voice library of children's literary materials based on big data described in the embodiments of the present application, if it is less than, then adjust the model parameters of the voice generation model, specifically including:

[0032] Divide the set matching degree threshold into a first matching degree threshold and a second matching degree threshold, where the first matching degree threshold is less than the second matching degree threshold;

[0033] If the matching degree is greater than the first matching degree threshold and less than the second matching degree threshold, then generate a first adjustment coefficient, and adjust the model parameters of the voice generation model according to the first adjustment coefficient;

[0034] If the matching degree is greater than the second matching degree threshold, then generate a second adjustment coefficient, and adjust the model parameters of the voice generation model according to the second adjustment coefficient.

[0035] In a second aspect, an embodiment of the present application provides a system for managing a voice library of children's literary materials based on big data. The system includes: a memory and a processor. The memory includes a program for the method for managing a voice library of children's literary materials based on big data. When the program for the method for managing a voice library of children's literary materials based on big data is executed by the processor, the following steps are implemented:

[0036] Obtain literary materials, input the literary materials into a voice generation model, and output voice information;

[0037] Parse and process the voice information to obtain parsed information, and analyze the matching degree between the voice information and the literary materials according to the parsed information;

[0038] Judge whether the matching degree is greater than a set matching degree threshold;

[0039] If it is greater than, then classify the voice information to obtain a voice category, and store the voice information according to the voice category;

[0040] If it is less than, then adjust the model parameters of the voice generation model.

[0041] Optionally, in the system for managing a voice library of children's literary materials based on big data described in the embodiments of the present application, obtaining literary materials, inputting the literary materials into a voice generation model, and outputting voice information specifically includes:

[0042] Establish a training set through big data, and iteratively train the voice generation model through the training set to obtain a training result;

[0043] Determine whether the training result converges;

[0044] If it converges, input the literary materials into the trained speech generation model to obtain speech information;

[0045] If it does not converge, generate extended information, expand the data in the training set according to the extended information, and perform secondary training on the speech generation model.

[0046] Optionally, in the big data-based children's literary materials speech library management system described in the embodiments of the present application, the speech information is parsed to obtain parsing information, which specifically includes:

[0047] Obtain the speech information, perform slicing processing on the speech information to obtain a number of slices;

[0048] Perform semantic analysis on the text according to the sentences, grammar, and voice in the literary materials, extract the semantic information in the slices respectively, and convert it into text;

[0049] Connect the text and determine whether the text description is smooth;

[0050] If it is smooth, generate parsing information;

[0051] If it is not smooth, adjust the text connection relationship.

[0052] In a third aspect, the embodiments of the present application also provide a computer-readable storage medium, which includes a program for the big data-based children's literary materials speech library management method. When the program for the big data-based children's literary materials speech library management method is executed by a processor, the steps of the big data-based children's literary materials speech library management method described in any one of the above are implemented.

[0053] As can be seen from the above, a big data-based children's literary materials speech library management method, system, and medium provided by the embodiments of the present application obtain literary materials, input the literary materials into a speech generation model, and output speech information; parse the speech information to obtain parsing information, and analyze the matching degree between the speech information and the literary materials according to the parsing information; determine whether the matching degree is greater than a set matching degree threshold; if it is greater, classify the speech information to obtain a speech category, and store the speech information according to the speech category; if it is less, adjust the model parameters of the speech generation model; convert the literary materials into speech through the speech generation model, so as to realize children's early education voice prompts. After the speech information is converted, the speech information is reversely parsed, and the matching degree between the speech information and the literary materials is analyzed, so as to complete accurate conversion and improve the conversion accuracy. Description of the Drawings

[0054] To more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, without creative efforts, other relevant drawings can also be obtained based on these drawings.

[0055] Figure 1 It is a flowchart of a method for managing a speech library of children's literature materials based on big data provided by an embodiment of the present application;

[0056] Figure 2 It is a flowchart of a method for training a speech generation model of a method for managing a speech library of children's literature materials based on big data provided by an embodiment of the present application;

[0057] Figure 3 It is a flowchart of a method for generating parsing information of a method for managing a speech library of children's literature materials based on big data provided by an embodiment of the present application. Specific Embodiments

[0058] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, rather than all of them. The components of the embodiments of the present application described and illustrated in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the present application to be protected, but only represents the selected embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present application without creative efforts fall within the scope of protection of the present application.

[0059] It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings. At the same time, in the description of the present application, terms such as "first" and "second" are only used for distinguishing descriptions and cannot be understood as indicating or implying relative importance.

[0060] Please refer to Figure 1 , Figure 1 which is a flowchart of a method for managing a speech library of children's literature materials based on big data in some embodiments of the present application. This method for managing a speech library of children's literature materials based on big data is used in a terminal device and includes the following steps:

[0061] S101, obtain literary materials, input the literary materials into a speech generation model, and output speech information;

[0062] S102, Parse and process the voice information to obtain parsed information, and analyze the matching degree between the voice information and literary materials according to the parsed information;

[0063] S103, Determine whether the matching degree is greater than a set matching degree threshold;

[0064] S104, If it is greater, classify the voice information to obtain a voice category, and store the voice information according to the voice category;

[0065] S105, If it is less, adjust the model parameters of the voice generation model.

[0066] It should be noted that through the voice generation model to perform voice translation on literary materials to achieve the convenience of children's education. During the voice generation process, in order to ensure the accuracy of voice translation, the voice information is reversely parsed to analyze whether the expression of the voice information is accurate, so as to dynamically adjust the model parameters and improve the output accuracy of the voice generation model.

[0067] Please refer to Figure 2 , Figure 2 is a flowchart of a voice generation model training method for a voice library management method of children's literary materials based on big data in some embodiments of the present application. According to the embodiments of the present invention, obtain literary materials, input the literary materials into the voice generation model, and output voice information, specifically including:

[0068] S201, Establish a training set through big data, and perform iterative training on the voice generation model through the training set to obtain a training result;

[0069] S202, Determine whether the training result converges;

[0070] S203, If it converges, input the literary materials into the trained voice generation model to obtain voice information;

[0071] S204, If it does not converge, generate extension information, expand the data in the training set according to the extension information, and perform secondary training on the voice generation model.

[0072] It should be noted that through the training set generated by big data to continuously train the voice generation model, improve the accuracy of the output result of the voice generation model, ensure the accuracy of literary material translation, and enable the voice information output by the voice generation model to accurately translate and output literary material information.

[0073] Please refer to Figure 3 , Figure 3It is a flowchart of an analysis information generation method for a voice library management method of children's literature materials based on big data in some embodiments of the present application. According to the embodiments of the present invention, the voice information is parsed to obtain parsing information, specifically including:

[0074] S301, obtain the voice information, perform slicing processing on the voice information to obtain a plurality of slices;

[0075] S302, perform semantic analysis on the text according to the sentences, grammar, and voice in the literature materials, extract the semantic information in the slices respectively, and convert it into text;

[0076] S303, connect the text and judge whether the text description is smooth;

[0077] S304, if it is smooth, generate parsing information;

[0078] S305, if it is not smooth, adjust the text connection relationship.

[0079] It should be noted that the voice information is sliced, semantic analysis is performed on each slice, so as to convert the semantic information into text, judge whether the text description is smooth, and thus accurately judge the voice information according to the analysis results.

[0080] According to the embodiments of the present invention, analyze the matching degree between the voice information and the literature materials according to the parsing information, specifically including:

[0081] Reverse-parsing the voice information according to the parsing information, performing semantic analysis on the parsing information to obtain semantic information;

[0082] Obtain the literature materials, perform analysis according to the literature materials to obtain text information;

[0083] Perform semantic matching between the text information and the semantic information to judge the semantic difference degree;

[0084] Generate the matching degree between the voice information and the literature materials according to the voice difference degree.

[0085] It should be noted that the voice information of the voice information is matched with the text information of the literature materials to judge whether the voice information can accurately express the literature materials, so that the voice information can highly match the literature materials.

[0086] According to the embodiments of the present invention, if it is greater than, classify the voice information to obtain a voice category, and store the voice information according to the voice category, specifically including:

[0087] Obtain the voice information, perform semantic parsing on the voice information to obtain semantic information;

[0088] Match the corresponding literary materials according to semantic information, classify them according to the content of the literary materials, and obtain category information;

[0089] Store the voice information of literary materials in the same category range;

[0090] Generate a title header according to the category of the literary materials, match the title header to the corresponding category range, and generate a storage area with the title header.

[0091] It should be noted that literary materials are matched according to the semantics of voice information, classified according to the content of different literary materials, stored in different storage areas, and a title header is established for the storage area to facilitate the retrieval of literary materials.

[0092] According to an embodiment of the present invention, if it is less than, adjust the model parameters of the voice generation model, specifically including:

[0093] Divide the set matching degree threshold into a first matching degree threshold and a second matching degree threshold, and the first matching degree threshold is less than the second matching degree threshold;

[0094] If the matching degree is greater than the first matching degree threshold and less than the second matching degree threshold, generate a first adjustment coefficient, and adjust the model parameters of the voice generation model according to the first adjustment coefficient;

[0095] If the matching degree is greater than the second matching degree threshold, generate a second adjustment coefficient, and adjust the model parameters of the voice generation model according to the second adjustment coefficient.

[0096] It should be noted that different matching degrees generate different adjustment coefficients, and the model parameters are adjusted according to different adjustment coefficients to improve the accuracy of the model parameters and ensure the accuracy of the voice generation model.

[0097] According to an embodiment of the present invention, it further includes: obtaining retrieval information, retrieving the title header of the storage area according to the retrieval information, and obtaining the corresponding storage area;

[0098] Obtain the voice information and the content of the literary materials in the storage area;

[0099] Analyze the retrieval accuracy according to the voice information and the content of the literary materials;

[0100] Judge whether the retrieval accuracy is greater than or equal to the set accuracy threshold;

[0101] If it is greater than or equal to, output the content of the literary materials and the voice information in the storage area;

[0102] If it is less than, generate correction information, and adjust the title header of the storage area according to the correction information.

[0103] It should be noted that when educating children of different ages, the title headers of the storage area are retrieved by retrieving information, so as to quickly obtain literary materials. During the process of obtaining materials, it is analyzed whether the literary materials in the retrieved storage area are the desired materials, and then the title headers are dynamically adjusted according to the retrieval results to ensure a higher matching degree between the title headers and the literary materials in the storage area.

[0104] Secondly, the embodiment of the present application provides a speech database management system for children's literary materials based on big data. The system includes: a memory and a processor. The memory includes a program of a method for managing a speech database of children's literary materials based on big data. When the program of the method for managing a speech database of children's literary materials based on big data is executed by the processor, the following steps are implemented:

[0105] Obtain literary materials, input the literary materials into a speech generation model, and output speech information;

[0106] Parse and process the speech information to obtain parsing information, and analyze the matching degree between the speech information and the literary materials according to the parsing information;

[0107] Judge whether the matching degree is greater than a set matching degree threshold;

[0108] If it is greater than, classify the speech information to obtain a speech category, and store the speech information according to the speech category;

[0109] If it is less than, adjust the model parameters of the speech generation model.

[0110] It should be noted that through the speech generation model, speech translation of literary materials is carried out to realize the convenience of children's education. During the process of speech generation, in order to ensure the accuracy of speech translation, the speech information is reversely parsed to analyze whether the expression of the speech information is accurate, and then the model parameters are dynamically adjusted to improve the output accuracy of the speech generation model.

[0111] According to the embodiment of the present invention, obtaining literary materials, inputting the literary materials into a speech generation model, and outputting speech information specifically includes:

[0112] Establish a training set through big data, and iteratively train the speech generation model through the training set to obtain a training result;

[0113] Judge whether the training result converges;

[0114] If it converges, input the literary materials into the trained speech generation model to obtain speech information;

[0115] If it does not converge, generate extension information, expand the data in the training set according to the extension information, and perform secondary training on the speech generation model.

[0116] It should be noted that by continuously training the speech generation model with a training set generated by big data, the accuracy of the output result of the speech generation model is improved, the accuracy of literary material translation is ensured, and the speech information output by the speech generation model can accurately translate the literary material information.

[0117] According to an embodiment of the present invention, the speech information is parsed to obtain parsed information, which specifically includes:

[0118] Obtain the speech information, slice the speech information to obtain a number of slices;

[0119] Perform semantic analysis on the text according to the sentences, grammar, and voice in the literary material, extract the semantic information in the slices respectively, and convert it into text;

[0120] Connect the text and determine whether the text description is smooth;

[0121] If it is smooth, generate parsed information;

[0122] If it is not smooth, adjust the text connection relationship.

[0123] It should be noted that the speech information is sliced, semantic analysis is performed on each slice, so as to convert the semantic information into text, determine whether the text description is smooth, and thus judge the accuracy of the speech information according to the analysis result.

[0124] According to an embodiment of the present invention, the matching degree between the speech information and the literary material is analyzed according to the parsed information, which specifically includes:

[0125] Reverse-parsing the speech information according to the parsed information, performing semantic analysis on the parsed information to obtain semantic information;

[0126] Obtain the literary material, analyze it according to the literary material to obtain text information;

[0127] Perform semantic matching between the text information and the semantic information to judge the semantic difference degree;

[0128] Generate the matching degree between the speech information and the literary material according to the speech difference degree.

[0129] It should be noted that the speech information of the speech information is matched with the text information of the literary material to judge whether the speech information can accurately express the literary material, so that the speech information can highly match the literary material.

[0130] According to an embodiment of the present invention, if it is greater, the speech information is classified to obtain a speech category, and the speech information is stored according to the speech category, which specifically includes:

[0131] Obtain voice information, perform semantic analysis on the voice information to obtain semantic information;

[0132] Match corresponding literary materials according to the semantic information, classify them according to the content of the literary materials to obtain category information;

[0133] Store the voice information of the literary materials of the same category within the same category range;

[0134] Generate a title header according to the category of the literary materials, match the title header to the corresponding category range, and generate a storage area with the title header.

[0135] It should be noted that, according to the semantics of the voice information, literary materials are matched, classified according to the content of different literary materials, different categories of literary materials are stored in different storage areas, and a title header is established for the storage area to facilitate the retrieval of literary materials.

[0136] According to an embodiment of the present invention, if it is less than, then adjust the model parameters of the voice generation model, specifically including:

[0137] Divide the set matching degree threshold into a first matching degree threshold and a second matching degree threshold, and the first matching degree threshold is less than the second matching degree threshold;

[0138] If the matching degree is greater than the first matching degree threshold and less than the second matching degree threshold, then generate a first adjustment coefficient, and adjust the model parameters of the voice generation model according to the first adjustment coefficient;

[0139] If the matching degree is greater than the second matching degree threshold, then generate a second adjustment coefficient, and adjust the model parameters of the voice generation model according to the second adjustment coefficient.

[0140] It should be noted that different matching degrees generate different adjustment coefficients, and the model parameters are adjusted according to different adjustment coefficients to improve the accuracy of the model parameters and ensure the accuracy of the voice generation model.

[0141] According to an embodiment of the present invention, it further includes: obtaining retrieval information, retrieving the title header of the storage area according to the retrieval information to obtain the corresponding storage area;

[0142] Obtain the voice information and the content of the literary materials in the storage area;

[0143] Analyze the retrieval accuracy according to the voice information and the content of the literary materials;

[0144] Judge whether the retrieval accuracy is greater than or equal to the set accuracy threshold;

[0145] If it is greater than or equal to, then output the content of the literary materials and the voice information in the storage area;

[0146] If it is less than, correction information is generated, and the title of the storage area is adjusted according to the correction information.

[0147] It should be noted that when educating children of different ages, the title of the storage area is retrieved by retrieving information, so as to quickly obtain literary materials. During the process of obtaining materials, it is analyzed whether the literary materials in the retrieved storage area are the desired materials, and then the title is dynamically adjusted according to the retrieval result to ensure a higher matching degree between the title and the literary materials in the storage area.

[0148] The third aspect of the present invention provides a computer-readable storage medium, which includes a program for managing a speech library of children's literary materials based on big data. When the program for managing the speech library of children's literary materials based on big data is executed by a processor, the steps of the method for managing the speech library of children's literary materials based on big data as described in any one of the above are implemented.

[0149] A method, system and medium for managing a speech library of children's literary materials based on big data disclosed by the present invention obtain literary materials, input the literary materials into a speech generation model to output speech information; parse and process the speech information to obtain parsing information, and analyze the matching degree between the speech information and the literary materials according to the parsing information; judge whether the matching degree is greater than a set matching degree threshold; if it is greater than, classify the speech information to obtain a speech category, and store the speech information according to the speech category; if it is less than, adjust the model parameters of the speech generation model; convert the literary materials into speech through the speech generation model, so as to realize children's early education speech prompts. After the speech information is converted, the speech information is reversely parsed, and the matching degree between the speech information and the literary materials is analyzed, so as to complete accurate conversion and improve the conversion accuracy.

[0150] In several embodiments provided by the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only illustrative. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored, or not executed. In addition, the coupling, direct coupling or communication connection between the components shown or discussed with each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0151] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0152] In addition, each functional unit in the embodiments of the present invention may be fully integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above integrated unit may be implemented in the form of hardware, or in the form of a combination of hardware and software functional units.

[0153] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions. The foregoing program can be stored in a readable storage medium. When the program is executed, it performs the steps including the above method embodiments; and the foregoing storage medium includes: removable storage devices, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical discs and other various media that can store program codes.

[0154] Alternatively, if the above integrated unit of the present invention is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a readable storage medium. Based on such an understanding, the technical solution of the embodiments of the present invention essentially or the part that contributes to the prior art can be embodied in the form of a software product. The software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in the embodiments of the present invention. And the foregoing storage medium includes: removable storage devices, ROM, RAM, magnetic disks or optical discs and other various media that can store program codes.

Claims

1. A method for managing a voice library of children's literature materials based on big data, characterized in that: include: Obtain literary materials, input the literary materials into the speech generation model, and output speech information; The voice information is parsed and processed to obtain parsed information, and the matching degree between the voice information and the literary data is analyzed according to the parsed information; Determining whether the matching degree is greater than a set matching degree threshold; If it is greater than, the voice information is classified to obtain the voice category, and the voice information is stored according to the voice category; If it is less than, the model parameters of the speech generation model are adjusted.

2. The method for managing a children's literature voice database based on big data according to claim 1, characterized in that: Obtain literary materials, input the literary materials into the speech generation model, and output speech information, including: Establish a training set through big data, iteratively train the speech generation model through the training set, and obtain the training results; Determine whether the training results converge; If convergence occurs, the literary data is input into the trained speech generation model to obtain speech information; If it does not converge, extended information is generated, and the data in the training set is expanded according to the extended information, and the speech generation model is trained again.

3. The method for managing a children's literature voice database based on big data according to claim 2, characterized in that: The voice information is analyzed and processed to obtain analysis information, which specifically includes: Acquire voice information, and slice the voice information to obtain a plurality of slices; According to the sentences, grammar and voice in the literary materials, the text is semantically analyzed, and the semantic information in the slices is extracted and converted into text; Connect the text and determine whether the text description is fluent; If it is fluent, generate parsing information; If it is not smooth, adjust the text connection relationship.

4. The method for managing a children's literature voice database based on big data according to claim 3 is characterized in that: According to the parsing information, the matching degree between the voice information and the literary data is analyzed, including: Reversely analyze the speech information according to the analysis information, and perform semantic analysis on the analysis information to obtain semantic information; Obtain literary materials, analyze them based on the literary materials, and obtain textual information; Semantically match text information with semantic information to determine the degree of semantic difference; The matching degree between the speech information and the literary material is generated based on the speech difference.

5. The method for managing a children's literature voice database based on big data according to claim 4 is characterized in that: If it is greater than, the voice information is classified to obtain the voice category, and the voice information is stored according to the voice category, specifically including: Acquire voice information, perform semantic analysis on the voice information, and obtain semantic information; Match the corresponding literary materials according to the semantic information, classify the literary materials according to their contents, and obtain category information; The voice information of the same category of literary materials is stored in the same category interval; A title header is generated according to the category of the literary material, the title header is matched to a corresponding category interval, and a storage area with the title header is generated.

6. The method for managing a children's literature voice database based on big data according to claim 5 is characterized in that: If it is less than, the model parameters of the speech generation model are adjusted, including: The set matching degree threshold is divided into a first matching degree threshold and a second matching degree threshold, wherein the first matching degree threshold is less than the second matching degree threshold; If the matching degree is greater than a first matching degree threshold and less than a second matching degree threshold, a first adjustment coefficient is generated, and a model parameter of the speech generation model is adjusted according to the first adjustment coefficient; If the matching degree is greater than a second matching degree threshold, a second adjustment coefficient is generated, and the model parameters of the speech generation model are adjusted according to the second adjustment coefficient.

7. A children's literature material voice library management system based on big data, characterized in that: The system includes: a memory and a processor, wherein the memory includes a program of a method for managing a voice library of children's literature materials based on big data, and when the program of the method for managing a voice library of children's literature materials based on big data is executed by the processor, the following steps are implemented: Obtain literary materials, input the literary materials into the speech generation model, and output speech information; The voice information is parsed and processed to obtain parsed information, and the matching degree between the voice information and the literary data is analyzed according to the parsed information; Determining whether the matching degree is greater than a set matching degree threshold; If it is greater than, the voice information is classified to obtain the voice category, and the voice information is stored according to the voice category; If it is less than, the model parameters of the speech generation model are adjusted.

8. The children's literature material voice library management system based on big data according to claim 7 is characterized in that: Obtain literary materials, input the literary materials into the speech generation model, and output speech information, including: Establish a training set through big data, iteratively train the speech generation model through the training set, and obtain the training results; Determine whether the training results converge; If convergence occurs, the literary data is input into the trained speech generation model to obtain speech information; If it does not converge, extended information is generated, and the data in the training set is expanded according to the extended information, and the speech generation model is trained again.

9. The children's literature material voice library management system based on big data according to claim 8 is characterized in that: The voice information is analyzed and processed to obtain analysis information, which specifically includes: Acquire voice information, and slice the voice information to obtain a plurality of slices; According to the sentences, grammar and voice in the literary materials, the text is semantically analyzed, and the semantic information in the slices is extracted and converted into text; Connect the text and determine whether the text description is fluent; If it is fluent, generate parsing information; If it is not smooth, adjust the text connection relationship.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a children's literature material voice library management method program based on big data. When the children's literature material voice library management method program based on big data is executed by a processor, the steps of the children's literature material voice library management method based on big data as described in any one of claims 1 to 6 are implemented.

Images