CN115994536B - Text information processing method, system, equipment and computer storage medium - Google Patents
Text information processing method, system, equipment and computer storage medium Download PDFInfo
- Publication number
- CN115994536B CN115994536B CN202310294417.2A CN202310294417A CN115994536B CN 115994536 B CN115994536 B CN 115994536B CN 202310294417 A CN202310294417 A CN 202310294417A CN 115994536 B CN115994536 B CN 115994536B
- Authority
- CN
- China
- Prior art keywords
- text information
- target
- content
- word
- content type
- 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.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The application discloses a text information processing method, a text information processing system, text information processing equipment and a computer storage medium, and relates to the technical field of AIGC (automatic guided way) and aims at acquiring target text information to be processed; word segmentation processing is carried out on the target text information to obtain target words; parsing the content type described by the target text information based on the target word; if the content type is static content, generating a target picture corresponding to the target text information based on the text generation image model; and if the content type is dynamic content, generating a target video corresponding to the target text information based on the text generation video model. According to the method and the device, when the content type is static content, the target picture corresponding to the target text information is generated based on the AIGC, and when the content type is dynamic content, the target video corresponding to the target text information is generated based on the AIGC, so that the automatic application of the AIGC to convert the target text information into the corresponding target picture or target video is realized, and the autonomous processing capacity of the AIGC is improved.
Description
Technical Field
The present application relates to the technical field of AI-Generated Content (AIGC, artificial intelligence generated content), and more particularly, to a text information processing method, system, apparatus, and computer storage medium.
Background
With the development of artificial intelligence theory and technology, the AIGC can generate drawings in a short time, such as matching vivid pictures and videos for digital works (news, poetry, novel, papers, patents, etc.) of text types, so that the work description can be more vivid.
However, in the conventional AIGC, the user decides whether to convert a text into a picture or a video, that is, when the user decides to convert the text into the picture, the text generation image model in the AIGC is applied to generate the picture corresponding to the text, and when the user decides to convert the text into the video, the text generation video model in the AIGC is applied to generate the video corresponding to the text, so that the application of the conventional AIGC is independent of the operation of the user, and needs to rely on the user, and has poor autonomous processing capability.
In view of the above, how to improve the autonomous processing capability of the AIGC is a problem to be solved by those skilled in the art.
Disclosure of Invention
The purpose of the application is to provide a text information processing method which can solve the technical problem of how to improve the autonomous processing capability of AIGC technology to a certain extent. The application also provides a text information processing system, a text information processing device and a computer readable storage medium.
In order to achieve the above object, the present application provides the following technical solutions:
a text information processing method, comprising:
acquiring target text information to be processed;
word segmentation processing is carried out on the target text information to obtain target words;
resolving the content type described by the target text information based on the target word;
if the content type is static content, generating a target picture corresponding to the target text information based on a text generation image model;
and if the content type is dynamic content, generating a target video corresponding to the target text information based on a text generation video model.
In some embodiments, the parsing the content type described by the target text information based on the target word includes:
predicting probability values of all known words of the target word belonging to a target dictionary base;
determining the content type described by the target text information based on a target part of speech of the known word and the probability value;
wherein the target part of speech includes the known word corresponding to the static content or the known word corresponding to the dynamic content.
In some embodiments, the determining the content type described by the target text information based on the target part of speech of the known word and the probability value comprises:
Adding the probability values of the known words corresponding to the static content to obtain a first probability value;
adding the probability values of the known words corresponding to the dynamic content to obtain a second probability value;
if the first probability value is larger than the second probability value, determining that the content type described by the target text information is the static content;
and if the first probability value is smaller than the second probability value, determining that the content type described by the target text information is the dynamic content.
In some embodiments, the determining the content type described by the target text information based on the target part of speech of the known word and the probability value comprises:
for each target word, determining the known word corresponding to the probability value with the largest value as the selected word of the target word;
determining a first quantity value of the known words belonging to the static content in the selected words;
determining a second quantity value of the known words belonging to the dynamic content in the selected words;
calculating a difference between the first quantity value and the second quantity value;
If the difference value is larger than a first preset value, determining that the content type described by the target text information is the static content;
and if the difference value is smaller than the first preset value, determining that the content type described by the target text information is the dynamic content.
In some embodiments, the determining the content type described by the target text information based on the target part of speech of the known word and the probability value comprises:
for each target word, determining the known word corresponding to the probability value with the largest value as the selected word of the target word;
determining a second quantity value of the known words belonging to the dynamic content in the selected words;
if the second quantity value is smaller than a second preset value, determining that the content type described by the target text information is the static content;
and if the second quantity value is larger than the second preset value, determining that the content type described by the target text information is the dynamic content.
In some embodiments, the predicting the probability value that the target word belongs to each known word in a target dictionary base, determining the content type described by the target text information based on the target part of speech of the known word and the probability value, comprises:
Inputting the target words into a pre-trained language neural network model;
acquiring the content type output by the language neural network model;
wherein the language neural network model is used for: predicting probability values of the target words belonging to the known words in a target dictionary base, and determining the content type described by the target text information based on the target parts of speech of the known words and the probability values.
In some embodiments, the type of language neural network model includes a prompt learning model.
In some embodiments, the obtaining the target text information to be processed includes:
acquiring initial text information to be processed;
translating the initial text information to obtain translated text information;
and obtaining the target text information based on the translation text information.
In some embodiments, the obtaining the target text information based on the translated text information includes:
and expanding the translation text information to obtain the target text information.
In some embodiments, the expanding the translated text information to obtain the target text information includes:
Extracting target keywords in the translation text information;
selecting target description words similar to the target keywords from a preset content extension description knowledge base;
and generating the target text information based on the target descriptive language and the translation text information.
In some embodiments, the generating the target text information based on the target descriptive language and the translated text information includes:
selecting a target qualifier corresponding to the target keyword from preset image quality experience qualifiers;
and generating the target text information based on the target description language, the target qualifier and the translation text information.
In some embodiments, the obtaining the target text information to be processed includes:
acquiring initial text information to be processed;
splitting the initial text information to obtain split text information;
and obtaining the target text information based on the split text information.
In some embodiments, the splitting the initial text information to obtain split text information includes:
and carrying out equal-length splitting on the initial text information based on a natural language processing algorithm to obtain the split text information.
In some embodiments, the obtaining the target text information based on the split text information includes:
for each split text message, generating a total semantic vector of the split text message, generating paragraph semantic vectors of all text paragraphs in the split text message, calculating similarity values of all paragraph semantic vectors and the total semantic vector, and determining the text paragraphs corresponding to the similarity values with the largest target number as target text paragraphs;
and generating the target text information based on the target text paragraph.
In some embodiments, the computing similarity values for each of the paragraph semantic vectors and the total semantic vector comprises:
and calculating the similarity value of each paragraph semantic vector and the total semantic vector based on a cosine similarity method.
In some embodiments, the generating the target text information based on the target text passage includes:
extracting a target sentence in the target text paragraph;
and taking the target sentence as the target text information.
In some embodiments, the generating the total semantic vector of the split text information and generating the paragraph semantic vector of each text paragraph in the split text information include:
Generating the total semantic vector of the split text information through a Sentence-BERT model, and generating the paragraph semantic vector of each text paragraph in the split text information.
A text information processing system comprising:
the acquisition module is used for acquiring target text information to be processed;
the processing module is used for carrying out word segmentation processing on the target text information to obtain target words;
the analysis module is used for analyzing the content type described by the target text information based on the target word;
the generation module is used for generating a target picture corresponding to the target text information based on a text generation image model if the content type is static content; and if the content type is dynamic content, generating a target video corresponding to the target text information based on a text generation video model.
A text information processing apparatus comprising:
a memory for storing a computer program;
a processor for implementing the steps of any one of the text information processing methods described above when executing the computer program.
A computer-readable storage medium having stored therein a computer program that is executed by a processor to implement the steps of any of the text information processing methods described above.
According to the text information processing method, target text information to be processed is obtained; word segmentation processing is carried out on the target text information to obtain target words; parsing the content type described by the target text information based on the target word; if the content type is static content, generating a target picture corresponding to the target text information based on the text generation image model; and if the content type is dynamic content, generating a target video corresponding to the target text information based on the text generation video model. According to the method and the device, the content type described by the target text information can be analyzed according to the target words obtained by word segmentation of the target text information, the target picture corresponding to the target text information can be generated based on the AIGC when the content type is static content, and the target video corresponding to the target text information can be generated based on the AIGC when the content type is dynamic content, so that the automatic application of the AIGC to convert the target text information into the corresponding target picture or target video is realized, and the autonomous processing capacity of the AIGC on the text information is improved. The text information processing system, the text information processing device and the computer readable storage medium also solve the corresponding technical problems.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings that are required to be used in the embodiments or the description of the prior art will be briefly described below, and it is obvious that the drawings in the following description are only embodiments of the present application, and that other drawings may be obtained according to the provided drawings without inventive effort to a person skilled in the art.
Fig. 1 is a flowchart of a text information processing method provided in an embodiment of the present application;
fig. 2 is another flowchart of a text information processing method provided in an embodiment of the present application;
fig. 3 is another flowchart of a text information processing method according to an embodiment of the present application;
FIG. 4 is a schematic diagram of a process for translating and expanding initial text information in the present application;
FIG. 5 is another flowchart of a text message processing method according to an embodiment of the present application;
fig. 6 is a schematic structural diagram of a text information processing system according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a text information processing apparatus according to an embodiment of the present application;
fig. 8 is another schematic structural diagram of a text information processing apparatus according to an embodiment of the present application.
Detailed Description
The following description of the embodiments of the present application will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are only some, but not all, of the embodiments of the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden from the present disclosure, are within the scope of the present disclosure.
Referring to fig. 1, fig. 1 is a flowchart of a text information processing method according to an embodiment of the present application.
The text information processing method provided by the embodiment of the application can comprise the following steps:
step S101: and acquiring target text information to be processed.
In practical application, the target text information to be processed may be obtained first, and the content, the source, etc. of the target text information may be determined according to practical needs, for example, the target text information may be scene description information from a user client, may be real-time hot news from newspaper, may be abstract from a book, etc., which is not limited herein specifically.
Step S102: and performing word segmentation processing on the target text information to obtain target words.
In practical application, after the target text information to be processed is obtained, word segmentation processing can be performed on the target text information to obtain each target word in the target text information, in a specific application scenario, word segmentation processing can be performed on the target text information to obtain a target word, and the like, which is not particularly limited herein.
Step S103: the content type described by the target text information is parsed based on the target terms.
In practical application, after word segmentation is performed on the target text information to obtain the target word, the content type described by the target text information can be analyzed based on the target word, that is, whether the content described by the target text information is static or dynamic is analyzed based on the target word, so that the target text information can be correspondingly processed based on the content type.
Step S104: and if the content type is static content, generating a target picture corresponding to the target text information based on the text generation image model.
In practical application, after the content type described by the target text information is analyzed based on the target word, and in the case that the content type is static content, the target picture corresponding to the target text information may be generated based on the text generation image model in the AIGC, specifically, the target picture corresponding to the target text information may be generated by applying the stable diffusion model or the dall·e2 model or the Imagen model in the AIGC, which is not specifically limited herein.
Step S105: and if the content type is dynamic content, generating a target video corresponding to the target text information based on the text generation video model.
In practical application, after the content type described by the target text information is analyzed based on the target word, and in the case that the content type is dynamic content, the target video corresponding to the target text information may be generated based on the text generation video model in the AIGC, specifically, the target video corresponding to the target text information may be generated by applying the cgvideo model in the AIGC, etc., which is not particularly limited herein.
In the specific application scenario, after the target picture or the target video corresponding to the target text information is generated based on the AIGC, style migration, image enhancement, customization and other processes may be performed on the target picture or the target image, so that the target picture or the target video may further meet the use requirement of the user.
According to the text information processing method, target text information to be processed is obtained; word segmentation processing is carried out on the target text information to obtain target words; parsing the content type described by the target text information based on the target word; if the content type is static content, generating a target picture corresponding to the target text information based on the text generation image model; and if the content type is dynamic content, generating a target video corresponding to the target text information based on the text generation video model. According to the method and the device, the content type described by the target text information can be analyzed according to the target words obtained by word segmentation of the target text information, the target picture corresponding to the target text information can be generated based on the AIGC when the content type is static content, and the target video corresponding to the target text information can be generated based on the AIGC when the content type is dynamic content, so that the automatic application of the AIGC to convert the target text information into the corresponding target picture or target video is realized, and the autonomous processing capacity of the AIGC on the text information is improved.
Referring to fig. 2, fig. 2 is another flowchart of a text information processing method according to an embodiment of the present application.
The text information processing method provided by the embodiment of the application can comprise the following steps:
step S201: and acquiring target text information to be processed.
Step S202: and performing word segmentation processing on the target text information to obtain target words.
Step S203: predicting probability values of all known words belonging to target words in a target dictionary base;
step S204: determining a content type described by the target text information based on the target part of speech and the probability value of the known word; the target part of speech comprises known words corresponding to static content or known words corresponding to dynamic content.
In practical application, in the process of analyzing the content type described by the target text information based on the target word, the content type described by the target text information can be analyzed by means of a preset target dictionary base for determining the content type described by the text information, and specifically, the probability value of each known word belonging to the target word in the target dictionary base can be predicted; determining a content type described by the target text information based on the target part of speech and the probability value of the known word; the target part of speech comprises known words corresponding to static content or known words corresponding to dynamic content.
In a specific application scene, in the process of determining the content type described by the target text information based on the target part of speech and the probability value of the known words, the content type described by the target text information can be estimated from the whole based on the calculated probability value, namely, the probability values of the known words corresponding to the static content can be added to obtain a first probability value; adding probability values of known words corresponding to the dynamic content to obtain a second probability value; if the first probability value is larger than the second probability value, determining that the content type described by the target text information is static content; and if the first probability value is smaller than the second probability value, determining that the content type described by the target text information is dynamic content.
In a specific application scenario, in the process of determining the content type described by the target text information based on the target part of speech and the probability value of the known word, the content type described by the text information can be evaluated only according to the corresponding word of the target word in the target dictionary base, that is, for each target word, the known word corresponding to the probability value with the largest value can be determined as the selected word of the target word; determining a first quantity value of known words corresponding to static content in the selected words; determining a second quantity value of known words corresponding to dynamic content in the selected words; calculating a difference between the first quantity value and the second quantity value; if the difference value is larger than a first preset value, determining that the content type described by the target text information is static content; if the difference value is smaller than the first preset value, determining that the content type described by the target text information is dynamic content. It should be noted that, the specific value of the first preset value may be determined according to a specific application scenario, which is not specifically limited herein.
In a specific application scenario, in the process of determining the content type described by the target text information based on the target part of speech and the probability value of the known words, when the number of words describing the dynamic content contained in the target text information reaches a set value, the dynamic content described by the target text information is considered, that is, for each target word, the known word corresponding to the probability value with the largest value can be determined as the selected word of the target word; determining a second quantity value of known words corresponding to dynamic content in the selected words; if the second quantity value is smaller than a second preset value, determining that the content type described by the target text information is static content; and if the second quantity value is larger than a second preset value, determining that the content type described by the target text information is dynamic content. It should be noted that, the specific value of the second preset value may be determined according to a specific application scenario, which is not specifically limited herein.
In a specific application scene, in the process of predicting probability values of all known words in a target dictionary library, determining content types described by target text information based on target parts of speech and the probability values of the known words, the content types can be rapidly predicted by means of a neural network model, namely, the target words are input into a pre-trained language neural network model; obtaining the content type output by the language neural network model; wherein, language neural network model is used for: and predicting probability values of the target words belonging to all known words in the target dictionary base, and determining the content type described by the target text information based on the target part of speech and the probability values of the known words. It should be noted that, the type of the language neural network model may be determined according to a specific application scenario, for example, the language neural network model may include a prompt learning model, and the prompt learning model may include an input layer, a multi-head self-attention layer connected to the input layer, a first addition normalization layer connected to the input layer and the multi-head self-attention layer, a forward network connected to the first addition normalization layer, a second addition normalization layer connected to the forward network and the first addition normalization layer, an output connected to the second addition normalization layer, and the application does not limit a specific architecture of the prompt learning model herein.
It should be noted that, the content of the target dictionary library may be determined according to actual needs, for example, words corresponding to static content in the target dictionary library may include apples, fruits, mountains, landscapes, animals, hotels, rabbits, kittens, rockets, puppies, houses, buildings, water cups, stars, foods, moon, sun, mountains, etc., and words corresponding to dynamic content in the target dictionary library may include braiding, writing, running, investment, long jump, events, sports, attacks, flying, falls, skiing, throwing, rapture, exercise, occurrence, contention, sanctions, frying, reading, browsing, scoring, clicking, pentium, jumping, modification, rising, falling, sitting, etc.
Step S205: and if the content type is static content, generating a target picture corresponding to the target text information based on the text generation image model.
Step S206: and if the content type is dynamic content, generating a target video corresponding to the target text information based on the text generation video model.
Referring to fig. 3, fig. 3 is another flowchart of a text information processing method according to an embodiment of the present application.
The text information processing method provided by the embodiment of the application can comprise the following steps:
Step S301: and acquiring the initial text information to be processed.
Step S302: and translating the initial text information to obtain translated text information.
Step S303: and obtaining target text information based on the translated text information.
In practical application, due to the diversity of languages, the language versions of the text information are more, so that the AIGC can process the text information in different languages, and in the process of acquiring the target text information, the initial text information to be processed can be acquired; translating the initial text information to obtain translated text information, such as Chinese translated text information; and obtaining target text information based on the translated text information. The method of translating the initial text information may be determined according to actual needs, for example, the translation text information may be obtained by translating the initial text information by means of a neural machine translation model based on a transducer.
In a specific application scene, considering that the content of the initial text information to be processed is single, for example, the content of the initial text information is a 'armor-penetrating horse', if a picture or a video is generated only according to the content, the content is single, and the user requirement cannot be met, in order to solve the problem, in the process of obtaining the target text information based on the translation text information, the translation text information can be expanded to obtain the target text information, and the process can be as shown in fig. 4, and in particular, the target keywords in the translation text information can be extracted; selecting target description words similar to the target keywords from a preset content extension description knowledge base; generating target text information based on the target description language and the translation text information, and selecting a target definition word corresponding to the target keyword from preset image quality experience definition words in the process of generating the target text information based on the target description language and the translation text information, wherein the target definition word can comprise 4k,8k, cartoon wind, reality sense, impression sense, van Gao creation and the like; and generating target text information based on the target description language, the target qualifier and the translation text information. The initial text information with the content of 'armor-penetrating horses' can be expanded into target text information with the content of 'personified male peoples Ma Qishi, portrait, exquisite armor, movie lights, complex wire designs, 4k,8k, illusion engines, octane number rendering', and the like; for another example, after the content of the initial text information is "a flashing purple lotus flower", the content of the target text information after the processing of the scheme of the application may be "transparent high definition CG, light purple fire purple big lotus flower, lightning, lamplight effect, 8K, ultra-high definition, artistic trend of snow storm style, noctilucent fog and noctilucent, volume light of volume light-ar 9:166" and the like, which are not particularly limited herein.
Step S304: and performing word segmentation processing on the target text information to obtain target words.
Step S305: the content type described by the target text information is parsed based on the target terms.
Step S306: and if the content type is static content, generating a target picture corresponding to the target text information based on the text generation image model.
Step S307: and if the content type is dynamic content, generating a target video corresponding to the target text information based on the text generation video model.
Referring to fig. 5, fig. 5 is another flowchart of a text information processing method according to an embodiment of the present application.
The text information processing method provided by the embodiment of the application can comprise the following steps:
step S401: and acquiring the initial text information to be processed.
Step S402: splitting the initial text information to obtain split text information.
Step S403: and obtaining target text information based on the split text information.
In practical application, in the process of acquiring the target text information to be processed, considering that the content of the acquired initial text information is more, the AIGC can only generate pictures of one picture or videos of a plurality of pictures, namely the AIGC has weaker processing capability on the large text information, and in order to solve the problem, the initial text information can be automatically split after the initial text information to be processed is acquired, so as to obtain split text information; and obtaining target text information based on the split text information.
In a specific application scenario, in the process of splitting the initial text information to obtain split text information, equal-length splitting can be performed on the initial text information based on a natural language processing algorithm to obtain split text information, and of course, other splitting modes, such as splitting based on content continuity, can also be used, and the application is not limited in detail herein.
In a specific application scene, in the process of obtaining target text information based on split text information, for each split text information, a total semantic vector of the split text information can be generated, paragraph semantic vectors of all text paragraphs in the split text information are generated, similarity values of all paragraph semantic vectors and the total semantic vector are calculated, and text paragraphs corresponding to the similarity value with the largest target number of values are determined as target text paragraphs, so that text paragraphs describing core contents of the split text information are screened out; target text information is generated based on all of the target text paragraphs.
In a specific application scenario, in the process of calculating the similarity value between each paragraph semantic vector and the total semantic vector, the similarity value between each paragraph semantic vector and the total semantic vector can be quickly calculated based on a cosine similarity method, and the like, which is not specifically limited herein.
In a specific application scene, in the process of generating target text information based on a target text paragraph, core sentences in the target text paragraph can be extracted as target sentences, for example, the target text paragraph is extracted according to a preset rule to obtain the target sentences; the target sentence is used as the target text information, so that the core content of the initial text information can be simply, conveniently and accurately described by the target sentence.
In a specific application scene, in the process of generating the total semantic vector of the split text information and the paragraph semantic vector of each text paragraph in the split text information, the total semantic vector of the split text information can be quickly generated through a Sentence-BERT model, and the paragraph semantic vector of each text paragraph in the split text information and the like are generated.
Step S404: and performing word segmentation processing on the target text information to obtain target words.
Step S405: the content type described by the target text information is parsed based on the target terms.
Step S406: and if the content type is static content, generating a target picture corresponding to the target text information based on the text generation image model.
Step S407: and if the content type is dynamic content, generating a target video corresponding to the target text information based on the text generation video model.
It should be noted that, in a specific application scenario, after the initial text information to be processed is obtained, it may be necessary to translate, expand and split the initial text information to obtain the target text information, that is, the text information processing method that is described in the embodiments of the present application may be combined according to specific application requirements, so as to obtain the text information processing method that meets the user requirements, and the process of the present application is not specifically limited herein.
Referring to fig. 6, fig. 6 is a schematic structural diagram of a text information processing system according to an embodiment of the present application.
The text information processing system provided in the embodiment of the application may include:
an obtaining module 101, configured to obtain target text information to be processed;
the processing module 102 is used for performing word segmentation processing on the target text information to obtain target words;
a parsing module 103, configured to parse the content type described by the target text information based on the target word;
the generating module 104 is configured to generate a target picture corresponding to the target text information based on the text generation image model if the content type is static content; and if the content type is dynamic content, generating a target video corresponding to the target text information based on the text generation video model.
The text information processing system provided in the embodiment of the present application, the parsing module may include:
the prediction unit is used for predicting probability values of all known words in the target dictionary base of the target words;
a determining unit for determining the content type described by the target text information based on the target part of speech and the probability value of the known word;
the target part of speech comprises known words corresponding to static content or known words corresponding to dynamic content.
The text information processing system provided in the embodiment of the present application, the determining unit may be specifically configured to: adding probability values of known words corresponding to the static content to obtain a first probability value; adding probability values of known words corresponding to the dynamic content to obtain a second probability value; if the first probability value is larger than the second probability value, determining that the content type described by the target text information is static content; and if the first probability value is smaller than the second probability value, determining that the content type described by the target text information is dynamic content.
The text information processing system provided in the embodiment of the present application, the determining unit may be specifically configured to: for each target word, determining the known word corresponding to the probability value with the largest value as the selected word of the target word; determining a first quantity value of known words corresponding to static content in the selected words; determining a second quantity value of known words corresponding to dynamic content in the selected words; calculating a difference between the first quantity value and the second quantity value; if the difference value is larger than a first preset value, determining that the content type described by the target text information is static content; if the difference value is smaller than the first preset value, determining that the content type described by the target text information is dynamic content.
The text information processing system provided in the embodiment of the present application, the determining unit may be specifically configured to: for each target word, determining the known word corresponding to the probability value with the largest value as the selected word of the target word; determining a second quantity value of known words corresponding to dynamic content in the selected words; if the second quantity value is smaller than a second preset value, determining that the content type described by the target text information is static content; and if the second quantity value is larger than a second preset value, determining that the content type described by the target text information is dynamic content.
The text information processing system provided in the embodiment of the present application, the parsing module may be specifically configured to: inputting target words into a pre-trained language neural network model; obtaining the content type output by the language neural network model; wherein, language neural network model is used for: and predicting probability values of the target words belonging to all known words in the target dictionary base, and determining the content type described by the target text information based on the target part of speech and the probability values of the known words.
The text information processing system provided by the embodiment of the application comprises a prompt learning model.
The text information processing system provided in the embodiment of the present application, the obtaining module may include:
the first acquisition unit is used for acquiring initial text information to be processed;
the translation unit is used for translating the initial text information to obtain translated text information;
and the first processing unit is used for obtaining target text information based on the translated text information.
The text information processing system provided in the embodiment of the present application, the first processing unit may be specifically configured to: and expanding the translation text information to obtain target text information.
The text information processing system provided in the embodiment of the present application, the first processing unit may be specifically configured to: extracting target keywords in the translation text information; selecting target description words similar to the target keywords from a preset content extension description knowledge base; target text information is generated based on the target descriptive and the translated text information.
The text information processing system provided in the embodiment of the present application, the first processing unit may be specifically configured to: selecting a target qualifier corresponding to the target keyword from preset image quality experience qualifiers; and generating target text information based on the target description language, the target qualifier and the translation text information.
The text information processing system provided in the embodiment of the present application, the obtaining module may include:
the second acquisition unit is used for acquiring initial text information to be processed;
the splitting unit is used for splitting the initial text information to obtain split text information;
and the second processing unit is used for obtaining target text information based on the split text information.
The text information processing system provided in the embodiment of the present application, the splitting unit may be specifically configured to: and performing equal-length splitting on the initial text information based on a natural language processing algorithm to obtain split text information.
The text information processing system provided in the embodiment of the present application, the second processing unit may be specifically configured to: for each split text message, generating a total semantic vector of the split text message, generating paragraph semantic vectors of all text paragraphs in the split text message, calculating similarity values of all paragraph semantic vectors and the total semantic vector, and determining the text paragraphs corresponding to the similarity value with the largest target number as target text paragraphs; target text information is generated based on the target text passage.
The text information processing system provided in the embodiment of the present application, the second processing unit may be specifically configured to: and calculating the similarity value of each paragraph semantic vector and the total semantic vector based on a cosine similarity method.
The text information processing system provided in the embodiment of the present application, the second processing unit may be specifically configured to: extracting a target sentence in a target text paragraph; and taking the target sentence as target text information.
The text information processing system provided in the embodiment of the present application, the second processing unit may be specifically configured to: generating a total semantic vector of the split text information through a Sentence-BERT model, and generating paragraph semantic vectors of all text paragraphs in the split text information.
The application also provides text information processing equipment and a computer readable storage medium, which have the corresponding effects of the text information processing method. Referring to fig. 7, fig. 7 is a schematic structural diagram of a text information processing apparatus according to an embodiment of the present application.
The text information processing device provided in the embodiment of the present application includes a memory 201 and a processor 202, where the memory 201 stores a computer program, and the processor 202 implements the steps of the text information processing method described in any of the embodiments above when executing the computer program.
Referring to fig. 8, another text information processing apparatus provided in an embodiment of the present application may further include: an input port 203 connected to the processor 202 for transmitting an externally input command to the processor 202; a display unit 204 connected to the processor 202, for displaying the processing result of the processor 202 to the outside; and a communication module 205 connected to the processor 202, for implementing communication between the text information processing device and the outside. The display unit 204 may be a display panel, a laser scanning display, or the like; communication means employed by the communication module 205 include, but are not limited to, mobile high definition link technology (HML), universal Serial Bus (USB), high Definition Multimedia Interface (HDMI), wireless connection: wireless fidelity (WiFi), bluetooth communication, bluetooth low energy communication, ieee802.11s based communication.
The embodiment of the application provides a computer readable storage medium, in which a computer program is stored, where the computer program when executed by a processor implements the steps of the text information processing method described in any of the embodiments above.
The computer readable storage medium referred to in this application includes Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The description of the relevant parts in the text information processing system, the text information processing device and the computer readable storage medium provided in the embodiments of the present application refers to the detailed description of the corresponding parts in the text information processing method provided in the embodiments of the present application, and will not be repeated here. In addition, the parts of the above technical solutions provided in the embodiments of the present application, which are consistent with the implementation principles of the corresponding technical solutions in the prior art, are not described in detail, so that redundant descriptions are avoided.
It is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (19)
1. A text information processing method, characterized by comprising:
acquiring target text information to be processed;
word segmentation processing is carried out on the target text information to obtain target words;
resolving the content type described by the target text information based on the target word;
if the content type is static content, generating a target picture corresponding to the target text information based on a text generation image model;
if the content type is dynamic content, generating a target video corresponding to the target text information based on a text generation video model;
wherein the parsing the content type described by the target text information based on the target word includes:
Predicting probability values of all known words of the target word belonging to a target dictionary base;
determining the content type described by the target text information based on a target part of speech of the known word and the probability value;
wherein the target part of speech includes the known word corresponding to the static content or the known word corresponding to the dynamic content.
2. The text information processing method according to claim 1, wherein the determining the content type described by the target text information based on the target part of speech of the known word and the probability value includes:
adding the probability values of the known words corresponding to the static content to obtain a first probability value;
adding the probability values of the known words corresponding to the dynamic content to obtain a second probability value;
if the first probability value is larger than the second probability value, determining that the content type described by the target text information is the static content;
and if the first probability value is smaller than the second probability value, determining that the content type described by the target text information is the dynamic content.
3. The text information processing method according to claim 1, wherein the determining the content type described by the target text information based on the target part of speech of the known word and the probability value includes:
for each target word, determining the known word corresponding to the probability value with the largest value as the selected word of the target word;
determining a first quantity value of the known words belonging to the static content in the selected words;
determining a second quantity value of the known words belonging to the dynamic content in the selected words;
calculating a difference between the first quantity value and the second quantity value;
if the difference value is larger than a first preset value, determining that the content type described by the target text information is the static content;
and if the difference value is smaller than the first preset value, determining that the content type described by the target text information is the dynamic content.
4. The text information processing method according to claim 1, wherein the determining the content type described by the target text information based on the target part of speech of the known word and the probability value includes:
For each target word, determining the known word corresponding to the probability value with the largest value as the selected word of the target word;
determining a second quantity value of the known words belonging to the dynamic content in the selected words;
if the second quantity value is smaller than a second preset value, determining that the content type described by the target text information is the static content;
and if the second quantity value is larger than the second preset value, determining that the content type described by the target text information is the dynamic content.
5. The text information processing method according to claim 1, wherein predicting a probability value that the target word belongs to each known word in a target dictionary base, determining the content type described by the target text information based on a target part of speech of the known word and the probability value, comprises:
inputting the target words into a pre-trained language neural network model;
acquiring the content type output by the language neural network model;
wherein the language neural network model is used for: predicting probability values of the target words belonging to the known words in a target dictionary base, and determining the content type described by the target text information based on the target parts of speech of the known words and the probability values.
6. The text information processing method of claim 5, wherein the type of language neural network model includes a prompt learning model.
7. The text information processing method according to claim 1, wherein the acquiring the target text information to be processed includes:
acquiring initial text information to be processed;
translating the initial text information to obtain translated text information;
and obtaining the target text information based on the translation text information.
8. The text information processing method of claim 7, wherein the obtaining the target text information based on the translated text information includes:
and expanding the translation text information to obtain the target text information.
9. The text information processing method of claim 8, wherein expanding the translated text information to obtain the target text information includes:
extracting target keywords in the translation text information;
selecting target descriptors meeting similar conditions with the target keywords from a preset content extension description knowledge base;
and generating the target text information based on the target descriptive language and the translation text information.
10. The text information processing method according to claim 9, wherein the generating the target text information based on the target descriptive language and the translated text information includes:
selecting a target qualifier corresponding to the target keyword from preset image quality experience qualifiers;
and generating the target text information based on the target description language, the target qualifier and the translation text information.
11. The text information processing method according to any one of claims 1 to 10, wherein the acquiring the target text information to be processed includes:
acquiring initial text information to be processed;
splitting the initial text information to obtain split text information;
and obtaining the target text information based on the split text information.
12. The text information processing method of claim 11, wherein splitting the initial text information to obtain split text information includes:
and carrying out equal-length splitting on the initial text information based on a natural language processing algorithm to obtain the split text information.
13. The text information processing method of claim 11, wherein the obtaining the target text information based on the split text information includes:
For each split text message, generating a total semantic vector of the split text message, generating paragraph semantic vectors of all text paragraphs in the split text message, calculating similarity values of all paragraph semantic vectors and the total semantic vector, and determining the text paragraphs corresponding to the similarity values with the largest target number as target text paragraphs;
and generating the target text information based on the target text paragraph.
14. The text information processing method of claim 13, wherein said calculating a similarity value of each of said paragraph semantic vectors to said total semantic vector comprises:
and calculating the similarity value of each paragraph semantic vector and the total semantic vector based on a cosine similarity method.
15. The text information processing method of claim 13, wherein the generating the target text information based on the target text passage comprises:
extracting a target sentence in the target text paragraph;
and taking the target sentence as the target text information.
16. The method for processing text information according to claim 13, wherein generating the total semantic vector of the split text information and generating the paragraph semantic vector of each text paragraph in the split text information includes:
Generating the total semantic vector of the split text information through a Sentence-BERT model, and generating the paragraph semantic vector of each text paragraph in the split text information.
17. A text information processing system, comprising:
the acquisition module is used for acquiring target text information to be processed;
the processing module is used for carrying out word segmentation processing on the target text information to obtain target words;
the analysis module is used for analyzing the content type described by the target text information based on the target word;
the generation module is used for generating a target picture corresponding to the target text information based on a text generation image model if the content type is static content; if the content type is dynamic content, generating a target video corresponding to the target text information based on a text generation video model;
wherein, the parsing module includes:
the prediction unit is used for predicting probability values of all known words in the target dictionary base of the target words;
a determining unit configured to determine the content type described by the target text information based on a target part of speech of the known word and the probability value;
Wherein the target part of speech includes the known word corresponding to the static content or the known word corresponding to the dynamic content.
18. A text information processing apparatus characterized by comprising:
a memory for storing a computer program;
processor for implementing the steps of the text information processing method according to any one of claims 1 to 16 when executing said computer program.
19. A computer-readable storage medium, in which a computer program is stored, the computer program being executed by a processor to implement the steps of the text information processing method according to any one of claims 1 to 16.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310294417.2A CN115994536B (en) | 2023-03-24 | 2023-03-24 | Text information processing method, system, equipment and computer storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310294417.2A CN115994536B (en) | 2023-03-24 | 2023-03-24 | Text information processing method, system, equipment and computer storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115994536A CN115994536A (en) | 2023-04-21 |
| CN115994536B true CN115994536B (en) | 2023-07-14 |
Family
ID=85995431
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310294417.2A Active CN115994536B (en) | 2023-03-24 | 2023-03-24 | Text information processing method, system, equipment and computer storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115994536B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116233491B (en) * | 2023-05-04 | 2023-07-18 | 阿里巴巴达摩院(杭州)科技有限公司 | Video generation method and server |
| CN116304356B (en) * | 2023-05-11 | 2023-07-18 | 环球数科集团有限公司 | Scenic spot multi-scene content creation and application system based on AIGC |
| CN116306317B (en) * | 2023-05-12 | 2023-07-28 | 环球数科集团有限公司 | Automatic AIGC modeling system based on artificial intelligence |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114398482A (en) * | 2021-12-06 | 2022-04-26 | 腾讯数码（天津）有限公司 | Dictionary construction method and device, electronic equipment and storage medium |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108509465B (en) * | 2017-02-28 | 2022-03-15 | 阿里巴巴集团控股有限公司 | Video data recommendation method and device and server |
| CN107506377A (en) * | 2017-07-20 | 2017-12-22 | 南开大学 | This generation system is painted in interaction based on commending system |
| CN110263324B (en) * | 2019-05-16 | 2021-02-12 | 华为技术有限公司 | Text processing method, model training method and device |
| US11907664B1 (en) * | 2021-04-30 | 2024-02-20 | Capital One Services, Llc | Computer-based systems involving machine learning associated with generation of predictive content for data structure segments and methods of use thereof |
| CN113822067A (en) * | 2021-08-17 | 2021-12-21 | 深圳市东信时代信息技术有限公司 | Key information extraction method and device, computer equipment and storage medium |
| CN113705198B (en) * | 2021-10-21 | 2022-03-25 | 北京达佳互联信息技术有限公司 | Scene graph generation method and device, electronic equipment and storage medium |
| CN114610948A (en) * | 2022-03-11 | 2022-06-10 | 北京明略昭辉科技有限公司 | Video classification method and device, electronic equipment and storage medium |
| CN114638232A (en) * | 2022-03-22 | 2022-06-17 | 北京美通互动数字科技股份有限公司 | Method and device for converting text into video, electronic equipment and storage medium |
-
2023
- 2023-03-24 CN CN202310294417.2A patent/CN115994536B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114398482A (en) * | 2021-12-06 | 2022-04-26 | 腾讯数码（天津）有限公司 | Dictionary construction method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115994536A (en) | 2023-04-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115994536B (en) | Text information processing method, system, equipment and computer storage medium | |
| CN109145152B (en) | Method for adaptively and intelligently generating image-text video thumbnail based on query word | |
| CN104735468B (en) | A kind of method and system that image is synthesized to new video based on semantic analysis | |
| CN107818085B (en) | Answer selection method and system for reading understanding of reading robot | |
| JP6333745B2 (en) | Translation result providing method, translation result providing system, program, and file distribution system | |
| KR20190125863A (en) | Multilingual translation device and multilingual translation method | |
| CN108304376B (en) | Text vector determination method and device, storage medium and electronic device | |
| EP3885966B1 (en) | Method and device for generating natural language description information | |
| CN109978139B (en) | Method, system, electronic device and storage medium for automatically generating description of picture | |
| KR20200087977A (en) | Multimodal ducument summary system and method | |
| CN111428025A (en) | Text summarization method and device, electronic equipment and storage medium | |
| CN111444725A (en) | Statement generation method, device, storage medium and electronic device | |
| CN110990563A (en) | Artificial intelligence-based traditional culture material library construction method and system | |
| CN109446496A (en) | A kind of conversion method, device, equipment and the storage medium of test language file | |
| CN113449081A (en) | Text feature extraction method and device, computer equipment and storage medium | |
| CN116579304A (en) | Editing method and device for multi-mode document | |
| CN111814496B (en) | Text processing method, device, equipment and storage medium | |
| CN112749364B (en) | Webpage generation method, device, equipment and storage medium based on artificial intelligence | |
| CN114973086A (en) | Video processing method and device, electronic equipment and storage medium | |
| CN112035622A (en) | Integrated platform and method for natural language processing | |
| CN110309517B (en) | Expression document processing method, device, system and storage medium | |
| CN114021541A (en) | Presentation generation method, device, equipment and storage medium | |
| CN110750989A (en) | Statement analysis method and device | |
| CN114399782B (en) | Text image processing method, apparatus, device, storage medium, and program product | |
| CN111401057B (en) | Semantic analysis method, storage medium and terminal 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |