CN111325040A

CN111325040A - Language and literature translation auxiliary device

Info

Publication number: CN111325040A
Application number: CN202010172511.7A
Authority: CN
Inventors: 宋银霞; 郭保红; 贾开吉
Original assignee: Jiaozuo university
Current assignee: Jiaozuo university
Priority date: 2020-03-12
Filing date: 2020-03-12
Publication date: 2020-06-23

Abstract

The invention discloses a language translation auxiliary device, and particularly relates to the field of translation equipment. The invention can carry out multi-mode input by setting multi-mode input in a mode of keyboard, touch handwriting or character image extraction, and can carry out input in a mode of photographing identification or handwriting input when the uncommon character in the language is required to be identified, thereby enlarging the use scene, facilitating research and study of ancient workers or enthusiasts and improving the practicability of the ancient workers or enthusiasts.

Description

Language and literature translation auxiliary device

Technical Field

The invention relates to the technical field of translation equipment, in particular to a language translation auxiliary device.

Background

The teaching instrument refers to an instrument for teaching or scientific research in teaching and scientific research units. The teaching instrument can be divided into two main categories of experiment and practical training and is mainly applied to various professions of the technical and industrial categories.

The literature is a written language processed based on ancient Chinese. The processing may already be done in written language written in the earliest time in spoken language. The literature is an article composed of written languages in ancient China, and mainly comprises written languages formed on the basis of the oral languages in the period of the ancestry of the ancestors of the China. In spring and autumn warring time, articles for recording characters are not invented, the characters are recorded by bamboo slips, silk and other articles, the silk is expensive, the bamboo slips are heavy, the number of recorded characters is limited, and unimportant characters can be deleted in order to record more things on one roll of bamboo slips. Later, when the paper is used on a large scale, the use habit of the official documents from beginning to end of the governing grade is already established, and the literary characters are developed into symbols for reading and reading. The language is a word, which is characterized by writing based on characters, emphasizing the classical language, parallel-Li-parallel-link, and tone rule and arrangement without using marks, including many languages such as strategies, poems, words, songs, eight-ply, parallel-language and ancient languages. In order to facilitate reading and understanding, the literary languages in modern books are marked with punctuation marks.

However, the characters of the language are unsmooth and tedious, the famous sentences of the language are numerous and complicated, and the simple memorandum is boring and uninteresting, so that ancient workers or enthusiasts feel very arduous when studying the language, therefore, the mutual translation of the language and the white language is a topic with profound meaning and obvious value, and is a new application field of artificial intelligence technology such as natural language processing, machine translation and the like. The main purpose of the language and white language translation is to apply the natural language processing technology and the machine translation technology, input the digitized language and text into the computer, the computer automatically outputs the corresponding white language translation text, and the digitized white language text is input into the computer, and the computer automatically outputs the corresponding language and text translation text.

With the continuous development of network information technology, particularly the rapid development of technologies such as big data, cloud computing, mobile internet and the like in China, a large amount of white words and language information is accumulated on the internet, the demands of people are not limited to the reading and understanding of the existing language, more and more people want to be able to use and write the language belonging to the people in different occasions, and meanwhile, due to the fact that the language has a learning threshold, along with the continuous development of artificial intelligence, machine learning and natural language processing technologies and the continuous deepening of the application range of an information system, particularly the gradual maturity of a machine translation technology, many people want to carry out a language and white words translation auxiliary device by means of the device.

Therefore, it is desirable to provide a device for assisting the translation of a Chinese language.

Disclosure of Invention

In order to overcome the above defects in the prior art, embodiments of the present invention provide an apparatus for assisting in translating a language and literature, which can perform multi-mode input by setting multi-mode input, and performing multi-mode input by using a keyboard, touch handwriting or character image extraction, and can perform input by using a photographing recognition or handwriting input mode when a user needs to recognize a uncommon word in a language and literature, thereby enlarging a use scene, facilitating research and study by ancient workers or enthusiasts, and improving the practicability of the ancient workers or enthusiasts; in addition, the cloud computing platform is constructed, and the vocabulary analysis and the semantic analysis are operated through the cloud server, so that the operation amount and the manufacturing cost of the translation auxiliary device body equipment are reduced, the mass production and the manufacturing of the translation auxiliary device can be rapidly realized, the popularization and the popularization of the translation auxiliary device are facilitated, the operation module can directly realize the point-by-point and plane optimization and upgrade through changing the cloud algorithm, the maintenance cost is reduced, and the problems in the background technology are solved.

In order to achieve the purpose, the invention provides the following technical scheme: the utility model provides a literary composition translation auxiliary device, includes translation device, cloud ware, the translation device includes the fuselage, the fixed surface of fuselage installs touch-control display screen, keyboard, the inside fixed mounting of fuselage has PCB control panel, lithium cell, the inside of fuselage is equipped with accomodates the groove and has inserted the stylus, the bottom fixed mounting of fuselage has charging hole, speaker, one side joint of fuselage has the SIM card to hold in the palm, the surface integration of PCB control panel installs GPRS communication module, camera, central processing unit, analog-to-digital converter, ROM module and RAM module, the inside of cloud ware is equipped with cloud computing module, cloud computing module includes lexical analysis, semantic analysis, machine learning and result output.

In a preferred embodiment, the stylus pen is composed of a telescopic long rod and a capacitor head, the capacitor head is a capacitive contact pen head, the touch display screen is a capacitive touch display screen, a camera lens is arranged on the back surface of the machine body, the camera lens is located on the back side of the camera, and the output ends of the touch display screen, the keyboard and the camera are electrically connected with the input end of the PCB control board.

In a preferred embodiment, the translation device carries out cloud interconnection through GPRS communication module and cloud server, the inside of PCB control panel is provided with power amplifier circuit, the input of GPRS communication module passes through power amplifier circuit and the output electric connection that the SIM card held in the palm, the inside joint that the SIM card held in the palm has the SIM card.

In a preferred embodiment, the information processing system includes an information acquisition module, an identification conversion module, an electrical signal conversion module and a signal output module, the information acquisition module includes a character identification module, an input end of the character identification module is electrically connected with an output end of the touch display screen, the keyboard and the camera, the identification conversion module is composed of a central processing unit, the electrical signal conversion module is composed of a stylus pen, and the signal output module is composed of a GPRS communication module.

In a preferred embodiment, the machine learning comprises a traditional logic research, a cognitive model and a theoretical analysis, the machine learning performs auxiliary operations on the vocabulary analysis and the semantic analysis, the machine learning can control the algorithm optimization updating of the vocabulary analysis and the semantic analysis, the machine learning is used for carrying out the traditional logic research and the cognitive model on the operation results of the vocabulary analysis and the semantic analysis, and the operation results of the traditional logic research, the cognitive model and the theoretical analysis are used for feedback adjustment of the operation of the vocabulary analysis and the semantic analysis.

In a preferred embodiment, the traditional logic study is comprised of an auxiliary machine learning model, the traditional logic study is configured to assign a first score to the unlabeled observation, the cognitive model is comprised of a target machine learning model, the cognitive model is configured to give a second score to the unlabeled observation, the traditional logic study and the cognitive model are from different machine learning model classes, the cognitive model is a limited capacity machine learning model, the theoretical analysis is comprised of a comparison component configured to compare the first score and the second score to determine a probability that the cognitive model has returned a false positive or false negative result, the comparison component of the first score and the second score is further configured to perform a comparison comprising: determining a magnitude of a difference between the first fraction and the second fraction; determining that the target machine learning model has returned a false positive when the magnitude is negative; and determining that the target machine learning model has returned a false negative when the magnitude is positive.

In a preferred embodiment, the word recognition module comprises: when the scanning strategy acquires information of a first character image in real time, the resolution strategy extracts a definition value of the first character image and compares the definition value with a definition threshold of the character image, and when the definition value of the first character image is lower than the definition threshold of the character image, the character image processing unit outputs re-shooting information; otherwise, numbering the first character image and sending second character image extraction information to the camera; when the scanning strategy acquires information of a second character image in real time, the resolution strategy extracts a definition value of the second character image and compares the definition value with a definition threshold of the character image, and when the definition value of the second character image is lower than the definition threshold of the character image, re-shooting information is output; otherwise, numbering the second character image, comparing the definition values of the numbered first character image and the second character image, and outputting the current character image information with the highest definition value.

In a preferred embodiment, the outputting of the result comprises outputting a translation of the language to be translated in the current collected information, a comment, and a word or a sentence with a similar meaning to the language to be translated in the current collected information.

The invention has the technical effects and advantages that:

1. according to the invention, by setting multi-mode input, the multi-mode input can be carried out in a mode of keyboard, touch handwriting or character image extraction, when the uncommon word in the language is required to be recognized, the input can be carried out in a mode of photographing recognition or handwriting input, the use scene is enlarged, the research and study of ancient workers or enthusiasts are facilitated, and the practicability of the ancient workers or enthusiasts is improved;

2. according to the invention, the cloud computing platform is constructed, and the vocabulary analysis and the semantic analysis are operated through the cloud server, so that the operation amount and the manufacturing cost of the translation auxiliary device body equipment are reduced, the mass production and the manufacturing of the translation auxiliary device can be rapidly realized, the popularization and the popularization of the translation auxiliary device are facilitated, the operation module can directly realize the point-by-point and area optimization and upgrade by changing the cloud algorithm, and the maintenance cost is reduced;

3. according to the invention, through learning optimization carried out at the cloud, the multilayer perceptron is utilized to simulate brain work, a deep neural network is constructed, simple features are learned, complex features are established, learning mapping and output are realized, all levels are continuously optimized in the training process, the energizing of data processing is more and more obvious, the translation accuracy is gradually improved, and the translation assisting device is enabled to be intelligent and the translation accuracy is gradually improved as the use time of the translation system is prolonged and the number of guidance times of prediction is increased along with the increase of accumulated translation amount.

Drawings

Fig. 1 is a schematic view of the overall structure of the present invention.

Fig. 2 is a schematic view of the appearance structure of the present invention.

FIG. 3 is a diagram illustrating a translation operation structure according to the present invention.

FIG. 4 is a diagram of an information processing system according to the present invention.

FIG. 5 is a structural diagram of the machine learning intervention of the present invention.

The reference signs are: 1. a translation device; 2. a body; 3. a touch display screen; 4. a keyboard; 5. a stylus pen; 6. a PCB control board; 7. a lithium battery; 8. a charging hole; 9. a speaker; 10. a GPRS communication module; 11. a SIM card holder; 12. a camera; 13. a central processing unit; 14. a power amplification circuit; 15. an analog-to-digital converter; 16. a ROM module; 17. a RAM module; 18. a receiving groove; 19. an information processing system; 20. A cloud server; 21. a cloud computing module; 22. analyzing vocabularies; 23. semantic analysis; 24. machine learning; 25. outputting the result; 51. a capacitor head; 241. traditional logic research; 242. a cognitive model; 243. And (4) theoretical analysis.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The device for assisting the language translation shown in the accompanying drawings 1-5 comprises a translation device 1 and a cloud server 20, wherein the translation device 1 comprises a machine body 2, a touch display screen 3 and a keyboard 4 are fixedly installed on the surface of the machine body 2, a PCB control board 6 and a lithium battery 7 are fixedly installed inside the machine body 2, a storage groove 18 is formed inside the machine body 2, a handwriting pen 5 is inserted in the storage groove, a charging hole 8 and a loudspeaker 9 are fixedly installed at the bottom of the machine body 2, a SIM card holder 11 is clamped on one side of the machine body 2, a GPRS communication module 10, a camera 12, a central processing unit 13, an analog-to-digital converter 15, a ROM module 16 and a RAM module 17 are installed on the surface of the PCB control board 6 in an integrated mode, a cloud computing module 21 is arranged inside the cloud server 20, and the cloud computing module 21 comprises a vocabulary analysis 22, a semantic analysis 23.

The implementation mode is specifically as follows: by setting the multi-mode input, the multi-mode input can be carried out through the keyboard 4 and a touch handwriting or character image extraction mode, when the uncommon word in the language is required to be recognized, the input can be carried out by using a photographing recognition or handwriting input mode, the use scene is enlarged, the ancient workers or enthusiasts can conveniently study and study, and the practicability of the ancient workers or enthusiasts is improved; in addition, according to the invention, by constructing a cloud computing platform, vocabulary analysis and semantic analysis are operated through the cloud server 20, so that the operation amount and the manufacturing cost of the translation auxiliary device body equipment are reduced, mass production and manufacturing of the translation auxiliary device can be rapidly realized, the popularization and the popularization of the translation auxiliary device are facilitated, and the operation module can directly realize point-by-point and plane optimization and upgrade by changing a cloud algorithm, so that the maintenance cost is reduced.

Wherein, the stylus pen 5 comprises scalable stock and electric capacity head 51, electric capacity head 51 is capacitanc contact pen head, touch display screen 3 is capacitanc touch display screen, the lens of making a video recording has been seted up at the back of fuselage 2, the lens of making a video recording is located the dorsal part of camera 12, touch display screen 3, keyboard 4, the output of camera 12 and PCB control panel 6's input electric connection, accessible keyboard 4, the mode of touch-control handwriting or text image extraction carries out multi-mode input, can use the mode of taking a picture discernment or handwriting input to input when the rarely used word in the text of needing discernment, enlarge the use scene.

Wherein, translation device 1 carries out cloud interconnection through GPRS communication module 10 and cloud server 20, and the inside of PCB control panel 6 is provided with power amplifier circuit 14, and the input of GPRS communication module 10 passes through power amplifier circuit 14 and holds in the palm 11 output electric connection with the SIM card, and the inside joint that the SIM card held in the palm 11 has the SIM card, thereby the calculation volume and the manufacturing cost of the supplementary device body equipment of translation of construction high in the clouds computing platform reduction.

Wherein, information processing system 19 includes information acquisition module, discernment conversion module, signal of telecommunication conversion module and signal output module, information acquisition module includes the character recognition module, the input of character recognition module and touch-control display screen 3, keyboard 4, camera 12's output electric connection, discernment conversion module comprises central processing unit 13, signal of telecommunication conversion module comprises stylus 5, signal of telecommunication output module comprises GPRS communication module 10, carry out the discernment of characters and the conversion generates the signal of telecommunication.

The machine learning 24 comprises a traditional logic research 241, a cognitive model 242 and a theoretical analysis 243, the machine learning 24 executes auxiliary operation on the vocabulary analysis 22 and the semantic analysis 23, the machine learning 24 can control the algorithm optimization updating of the vocabulary analysis 22 and the semantic analysis 23, the machine learning 24 is used for conducting the traditional logic research 241 and the cognitive model 242 on the operation results of the vocabulary analysis 22 and the semantic analysis 23, the operation results of the traditional logic research 241, the cognitive model 242 and the theoretical analysis 243 are used for feedback adjustment of the operation of the vocabulary analysis 22 and the semantic analysis 23, and the multilayer perception brain is used for simulating work and building a deep neural network to learn simple features, build complex features, learn mapping and output.

Wherein the traditional logic study 241 is comprised of an auxiliary machine learning model, the traditional logic study 241 is configured to assign a first score to the unlabeled observation, the cognitive model 242 is comprised of a target machine learning model, the cognitive model 242 is configured to assign a second score to the unlabeled observation, the traditional logic study 241 and the cognitive model 242 are from different machine learning model classes, the cognitive model 242 is a limited capacity machine learning model, the theoretical analysis 243 is comprised of a comparison component, the theoretical analysis 243 is configured to compare the first score and the second score to determine a probability that the cognitive model 242 has returned a result of a false positive or a false negative, the comparison component of the first score and the second score is further configured to perform a comparison comprising: determining a magnitude of a difference between the first score and the second score; when the magnitude is negative, determining that the target machine learning model has returned a false positive; and when the amplitude is positive, determining that the target machine learning model has returned a false negative, performing feedback learning, and increasing the number of guidance times of prediction along with the increase of the translation amount so that the translation assisting device is intelligentized and has gradually improved accuracy.

Wherein, the character recognition module includes: when the scanning strategy acquires the information of the first character image in real time, the resolution strategy extracts the definition value of the first character image and compares the definition value with the definition threshold of the character image, and when the definition value of the first character image is lower than the definition threshold of the character image, the character image processing unit outputs re-shooting information; otherwise, numbering the first character image and sending second character image extraction information to the camera 12; when the scanning strategy acquires the information of the second character image in real time, the resolution strategy extracts the definition value of the second character image and compares the definition value with the definition threshold of the character image, and when the definition value of the second character image is lower than the definition threshold of the character image, the re-shooting information is output; otherwise, the second character image is numbered, the definition values of the numbered first character image and the second character image are compared, the current character image information with the highest definition value is output, and the character image identification accuracy is guaranteed.

The result output 25 includes outputting the translated text and the comment of the to-be-translated language text in the current collected information, and the vocabulary or the sentence with the meaning similar to that of the to-be-translated language text in the current collected information, so that multi-information output is realized, and the language text workers are assisted in learning and researching.

The working principle of the invention is as follows:

firstly, selecting any mode of a keyboard 4, touch handwriting or character image extraction according to a use scene, inputting a language sentence or a vocabulary, then recognizing and converting characters through an information processing system 19 in a translation device 1, converting character or image signals into digital signals, transmitting the digital signals to a cloud server 20 through a GPRS communication module 10, operating the cloud server 20, performing learning optimization through a cloud end, simulating brain work by using a multilayer perceptron, constructing a deep neural network to learn simple features, establishing complex features, learning mapping and outputting, continuously optimizing all levels in a training process, enabling the energizing of data processing to be more and more obvious, gradually improving the translation accuracy, and waiting for the translation system to increase the use duration and the number of predicted guidance times along with the increase of accumulated translation amount, the translation auxiliary device is enabled to be intelligent, the accuracy is gradually improved, the output result is finally fed back to the translation device 1, and the translated text and the annotation of the language text to be translated and the vocabulary or the sentence with the similar meaning to the language text to be translated in the current collected information are displayed and output through the touch display screen 3.

The points to be finally explained are: first, in the description of the present application, it should be noted that, unless otherwise specified and limited, the terms "mounted," "connected," and "connected" should be understood broadly, and may be a mechanical connection or an electrical connection, or a communication between two elements, and may be a direct connection, and "upper," "lower," "left," and "right" are only used to indicate a relative positional relationship, and when the absolute position of the object to be described is changed, the relative positional relationship may be changed;

secondly, the method comprises the following steps: in the drawings of the disclosed embodiments of the invention, only the structures related to the disclosed embodiments are referred to, other structures can refer to common designs, and the same embodiment and different embodiments of the invention can be combined with each other without conflict;

and finally: the above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that are within the spirit and principle of the present invention are intended to be included in the scope of the present invention.

Claims

1. A literary language translation auxiliary device comprises a translation device (1) and a cloud server (20), and is characterized in that: translation device (1) is including fuselage (2), the fixed surface of fuselage (2) installs touch-control display screen (3), keyboard (4), the inside fixed mounting of fuselage (2) has PCB control panel (6), lithium cell (7), the inside of fuselage (2) is equipped with accomodates groove (18) and has inserted stylus (5), the bottom fixed mounting of fuselage (2) has hole (8), speaker (9) of charging, one side joint of fuselage (2) has SIM card to hold in the palm (11), the surface integration of PCB control panel (6) installs GPRS communication module (10), camera (12), central processing unit (13), adc (15), ROM module (16) and RAM module (17), the inside of cloud server (20) is equipped with cloud computing module (21), cloud computing module (21) include vocabulary analysis (22), Semantic analysis (23), machine learning (24) and result output (25).

2. A literary expression translation assisting apparatus according to claim 1, wherein: stylus (5) comprise scalable stock and electric capacity head (51), electric capacity head (51) are capacitanc contact pen point, touch-control display screen (3) are capacitanc touch-control display screen, the lens of making a video recording has been seted up at the back of fuselage (2), the lens of making a video recording is located the dorsal part of camera (12), the output of touch-control display screen (3), keyboard (4), camera (12) and the input electric connection of PCB control panel (6).

3. A literary expression translation assisting apparatus according to claim 1, wherein: translation device (1) carries out cloud interconnection through GPRS communication module (10) and cloud server (20), the inside of PCB control panel (6) is provided with power amplifier circuit (14), the input of GPRS communication module (10) passes through power amplifier circuit (14) and the output electric connection that the SIM card held in the palm (11), the inside joint that the SIM card held in the palm (11) has the SIM card.

4. A literary expression translation assisting apparatus according to claim 1, wherein: information processing system (19) include information acquisition module, discernment conversion module, signal of telecommunication conversion module and signal output module, information acquisition module includes the character recognition module, the input of character recognition module and the output electric connection of touch-control display screen (3), keyboard (4), camera (12), discernment conversion module comprises central processing unit (13), signal of telecommunication conversion module comprises stylus pen (5), signal output module comprises GPRS communication module (10).

5. A literary expression translation assisting apparatus according to claim 1, wherein: the machine learning (24) comprises traditional logic research (241), a cognition model (242) and theoretical analysis (243), the machine learning (24) performs auxiliary operation on vocabulary analysis (22) and semantic analysis (23), the machine learning (24) can control the algorithm optimization updating of the vocabulary analysis (22) and the semantic analysis (23), the machine learning (24) is used for carrying out the traditional logic research (241) and the cognition model (242) on the operation results of the vocabulary analysis (22) and the semantic analysis (23), and the operation results of the traditional logic research (241), the cognition model (242) and the theoretical analysis (243) are used for feeding back and adjusting the operation of the vocabulary analysis (22) and the semantic analysis (23).

6. An apparatus for assisting the translation of a literary expression according to claim 5, wherein: the traditional logic study (241) is comprised of an auxiliary machine learning model, the traditional logic study (241) is configured to assign a first score to an unlabeled observation, the cognitive model (242) is comprised of a target machine learning model, the cognitive model (242) is configured to give a second score to the unlabeled observation, the traditional logic study (241) and the cognitive model (242) are from different machine learning model classes, the cognitive model (242) is a finite capacity machine learning model, the theoretical analysis (243) is comprised of a comparison component, the theoretical analysis (243) is configured to compare the first score and the second score, to determine the probability that the cognitive model (242) has returned a false positive or false negative result, the comparing component of the first score and the second score is further configured to perform a comparison comprising: determining a magnitude of a difference between the first fraction and the second fraction; determining that the target machine learning model has returned a false positive when the magnitude is negative; and determining that the target machine learning model has returned a false negative when the magnitude is positive.

7. An apparatus for assisting the translation of a literal language according to claim 4, characterized in that: the character recognition module comprises: when the scanning strategy acquires information of a first character image in real time, the resolution strategy extracts a definition value of the first character image and compares the definition value with a definition threshold of the character image, and when the definition value of the first character image is lower than the definition threshold of the character image, the character image processing unit outputs re-shooting information; otherwise, the first character image is numbered, and second character image extraction information is sent to the camera (12); when the scanning strategy acquires information of a second character image in real time, the resolution strategy extracts a definition value of the second character image and compares the definition value with a definition threshold of the character image, and when the definition value of the second character image is lower than the definition threshold of the character image, re-shooting information is output; otherwise, numbering the second character image, comparing the definition values of the numbered first character image and the second character image, and outputting the current character image information with the highest definition value.

8. A literary expression translation assisting apparatus according to claim 1, wherein: and the result output (25) comprises the step of outputting the translation and the annotation of the language to be translated in the current acquisition information and words or sentences similar to the meaning of the language to be translated in the current acquisition information.