TWI834293B - Natural language processing method and its system and application - Google Patents

Abstract

A natural language processing method, including: comparing a natural language through diachrony and/or synchrony, and grasping the cognitive framework of the natural language. The cognitive framework level of the natural language is generally semantics>grammar>speech>linguistic; edited this The symbolic signifier of natural language should include the phonetic signifier and the linguistic signifier. The phonetic signifier and the linguistic signifier are divided so that the symbolic signifier of natural language = phonetic signifier > linguistic signifier; this natural language The phonetic structure is formalized by including: custom symbol selection; specific permutation and combination design; and the operation of symbols and/or symbol combinations on phonemes, so that each segment corresponds to a unique symbol and/or symbol group, so that this The pronunciation of the signifier is consistent with the phonetic sequence and its pinyin structure is a segmental structure, an unchangeable language form; the set of sign elements can also form an independent, self-consistent and complete axiomatic state.

Description

Natural language processing methods and their systems and applications

The present invention relates to a natural language processing method and its system and application; in particular, it relates to a technical method for practicing the systematization of natural language symbol form axioms and its artificial system and application of results.

Natural Language Processing (NLP) is a branch of linguistics and artificial intelligence. This field explores how to process and use natural language; natural language processing includes many aspects and steps, basically including cognition, understanding, generation and other parts. Natural language cognition and understanding is to allow the computer to turn the input language into interesting symbols and relationships, and then process it according to the purpose to generate various results and applications.

Natural Language Processing Methods and Knowledge Technology In the late 1980s, natural language processing introduced the algorithm of machine learning. Instead of using programming language to command the computer all the rules in the 1950s, it established an algorithm model to let the computer learn to follow the rules. In the training data, look for specific patterns and trends in the data. There are also corpora established in the form of other non-axiomatic symbols using vector search rules. However, the calculation of vectors consumes a large amount of system resources and the effect is not effective. The final homogeneous translation of this invention is not a general non-axiomatic system set algorithm. can be accomplished.

Take "Chinese" as an example, but it is not limited to this. The "phonetic structure" of "Chinese contemporary synchronic speech" is the "phonetic sequence", which refers to the "phonetic structure" and has nothing to do with the differences in the use of symbolic characters; and then use "Symbols" represent pronunciation, forming a "Pinyin system." Taking "Chinese" again as an example, the "linguistic reference" used in today's "phonetic symbols" is out of date, lacks international versatility, and is not a traditional symbol element of computer science (26 Latin letters and 0123456789); while "Hanyu Pinyin" is due to " Symbol deformation" and/or "foley" are separated from the "speech structure" of the original.

In the natural language processing industry, there is currently no axiomatic approach to the pinyin structure and form of language symbols. However, according to this approach, symbol components can also build their own character libraries or perform word vector analysis.

The present invention provides a natural language processing method, its system and application, and systematizes the axioms of the pinyin structure and form of language symbols, including: comparing more than one natural language diachronically and/or synchronically to grasp the natural language cognitive framework possessed by human beings. Among them, human "language cognition" has a relative cognitive hierarchy order between "sounds" and "language level" (words), because humans interpret "language" with "contemporary synchronic speech", so the "voice level" " Closer to the core of language cognition, it is the "cognitive framework" of human beings for natural language. Among them, it can be found that the order of human cognitive hierarchy for natural language cognitive framework is semantics (core)>grammar>speech>linguistic.

It is known that Saussure's semiotics proposes that language signs have both signifiers and signifieds. From a linguistic point of view, Saussure pointed out that "signs" should be composed of two parts, one is the "signifier" and the other is the "signifier". "Signified" (signified), the so-called "signifier" is a "sound-image" (sound-image), and "signified" is a "concept" that connects sound-images. All symbols should have " "Signifier" and "signified" are indispensable. This invention integrates the knowledge content of the language cognitive framework and corrects the shortcomings of Saussure's semiotics. The symbolic signifier of natural language should include the phonetic signifier and the linguistic signifier. The invention divides the phonetic signifier and the linguistic signifier to make the meaning of the natural language symbol Signified = phonetic signifier > linguistic signifier. The phonetic signifier has linearity, is of auditory nature, only expands in time, and has characteristics borrowed from time, including: (a) it embodies a length, (b) this length can only be measured in one direction: it is A straight line. Among them, the straight-line nature of the phonetic signifier is the phonetic sequence. Observing the pinyin structure of the natural language phonetic sequence itself, and summarizing the natural law characteristics of the natural language phonetic sequence, we can assume that the axioms of the phonetic sequence are segment structures, which cannot be changed. complex. Among them, "segment structure": For example, Chinese Mandarin has 37 phonemes, including 21 initial consonants, 3 medial consonants, 13 finals, and 4 tones. They are arranged in order as initial consonant segment, medial consonant segment, and final vowel. There are four segments, namely segment and tone segment, and all Chinese characters are pronounced sequentially from the initial segment to the tone segment; "Unshakable": For example, the four segments of Chinese Mandarin appear and appear according to the individual needs of each Chinese character, but The sequence of the segments is fixed, and in the pronunciation of each Chinese character, the phonemes and tones of the same segment do not appear repeatedly.

Because human beings' "cognitive framework" for natural language exists at the "speech level", there is "randomness" between the "meaning referent" and "speech signifier" of natural language, which can be completely controlled by non-human beings. However, there is "arbitrariness" between "phonetic signifier" and "linguistic referent". The present invention uses this arbitrariness to convert the linguistic signifier symbol element to systematize the axioms of the pinyin structure form of language symbols, by including: formalization; The right to select the symbol elements of a custom language symbol set; the right to design the arrangement and combination of symbol elements in a custom language symbol set; and the display rules of symbol elements and/or symbol element groups are consistent with the natural law characteristics of phonetic sequences; maintaining one phoneme corresponding to one symbol The operation of components and/or symbol component groups to produce an artificial system with recognition and/or discriminability, so that the language of this natural language symbol collection can refer to the arrangement and combination state of symbol components and/or symbol component groups, which is mandatory, unique, closed and independent. In line with the natural law characteristics of phonetic sequence, its pinyin structure is a segmental symbol, an unchanging form of language expression, and axiomatic; verify the self-defined language symbol set of the present invention and all symbol elements and/or symbol element groups within it The relationship is self-consistent, independent and complete. Therefore, the segment structure is an axiom of the language symbol set, and it is the axiom of the language symbol set. Moreover, the overall language set is the symbol element and/or symbol element group of the language symbol set. It is structured by repeated permutations and combinations. Therefore, the segment structure will never change or repeat. This is the axiom of natural language. This is used to prove that the segment structure will never change or return. The axiom of natural language is true, and many kinds of The transformation of the symbol elements and/or symbol element groups of the natural language symbol set is consistent so that the multiple natural language symbol sets are equal and the elements of the symbol elements and/or symbol element groups are homogeneously assembled to achieve multi-language conversion and translation.

The natural language processing method of the present invention is used as a connection hub between the human side and the computer side. The human side uses language symbols to approximate pronunciation and foley without symbol deformation rules to meet human reading comprehension needs. The computer side uses language symbols to systematize the pinyin structure form axioms. The collection of language symbols presents segmental structures, and the unwavering formal axiom system facilitates computer-side artificial intelligence machines to understand human natural language. A system (English: system) generally refers to a group of related individuals that operate according to certain rules and can complete work that individual components cannot complete alone. The system of the present invention is an artificial system generated based on the natural law characteristics of speech sequence. The applications of the present invention include the application of natural language processing methods and system applications and the application of the results of the present invention, which further include the application of embodiments of the present invention in the field of artificial intelligence and pan-technologies.

In this embodiment, modifying the symbolic signifier of the natural language includes dividing the phonetic signifier and the linguistic signifier and adding a new custom collection symbol element of the linguistic signifier.

In this embodiment, the phonetic signifier has linearity, is auditory in nature, only expands in time, and has characteristics borrowed from time, including: (a) it embodies a length, (b) this length can only be in one direction. Measuring in degree: it is a straight line. By integrating its straight-line properties and phonetic sequence, a phonetic sequence pinyin structure is proposed, so that the phonetic sequence characteristics of this natural language are segmental structures, which cannot be changed.

In this embodiment, it further includes: making all symbol elements and/or component groups of the symbol set included in this language form self-consistent (compatible); using this language form as an axiom of the symbol set, forcing each to maintain The independence of each symbolic element and/or element group requires that the last segment must have a symbolic element and/or element group to close the entire structure, forming an independent, unique and closed form; and to make this language form conform to the segmental structure and remain unchanged. No longer, one of its characteristics is that the pinyin structure shows phonological relationships.

In this embodiment, it further includes: one segment, one symbol or one symbol group to satisfy the language form.

In this embodiment, it further includes: taking this language form as an axiom of a symbol set, it is mandatory to maintain the independence of each symbol element, and it is required that the last segment must have an element to close the entire structure to achieve self-consistency and completeness.

The present invention also provides a natural language system that utilizes the above-mentioned natural language processing method, including: a systematic set of language symbols, pinyin structures, forms and axioms and/or an artificial system invented based on the natural laws of language pronunciation.

In this embodiment, the formalized form of the symbol pinyin structure includes: symbol element selection, wherein the symbol selection includes symbol elements used in computer science; the formalized form of the symbol pinyin structure includes: symbol elements and/or symbol element combinations; phonemes; and the right to design the arrangement and combination of identifiable symbol groups.

The results of the above-mentioned natural language processing method of the present invention can also be applied to computer-readable recording media, including: transcription code and/or audio and symbol text conversion; computer program products, including: speech recognition; voice control; translation; audio conversion Text; and/or text-to-audio; and/or natural language processing in the field of artificial intelligence, including: audio integrated circuit design and/or artificial intelligence speech technology for natural language understanding.

In one embodiment, the formal axiomatic systematization of the pinyin structure of natural language symbols includes a systematic set of features of language symbol formalization; axiomatic features; and systematization features of formal axiomatics, wherein symbol elements and/or element groups It also includes equal conversion of symbol components or information modes.

In one embodiment, the symbol components and/or component groups further include: consistent translation of audio symbols, text and/or code in different languages through symbols, and homogeneous translation includes: enhanced processing of segment relationships of natural language through deep learning. (rapports syntagmatiques) and/or rapports associatifs (rapports associatifs) audio symbols, text and/or code.

In this embodiment, the systemization of the axioms of the pinyin structure and form of natural language symbols includes technical methods and artificial systems and application of results.

According to a preferred embodiment of the present invention, a natural language processing method is provided, including: comparing a natural language through diachrony and/or synchrony, and grasping the cognitive framework of the natural language, wherein the semantics > grammar > of the natural language Speech > language; modify the sign referent of this natural language, where the sign signifier of this natural language includes the phonetic signifier and the language signifier, and divide the phonetic signifier and the language signifier so that the sign referent = phonetic signifier > the language signifier; The pronunciation structure of this natural language is made by including: symbol selection; specific arrangement design rights; and operations that are consistent with phonemes, so that the pronunciation of this natural language conforms to the phonetic sequence and its pinyin structure is a segmental structure, which is unchangeable. Complex; and systematizing the axioms of the pinyin structure form of natural language symbols into a language form using one segment, one symbol or one symbol group.

In this embodiment, it further includes: treating the language form as a set of symbols, forcing the independence of each element, and requiring that the last segment must have an element to close the entire structure to achieve self-consistency and completeness.

In this embodiment, the symbol selection includes symbol elements used in computer science.

According to another preferred embodiment of the present invention, a natural language processing method is provided to implement the systematization of the axioms of the pinyin structure of natural language symbols, including: comparing more than one natural language through diachrony and/or synchrony, and grasping the characteristics of human beings. The natural language cognitive framework, in which the core order of human cognitive framework for natural language is semantics > grammar > phonetics > language; modify the symbolic referent of this natural language, where the symbolic signifier of this natural language includes the phonetic signifier and the linguistic signifier. Refers to, divides the phonetic signifier and the language signifier so that the sign refers = phonetic signifier > the language signifier; according to the pinyin structure of the language sequence, the segment is constructed, and the completion of the unwavering rules includes: formal symbol selection; specific Arrangement design rights; and operations corresponding to phonemes; and systematizing the axioms of the pinyin structure form of natural language symbols into a language form using one segment, one symbol or one symbol group.

In this embodiment, modifying the sign referent of the natural language includes adding a new linguistic signifier sign element.

In this embodiment, the phonetic signifier has linearity, is auditory in nature, only expands in time, and has characteristics borrowed from time, including: (a) it embodies a length, (b) this length can only be in one direction. Measuring in degree: it is a straight line. By integrating its straight-line properties and phonetic sequence, a phonetic sequence pinyin structure is proposed, so that the phonetic sequence characteristics of this natural language are segmental structures, which cannot be changed.

In this embodiment, the formalized form of the symbol pinyin structure includes: symbol element selection, wherein the symbol selection includes symbol elements used in computer science; the formalized form of the symbol pinyin structure includes: symbol elements and/or symbol element combinations; phonemes; and the right to design the arrangement and combination of identifiable symbol groups.

In this embodiment, it further includes: making all symbol elements and/or component groups of the symbol set included in this language form self-consistent (compatible); using this language form as an axiom of the symbol set, forcing each to be maintained The independence of each symbolic element and/or element group requires that the last segment must have a symbolic element and/or element group to close the entire structure, forming an independent, unique and closed form; and to make this language form conform to the segmental structure and remain unchanged. No longer, its characteristic pinyin structure shows phonological relationships.

In this embodiment, the systematization of the natural language symbol pinyin structure form axioms includes artificial system technology methods and application of results.

In this embodiment, the formal axiomatic systematization of the natural language symbol pinyin structure includes language symbol formalization; axiomatic; and a systematic approach to formal axiomatization, in which symbol elements and/or element groups further include equal conversion of symbol elements or messages. form.

In this embodiment, the symbol components and/or component groups further include: consistent translation of audio symbols, text and/or code in different languages through symbols, and homogeneous translation includes: enhanced processing of segment relationships of natural language through deep learning. (rapports syntagmatiques) and/or rapports associatifs (rapports associatifs) audio symbols, text and/or code.

In this embodiment, the symbol elements and/or symbol element groups of multiple natural language symbol sets can be converted into the same form so that the multiple natural language symbol sets are equal, and the symbol elements and/or symbol element groups of the multiple natural language symbol sets are homogeneously assembled to achieve multi-language conversion and translation. .

The present invention also provides a natural language system that utilizes the above-mentioned natural language processing method, including: a systematic set of language symbols, pinyin structures, forms and axioms and/or an artificial system invented based on the natural laws of language pronunciation.

The applications of the present invention include the application of natural language processing methods and system applications and the application of the results of the present invention, which further include the application of embodiments of the present invention in the field of artificial intelligence and pan-technologies.

The results of the above-mentioned natural language processing method of the present invention can also be applied to computer-readable recording media, including: transcription code and/or audio and symbol text conversion; computer program products, including: speech recognition; voice control; translation; audio conversion Text; and/or text-to-audio; and/or natural language processing in the field of artificial intelligence, including: audio integrated circuit design and/or artificial intelligence speech technology for natural language understanding.

Please refer to FIG. 1 , which is a schematic diagram of the relationship between a preferred natural language processing method and its system and applications of the present invention. The human terminal 102 uses natural language 104 and the natural language processing method 106 of the present invention as a connection hub between the human terminal 102 and the computer terminal 114 (including artificial intelligence 112). The human terminal 102 approximates pronunciation through language symbols and onomatopoeia without symbol deformation rules. To meet the needs of human reading comprehension, the computer terminal 114 systematizes the formal axioms of the pinyin structure of language symbols so that the collection of language symbols presents segmental structures. The unwavering formal axiom system facilitates the computer terminal 114 artificial intelligence 112 machine to understand human natural language. The system 108 of the present invention is an artificial system generated based on the characteristics of the natural laws of phonetic sequence, including: a systematic set of axioms of the language symbol pinyin structure form of the natural language processing method 106 of the present invention and/or inventions based on the natural laws of language phonetic sequence. It is a system. The application 110 of the present invention includes the application of the natural language processing method 106 and its system 108 and the application of the results of the present invention, which further includes the application of the embodiments of the present invention in the field of artificial intelligence and pan-technologies.

Please refer to FIG. 1A , which is a schematic flow chart of one of the preferred natural language processing methods (systematization of practical language symbols, pinyin structures, forms and axioms) of the present invention. In step 160, more than one natural language is compared diachronically and/or synchronically to grasp the natural language cognitive framework possessed by humans. Among them, human "language cognition" has a relative cognitive hierarchy between "sounds" and "language level" (words), because humans interpret "language" with "contemporary synchronic speech", so the "voice level" " Closer to the core of language cognition, it is the "cognitive framework" of human beings for natural language. Among them, it can be found that the order of human cognitive hierarchy for natural language cognitive framework is semantics (core)>grammar>speech>linguistic.

It is known that Saussure's semiotics proposes that language signs have both signifiers and signifieds. From a linguistic point of view, Saussure pointed out that "signs" should be composed of two parts, one is the "signifier" and the other is the "signifier". "Signified" (signified), the so-called "signifier" is a "sound-image" (sound-image), and "signified" is a "concept" that connects sound-images. All symbols should have " "Signifier" and "signified" are indispensable. In step 170, the present invention integrates the knowledge content of the language cognitive framework and corrects the shortcomings of Saussure's semiotics. The symbolic signifier of natural language should include the phonetic signifier and the linguistic signifier. The present invention divides the phonetic signifier and the linguistic signifier into two categories: The signified meaning of a natural language symbol = phonetic signifier > linguistic signifier; the phonetic signifier is linear, auditory in nature, only expands in time, and has characteristics borrowed from time, including: (a) it embodies a length, ( b) This length can only be measured in one direction: it is a straight line. Among them, the straight-line nature of the phonetic signifier is the phonetic sequence. Observing the pinyin structure of the natural language phonetic sequence itself, and summarizing the natural law characteristics of the natural language phonetic sequence, we can assume that the axioms of the phonetic sequence are segment structures, which cannot be changed. Repeat, as shown in step 180. Among them, "segment structure": For example, Mandarin Chinese has 37 phonemes, including 21 initial consonants, 3 medial consonants, 13 finals, and 4 tones. They are arranged in order as initial consonant segment, medial consonant segment, and final vowel. There are four segments, namely segment and tone segment, and all Chinese characters are pronounced sequentially from the initial segment to the tone segment; "Unchangeable": For example, the four segments of Chinese Mandarin appear and appear according to the individual needs of each Chinese character, but The sequence of the segments is fixed, and in the pronunciation of each Chinese character, the phonemes and tones of the same segment do not appear repeatedly.

Because human beings' "cognitive framework" for natural language exists at the "speech level", there is "randomness" between the "meaning referent" and "speech signifier" of natural language, which can be completely controlled by non-human beings. However, there is "arbitrariness" between "phonetic signifier" and "linguistic referent". Therefore, in step 180, the present invention uses this method to arbitrarily convert the linguistic signifier symbol element to systematize the axioms of the pinyin structure form of language symbols. Including: formalization; the right to select symbol elements of a custom language symbol set; the right to design the arrangement and combination of symbol elements in a custom language symbol set; and the symbol element and/or symbol element group display rules are consistent with the natural law characteristics of the phonetic sequence; in the steps 190. The operation of maintaining a phoneme corresponding to a symbol element and/or symbol element group to produce an artificial system with recognition and/or discriminability, so that the language of this natural language symbol set can refer to the symbol element and/or symbol element group. The permutation and combination state is forced to be unique and closed, independent and combined with the natural law characteristics of phonetic sequence. Its pinyin structure is a segmental symbol, an unchanging language expression form, and axiomatic; verify the custom language symbol set of the present invention and all the characters in it. The relationship between semiotic elements and/or semiotic element groups is self-consistent, independent and complete. Therefore, the segment structure will never move or converge, which is the axiom of the language symbol set, and because the overall language set is a language symbol set It is structured by repeated arrangement and combination of symbolic elements and/or groups of symbolic elements. Therefore, the structure of segmental symbols will never be moved or repeated. This is the axiom of natural language. This is used to prove that the structure of segmental symbols will never be moved or repeated in natural language. Assume that the axiom is true, and the symbol elements and/or symbol element groups of multiple natural language symbol sets can be converted into the same set, so that the multiple natural language symbol sets are equal, and the symbol elements and/or symbol element group elements are homogeneous and co-assembled to achieve multi-language conversion and translation. .

The natural language processing method of the present invention is used as a connection hub between the human side and the computer side. The human side uses language symbols to approximate pronunciation and foley without symbol deformation rules to meet human reading comprehension needs. The computer side uses language symbols to systematize the pinyin structure form axioms. The collection of language symbols presents segmental structures, and the unwavering formal axiom system facilitates computer-side artificial intelligence machines to understand human natural language. A system generally refers to a group of related individuals that operate according to certain rules and can complete work that individual components cannot complete alone. The system of the present invention is an artificial system generated based on the natural law characteristics of speech sequence. The applications of the present invention include the application of natural language processing methods and system applications and the application of the results of the present invention, which further include the application of embodiments of the present invention in the field of artificial intelligence and pan-technologies.

Please refer to Figure 2, which shows the development intention of human understanding of the cognitive framework of natural language, including: language definition 202; academic integration 204 (comparative treatment of modern linguistics and Chinese linguistics); Shimai 206 (Shimai and Chinese Evolutionary relationship comparison); cognitive stage sequence 208 (language cognitive stratification); cognitive framework 210; and Chinese phonetic sequence 212. The explanations are as follows.

Definition of “Language”: Language is a system of communication, organized by pronunciation, language, and grammar; in other words, language is a communication system used by a specific group of people (“Language” in British English: “A system of communication consists of sounds, words, and grammar, or the system of communication used by people in a particular country or type of work.").

Please refer to Figure 2A, which is a schematic diagram of comparative processing between “modern linguistics” and “Chinese linguistics”. In the research field of "Chinese linguistics", some scholars will use the term "language and characters" to "compare and process" the correlation between "character pronunciation and grapheme". The "language" in "Language and Characters" traces back to "Chinese phonology" and is "diachronic study of phonetics" (Sounds). The "character" in "Language and Characters" traces back to "Chinese phonology" and is "Sounds". A Diachronic Study of Glyphs” (Words). However, the term "language and writing" is not appropriate when comparing "modern linguistics". According to the above quotations, we know that in the term "modern linguistics", the so-called "language" ( Language) actually covers two different levels of language concepts: "Sounds" and "Words". Therefore, if we want to integrate and compare the "language cognition" of "modern linguistics" and "Chinese linguistics", "language and characters" should be revised to "speech and language", so as not to be used in the discussion. Falling into conceptual misunderstandings.

Through the above conceptual clarification, we have successfully integrated and connected the language concepts discussed in "modern linguistics" and "Chinese linguistics". This process also connects the "timeline" of China's "tradition" to "modernity", allowing us to more clearly observe the "evolution of Chinese" that has occurred in the "diachrony" of the Chinese nation. Please refer to Figure 2B , which is a schematic diagram comparing the relationship between "time pulse" and "Chinese evolution". The timing (diachronic/history) ranges from ancient times (Shang, Zhou, Qin and Han), medieval times (Wei, Jin, Sui and Tang), modern times (Song, Yuan, Ming and Qing), modern times (Chinese nation) to contemporary times (synchronic), which are compared with phonetics (phonology/phonology) and The evolution of Chinese language (Chinese characters/writing), from ancient times (the phonology of the Book of Songs and oracle bones and golden seal scripts), to the middle ages (the Guangyun phonology and official script and cursive script), to the modern times (the phonetic phonology and regular script and running script of the Central Plains) and traditional and simplified) to contemporary (synchronic speech and synchronic language).

Now, we return to the time point we are in - contemporary times, and re-examine the relationship between Chinese "speech" and "language" from a "synchronic" perspective. Please refer to Figure 2C, which is a schematic circle of language cognitive layering. Target map, we made a major discovery: human "language cognition" has a relative order of "speech level" (Sounds) and "language level" (Words), because "speech" or "language" are constantly changing It has evolved over time. However, we living in the contemporary era, whether we are conducting "speech origin research" with "Chinese phonology" or "language origin research" with "Chinese kanjiology", cannot escape from " Interpretation of "Contemporary Synchronic Speech", it can be seen that the "speech level" (Sounds) is closer to the core of language cognition (Core), and is a "cognitive framework" in a relative sense, and this cognitive framework is explained here as people's interpretation of external reality Psychological archetypes of the world that serve as the basis for understanding, identifying, and defining experiences.

To further explain, a natural language is compared diachronically and/or synchronically to grasp the cognitive framework of this natural language, where the natural language's semantics (core) > grammar > phonetics > language.

For example: The way people in the Zhou Dynasty used language may be to speak the dialect of the phonological system of the Book of Songs orally and write in gold inscriptions by hand. However, when we explore the source and study the language of the Zhou Dynasty, we cannot directly interpret the Jin dynasty using the phonetic pronunciation of the Book of Songs like the people of the Zhou Dynasty. text, because our "linguistic cognition" relies on "contemporary synchronic speech" to structure meaning. This is why no matter which classical dynasty is used as the performance background of modern drama, the "spoken language" of the drama characters is still the "contemporary synchronic voice". We can also find clues and confirm it from our own life process. For example, in childhood, we first learn to talk, and later we read and write. What's more, in an era when education was not common, some people could not read words in their entire lives and were called "illiterates". However, these people were still able to use language for listening and speaking responses, which further reflected human "language cognition" ” The fact that “speech” precedes “language” in class order. In other words, the existence of "spoken language" in language is the best evidence that "speech" precedes "language" in "language cognition".

In view of this, please refer to Figure 2D, which is a schematic diagram of a cognitive framework. Since speech has a sound sequence, and through the above-mentioned concepts and empirical hypothesis axioms, the cognitive framework of the natural law of speech sequence is explained here as a psychological prototype for people to interpret the external real world, and is used as the basis for understanding, identifying and defining behavioral experiences. . And Figure 2E is a schematic diagram of Chinese phonetic sequence.

Please refer to FIG. 3 , which is a schematic diagram of the invention steps of a preferred Saussurean semiotics modification of the present invention. On the other hand, since the end of the 19th century, semiotics has initially emerged as a discipline to study language. However, nowadays semiotics that generally talks about language has mostly been referred to as linguistics. Today’s semiotics refers to structuralist semiotics. (structuralist semiotics). The characteristic of Saussure's semiotics is that a single sign (Sign) is divided into two parts: signifier and signified, as shown in Semiotics of Conventions 302. The signifier is the phonetic image of the symbol; the signified is the conceptual part of the symbol. A whole composed of two parts is called a symbol. The relationship between the signifier and the signified is arbitrary and not necessarily connected. For example, the pronunciation and word combination of "tree" in English refer to the concept of "a leafy plant with wooden branches as the main body" due to conventional habits. Linguistic symbol is a two-sided psychological entity: concept and sound image. Saussure called the combination of a concept and a sound image a sign, the concept a "signified" (signifié), and the sound image a "signifier" (significant).

Combining the above-mentioned "cognitive framework" of "language cognition" and Saussure's explanation of "symbol", because "speech" precedes "language" in "language cognition", the "signifier" can actually be divided into "Phonetic signifier" and "linguistic signifier", as shown in Modified Semiotics 304, and "phonological signifier" is the subject of "sound image". Because natural language has "spoken language" and phonetic cognition precedes linguistic cognition, the necessary function of "linguistic signifier" is to summon "phonological signifier". Human beings' "cognitive framework" for natural language exists at the "speech level", so there is "randomness" between the "speech signifier" and "meaning signifier" of natural language, which can be completely controlled by non-human beings. However, there is "arbitrariness" between "phonetic signifier" and "linguistic referent", because "spoken language" is a comprehensible existence, so language programming can be carried out through the method of "summoning voice" to form a "systematic" "Synchronic Language". For example: "one sound, one symbol" of the cryptographic mapping method, one phoneme corresponds to one symbol, and the phonetic structure of "phonetic reference" corresponds to the text symbol of "linguistic reference".

Next, revise the symbolic signifier of this natural language, where the symbolic signifier of this natural language includes the phonetic signifier and the linguistic signifier, and divide the phonetic signifier and the linguistic signifier so that the sign refers = phonetic signifier > linguistic signifier. Linguist Saussure pointed out from a linguistic point of view that "sign" should be composed of two parts, one is the "signifier" and the other is the "signified". The so-called "signifier" is " "Sound-image" (sound-image), and "signified" is the "concept" that connects sound-images. All symbols should have a "signifier" and a "signified", both of which are indispensable.

In addition, modifying the linguistic signifier is because human beings' "cognitive framework" for natural language exists at the "speech level", so there is "randomness" between the "meaning signifier" and "speech signifier" of natural language, which cannot be artificially modified. fully control. However, there is "arbitrariness" between "phonological signifier" and "linguistic referent". The present invention uses this arbitrariness to convert the linguistic signifier symbol element to practice the axioms of the pinyin structure form of the language symbol and systematize the phonetic structure form of this natural language. Transformation makes the pronunciation of this natural language conform to the phonetic sequence and its pinyin structure becomes a segmental structure, which is an unchangeable language form. Among them, by including: symbol selection; specific arrangement design rights; and operations consistent with phonemes, the pronunciation of this natural language is consistent with the phonetic sequence and its pinyin structure is a segmental structure, an unchangeable language form. Among them, symbol selection includes symbol components used in computer science. In one embodiment, for example, using contemporary standard keyboards and computer input methods, 26 basic Latin (English) characters are used to mark 37 phonemes in Chinese, and tones are presented as 1, 2, 3, and 4; in the phonetic signifier In terms of language signifiers, one sound (phoneme) and one symbol (symbol) must be matched, that is, the onomatopoeia of the characters is given priority, and the onomatopoeia is given priority in the Chinese pronunciation order of the English pronunciation. , and those who cannot grasp the onomatopoeic association regard the characters as symbols. The symbols are replaced by phonetic sounds based on psychological cognition to meet the pinyin needs, and the permutations and combinations in mathematical theory are used to maintain the uniqueness of the symbols in the system. This embodiment is used to illustrate a preferred implementation method of the present invention and is not intended to limit the implementation of the present invention.

Please refer to Figure 4, which is a schematic diagram of a preferred linguistic axiom that the segmental structure of the segment cannot be moved. Saussure believes that the phonetic signifier has linearity, is of auditory nature, only expands in time, and has characteristics borrowed from time: (a) it embodies a length, (b) this length can only be measured in one direction : It is a straight line. In view of this, integrating its linear nature and phonetic sequence (natural law phenomenon of Pinyin structure 402), a phonetic sequence Pinyin structure is proposed, so that the phonetic sequence characteristics of this natural language language are segmental constructions, which cannot be moved or changed, such as the Pinyin structure is artificially processed 404 is shown. And because language is a large collection of symbol sets, the sound segment constructs a symbol and never moves or recovers. In terms of convergence, it is an axiom of symbol collection; in terms of divergence, it is an axiom of linguistics.

The inventor would like to emphasize here that what it refers to is the formalization of phonetic structure and is not limited to differences in the use of text symbols. Among them, the shape of the formalization of phonetic structure: is the symbol element selection, where the symbol selection includes the symbol elements used in computer science. The formal formula of phonetic structure: it is a symbol element and/or a group of symbol elements, combined with phonemes, and the arrangement and combination design of the symbol group can be recognized. Its characteristics: no symbol deformation. Furthermore, in a preferred embodiment, the symbols forming the symbols of the above-mentioned sound segments are in the shape of sound, which is the smallest unit of the semantics of this natural language. In addition, apart from the relationship between cause and effect, the above steps do not limit their order. Please refer to Table 1, which is a comparison table of a preferred Chinese Pinyin embodiment of the present invention. [Table 1] A comparison table of a preferred Chinese Pinyin embodiment of the present invention. Chinese Pinyin Chinese examples Pinyin of the present invention initial consonant b glass b p slope p m touch m f Buddha f d Got it d t Special t n ne ㄋ n l Leㄌ l g Brother g k Branch k h drink h j base j q bully q x Hope c zh know zv ch Chi ㄔ xv sh Poetry sv r Japan v z capital z c Female x s Thinking s stepmother i clothing i u black ㄨ w ü roundabout y vowels a Ah r o Oh ㄛ o e goose u ê Hey e ai Sorry ai ei eh ㄟ ae ao boil au ou Europe ao an An ㄢ am en Yes an ang Ang ㄤ arm eng Henry arn er son ㄦ u

Among them, the initial consonant: v, the first one is ㄖ, and the second one is retroflex; mediates: i, w, y; finals: two characters a+(), three characters ar+(); tones: 1, 2, 3 or 4 sounds.

Among them, the pinyin symbol law of the present invention includes: a four-layer structure, which is unchanging, that is, the phonetic structure contains 37 phonemes, which are 21 initials, 3 mediators, 13 finals, and the tones have 4 tones, arranged in order There are four sound layers: initial consonant layer, medial layer, final layer, and tone layer. All Chinese characters are combined and pronounced in sequence from the initial consonant layer to the tone layer; the four sound layers appear and appear according to the individual needs of each Chinese character, but the order of the sound layers It is fixed and unmovable, and in the pronunciation of each Chinese character, the phonemes and tones of the homophonic layer do not appear repeatedly.

In addition, the steps of systematizing the formal axioms according to the present invention further include: treating the above-mentioned language form as an axiom of a symbol set, forcing the independence of each element to be maintained, and requiring that the last segment must have an element to close the overall structure. Achieve self-consistency and completeness. One segment, one symbol or one symbol group satisfies the above language form. Among them, Saussure believes that the phonetic signifier has linearity, is of auditory nature, only expands in time, and has characteristics borrowed from time: (a) it embodies a length, (b) this length can only be in one dimension Upper measurement: It is a straight line. In view of this, speech has a phonetic sequence, and through the above-mentioned concepts and empirical assumptions, the language form of the present invention is a segmental structure, which cannot be changed. Taking Chinese as an example (but not limited to this), the four-segment structure , unwavering. And because language is a large collection of symbol sets, sound segments constitute symbols, and it is an axiom of symbol collection in terms of convergence and linguistics axiom in terms of divergence. Except for the relationship between cause and effect, the above steps do not limit their order.

From another perspective, a mathematical model of an axiomatic system is a well-defined set that gives meaning to undefined terms that appear in the system in a way that is consistent with the relationships defined in the system. The existence of specific models can prove the self-consistency (compatibility) of the system. Models can also be used to show the independence of an axiom in a system. By constructing an efficient model of a subsystem excluding a particular axiom, we show that the omitted axiom is independent if its correctness cannot be derived from the subsystem. Two models are said to be isomorphic if their elements can establish a one-to-one correspondence, and in a way that preserves the relationship between them. An axiomatic system in which each model is isomorphic to another is called categorical, and the categorizable property ensures the completeness of the system. From this perspective, the above-mentioned Pinyin structure axiomatization of the present invention is explained, that is: all symbol elements and/or element groups of the symbol set included in the above-mentioned language form are self-consistent (compatibility); the above-mentioned language form is regarded as The axioms of symbol sets force the maintenance of the independence of each symbol component and/or component group, requiring that the last segment must have a symbol component and/or component group to close the entire structure, forming an independent, unique and closed structure; and making the above language The form conforms to the segmental structure and remains unchanged, and its characteristic pinyin structure shows the phonological relationship. The above-mentioned system (English: system; German: System; French: système; Spanish: sistema) generally refers to a group of related individuals that operate according to certain rules and can complete work that individual components cannot complete alone. Systems are divided into two categories: natural systems and man-made systems. The systematization of language symbols, pinyin structures, forms and axioms of the present invention is based on artificial system technical methods and application of results.

In the present invention, the formal axioms of natural language are systematized as the connection hub between human beings and computer terminals. The human terminal meets human reading comprehension needs through Chinese and English approximate pronunciation (foley), and the computer terminal uses one segment, one symbol/one symbol group. It satisfies the preservation of the form of language (segment construction, unchanging) and plans specific permutations and combinations to facilitate machine understanding of the form (the axioms of permutations and combinations).

What the inventor wants to explain here is that according to the citation of "Modern Philology": "Language is a communication system organized by speech, language, and grammar." If the "cognitive framework" of language cognition exists in "speech" , then grammar (grammar) is not just a "rule" in "written language", but should refer to the existence of "laws" in the "speech structure", and "laws" are different from rules and are not made by humans, but are based on nature. An inductive statement of a phenomenon. In addition, a strictly complete axiom system must have three basic requirements: self-consistency (compatibility), independence and completeness. This part is within the scope of common knowledge, so the inventor has not elaborated on it except the relevant parts. Ink it.

Next, compare more than one natural language diachronically and/or synchronically to grasp the natural language cognitive framework possessed by humans. The core order of human cognitive framework for natural language is semantics > syntax > phonetics > language. For descriptions of diachronic and/or synchronic comparisons and the core sequence of the cognitive framework, please refer to the above descriptions of Figures 2A to 2C and will not be repeated here.

Next, the sign referent of this natural language is modified, where the sign signifier of this natural language includes a phonetic signifier and a linguistic signifier, and the phonetic signifier and the linguistic signifier are divided so that the sign referent = phonetic signifier > linguistic signifier. Among them, the modified sign referent of this natural language includes a new linguistic signifier sign element.

Then, the symbols are constructed for the segments according to the pinyin structure of the language sequence, and the unwavering rules are completed, including: formal symbol selection; specific arrangement design rights; and operations consistent with phonemes. Among them, the phonetic signifier has linearity, is auditory in nature, only expands in time, and has characteristics borrowed from time, including: (a) it embodies a length, (b) this length can only be measured in one direction: It is a straight line. By integrating its linear properties and phonetic sequence, a phonetic sequence pinyin structure is proposed, so that the phonetic sequence characteristics of this natural language language are segmental structures, which cannot be changed. And because language is a large collection of symbol sets, the sound segment constructs a symbol and never moves or recovers. In terms of convergence, it is an axiom of symbol collection; in terms of divergence, it is an axiom of linguistics.

Then, the pinyin structure of the natural language symbol is formalized into a language form by using one segment and one symbol or one symbol group. Among them, the formalized form of the symbol pinyin structure includes: symbol element selection, where the symbol selection includes symbol elements used in computer science; the formalized form of the symbol pinyin structure includes: symbol elements and/or symbol element groups; combined with phonemes; and The right to design the arrangement and combination of identifiable symbol groups.

In the above steps, the pinyin structure of the present invention is axiomatized. Among them, the mathematical model of an axiomatic system is a well-defined set that gives meaning to undefined terms that appear in the system in a way that is consistent with the relationships defined in the system. The existence of specific models can prove the self-consistency (compatibility) of the system. Models can also be used to show the independence of an axiom in a system. By constructing an efficient model of a subsystem excluding a particular axiom, we show that the omitted axiom is independent if its correctness cannot be derived from the subsystem. Two models are said to be isomorphic if their elements can establish a one-to-one correspondence, and in a way that preserves the relationship between them. An axiomatic system in which each model is isomorphic to another is called categorical, and the categorizable property ensures the completeness of the system. From this perspective, all symbolic elements and/or element groups of the symbolic set included in this language form are self-consistent (compatible). Taking this language form as an axiom for a set of symbols, it is mandatory to maintain the independence of each symbol component and/or component group. It is required that the last segment must have a symbol component and/or component group to close the entire structure, forming an independent, unique and closed structure. . This language form conforms to the segmental structure, and its characteristic pinyin structure shows the phonological relationship. The above-mentioned system (English: system; German: System; French: système; Spanish: sistema) generally refers to a group of related individuals that operate according to certain rules and can complete work that individual components cannot complete alone. . Systems are divided into two categories: natural systems and man-made systems. The systematization of language symbols, pinyin structures, forms and axioms of the present invention is based on artificial system technical methods and application of results.

Please refer to Figure 5, which is a schematic diagram of a systemized set theory of axiomatic set theory of pinyin structure form and axiom of one of the preferred language symbols of the present invention. The systematization of the form axiom of the pinyin structure of natural language symbols in the present invention includes the language symbol formalization 502; the axiomization 504; and the systematization 506 method of the form axiomization, in which the symbol elements and/or element groups further include equal conversion of symbol elements or audio forms. .

The symbol components and/or component groups of the present invention further include: consistent translation of audio symbols, text and/or program codes in different languages through symbols, and homogeneous translation includes: enhanced processing of natural language syntagmatiques (rapports syntagmatiques) through deep learning. ) and/or rapports associatifs audio symbols, text and/or code.

The pinyin symbols of the present invention are unique under the law. Each combination of symbols can be regarded as a unique existence under the law. The overall system of language is like a big puzzle, and each small puzzle piece that makes up the big puzzle must be specific. Uniquely, the pinyin of the present invention has no symbol deformation, so it is a breakthrough in language information processing and can smoothly convert the pinyin into digital components.

Please refer to Figure 6, which is a diagram of a preferred embodiment of the present invention. The natural language processing 610 method, system and results of the present invention are used as a connection hub between the human terminal 612 and the computer terminal 614. The human terminal 612 obtains the human terminal results 622 through natural language processing through language symbols that approximate pronunciation and have no symbol deformation rules. Foley meets the needs of human reading comprehension, and the computer terminal 614 uses the computer terminal results of natural language processing 624 to systematize the pinyin structure of language symbols through formal axioms, so that the collection of language symbols presents the segment structure. The unchangeable formal axiom system is convenient for the computer terminal 614 artificial intelligence machine understands human natural language. In this embodiment, the symbol elements and/or symbol element groups of multiple natural language symbol sets can be converted into the same form, so that the symbol elements and/or symbol element groups of the multiple natural language symbol sets are homogeneously collected and stored in one /Multi-language common symbol pool 620, therefore, when different natural languages are processed by the natural language processing 610 of the present invention, the corresponding symbol elements and/or symbol elements can be retrieved from the /Multi-language common symbol pool 620 Group to achieve multi-language conversion and translation.

The following shows the phonological structure:

The inventor illustrates this with an embodiment. Please refer to Table 2, which is a comparison table of one embodiment of the present invention. The rhyme ends with ㄚ. The quotation here is transcribed in Chinese pinyin to rhyme with a, and the citation in the present invention is transcribed in pinyin to rhyme with r. In other parts, the word "sunset" has deformation problems in Chinese pinyin (the medial i is deformed in the word "yang"), but the pinyin i of the present invention is clear at a glance. [Table 2] Comparison table of one embodiment of the present invention Tian Jing Sha Qiu Si (Lyrics: Ma Zhiyuan) Withered vines and old trees, dim crows‖ Small bridges and flowing water, people's houses‖ The west wind and thin horses on the ancient road‖ The sunset‖ The heartbroken people are at the end of the world‖ phonetic symbols ㄎㄨㄊㄥˊㄌㄠˇㄕㄨˋㄏㄨㄣㄧㄚ‖ ㄒㄧㄠˇㄑㄧㄠˊㄌㄧㄡˊㄕㄨㄟˇㄖㄣˊㄐㄧㄚ‖ ㄍㄨˇㄉ ㄠˋㄒㄧㄈㄥ ㄕㄡˋㄇㄚˇ‖ ㄒㄧˋㄧㄤˊㄒㄧㄒㄧㄚˋ‖ ㄉㄨㄢˋㄔㄤˊㄖㄣˊㄗㄞˋㄊㄧㄢㄧㄚˊ‖ Chinese Pinyin ku ¹ teng ² lao ³ shu ⁴ hun ¹ ya ¹ ‖ xiao ³ qiao ² liu ² shui ³ ren ² jia ¹ ‖ gu ³ dao ⁴ xi ¹ feng ¹ shou ⁴ ma ³ ‖ xi ⁴ yang ² xi ¹ xia ¹ ‖ duan ⁴ chang ² ren ² zai ⁴ tian ¹ ya ² ‖ Pinyin of the present invention kw ¹ tarn ² lau ³ svw ⁴ hwan ¹ ir ¹ ‖ ciau ³ qiau ² liao ² svwae ³ van ² jir ¹ ‖ gw ³ dau ⁴ ci ¹ farn ¹ svao ⁴ mr ^{3 ‖} ci ⁴ iarn ^{2 ci 1 cir 1} ‖ dwam ⁴ xvarm ² van ² zai ⁴ tiam ¹ ir ² ‖

The inventor illustrates this with another embodiment. Please refer to Table 3, which is a comparison table of another embodiment of the present invention. The rhyme ends ㄝ, and the quotation here is transcribed in Chinese pinyin to rhyme ê, and the rhyme is e in the pinyin of the present invention, and ê is not convenient for keyboard input. [Table 3] Comparison table of another embodiment of the present invention Excerpts from the Ages (Lyrics: Xu Song) Summer cicada and winter snow‖ are just a glimpse of reincarnation‖ Enlightenment and cultivation‖ Regardless of fate and calamity in life‖ phonetic symbols ㄒㄧㄚˋㄔㄢˊㄉㄨㄥㄒㄩㄝˇ‖ ㄅㄨˋㄍㄨㄛˋㄌㄨㄣˊㄏㄨㄟˊㄧˋㄆㄧㄝˇ‖ ㄨˋㄉㄠˋㄒㄧㄡ ㄌㄧㄢˋ‖ ㄅㄨˊㄨㄣˋㄧˋㄕㄥㄩㄢˊㄐㄧㄝˊ‖ Chinese Pinyin xia ⁴ chan ² dong ¹ xuê ³ ‖ bu ² guo ⁴ lun ² hui ² yi ⁴ piê ³ ‖ wu ⁴ dao ⁴ xiu ¹ lan ⁴ ‖ bu ² wen ⁴ yi ⁴ sheng ¹ yuan ² jiê ² ‖ Pinyin of the present invention cir ⁴ xvam ² dwarn ¹ cye ³ ‖ bw ² gwo ⁴ lwan ² hwae ² i ⁴ pie ³ ‖ w ⁴ dau ⁴ ciao ¹ liam ⁴ ‖ bw ² wan ⁴ i ⁴ svarn ¹ yam ² jie ² ‖

The inventor illustrates this with yet another embodiment. Please refer to Table 4, which is a comparison table of yet another embodiment of the present invention. The rhyme endings ㄟ and ㄝ, "zui", "month", "regret", and "stele" all rhyme, and the Chinese pinyin rhyme is mixed. The pinyin rhyme of the present invention is relatively clear. In addition, "zui" and "微" in the first paragraph have a similar pronunciation relationship, which cannot be seen in the Chinese pinyin zui ⁴ and wei ² , but the pinyin zwae ⁴ and wae ² of the present invention are clear at a glance. [Table 4] Comparison table of another embodiment of the present invention Excerpt from "Fa Ruxue" (lyrics: Fang Wenshan) Drunk in the mortal world‖ The years of being slightly tipsy‖ I have no regrets‖ Carving a monument to love you forever‖ phonetic symbols ㄏㄨㄥˊㄔㄣˊㄗㄨㄟˋ‖ ㄨㄟˊㄒㄩㄣㄉㄜㄙㄨㄟˋㄩㄝˋ‖ ㄨㄛˇㄩㄥˋㄨˊㄏㄨㄟˇ‖ ㄎㄜˋㄩ ㄥˇㄕˋㄞ ˋㄋㄧˇㄉㄜㄅㄟ‖ Chinese Pinyin hong ² chen ² zui ⁴ ‖ wei ² xun ¹ de ¹ sui ⁴ yue ⁴ ‖ wo ³ yong ⁴ wu ² hui ³ ‖ ke ⁴ yong ³ shi ⁴ ai ⁴ ni ³ de ¹ bei ¹ ‖ Pinyin of the present invention hwarn ² xvan ² zwae ⁴ ‖ wae ² cyan ¹ du ¹ swae ⁴ ye ⁴ ‖ wo ³ yarn ⁴ w ² hwae ³ ‖ ku ⁴ yarn ³ sv ⁴ ai ⁴ ni ³ du ¹ bae ¹ ‖

The inventor illustrates this with yet another embodiment. Please refer to Table 5, which is a comparison table of yet another embodiment of the present invention. The rhyme endings ㄠ and ㄜ are translated into Chinese Pinyin as "的", "好", and "流". The finals are "e", "ao", and "ao" respectively. The rhyme relationship cannot be seen. The finals are translated into Pinyin of the present invention. They are "u", "au", and "au" to see the rhyme relationship. [Table 5] Comparison table of yet another embodiment of the present invention An accidental excerpt (lyrics: Xu Zhimo) You have your ‖ I have mine ‖ direction ‖ whether you remember it ‖ or better you forget ‖ the light that shines on each other during this intersection ‖ phonetic symbols ㄋㄧˇㄧㄡˇㄋㄧˇㄉㄜ‖ ㄨㄛˇㄧㄡˇㄨㄛˇㄉㄜ‖ ㄈㄤㄒㄧㄤˋ‖ ㄋㄧˇㄐㄧˋㄉㄜˊㄧㄝˇㄏㄠˇ ‖ ㄗㄨㄟ ㄋ ㄋ ˇ ˇ ㄧ ㄏㄠ ㄨㄤ ㄉ ㄉ ㄉ ˋ‖ ㄗㄞ ㄓㄜ ˋ ㄐ ㄐ ㄐ ㄏ ㄨㄟ ˋ ㄏㄨ ˋ ˋ ㄉㄜ ㄍ ㄤ ㄤ ㄧ ㄤ ˋ‖ ㄤ ㄤ ㄤ Chinese Pinyin ni ³ you ³ ni ³ de ¹ ‖ wo ³ you ³ wo ³ de ¹ ‖ fang ¹ xiang ⁴ ‖ ni ³ ji ⁴ de ² ye ³ hao ³ ‖ zui ⁴ hao ³ ni ³ wang ⁴ diao ⁴ ‖ zai ⁴ zhe ⁴ jiao ¹ hui ⁴ shi ² hu ⁴ fang ⁴ de ¹ guang ¹ liang ⁴ ‖ Pinyin of the present invention ni ³ iao ³ ni ³ du ¹ ‖ wo ³ iao ³ wo ³ du ¹ ‖ farm ¹ ciarm ⁴ ‖ ni ³ ji ⁴ du ² ie ³ hau ³ ‖ ni ³ zwae ⁴ hau ³ warm ⁴ diau ⁴ ‖ zai ⁴ zvu ⁴ jiau ¹ hwae ⁴ sv ² hw ⁴ farm ⁴ du ¹ gwarm ¹ liarm ⁴ ‖

Please refer to Table 6, which is a comparison table of another preferred Chinese pinyin embodiment of the present invention. Compared with Table 1, except for initial consonants (ㄅ~ㄙ), mediates (ㄧ, ㄨ, ㄩ) and finals (ㄚ~ In addition to the pinyin of ㄥ), it also lists the tones (1~4 sounds) and the pairs of medial vowels (Qi Te Hu, He Kou Hu, Pinch Lip Hu) and finals (monophthong finals, diphthong finals, nasal finals) classification. In short, Table 1 and Table 6 are systems and results produced by using the natural language processing method of the present invention, and they all have the properties and characteristics of the natural language processing method of the present invention (see the above description for details, see below (This will not be described again), in addition to being applied to the implementation examples in Tables 2 to 5 above, it can also be applied to computer-readable recording media, including: transcribing program codes and/or converting audio and symbol text; computer program products, Including: speech recognition; voice control; translation; audio to text; and/or text to audio; and/or natural language processing in the field of artificial intelligence, including: audio integrated circuit design and/or artificial intelligence speech technology for natural language understanding, In order to solve the shortcomings and inconveniences in the implementation of current products, the present invention is not limited here. In addition, the application of the present invention also includes the application of the natural language processing method and its system application and the application of the results of the present invention, which further includes the application of the embodiments of the present invention in the field of artificial intelligence and pan-technologies. [Table 6] Comparison table of one of the preferred Chinese pinyin embodiments of the present invention initial consonant b glass p slope m touch f buddha d got ㄉ t special n ne ㄋ l Leㄌ g brother k branch h drink j base q bully c hope zv know xv 萩ㄔ sv poem v 日ㄖ z capital x female s thinking medial vowel Open your mouth to call (none) Qiqihu i clothes one Close the mouth and call w ㄨ Pinch lips and call out y 肂 ㄩ single vowel vowel r r ah ㄚ ir wr frog ㄨㄚ o o Oh ㄛ wo nest ㄨㄛ u u goose ㄜ e e eh ㄝ ie Yeah 一ㄝ ye about ㄩㄝ diphthong vowels ai ai mourn ㄞ wai crooked ㄨㄞ ae ae eh ㄟ wae 伟 ㄨㄟ au au boil ㄠ iau waist 一ㄠ ao ao 欧ㄡ iao 殮一ㄡ nasal vowels am am 安ㄢ iam smoke a ㄢ wam bend ㄨㄢ yam injustice ㄩㄢ an an en ㄣ ian because of one ㄣ wan Wen ㄨㄣ yan dizzy ㄩㄣ arm arm ANG ㄤ iarm 杨一ㄤ warm Wow ㄨㄤ arn arn heng ㄥ iarn 英一ㄥ warn von ㄨㄥ yarn 鍍 ㄩㄥ tone 1, 2, 3, 4 sounds

Although the present invention has been disclosed using the above preferred embodiments, they are not intended to limit the present invention. Anyone skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention. Therefore, the present invention is The scope of protection shall be determined by the appended patent application scope.

102:Human side 104:Natural language 106: Systematization of the axioms of the pinyin structure and form of language symbols 108:System 110:Application 112:Artificial intelligence 114:PC version 160~190: steps 202: Language definition 204:Academic Integration 206: Clock 208: Cognitive stage sequence 210:Cognitive Framework 212:Chinese phonetic sequence 302: Semiotics of familiarity 304: Correct the semiotics part 402: Phenomenon of natural laws of Pinyin structure 404: Pinyin structure is artificially processed 502:Formal scope 504: Axiomatic range 506: Scope of formal axiom systematization 610:Natural Language Processing 612:Human side 614:PC 620: One/multi-language symbol pool 622: Human side results of natural language processing 624: Computer-side results of natural language processing

Figure 1 is a schematic diagram of the relationship between a preferred natural language processing method and its system and applications of the present invention; Figure 1A is a schematic flow chart of a preferred natural language processing method of the present invention; Figure 2 shows the development intention of human understanding of natural language cognitive framework research; Figure 2A is a schematic diagram of comparative processing between modern linguistics and Chinese linguistics; Figure 2B is a schematic diagram comparing the relationship between time pulse and the evolution of Chinese; Figure 2C is a target diagram illustrating the layering of language cognition; Figure 2D is a schematic diagram of a cognitive framework; Figure 2E is a schematic diagram of Chinese phonetic sequence; Figure 3 is a schematic diagram of the inventive steps of a preferred Saussurean semiotics correction of the present invention; Figure 4 is a schematic diagram of one of the preferred linguistic axioms of the present invention: the segmental structure of the segment cannot be moved; Figure 5 is a schematic diagram of the systemized set theory of pinyin structure form and axiom of one of the preferred language symbols of the present invention; and Figure 6 is a relationship diagram of a preferred embodiment of the present invention.

160~190: steps

Claims (11)

A natural language processing method that implements the systematic axioms of the pinyin structure of natural language symbols, including: comparing more than one natural language through diachrony and/or synchrony, and grasping the natural language cognitive framework of human beings, in which human beings have natural language cognition framework. The order of the cognitive hierarchy of the framework is semantics (core)>grammar>speech>linguistic; modify the symbolic signifier of the natural language, where the symbolic signifier of the natural language includes the phonetic signifier and the linguistic signifier, dividing the phonetic signifier and the linguistic signifier. The signified signifier = phonetic signifier > linguistic signifier; construct symbols for segments according to the pinyin structure of the language sequence, and the unwavering rules include: formalization; customization of language symbol collection symbol element selection; customization The right to design the arrangement and combination of symbol elements in language symbol sets; and the arrangement rules of symbol elements and/or symbol element groups are consistent with the natural law characteristics of phonetic sequences; and the pinyin structure of the natural language symbols is implemented with one symbol per segment or one symbol group. The axioms are systematized into a language form; the phonetic signifier has linearity, is auditory in nature, only expands in time, and has characteristics borrowed from time, including: (a) it embodies a length, (b) this length can only be in Measured in one dimension: it is a straight line. By integrating its linear properties and phonetic sequence, a phonetic sequence pinyin structure is proposed, so that the phonetic sequence characteristics of the natural language are segmental structures, which remain unchanged. For example, the natural language processing method of claim 1, wherein the modified symbol referent of the natural language includes a newly added linguistic signifier custom collection symbol element. Such as the natural language processing method of claim 1, wherein the formalized form of the symbol pinyin structure includes: symbol component selection, wherein the symbol selection includes symbol components used in computer science; the formalized formula of the symbol pinyin structure includes: symbol components and/or Symbol component groups; combined with phonemes; and the right to design the arrangement and combination of identifiable symbol groups. For example, the natural language processing method of claim 1 further includes: making all symbol elements and/or component groups of the symbol set included in the language form self-consistent (consistent); using the language form as an axiom of the symbol set , it is mandatory to maintain the independence of each symbolic component and/or component group, requiring that the last segment must have a symbolic component and/or component group to close the entire structure, forming an independent, unique and closed structure; and making the language form conform to the segment structure. The symbol never moves or recovers. One of its characteristics is that the pinyin structure shows the phonological relationship. For example, the natural language processing method of claim 1, wherein the systematization of the natural language symbol pinyin structure form axioms includes artificial system technology methods and application of results. For example, the natural language processing method of claim 1, wherein the formal axiomatic systematization of the pinyin structure of natural language symbols includes the formalization of language symbols; Convert symbol components or audio formats. For example, the natural language processing method of claim 1, wherein the symbol components and/or component groups further include: the consistent translation of audio symbols, text and/or codes in different languages through symbols, and the homogeneous translation includes: enhanced processing of natural language through deep learning The audio symbols, text and/or codes of the language's syntagmatiques and/or rapports associatifs. For example, the natural language processing method of claim 1 further includes converting the symbol elements and/or symbol element groups of multiple natural language symbol sets into the same set, so that the multiple natural language symbol sets are equal, and the elements of the symbol elements and/or symbol element groups are homogeneous. Achieve multi-language conversion and translation. A natural language system using the natural language processing method as claimed in claim 1, including: a systematized set of language symbols, pinyin structures, formal axioms and/or an artificial system invented based on the natural laws of language pronunciation. An application using the natural language processing method as claimed in claim 1 includes its processing method application, its system application and its result application, which further includes the application of the embodiments of the present invention in the field of artificial intelligence and pan-technologies. A result application using the natural language processing method of claim 1 includes: computer-readable recording media, including: transcription code and/or audio and symbol text conversion; computer program products, including: speech recognition; voice control; translation ; Audio to text; and/or text to audio; and/or natural language processing in the field of artificial intelligence, including: audio integrated circuit design and/or artificial intelligence speech technology for natural language understanding, including: audio chips, software Hard firmware design, quantum assembly sequencing, various electronic and mechanical material design, factor engineering sequence, factor engineering method information processing.