What Does a Voice Teacher Do?
Phonetics is a branch of linguistics. The study of the sounds of human speech. Mainly study the pronunciation mechanism of language, phonetic characteristics and the law of change in speech.
Applied phonetics
discuss
- Chinese name
- Applied phonetics
- Foreign name
- phonetics
- Direction
- Study human speech sounds
- Affiliation
- A branch of linguistics
- Phonetics is a branch of linguistics. The study of the sounds of human speech. Mainly study the pronunciation mechanism of language, phonetic characteristics and the law of change in speech.
- Phonetics
Phonetics is a branch of linguistics. The study of the sounds of human speech. Mainly study the pronunciation mechanism of language, phonetic characteristics and the law of change in speech. Because its research content is related to pronunciation actions (physiological phenomena), speech characteristics (physical phenomena) and sense of hearing (psychological effects), and different human languages collectively have their own phonetic characteristics, so the study of modern phonetics needs Knowledge of natural and social sciences as the foundation.
The term phonetics is derived from the Greek s in the West, which means sound. The scope of early research was relatively extensive. In addition to the study of phonetic characteristics, it also includes the pronunciation or pinyinology of phonetics, and phonetics. The description and classification of phonetics in traditional Chinese phonology also belong to the category of phonetics. However, since the development of modern scientific phonetics, the classification has become more and more detailed, and the definition has become stricter. Phonetics refers to the study of the characteristics of phonetics itself.
Research Scope and Objects Most of the early phonetic studies were only for the needs of language teaching and interest in language research. In recent years, due to the improvement of medical equipment, people can observe the movements and functions of articulation organs, and have developed physiological phonetics. Due to the development of acoustic instruments, many linguistic phenomena that could only be heard by ears are now not only visually detectable, but also artificially synthesized, so there is acoustic phonetics. Due to the improvement of psychological testing methods, the research on thinking and auditory neurophysiology has become more and more advanced, and the laws of speech in speech control and auditory feedback have been analyzed more and more deeply. Perceptive phonetics (or psychophonetics) has also been developed and developed into nerve Phonetics. These are the three major branches from the perspective of research methods. Recently, due to the advancement of the information age and the need for man-machine dialogue, the study of speech in isolation has been unable to meet the requirements. Because human language is not a concatenation of isolated sounds, it is a series of interdependent and variable sounds. At the same time, language cannot be separated from the social environment and personal language habits. The study of speech cannot leave a specific language. Laws (including grammar, rhetoric, phonetic changes, etc.), so linguistic phonetics was also proposed.
The research objects of phonetics are traditionally limited to vowels, consonants, tones, stresses, and rhythms and phonetic changes. These are all qualitative studies. Recently, due to the advancement of analytical methods and the urgent need for speech information processing, comprehensive research and quantitative research on the characteristics of speech prosody have been strengthened.
In terms of the application of phonetics, in addition to teaching and linguistic research in recent years, it has spread to various subjects related to human language, such as speech correction, communication engineering, automatic control, and artificial intelligence. With the development of the fifth-generation computer and the exploration of artificial intelligence and human-computer dialogue, the theory and results of traditional phonetics can no longer meet the requirements, so the fifth-generation phonetics or speech engineering has recently been proposed. It integrates the knowledge in the fields of physiology, acoustics, perception and linguistics to explore the variables and invariants of human speech in order to serve the technological revolution.
International Phonetic Alphabet In order to make the recorded phonetic symbols more consistent, the International Phonetic Association developed a set of "International Phonetic Alphabet" in 1886, which was published in the Association's publication "Voice Teacher" in 1888 and has since been adopted by world linguists. During this period, several changes were made, and the most recent revision was in 1979. These phonetic symbols and some additional symbols basically meet the requirements for describing speech in various languages of the world.
The pronunciation mechanism of speech Voice is produced by the human articulation organ. After it is emitted, it becomes a sound wave. It is transmitted to the other party's ears and is received and understood by the other party to form a set of links called "verbal communication." Therefore, speech is produced by the articulation organ and received by the auditory organ. The research of these two parts belongs to the scope of physiology and psychology, and the characteristics of speech waves belong to the field of physical acoustics. As for the control and understanding of speech, it belongs to the field of linguistics.
The vocal organs are divided into three parts: the lower throat, the upper throat and the upper throat. The lower larynx runs from the trachea to the lungs. The airflow exhaled from the lungs becomes the source of speech. The throat part is mainly the glottis and vocal cords. The vocal cords are two ligaments that act as a larynx valve. They close and open to form the glottis. When the glottis is wide open, the air is smooth, the glottis is closed, and the airflow bursts to make the vocal cords vibrate periodically to produce "music sounds". The upper part of the larynx includes three areas: the pharyngeal cavity, the oral cavity, and the nasal cavity, which are mainly used to regulate speech.
The oral cavity and nasal cavity are the main organs that regulate pronunciation. The nasal cavity is basically fixed, while in the oral cavity, due to the expansion and contraction of the tongue, the lifting and sagging of the small tongue, the volume changes and different voices are produced. The oral cavity from the lip to the glottis is generally referred to as the "sound cavity" and is divided into several sections.
The auditory organ is divided into three parts: the outer ear, the middle ear and the inner ear. The outer ear has a caster, an outer ear canal, and a tympanic membrane. The ear wheel collects sound waves, sends them from the ear canal to the eardrum, and vibrates with the sound waves. The middle ear consists of a set of "listening bones" that transmit vibration to the inner ear in connection with the tympanic membrane. The main component of the inner ear is the cochlea, and the basement membrane inside it responds to changes in sound level.
When the vocal cords vibrate, a musical sound (also known as a voice) with periodic vibrations is called a voiced voice. When the glottis is opened, the airflow passing through the sound cavity is obstructed by various parts to generate irregular vibration noise, which is called unvoiced in pronunciation. The sound waves of speech travel between the speaker and the listener and reach each other's ears. The change in sound pressure acts on the auditory organ of the listener, and generates nerve impulses that are transmitted to the brain for identification and understanding. When the speaker is speaking, the voice is also transmitted to his own hearing organ and plays a monitoring role.
Vowels are also called vowels. In Mandarin, a, i, u, etc. are all vowels. Earlier people knew that vowels are unobstructed sounds in the mouth, but consonants are different. Later, using X-ray fluoroscopy, the movement of organs in various parts of the mouth when vowels were observed was observed. With the acoustic analysis, it is also known that the differences in vowel sounds are caused by different resonance values in the acoustic cavity. As for how vowels occur and how the vocal cords move, there are also instruments for analysis. So now the knowledge about vowels has been greatly enriched.
The production of vowels is, in short, caused by the closed vocal cords being impacted by the exhaled airflow, making periodic tremors, and having resonance through the adjustment of the mouth. The difference in vowel sounds is determined by the various parts of the oral cavity, including the height of the tongue, front and back, the roundness and spread of the lip, and the rise and fall of the lower jaw. The most important factor is the position of the tongue. Therefore, the previous description of vowel sounds was mainly determined by the "high point of the tongue", that is, the distance between the tip of the tongue, the surface of the tongue, or the base of the tongue and the palate, and the roundness and spread of the lips were determined as secondary factors. Recently, according to experiments, the physiological parameters of vowels are divided more, such as: the height of the tongue, the anteroposterior degree of the tongue, the arch of the tongue (concave and convexity), the width of the throat, and the height of the throat, Vertical height of lips (opening degree), width of lip width (lip spreading), lip length of front lips (round lip degree), soft palate sag (degree of nasalization), vocal cord closure degree (degree of air sound) )Wait.
Vowel tongue maps Phonologists make a triangular or quadrangular map, and mark the position of the vowels of a language before and after (mainly the reference tongue position) on the map, pointing out the characteristics of each vowel, which can help Learners practice their own pronunciation, which is also convenient for teaching. Known worldwide, there is the standard vowel chart of the British phonetician D. Jones. This picture includes only 8 vowels, 1 to 5 are unrounded vowels, and 6 to 8 are rounded vowels. It was later revised by linguists to mark more vowel positions in a map. The left of each straight line in the figure is a round lip sound, and the right is a round lip sound. Vocal sound maps similar to tongue maps can also be made based on the acoustic parameters of the vowels.
Classification of vowels The vowels can be divided into four levels of high, second highest, second lowest, and low according to the rise and fall of the tongue, such as i, e, , a and u, o, , , from high to low. The higher the tongue, the closer the mouth, and the lower the mouth, the more open. In addition, the front and back of the tongue can be divided into three positions: front, center, and back. As shown in Figure 6, the left vowels are on the left, the rear vowels are on the right, and the central vowels are in the middle inverted triangle. It is the central vowel [], also called mixed vowel.
There are at least four or five vowel phonemes in general languages, and more than a dozen. In normal speech, when the vowel is made, the tongue does not slide and is called a unit sound. In some languages, there are often two sets of vowels that are tightly opposed or long and short, and are distinguishing. The long ones are mostly tight vowels, while the short ones are mostly loose vowels. Also, the soft palate is drooping when vowels are pronounced, and the nasalized vowels are called nasal vowels. In Beijing speech, some vowels are r-colored, which are called rolling tongues (or tongues).
A compound vowel is a compound vowel when two or more vowels are combined in a syllable. In Chinese, two to three vowels are usually combined in one syllable. Two vowels are called second vowels, such as ai, ao, ia, and ua in Mandarin; three are called triple vowels, such as iao and uai in Mandarin. The strength of these vowels combined is not Equally, often one is stronger and the other weaker. Therefore, according to the position of this strong vowel in the syllable, it can be divided into pre-, post-, or middle-voice complex vowels. For example, ai, ou, etc. are the front ring, ia, uo, etc. are the rear ring, and iao, uai, etc. are the middle ring. Generally speaking, the tongue position of a vowel in a syllable is lower than that of an unvoiced vowel. If the first vowel in a compound vowel is i, u, or ü, it is called a vowel.
When the vowels of the compound vowel are connected together, the tongue position gradually slides from the previous vowel to the next vowel instead of jumping. Moreover, in Mandarin, the last vowel or the last vowel of the complex vowel is often read as "not at home", that is, the tongue position is originally higher and lower, and the latter is more central. Such as [ai] [ae], [ou] [o ].
Consonants are also called consonants. In Mandarin, such as b, p, m, f, etc. are all consonants. The pronunciation method is that, because there are obstacles in the mouth, the exhaled airflow bursts into sound or friction sound through these obstacles. The sequence of articulation actions can be divided into three stages. At the beginning, the stage where the articulation organs are placed is called resistance, and the stage that has been acting but not yet uttered is called resistance, and when the sound is emitted, it is called resistance elimination. The airflow of consonants generally comes from the lungs and passes through the glottis and the sound cavity, making sounds from obstruction. Two different sound sources that form a consonant due to the opening or closing of the glottis. The glottis is open, and the vocal cords do not tremble, becoming burst or friction noise. They are called clear consonants. The vocal cords vibrate to produce musical sounds, which are issued at the same time (or in advance) as bursts or frictions, and are called voiced consonants.
Consonants produce different timbre due to different pronunciation parts (obstructive parts) in the oral cavity. Therefore, the pronunciation parts are also an important basis for analyzing consonants. The obstacles in the oral cavity are generally composed of static organs and moving organs, mostly in the upper part, such as the upper lip, upper teeth, and palate, and more in the lower part, such as the lower lip, lower teeth, lower jaw, and tongue. Although the small tongue at the back of the soft palate is located at the upper part, it is very flexible. It can move up and down to open and close the pharyngeal passage, thereby determining whether it is an accent or a nasal sound.
The classification of consonants is mostly based on both the pronunciation method and the pronunciation part. Since the consonants of different languages in the world are different from each other, this is a detailed description. Therefore, it is impossible to develop a set of tables that includes all consonants in the world's languages. The consonants in the International Phonetic Alphabet Table have summarized most of them. Phonetics are always revised and supplemented when analyzing the speech of a particular language.
Stopped sound is also called burst sound. The moving part of the vocal organ moves closer to the static part, causing occlusion (resistance). The airflow flows out from the open glottis, opens the occlusion, and explodes (removes resistance). After removing the obstruction, the glottis immediately closes, and the vocal cords vibrate to connect the vowels without aspirating sounds. After removing the obstruction, the glottis is still open for a short period of time to allow the airflow to continue to flow, and then the vowels are aspirated. The former as mandarin
- , [T], [k], the latter being [p '], [t'], [k '].
The fricative sound of the two parts of the mouth is close to the organs, forming a gap. When the exhaled air passes through, it produces noise and becomes friction sounds, such as [f] and [s] in Mandarin and [v] and [] in English.
The vocal fricative action is first occluded and then rubbed. At some point, the two parts of the organs move closer to each other, such as a stopper, which hinders the slight relaxation when the obstacle is removed, and allows the airflow to pass through such as a fricative. However, the difference between it and fricative is that it first occludes and then relaxes when it becomes blocked; the difference between it and block is that it does not immediately release the obstacle and leaves a slight gap when removing it. The non-aspirating fricative is immediately followed by vowels, such as [], [] in Mandarin. The fricative is sometimes treated as a consonant in Western languages. Aspirator fricative is that the lungs still have airflow after removing the obstruction, and then obstruct and then connect the vowels, such as [ '], ['] in Mandarin.
Nasal sounds. The two parts of the mouth move closer and closer to each other, forming a blockage. The soft palate sags and opens the nasal passages. The vocal cords tremble, and the airflow divides in two ways. Different resonance sounds, such as [m], [n] in Mandarin, and [] in Shanghai.
In the side-tone mouth, the tip of the tongue or the back of the tongue abuts the upper back of the teeth, gums or hard palate, leaving a gap on both sides (sometimes one side) of the tongue. The soft palate rises, the vocal cords vibrate and pronounce, the airflow flows out of the teeth, and then releases, as in [1] in Mandarin. Some side sounds have friction at the same time, which is called side fricative, such as [] in Guangdong Taishan dialect.
Trills and flashes The moving parts of the lower lip, tongue tip, or small tongue come into contact with the quiet parts of the upper lip, gums, palate, etc., and vibrate, making the vocal cords tremble. A vibrato is called a vibrato, and a flicker is called when it touches once. The former as Russian
- The latter is like [] in English.
Half vowels are also called tones. Some high vowels, such as [u], [y], have a small distance between the moving part (high point of the tongue) and the quiet part during pronunciation, causing slight slit friction. Such as [j] or [w] in Mandarin.
Most of the above sounds come from the lung airflow. In addition, there are several consonants. When pronounced, the airflow does not come from the lungs, but the air pressure in the mouth is driven by the movement of the throat and the movement of the tongue, which becomes the power to open and occlude. There are three types of sounds driven by the pressure in the mouth:
Squeeze the throat: Obstructions such as stop sounds, glottis and obstructions on the glottis are closed during articulation, and the throat moves upwards, pushing the air in the closed mouth to form pressure. The occlusion was released, and the airflow outside the glottis escaped. Such as [p '], [t'], [k '] in American Indian.
Constriction sound: the glottis is tightly closed, the obstruction in the mouth is the same as the stop sound, and the throat is lowered, so that the air in the closed mouth is decompressed and the sound is removed. Such as [], [], [] in African and American Indian languages.
Mouth sound: The root of the tongue rises against the soft palate and blocks the mouth and glottis. At this time, the closed lips or the tip of the tongue and the upper gingival crest are quickly removed, causing the air pressure in the mouth to become low and inward at the obstruction. Pop sounds (similar to imitating the click of a horseshoe with the tip of a tongue). It is pronounced in African Zulu, such as [].
The characteristics of consonants in modern phonetics are becoming more and more meticulous. In addition to the pronunciation part and the pronunciation method, it also analyzes the "sound type", which means that in addition to the processes on the glottis, such as voiced and aspirated, voiced consonants in some languages, Different vocal patterns due to different vocal cord tremors also play a role in distinguishing meaning. Such as low voice, also known as breath, vocal cords are not closed when the vocal cords flutter; larynx is also known as creaks, part of the vocal cords vibrate normally and part of them move slowly; in addition, there are ear speeches, the vocal cords do not vibrate, and the airflow is constricted by the throat Noise is caused by friction.
Prosodic features are also known as supersegmental features, and are also traditionally called suprasegmental. This is in addition to the vowel, consonant and other timbre characteristics, including pitch, intensity, sound length and their interrelationships. They are expressed in tones as tone, intonation, stress, rhythm.
In addition to vowels and consonants, tone is also the main component of speech. Tone is expressed by the change of pitch. The vocal tremor is controlled to be fast and slow, making the pitch higher or lower. In a natural language, the range of pitch fluctuations is roughly stable.
The languages of the world can be divided into non-tonal languages and tonal languages. The tones of non-tonal languages are responsible for the tone function, while the tones in the tonal language (represented in the tones) play the same role as the consonants and vowels. For example, "mama" m ," m "má," horse "m, and" scolding "mà in Mandarin have the same pronunciation, but they have different meanings due to different tones. There are two types of pitch modes for different tonal languages. One is the scale type, which uses only flat tones that are divided into levels without the use of ascending and descending to distinguish the tone types. Tonality (Languages generally have flat tones). The tone range is wide, the difference between the tone values is larger, and the narrower is smaller. Different people have different tonal ranges, and the same person's tonal range varies in different tones. Therefore, the pitch of a voice is a relative level rather than an absolute value.
The single-word tone value of Putonghua was tested by Liu Fu and Zhao Yuan experimentally in the early days. Recently, it has been tested by acoustic instruments, and the tone value has been much more accurate.
Tone symbols In the past, linguists often used high and low curves or tabs to describe tones. The standard notation, or standard letter, published by Zhao Yuanren in 1930 is now used by most linguists. This is a five-degree standard tone symbol. The average relative pitch of the tone is divided into low, half low, medium, half high, and high five degrees, which are represented by 1, 2, 3, 4, and 5, respectively. The tone number uses a vertical line as a comparison line, and a horizontal line or a zigzag line on the left indicates the tone level and arch (shape) of the tone. For example, the four tones of Mandarin: if the vowels in a language are different in length, the short-tone symbol can shorten the horizontal line by half, etc. The soft-sounding symbols in Mandarin use dots instead of horizontal lines. Such as. When there is a tone change, put the horizontal line or the dot to the right of the vertical line. For example, when the original high-level tone becomes high-descending, write, and when the original high-level tone becomes medium-low, write.
When two or more syllables are spoken together, the original single word tone often changes. For example, in spoken Mandarin Chinese, the two upper voices are continuous, and the upper and lower voices become flat. For example, "good rice" (good rice 214) = "mm"; "minimum" (minimum 214 = "riding"). Three consecutive sounds are usually divided into single and double cells or double single cells according to the grammatical structure. The former such as "old factory director", "factory" becomes Yangping (factory), and "old" becomes half-top (old), the latter such as "factory's room", "factory" becomes Yangping (factory), and " "Long" is the over-shaping (long) of the short to medium. Generally speaking, two-character continuous tone has a basic continuous tone pattern. Conjunctive tones with three or more characters are a combination of one-character and two-character consonant except for a few individual rules.
The tone change of intonation sentences is called intonation. Non-tonal languages (such as English, German, etc.) are mainly based on intonation changes, which vary according to the tone and mood. For example, in English, the ending of a declarative sentence generally decreases, while the ending of a question sentence rises. Tone languages (such as Mandarin Chinese) use single-character tones and two-character consonants as the basic unit. In a normal-speed sentence, even if there is a different tone, these basic tones can still maintain their original camber (rising zigzag), but in the statement sentence, the tone level at the end of the sentence is lower than the camber; In the middle of the sentence, the tone level is raised instead of the arch. So the absolute adjustment value can be changed, but the general adjustment type is unchanged. Only in accelerated statements can there be major changes in tuning.
The intensity of the stress speech is expressed by the strength of the sounding airflow. The weight of a word in a sentence indicates whether it is strengthened or not. In non-tonal languages, the syllable stress of a word generally has the function of distinguishing the meaning of a word, which is called word stress. For example, 'object (object, noun), ob'ject (objection, verb) in English. In tonal language, stress generally represents the mood, but does not distinguish the meaning of the word. In a sentence, stress often falls on a focused meaning group (combination of words), and sometimes it can change the meaning of the sentence, which is called logical stress.
A syllable is divided into several words. A word contains one or several phonetic units, called a syllable, which is composed of more than one phoneme. For example, the word communication in English has five syllables, and the word come has only one syllable; while a single word in Chinese is a syllable, and a word contains one to five syllables, but most of them are two syllables. The definition of a syllable is difficult in one or two sentences. In the past, linguists have various definitions of syllables, such as: "chest beat theory", a syllable has a chest beat; "speak peak theory", each syllable has only one sound peak; The strongest vowel or consonant is at the core; "muscle tension theory", muscle tension and weakening become one syllable at a time when speaking. But these claims are not comprehensive. Because in actual continuous speech, the result of dividing syllables according to these statements may be inconsistent. Therefore, from the perspective of phonetic physiology alone, the syllable division of the pronunciation angle and the auditory angle will be different. If the analysis is based on the phonetic description or phoneme classification, there will be more differences.
However, from the perspective of general acoustic phenomena, it can be considered that: at least one vowel (sometimes a consonant) in a syllable plays a central role; there is a weaker boundary between syllables and syllables in terms of sound intensity point. However, if the two syllables are connected by vowels without a consonant to separate them, the boundary is not obvious, such as "Xi'an" and "xian" in Mandarin are spelled xian.
Rhythm generally refers to the length of each syllable in a sentence. The rhythm of the sentence is composed of the intentional groups in the sentence, which is equivalent to the beat in music. Each "beat" of language includes one to three syllables, with two syllables being the most common. The length of the beat often varies depending on the speed of the entire sentence and the priority of the meaning group, not as strict as the music beat. The following is an analysis of the acoustic length of a Mandarin sentence: The bottom horizontal line in the sentence represents a meaning group and becomes a beat. The number in the sentence is the length of the meaning group, and the unit is second. It can be seen from this sentence that the length of the two-word group is basically 0.3 seconds, and the three-word groups are 0.5 seconds, while the lightly read words only occupy 0.1 seconds. These two trigrams are both focused words, so they are relatively long compared to the two-characters.
Changes in speech One sound is connected to another sound. Due to the naturalness of the pronunciation action or other reasons, it often affects each other and changes the original pronunciation. This is called assimilation. The front assimilation affects the back assimilation and vice versa. Assimilate like a duck in Mandarin [ji tan] [ji dan]. The reverse assimilation is like "South Gate" [nanmen] [nammen], and the assimilation is assimilated as "Sanpa" [sanpa] [samba].
When two similar syllables are connected, in order to avoid repetition or monotony, one of the syllables becomes the other, which is called alienation. For example, if the two previous sounds are read in succession, the previous one becomes yangping, that is, a type of alienation, and there is also weakening, such as Beijing dialect "Cotton" [minxua] [minxu ] ("" becomes lighter) Read while a ). The soft tone of the tone is also a weakening phenomenon. In addition, there are sound changes such as increasing, reducing, leaking, transposing, replacing, and converting. Many of these phonological changes are due to reasons for saving articulation or rhythm, while others are the result of historical evolution or language errors.
Phonemes The smallest unit of speech commonly used in each language group, such as vowels and consonants, has its inherent sound quality. These sound quality have a certain range of variation between different speakers or between different speech environments. For example, an [e] can be read open or closed, and an [l] can be read brighter or darker. Although there is a change, the distinguishing function is the same as the original normal sound. Each cluster is called a phoneme, and each changed phoneme is called a phoneme variant. Experiments have shown that there is no clear boundary between two adjacent phonemes, and the variants of the two phonemes have overlapping phenomena.
Distinguishing features In the early 1950s, western phoneticians, in view of the fact that the specifications adopted in previous speech analysis, could not express the smallest difference in speech, thus developed a speech analysis theory of distinguishing features. The pioneers are R. Jacobson and others. They believe that the speech of all languages can be divided into a number of minimum pairs according to their physiological and acoustic characteristics. Such as: vowel / non-vowel, consonant / non-consonant, nasal / accent, agglomeration / dispersion, suddenness / continuity, roughness / softness, Emergency stop / non-emergency stop, dullness / unvoicing, tension / relaxation, dullness / sharpness, (11) downward trend / flat trend, (12) upward trend / Flatness. Later, some people thought that these projects were not comprehensive enough and gradually revised them. N. Chomsky and M. Harley set a few more. Among them, except for the same as the above, according to the pronunciation part: before the tongue / non-lingual, tongue top / non-lingual top, long seam / non-long seam, back tongue / non-back tongue, tongue surface Raise / not raise the tongue surface, lower the tongue surface / not lower the tongue surface; They also divide speech into two categories: resonance / obstruction. Some people in China have added several items based on the characteristics of Chinese phonetics, such as the opening / closing of vowels, Qi / pinch, Hong / Shui, consonant convection (non-aspirated / aspirated), and tonal up / down, Flat / Curved, High / Low etc.
The smallest difference in speech may not be summarized by the dichotomous feature in some cases. For example, there are half-height and half-low between the tongue height; half-nasal sound and different degrees of nasalization between accent and nasal sound. Therefore, some people have proposed different polynomial features and other theories. In actual language, from one phoneme to another phoneme, there can be countless phoneme variants in the middle, that is, it can be divided into countless phoneme units, and each pair of adjacent phoneme units form oppositions. Therefore, from the perspective of dialectical relationship, each pair of three-dimensional pairs can be divided into countless features, and each of these features is an opposite relationship, which can be called N-even relationship. Using this concept to analyze speech can express the interrelationship of speech.