CN1307273A - Intelligent phonetic input system and method - Google Patents
Intelligent phonetic input system and method Download PDFInfo
- Publication number
- CN1307273A CN1307273A CN 00111631 CN00111631A CN1307273A CN 1307273 A CN1307273 A CN 1307273A CN 00111631 CN00111631 CN 00111631 CN 00111631 A CN00111631 A CN 00111631A CN 1307273 A CN1307273 A CN 1307273A
- Authority
- CN
- China
- Prior art keywords
- pronunciation
- group
- phrase
- input
- words
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims description 77
- 150000001875 compounds Chemical class 0.000 claims description 30
- 230000008676 import Effects 0.000 claims description 14
- 230000008859 change Effects 0.000 claims description 6
- 230000008878 coupling Effects 0.000 claims description 2
- 238000010168 coupling process Methods 0.000 claims description 2
- 238000005859 coupling reaction Methods 0.000 claims description 2
- 230000008569 process Effects 0.000 description 8
- 210000004556 brain Anatomy 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000002490 cerebral effect Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000006837 decompression Effects 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 239000000446 fuel Substances 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000005096 rolling process Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Landscapes
- Document Processing Apparatus (AREA)
Abstract
Symbols the operator inputs is first divided into several pronunciations, and the word information with the same initial consonant combination is then selected from data base based on the input initial consonant combination, where one word information consists of Chinese character corresponding to each initial consonant. The pronunciation of each word information is composed with the input pronunciation, and while matching the word information is displayed for the operator to select.
Description
The present invention relates to a kind of phonetic input system and correlation technique thereof, particularly a kind of intelligent phonetic input system and the method that can select words by the input initial consonant automatically.
Progress along with electronics technology is maked rapid progress because all trades and professions all towards electronization, robotization effort, are added adding fuel to the flames of recent network use, causes PC to be popularized in a large number.Yet,,, a lot of difficulties are arranged still by keyboard inputting word information in computer for most people.Generally speaking, because Chinese phonetic notation rule, be the Chinese pinyin mode that Chinese are familiar with, when carrying out words to computer or associated electrical disposal system when importing, phonetic input method is the most frequently used input mode.For example, when input " China " two Chinese characters, can on keyboard, key in " ㄓ ㄨ ㄥ ㄍ ㄨ ㄛ Pie ", or when wanting input " computer ", can on keyboard, key in " ㄉ one ㄢ
ㄋ ㄠ
", can select correct Chinese character by the word string that is occurred on the computer screen then with above-mentioned pronunciation.Yet, use phonetic input method to carry out loading routine, still have many shortcomings.Wherein, in Chinese phonetic notation rule, the pronunciation of a Chinese character is made up of initial consonant, simple or compound vowel of a Chinese syllable and tone, thereby when carrying out loading routine, often will press on keyboard for several times, but just key goes out required Chinese character.And,, thereby after keying in phonetic symbol, need carry out word selection again, to obtain required words because several different Chinese characters often can be represented in same pronunciation.Thus, the also feasible speed of using phonetic input method to carry out loading routine is slower.
In addition, because Chinese phonetic input method needs the operator to import correct Chinese phonetic notating symbol, thereby, when being difficult for grasping, tend to cause input error when the pronunciation that will import words.For example cerebral " ㄌ and ㄖ ", " ㄓ and ㄗ " and " and ㄙ ", simple or compound vowel of a Chinese syllable " ㄛ and ㄜ ", " ㄢ and ㄤ " reach " ㄣ and ㄥ ", or Chinese tone one, two, three, the four tones of standard Chinese pronunciation and softly or the like, all obscured by the operator easily, and cause identification error thus and need re-enter.Thereby, for the operator that Chinese phonetic notation rule is known in right and wrong Changshu not, can feel to use traditional Chinese phonetic input method and inconvenient.
Purpose of the present invention is providing a kind of Chinese phonetic input method that omits tone, simple or compound vowel of a Chinese syllable.
Intelligent phonetic input method of the present invention comprises the following steps: to import at least one phonetic symbol at least; This phonetic symbol is divided into one group of pronunciation at least; Organize the initial consonant of this pronunciation according to each, and determine an initial consonant combination; Carry out search program, from database, to select to comprise the word information of this initial consonant combination; Relatively whether the pronunciation of this word information comprises this and respectively organizes pronunciation; And show and comprise this this word information of respectively organizing pronunciation for you to choose.
In said method, after this at least one phonetic symbol is divided at least one group of pronunciation, should convert at least one group of scale-of-two pronunciation coding to by at least one group of pronunciation.
Above-mentioned database comprises: a group index district, divide according to this at least one group of pronunciation; The secondary index district, corresponding with this one-level index area, and according to this initial consonant combination is divided; Individual character and phrase block of information, corresponding with this secondary index district, be used to provide the word information that conforms to this initial consonant combination; Reach individual character or phrase information and form the district, be used to deposit the pronunciation coding of this word information.
In said method, relatively the step of the pronunciation of this word information and this at least one group of pronunciation also comprises the following steps: to determine simplifying procedures of this at least one group of pronunciation, to confirm the simplification situation of this each its initial consonant of group pronunciation, simple or compound vowel of a Chinese syllable and tone; Simplify procedures according to this, each group pronunciation of this word information is simplified, and obtain the simplification pronunciation of this word information; And relatively whether the simplification pronunciation of this word information comprises this at least one group of pronunciation.
The present invention also provides a kind of program of using the intelligent phonetic input method to carry out self-built phrase storehouse, and this program comprises the following step at least: the input needs increase the phonetic symbol of phrase; This phonetic symbol is divided at least one pronunciation; According to the order of this at least one pronunciation, provide prepare word for affirmation successively, and produce one group of phrase words; Determine the position that this group phrase words is deposited in this database; And should organize the phrase words is stored in this database.Before storing this group phrase words, also carry out one and check step, to check whether the existing words and phrases identical with this group phrase words exist this deposit position.After storing this group phrase words, also carry out a change program, with start address and the word quantity of changing index area in this database.
The present invention also provides another kind of intelligent phonetic input method, comprises the following steps: to import at least one group of phonetic symbol at least; This phonetic symbol is divided at least one group of pronunciation; Change this each group pronunciation and be the input pronunciation coding, wherein each is organized this input pronunciation coding and is the scale-of-two pronunciation coding; Organize the initial consonant of this input pronunciation coding according to each, form the initial consonant combination; According to this initial consonant combination, carry out search program; Determine each to organize simplifying procedures of this input pronunciation coding, to confirm the simplification situation of this its initial consonant of input pronunciation coding, simple or compound vowel of a Chinese syllable and tone; Simplify procedures according to this, each group retrieval pronunciation coding of this word information is simplified; Relatively whether the retrieval pronunciation coding of this word information is organized this input pronunciation coding and is consistent with each; And show and organize this word information that this input pronunciation coding is consistent with each,
Wherein, above-mentioned search program comprises: carry out first search program, make up the same number of word information to mark off with this initial consonant; Carry out second search program, to mark off the word information that is consistent with this initial consonant combination; And from storing the address that this word information is handled, read this word information, wherein each is organized this word information and all comprises this initial consonant combination.
Said method also comprises the program in a self-built phrase storehouse, and this program comprises the following step at least: the input needs increase the phonetic symbol of phrase; This phonetic symbol is divided at least one pronunciation; According to the order of this at least one pronunciation, provide prepare word to confirm successively, to produce one group of phrase words for the operator; Determine the position that this group phrase words is deposited in this database; And should organize the phrase words is stored in this database.The present invention also provides a kind of intelligent phonetic input system, it is characterized in that, this system comprises at least: input media is used to provide the operator to carry out the phonetic notation loading routine; The pronunciation detachment device responds this input media, is used for the input phonetic notation is divided at least one group of pronunciation, and distinguishes at least one group of initial consonant that this at least one group of pronunciation comprises; Searching system is connected and responds this with a database and read detachment device, so that read at least one words corresponding to this at least one group of initial consonant from this database; Display device responds this searching system, is used to show this at least one words by retrieval, recognizes and selects so that this operator to be provided.
Said system also comprises a scale-of-two transcoder, responds this pronunciation detachment device, and is coupled in this searching system, is used for this at least one group of pronunciation is converted to binary pronunciation coding.
Said system comprises that also one simplifies comparison system, respond this scale-of-two transcoder and this searching system, and be coupled, be used to determine the simplified condition of this pronunciation coding with this display device, and the words that this searching system read is simplified with this simplified condition, compared with this pronunciation coding again.Said system also comprises a words enlargement module, responds this simplification comparison system, and is coupled with this database, is used to carry out self-built phrase program, to increase the phrase data in the database.Intelligent phonetic input mode provided by the present invention has a lot of advantages.Wherein,, be used as retrieving the foundation of comparison, can omit tone, simple or compound vowel of a Chinese syllable and reach the purpose of quick input by the initial consonant part of using the Chinese to decide pronunciation.For example, no matter key in " ㄓ ㄨ ㄥ ㄍ ㄨ ㄛ ", " ㄓ ㄍ ㄨ ㄛ " " ㄓ ㄨ ㄥ ㄍ " or " ㄓ ㄍ ", the option of " China " these two Chinese characters can appear on screen all.Similarly, as long as input " ㄉ ㄋ " just can be imported " computer ".Thus, can significantly reduce input simple or compound vowel of a Chinese syllable and spent time of tone, and improve input efficiency effectively.Simultaneously, also can reduce owing to be unfamiliar with the time waste that the phonetic notation rule is caused.In addition, the intelligent phonetic input method also has phrase, the short sentence function of coupling automatically.Wherein, in the pattern of input phrase, short sentence, only need key in a few phonetic notation, just can import one section long phrase or short sentence.For example, as long as key in " ㄓ ", just can import the short sentence of " the water sky of the Yellow River comes up ".Because being respectively this section literal, " ", " " and " ㄓ " start triliteral initial consonant.Thus, obviously can significantly shorten the time of input in batch.And, because the intelligent phonetic input also has the function of self-built phrase, therefore can provide operator oneself to build phrase, and reach the expansion dictionary, the convenient function of input in the future.
The invention will be further described by the following examples and in conjunction with the accompanying drawings.
Fig. 1 shows the mode of carrying out 16 binary codings and storage according to Chinese-character pronunciation provided by the present invention; Fig. 2 shows according to the Database Systems that are used to carry out the intelligent phonetic input method provided by the present invention; Fig. 3 shows the main process flow diagram that carries out the intelligent phonetic input method according to the present invention; Fig. 4 shows the application example that carries out the intelligent phonetic input method according to the present invention; Fig. 5 A-B shows when carrying out the intelligent phonetic input method according to the present invention, shown result on screen; Fig. 6 shows the related procedure of carrying out the input of short sentence or phrase according to the present invention; Fig. 7 shows the related application example that carries out the input of short sentence or phrase according to the present invention; Fig. 8 shows the related procedure of carrying out the next self-built phrase of intelligent phonetic input method according to the present invention; When Fig. 9 A-D shows that carrying out the intelligent phonetic input method according to the present invention comes self-built phrase, shown result on screen; And Figure 10 shows the arrangement plan according to intelligent phonetic input system provided by the present invention.
The invention provides a kind of new intelligent input method and system.By using the initial consonant part of Chinese-character pronunciation, be used as retrieving the foundation of contrast, can be under the situation of omitting tone, simple or compound vowel of a Chinese syllable, and reach the purpose of quick input.Relevant of the present invention be described in detail as follows described.Please refer to Fig. 1, this figure is used to illustrate the new method of encoding and storing that Chinese character is read provided by the present invention.In conventional art, need to use 4-8 byte (Byte) to store a Chinese-character pronunciation, coded system used in the present invention has only been used 2 bytes (16), just can represent the pronunciation of a Chinese character.Generally speaking, phonetic symbol can be distinguished into " the ㄅ ㄆ ㄇ ㄈ ㄉ ㄊ ㄋ ㄌ ㄍ ㄐ ㄑ ㄒ ㄓ ㄔ ㄖ ㄗ ㄘ ㄙ " that uses as initial consonant, and " the ㄨ ㄩ ㄚ ㄛ ㄜ ㄝ ㄞ ㄟ ㄠ ㄡ ㄢ ㄣ ㄤ ㄥ ㄦ " that use as simple or compound vowel of a Chinese syllable.In addition, by use " one (high and level tone), two (
), three (
), the four tones of standard Chinese pronunciation (
), () softly ", can represent the tone of Chinese-character pronunciation.Wherein, owing to the pronunciation of a Chinese character is made up of initial consonant, simple or compound vowel of a Chinese syllable or combination simple or compound vowel of a Chinese syllable (for example ㄍ ㄨ ㄛ ' (state)) and tone.Therefore, as shown in Figure 1, in the present invention, use the 0-5 position to represent initial consonant; Use the 6-12 position to represent simple or compound vowel of a Chinese syllable; And use the 13-15 position to represent tone.It should be noted that Chinese character, for example " ㄞ to no initial consonant
(love) " word, in the initial consonant part, also give coding corresponding to " no initial consonant ", do not possess the situation of initial consonant to show this Chinese character.For example show among Fig. 1) " 000001 " coding, promptly being used for representing the state of this Chinese character initial consonant part is no initial consonant.Relatively, for the Chinese character that does not have simple or compound vowel of a Chinese syllable, for example "
(love) " word, also " 0000001 " with the no simple or compound vowel of a Chinese syllable of representative is used as its coding in the simple or compound vowel of a Chinese syllable part.In addition, it should be noted that when the situation with combination simple or compound vowel of a Chinese syllable takes place, for example " ㄨ ㄜ
(I) " word, then can constitute its coding of each simple or compound vowel of a Chinese syllable of combination simple or compound vowel of a Chinese syllable, made up and constituted its corresponding combination simple or compound vowel of a Chinese syllable coding.For example above-mentioned " ㄨ ㄜ
(I) " combination simple or compound vowel of a Chinese syllable " ㄨ ㄜ " in the word can use the coding " 01000000 " of representative " ㄨ " to represent with the coding " 0100011 " that combines of the coding " 0000011 " of representative " ㄛ ".
Thus, by the correlative coding of initial consonant among Fig. 1 (0-5 position), simple or compound vowel of a Chinese syllable (6-12 position) and tone (13-15), can express selected Chinese character.For example, when input " computer " two words, because its pronunciation is " ㄅ-ㄢ
" and " ㄋ ㄠ
", therefore can be by learning " ㄉ one ㄢ among Fig. 1
" can to show be 000110 to its initial consonant " ㄉ "; And " ㄢ " is the combination simple or compound vowel of a Chinese syllable, is made up of jointly " one " and " ㄢ ", and its coding can be expressed as 0011010; In addition, its tone is
So, can be expressed as 101.Thus, can obtain whole " ㄉ-ㄢ
" word two into be encoded to 0001100011010101, and can be in computer system (0 * 18D5) is represented with the coding that occupies 2 bytes.Similarly, above-mentioned " ㄋ ㄠ
" also can 0010000001000100 two into coding represent, and in computer system, be converted into and use the coding (0 * 2044) of two bytes to represent.Thus, the coded system that the application of the invention provided, saving storage element that can be a large amount of.
Then, please refer to Fig. 2, the figure illustrates among the present invention, during the operative intelligence phonetic input method, carry out the required database of access Chinese character 100.This database 100 can be formed district 108 and be divided into one-level index area 102, two utmost point index areas 104, individual character and phrase block of information 106 with individual character or phrase information according to the different levels of retrieving.Wherein, one-level index area 102 mainly is to distinguish according to the number of words of input words.For example, individual character, two words groups, three words groups can be divided in the words of input ...After the number of words of confirming the input words, then can enter secondary index district 104.104 in this secondary index district is divided according to different initial consonant combinations.For example, when in one-level index area 102, after the decision input words number of words, just can in secondary index district 104, retrieve the identical initial consonant combination phrase of number of words.That is work as the words imported via the one-level index, determine that the input words has n Chinese character after, can go out the phrase of forming by n initial consonant by two utmost point indexed searchs.Then,, can enter individual character and phrase block of information 106,, provide the phrase information that conforms to the initial consonant combination with according to different initial consonant combinations according to the division of secondary index.Then, enter individual character or phrase information again and form district 108, to extract the voice coding of above-mentioned phrase information.Wherein, individual character or phrase information are formed district 108, are actual position of depositing individual character or phrase data in the database 100.
It should be noted that in the practical operation of computer, often, pronunciation coding (has been occupied 2 via encrypting and compression process
*The N byte space, wherein N is the number of words of this words) be converted to the Chinese character encrypted code, taken up space with further saving, and be stored in the computer system.Therefore selecting certain group individual character or phrase, and when extracting its pronunciation coding program, can select its corresponding Chinese character encrypted code earlier, and, be reduced to the pronunciation sign indicating number via deciphering and decompression process.In one embodiment, be " during computer two words, in one-level index area 102, it can be divided into phrase when importing words with 2 Chinese characters.Then, in secondary index district 104, then can the initial group cooperation be the differentiation foundation, and mark off the phrase district that initial group is combined into " ㄉ " and " ㄠ ".Then, in individual character and phrase block of information 106, can select corresponding phrase information according to above-mentioned initial consonant combination " ㄉ " " ㄠ ", for example " brain ", " computer ", " then " or the like ...Then, form in the district 108 in individual character or phrase information, again will with above-mentioned phrase information (" brain ", " computer ", " then " ...) corresponding Chinese character encrypted code, and can be extracted via decrypting process.Then, please refer to Fig. 3, this figure is shown to be the main flow process of intelligent phonetic input method of the present invention.Wherein, at first carry out step 200, the input phonetic symbol.That is, use the interface, in the phonetic notation input operation system of the Chinese character words and phrases that the operator will be imported via the operator.Then, carry out step 205,, divide into many group pronunciations the phonetic symbol of being imported.Carry out step 210 again, determine that the initial consonant of each group pronunciation so can obtain a plurality of initial consonants corresponding to many group pronunciations in these many group pronunciations.Then, carry out step 215, from database, select to have comprised the phrase information of these a plurality of initial consonants.Wherein, this phrase information is made up of a plurality of Chinese character that corresponds respectively to a plurality of initial consonants.Carry out step 220 again,, compare through many groups pronunciation of distinguishing,, whether comprised many groups pronunciation fully to determine the Chinese-character pronunciation of all phrase information with above-mentioned with the Chinese-character pronunciation of all phrase information.Then, carry out step 225, show the phrase information that is consistent with above-mentioned many group pronunciations, select for the operator.What specify is, because whole intelligent phonetic input method in examining the process of contrast, is only used the initial consonant part of each Chinese-character pronunciation.Therefore, when the operator does not very determine the Chinese phonetic notation of certain Chinese character, only need import the initial consonant of this Chinese character, just can obtain having all Chinese characters of this initial consonant.Thus, even the operator has omitted the simple or compound vowel of a Chinese syllable and the tone of each Chinese-character pronunciation, but by intelligent phonetic input method provided by the present invention, still can demonstrate the possible phrase that is made of these initial consonants when the input phonetic notation,, the operator selects so that being provided.
Note the detailed process of input method please refer to Fig. 4 as for intelligence.The figure illustrates its application example of quick input mode of above-mentioned omission tone and simple or compound vowel of a Chinese syllable.For example, when two words of operator's desire input computer, can be via the inputting interface of keyboard, the phonetic symbol (step 300) of input as " ㄉ ㄋ ㄠ ".Then, by the phonetic notation of input is distinguished, and determine the phonetic symbol of input, can be divided into many group pronunciations (step 305).When for example importing phonetic notation, it can be divided into " ㄉ " and " ㄋ ㄠ " two groups of pronunciations for " ㄉ ㄋ ㄠ ".Then, will distinguish good pronunciation, according to the present invention in new encode Chinese characters for computer mode, convert one group 16 scale-of-two pronunciation coding (step 310) to.Then, again according to the consonant information in the generation scale-of-two pronunciation coding, the Database Systems among Fig. 2 100 are carried out search program.To read use " ㄉ " and " ㄋ " phrase information, storage address in Database Systems and quantity (step 315) as initial consonant.Subsequently, according to the phonetic notation feature of operator's input, determine the simplification situation of each group input pronunciation, and its pronunciation of phrase information that retrieves in the step 315 is carried out identical simplify procedures (step 320).In this example, because phonetic symbol that the operator imported is " ㄉ ㄋ ㄠ ", so being characterized as of its input: first pronunciation " ㄉ " has omitted simple or compound vowel of a Chinese syllable and tone, and second pronunciation " ㄋ ㄠ " then only omitted tone.Therefore, can carry out identical simplifying procedures, removing the simple or compound vowel of a Chinese syllable and the tone of first pronunciation in the phrase information, and remove the tone of second pronunciation the phrase information that obtains in the step 315.Then, the pronunciation sign indicating number of compare operation person input with from database by retrieval with simplify procedures after phrase information pronunciation sign indicating number whether conform to (step 325).In this example, can be with the phrase information pronunciation sign indicating number after simplifying procedures, the pronunciation sign indicating number of being imported with the operator (so " ㄉ ㄋ ㄠ " in the example) compares.When the pronunciation sign indicating number of input with by database in the pronunciation sign indicating number read when identical, the phrase information that then will have this pronunciation sign indicating number is shown in the candidate regions on the screen, for operator's selection (step 320).Opposite, when the pronunciation sign indicating number of being imported, when inequality, then skip this phrase information, and be not presented on the screen with the pronunciation sign indicating number of being simplified by phrase information in the database.Then, judge whether all phrase information have disposed (step 335), finish, then skip back to step 320 again, again the next phrase information that is retrieved in the pronunciation sign indicating number of compare operation person's input and the step 315 if still be untreated.
Therefore when the operator imports " ㄉ ㄋ ㄠ ", can occur on the screen as the result as shown among the A5 figure.Wherein, when intelligent phonetic input system is retrieved word selection according to above-mentioned flow process, can 2. quarrel and fight noisily, 3. brain comprising 1. computers ... Deng the possible phrase information that comprises " ㄉ ㄋ ㄠ " phonetic notation, all be shown on the screen, select for the operator.Relative, when the operator only imports " ㄉ ㄋ ", intelligent phonetic input system also can show as comprise among Fig. 5 B 1. then, 2. computer, 3. brain, 4. give repeated exhortations, 5. quarrel and fight noisily ... Deng corresponding phrase information, select for the operator.It should be noted that intelligent phonetic input method provided by the present invention, except can be according to the phonetic symbol of operator input, behind its pronunciation number of decision, provide with words that each its initial consonant of group pronunciation is consistent outside, also can be especially at the input of short sentence or phrase.Please refer to Fig. 5, this figure display application the inventive method is at the detailed process of short sentence or phrase input.At first, via keyboard input phonetic symbol (step 400).Then, as above-mentioned, by with the input phonetic notation classified, and the decision be divided into how much organize pronunciation (step 405).Then, with ready-portioned phonetic notation, be converted to 16 binary coding (step 410).Then, again according to the consonant information in the generation binary coding, Database Systems among Fig. 2 100 are carried out search program, to read all phrases of comprising above-mentioned consonant information and having different length, short sentence, storage address and quantity (step 415) in Database Systems.Subsequently, with phrase, the short sentence that retrieves,, divide successively from long to short (step 420) according to its different length.Import the feature of phonetic notation again according to the operator,, carry out identical simplifying procedures, so that compare (step 425) with phonetic notation that the operator imports with the data that retrieve in the step 420.Then, the pronunciation sign indicating number of relatively being imported with through retrieval and the phrase information after simplifying procedures whether read yard identical (step 430).Wherein, it should be noted that pronunciation sign indicating number when input when having N pronunciation, whether the top n pronunciation sign indicating number that only need relatively retrieve phrase information identical getting final product, so retrieve the pronunciation number of codes (being its number of words) that phrase information is had, also can be greater than N.When the pronunciation sign indicating number after pronunciation sign indicating number and the phrase information of input are simplified was identical, the phrase information that retrieval can be come out was presented in the candidate regions on the screen, and the confession operator selects (step 435) for use.Then, whether decision phrase, the short sentence information with identical number of words handled at present all disposes (step 440), finishes if still be untreated, and then skips back to step 425 again.Finish if will have phrase, the short sentence information processing of same number of words, then proceed to handle (step 445), also can knock-on, to handle word information to step 420 at the less phrase of number of words, short sentence information.When handled phrase, when its pronunciation number of short sentence is imported the pronunciation number of codes less than the operator, then stop handling procedure.
Please refer to Fig. 7, the figure illustrates use the inventive method, the related example of input phrase, short sentence pattern.Wherein, when operator input " ㄥ ㄋ ",, it can be divided into " ㄥ ", " ㄋ ", " " three pronunciations, and it is simplified feature and is similarly and has omitted simple or compound vowel of a Chinese syllable and tone through above-mentioned steps 405.Therefore, after database is retrieved, phrase, the short sentence of read with different numbers of words through the identical back of simplifying procedures (in this example, simple or compound vowel of a Chinese syllable and tone being removed in first to the 3rd group of pronunciation), can be compared with input pronunciation sign indicating number.For example, during input " ㄥ ㄋ ", the short sentence of " woulding you please use the intelligent phonetic input method " will appear on the screen.And, by the rolling mouse device, push the arrow of turning right, can select other required short sentence and phrase.
In addition, intelligent phonetic input method provided by the present invention also has the function of self-built phrase, so that increase the data that store in the database.Please refer to Fig. 8, this figure shows the related procedure of self-built phrase.Wherein, the operator can import phonetic symbol (500) by keyboard.Same, the phonetic symbol of being imported is distinguished the program of pronunciation and transcoding, and produced many groups pronunciation sign indicating number (505).Then, can select prepare word, and, write down this words pronunciation (510) completely, till self-built program finishes according to the order of input pronunciation according to the selected words of operator.Shown in Fig. 9 A to Fig. 9 C, wherein when operator's input " ㄗ ㄒ ㄩ ㄣ ㄐ one ㄚ ㄉ one ㄢ ", the intelligent phonetic system can according to the pronunciation order, show with phrase or individual character, and select for the operator at input pronunciation sign indicating number.In Fig. 9 A, shown " information " of representative " ㄗ ㄒ ㄩ ㄣ " pronunciation sign indicating number.Then, in Fig. 9 B, then show in regular turn " ㄐ one ㄚ " (family), and in Fig. 9 C, shown " ㄉ one ㄢ " () word.So, can finish the word selection program in regular turn, and write down the complete pronunciation of each Chinese character in " information household appliances " simultaneously.Then, after finishing self-built phrase, can be according to preceding two initial consonants of self-built phrase and the number of words of phrase, and orient its position that in database, should deposit (515).Reexamine this position, to confirm whether there are (520) in the words and phrases identical with newly-built phrase, when not having identical words and phrases in the database, then the phrase that will newly build up is read sign indicating number completely together with it and is all inserted position common in the database (525).Simultaneously, make amendment, to change its start address and word quantity (530) at two utmost point index areas 102 in the database 100.Thus, when the operator imported " ㄗ ㄒ ㄐ ㄉ " on keyboard, the intelligent phonetic system will be by above-mentioned retrieval flow, and the screen-picture as " information household appliances " among Fig. 9 D occurred.
Please refer to Figure 10, the figure illustrates intelligent phonetic input system 600 provided by the present invention.Wherein, this system has an input media 610, is used to provide the operator to carry out the phonetic notation loading routine.And have a pronunciation detachment device 620, and be used to respond input media 610, the array pronunciation is divided in the phonetic notation that the operator imported, and determined each group initial consonant that pronunciation comprised in this array pronunciation.In addition, also have a scale-of-two transcoder 630, be connected to this pronunciation detachment device 620, be used for the array pronunciation that to be produced, be converted to the binary codings of 16 of many groups.It should be noted that each Chinese character can use the binary coding of 2 bytes to be represented with shown method among Fig. 2.Then, also have a searching system 640, be connected with a database 100 and respond above-mentioned scale-of-two transcoder 630, so that carry out concordance program according to the pronunciation sign indicating number of being imported.So, can in database 100, find start address corresponding and data bulk with importing the pronunciation sign indicating number.And have one and simplify comparison system 650, be used to determine the simplification feature of the pronunciation sign indicating number of importing, and, simplified with identical simplifying procedures with searching system 640 obtained data from database 100, compare with the pronunciation sign indicating number of being imported again.Wherein, the retrieval pronunciation sign indicating number after simplifying when being conformed to by the pronunciation sign indicating number that the operator imported, then can be shown the words and phrases of this retrieval pronunciation sign indicating number representative via a display device 660, select for use so that the operator to be provided.In addition, have a words and expand module 670, be coupled in and simplify comparison system 650 and database 100, be used to carry out self-built phrase program, to increase the phrase data in the database.Though the present invention with preferred embodiment explanation as above, so it is not to be used to limit the present invention's spirit and invention entity.To those skilled in the art, the modification of being done in not breaking away from spirit of the present invention and scope all should belong to protection scope of the present invention.
Claims (15)
1. intelligent phonetic input method, this method comprises the following steps: at least
Import at least one phonetic symbol;
This phonetic symbol is divided into one group of pronunciation at least;
Organize the initial consonant of this pronunciation according to each, and determine an initial consonant combination;
Carry out search program, from database, to select to comprise the word information of this initial consonant combination;
Relatively whether the pronunciation of this word information comprises this and respectively organizes pronunciation; And
Show and comprise this this word information of respectively organizing pronunciation for you to choose.
2. the method for claim 1 is characterized in that, after this at least one phonetic symbol is divided at least one group of pronunciation, should convert at least one group of scale-of-two pronunciation coding to by at least one group of pronunciation.
3. the method for claim 1 is characterized in that, above-mentioned database comprises:
One group index district divides according to this at least one group of pronunciation;
The secondary index district, corresponding with this one-level index area, and according to this initial consonant combination is divided;
Individual character and phrase block of information, corresponding with this secondary index district, be used to provide the word information that conforms to this initial consonant combination; And
Individual character or phrase information are formed the district, are used to deposit the pronunciation coding of this word information.
4. the method for claim 1 is characterized in that, relatively the step of the pronunciation of this word information and this at least one group of pronunciation also comprises the following steps:
Determine simplifying procedures of this at least one group of pronunciation, to confirm the simplification situation of this each its initial consonant of group pronunciation, simple or compound vowel of a Chinese syllable and tone;
Simplify procedures according to this, each group pronunciation of this word information is simplified, and obtain the simplification pronunciation of this word information; And
Relatively whether the simplification pronunciation of this word information comprises this at least one group of pronunciation.
5. a program of using the intelligent phonetic input method to carry out self-built phrase storehouse is characterized in that, this program comprises the following step at least:
Input needs to increase the phonetic symbol of phrase;
This phonetic symbol is divided at least one pronunciation;
According to the order of this at least one pronunciation, provide prepare word for affirmation successively, and produce one group of phrase words;
Determine the position that this group phrase words is deposited in this database; And
Should organize the phrase words is stored in this database.
6. method as claimed in claim 5 is characterized in that, before storing this group phrase words, also carries out one and checks step, to check whether the existing words and phrases identical with this group phrase words exist this deposit position.
7. method as claimed in claim 5 is characterized in that, after storing this group phrase words, also carries out a change program, with start address and the word quantity of changing index area in this database.
8. an intelligent phonetic input method is characterized in that, this method comprises the following steps: at least
Import at least one group of phonetic symbol;
This phonetic symbol is divided at least one group of pronunciation;
Change this each group pronunciation and be the input pronunciation coding, wherein each is organized this input pronunciation coding and is the scale-of-two pronunciation coding;
Organize the initial consonant of this input pronunciation coding according to each, form the initial consonant combination;
According to this initial consonant combination, carry out search program;
Determine each to organize simplifying procedures of this input pronunciation coding, with confirm this its initial consonant of input pronunciation coding,
The simplification situation of simple or compound vowel of a Chinese syllable and tone;
Simplify procedures according to this, each group retrieval pronunciation coding of this word information is simplified;
Relatively whether the retrieval pronunciation coding of this word information is organized this input pronunciation coding and is consistent with each; And
Show and organize this word information that this input pronunciation coding is consistent with each,
Wherein, above-mentioned search program comprises:
Carry out first search program, make up the same number of word information to mark off with this initial consonant;
Carry out second search program, to mark off the word information that is consistent with this initial consonant combination; And
From storing the address that this word information is handled, read this word information, wherein each is organized this word information and all comprises this initial consonant combination.
9. method as claimed in claim 8 is characterized in that, also comprises the program in a self-built phrase storehouse, and this program comprises the following step at least:
Input needs to increase the phonetic symbol of phrase;
This phonetic symbol is divided at least one pronunciation;
According to the order of this at least one pronunciation, provide prepare word to confirm successively, to produce one group of phrase words for the operator;
Determine the position that this group phrase words is deposited in this database; And
Should organize the phrase words is stored in this database.
10. method as claimed in claim 8 is characterized in that, before storing this group phrase words, comprises that also carrying out one checks step, to check whether the existing words and phrases identical with this group phrase words exist this deposit position.
11. method as claimed in claim 8 is characterized in that, after storing this group phrase words, also comprises and carries out a change program, with start address and the word quantity of changing index area in this database.
12. an intelligent phonetic input system is characterized in that, this system comprises at least:
Input media is used to provide the operator to carry out the phonetic notation loading routine;
The pronunciation detachment device responds this input media, is used for the input phonetic notation is divided at least one group of pronunciation, and
Distinguish at least one group of initial consonant that this at least one group of pronunciation comprises;
Searching system is connected and responds this with a database and read detachment device, so that read from this database
Get at least one words corresponding to this at least one group of initial consonant;
Display device responds this searching system, is used to show this at least one words by retrieval, to provide
This operator recognizes and selects.
13. system as claimed in claim 12 is characterized in that, also comprises a scale-of-two transcoder, responds this pronunciation detachment device, and is coupled in this searching system, is used for this at least one group of pronunciation is converted to binary pronunciation coding.
14. system as claimed in claim 12, it is characterized in that, comprise that also one simplifies comparison system, respond this scale-of-two transcoder and this searching system, and with this display device coupling, be used to determine the simplified condition of this pronunciation coding, and the words that this searching system read is simplified with this simplified condition, compare with this pronunciation coding again.
15. system as claimed in claim 12 is characterized in that, also comprises a words enlargement module, responds this simplification comparison system, and is coupled with this database, is used to carry out self-built phrase program, to increase the phrase data in the database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 00111631 CN1307273A (en) | 2000-01-28 | 2000-01-28 | Intelligent phonetic input system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 00111631 CN1307273A (en) | 2000-01-28 | 2000-01-28 | Intelligent phonetic input system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1307273A true CN1307273A (en) | 2001-08-08 |
Family
ID=4581540
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 00111631 Pending CN1307273A (en) | 2000-01-28 | 2000-01-28 | Intelligent phonetic input system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1307273A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103365925A (en) * | 2012-04-09 | 2013-10-23 | 高德软件有限公司 | Method for acquiring polyphone spelling, method for retrieving based on spelling, and corresponding devices |
| CN102147796B (en) * | 2010-02-05 | 2014-10-15 | 阿里巴巴集团控股有限公司 | Vocabulary searching method and device |
| CN108459734A (en) * | 2017-02-17 | 2018-08-28 | 李建文 | Concentrated quick pinyin input method and system |
| TWI689829B (en) * | 2017-02-17 | 2020-04-01 | 李建文 | Concentrated fast Pinyin input method and system |
-
2000
- 2000-01-28 CN CN 00111631 patent/CN1307273A/en active Pending
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102147796B (en) * | 2010-02-05 | 2014-10-15 | 阿里巴巴集团控股有限公司 | Vocabulary searching method and device |
| CN103365925A (en) * | 2012-04-09 | 2013-10-23 | 高德软件有限公司 | Method for acquiring polyphone spelling, method for retrieving based on spelling, and corresponding devices |
| CN108459734A (en) * | 2017-02-17 | 2018-08-28 | 李建文 | Concentrated quick pinyin input method and system |
| TWI689829B (en) * | 2017-02-17 | 2020-04-01 | 李建文 | Concentrated fast Pinyin input method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US5649023A (en) | Method and apparatus for indexing a plurality of handwritten objects | |
| US5175803A (en) | Method and apparatus for data processing and word processing in Chinese using a phonetic Chinese language | |
| US5197810A (en) | Method and system for inputting simplified form and/or original complex form of Chinese character | |
| EP0841630A3 (en) | Apparatus for recognizing input character strings by inference | |
| CN101149645A (en) | Reduced keyboard disambiguating system | |
| CN101388012A (en) | Phonetic check system and method with easy confusion tone recognition | |
| CN102184167A (en) | Method and device for processing text data | |
| WO2009046612A1 (en) | System for synthetically cognizing entire semantic information and applications thereof | |
| US5331557A (en) | Audio-video coding system for Chinese characters | |
| CN101739143A (en) | Character inputting method and character inputting system | |
| CN100462901C (en) | GB phoneticize input method | |
| CN102867049A (en) | Chinese PINYIN quick word segmentation method based on word search tree | |
| WO2010043117A1 (en) | Digital encoding method and application thereof | |
| CN100476826C (en) | Chinese font sorting and searching method and device and information system | |
| CN1307273A (en) | Intelligent phonetic input system and method | |
| CN119886116B (en) | Chinese text correction method and device based on spell check and computer equipment | |
| CN1147811C (en) | Chinese character identifying method and system with correcting function | |
| CN1286421A (en) | Chinese-character phonetic letter input method with keypad | |
| CN1105985C (en) | Device and method for Chinese input by hand writing and speech sound | |
| CN114595665A (en) | Method for constructing binary extremely-short code word character and word coding set | |
| CN1079060A (en) | Sound-figure word code input system for Chinese character | |
| CN1466039A (en) | Electronic remote controller capable of inputting Chinese and various characters | |
| CN1384426A (en) | Dian code Chinese character input method for computer | |
| CN1027839C (en) | Computer keyboard for Chinese double-spelling Chinese character | |
| CN1269542A (en) | Association Chinese character input system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: GR Ref document number: 1081577 Country of ref document: HK |