TWI564736B - Method of merging single word and multiple words - Google Patents

Description

Method for combining single words and multiple words

The method for merging single words and multiple words into the input method is used for solving the problem that the word prediction result is wrong when the Chinese input is performed, and the correction can be quickly performed, and the correctness of the prediction is improved.

In response to the popularity of mobile communication devices, there are more and more opportunities for mobile communication devices to be transferred to the required applications as text input. Therefore, many operators have begun to specialize in various input methods for typing on users. Speed improvement.

There are many types of input methods, but the Chinese input is taken as an example. The most commonly used is the sound. For example, the phonetic type (phonetic symbol) is based on the pronunciation of the text, and each tone corresponds to a word key. When composing text, input them in order according to the order in which the words are pronounced. The advantage is that it is not necessary to remember the root of the word, it can be read, there is no complicated group rule, and the input is simple; but the disadvantage is that the word that can not be pronounced cannot be input, the homophone is too much, and the word selection is quite time consuming, and then There are input methods based on partial pinyin code. However, there are too many candidates in this way. When the number of input codes is too small and the prediction result is too much, it will take multiple pages to find the desired word.

In order to improve the above shortcomings, a pinyin-based input method, usually called the wisdom input method, is developed, which is to match the pronunciation of Chinese characters to the English letters of similar pronunciations, and to input the sentences to be in English letters. Pinyin corresponds to a Chinese character, and a series of inputting a lot of pinyin becomes a sentence, the system will automatically judge the sentence that the user wants to express. child. However, in Chinese, the expression of Chinese is composed of many single words combined with multiple words. It is often because the system cannot judge the single word for the pinyin, but it is wrong with the subsequent pinyin, which affects the judgment of the whole sentence. It takes more time, and picking the correct text from the candidate menu not only saves time, but also takes more time.

In view of the shortcomings derived from the previous usage methods, the inventor of this case has improved and innovated, and after painstaking research, he finally succeeded in researching and developing a method for combining single words and multiple words.

The main purpose of the present invention is to propose a method for combining single words and multiple words, which is to predict that a pinyin is a single word, and is not combined with the pinyin of the subsequent order, thereby reducing the error of word prediction. purpose.

The secondary object of the present invention is to propose a method for combining single words and multiple words, and when judging that a pinyin is a single word, at the same time, moving the selected cursor to the middle of the pinyin of the pinyin and the subsequent position, Provide users with a quick choice of text or words to enter.

The method for merging single words and multiple words can achieve the above object, and mainly comprises an input unit, and the user can continuously input one pinyin by using a keyboard or a soft/hard device that can replace the keyboard, the pinyin It can be the first code of a word or the complete pinyin of a word, and display the input pinyin item in an input pinyin display unit. The system will use a full sentence prediction unit to perform pinyin combination and predict the corresponding word. Each word can be combined into a specific word or sentence, and then displayed in a full sentence prediction result display unit; when the user continuously inputs a series of pinyin or the input pinyin is not complete, the whole sentence prediction unit A pinyin combination error may occur, which may result in a mistake in the word or sentence prediction. In this case, a single word prediction unit may be used to forcibly input a series of pinyin to perform a certain pinyin. The judgment of a single word, and does not combine the pinyin regarded as a single word with the pinyin located behind it, and displays the single word in a single word prediction result display unit, and is located in the single word After the wording, the pinyin continues to use the whole sentence prediction unit and the whole sentence prediction result display unit to continue the combination, prediction and display of the pinyin, so that the user selects the correct word by using the input unit and the input pinyin display unit. Or words, until the final sentence is completed, using an output unit to send the final sentence to the desired target unit, such as an application or software component that accepts the text, thereby achieving the purpose of improving the prediction accuracy and input efficiency.

11‧‧‧ Input unit

12‧‧‧Input Pinyin Display Unit

13‧‧‧full sentence prediction unit

14‧‧‧Full sentence prediction result display unit

15‧‧‧ single word prediction unit

16‧‧‧Single word prediction result display unit

17‧‧‧Output unit

FIG. 1 is a structural diagram of a method for combining single words and multiple words in the present invention; FIG. 2 is a flow chart of a method for combining single words and multiple words in the present invention; FIG. 3A to FIG. The present invention is a schematic diagram of a method for combining a single word and a multi-word word input; FIG. 4A to FIG. 4F are schematic diagrams showing the implementation of a phonetic symbol of a method for combining single word and multi-word words.

Please refer to FIG. 1 , which is a structural diagram of a method for combining single words and multiple words in the present invention, including: an input unit 11 for connecting an input interface of a mobile communication device or other computer device, and using You can use the keyboard or a soft/hard device that can replace the keyboard and input one by one. Pinyin or phonetic, and the result of the input is displayed on an input pinyin display unit 12; a full sentence prediction unit 13 is used to perform the combination of pinyin or phonetic input input by the input unit 11 to predict the corresponding word, and Each word can be combined into a specific word or sentence, and the predicted result is displayed in a full sentence prediction result display unit 14 for the user to confirm and select the correct text; a single word prediction unit 15, the system When the whole sentence prediction unit 13 generates a pinyin or phonetic combination recognition error, after moving the cursor to the first predicted error symbol in the pinyin or phonetic sequence, a series of pinyin or phonetic input input by the input unit 11 is forcibly performed. A judgment that the predicted error symbol is a single word, and uses the single word as a priority recognition, and does not combine the pinyin or phonetic of the single word with the pinyin or phonetic behind it, and Displaying the single word in a single word prediction result display unit 16 allows the user to quickly select the correct single word word; and the pinyin behind the single word continues to use The full sentence prediction unit 13 and the full sentence prediction result display unit 14 continue to perform combination, prediction and display of pinyin, and can automatically update the determined words, words or sentences immediately, and let the user utilize the input unit 11 and the The input pinyin display unit 12 picks the correct word or word until the final sentence is completed; an output unit 17 sends the finally selected sentence to the desired target unit, such as an application or software component that accepts the text.

Please refer to FIG. 2 , which is a flowchart of a method for combining single words and multiple words in the present invention, comprising the following steps: Step 1: input a partial pinyin or phonetic of a sentence, or a series of times. Input the complete pinyin or phonetic, and display the entered pinyin or phonetic 201; Step 2: automatically combine the input pinyin or phonetic, and predict the pair The word, word or sentence, or the identification 202 which may be a single word (candidate); Step 3: Display the identification result of step 2 (the whole sentence prediction result and the single word prediction result) 203; Step 4 : Let the user judge whether the predicted result is correct 204, if it is correct, go to step 8; if the predicted result is not correct, go to step 5; Step 5: If the predicted result is not correct, move the cursor to the pinyin or phonetic sequence After the first predicted error symbol, the system automatically determines that the symbol corresponds to a single word and displays the result of the single word prediction 205; Step 6: Let the user determine whether the single word prediction result is correct 206; if not correct , then go back to step 5 to continue to judge the correct single word, or continue to select the subsequent single word (because a sentence may have many single words and multiple words), and continue to judge the corresponding sentence; If it is correct, proceed to step 7; Step 7: According to the correct single word selected by the user, and change the latest result predicted by the instant sentence to the correct sentence, let the user do After confirming the selection 207: Step eight: the input word, words and sentences to the target output, such as application programs or software components 208 to accept text, and ends the process.

Please refer to FIG. 3A to FIG. 3G, which are schematic diagrams of the method for combining the input of single words and multiple words according to the present invention: the user wants to input the following sentence: Today's reading theme

JT DDSHZT

DDSHZT candidate shows: most / metropolitan / get / reach ... and other words, are not single words, it is forced to use the single word recognition mechanism as a priority to re-process the "reading" identification, while the cursor Move to the pinyin "D" and "D" in this embodiment, and the input text will become as follows: Today's book club theme

JT D DSHZT

The candidate word of D shows: single/pair/to/ground/("" is very behind), ... and so on, at this time, the user can directly input DE into pinyin, and narrow the range of the selected word to / De/de, select "" to complete part of the input display, and record "today" as a vocabulary, the operation is as follows: today's reading club theme

JTD DSHZT

DSHZT candidate words: board / time / reading ... words (do not show "book club"), select "read", the input text will be displayed as follows: today's book club theme

JTDDS HZT

HZT candidate display: or / Hangzhou / patient ..., at this time press the word key to force the "meeting theme" to be recognized as a single word priority processing, while moving the cursor to the pinyin "H" and "Z" in this embodiment "The input text will become as follows: Today's reading club theme

JTDDS H ZT

H candidate words display: and / / / line / after ("will be very behind" ... and other single words, at this time the user can directly input the HUI in pinyin, narrow the selection of the word to the meeting / back / Swing, After selecting “Meeting”, the input text will be displayed as follows, and the “Reading Club” will be recorded as the vocabulary: Today’s Reading Club Theme

JTDDSH ZT

The candidate words of ZT display: words such as theme/status/topic..., and finally select "topic" to complete the input of words and sentences, and the sentence finally sent to the target is: the theme of today's reading club.

Please refer to FIG. 4A to FIG. 4G, which are schematic diagrams showing the implementation of a phonetic symbol of a method for combining single words and multiple words. In the picture, the user wants to enter the following sentence: Today's TV repair

ㄐㄊ ㄉㄉㄕㄐㄨㄒ

候选 候选 candidate display: location / land / get / set ... and other words, are not single words, then force the use of single word recognition mechanism as a priority to re-process the "TV" identification The user only needs to move the cursor to the phonetic "ㄉ" and "ㄉ" in this embodiment at the same time, an arrow will appear in the figure to remind the user to correct the word, and the input text will become as follows Said: Today's TV repair

ㄐㄊ ㄉ←ㄉㄕㄐㄨㄒ

候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选Words, select "" to complete some of the input and record "today" as a vocabulary, as shown below: Today's TV repair

ㄐㄊㄉ ㄉㄕㄐㄨㄒ

候选 候选 candidate display: TV / TV drama / short time / big world / ... and other words (not showing "TV"), after selecting "TV", the input text will be displayed as follows: today TV repair

ㄐㄊㄉㄉㄕ ㄐㄨㄒ

Candidate word display: Tonight / housework / no ..., at this time press the word key to force the "machine maintenance" to be recognized as a single word priority processing, while moving the cursor to the pinyin in this embodiment "ㄐ Between "and", an arrow will appear in the picture to remind the user to correct the word. The input text will become as follows: TV repair today

ㄐㄊㄉㄉㄕ ㄐ←ㄨㄒ

候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选 候选After the "machine", the input text will be displayed as described below, and the record "television" is the vocabulary: today's TV repair

ㄐㄊㄉㄉㄕㄐ ㄨㄒ

The candidate words of ㄨㄒ display: unlimited / status / topic / repair ... and other words, the final selection of "maintenance" to complete the text and sentence input, and finally sent to the target sentence is: today's book club theme.

The method for combining single words and multiple words provided by the invention, and other methods When the conventional techniques are compared with each other, the following advantages are obtained: 1. By using the method, the recognition error of the candidate word list before the pinyin or the phonetic combination can be improved, thereby achieving the purpose of increasing the text input speed.

The detailed description of the preferred embodiments of the present invention is intended to be limited to the scope of the invention, and is not intended to limit the scope of the invention. The patent scope of this case.

To sum up, this case is not only innovative in terms of technical thinking, but also able to enhance the above-mentioned multiple functions compared with conventional articles. It should fully comply with the statutory invention patent requirements of novelty and progressiveness, and apply in accordance with the law. I urge you to approve this article. Invention patent application, in order to invent invention, to the sense of virtue.

Claims (5)

A method for combining input of a single word and a multi-word in an electronic device having an input unit and a display unit, the method comprising: inputting a first sequence symbol through the input device and displaying the same on the display unit, wherein Each of the first sequence symbols is a first uttered symbol of a corresponding word; generating a first prediction for converting the first sequence symbol to a first multi-word or a first sentence; The first prediction is an error, indicating a symbol of the first predicted error in the first sequence to convert the symbol of the first predicted error into a single word and displaying the predicted result of the single word; determining the single Whether the prediction result of the single word is correct; if not, determining to select the correct single word; and generating a second prediction according to the correct single word selected by the user, for converting the first sequence symbol in the first The sequence symbol after the symbol of the predicted error is a second multi-word or a second sentence. The method of claim 1, wherein each of the symbols of the first sequence is a pinyin or a phonetic symbol. The method of claim 1, wherein the symbol indicating the first predicted error in the first sequence is to move the cursor between the symbol of the first predicted error and the next symbol. s position. The method of claim 1, wherein the symbol indicating the first predicted error in the first sequence is directly movable to a first symbol of the sequence by a designated button The position between the two symbols. The method of claim 1, wherein the converting the symbol of the first predicted error to a single word further comprises inputting a remaining pronunciation symbol corresponding to the single word, wherein the remaining pronunciation symbol Combined with the first predicted error symbol to identify the single word.