Background technology
Input method, people are undesirable the most all the time also is the problem of difficult effective solution, nothing more than being exactly input speed and easy-to-use these two.General most input method all is to lay particular stress on the dealing with problems of input speed, utilize its independently developed uniqueness purpose that input mode reaches increases input speed of divining by means of characters, in the face of these input methods user often must and recite its special input mode of divining by means of characters through one section not very short time self training, the sense of accomplishment of enjoying the input speed increase that can be real, but this threshold that enters use for general most common user is quite high, and on input is used, also seem quite not convenient, can't attract general user to use.
Some input method is just by providing the mode of built-in dictionary, wish to solve simultaneously input speed and easy-to-use on problem, when utilizing this mode user when importing, to run into the sentence that is had in the built-in dictionary, just the time that can save input again directly reaches the purpose of quick input by the mode of choosing, but because the actual operation mode of these input methods all is after the input of accepting the user mostly, by the mode of finding out possible words and phrases in the dictionary and utilizing candidate list with project one by one bar list and allow the user select, and the mode of selecting can be finished the action of choosing input at last after often all must the user clicking with keyboard input or mouse by manual mode, this must often use the user of words and phrases in the dictionary for those, the frequent input action of manually choosing causes them using one on the dictionary wish to hinder greatly, the mistake of utilizing manual mode to carry out that choosing of candidate's words and phrases also may be easy to simultaneously to choose because of wholwe-hearted for the moment or carelessness, in the end being used for time of deletion error input words and phrases may be more, may have influence on the velocity efficiency of whole input on the contrary.
What speech recognition science and technology had developed at present is quite flourishing, though comprehensive speech recognition still has some technical difficulty that temporarily can't overcome, as if also has one section gap in the identification of distance realization full voice, but but be out of question already for the speech recognition technology on the fundamental element that is applied in a few fixed, therefore to how to pass through the design and the execution of computer program, develop a kind of can be in input method system in conjunction with the advantage of speech recognition technology, auxiliary user carries out the input of choosing of candidate's words and phrases, in order to reduce the program that the user manually imports repeatedly, reduce the manually wrong incidence of input, and then improving the speed of user in whole words and phrases input, this is only, and present many input method users institute very expects.
Summary of the invention
In view of this, the present invention addresses the above problem a kind of Chinese character input method and system thereof with speech recognition subsidiary function that propose, fundamental purpose is to use the speech recognition technology, carry out choosing in the action of candidate's words and phrases or candidate's classification the user, provide mode to reach the number of times that reduces manually input, the generation of avoiding mistake and the purpose that improves whole input speed with phonetic entry.
Institute thinks and reaches above-mentioned purpose that the input in Chinese system with speech recognition subsidiary function provided by the present invention comprises: the words and phrases database, receive recognition module, data acquisition module, and output display module.
In addition, Chinese character input method with speech recognition subsidiary function provided by the present invention, comprise the following step: but in Edit Document, activate the initial interface of input method earlier, show the class categories inventory then, utilize phonetic entry candidate words and phrases numbering to carry out the action of choosing of class categories, show category item words and phrases wherein, utilize voice selecting candidate words and phrases numbering and acquisition corresponding class project words and phrases, but and the category item words and phrases are outputed in the Edit Document.
Relevant detailed content of the present invention and technology, existing accompanying drawings is as follows:
Embodiment
The present invention relates to a kind of Chinese character input method and system thereof with speech recognition subsidiary function, by Fig. 1 system architecture of the present invention is described, this figure is the have Chinese character input method of speech recognition subsidiary function and the system construction drawing of system thereof provided by the present invention, is described as follows:
(1) the words and phrases database 210, are responsible for storing all class categories content and category item words and phrases.This words and phrases database 210 is a kind of relevant database, sets up the association index between each class categories and each category item words and phrases, can be more accurately and fast on data search.
(2) the data acquisition module 220, mainly according to receiving the speech conversion result that recognition module 230 is sent, in words and phrases database 210 data item of correspondence come out to acquisition.The state of can record in this data acquisition module 220 at present shown class categories inventory, when obtaining speech conversion as a result the time, this module 220 can contrast corresponding data item actual in the database, and then acquisition in the words and phrases database 210.
(3) receive recognition module 230, receive the voice that the user imported, carry out the work of identification and conversion then by phonetic entry peripheral unit (as: microphone, radio ...).Is wherein the program of identification mainly to find out what is the real candidate words and phrases numbering that the user imports the voice representative? be can identification in the zone of reasonableness real candidate words and phrases numbering? if user's input voice exceed the scope of candidate's words and phrases numbering, or because unclear causing of voice fuzzy can't carry out identification, then this reception recognition module 230 can be carried out the prompting of error message, and requires the user to carry out the input of voice more again.As for the part of conversion, then be after the input voice of confirming the user are the cognizable real candidate words and phrases numbering of zone of reasonableness, the simulating signal content is made it to become the accessible digital signal of computing machine by conversion.
(4) output display module 240, be mainly used to show class categories inventory (wherein the class categories inventory comprises phonetic entry district and project viewing area two parts), in addition, when the user selectes the category item words and phrases that will import at last, this output display module 240 will be automatically outputs to the words and phrases of correspondence in the document that the user using, and can reduce real input time of user.
The part of Fig. 2 then illustrates phonetic entry flow process of the present invention, and this figure is the have Chinese character input method of speech recognition subsidiary function and the phonetic entry process flow diagram of system thereof provided by the present invention, is described as follows:
At first, but the user can be in employed Edit Document (as: WORD, Notepad...), and (step 300) opened at the initial interface 100 of input method of the present invention; Input method system 200 of the present invention afterwards can demonstrate the picture (step 310) of a class categories inventory, comprises a phonetic entry district and a project viewing area in this picture, is available for users to begin to carry out the action of phonetic entry; This moment the user can utilize voice according to candidate's words and phrases shown on the class categories inventory number carry out classification choose action (step 320); Can input method system of the present invention then just be judged the class categories that still has other outside the selected project of user? (step 330) if any, then flow process is got back to the place of step 310 once again, continue to show further class categories inventory, otherwise, expression has entered into the part of the last category item words and phrases of this classification, just then carries out the demonstration work (step 340) of category item words and phrases; This moment, the user can be again selects oneself to want candidate's words and phrases numberings (step 350) of the category item words and phrases imported by voice; At last, but the present invention just these classification project words and phrases are given and are outputed in the employed Edit Document of user (step 360) automatically, finish the flow process of utilizing phonetic entry to choose words and phrases.
Then, illustrate with Fig. 3 how the present invention utilizes the input of voice to carry out the acquisition of data, this figure is the have Chinese character input method of speech recognition subsidiary function and the data acquisition process flow diagram of system thereof provided by the present invention, further specifies as follows:
At first, input method system 200 of the present invention must receive the voice content (step 400) that the user imports by the phonetic entry peripheral unit earlier; Then begin to carry out the program (step 410) of speech recognition, wherein the program of identification mainly be to find out the voice of user's input be can identification in the zone of reasonableness real candidate words and phrases numbering? (step 420) is if user's input voice exceed the scope of candidate's words and phrases numbering, or because unclear causing of voice fuzzy can't carry out identification, then carry out the prompting (step 470) of error message, and get back to the part of step 400, require the user to carry out the input of voice more again, otherwise just proceed the action of speech conversion, the simulating signal content is made it to become the accessible digital signal of computing machine by conversion; Produce computing machine and can carry out the digital signal (step 430) of processing, the data item that just will correspond on the class categories inventory according to the represented content of digital signal captures come out (step 440) in words and phrases database 210 afterwards; Carry out display action (step 450) by output display module 240 then; At this moment, if input method system 200 of the present invention can be judged when content displayed has arrived last category item words and phrases, whether selected wherein one and of user with this classification project words and phrases output? (step 460) if not words, class categories exists to represent to also have further by shown class categories inventory, then flow process can be got back to the part of step 400 once again, continue to receive user's phonetic entry, otherwise, current data acquisition flow process has been finished in expression, finishes whole flow process.
At last, carry out the explanation of specific embodiments of the invention below by Fig. 4 a, Fig. 4 b and Fig. 4 c:
During beginning, the user can see the class categories inventory that is presented as Fig. 4 a, wherein the part in the project viewing area demonstrates the basic classification project (as: life, nature, life, life ...) of all words and phrases in the words and phrases database 210, this moment, the user can utilize the mode of voice to import, for instance, when if the user wants input about the words and phrases under " life " classification, the user can import candidate's words and phrases numberings (i.e. " 1 ") of " life " classification in the mode of voice; Then the picture that occurs will be shown in Fig. 4 b, in the project viewing area, can demonstrate each category item (comprising: diet, work, family, amusement, love ...) under all " life " classifications, in the phonetic entry district, then demonstrate the residing words and phrases class categories of present user state; At this moment, if the user wish to import be the words and phrases of " work " aspect the time, the user can number (i.e. " 2 ") by candidate's words and phrases that the mode of voice be imported representative " work " classification; Similarly, system can be shown each category item under " work " classification, if when under " work " classification, not having the class categories of other thinner portion, system just can give category item words and phrases all under " work " classification and show (as shown in Fig. 4 c), at this time the user can utilize candidate's words and phrases numbering of phonetic entry correspondence that the own words and phrases of being wanted are chosen out equally, system can be automatically come out corresponding class project words and phrases to acquisition, output to automatically then the user using in editor's the document.
Though preferred embodiment of the present invention discloses as above; yet be not in order to limit the scope of the invention; any technician who is familiar with this skill; without departing from the spirit and scope of the present invention; a little change and the modification done all should be included within the scope that claims of the present invention protect.