CN1924770A - Word input device and method combining association input method and voice recognition method - Google Patents
Word input device and method combining association input method and voice recognition method Download PDFInfo
- Publication number
- CN1924770A CN1924770A CN 200510099856 CN200510099856A CN1924770A CN 1924770 A CN1924770 A CN 1924770A CN 200510099856 CN200510099856 CN 200510099856 CN 200510099856 A CN200510099856 A CN 200510099856A CN 1924770 A CN1924770 A CN 1924770A
- Authority
- CN
- China
- Prior art keywords
- words
- word
- input
- ordering
- prepare
- 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.)
- Granted
Links
Images
Landscapes
- Machine Translation (AREA)
Abstract
This invention discloses one movable communication device work input device and its method, wherein, the input device comprises one keyboard to input one code for one work according to one association method; one microphone to input one sound by sound identification method; one micro processor coupled to the keyboard and microphone to generate at least one prepare words according to the association method and to select the words according to the sound identification method.
Description
Technical field
The present invention relates to a kind of input device and method that is used for device for mobile communication, the input device and the method for the device for mobile communication of particularly a kind of combining association formula input method and sound identification method.
Background technology
At present device for mobile communication (for example: a mobile phone or a personal digital assistant device, the PDA) character input method on, it is very not convenient to be still concerning the user.In traditional input method, the user can reach just press many buttons possibly in order to import an English alphabet or phonetic notation.
Recently association method such as T9 rather in vogue for each English alphabet or phonetic notation, only need by a key, and last English words or the Chinese words that shows, be the method for utilizing dictionary to search, find out all possible combination, bar is listed again provides the user to select.Please refer to Fig. 1, it shows the synoptic diagram of character input keyboard of the mobile phone of a prior art.Association method such as T9 can be described as follows according to this Fig. 1.For example: me:63+ chooses.And for example, sky (ㄊ ㄧ ㄢ ㄨ ㄥ): 209+ chooses, and 309+ chooses.And when pressing this " * " key, can show that then a plurality of special symbols commonly used are for choosing; As when pressing this " # " key, then changeable different input method, the switching of for example input in Chinese (phonetic symbol) and English input (English alphabet/arabic numeral).
Please refer to Fig. 2, it shows the association method of a prior art such as the process flow diagram of T9.Wherein, first step is: " input alphabet of button or phonetic notation " (step 21), its represent a user just on a keyboard button import an English alphabet or a phonetic symbol, can import a words, this words can be a Chinese words, an English word string or an English-word.This word string is formed by connecting by at least one English alphabet.And the complete phonetic symbol of a Chinese words of being keyed in is used to search the prepare word of the Chinese words of this desire input.Second step of this input method is: " whether check this first step finishes " (step 22), if finished and then carried out third step, " queries dictionary is listed prepare word " (step 23), it inquires about a dictionary 25 according to complete letter, arabic numeral or the phonetic notation of keying in, listing at least one prepare word of this desire input words, and these prepare words are listed by an ordering of choosing (for example one of produce according to the institutes such as frequency of utilization of each words and sort); But, then get back to first step (step 21) if step 1 is not finished.The 4th step then is: choose (step 24) by key up and down, the prepare word that its expression is come out when first regulations be not the user want prepare word the time, then need see through by about key choose one by one.Certainly, if when the prepare word that comes out of first regulations is the prepare word that the user wants, then can directly choose.
This mode has been simplified the program of input, most words can both be finished with less button, but if last possible combination is a lot, and the prepare word that comes out of first regulations be not the user want prepare word the time, still see through up and down key and choose one by one.For example: 63:of, me...; 209: electricity, moving, day, wheel ...; 309: light, change rod, sky ....Then during desire input " me ", press 63 keys, then once by key down.And during desire input " sky ", press 209 keys, by key down three times, press 309 keys again, by key down four times.And, as if carrying out Chinese, English simultaneously, when numeral is imported, still needing the manual switchover input method, also be very not convenient to Chinese pattern, English mode, figure pattern.
Therefore, concerning the user, if can provide easier mode to choose prepare word, the program of whole literal input will be convenient.In addition, one user sees through the input media as microphone, phonetic entry one device for mobile communication with the words of desire input, and through a speech recognition technology in addition after the identification, in a dictionary, search prepare word again, for you to choose, also be another prior art of this area, i.e. " sound identification method ".But its problem is, when the correct discrimination power of voice, when relatively low, then may there is no the words that user institute desire is selected in the prepare word of Chu Xianing, and perhaps prepare word quantity is comparatively huge, so that the user is time-consuming because of choosing, and that sensation is used is inconvenient.But if above-mentioned sound identification method is incorporated in this association method, use the voice of this input to confirm, these prepare words that this association method produced are produced an ordering of choosing again according to the similarity of itself and this voice, then can in a plurality of prepare words, find the user to want the words of importing apace.In addition, also can directly say the pronunciation of Chinese, English or numeral, and directly from prepare word, choose, and not need to switch the digital input pattern of Chinese and English according to this pronunciation.Thus, the advantage of combining association formula input method and sound identification method can provide the more convenient literal fast of user inputting interface really.
Thus, the inventor proposes the present invention's " input device of combining association formula input method and sound identification method and method " in view of the shortcoming of prior art.
Summary of the invention
Fundamental purpose of the present invention is to provide the input device and the method for a kind of combining association formula input method and sound identification method, can be applicable to a device for mobile communication, because of the simple loading routine and the sound identification method of its combining association formula input method are chosen the two-fold advantage of words with voice, so the more convenient literal fast of user inputting interface can be provided.
Another fundamental purpose of the present invention is to provide a kind of input device of device for mobile communication, comprise: a keyboard, be used for keying in a code of a words according to an association method, one microphone, be used for importing voice according to a sound identification method, one microprocessor is coupled in this keyboard and this microphone, and this microprocessor produces at least one prepare word of this words and compares these voice and this prepare word to choose this words according to this sound identification method according to this association method.
According to above-mentioned conception, this microprocessor also comprises: a storer comprises: a database comprises: association type input dictionary, produce this prepare word according to this code, and an acoustic model, association type input algorithm, and a speech recognition algorithm, an and micro-control unit, be coupled in this storer, this micro-control unit is carried out this association type input algorithm and this speech recognition algorithm, and compares these voice and this prepare word to choose this words according to this acoustic model.
According to above-mentioned conception, this input device also comprises a display screen, and this display screen is coupled in this microprocessor, and shows wherein arbitrary from this code, this prepare word and this words of this keyboard key entry.
According to above-mentioned conception, this words is that an individual character of a literal and a word string both one of them, arabic numeral and a special symbol commonly used are wherein arbitrary.
According to above-mentioned conception, this individual character is that a Chinese words, a Japan word, a Korean word, an English-word, a German individual character, a French individual character, a Spanish individual character, an Arabic individual character, Russian individual character, an Italian individual character, a Portuguese individual character, a Dutch individual character, a Greek individual character, Czech literary composition individual character and a Danish individual character are wherein arbitrary.
According to above-mentioned conception, this word string is made up of at least one letter of this literal, and this word string is that an English word string, a German word string, a method text strings, a Spanish word string, an Arabic word string, Russia's text strings, an Italian word string, a Portuguese word string, a Dutch word string, a Greek word string, Czech's text strings and a Danish word string are wherein arbitrary.
According to above-mentioned conception, it is wherein arbitrary that this keyboard once can be keyed in a letter, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number of at least one literal.
According to above-mentioned conception, this phonetic symbol be the phonetic symbol of phonetic symbol, a Japanese of a Chinese and a Korean the phonetic symbol three one of them.
According to above-mentioned conception, this association type input dictionary is used to store this literal a plurality of should letter, a plurality of this phonetic symbol, the individual character of a plurality of these literal, a plurality of this arabic numeral sign indicating number should be used special symbol always with a plurality of.
According to above-mentioned conception, this code be at least one should letter, at least one this phonetic symbol and arabic numeral are wherein arbitrary.
According to above-mentioned conception, this association type input algorithm produces one first according to the arrangement from high to low of the frequency of utilization of this at least one prepare word each prepare word wherein and chooses ordering.
According to above-mentioned conception, this speech recognition algorithm first is chosen ordering and is compared the similarity of these voice and this prepare word one pronunciation with this acoustic model according to this, chooses ordering and chooses this words according to this to produce one second.
According to above-mentioned conception, this display screen is used for respectively first choosing ordering and second choosing ordering and show this at least one prepare word in regular turn with this according to this.
According to above-mentioned conception, two input methods that this code institute of twice key entry uses respectively before and after this keyboard certainly need not be switched input method not simultaneously, and only need directly be imported this two code according to this two input method respectively and get final product on this keyboard.
According to above-mentioned conception, these voice are a pronunciation of this words of sending of a user.
According to above-mentioned conception, this device for mobile communication is a mobile phone and a personal digital assistant device (PDA) both one of them.
Of the present invention time a fundamental purpose is to provide a kind of input device that is used for device for mobile communication, comprise: a keyboard, in order to key in a code of a words, one microphone, in order to import voice, one microprocessor, be coupled in this keyboard and this microphone, this microprocessor comprises: a storer, comprise: a database, comprise: association type input dictionary produces an at least one prepare word of this words and an acoustic model according to this code, one association type input algorithm, produce one first of this prepare word and choose ordering, and a speech recognition algorithm, produce one second of this prepare word and choose ordering, an and micro-control unit, be coupled in this storer, this micro-control unit is carried out this association type input algorithm and this speech recognition algorithm, and first chooses to sort and compare these voice and this prepare word and second choose ordering for choosing this words to produce this according to this acoustic model and this.
According to above-mentioned conception, it is wherein arbitrary that this keyboard is used for once keying in an English alphabet, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number.
According to above-mentioned conception, this dictionary is used to store a plurality of phonetic symbols, a plurality of English alphabet, a plurality of Chinese words, a plurality of English words, a plurality of arabic numeral sign indicating number and a plurality of special symbol commonly used.
According to above-mentioned conception, this words is that a Chinese words, an English word string, an English words, arabic numeral and a conventional sign are wherein arbitrary.
According to above-mentioned conception, this English word string is made up of at least one English alphabet.
Another fundamental purpose of the present invention is to provide a kind of character input method that is used for device for mobile communication, be applied to an input media, this input media comprises: a keyboard, a database tool one association type input dictionary and an acoustic model, a microphone and a display screen, this method comprises the following step: (a) certainly this keyboard once to key in an English alphabet, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number wherein arbitrary; (b) show wherein arbitrary of this English alphabet, this phonetic symbol, these a plurality of special symbols commonly used and this arabic numeral sign indicating number keyed in this display screen; (c) then carry out next step if finished the key entry of a code of a words, if not, then carry out step (a); (d) search this dictionary according to this code and show this at least one prepare word in regular turn to choose ordering according to one first at this display screen; (e) this microphone is imported voice certainly; (f) see through this acoustic model these voice and the pronunciation of one of this at least one prepare word first are chosen to sort and compared according to this, choose ordering to produce one second; (g) second choose ordering and show this at least one prepare word in regular turn and choose this words according to this according to this at this display screen; And (h) show that this words is in this display screen.
A fundamental purpose more of the present invention is to provide a kind of character input method that is used for device for mobile communication, comprises the following step: (a) according to an association method, key in one of words code successively; (b) on this device for mobile communication, show this code of having keyed in; (c) if finished the key entry of this code, then carry out next step, if not, then carry out step (a); (d) at least one prepare word and one first that produces this words according to this association method and this code is chosen ordering; (e) first choose ordering and on this device for mobile communication, show this at least one prepare word successively according to this; (f) input one voice; (g) see through a sound identification method one pronunciation of these voice and this at least one prepare word first is chosen that sorting compares chooses ordering to produce one second of this words according to this; And (h) second choose ordering and show successively that on this device for mobile communication this at least one prepare word also chooses this words according to this according to this.
Description of drawings
Fig. 1: it shows the synoptic diagram of character input keyboard of the mobile phone of a prior art;
Fig. 2: it shows the process flow diagram of the association method of a prior art;
Fig. 3: it shows the process flow diagram of the preferred embodiment of the character input method that is used for device for mobile communication of conception according to the present invention;
Fig. 4: it shows the detailed process flow diagram of choosing step as the speech recognition among Fig. 3; And
Fig. 5: it shows the structural representation of the preferred embodiment of the input device that is used for device for mobile communication of conception according to the present invention.
Embodiment
As previously mentioned, association method of the prior art such as T9 are for each English alphabet or phonetic notation, only need, and last English words or the Chinese words that shows is the method for utilizing dictionary to search by a key, find out all possible candidate contamination, bar is listed again provides the user to select.The present invention then is the two-fold advantage of combining association formula input method and sound identification method, remove earlier and to import the code of a words according to the association method button (it can be at least one phonetic symbol, a phonetic symbol for example, an at least one letter or a numeral are wherein arbitrary, in order to represent the words of this desire input), and list outside the prepare word, and provide a phonetic entry by the user, with by sound identification method in prepare word, first (for example chose ordering according to original association method produces, ordering from high to low according to the frequency of utilization of each words) compares these voice and each prepare word and choose ordering (from high to low), choose the words of this input for the user by the similarity of pronunciation to produce one second.Because of the present invention in according to limited the prepare word that association method produced, confirmed again that with sound identification method the degree of accuracy of its speech recognition can be brought up to more than 95 percent, choose the words of its desire input so the user can be very fast.
The present invention uses character input method proposed by the invention because of having above-mentioned advantage, and the program of whole literal input will better convenience.Please refer to Fig. 1, for example: 63:of, me...; 209: electricity, moving, day, wheel ...; 309: light, change rod, sky ...; When desire input " me ", press 63 keys, say " me " again.During desire input " sky ", press 209 keys, say " my god ", press 309 keys again, say " sky " (or press 209,309 keys, say " sky " again).
And, can not need the input of manual switchover Chinese and English when carrying out Chinese and English numeral when mixing input, and after button imported, say with voice and to get final product.For example: 28: platform, at au bt ct, 28...; 27: reach as ar bs cs, 27...; 209: electricity, a w, a x, a y, b y c y, 209...; During desire input " as Delta electricity at 209 ", press 27 keys, say " as ", press 2827209 keys again, say " Delta ", press 28 keys again, say " at ", press 209 again, say " 209 ".
Please refer to Fig. 3, it shows the process flow diagram of the preferred embodiment of the character input method that is used for device for mobile communication of conception according to the present invention.Wherein, first step is: " input alphabet of button or phonetic notation " (step 31), its represent a user just on a keyboard button import an English alphabet (comprising an arabic numeral sign indicating number) or a phonetic symbol, can import a words, this words can be a Chinese words, arabic numeral, a plurality of conventional sign, an English word string or an English-word.This word string is formed by connecting by at least one English alphabet, and the complete phonetic symbol of a Chinese words of being keyed in, and is used to search the prepare word of the Chinese words of this desire input.Second step of this input method is: " whether check this first step finishes " (step 32), if finished and then carried out third step, " queries dictionary is listed prepare word " (step 33), it inquires about a dictionary 25 according to complete letter, arabic numeral or the phonetic notation of keying in, listing at least one prepare word of this desire input words, and these prepare words are listed by an ordering of choosing (a for example ordering that is produced according to the frequency of utilization of each words etc.); But, then get back to first step (step 31) if step 1 is not finished.The 4th step then is: " speech recognition is chosen " (step 34), it provides a phonetic entry by the user, and according to by sound identification method at prepare word that step 33 produced, first (for example chose ordering according to original association method produces, ordering from high to low according to the frequency of utilization of each words) compares these voice and each prepare word and choose ordering (mark by the similarity of pronunciation is arranged from high to low), choose the words of this input for the user to produce one second.The soprano that wherein sorts that is to say a most probable words.If the user is desired, then directly output; If not, then allow the user choose with button.
Please referring to Fig. 4, it chooses the detailed process flow diagram of step (step 34) for this speech recognition among Fig. 3, this speech recognition is chosen step (step 34) and also further comprised following step: its first step is: " speech recognition rearranges prepare word " (step 341), promptly choose ordering to produce one second according to this first similarity of choosing ordering this phonetic entry of comparison and one pronunciation of each prepare word that original association method produced.Its second step is: " listing the most probable words " (step 342), ordering (mark) soprano among this second ordering promptly is a most probable words, promptly be listed in first cis-position this moment, chooses for this user.Its third step is: " correctness " (step 343), even this most probable words of listing is " correctly ", then the user can choose and import in this/english word, if not, then entering the 4th step (step 344) " chooses by key up and down ", even this most probable words of listing is not optant that this user desires, then this user can be this moment by key up and down to choose the words of its desire input.As previously mentioned, utilization speech recognition technology, in a spot of vocabulary number, discrimination power generally can reach more than 95%, and therefore the identification result of the process most probable words that step 342 produced is the words that the user will import usually in the present invention.Even in 5% mistake, still have most correct words can appear at second cis-position or the 3rd cis-position, also can allow the user choose easily by individual key.Thus, both advantages chosen in input of combining association formula button and voice, and the more convenient literal fast of user inputting interface can be provided.
In sum, the character input method that is used for device for mobile communication proposed by the invention comprises the following step: (a) according to an association method, key in a code of a words successively; (b) on this device for mobile communication, show this code of having keyed in; (c) if finished the key entry of this code, then carry out next step, if not, then carry out step (a); (d) at least one prepare word and one first that produces this words according to this association method and this code is chosen ordering; (e) first choose ordering and on this device for mobile communication, show this at least one prepare word successively according to this; (f) input one voice; (g) see through a sound identification method one pronunciation of these voice and this at least one prepare word first is chosen that sorting compares chooses ordering to produce one second of this words according to this; And (h) second choose ordering and show successively that on this device for mobile communication this at least one prepare word also chooses this words according to this according to this.
Please referring to Fig. 5, it shows the structural representation of the preferred embodiment of the input device that is used for device for mobile communication of conception according to the present invention.The input device 1 of a device for mobile communication wherein comprises: a keyboard 11 is used for keying in according to an association method code of a words; One microphone 12 is used for importing voice according to a sound identification method; One microprocessor 13 is coupled in this keyboard 11 and this microphone 12, and this microprocessor 13 produces at least one prepare word of this words and compares these voice and this prepare word to choose this words according to this sound identification method according to this association method.Wherein this microprocessor 13 also comprises: a storer 131, comprise: a database 1311, comprise: association type input dictionary 13111, produce this prepare word according to this code, an and acoustic model 13112, one association type input algorithm 1312, an and speech recognition algorithm 1313, an and micro-control unit 132, be coupled in this storer 131, this micro-control unit 132 is carried out this association types input algorithm 1312 and this speech recognition algorithm 1313, and according to these acoustic model 13112 these voice of comparison and this prepare word to choose this words.In addition, this input device 1 also comprises a display screen 14, and this display screen 14 is coupled in this microprocessor 13, and shows wherein arbitrary from this code, this prepare word and this words of these keyboard 11 key entries.In addition, this words is that an individual character of a literal and a word string both one of them, arabic numeral and a special symbol commonly used are wherein arbitrary.Above-mentioned this literal for any literal of can association method importing a device for mobile communication one of them (as in/English etc., kind of literal can be imported a device for mobile communication by association method surplus the whole world amounted to existing 40 at present).Therefore, it is wherein arbitrary that this individual character can be a Chinese words, a Japan word, a Korean word, an English-word, a German individual character, a French individual character, a Spanish individual character, an Arabic individual character, Russian individual character, an Italian individual character, a Portuguese individual character, a Dutch individual character, a Greek individual character, Czech literary composition individual character and a Danish individual character etc.And a word string is by wherein at least one should the composition by letter of the literal (with the literal that letter is write, for example English has 26 letters) of a plurality of letters of a tool.Therefore, it is wherein arbitrary that this word string can be an English word string, a German word string, a method text strings, a Spanish word string, an Arabic word string, Russia's text strings, an Italian word string, a Portuguese word string, a Dutch word string, a Greek word string, Czech's text strings and a Danish word string etc.In addition, this keyboard 11 once can key in above-mentioned can be wherein arbitrary by a letter, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number of at least one literal of association method input.And indicate being transfused to of phonetic symbol of words pronunciation, in order to as the code of representing this words.For example, its can be the phonetic symbol of phonetic symbol, a Japanese of a Chinese and a Korean the phonetic symbol three one of them.Wherein, this association type input dictionary 13111 be used to store above-mentioned can import at least one this literal a plurality of of a device for mobile communication by association method should letter, a plurality of this phonetic symbol, a plurality of individual characters of this literal, a plurality of this arabic numeral sign indicating number should use special symbol always with a plurality of.This above-mentioned code, its be at least one should letter, at least one this phonetic symbol and arabic numeral are wherein arbitrary.This association type input algorithm 1312 produces one first according to the arrangement from high to low of the mark (for example, can be produced by frequency of utilization) of wherein each of this at least one prepare word and chooses ordering.This speech recognition algorithm 1,313 first (is for example chosen similarity that ordering pronounces with this acoustic model 13112 these voice of comparison and this prepare word one according to this, with the height of a produced mark with hang down represent the big and little of its similarity), choose ordering and choose this words according to this to produce one second.In addition, this display screen 14 also is used for respectively first choosing ordering and second choosing ordering and show this at least one prepare word in regular turn with this according to this.When this code of this keyboard twice key entry in 11 front and back certainly two input methods used respectively not simultaneously (not for example, for the first time be that Chinese character coding input method is English input method for the second time), need not switch input method, only need on this keyboard, directly to import this two code by this two input method respectively and get final product.Above-mentioned these voice from 12 inputs of this microphone are a pronunciation of this words of being sent of a user.Input device 1 of the present invention, can apply to any can with the device for mobile communication that cooperates.For example, it can be a mobile phone and a personal digital assistant device (PDA) both one of them.
Please referring again to Fig. 5, below will analyze the operation principles of this input device that is used for device for mobile communication 1 with regard to the function of the interaction of the wherein each assembly of this preferred embodiment and institute's tool thereof.This input device 1 comprises: a keyboard 11, in order to key in a code of a words; One microphone 12 is in order to import voice; One microprocessor 13 is coupled in this keyboard 11 and this microphone 12, and this microprocessor 13 comprises: a storer 131 comprises: a database 1311 comprises: an association type is imported dictionary 13111, produces at least one prepare word of this words according to this code; An and acoustic model 13112; One association type input algorithm 1312 produces one first of this prepare word and chooses ordering; And a speech recognition algorithm 1313, produce one second of this prepare word and choose ordering; An and micro-control unit 132, be coupled in this storer 131, this micro-control unit 132 is carried out this association types input algorithm 1312 and this speech recognition algorithm 1313, and first chooses ordering these voice of comparison and this prepare word and second choose ordering for choosing this words to produce this according to this acoustic model 13112 and this.The function of this display screen 14 is then identical with aforementioned person.
Usually in a country or an area, the literal of its employed device for mobile communication institute desire input should be the international language (as English) that passes through in the most general a kind of literal (as Chinese) of this country or this area institute and a kind of world, because the text type that can import when this input device the more, then system architecture or relevant software are then more complicated, and relevant costs such as manufacturing are also higher.So, as with in above-mentioned the input/English device for mobile communication is an example, it is wherein arbitrary that then this keyboard 11 is used for once keying in an English alphabet, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number.This association type input dictionary 13111 is used to store a plurality of phonetic symbols, a plurality of English alphabet, a plurality of Chinese words, a plurality of English words, a plurality of arabic numeral sign indicating number and a plurality of special symbol commonly used.This words is that a Chinese words, an English word string, an English words, arabic numeral and a conventional sign are wherein arbitrary.And should the English word string be formed by at least one English alphabet.
Please refer to the structure of the preferred embodiment 1 of the input device that is used for device for mobile communication shown in Figure 5, as being that this method is analyzed on the basis with each element that directly relates to the proposed character input method of the present invention, but then its operation principles and step division are as follows.This is used for the character input method of device for mobile communication, be applied to an input media 1, this input media 1 comprises: a keyboard 11, a database 1311 tools one association type input dictionary 13111 and an acoustic model 13112, a microphone 12 and a display screen 14, this method comprises the following step: (a) certainly this keyboard 11 once to key in a letter, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number wherein arbitrary; (b) show that this letter, this phonetic symbol, these a plurality of special symbols commonly used and this arabic numeral sign indicating number keyed in are wherein arbitrary in this display screen 14; (c) then carry out next step if finished the key entry of a code of a words, if not, then carry out step (a); (d) search this dictionary 13111 according to this code and show this at least one prepare word in regular turn to choose ordering according to one first at this display screen 14; (e) this microphone 12 is imported voice certainly; (f) see through this acoustic model 13112 pronunciation of these voice and this at least one prepare word first is chosen to sort and compared according to this, choose ordering to produce one second; (g) second choose ordering and show this at least one prepare word in regular turn and choose this words according to this according to this at this display screen 14; And (h) show that this words is in this display screen.
By above-mentioned explanation as can be known, the invention reside in input device and method that a kind of combining association formula input method and sound identification method are provided, can be applicable to a device for mobile communication, because of the simple loading routine and the sound identification method of its combining association formula input method are chosen the two-fold advantage of words with voice, so the more convenient literal fast of user inputting interface can be provided.
So, even if the present invention is described in detail by the above embodiments and can makes many variations on the basis that does not break away from spirit of the present invention by those of ordinary skills, the scope that the present invention advocated must be as the criterion so that the claim in claims is described.
Claims (10)
1. the input device of a device for mobile communication comprises:
One keyboard is used for the code that foundation one association method is keyed in a words;
One microphone is used for importing voice according to a sound identification method;
One microprocessor, be coupled in described keyboard and described microphone, described microprocessor produces at least one prepare word of described words and compares described voice and described prepare word to choose described words according to described sound identification method according to described association method.
2. input device as claimed in claim 1 is characterized in that,
Described microprocessor also comprises:
One storer comprises:
One database comprises:
One association type input dictionary produces described prepare word according to described code; And
One acoustic model;
One association type input algorithm; And
One speech recognition algorithm; And
One micro-control unit is coupled in described storer, and described micro-control unit is carried out described association type input algorithm and described speech recognition algorithm, and compares described voice and described prepare word to choose described words according to described acoustic model; And/or
Described input device also comprises a display screen, and described display screen is coupled in described microprocessor, and demonstration is wherein arbitrary from described code, described prepare word and described words that described keyboard is keyed in, wherein:
Described association type input dictionary is used to store the individual character of a plurality of described letter of described literal, a plurality of described phonetic symbol, a plurality of described literal, a plurality of described arabic numeral sign indicating number and a plurality of described special symbol commonly used;
Described code is that at least one described letter, at least one described phonetic symbol and arabic numeral are wherein arbitrary;
Described association type input algorithm produces first according to the arrangement from high to low of the frequency of utilization of described at least one prepare word each prepare word wherein and chooses ordering;
Described speech recognition algorithm is chosen ordering according to described first and is compared the similarity of described voice and described prepare word pronunciation with described acoustic model, chooses ordering and chooses described words according to this to produce second; And/or described display screen is used for respectively choosing ordering and described second according to described first and chooses ordering and show this at least one prepare word in regular turn.
3. input device as claimed in claim 1 is characterized in that,
This words is that an individual character of a literal and a word string both one of them, arabic numeral and a special symbol commonly used are wherein arbitrary;
This individual character is that a Chinese words, a Japan word, a Korean word, an English-word, a German individual character, a French individual character, a Spanish individual character, an Arabic individual character, Russian individual character, an Italian individual character, a Portuguese individual character, a Dutch individual character, a Greek individual character, Czech literary composition individual character and a Danish individual character are wherein arbitrary; And/or
This word string is made up of at least one letter of this literal, and this word string is that an English word string, a German word string, a method text strings, a Spanish word string, an Arabic word string, Russia's text strings, an Italian word string, a Portuguese word string, a Dutch word string, a Greek word string, Czech's text strings and a Danish word string are wherein arbitrary.
4. input device as claimed in claim 1 is characterized in that, it is wherein arbitrary that described keyboard once can be keyed in a letter, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number of at least one literal; And/or
Described phonetic symbol be the phonetic symbol of phonetic symbol, a Japanese of a Chinese and a Korean the phonetic symbol three one of them.
5. input device as claimed in claim 1 is characterized in that,
Two input methods of using respectively when the described code institute of twice key entry before and after described keyboard need not be switched input method not simultaneously, only need on described keyboard respectively described two input methods of foundation directly import described two codes and get final product;
Described voice are the pronunciation of one of described words of sending of a user; And/or
Described device for mobile communication be a mobile phone and a personal digital assistant device one of them.
6. input device that is used for device for mobile communication comprises:
One keyboard is in order to key in one of words code;
One microphone is in order to import voice;
One microprocessor is coupled in described keyboard and described microphone, and described microprocessor comprises:
One storer comprises:
One database comprises:
One association type is imported dictionary, produces at least one prepare word of described words according to described code; And
One acoustic model;
One association type input algorithm produces one of described prepare word first and chooses ordering; And
One speech recognition algorithm produces one of described prepare word second and chooses ordering; And
One micro-control unit, be coupled in described storer, described micro-control unit is carried out described association type input algorithm and described speech recognition algorithm, and chooses to sort according to described acoustic model and described first and compare described voice and described prepare word is chosen ordering for choosing described words to produce described second.
7. input device as claimed in claim 6 is characterized in that,
It is wherein arbitrary that described keyboard is used for once keying in an English alphabet, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number; And/or
Described dictionary is used to store a plurality of phonetic symbols, a plurality of English alphabet, a plurality of Chinese words, a plurality of English words, a plurality of arabic numeral sign indicating number and a plurality of special symbol commonly used.
8. input device as claimed in claim 6 is characterized in that described words is that a Chinese words, an English word string, an English words, arabic numeral and a conventional sign are wherein arbitrary, and wherein said English word string is made up of at least one English alphabet.
9. character input method that is used for device for mobile communication, be applied to an input media, described input media comprises: a keyboard, a database have association type input dictionary and an acoustic model, a microphone and a display screen, and described method comprises the following step:
(a) certainly this keyboard once to key in a letter, a phonetic symbol, a plurality of special symbol commonly used and an arabic numeral sign indicating number wherein arbitrary;
(b) show that described letter, described phonetic symbol, described a plurality of special symbols commonly used and the described arabic numeral sign indicating number keyed in are wherein arbitrary in this display screen;
(c) then carry out next step if finished the key entry of a code of a words, if not, then carry out step (a);
(d) search described dictionary according to described code and show described at least one prepare word in regular turn to choose ordering according to one first at described display screen;
(e) import voice from described microphone;
(f) through described acoustic model the pronunciation of one of described voice and described at least one prepare word is chosen ordering according to described first and compare, choose ordering to produce one second;
(g) choosing ordering at described display screen according to described second shows described at least one prepare word in regular turn and chooses described words according to this; And
(h) show that described words is in described display screen.
10. character input method that is used for device for mobile communication comprises the following step:
(a), key in a code of a words successively according to an association method;
(b) on described device for mobile communication, show the described code of having keyed in;
(c) if finished the key entry of described code, then carry out next step, if not, then carry out step (a);
(d) at least one prepare word and one first that produces described words according to described association method and described code is chosen ordering;
(e) choose ordering according to described first and on described device for mobile communication, show described at least one prepare word successively;
(f) input one voice;
(g) see through a sound identification method and one pronunciation of described voice and described at least one prepare word is chosen according to described first sorting compares chooses ordering to produce one second of described words; And
(h) choosing ordering according to described second shows described at least one prepare word successively and chooses described words according to this on described device for mobile communication.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200510099856XA CN100561411C (en) | 2005-08-31 | 2005-08-31 | The input device of combining association formula input method and sound identification method and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB200510099856XA CN100561411C (en) | 2005-08-31 | 2005-08-31 | The input device of combining association formula input method and sound identification method and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1924770A true CN1924770A (en) | 2007-03-07 |
| CN100561411C CN100561411C (en) | 2009-11-18 |
Family
ID=37817436
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB200510099856XA Expired - Fee Related CN100561411C (en) | 2005-08-31 | 2005-08-31 | The input device of combining association formula input method and sound identification method and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100561411C (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010124515A1 (en) * | 2009-04-30 | 2010-11-04 | 广东国笔科技股份有限公司 | Control system and method of semantic real-time analysis and association |
| CN102221985A (en) * | 2010-04-16 | 2011-10-19 | 韦宏伟 | Chinese and control command voice recognition input method and device |
| CN102541252A (en) * | 2010-12-26 | 2012-07-04 | 上海量明科技发展有限公司 | Method and system for voice assisted character inputting |
| CN102566748A (en) * | 2010-09-30 | 2012-07-11 | 佳能株式会社 | Character input apparatus equipped with auto-complete function, and method of controlling the same |
| CN108376033A (en) * | 2017-09-08 | 2018-08-07 | 淄博职业学院 | A kind of english character string input method applied in English inputting method |
-
2005
- 2005-08-31 CN CNB200510099856XA patent/CN100561411C/en not_active Expired - Fee Related
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2010124515A1 (en) * | 2009-04-30 | 2010-11-04 | 广东国笔科技股份有限公司 | Control system and method of semantic real-time analysis and association |
| CN102221985A (en) * | 2010-04-16 | 2011-10-19 | 韦宏伟 | Chinese and control command voice recognition input method and device |
| CN102566748A (en) * | 2010-09-30 | 2012-07-11 | 佳能株式会社 | Character input apparatus equipped with auto-complete function, and method of controlling the same |
| CN102566748B (en) * | 2010-09-30 | 2015-05-06 | 佳能株式会社 | Character input apparatus equipped with auto-complete function, and method of controlling the same |
| CN102541252A (en) * | 2010-12-26 | 2012-07-04 | 上海量明科技发展有限公司 | Method and system for voice assisted character inputting |
| CN108376033A (en) * | 2017-09-08 | 2018-08-07 | 淄博职业学院 | A kind of english character string input method applied in English inputting method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100561411C (en) | 2009-11-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7395203B2 (en) | System and method for disambiguating phonetic input | |
| KR100656736B1 (en) | System and method for disambiguating phonetic input | |
| CN1133918C (en) | Symbol input | |
| CN1834865A (en) | Multi-character continuous inputting method of Chinese phonetic and notional phonetic alphabet with digitally coded on keypad | |
| CN100478852C (en) | Digital keyboard English and Chinese input method | |
| US20070038456A1 (en) | Text inputting device and method employing combination of associated character input method and automatic speech recognition method | |
| CN1924770A (en) | Word input device and method combining association input method and voice recognition method | |
| CN1704879A (en) | Method and apparatus for inputting Chinese characters and phrases | |
| CN1129838C (en) | Free-type Chinese-character enter method using keypad and its device | |
| CN1097766C (en) | Chinese-character 5-key input method | |
| CN1991743A (en) | Method and device for voice input method | |
| CN1283828A (en) | Database index method and system | |
| CN1510554A (en) | Embedded applied Chinese character inputting method | |
| CN1157919C (en) | Chinese character and word input method and system | |
| CN1744005A (en) | Mobile phone code mobile phone input platform | |
| CN1584810A (en) | Chinese inputting method | |
| CN1595341A (en) | Chinese-English number and five-sound input method and keyboard | |
| CN1752900A (en) | Chinese character input method combined with speech sound identification | |
| CN1243301C (en) | Chinese character computer/cell phone input integrated code | |
| CN1208711C (en) | Digital English inputting method and keyboard thereof | |
| CN1272693C (en) | Artificial phonetic digital input method | |
| CN1866173A (en) | Simple ten-numerical-key input method for Chinese, English and Korean | |
| CN1290306C (en) | Chinese characters inputting method | |
| CN1376961A (en) | Chinese-character 'Pronunciation-meaning code' input method | |
| CN1256453A (en) | Foreign language syllable input method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20091118 Termination date: 20210831 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |