CN100365551C - Words input method and apparatus for hand-held devices - Google Patents
Words input method and apparatus for hand-held devices Download PDFInfo
- Publication number
- CN100365551C CN100365551C CNB2004100428307A CN200410042830A CN100365551C CN 100365551 C CN100365551 C CN 100365551C CN B2004100428307 A CNB2004100428307 A CN B2004100428307A CN 200410042830 A CN200410042830 A CN 200410042830A CN 100365551 C CN100365551 C CN 100365551C
- Authority
- CN
- China
- Prior art keywords
- module
- sound
- input
- literal
- female
- 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.)
- Active
Links
Abstract
The present invention discloses a word input method and a device for hand held devices. When a word is input into one hand held device, speech input is combined with keyboard input, corresponding word information is identified through the speech input, and then, a desired word is selected from similar words through the keyboard input. The speech input can be the speech of initials and finals, strokes, radicals, etc., and accordingly, the word information can be the combination of the initials and the finals, the strokes, the radicals, etc. According to the present invention, the simple word information is identified just according to the speech input and a sample training is very simple; the operand of the hand held devices is slightly small and requirements for the operational capability of the hand held devices are low; the correct rate of the word input method provided by the present invention is slightly high; simultaneously, the method only uses the keyboard to input the words for one time so that word input speed is greatly improved. Next, the word input method provided by the present invention is called as a mixed input method.
Description
Technical field
The present invention relates to the literal input technology, be meant a kind of character input method and device especially at handheld device.
Background technology
Along with popularizing on a large scale of computer technology and development of Communication Technique and application, need be at mobile phone, personal digital assistant (PDA, Personal Digital Assistant), personal computer (PC, Personal Computer) carries out editing and processing to Word message more and more.For input characters on various terminal devices, at the type characteristic of literal, the ability of input equipment etc., designed diversified character input method, as input methods such as optical character identification (OCR, Optical Character Recognition), speech recognition input and various phonetic inputs.OCR identifies concrete literal after being meant and scanning at literal, and described literal comprises the literal of handwriting input.
Fig. 1 realizes synoptic diagram for phonitic entry method, as shown in Figure 1, the module that realizes phonitic entry method comprises microphone (MIC/microphone) module 101, adopts module 102, sound identification module 103 and literal output module 104, wherein, the voice signal that MIC module 101 is used for receiving is converted to simulating signal, and this simulating signal is sent to sampling module 102; The analog signal conversion that sampling module 102 is used for receiving is a digital signal, and this digital signal is sent to sound identification module 103; Sound identification module 103 is used for identifying corresponding literal according to the digital signal that receives, the literal that identifies may have a plurality of, can be by MIC module 101 or keyboard input or other methods such as contact based on context, select required literal, and required literal is sent to literal output module 104; Literal output module 104 is used to receive literal, finishes the interface between phonitic entry method and higher layer applications, as literal being offered short message, E-mail etc.
The specific implementation process of phonitic entry method may further comprise the steps: at first, by MIC module 101 received speech signals, voice signal is converted to simulating signal, and this simulating signal is sent to sampling module 102; Secondly, sampling module 102 will be a digital signal from the analog signal conversion of MIC module 101, and this digital signal is sent to sound identification module 103; Once more, sound identification module 103 is according to the digital signal from sampling module 102, as this digital signal is calculated, identify corresponding literal, the literal that identifies may have a plurality of, can select required literal by MIC module 101 or keyboard input, or further select required literal automatically, and required literal is sent to literal output module 104 by the contact between the contextual literal; At last, the literal that literal output module 104 receives from sound identification module 103 is finished the interface between phonitic entry method and higher layer applications, as literal being offered short message, E-mail etc.
Though phonitic entry method is very convenient when input characters, but, because the character library capacity of phonitic entry method is the quantity of literal substantially, therefore character library is very huge, sound identification module in the phonitic entry method need identify correct literal in huge character library, operand increases rapidly with the increase of character library capacity, performance requirement to equipment is higher, usually be applied on the terminals that arithmetic capability is powerful, memory size is big such as PC, and be not suitable for the handheld device that arithmetic capability is lower, memory size is less.
As seen from the above description, phonitic entry method is subjected to the restriction of existing voice recognition technology, and the accuracy of the literal that obtains by phonetic entry is not high, and under especially big in character library and the situation that arithmetic capability is limited, the accuracy of text-to-speech input is lower; Phonitic entry method requires than higher the arithmetic capability of terminal, and the big more requirement to arithmetic capability of character library is high more; Because the character library of phonitic entry method is the character library of Received Pronunciation literal, phonetic entry person's pronunciation but is subjected to the influence of various accents inevitably, in order to improve the accuracy of phonitic entry method, not only require phonetic entry person when the phonetic entry literal, pronunciation will be tried one's best accurately, and require before carrying out phonitic entry method, the importer is carried out voice training at sample, the workload of this voice training capacity same and character library is closely related, when character library is very huge, the workload of voice training will be very big also, so phonitic entry method is not applicable to handheld device.
The spelling input method of literal is many, and the spelling input method with handheld device is that example is introduced spelling input method below.
Fig. 2 realizes synoptic diagram for spelling input method, as shown in Figure 2, realize that the module of spelling input method comprises keyboard load module 201, the female composite module 202 of sound, phonetically similar word are selected module 203 and literal output module 204, wherein, keyboard load module 201 is used to finish the keyboard input of handheld device; The female composite module 202 of sound is used for according to the keyboard input information from keyboard load module 201, generates the female combination of corresponding sound, and selects the female combination of required sound; Phonetically similar word selects module 203 to be used for selecting required literal according to the keyboard input information of keyboard load module 201, and the literal of choosing is sent to literal output module 204; Literal output module 204 is used to receive literal, finishes the interface between phonitic entry method and higher layer applications, as literal being offered short message, E-mail etc.
The specific implementation process of spelling input method may further comprise the steps: at first, by keyboard load module 201 to the female composite module 202 input keyboard input informations of sound, the female composite module 202 of sound generates the female combination of corresponding sound according to the keyboard input information that receives, because the number of keys on the keyboard for hand held set is less, therefore, a button needs corresponding a plurality of different letters, like this, the combination of one group of button just may corresponding a plurality of sound mothers be made up, secondly, by keyboard load module 201 to the female composite module 202 input keyboard input informations of sound, the female composite module 202 of sound is selected the female combination of required sound according to the keyboard input information that receives, be sent to phonetically similar word then and select module 203, once more, because the female combination of one group of sound can corresponding a plurality of phonetically similar words, therefore, select module 203 input keyboard input informations by keyboard load module 201 to phonetically similar word, phonetically similar word selects module 203 to select required literal according to the keyboard input information that receives, and the literal of choosing is sent to literal output module 204, at last, literal output module 204 is selected the literal of module 203 from phonetically similar word, finish the interface between phonitic entry method and higher layer applications, as literal is offered short message, E-mail etc.
As seen from the above description, though the accuracy of spelling input method is very high, but implementation procedure more complicated, required step is many, import a literal, not only need to import a plurality of letters, but also need through carry out the process that female combination selection of sound and phonetically similar word are selected by keyboard, like this, will cause input speed very slow.
Summary of the invention
In view of this, the object of the present invention is to provide a kind of character input method of handheld device, improve the accuracy of literal input, and improve the input speed of literal.The present invention also provides a kind of input device of handheld device.
In order to achieve the above object, the invention provides a kind of character input method of handheld device, the method includes the steps of:
A1, import the voice messaging of initial consonant and simple or compound vowel of a Chinese syllable respectively, the voice messaging of initial consonant and simple or compound vowel of a Chinese syllable is converted to digital signal respectively, identify the female combination of sound according to this digital signal then;
B1, female combination generates phonetically similar word according to described sound, selects required literal in phonetically similar word, carries out literal output.
In phonetically similar word, select required literal to be described in the step B1: in phonetically similar word, to select required literal by keyboard or MIC input.
If identify the female combination of more than one sound according to digital signal, further comprise after the then described steps A 1: select the female combination of required sound.
The present invention also provides a kind of input device of handheld device, and this device comprises at least:
The MIC module is used for receiving respectively the voice signal of initial consonant and simple or compound vowel of a Chinese syllable, the initial consonant that receives and the voice signal of simple or compound vowel of a Chinese syllable is converted to simulating signal respectively, and this simulating signal is sent to sampling module;
Sampling module is used for the initial consonant that will receive and the simulating signal of simple or compound vowel of a Chinese syllable and is converted to digital signal respectively, and this digital signal is sent to the female identification module of sound;
The female identification module of sound is used for identifying the female combination of corresponding sound according to the digital signal of reception, and the female combination of sound is sent to phonetically similar word selection module;
Phonetically similar word is selected module, is used for according to the definite phonetically similar word of the female combination of the sound that receives, and according to the input information that receives, selects required literal;
The literal output module is used to export the literal of reception.
This device further comprises:
The keyboard load module is used for to the female identification module input keyboard of sound input information;
The female identification module of described sound is used for further selecting the female combination of required sound according to the input information that receives in the female combination of the sound that identifies.
Described phonetically similar word selects the input information of module reception from the MIC module.
This device further comprises:
The keyboard load module is used for input and transmits keyboard input information,
Described input information is from the keyboard load module.
Among the present invention, in handheld device, during input characters, phonetic entry and keyboard input are combined, identify related words information by phonetic entry, and then in similar character, select required literal by keyboard input, described phonetic entry can be the voice of initial consonant and simple or compound vowel of a Chinese syllable, also can be the voice of stroke, also can be voice of radical or the like, correspondingly, described Word message is the female combination of sound, also can be stroke combination, also can be combination of radical or the like.According to the proposed method, it is just passable only to need to identify simple word information according to phonetic entry, and sample training is very simple; And the operand of handheld device is smaller, and lower to the arithmetic capability requirement of handheld device, the accuracy of the character input method that the present invention proposes is also just than higher; Simultaneously, only need once to import, therefore improved the input speed of literal greatly by keyboard.Below character input method that the present invention is proposed be called mixed input method.
With the input voice is that the voice of initial consonant and simple or compound vowel of a Chinese syllable are example, among the present invention, only need to identify initial consonant and simple or compound vowel of a Chinese syllable is just passable according to phonetic entry, because the quantity of the relative literal of quantity of initial consonant and simple or compound vowel of a Chinese syllable is much smaller, therefore, required sample size reduces greatly, again in conjunction with the cooperation rule of initial consonant and simple or compound vowel of a Chinese syllable, thereby the accuracy that the sound mother is discerned is improved greatly, and operand again can be very not big, lower to the arithmetic capability requirement of handheld device; The sample that carries out voice training also only needs 21 initial consonants and 39 simple or compound vowel of a Chinese syllable, and do not need each literal in the huge standard character library is all carried out voice training, avoided speech recognition difficulty in the phonitic entry method, the literal correct rate for input is low, computing is complicated, need carry out shortcomings such as voice training to huge character library, is highly suitable for the limited situations of self performance such as handheld device.Owing to only need when selecting phonetically similar word, once import by keyboard, compare with spelling input method etc., simplified input step, the input speed of literal is increased substantially.Voice with stroke or radical are similar as the effect of input voice with the voice with initial consonant and simple or compound vowel of a Chinese syllable as the effect of input voice, therefore, repeat no more.
Description of drawings
Fig. 1 realizes synoptic diagram for phonitic entry method;
Fig. 2 realizes synoptic diagram for spelling input method;
Fig. 3 realizes synoptic diagram for Chinese words input method of the present invention.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, the present invention is described in further detail below in conjunction with accompanying drawing.
Among the present invention, in handheld device, during input characters, phonetic entry and keyboard input are combined, identify related words information by phonetic entry, and then in similar character, select required literal by keyboard input, described phonetic entry can be the voice of initial consonant and simple or compound vowel of a Chinese syllable, also can be the voice of stroke, also can be voice of radical or the like, correspondingly, described Word message is the female combination of sound, also can be stroke combination, also can be combination of radical or the like.
Fig. 3 realizes synoptic diagram for Chinese words input method of the present invention, as shown in Figure 3, the module that realizes character input method of the present invention comprises: MIC module 301, sampling module 302, the female identification module 303 of sound, keyboard load module 304, phonetically similar word are selected module 305 and literal output module 306, wherein, MIC module 301 is used for the initial consonant that will receive and the voice signal of simple or compound vowel of a Chinese syllable is converted to simulating signal, and this simulating signal is sent to sampling module 202; The analog signal conversion that sampling module 202 is used for receiving is a digital signal, and this digital signal is sent to the female identification module 303 of sound; The female identification module 303 of sound is used for identifying the female combination of corresponding sound according to the digital signal of reception, forms a complete phonetic, and the female combination of sound is sent to phonetically similar word selection module 305; Keyboard load module 304 is used for selecting module 305 input keyboard input informations to phonetically similar word; Phonetically similar word is determined in the female combination of sound that phonetically similar word selects module 305 to be used for providing according to the female identification module 303 of sound, and according to the keyboard input information of keyboard load module 304 input, select required literal, and the literal of choosing is sent to literal output module 306; Literal output module 306 is used to receive literal, finishes the interface between phonitic entry method and higher layer applications, as literal being offered short message, E-mail etc.
The specific implementation process of mixed input method may further comprise the steps: at first, receive the voice signal of initial consonant and simple or compound vowel of a Chinese syllable by MIC module 301, the voice signal of initial consonant and simple or compound vowel of a Chinese syllable is converted to simulating signal, for example, input characters " wind ", can import the voice of initial consonant " f " earlier, and then import the voice of simple or compound vowel of a Chinese syllable " eng ", and this simulating signal is sent to sampling module 302; Secondly, sampling module 302 will be digital signal from the analog signal conversion of MIC module 301, and this digital signal is sent to the female identification module 303 of sound; The female identification module 303 of sound identifies the female combination of corresponding sound according to the digital signal of reception, and the female combination of the sound that identifies is sent to phonetically similar word selection module 305; Once more, phonetically similar word selects module 305 to determine phonetically similar word according to the female combination of sound that the female identification module 303 of sound provides, and select module 305 input keyboard input informations to phonetically similar word by keyboard load module 304, phonetically similar word selects module 305 according to the keyboard input information that receives, in the phonetically similar word of determining, select required literal, the literal of choosing is sent to literal output module 306; At last, literal output module 306 is finished the interface between phonitic entry method and higher layer applications, as literal being offered short message, E-mail etc. from the literal of phonetically similar word selection module 305.
In addition, input information at the required literal of phonetically similar word selection module 305 selections also can be from MIC module 301, promptly select module 305 input speech input informations by MIC module 301 to phonetically similar word, phonetically similar word selects module 305 to select required literal according to the speech input information that receives.
If importer's pronunciation is very lack of standardization, then the female identification module 303 of sound can identify the female combination of a plurality of possible sound, at this moment, further, can be by keyboard load module 304 to the female identification module 303 input keyboard input informations of sound, the female identification module 303 of sound is selected the female combination of required sound according to the keyboard input information that receives in the female combination of the possible sound that identifies.
Because the quantity of initial consonant only is 21, the quantity of simple or compound vowel of a Chinese syllable only is 39, compare with 3755 first-level Chinese characters of standard character library, the sample size that needs in the female identification module 303 of sound greatly reduces, again in conjunction with the cooperation rule of initial consonant and simple or compound vowel of a Chinese syllable, thereby the accuracy that 303 couples of sound mothers of the female identification module of sound are discerned improves greatly, and operand again can be very not big, the sample that carries out voice training also only needs 21 initial consonants and 39 simple or compound vowel of a Chinese syllable, and does not need each literal in the huge standard character library is all carried out voice training.
In addition, the voice of the above MIC module 301 input are the voice of initial consonant and simple or compound vowel of a Chinese syllable, in fact, also can be the voice of stroke or radical, like this, female 303 of the identification modules of sound are replaced by stroke identification module or radical identification module, and phonetically similar word is selected 305 of modules to be replaced by similar character and selected module.
Above-described handheld device can be mobile phone, PDA or the like.
In a word, the above is preferred embodiment of the present invention only, is not to be used to limit protection scope of the present invention.
Claims (6)
1. the character input method of a handheld device is characterized in that, the method includes the steps of:
A1, import the voice messaging of initial consonant and simple or compound vowel of a Chinese syllable respectively, the voice messaging of initial consonant and simple or compound vowel of a Chinese syllable is converted to digital signal respectively, identify the female combination of sound according to this digital signal then;
B1, female combination generates phonetically similar word according to described sound, selects required literal in phonetically similar word, carries out literal output.
2. method according to claim 1 is characterized in that, selects required literal to be described in the step B1 in phonetically similar word: select required literal by keyboard or MIC input in phonetically similar word.
3. method according to claim 1 is characterized in that, if identify the female combination of more than one sound according to digital signal, further comprises after the then described steps A 1: select the female combination of required sound.
4. the input device of a handheld device, it is characterized in that this device comprises at least: the MIC module is used for receiving respectively the voice signal of initial consonant and simple or compound vowel of a Chinese syllable, the initial consonant of reception and the voice signal of simple or compound vowel of a Chinese syllable are converted to simulating signal respectively, and this simulating signal is sent to sampling module;
Sampling module is used for the initial consonant that will receive and the simulating signal of simple or compound vowel of a Chinese syllable and is converted to digital signal respectively, and this digital signal is sent to the female identification module of sound;
The female identification module of sound is used for identifying the female combination of corresponding sound according to the digital signal of reception, and the female combination of sound is sent to phonetically similar word selection module;
Phonetically similar word is selected module, is used for according to the definite phonetically similar word of the female combination of the sound that receives, and according to the input information that receives, selects required literal;
The literal output module is used to export the literal of reception.
5. method according to claim 4 is characterized in that, this device further comprises:
The keyboard load module is used for to the female identification module input keyboard of sound input information;
The female identification module of described sound is used for further selecting the female combination of required sound according to the input information that receives in the female combination of the sound that identifies.
6. according to claim 4 or 5 described methods, it is characterized in that described phonetically similar word selects the input information of module reception from the MIC module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100428307A CN100365551C (en) | 2004-05-26 | 2004-05-26 | Words input method and apparatus for hand-held devices |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100428307A CN100365551C (en) | 2004-05-26 | 2004-05-26 | Words input method and apparatus for hand-held devices |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1704877A CN1704877A (en) | 2005-12-07 |
| CN100365551C true CN100365551C (en) | 2008-01-30 |
Family
ID=35577159
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100428307A Active CN100365551C (en) | 2004-05-26 | 2004-05-26 | Words input method and apparatus for hand-held devices |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100365551C (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103294370A (en) * | 2012-03-05 | 2013-09-11 | 北京千橡网景科技发展有限公司 | Method and equipment for triggering keystroke operation |
| CN114356109A (en) * | 2020-09-27 | 2022-04-15 | 华为终端有限公司 | Character input method, electronic device and computer readable storage medium |
| CN114092945B (en) * | 2022-01-21 | 2022-04-19 | 广州文石信息科技有限公司 | Handwriting input method and device based on ink screen |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1357821A (en) * | 2000-12-15 | 2002-07-10 | 丽台科技股份有限公司 | Phonetic input method |
| CN1373406A (en) * | 2001-08-15 | 2002-10-09 | 安平 | Chinese-character 'speech recognition' input method for mobile telephone |
| CN1420421A (en) * | 2001-11-20 | 2003-05-28 | 邵跃波 | Precision voice input method |
-
2004
- 2004-05-26 CN CNB2004100428307A patent/CN100365551C/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1357821A (en) * | 2000-12-15 | 2002-07-10 | 丽台科技股份有限公司 | Phonetic input method |
| CN1373406A (en) * | 2001-08-15 | 2002-10-09 | 安平 | Chinese-character 'speech recognition' input method for mobile telephone |
| CN1420421A (en) * | 2001-11-20 | 2003-05-28 | 邵跃波 | Precision voice input method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1704877A (en) | 2005-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111292720B (en) | Speech synthesis method, device, computer readable medium and electronic equipment | |
| CN111429889B (en) | Method, apparatus, device and computer readable storage medium for real-time speech recognition based on truncated attention | |
| CN109741732B (en) | Named entity recognition method, named entity recognition device, equipment and medium | |
| CN111477216B (en) | Training method and system for voice and meaning understanding model of conversation robot | |
| US20080294433A1 (en) | Automatic Text-Speech Mapping Tool | |
| CN111967224A (en) | Method and device for processing dialog text, electronic equipment and storage medium | |
| EP1139332A3 (en) | Spelling speech recognition apparatus | |
| CN201919034U (en) | Network-based voice prompt system | |
| CN110910903B (en) | Speech emotion recognition method, device, equipment and computer readable storage medium | |
| CN111369971A (en) | Speech synthesis method, speech synthesis device, storage medium and electronic equipment | |
| CN110689876A (en) | Voice recognition method and device, electronic equipment and storage medium | |
| CN111435592B (en) | Voice recognition method and device and terminal equipment | |
| CN101876887A (en) | Voice input method and device | |
| CN110931000B (en) | Method and device for speech recognition | |
| CN112116907A (en) | Speech recognition model establishing method, speech recognition device, speech recognition equipment and medium | |
| CN101377726A (en) | Input method combining speech recognition with stroke recognition and terminal thereof | |
| CN111241853B (en) | Session translation method, device, storage medium and terminal equipment | |
| CN102970618A (en) | Video on demand method based on syllable identification | |
| CN102571882A (en) | Network-based voice reminding method and system | |
| CN100365551C (en) | Words input method and apparatus for hand-held devices | |
| CN106782546A (en) | Audio recognition method and device | |
| CN113486661A (en) | Text understanding method, system, terminal equipment and storage medium | |
| CN112489634A (en) | Language acoustic model training method and device, electronic equipment and computer medium | |
| CN112785667A (en) | Video generation method, device, medium and electronic equipment | |
| CN111414748A (en) | Traffic data processing method and device |
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 |