CN1357821A - Phonetic input method - Google Patents
Phonetic input method Download PDFInfo
- Publication number
- CN1357821A CN1357821A CN 00135351 CN00135351A CN1357821A CN 1357821 A CN1357821 A CN 1357821A CN 00135351 CN00135351 CN 00135351 CN 00135351 A CN00135351 A CN 00135351A CN 1357821 A CN1357821 A CN 1357821A
- Authority
- CN
- China
- Prior art keywords
- phonetic
- special
- input
- those
- sounds
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Document Processing Apparatus (AREA)
Abstract
The phonetic sound input method separates the pronunciation into special letter sounds or decomposed sounds, which are transferred into analog/digital converter via phonetic signal receiver and identified as input letters. It may be combined with keyboard input to input several simple signals. Special pronunciations for producing special words or special symbols may be customized or used to input characters or words for fast selection. The input letters are combined with simple signal to obtain several optional text, so as to select and output correct test. The present invention can simplity identification program without need of high level operation functions in text input.
Description
The present invention relates to a kind of method of phonetic entry, particularly a kind of method of phonetic phonetic entry.
The arriving of information age, various information products are constantly released, and towards aspect researchs such as simple operations easily, utilize the mode of phonetic entry to control computer, assign instruction, input characters, are a kind of methods that has more hommization.Use (Information Appliance in some information, IA) sentence of required input does not need too much on the device, and existing phonetic entry fado is directly to import with a succession of sentence, with speech as unit, utilize the feature of initial consonant, simple or compound vowel of a Chinese syllable to come recognizing voice, yet the discrimination of the literal that is identified but can't reach 100% discrimination in this way, will expend more time for the word that can't discern or speech and just can correctly import, and can't reach the facility of phonitic entry method.
Please refer to Fig. 1, be depicted as the hardware pipeline figure of known a kind of phonitic entry method, voice to pre-amplifier 104 inputs, utilize digital signal panel 106 that speech conversion is become numeral by microphone 102, import in the system 108 that includes processor.
Please refer to Fig. 2, be depicted as the system architecture diagram of known a kind of phonitic entry method, its step is as follows, at first phonetic entry, step 202 is through the language end point detector, voice are cut the audio frame, step 204 is carried out parameter extractor, through step 206 tone identification device and the quick candidate's sound of step 208 continuant table searcher, select several better suited candidate's sounds, output candidate sound table, step 210 is utilized quick candidate word searcher search, step 212 utilize again according to the context frequency of occurrences select the speech device, find out optimal literal, literal is exported.
But, a series of sentence is through after the identification, the discrimination of gained is extremely low, especially the employed language of country of non-English Department such as Chinese is an example with Chinese, and the vocabulary of Chinese has hundreds thousand of, only search for needed speech and just need the time for a long time, similar may have a pile and the speech that finds is close, the error rate that obtains words and phrases at last is very high, resulting recognition effect and good not as expection.Because vocabulary has a lot, add the utilization multivalence of vocabulary again, the function that train the computer self-teaching to proofread and correct is difficult to performance, and therefore the possibility of result that identifies can over and over again make a mistake.
In sum, known have a following shortcoming:
(1) continuous sentence will be splitted into several syllables, tell initial consonant, simple or compound vowel of a Chinese syllable etc. more one by one from syllable, come recognizing voice according to phonetic feature, vocabulary commonly used, contextual continuity more at last, so recognizer bothers very much.
(2) vocabulary of phonitic entry method is a lot, and it not is frequent use that many vocabulary are arranged, even use also because usage, meaning, use-pattern difference, is difficult to train computer to do function from normal moveout correction.
(3) because continuous sentence is not it can be split easily, even after splitting, discerning its initial consonant, simple or compound vowel of a Chinese syllable also is to be difficult to its identification, though recognizer trouble, therefore the ability of identification does not improve, add the zero offset capability of computer and can not bring into play effect very accurately, the discrimination of the phonetic entry of gained is low.
In view of this, the object of the present invention is to provide a kind of phonitic entry method, can use is easily only needing on the information products of the brief sentence of input, utilize the mode of phonetic letter, required identification literal is simply extremely only remained tens recognition units such as letter from hundreds thousand of words, and come sounding, and, make processor easily to discern and why sound in conjunction with the input of keyboard with single special sound such as letter or special speech.
The present invention is to provide a kind of phonetic phonitic entry method exactly, and its step is as follows: import several and decompose sound, express in the phonetic mode of single special sounds such as letter, special speech; Obtain several input alphabets; Or use special pronunciation, to import several special symbols; Sound and special pronunciation are decomposed in the identification input; In conjunction with these input alphabets, to obtain several candidate characters; Select at least one correct literal.Or decomposing sound and special pronunciation obtains switching input method from identification input, several decompose sounds to return input, express in the phonetic mode of single special sounds such as letter, special speech.The required device of the method comprises: a voice signal receiver, an analog/digital converter, a processor and an output device.
The voice that will import are splitted into the decomposition sound of single special sounds such as letter or special speech, or use special pronunciation, to import several special symbols, by the voice signal receiver, import analog/digital converter into, identification is decomposed sound and is become input alphabet, again by processor with input alphabet in conjunction with bunchiness, obtain a plurality of candidate characters, from candidate character, select correct literal, can correct literal be exported by output device at last.Owing to discern and enter as tree-shaped Database Systems from primary letters, discern search database immediately immediately and can reach, removing the problem that can effectively solve identification also increases the speed of identification and the speed of search widely.
From the phonitic entry method of phonetic or basic composition letter is the maximum characteristic of this invention, also will thoroughly and effectively improve the identification bottleneck and the recognition speed of phonetic entry.Be aided with functions such as intelligent study, classification, record, judgement again and with efficient again with lifting.
Step of the present invention also can be: import several and decompose sound, express in the phonetic mode of single special sounds such as letter, special speech; Obtain several input alphabets; Or use special pronunciation, to import several special symbols; Use special key, to import several simple signals; Sound and special pronunciation are decomposed in the identification input; In conjunction with these input alphabets, to obtain several candidate characters; Select at least one correct literal.Or decomposing sound and special pronunciation obtains switching input method from identification input, several decompose sounds to return input, express in the phonetic mode of single special sounds such as letter, special speech.The required device of the method comprises: a voice signal receiver, an analog/digital converter, a keyboard, a processor and an output device.
The voice that will import are splitted into the decomposition sound of single special sounds such as letter or special speech, or use special pronunciation, to import several special symbols, by the voice signal receiver, import analog/digital converter into, identification is decomposed sound and is become input alphabet, and mixes keyboard and use, and imports several simple signals, by processor input alphabet is combined with simple signal again, obtain a plurality of candidate characters, from candidate character, select correct literal, can correct literal be exported by output device at last.
For above-mentioned and other purposes of the present invention, feature and advantage can be become apparent, a preferred embodiment cited below particularly, and conjunction with figs. are described in detail below:
The simple declaration of accompanying drawing:
Fig. 1 is the hardware pipeline figure of known a kind of phonitic entry method;
Fig. 2 is the system architecture diagram of known a kind of phonitic entry method;
Fig. 3 is the system architecture diagram of method of a kind of phonetic phonetic entry of a preferred embodiment of the present invention;
Fig. 4 is the hardware pipeline figure of method of a kind of phonetic phonetic entry of a preferred embodiment of the present invention; And
Fig. 5 is according to a preferred embodiment of the present invention, the use process flow diagram of the method for a kind of phonetic phonetic entry of phonetic entry under English mode.
Please refer to Fig. 3, the system architecture diagram of the method for a kind of phonetic phonetic entry of a preferred embodiment of the present invention that shown is.
The step of the method for phonetic phonetic entry is as follows: at first several decompose sound to step 302 for input, express in the phonetic mode of single special sounds such as letter, special speech; Obtain several input alphabets to obtain step 304; Or step 308 is for using special pronunciation, to import several special symbols; Next be that sound and special pronunciation are decomposed in step 310 identification input, the efficient that these can promote input from special pronunciations of ordering, and can judge whether to make up primary letters or syllable to finish a standard word; Step 314 is in conjunction with these input alphabets, to obtain several candidate characters; Final step 316 is selected at least one correct literal.Or these are from the changeable input pattern of ordering of special pronunciation, obtain step 312 from step 310 and switch input method, and several decompose sound to return step 302 input, express in the phonetic mode of single special sounds such as letter, special speech.
In addition because spelling such as many names or place name or phonetic are fixing, and it is essential utilizing abundant dictionary and obtaining correct speech, this also is that this invention utilizes intelligent data character library dictionary to be aided with intelligent input again to select to judge and reach.
With " Taibei " is example, the standard English in the Taibei is TAIPEI, if use the Chinese phonetic alphabet of Chinese, at first the voice " Taibei " that will import being combined into the decomposition sound of expressing with single special sounds such as letter, the phonetic mode of special speech is " T; A; I " representative " platform " and " B, E, I " representative " north "; Next discern these decomposition sounds " T, A, I, B, E, I " and become several input alphabets; " TAI " BEI " obtains candidate character " platform " and a plurality of candidate characters such as " north ", selects correct literal " platform " and " north " in conjunction with input alphabet one-tenth.But, user's input " T, A, I, P, E, I " becomes several input alphabets if next discerning these decomposition sounds " T, A, I, P, E, I "; Become " TAI " " PEI " in conjunction with input alphabet, obtain candidate character " platform " and a plurality of candidate characters such as " accompanying ", do not conform to user's expection, prev word or preceding numeral and the current word of importing are combined into the words and phrases phrase of a complete or part and are judged that will to list words and phrases such as " Taibei " after the selection selective via intelligent determination methods, and the user can select correct literal " Taibei ".The identification of this hybrid standard phonetic and usage or special proper noun also for this reason one the invention a big characteristic.
The step of the method for phonetic phonetic entry also can add keyboard to be used, and makes step be: at first several decompose sound to step 302 for input, express in the phonetic mode of single special sounds such as letter, special speech; Obtain several input alphabets to obtain step 304; Step 306 is for using special key, to import several simple signals; Or step 308 is for using special pronunciation, to import several special symbols; Next be that sound and special pronunciation and special key are decomposed in step 310 identification input, the efficient that these can promote input from the special pronunciation of ordering and special key, and can judge whether to make up primary letters or syllable to finish a standard word; Step 314 is in conjunction with these input alphabets, to obtain several candidate characters; Final step 316 is selected at least one correct literal.Or these are from special pronunciation and the changeable input pattern of special key ordered, obtain step 312 from step 310 and switch input method, and several decompose sound to return step 302 input, express in the phonetic mode of single special sounds such as letter, special speech.
State's word phonetic notation phonetic with Chinese is example, at first the voice " Taibei " that will import are combined into the decomposition sound of expressing in the phonetic mode of single special sounds such as letter, special speech for " ㄊ, ㄞ, ', ㄅ, ㄟ, ˇ "; Next discern these decompose sounds " ㄊ, ㄞ, ', ㄅ, ㄟ, ˇ " become several input alphabets; Can mix keyboard in the step uses, " platform " as the Taibei is two in phonetic notation phonetic, then can use numerical key " 2 " to substitute, and " north " in the Taibei is three in phonetic notation phonetic, then can use numerical key " 3 " to substitute, keyboard that can also other kind of mode uses, and input alphabet and button are combined into " " ㄊ, ㄞ, 2, ㄅ, ㄟ, 3 "; obtain candidate character a plurality of candidate characters such as " Taibei "; select correct literal " Taibei ", also can use from ordering or default special key or symbol decide the modulation in tone of word and word space or decision pronunciation, as being two in phonetic notation phonetic with use numerical key " 2 " expression " platform ".
Please refer to Fig. 4, the hardware pipeline figure of the method for a kind of phonetic phonetic entry of a preferred embodiment of the present invention that shown is.
The decomposition sound disassembled in literal, with single special sounds such as letters, the phonetic mode of special speech or use special pronunciation, or use special pronunciation, to import several special symbols, by 402 inputs of voice signal receiver, this voice signal receiver 402 can be microphone, utilize analog/digital converter 404 that speech conversion is become numeral, make that decomposing sound is identified as input alphabet, if needed, also can mix keyboard 406 and use, key in simple signal, import in the processor 408 with the input alphabet combination and with the simple signal of keying in, wherein processor 408 can be host computer, microcontrollers etc. produce a plurality of candidate characters, select at least one correct literal again, at last can be by output device 410 outputs, wherein output device 410 can be PDA, IA, information products such as mobile phone.
A big bottleneck really of mobile phone input now, complicated input mode dare not be flattered the people, though new-type PDA has the right writing style of pen type handwriting input and simplified Chinese character, the complex form of Chinese characters, reaches other fonts or the like, complicated situation is from being nothing difficult, make things convenient for the user for asking, effectively phonetic entry will be best input method, and through one inventing and will do revolutionary contribution to input thus.
Please refer to Fig. 5, the use process flow diagram of the method for a kind of phonetic phonetic entry of a preferred embodiment of the present invention that shown is.At first step 502 resolves into several letters with first individual character and word for word reads; Next step 504 input first steering order makes it export first space or first special symbol; Get off is that step 506 resolves into several letters with second individual character and word for word reads again; Step 508 makes it import second space or one second special symbol for input second steering order again; Final step 510 repeats above-mentioned steps, up to the input of finishing words and expressions.Wherein one of them is assigned in order to utilize special pronunciation and special key for first steering order and second steering order.
In addition, though English phonetic entry can reach 80% discrimination, but still can't reach hundred-percent accuracy rate, similarly be when using general phonitic entry method, directly read out the sound of " the world ", through after the speech recognition, and the result is exported, but may be because problems such as the relation of similar sound or the cacoepies of sounder own, and " the word " this word appears, once wrong literal occurs, then need use the mode of the traditional keyboard input word that corrects mistakes, and often not have to be equipped with the keyboard that comprises 26 English alphabets on communication product, be example with present general mobile phone, 26 English alphabets are imported by the repeated presses numerical key entirely, utilize by numerical key " 8 " an input " t " of getting off, by numerical key " 4 " two input " h " of getting off, again by numerical key " 3 " two input " e " of getting off, so far the beginning is finished the literal input of " the ", and here we can see that has certain degree inconvenience in the use.Method provided by the invention is under English input pattern, system only needs to discern 26 English alphabets, with and/or represent a few special pronunciation of special symbol, difference between each letter is quite big, therefore do not worry that the result understands the letter of output error, with input " the world " is example, say respectively earlier " t ", " h ", " e ", pause between the English-word can be used special pronunciation, or cooperate special key to come input control order, say " w " afterwards more respectively, " o ", " r ", " l ", " d ", so can correctly discern and this word of array output " the world ", repeat above-mentioned step, just can finish the input of words and expressions, so word for word read the method that English alphabet is edited words and expressions, discrimination almost can reach absolutely.Yet those skilled in the art should know, the present invention is not limited in Chinese and English phonetic entry, every can should be still in spiritual scope of the present invention with the language that the phonetic mode is expressed.
The present invention not only can discern the place name of some specific mosaics or phonetic by intelligent character library dictionary, name or the like, and select these correct literal or words and phrases; The efficient that also can be further can promote input by the special key of ordering certainly or special pronunciation; More can judge whether to make up primary letters or syllable to finish a standard word by the special key of ordering certainly or special pronunciation, also only input alphabet such as ABC etc., Chinese occurs most probably simultaneously with English words after all, and the pattern of switching input method by special pronunciation or button is input in Chinese or English input or other literary composition inputs.
So use the present invention and have following advantage:
When (1) the present invention imports voice,, do not need complicated identification skill and flow process, can shorten its recognition time to use the phonetic of simplifying most.
(2) the recognition unit number of identification required for the present invention is few, does not need the processor of high calculation function, even therefore the processor of low calculation function also can use.
(3) the recognition unit number of identification required for the present invention is few, often uses down, and the self-teaching of processor can be proofreaied and correct correctly.
By above-mentioned advantages, can make discrimination of the present invention improve greatly, the voice correctness of being imported also raises greatly; Compared to the present invention, though it is very fast that known audio recognition method is imported the speed of complete words and expressions, need utilize traditional keyboard entry method to get back to wrong individual character when but mistake occurring with cursor, the deletion error word is imported correct word again and is revised, confirm so repeatedly to revise to expend many times, and the present invention directly carries out phonetic entry in the mode of phonetic, the situation that need not worry identification error takes place, therefore be applied in when importing abbreviated message on the IA product, more can highlight its convenience.
Though the present invention with a preferred embodiment openly as above; right its is not in order to limit the present invention; any those skilled in the art; without departing from the spirit and scope of the present invention; when can being used for a variety of modifications and variations, so protection scope of the present invention is as the criterion when looking the appended claims person of defining.
Claims (16)
1. the method for a phonetic phonetic entry, its step is as follows:
Import a plurality of decomposition sounds, those decompose sound and express in the phonetic mode of single special sounds such as letter, special speech;
Obtain a plurality of input alphabets;
Use a plurality of special pronunciations, to import a plurality of special symbols;
Discern those and decompose sound and those special pronunciations;
In conjunction with those input alphabets, to obtain a plurality of candidate characters and switching input method either-or; And
Select at least one correct literal.
2. the method for phonetic phonetic entry as claimed in claim 1 is wherein selected step to switch the input rule and is got back to those decomposition sounds of input, and those decompose sounds and express in the phonetic mode of single special sounds such as letter, special speech.
3. the method for phonetic phonetic entry as claimed in claim 1, wherein the required device of the method for this phonetic phonetic entry comprises: a voice signal receiver, an analog/digital converter, a processor and an output device.
4. the method for phonetic phonetic entry as claimed in claim 1 can be discerned place name, name of a plurality of specific mosaics and phonetic or the like by an intelligent character library dictionary, and select this correct literal.
5. the method for phonetic phonetic entry as claimed in claim 1, the efficient that can promote input by those special pronunciations of ordering certainly.
6. the method for phonetic phonetic entry as claimed in claim 5 wherein by those special pronunciations of ordering certainly, judges whether to make up primary letters or syllable to finish a standard word.
7. the method for phonetic phonetic entry as claimed in claim 6 is wherein switched an input pattern by this special pronunciation.
8. the method for a phonetic phonetic entry, its step is as follows:
Import a plurality of decomposition sounds, those decompose sound and express in the phonetic mode of single special sounds such as letter, special speech;
Obtain a plurality of input alphabets;
Use a plurality of special pronunciations, to import a plurality of special symbols;
Use a plurality of special key, to import a plurality of simple signals;
Discern those and decompose sound, those special pronunciations and those special key;
In conjunction with those input alphabets, to obtain a plurality of candidate characters and switching input method either-or; And
Select at least one correct literal.
9. the method for phonetic phonetic entry as claimed in claim 8 is wherein selected step to switch the input rule and is got back to those decomposition sounds of input, and those decompose sounds and express in the phonetic mode of single special sounds such as letter, special speech.
10. the method for phonetic phonetic entry as claimed in claim 8, wherein the required device of the method for this phonetic phonetic entry comprises: a voice signal receiver, a keyboard, an analog/digital converter, a processor and an output device.
11. the method for phonetic phonetic entry as claimed in claim 8 can be discerned place name, name of a plurality of specific mosaics and phonetic or the like by an intelligent character library dictionary, and select this correct literal.
12. the method for phonetic phonetic entry as claimed in claim 8, the efficient that can promote input by a plurality of special key of ordering certainly and a plurality of special pronunciation.
13. the method for phonetic phonetic entry as claimed in claim 12 wherein by those special key of ordering certainly and those special pronunciation either-ors, judges whether to make up primary letters or syllable to finish a standard word.
14. the method for phonetic phonetic entry as claimed in claim 13 is wherein switched an input pattern by this special pronunciation and this button either-or.
15. the method for a voice phonetic input comprises the following steps:
One first individual character is resolved into a plurality of letters word for word to be read;
Import one first steering order, make it import one first space and one first special symbol either-or;
One second individual character is resolved into a plurality of letters word for word to be read;
Import one second steering order, make it import second space and one second special symbol either-or; And
Repeat above-mentioned those steps, up to the input of finishing words and expressions.
16. the method for voice phonetic input as claimed in claim 15, wherein this first steering order and this second steering order utilize a special pronunciation and a special key either-or to assign.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 00135351 CN1357821A (en) | 2000-12-15 | 2000-12-15 | Phonetic input method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 00135351 CN1357821A (en) | 2000-12-15 | 2000-12-15 | Phonetic input method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN1357821A true CN1357821A (en) | 2002-07-10 |
Family
ID=4596671
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 00135351 Pending CN1357821A (en) | 2000-12-15 | 2000-12-15 | Phonetic input method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1357821A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100365551C (en) * | 2004-05-26 | 2008-01-30 | 华为技术有限公司 | Words input method and apparatus for hand-held devices |
| CN101067780B (en) * | 2007-06-21 | 2010-06-02 | 腾讯科技（深圳）有限公司 | Character inputting system and method for intelligent equipment |
| CN103595852A (en) * | 2012-08-14 | 2014-02-19 | 中兴通讯股份有限公司 | A voice auxiliary input method and a voice auxiliary input apparatus |
| CN105096945A (en) * | 2015-08-31 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Voice recognition method and voice recognition device for terminal |
| CN113743053A (en) * | 2021-08-17 | 2021-12-03 | 上海明略人工智能(集团)有限公司 | Alphabet vector calculation method, system, storage medium and electronic device |
-
2000
- 2000-12-15 CN CN 00135351 patent/CN1357821A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100365551C (en) * | 2004-05-26 | 2008-01-30 | 华为技术有限公司 | Words input method and apparatus for hand-held devices |
| CN101067780B (en) * | 2007-06-21 | 2010-06-02 | 腾讯科技（深圳）有限公司 | Character inputting system and method for intelligent equipment |
| CN103595852A (en) * | 2012-08-14 | 2014-02-19 | 中兴通讯股份有限公司 | A voice auxiliary input method and a voice auxiliary input apparatus |
| CN105096945A (en) * | 2015-08-31 | 2015-11-25 | 百度在线网络技术（北京）有限公司 | Voice recognition method and voice recognition device for terminal |
| CN113743053A (en) * | 2021-08-17 | 2021-12-03 | 上海明略人工智能(集团)有限公司 | Alphabet vector calculation method, system, storage medium and electronic device |
| CN113743053B (en) * | 2021-08-17 | 2024-03-12 | 上海明略人工智能(集团)有限公司 | Letter vector calculation method, system, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US6490563B2 (en) | Proofreading with text to speech feedback | |
| JP4829901B2 (en) | Method and apparatus for confirming manually entered indeterminate text input using speech input | |
| TWI293455B (en) | System and method for disambiguating phonetic input | |
| US5787230A (en) | System and method of intelligent Mandarin speech input for Chinese computers | |
| JP4734155B2 (en) | Speech recognition apparatus, speech recognition method, and speech recognition program | |
| US7848926B2 (en) | System, method, and program for correcting misrecognized spoken words by selecting appropriate correction word from one or more competitive words | |
| CN101067780B (en) | Character inputting system and method for intelligent equipment | |
| KR100769029B1 (en) | Method and system for voice recognition of names in multiple languages | |
| US6876967B2 (en) | Speech complementing apparatus, method and recording medium | |
| CN101133411A (en) | Fault-tolerant romanized input method for non-roman characters | |
| JP5703491B2 (en) | Language model / speech recognition dictionary creation device and information processing device using language model / speech recognition dictionary created thereby | |
| JPH10503033A (en) | Speech recognition method and device based on new word modeling | |
| CA2613154A1 (en) | Dictionary lookup for mobile devices using spelling recognition | |
| US20070038456A1 (en) | Text inputting device and method employing combination of associated character input method and automatic speech recognition method | |
| CN1359514A (en) | Multimodal data input device | |
| CN100561411C (en) | The input device of combining association formula input method and sound identification method and method | |
| CN1357821A (en) | Phonetic input method | |
| KR100599873B1 (en) | device for input various characters of countrys using hangul letters and method thereof | |
| JPH08166966A (en) | Dictionary retrieval device, database device, character recognizing device, speech recognition device and sentence correction device | |
| CN1965349A (en) | Multimodal disambiguation of speech recognition | |
| JP2002189490A (en) | Method of pinyin speech input | |
| JP4622861B2 (en) | Voice input system, voice input method, and voice input program | |
| JP2004309928A (en) | Speech recognition device, electronic dictionary device, speech recognizing method, retrieving method, and program | |
| JP2008243076A (en) | Interpretation device, method and program | |
| JP2007535692A (en) | System and method for computer recognition and interpretation of arbitrarily spoken characters |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C06 | Publication | ||
| PB01 | Publication | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |