CN103219007A - Voice recognition method and voice recognition device - Google Patents

Voice recognition method and voice recognition device Download PDF

Info

Publication number
CN103219007A
CN103219007A CN2013101020652A CN201310102065A CN103219007A CN 103219007 A CN103219007 A CN 103219007A CN 2013101020652 A CN2013101020652 A CN 2013101020652A CN 201310102065 A CN201310102065 A CN 201310102065A CN 103219007 A CN103219007 A CN 103219007A
Authority
CN
China
Prior art keywords
character information
information
numerical information
numerical
correspondence
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
Application number
CN2013101020652A
Other languages
Chinese (zh)
Inventor
谢东来
李宏图
宋吉江
官巍
王健
马泽威
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN2013101020652A priority Critical patent/CN103219007A/en
Publication of CN103219007A publication Critical patent/CN103219007A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Machine Translation (AREA)

Abstract

The invention discloses a voice recognition method and a voice recognition device, and belongs to the field of voice recognition. The method comprises the following steps of recognizing the received voice information as character information; judging whether digital information corresponding to the character information exists in preset grammar rules or not; and obtaining recognition result according to the digital information when judging that the digital information corresponding to the character information exists in the preset grammar rules. The device comprises a recognition module, a first judgment module and a processing module. The process of recognizing natural language of human being by a machine is realized, the training work of flight command can be carried out by a flight commander, and a computer-generated intelligent airplane is controlled according to commands recognized by the machine; and the voice recognition method can be suitable for different speakers, so that voice recognition can be finished by different people through the voice recognition method.

Description

Audio recognition method and device
Technical field
The present invention relates to field of speech recognition, particularly a kind of audio recognition method and device.
Background technology
Along with the sustainable development of computing machine and infotech, interactive voice has become the necessary means of man-machine interaction.As one of important technology of interactive voice, speech recognition technology reaches its maturity, and is widely used through the development of nearly half a century.Speech recognition technology is the technology that machine changes voice signal into by identification and understanding process corresponding text or order.
Existing speech recognition technology is generally the text identification technology, mates by using common dictionary, and the content of common dictionary is many, and having a large vocabulary of covering is applicable to the dictation at random that does not have specific objective.
In realizing process of the present invention, the inventor finds that there is following problem at least in prior art: the text identification technology causes accuracy of identification to reduce owing to have a large vocabulary, and recognition speed is slow.
Summary of the invention
In order to solve prior art problems, the embodiment of the invention provides a kind of audio recognition method and device.Described technical scheme is as follows:
On the one hand, provide a kind of audio recognition method, described method comprises:
The voice messaging that receives is identified as character information;
Judge the numerical information that whether has described character information correspondence in the preset grammatical rule;
When judging described preset grammatical rule and have the numerical information of described character information correspondence, obtain recognition result according to described numerical information.
Wherein, describedly judge that before the numerical information that whether has described character information correspondence in the preset grammatical rule, described method also comprises:
Judge whether described character information is the command character information of presetting;
When judging described character information, carry out and judge the operation that whether has the numerical information of described character information correspondence in the preset grammatical rule for default command character information.
Wherein, describedly judge that whether described character information is the command character information of presetting, and comprising:
Described character information is split as a plurality of single characters;
Judge one by one whether described a plurality of single character is included in the default command character information;
When described a plurality of single characters all were included in the default command character information, then judging described character information was the command character information of presetting.
Further, described method also comprises:
When a certain single character in described a plurality of single characters was not included in the default command character information, then judging described character information was not the command character information of presetting.
Wherein, the corresponding relation of store character information and numerical information in the described preset grammatical rule is describedly judged the numerical information that whether has described character information correspondence in the preset grammatical rule, comprising:
In the corresponding relation of the character information of described syntax rule storage and numerical information, search the numerical information whether described character information has correspondence;
If find corresponding numerical information, then judge the numerical information that has described character information correspondence in the described preset grammatical rule;
If search numerical information, then judge the numerical information that does not have described character information correspondence in the described preset grammatical rule less than correspondence.
Wherein, obtain recognition result according to described numerical information, comprising: according to described numerical information, obtain character information from described syntax rule, the described character information that obtains comprises subject information and/or action message:
The described character information that obtains is defined as recognition result.
Wherein, described preset grammatical rule is kept in the expandable mark language XML file.
On the other hand, provide a kind of speech recognition equipment, described device comprises:
Identification module, the voice messaging that is used for receiving is identified as character information;
First judge module is used for judging whether the preset grammatical rule exists the numerical information of described character information correspondence;
Processing module is used for obtaining recognition result according to described numerical information when judging described preset grammatical rule and have the numerical information of described character information correspondence.
Wherein, described device also comprises second judge module, be used to judge whether described character information is the command character information of presetting, when judging described character information, carry out and judge the operation that whether has the numerical information of described character information correspondence in the preset grammatical rule for default command character information.
Wherein, described judging unit comprises:
Split subelement, be used for described character information is split as a plurality of single characters;
Judgment sub-unit is used for judging one by one whether described a plurality of single character is included in default command character information;
The judged result subelement is used for when described a plurality of single characters all are included in default command character information, and then judging described character information is the command character information of presetting.
Further, described judged result subelement also is used for: when a certain single character of described a plurality of single characters was not included in the default command character information, then judging described character information was not the command character information of presetting.
Wherein, the corresponding relation of store character information and numerical information in the described preset grammatical rule, described first judge module comprises:
Whether search the unit, be used at the character information of described syntax rule storage and the corresponding relation of numerical information, searching described character information has corresponding numerical information;
The judged result unit, be used for when the described result who searches the unit when being, judge the numerical information that has described character information correspondence in the described preset grammatical rule; When the described result who searches the unit for not the time, judge the numerical information that does not have described character information correspondence in the described preset grammatical rule.
Wherein, described processing module comprises acquiring unit and determining unit;
Described acquiring unit, be used for when judging described preset grammatical rule and have the numerical information of described character information correspondence, according to described numerical information, from described syntax rule, obtain character information, the described character information that obtains comprises subject information and/or action message;
Described determining unit is used for the described character information that obtains is defined as recognition result.
Wherein, described preset grammatical rule is kept in the expandable mark language XML file.
The beneficial effect that the technical scheme that the embodiment of the invention provides is brought is:
By voice messaging is identified as character information, and whether there is the numerical information of this character information correspondence according to the preset grammatical rule judgment, when having corresponding numerical information, to be defined as recognition result according to the character information that numerical information gets access to, get rid of there not being the character information of corresponding numerical information, realized filtration treatment to the character information that identifies, can be according to the recognition result that the character information after filtering obtains by machine recognition.The recognition result that the aircraft controller can obtain according to this speech recognition equipment, the intelligent aircraft that control computer generates is finished the training work of commander's flight.
Description of drawings
In order to be illustrated more clearly in the technical scheme in the embodiment of the invention, the accompanying drawing of required use is done to introduce simply in will describing embodiment below, apparently, accompanying drawing in describing below only is some embodiments of the present invention, for those of ordinary skills, under the prerequisite of not paying creative work, can also obtain other accompanying drawing according to these accompanying drawings.
Fig. 1 is the speech recognition technology synoptic diagram that provides in the prior art;
Fig. 2 is the audio recognition method process flow diagram that the embodiment of the invention 1 provides;
Fig. 3 is the audio recognition method process flow diagram that the embodiment of the invention 2 provides;
Fig. 4 is the speech recognition equipment structural representation that the embodiment of the invention 3 provides;
Fig. 5 is another structural representation of the speech recognition equipment that provides of the embodiment of the invention 3.
Embodiment
For making the purpose, technical solutions and advantages of the present invention clearer, embodiment of the present invention is described further in detail below in conjunction with accompanying drawing.
Existing speech recognition technology generally is based on statistical model and discerns, and as shown in Figure 1, speech recognition technology can be divided into three parts, at first extracts phonetic feature, extracts time dependent phonetic feature sequence from speech waveform.Extract before the phonetic feature sequence, preferably can do front-end processing, as eliminating noise etc., thereby be convenient to accurately extract phonetic feature speech waveform; Be identifying then, by learning algorithm phonetic feature learnt to produce acoustic model, the phonetic feature that said extracted is gone out compares in acoustic model and mates, and obtains best recognition result; By language model above-mentioned recognition result is carried out Language Processing again and finish identification, comprised the grammer network that constitutes by voice command recognition or statistical method in the language model, can carry out speech processes such as syntax and semantics analysis by language model above-mentioned recognition result.
The embodiment of the invention provides a kind of audio recognition method, use developer oneself to write XML(Extensible Markup Language, extend markup language) file is as dictionary, and this XML file vocabulary is pointed, and compares with the common dictionary that contains each technical field, because the file vocabulary is few, shortened greatly and mated the required time, improved recognition speed, also improved accuracy of identification simultaneously, generally be used as the identification of brief, be convenient to user's operation.
Embodiment 1
Referring to Fig. 2, the embodiment of the invention provides a kind of audio recognition method, and this method comprises:
101: the voice messaging that receives is identified as character information;
102: judge whether the preset grammatical rule exists the numerical information of character information correspondence;
103: when judging the preset grammatical rule and have the numerical information of character information correspondence, obtain recognition result according to numerical information.
The method that the embodiment of the invention provides, by voice messaging is identified as character information, and whether there is the numerical information of this character information correspondence according to the preset grammatical rule judgment, when having corresponding numerical information, to be defined as recognition result according to the character information that numerical information gets access to, get rid of there not being the character information of corresponding numerical information, realized filtration treatment to the character information that identifies, can be according to the recognition result that the character information after filtering obtains by machine recognition.The recognition result that the aircraft controller can obtain according to this speech recognition equipment, the intelligent aircraft that control computer generates is finished the training work of commander's flight.
Embodiment 2
Referring to Fig. 3, the embodiment of the invention provides a kind of audio recognition method, and this method comprises:
201: receiving speech information;
Wherein, the execution subject of present embodiment is a speech recognition equipment, can trigger the reception that this speech recognition equipment carries out voice messaging by the user, also can initiatively receive the voice messaging that the user sends by speech recognition equipment.For example, receive window can be set on the speech recognition equipment or receive button, the user triggers this receive window or receives button, after speech recognition equipment receives user's trigger action, begins to receive the voice messaging that the user sends; Also the voice recognition module can be set on the speech recognition equipment, recognize sound when sending, receive this sound, be convenient to further this sound be discerned.
202: the voice messaging that receives is identified as character information;
Particularly, the recognition methods that the voice messaging that receives is identified as character information can have a variety of, in the embodiment of the invention, by Microsoft's voice kit 5.1, is called for short SAPI5, and the voice messaging that receives is identified as character information.
Wherein, SAPI5 comprises text voice transformational structure and speech recognition structure, and the text voice transformational structure synthesizes audio stream with text-string or file, and the speech recognition structure is identified as readable text-string or file with audio stream.The embodiment of the invention is utilized the speech recognition structure of SAPI5, and the voice messaging that receives is identified as character information.
203: judge whether character information is the command character information of presetting, when character information is the command character information of presetting, execution in step 204; When character information is not the command character information of presetting, execution in step 201;
Whether wherein, this step is priority service, be the command character information of presetting by judging character information, undesirable character information is no longer carried out the operation of back, can shorten the required time of follow-up coupling greatly, improve recognition speed, also improve accuracy of identification simultaneously.
Default command character information is the Command order in the embodiment of the invention, that is to say to judge whether character information is the Command order, particularly:
At first, character information is split as a plurality of single characters, the order of a plurality of single characters after the fractionation can be random order, can ignore the sequencing of single character like this;
Judge one by one whether each the single character in this character information is included in the default command character information;
When judging single character and be included in the default command character information, then continue to judge whether next single character is included in the default command character information, when judging all single characters that this character information comprises and all be included in the default command character information, then drawing this character information is the command character information of presetting;
When judging a certain single character and be not included in the default command character information, then drawing this character information is not the command character information of presetting.
That is to say, when present embodiment is judged character information and is not the command character information of presetting, then the identifying operation of this voice messaging is finished, continue to receive next voice messaging, speech recognition is carried out in the operation of the next voice messaging repeating step 201 that receives.
204: judge the numerical information that whether has the character information correspondence in the preset grammatical rule, if exist, then execution in step 205; If do not exist, then step finishes;
In the present embodiment, after speech recognition equipment was discerned successfully to voice messaging, at first the form of expression was a character information, according to syntax rule this character information is changed again, obtain the numerical information of character information correspondence after the conversion, this step can be dwindled identification range, improves the success ratio of identification.
In this step, store the corresponding relation of character information and numerical information in the syntax rule in advance, preferably, present embodiment can define syntax rule in the XML file, thereby judge whether to exist the numerical information of character information correspondence according to the syntax rule in the XML file, and then handle it and judge, the operation below carrying out according to the meaning of numerical information.
Wherein, judge the numerical information that whether has the character information correspondence in the preset grammatical rule, comprising:
In the corresponding relation of the character information of syntax rule storage and numerical information, search the numerical information whether character information has correspondence; If find corresponding numerical information, then judge the numerical information that has the character information correspondence in the preset grammatical rule; If search numerical information, then judge the numerical information that does not have the character information correspondence in the preset grammatical rule less than correspondence.
The vocabulary that comprises in the syntax rule of embodiment of the invention setting is few, syntax rule can be kept in the XML file, as being kept in the Command.xml file, because the vocabulary that comprises in the syntax rule is less, can shorten the required time of retrieval in the matching process greatly, improve recognition speed, also can improve accuracy of identification simultaneously, be applicable to identification for brief order voice messaging.For example, the embodiment of the invention can be preset syntax rule as follows, but the embodiment of the invention not with this as qualification.
Figure BDA00002974317500071
Figure BDA00002974317500081
Figure BDA00002974317500091
Wherein, the character information of storing in the syntax rule and the corresponding relation of numerical information can be for shown in as shown in table 1 or the tables 2, and the corresponding relation of character information and numerical information is as follows:
Table 1
Character information Numerical information
Waist moves waist 1
Waist moving two 2
Waist moving three 3
Waist moving four 4
Waist moving five 5
Waist moving six 6
Waist is moving to turn 7
Waist moving eight 8
Waist moving nine 9
…… ……
Table 2
Character information Numerical information
Can drive 1
Can skid off 2
Can advance runway 3
Can take off 4
Can go No. one 5
Can withdraw from 6
…… ……
Wherein, for the numerical information in the table 2 " 5 ", corresponding characters information can also be expressed as " can add a bit " or " can add No. one ", and the concrete expression form of character information can preestablish, and present embodiment does not limit.
Particularly,, then judge the numerical information that has this character information correspondence in this syntax rule if in the syntax rule when having stored the numerical information corresponding with this character information, otherwise, there is not the numerical information of this character information correspondence.For example, if the character information that step 202 identification obtains is " can skid off ", can judge the numerical information " 2 " that has this character information correspondence according to table 1; If the character information that step 202 identification obtains is " can draw ", then can judge the numerical information that does not have this character information correspondence according to table 1, identifying finishes, like this, the character information that identifies in the step 202 can be excluded a part, can dwindle identification range, improve the success ratio of identification.
205: obtain this numerical information corresponding characters information according to syntax rule, and the character information that obtains is defined as recognition result.
According to the default numerical information and the corresponding relation of character information, can obtain numerical information corresponding characters information, thereby with the character information that obtains as recognition result.For example, the numerical information that above-mentioned steps is obtained is " 1 ", and when default corresponding relation was table 1, recognition result was " waist moves waist "; The numerical information that above-mentioned steps is obtained is " 4 ", and when default corresponding relation was table 2, recognition result was " can take off ".
Preferably, the character information of the embodiment of the invention comprises subject information and/or action message, the character information that obtains is defined as recognition result is specially: when character information includes only subject information, this subject information is defined as recognition result; When character information includes only action message, this subject information is defined as recognition result; When character information comprises subject information and action message, this subject information and action message are made up, obtain recognition result.
In the present embodiment, subject information is Pilot information, and action message is a Com information, and wherein Pilot information is meant airplane information, and Com information is meant the action message of aircraft.
For example: the character information that step 204 identification obtains is " waist moves waist " and " can take off ", the corresponding relation of character information and numerical information is table 1 and table 2 in the syntax rule, so, in syntax rule, judge character information " waist moves waist " and have corresponding numerical information " 1 ", " can take off " and have corresponding numerical information " 4 ", so, subject information and action message are made up, the recognition result that obtains is " the moving waist of waist can take off ", just can obtain pilot commander's password according to this recognition result.
The method that the embodiment of the invention provides, by voice messaging is identified as character information, and whether there is the numerical information of this character information correspondence according to the preset grammatical rule judgment, when having corresponding numerical information, to be defined as recognition result according to the character information that numerical information gets access to, get rid of there not being the character information of corresponding numerical information, realized filtration treatment to the character information that identifies, can be according to the recognition result that the character information after filtering obtains by machine recognition.The recognition result that the aircraft controller can obtain according to this speech recognition equipment, the intelligent aircraft that control computer generates is finished the training work of commander's flight.
Embodiment 3
Referring to Fig. 4, the embodiment of the invention provides a kind of speech recognition equipment, and described device comprises:
Identification module 301, the voice messaging that is used for receiving is identified as character information;
First judge module 302 is used for judging whether the preset grammatical rule exists the numerical information of described character information correspondence;
Processing module 303 is used for obtaining recognition result according to described numerical information when judging described preset grammatical rule and have the numerical information of described character information correspondence.
Wherein, referring to Fig. 5, described device also comprises second judge module 304, comprising:
Judging unit 3041 is used to judge whether character information is the command character information of presetting;
Performance element 3042 is used for carrying out and judging the operation that whether has the numerical information of character information correspondence in the preset grammatical rule when judgment unit judges goes out character information for default command character information.
Wherein, judging unit 3041 comprises:
Split subelement, be used for character information is split as a plurality of single characters;
Judgment sub-unit is used for judging one by one whether a plurality of single characters are included in default command character information;
The judged result subelement is used for when a plurality of single characters all are included in default command character information, and then judging character information is the command character information of presetting.
Wherein, the judged result subelement also is used for: when a certain single character of a plurality of single characters was not included in the default command character information, then judging character information was not the command character information of presetting.
Wherein, referring to Fig. 5, the corresponding relation of store character information and numerical information in the preset grammatical rule, first judge module 302 comprises:
Whether search unit 3021, be used at the character information of syntax rule storage and the corresponding relation of numerical information, searching character information has corresponding numerical information;
Judged result unit 3022, be used for when the result who searches the unit when being, judge the numerical information that has the character information correspondence in the preset grammatical rule; When the result who searches the unit for not the time, judge the numerical information that does not have the character information correspondence in the preset grammatical rule.
Wherein, processing module 303 comprises acquiring unit 3031 and determining unit 3032;
Acquiring unit 3031 is used for according to numerical information, obtaining character information from syntax rule when judging the preset grammatical rule and have the numerical information of character information correspondence, and the character information that obtains comprises subject information and/or action message;
Determining unit 3032, the character information that is used for obtaining is defined as recognition result.
Wherein, the preset grammatical rule is kept in the expandable mark language XML file.
The device that the embodiment of the invention provides, by voice messaging is identified as character information, and whether there is the numerical information of this character information correspondence according to the preset grammatical rule judgment, when having corresponding numerical information, to be defined as recognition result according to the character information that numerical information gets access to, realized the process of machine recognition people's natural language, recognition result can be used for the recognition result that the aircraft controller can identify according to this speech recognition equipment, the intelligent aircraft that control computer generates, the training work of commander's flight.This recognition methods goes for different speaker, and different can finish speech recognition by this audio recognition method per capita.
Need to prove: the speech recognition equipment that the foregoing description provides is when carrying out the speech recognition business, only the division with above-mentioned each functional module is illustrated, in the practical application, can as required the above-mentioned functions distribution be finished by different functional modules, the inner structure that is about to device is divided into different functional modules, to finish all or part of function described above.In addition, speech recognition equipment that the foregoing description provides and audio recognition method embodiment belong to same design, and its specific implementation process sees method embodiment for details, repeats no more here.
The invention described above embodiment sequence number is not represented the quality of embodiment just to description.
The all or part of step that one of ordinary skill in the art will appreciate that realization the foregoing description can be finished by hardware, also can instruct relevant hardware to finish by program, described program can be stored in a kind of computer-readable recording medium, the above-mentioned storage medium of mentioning can be a ROM (read-only memory), disk or CD etc.
The above only is preferred embodiment of the present invention, and is in order to restriction the present invention, within the spirit and principles in the present invention not all, any modification of being done, is equal to replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (14)

1. an audio recognition method is characterized in that, described method comprises:
The voice messaging that receives is identified as character information;
Judge the numerical information that whether has described character information correspondence in the preset grammatical rule;
When in judging described preset grammatical rule, having the numerical information of described character information correspondence, obtain recognition result according to described numerical information.
2. the method for claim 1 is characterized in that, describedly judges that before the numerical information that whether has described character information correspondence in the preset grammatical rule, described method also comprises:
Judge whether described character information is the command character information of presetting;
When judging described character information, carry out and judge the operation that whether has the numerical information of described character information correspondence in the preset grammatical rule for default command character information.
3. method as claimed in claim 2 is characterized in that, describedly judges that whether described character information is the command character information of presetting, and comprising:
Described character information is split as a plurality of single characters;
Judge one by one whether described a plurality of single character is included in the default command character information;
When described a plurality of single characters all were included in the default command character information, then judging described character information was the command character information of presetting.
4. method as claimed in claim 3 is characterized in that, described method also comprises:
When a certain single character in described a plurality of single characters was not included in the default command character information, then judging described character information was not the command character information of presetting.
5. the method for claim 1 is characterized in that, the corresponding relation of store character information and numerical information in the described preset grammatical rule is describedly judged the numerical information that whether has described character information correspondence in the preset grammatical rule, comprising:
In the corresponding relation of the character information of described syntax rule storage and numerical information, search the numerical information whether described character information has correspondence;
If find corresponding numerical information, then judge the numerical information that has described character information correspondence in the described preset grammatical rule;
If search numerical information, then judge the numerical information that does not have described character information correspondence in the described preset grammatical rule less than correspondence.
6. the method for claim 1 is characterized in that, obtains recognition result according to described numerical information, comprising:
According to described numerical information, from described syntax rule, obtain character information, the described character information that obtains comprises subject information and/or action message:
The described character information that obtains is defined as recognition result.
7. the method for claim 1 is characterized in that, described preset grammatical rule is kept in the expandable mark language XML file.
8. a speech recognition equipment is characterized in that, described device comprises:
Identification module, the voice messaging that is used for receiving is identified as character information;
First judge module is used for judging whether the preset grammatical rule exists the numerical information of described character information correspondence;
Processing module is used for obtaining recognition result according to described numerical information when judging described preset grammatical rule and have the numerical information of described character information correspondence.
9. device as claimed in claim 8 is characterized in that, described device also comprises second judge module, comprising:
Judging unit is used to judge whether described character information is the command character information of presetting;
Performance element is used for carrying out and judging the operation that whether has the numerical information of described character information correspondence in the preset grammatical rule when described judgment unit judges goes out described character information for default command character information.
10. device as claimed in claim 9 is characterized in that, described judging unit comprises:
Split subelement, be used for described character information is split as a plurality of single characters;
Judgment sub-unit is used for judging one by one whether described a plurality of single character is included in default command character information;
The judged result subelement is used for when described a plurality of single characters all are included in default command character information, and then judging described character information is the command character information of presetting.
11. device as claimed in claim 10, it is characterized in that, described judged result subelement also is used for: when a certain single character of described a plurality of single characters was not included in the default command character information, then judging described character information was not the command character information of presetting.
12. device as claimed in claim 8 is characterized in that, the corresponding relation of store character information and numerical information in the described preset grammatical rule, and described first judge module comprises:
Whether search the unit, be used at the character information of described syntax rule storage and the corresponding relation of numerical information, searching described character information has corresponding numerical information;
The judged result unit, be used for when the described result who searches the unit when being, judge the numerical information that has described character information correspondence in the described preset grammatical rule; When the described result who searches the unit for not the time, judge the numerical information that does not have described character information correspondence in the described preset grammatical rule.
13. device as claimed in claim 8 is characterized in that, described processing module comprises acquiring unit and determining unit;
Described acquiring unit, be used for when judging described preset grammatical rule and have the numerical information of described character information correspondence, according to described numerical information, from described syntax rule, obtain character information, the described character information that obtains comprises subject information and/or action message;
Described determining unit is used for the described character information that obtains is defined as recognition result.
14. device as claimed in claim 8 is characterized in that, described preset grammatical rule is kept in the expandable mark language XML file.
CN2013101020652A 2013-03-27 2013-03-27 Voice recognition method and voice recognition device Pending CN103219007A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN2013101020652A CN103219007A (en) 2013-03-27 2013-03-27 Voice recognition method and voice recognition device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN2013101020652A CN103219007A (en) 2013-03-27 2013-03-27 Voice recognition method and voice recognition device

Publications (1)

Publication Number Publication Date
CN103219007A true CN103219007A (en) 2013-07-24

Family

ID=48816733

Family Applications (1)

Application Number Title Priority Date Filing Date
CN2013101020652A Pending CN103219007A (en) 2013-03-27 2013-03-27 Voice recognition method and voice recognition device

Country Status (1)

Country Link
CN (1) CN103219007A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104123939A (en) * 2014-06-06 2014-10-29 国家电网公司 Substation inspection robot based voice interaction control method
CN104485106A (en) * 2014-12-08 2015-04-01 畅捷通信息技术股份有限公司 Voice recognition method, voice recognition system and voice recognition equipment
CN105810200A (en) * 2016-02-04 2016-07-27 深圳前海勇艺达机器人有限公司 Man-machine dialogue apparatus and method based on voiceprint identification
CN105818983A (en) * 2016-03-18 2016-08-03 普宙飞行器科技(深圳)有限公司 Operation method for unmanned aerial vehicle and unmanned aerial vehicle system
WO2017156893A1 (en) * 2016-03-18 2017-09-21 深圳Tcl数字技术有限公司 Voice control method and smart television
CN108334109A (en) * 2015-03-30 2018-07-27 安溪钟泰专利技术转移有限公司 A kind of phonetic controller
WO2018141140A1 (en) * 2017-02-06 2018-08-09 中兴通讯股份有限公司 Method and device for semantic recognition
CN108766426A (en) * 2018-05-31 2018-11-06 中国舰船研究设计中心 A kind of naval vessels intelligent sound interaction command system
CN109801630A (en) * 2018-12-12 2019-05-24 平安科技(深圳)有限公司 Digital conversion method, device, computer equipment and the storage medium of speech recognition
CN110164224A (en) * 2019-05-14 2019-08-23 重庆埃默科技有限责任公司 A kind of phonic plotting method, apparatus and system

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1462428A (en) * 2001-03-30 2003-12-17 索尼公司 Sound processing apparatus
CN1512484A (en) * 2002-12-27 2004-07-14 联想(北京)有限公司 Organizing and identifying method for natural language
CN1540626A (en) * 2003-04-25 2004-10-27 台达电子工业股份有限公司 Device and method for implementing voice recognition of mixed multinational languages by using model of double tone
CN102938945A (en) * 2012-10-17 2013-02-20 上海移远通信技术有限公司 Wireless communication module and control method thereof

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1462428A (en) * 2001-03-30 2003-12-17 索尼公司 Sound processing apparatus
CN1512484A (en) * 2002-12-27 2004-07-14 联想(北京)有限公司 Organizing and identifying method for natural language
CN1540626A (en) * 2003-04-25 2004-10-27 台达电子工业股份有限公司 Device and method for implementing voice recognition of mixed multinational languages by using model of double tone
CN102938945A (en) * 2012-10-17 2013-02-20 上海移远通信技术有限公司 Wireless communication module and control method thereof

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104123939A (en) * 2014-06-06 2014-10-29 国家电网公司 Substation inspection robot based voice interaction control method
CN104485106A (en) * 2014-12-08 2015-04-01 畅捷通信息技术股份有限公司 Voice recognition method, voice recognition system and voice recognition equipment
CN104485106B (en) * 2014-12-08 2018-08-28 畅捷通信息技术股份有限公司 Audio recognition method, speech recognition system and speech recognition apparatus
CN108334109A (en) * 2015-03-30 2018-07-27 安溪钟泰专利技术转移有限公司 A kind of phonetic controller
CN108334109B (en) * 2015-03-30 2021-02-12 绵阳硅基智能科技有限公司 Voice control device
CN105810200A (en) * 2016-02-04 2016-07-27 深圳前海勇艺达机器人有限公司 Man-machine dialogue apparatus and method based on voiceprint identification
CN105818983B (en) * 2016-03-18 2019-10-01 普宙飞行器科技(深圳)有限公司 The control method and UAV system of unmanned plane
CN105818983A (en) * 2016-03-18 2016-08-03 普宙飞行器科技(深圳)有限公司 Operation method for unmanned aerial vehicle and unmanned aerial vehicle system
WO2017156893A1 (en) * 2016-03-18 2017-09-21 深圳Tcl数字技术有限公司 Voice control method and smart television
WO2018141140A1 (en) * 2017-02-06 2018-08-09 中兴通讯股份有限公司 Method and device for semantic recognition
CN108766426A (en) * 2018-05-31 2018-11-06 中国舰船研究设计中心 A kind of naval vessels intelligent sound interaction command system
CN108766426B (en) * 2018-05-31 2021-09-07 中国舰船研究设计中心 Intelligent voice interaction command system for naval vessel
CN109801630A (en) * 2018-12-12 2019-05-24 平安科技(深圳)有限公司 Digital conversion method, device, computer equipment and the storage medium of speech recognition
CN109801630B (en) * 2018-12-12 2024-05-28 平安科技(深圳)有限公司 Digital conversion method, device, computer equipment and storage medium for voice recognition
CN110164224A (en) * 2019-05-14 2019-08-23 重庆埃默科技有限责任公司 A kind of phonic plotting method, apparatus and system

Similar Documents

Publication Publication Date Title
CN103219007A (en) Voice recognition method and voice recognition device
CN107993665B (en) Method for determining role of speaker in multi-person conversation scene, intelligent conference method and system
CN105374356B (en) Audio recognition method, speech assessment method, speech recognition system and speech assessment system
CN105872687A (en) Method and device for controlling intelligent equipment through voice
CN103065630B (en) User personalized information voice recognition method and user personalized information voice recognition system
CN104078044A (en) Mobile terminal and sound recording search method and device of mobile terminal
CN108711421A (en) A kind of voice recognition acoustic model method for building up and device and electronic equipment
CN107305541A (en) Speech recognition text segmentation method and device
CN104538034A (en) Voice recognition method and system
CN104036774A (en) Method and system for recognizing Tibetan dialects
CN107369439A (en) A kind of voice awakening method and device
CN107767861A (en) voice awakening method, system and intelligent terminal
CN101826325A (en) Method and device for identifying Chinese and English speech signal
CN109616096A (en) Construction method, device, server and the medium of multilingual tone decoding figure
CN104143331A (en) Method and system for adding punctuations
CN110691258A (en) Program material manufacturing method and device, computer storage medium and electronic equipment
CN111415537A (en) Symbol-labeling-based word listening system for primary and secondary school students
CN103903617A (en) Voice recognition method and electronic device
CN105869622B (en) Chinese hot word detection method and device
CN104952446A (en) Digital building presentation system based on voice interaction
CN103902193A (en) System and method for operating computers to change slides by aid of voice
CN105206263A (en) Speech and meaning recognition method based on dynamic dictionary
US20150006175A1 (en) Apparatus and method for recognizing continuous speech
CN103474062A (en) Voice identification method
CN104424942A (en) Method for improving character speed input accuracy

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C12 Rejection of a patent application after its publication
RJ01 Rejection of invention patent application after publication

Application publication date: 20130724