CN104216906A - Voice searching method and device - Google Patents
Voice searching method and device Download PDFInfo
- Publication number
- CN104216906A CN104216906A CN201310217287.9A CN201310217287A CN104216906A CN 104216906 A CN104216906 A CN 104216906A CN 201310217287 A CN201310217287 A CN 201310217287A CN 104216906 A CN104216906 A CN 104216906A
- Authority
- CN
- China
- Prior art keywords
- pinyin information
- fuzzy
- phonetic
- information
- pinyin
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9537—Spatial or temporal dependent retrieval, e.g. spatiotemporal queries
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/63—Querying
- G06F16/632—Query formulation
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Multimedia (AREA)
- Remote Sensing (AREA)
- Document Processing Apparatus (AREA)
Abstract
The invention discloses a voice searching method and device. The voice searching method comprises the steps that voice information used for searching is recognized into text information; the text information is converted into first pinyin information; a searching result matched with the first pinyin information is obtained locally or from a server. According to the voice searching method and device, the technical problem that in the prior art, a correct searching result can not be obtained due to incorrect voice recognition is effectively solved, and the technical effect that correct searching results can be obtained is achieved.
Description
Technical field
The present invention relates to internet arena, in particular to a kind of voice search method and equipment.
Background technology
Speech recognition is exactly the technology that voice signal is changed into corresponding text or order by the process that allows machine pass through to identify and understand.Recent two decades comes, and speech recognition technology achieves marked improvement, and speech recognition has also been used in increasing field, such as: the browser, vehicle-mounted navigational system etc. of mobile phone, band speech recognition.
Can be there is polyphone, situation that phonetically similar word is confused in speech recognition, if so just cause the voice itself inputted to be polyphone or phonetically similar word, be easy to the situation occurring identification error or identify less than the content of phonetic entry unavoidably.The settling mode of current existence be according to the context of input content determine to should polyphone or phonetically similar word should be corresponding Chinese character, if but input content be do not exist contextual, such as, in onboard navigation system, if need to search certain place, what then input is exactly the name in this place, and this word does not have the context of its correspondence.Therefore, if do not have an input word in contextual situation this, and this word exists in multiple homonym or this word and has polyphone, then the voice that still cannot solve input are by the way difficult to the problem be identified.Because speech recognition is inaccurate, thus it is incorrect to cause carrying out searching for the Search Results obtained based on the voice messaging identified.
For the problems referred to above, at present effective solution is not yet proposed.
Summary of the invention
The fundamental purpose of the application is to provide a kind of voice search method and equipment, at least to solve the technical matters that cannot obtain correct Search Results caused because speech recognition is inaccurate in correlation technique.
According to an aspect of the present invention, provide a kind of voice search method, comprising: the voice messaging being used for searching for is identified as text message; Described text message is converted to the first Pinyin information; Local or obtain the Search Results mated with described first Pinyin information to server.
Alternatively, if the Search Results mated with the first Pinyin information cannot be got, then obtain the fuzzy pinyin information corresponding with the first Pinyin information, wherein, the initial consonant of the phonetic in fuzzy pinyin information is the fuzzy phoneme of the initial consonant of the phonetic in the first Pinyin information, and/or the simple or compound vowel of a Chinese syllable of the phonetic in fuzzy pinyin information is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in the first Pinyin information; Obtain the Search Results with fuzzy pinyin information matches.
Alternatively, obtain the fuzzy pinyin information corresponding with the first Pinyin information to comprise: arrange according to user individual preference and obtain the fuzzy pinyin information corresponding with the first Pinyin information.
Alternatively, obtain the fuzzy pinyin information corresponding with the first Pinyin information and comprise one of following mode: the initial consonant of the phonetic in the fuzzy pinyin information of acquisition is the fuzzy phoneme of the initial consonant of the phonetic in the first Pinyin information and the simple or compound vowel of a Chinese syllable of the phonetic in the fuzzy pinyin information obtained is the simple or compound vowel of a Chinese syllable of the phonetic in the first Pinyin information; The initial consonant of phonetic in the fuzzy pinyin information obtained is the initial consonant of phonetic in the first Pinyin information and the simple or compound vowel of a Chinese syllable of the phonetic in the fuzzy pinyin information obtained is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of phonetic in the first Pinyin information; The initial consonant of phonetic in the fuzzy pinyin information obtained is the fuzzy phoneme of initial consonant of phonetic in the first Pinyin information and the simple or compound vowel of a Chinese syllable of the phonetic in the fuzzy pinyin information obtained is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of phonetic in the first Pinyin information.
Alternatively, user individual preference arranges and comprises user region indication information, the fuzzy phoneme of the initial consonant of the phonetic in the first Pinyin information is the fuzzy phoneme corresponding with user region indication information, and/or the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in the first Pinyin information is the fuzzy phoneme corresponding with user region indication information.
Alternatively, local or obtains to server the Search Results mated with the first Pinyin information and comprise: be used for representing that the text message of the title of entry is converted to the second Pinyin information by the database of this locality or server; Judge that whether the first Pinyin information is identical with the second Pinyin information; If the first Pinyin information is identical with the second Pinyin information, then the name of entry is referred to as Search Results and shows.
Alternatively, the text message being used for the title representing entry in database is converted to the second Pinyin information to comprise: the text message of the title being used for representing entry in database is converted to the second Pinyin information by the search engine provided by third party.
Alternatively, the information of the entry record point of interest in database, wherein, the information of point of interest comprises: the title of point of interest, the positional information of point of interest, the address of point of interest.
According to a further aspect in the invention, provide a kind of phonetic search equipment, comprising: speech recognition equipment, for the voice messaging being used for searching for is identified as text message; Signal conditioning package, for being converted to the first Pinyin information by described text message; Searcher, for local or obtain the Search Results mated with described first Pinyin information to server.
Alternatively, signal conditioning package is also for getting the Search Results mated with the first Pinyin information, obtain the fuzzy pinyin information corresponding with the first Pinyin information, wherein, the initial consonant of the phonetic in fuzzy pinyin information is the fuzzy phoneme of the initial consonant of the phonetic in the first Pinyin information, and/or the simple or compound vowel of a Chinese syllable of the phonetic in fuzzy pinyin information is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in the first Pinyin information; Searcher is also for obtaining the Search Results with fuzzy pinyin information matches in this locality or to server.
Alternatively, equipment also comprises: display device; Signal conditioning package is also for being converted to the second Pinyin information by the text message being used for the title representing entry in the database in local or server; Searcher is for judging that whether the first Pinyin information is identical with the second Pinyin information; When the first Pinyin information is identical with the second Pinyin information, the name of entry is referred to as Search Results and sends to display device; Display device is used for display of search results.
In embodiments of the present invention, first the voice messaging received is converted to the Pinyin information of its correspondence, then by Search Results that the search of this Pinyin information matches.Due in the process obtaining Search Results, comparison be Pinyin information, instead of character itself, like this, when making voice messaging be wrongly recognized into identical other characters of phonetic in the existence due to phonetically similar word, still can adopt correct Pinyin information to search for the Search Results mated with this Pinyin information, thus solve the technical matters that cannot obtain correct Search Results caused because speech recognition is inaccurate, reach the technique effect that can obtain correct Search Results.
Accompanying drawing explanation
Accompanying drawing described herein is used to provide further understanding of the present application, and form a application's part, the schematic description and description of the application, for explaining the application, does not form the improper restriction to the application.In the accompanying drawings:
Fig. 1 is a kind of structured flowchart of the phonetic search equipment according to the embodiment of the present invention;
Fig. 2 is the another kind of structured flowchart of the phonetic search equipment according to the embodiment of the present invention;
Fig. 3 is a kind of process flow diagram of the voice search method according to the embodiment of the present invention; And
Fig. 4 is a kind of process flow diagram of the Search Results mated with Pinyin information according to the acquisition of the embodiment of the present invention.
Embodiment
Hereinafter also describe the application in detail with reference to accompanying drawing in conjunction with the embodiments.It should be noted that, when not conflicting, the embodiment in the application and the feature in embodiment can combine mutually.
Embodiment 1
The phonetic search equipment that the present embodiment provides, comprises as shown in Figure 1: speech recognition equipment 102, signal conditioning package 104 and searcher 106.Below this structure is specifically described:
1) speech recognition equipment 102, for being identified as text message by the voice messaging being used for searching for.Alternatively, the voice messaging of the voice messaging received or input is converted into text message by speech recognition equipment 102, such as: if user needs the position of searching point of interest " Guang Lin mansion " in the process of navigation, then can by the microphone input voice information " Guang Lin mansion " in in-vehicle electronic system, then, above-mentioned voice messaging can be identified as text message by speech recognition equipment 102.It should be noted that, above-mentioned navigation scenarios is only an example, and the voice messaging of input can also be identified as corresponding text message by speech recognition equipment 102 under other scenes, such as, under the scene that vehicle does not start, or, in vehicle operating but under not opening the scene of navigation.
2) signal conditioning package 104, is coupled with speech recognition equipment 102, for above-mentioned text message is converted to the first Pinyin information.Such as, above-mentioned text message can be converted to the Pinyin information " guang lin da sha " of its correspondence by signal conditioning package 104.
3) searcher 106, is coupled with signal conditioning package 104, for local or obtain the Search Results mated with described first Pinyin information to server.Alternatively, corresponding with the Search Results that the first Pinyin information mates Pinyin information is identical with the first Pinyin information.Alternatively, multiple content to be matched can be saved, so that searcher 106 can obtain the Search Results mated with described first Pinyin information in above-mentioned database in the database of or server local at phonetic search equipment in advance.Such as, phonetic search equipment has a map data base in this locality, correspondingly store multiple entry in this map data base, and each entry comprises: the title (such as, the title of point of interest) of entry and corresponding address, shown in table specific as follows:
Table 1
| The title of point of interest | Address |
| Guang Lin mansion | No. 1, Beijing Road |
| Zhong Guang hotel | No. 1, Nanjing Road |
| Shanghai Communications University | No. 1, Tianjin road |
Like this after the above-mentioned Pinyin information of acquisition " guang lin da sha ", searcher 106 can obtain the title of the point of interest mated with Pinyin information " guang lin da sha " in map data base, here, the phonetic of the title of the point of interest matched is " guang lin da sha ".
Alternatively, can as follows local or obtain the Search Results mated with described first Pinyin information to server: the text message of the title being used for representing entry in the database in described this locality or server is converted to the second Pinyin information by signal conditioning package 104; Searcher 106 judges that whether described first Pinyin information is identical with described second Pinyin information; When described first Pinyin information is identical with described second Pinyin information, the name of this entry is referred to as described Search Results and returns by searcher 106.
Alternatively, the text message of the title being used for representing entry in described database is converted to described second Pinyin information by the search engine that signal conditioning package 104 can be provided by third party.Above-mentioned third party can be search server, namely provides the third-party server that phonetic is searched.
It should be noted that, the second Pinyin information also can be previously recorded in the database of this locality or server, such as, and can be as shown in the table:
Table 2
| The title of point of interest | Address | Pinyin information |
| Guang Lin mansion | No. 1, Beijing Road | guang?lin?da?sha |
| Zhong Guang hotel | No. 1, Nanjing Road | zhong?guang?jiu?dian |
| Shanghai Communications University | No. 1, Tianjin road | shang?hai?jiao?tong?da?xue |
In such a scenario, searcher 106 directly can obtain the Pinyin information of point of interest to be matched from the database of local or server, and the Pinyin information that this Pinyin information and signal conditioning package 104 are converted to is compared, if both are identical, then the name of the point of interest of correspondence are referred to as Search Results and return.
Alternatively, when returning Search Results, the name of point of interest being referred to as Search Results and returning, this is only a kind of example.The present invention is not limited only to this, other information in the entry can also mated with Pinyin information in return data storehouse, such as, and the address of the point of interest comprised in return recording item, the information such as the evaluation of point of interest.
Alternatively, as shown in Figure 2, the phonetic search equipment in the present embodiment can also comprise: display device 108.Mentioned above searching results is returned to display device 108 and shows by searcher 106.In the present embodiment, display device 108 can be, but not limited to as the touch-screen in in-vehicle electronic system.
It should be noted that, the phonetic search equipment in the present embodiment goes for the scene of text message that the voice messaging being used for searching for is identified as correct text message or comprises phonetically similar word by speech recognition equipment 102.Such as, when input voice information " Guang Lin mansion " is identified as " Guang Lin mansion " by speech recognition equipment 102, or, when input voice information " Guang Lin mansion " is identified as " Guang Lin mansion " by speech recognition equipment 102, the phonetic search equipment in the present embodiment can adopt above-mentioned mode to search correct Search Results.
In embodiments of the present invention, first the voice messaging received is converted to the Pinyin information of its correspondence, then by Search Results that the search of this Pinyin information matches.Due in the process obtaining Search Results, comparison be Pinyin information, instead of character itself, like this, even if make voice messaging be wrongly recognized into other identical characters of phonetic due to the existence of phonetically similar word, phonetic search equipment in the present embodiment still can adopt correct Pinyin information to search for the Search Results mated with this Pinyin information, thus solve the technical matters that cannot obtain correct Search Results caused because speech recognition is inaccurate, reach the technique effect that can obtain correct Search Results.
When the cacoepy of user is true, such as, the language message " lin " of input is said into " ling ", " continuous heavy rain " may be converted to " icepro " in process voice messaging being transformed into text message, thus the Search Results causing the search carried out based on the Pinyin information that text message is corresponding to obtain is inaccurate.To this, in order to ensure the correctness of Search Results further, the present embodiment additionally provides following scheme: when searcher 106 cannot in this locality or when obtaining to server the Search Results mated with described first Pinyin information, signal conditioning package 104 obtains the fuzzy pinyin information corresponding with described first Pinyin information, wherein, the initial consonant of the phonetic in described fuzzy pinyin information is the fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information, and/or, the simple or compound vowel of a Chinese syllable of the phonetic in described fuzzy pinyin information is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information, then, searcher 106 obtains the Search Results with described fuzzy pinyin information matches in described this locality or to described server.Alternatively, corresponding with the Search Results of described fuzzy pinyin information matches Pinyin information is identical with described first Pinyin information.
Alternatively, the fuzzy phoneme of initial consonant can include but not limited to: s and sh, c and ch, f and h, r and l.The fuzzy phoneme of simple or compound vowel of a Chinese syllable can include but not limited to: an and ang, en and eng, in and ing, ian and iang and uan and uang.
Such as, if the voice messaging being used for searching for " Guang Lin mansion " is identified as text message " Guang Ling mansion " by speech recognition equipment 102, the Pinyin information that then text information " Guang Ling mansion " is converted to by signal conditioning package 104 is " guang ling da sha ", cannot get the interest point name of mating with this Pinyin information like this from table 1.Under this scene, signal conditioning package 104 obtains the fuzzy pinyin information " guang lin da sha " of Pinyin information " guang ling da sha " further, then, searcher 106 can get the title " Guang Lin mansion " of the point of interest mated with fuzzy pinyin information " guang lin da sha " in Table 1, thus can obtain Search Results accurately when speech recognition equipment identification error.
As the optional scheme of one, one of in the following manner the fuzzy pinyin information corresponding with described first Pinyin information can be obtained:
1) initial consonant of the phonetic in the described fuzzy pinyin information obtained is the fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information and the simple or compound vowel of a Chinese syllable of phonetic in the described fuzzy pinyin information obtained is the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information;
Alternatively, the initial consonant of the one or more phonetics in the first Pinyin information all can be replaced to all possible fuzzy phoneme of this initial consonant, such as, Pinyin information " guang lin da sha " is converted to fuzzy pinyin information " kuang nin ta sa ", " kuang nin ta sha ", " kuang nin ha sa ", " kuang nin ta sha " etc.Alternatively, searcher 106 mates replacing all fuzzy pinyin information obtained with the database in local or server.
2) initial consonant of the phonetic in the described fuzzy pinyin information obtained is the initial consonant of the phonetic in described first Pinyin information and the simple or compound vowel of a Chinese syllable of phonetic in the described fuzzy pinyin information obtained is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of phonetic in described first Pinyin information;
Alternatively, the simple or compound vowel of a Chinese syllable of the one or more phonetics in the first Pinyin information all can be replaced to all possible fuzzy phoneme of this simple or compound vowel of a Chinese syllable, such as, Pinyin information " guang lin da sha " is converted to " guan ling dao shan ", " guan ling dao shang " etc.Alternatively, searcher 106 mates replacing all fuzzy pinyin information obtained with the database in local or server.
3) initial consonant of the phonetic in the described fuzzy pinyin information obtained is the fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information and the simple or compound vowel of a Chinese syllable of phonetic in the described fuzzy pinyin information obtained is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of phonetic in described first Pinyin information.
Alternatively, the initial consonant of the one or more phonetics in the first Pinyin information all can be replaced to all possible fuzzy phoneme of this initial consonant, and the simple or compound vowel of a Chinese syllable of the one or more phonetics in the first Pinyin information is all replaced to all possible fuzzy phoneme of this simple or compound vowel of a Chinese syllable, such as, Pinyin information " guang lin da sha " is converted to " kuang ning ta sa ", " kuang ning ta sha ", " kuang ning ha sa ", " kuang nin ta sha " etc.Alternatively, searcher 106 mates replacing all fuzzy pinyin information obtained with the database in local or server.
Certainly, aforesaid way is an example, and the present invention is not limited only to this.
Alternatively, if described user individual preference is arranged comprise user region indication information, the fuzzy phoneme of the initial consonant of the phonetic then in described first Pinyin information is that the fuzzy phoneme corresponding with described user region indication information is (preferred, the fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information is fuzzy phoneme corresponding with described user region indication information in all fuzzy phonemes of the initial consonant of the phonetic in described first Pinyin information), and/or, the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information is that the fuzzy phoneme corresponding with described user region indication information is (preferred, the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information is fuzzy phoneme corresponding with described user region indication information in all fuzzy phonemes of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information).
Such as, if user is Hunan people, then may exist " l " and " n " regardless of situation, user individual preference can arrange and comprise user region indication information " Hunan " by user in advance, in the present embodiment, user individual preference arranges and can be, but not limited to store in memory, such as, is stored in the storer of in-vehicle electronic system.Like this, signal conditioning package 104 increases entry in the mapping table of inside, to represent the fuzzy phoneme l and n being set to initial consonant each other.Like this, when the Pinyin information that voice messaging is corresponding is " lin ", Pinyin information " lin " can be replaced to fuzzy pinyin information " nin " according to this entry by signal conditioning package 104.
Alternatively, the entry in above-mentioned database have recorded the information of point of interest, and wherein, the information of described point of interest comprises: the title of point of interest, the positional information of point of interest, the address of point of interest.
Alternatively, the voice messaging in above-described embodiment can represent the title of a complete point of interest, such as, " Guang Lin mansion " or, voice messaging in above-described embodiment can represent the abbreviation of the title of a complete point of interest, such as, and " handing over large " (abbreviation of Shanghai Communications University).When voice messaging represents the abbreviation of the title of a complete point of interest, database can further include the abbreviation of the title of point of interest and corresponding Pinyin information, such as, as shown in the table, phonetic search equipment can search for the Search Results matched with Pinyin information in the manner described above, does not repeat them here.
Table 3
| The title of point of interest | Title is abridged | Address | Pinyin information | The Pinyin information of abbreviation |
| Guang Lin mansion | Wide continuous heavy rain | No. 1, Beijing Road | guang?lin?da?sha | guang?lin |
| Zhong Guang hotel | In wide | No. 1, Nanjing Road | zhong?guang?jiu?dian | zhong?guang |
| Shanghai Communications University | Hand over large | No. 1, Tianjin road | shang?hai?jiao?tong?da?xue | jiao?da |
In addition, when signal conditioning package 104 obtains the fuzzy pinyin information of Pinyin information, or, when searcher 106 is searched for according to the fuzzy pinyin information of Pinyin information or Pinyin information, it is also conceivable to the current residing position (regional information " Shanghai ", " Beijing " etc. such as, residing for this user) of user.With voice messaging be point of interest be abbreviated as example, when the position that user is current is in Beijing, the voice messaging of user's input is " Tsing-Hua University ", this voice messaging is identified as text message " front yard China " by speech recognition equipment 102, signal conditioning package 104 is when obtaining the fuzzy pinyin information of Pinyin information, consider that the current positional information of this user is for " Beijing ", the temperature comprising the point of interest of " Tsing-Hua University " in Beijing should be higher, therefore, Pinyin information " ting hua " corresponding to " front yard China " is converted to fuzzy pinyin information " qing hua ", and then search point of interest " Tsing-Hua University ".
Embodiment 2
Based on the language search equipment shown in Fig. 1 and Fig. 2, the embodiment of the present invention also proposed one voice search method alternatively, and as shown in Figure 3, the voice search method in the present embodiment comprises the following steps:
S302: the voice messaging being used for searching for is identified as text message;
Alternatively, step S302 with reference to the step performed by above-mentioned speech recognition equipment 102, can not repeat them here.
S304: described text message is converted to the first Pinyin information;
Alternatively, step S304 with reference to the step performed by above-mentioned signal conditioning package 104, can not repeat them here.
S306: local or obtain the Search Results mated with described first Pinyin information to server.Alternatively, the Pinyin information that the Search Results that mates of described and described first Pinyin information is corresponding is identical with described first Pinyin information.
Alternatively, step S306 with reference to the step performed by above-mentioned signal conditioning package 106, can not repeat them here.
In this step, multiple content to be matched can be saved, the Search Results mated with described first Pinyin information can be obtained in above-mentioned database in advance in the database of local or server.Such as, local or server has a map data base, and correspondingly store multiple entry in this map data base, each entry comprises: the title (such as, the title of point of interest) of entry and corresponding address, specifically as shown in table 1.
Alternatively, as shown in Figure 4, can as follows local or obtain the Search Results mated with described first Pinyin information to server:
S402: the text message being used for the title representing entry in the database in described this locality or server is converted to the second Pinyin information;
S404: judge that whether described first Pinyin information is identical with described second Pinyin information;
S406: when described first Pinyin information is identical with described second Pinyin information, the name of this entry is referred to as described Search Results and returns.
Alternatively, the text message of the title being used for representing entry in described database is converted to described second Pinyin information by the search engine that can be provided by third party.Above-mentioned third party can be search server, namely provides the third-party server that phonetic is searched.
It should be noted that, the second Pinyin information also can be previously recorded in the database of this locality or server, such as, and can be as shown in table 2.
In such a scenario, the Pinyin information of point of interest to be matched directly can be obtained from the database of local or server, and this Pinyin information and the Pinyin information be converted to are compared, if both are identical, then the name of the point of interest of correspondence are referred to as Search Results and return.Alternatively, mentioned above searching results can be returned to display device to show.
Alternatively, when returning Search Results, the name of point of interest being referred to as Search Results and returning, this is only a kind of example.The present invention is not limited only to this, other information in the entry can also mated with Pinyin information in return data storehouse, such as, and the address of the point of interest comprised in return recording item, the information such as the evaluation of point of interest.
In embodiments of the present invention, first the voice messaging received is converted to the Pinyin information of its correspondence, then by Search Results that the search of this Pinyin information matches.Due in the process obtaining Search Results, comparison be Pinyin information, instead of character itself, like this, when making voice messaging be wrongly recognized into identical other characters of phonetic in the existence due to phonetically similar word, still can adopt correct Pinyin information to search for the Search Results mated with this Pinyin information, thus solve the technical matters that cannot obtain correct Search Results caused because speech recognition is inaccurate, reach the technique effect that can obtain correct Search Results.
When the cacoepy of user is true, such as, the language message " lin " of input is said into " ling ", " continuous heavy rain " may be converted to " icepro " in process voice messaging being transformed into text message, thus the Search Results causing the search carried out based on the Pinyin information that text message is corresponding to obtain is inaccurate.To this, in order to ensure the correctness of Search Results further, the present embodiment additionally provides following scheme: when cannot local or when obtaining to server the Search Results mated with described first Pinyin information, obtain the fuzzy pinyin information corresponding with described first Pinyin information, wherein, the initial consonant of the phonetic in described fuzzy pinyin information is the fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information, and/or the simple or compound vowel of a Chinese syllable of the phonetic in described fuzzy pinyin information is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information; Then, obtain the Search Results with described fuzzy pinyin information matches in described this locality or to described server, wherein, the Pinyin information that the Search Results of described and described fuzzy pinyin information matches is corresponding is identical with described first Pinyin information.
Alternatively, the fuzzy phoneme of initial consonant can include but not limited to: s and sh, c and ch, f and h, r and l.The fuzzy phoneme of simple or compound vowel of a Chinese syllable can include but not limited to: an and ang, en and eng, in and ing, ian and iang and uan and uang.
As the optional scheme of one, obtain the fuzzy pinyin information corresponding with described first Pinyin information and can adopt mode 1 in above-described embodiment 1)-3) in one, do not repeat them here.
Certainly, aforesaid way is an example, and the present invention is not limited only to this.
Alternatively, if described user individual preference is arranged comprise user region indication information, the fuzzy phoneme of the initial consonant of the phonetic then in described first Pinyin information is that the fuzzy phoneme corresponding with described user region indication information is (preferred, the fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information is fuzzy phoneme corresponding with described user region indication information in all fuzzy phonemes of the initial consonant of the phonetic in described first Pinyin information), and/or, the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information is that the fuzzy phoneme corresponding with described user region indication information is (preferred, the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information is fuzzy phoneme corresponding with described user region indication information in all fuzzy phonemes of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information).
Alternatively, the entry in above-mentioned database have recorded the information of point of interest, and wherein, the information of described point of interest comprises: the title of point of interest, the positional information of point of interest, the address of point of interest.
Alternatively, the voice messaging in above-described embodiment can represent the title of a complete point of interest, such as, " Guang Lin mansion " or, voice messaging in above-described embodiment can represent the abbreviation of the title of a complete point of interest, such as, and " handing over large " (abbreviation of Shanghai Communications University).When voice messaging represents the abbreviation of the title of a complete point of interest, database can further include the abbreviation of the title of point of interest and corresponding Pinyin information, such as, as shown in table 3, the Search Results matched with Pinyin information can be searched in the manner described above, not repeat them here.
Obviously, those skilled in the art should be understood that, each module of above-mentioned the application or each step can realize with general calculation element, they can concentrate on single calculation element, or be distributed on network that multiple calculation element forms, alternatively, they can realize with the executable program code of calculation element, thus, they can be stored and be performed by calculation element in the storage device, and in some cases, step shown or described by can performing with the order be different from herein, or they are made into each integrated circuit modules respectively, or the multiple module in them or step are made into single integrated circuit module to realize.Like this, the application is not restricted to any specific hardware and software combination.
The foregoing is only the preferred embodiment of the application, be not limited to the application, for a person skilled in the art, the application can have various modifications and variations.Within all spirit in the application and principle, any amendment done, equivalent replacement, improvement etc., within the protection domain that all should be included in the application.
Claims (11)
1. a voice search method, is characterized in that, comprising:
The voice messaging being used for searching for is identified as text message;
Described text message is converted to the first Pinyin information;
Local or obtain the Search Results mated with described first Pinyin information to server.
2. method according to claim 1, is characterized in that,
If the Search Results mated with described first Pinyin information cannot be got, then obtain the fuzzy pinyin information corresponding with described first Pinyin information, wherein, the initial consonant of the phonetic in described fuzzy pinyin information is the fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information, and/or the simple or compound vowel of a Chinese syllable of the phonetic in described fuzzy pinyin information is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information;
Obtain the Search Results with described fuzzy pinyin information matches.
3. method according to claim 2, is characterized in that, obtains the fuzzy pinyin information corresponding with described first Pinyin information and comprises:
Arrange according to user individual preference and obtain the fuzzy pinyin information corresponding with described first Pinyin information.
4. method according to claim 2, is characterized in that, the described acquisition fuzzy pinyin information corresponding with described first Pinyin information comprises one of following mode:
The initial consonant of phonetic in the described fuzzy pinyin information obtained is the fuzzy phoneme of initial consonant of phonetic in described first Pinyin information and the simple or compound vowel of a Chinese syllable of the phonetic in the described fuzzy pinyin information obtained is the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information;
The initial consonant of phonetic in the described fuzzy pinyin information obtained is the initial consonant of phonetic in described first Pinyin information and the simple or compound vowel of a Chinese syllable of the phonetic in the described fuzzy pinyin information obtained is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of phonetic in described first Pinyin information;
The initial consonant of phonetic in the described fuzzy pinyin information obtained is the fuzzy phoneme of initial consonant of phonetic in described first Pinyin information and the simple or compound vowel of a Chinese syllable of the phonetic in the described fuzzy pinyin information obtained is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of phonetic in described first Pinyin information.
5. method according to claim 3, is characterized in that, described user individual preference arranges and comprises user region indication information; The fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information is the fuzzy phoneme corresponding with described user region indication information, and/or the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information is the fuzzy phoneme corresponding with described user region indication information.
6. method according to claim 1, is characterized in that, local or obtain to server the Search Results mated with described first Pinyin information and comprise:
The text message being used for the title representing entry in the database of described this locality or described server is converted to the second Pinyin information;
Judge that whether described first Pinyin information is identical with described second Pinyin information;
If described first Pinyin information is identical with described second Pinyin information, then the name of described entry is referred to as described Search Results and shows.
7. method according to claim 6, is characterized in that, the text message being used for the title representing entry is converted to the second Pinyin information comprises in described database:
The text message being used for the title representing entry in described database is converted to described second Pinyin information by the search engine provided by third party.
8. the method according to claim 6 or 7, is characterized in that, the information of the entry record point of interest in described database, and wherein, the information of described point of interest comprises: the title of point of interest, the positional information of point of interest, the address of point of interest.
9. a phonetic search equipment, is characterized in that, comprising:
Speech recognition equipment, for being identified as text message by the voice messaging being used for searching for;
Signal conditioning package, for being converted to the first Pinyin information by described text message;
Searcher, for local or obtain the Search Results mated with described first Pinyin information to server.
10. equipment according to claim 9, is characterized in that,
Described signal conditioning package is also for getting the Search Results mated with described first Pinyin information, obtain the fuzzy pinyin information corresponding with described first Pinyin information, wherein, the initial consonant of the phonetic in described fuzzy pinyin information is the fuzzy phoneme of the initial consonant of the phonetic in described first Pinyin information, and/or the simple or compound vowel of a Chinese syllable of the phonetic in described fuzzy pinyin information is the fuzzy phoneme of the simple or compound vowel of a Chinese syllable of the phonetic in described first Pinyin information;
Described searcher is also for obtaining the Search Results with described fuzzy pinyin information matches in described this locality or to described server.
11. equipment according to claim 9, is characterized in that, described equipment also comprises: display device;
Described signal conditioning package is also for being converted to the second Pinyin information by the text message being used for the title representing entry in the database in described this locality or server;
Described searcher is for judging that whether described first Pinyin information is identical with described second Pinyin information; When described first Pinyin information is identical with described second Pinyin information, the name of described entry is referred to as described Search Results and sends to described display device;
Described display device is for showing described Search Results.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310217287.9A CN104216906A (en) | 2013-05-31 | 2013-05-31 | Voice searching method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310217287.9A CN104216906A (en) | 2013-05-31 | 2013-05-31 | Voice searching method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104216906A true CN104216906A (en) | 2014-12-17 |
Family
ID=52098408
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310217287.9A Pending CN104216906A (en) | 2013-05-31 | 2013-05-31 | Voice searching method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104216906A (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104794218A (en) * | 2015-04-28 | 2015-07-22 | 百度在线网络技术（北京）有限公司 | Voice searching method and device |
| CN104819727A (en) * | 2015-04-13 | 2015-08-05 | 福建天晴数码有限公司 | Navigation method and system thereof |
| CN105279227A (en) * | 2015-09-11 | 2016-01-27 | 百度在线网络技术（北京）有限公司 | Voice search processing method and device of homonym |
| CN105589573A (en) * | 2015-12-18 | 2016-05-18 | 魅族科技（中国）有限公司 | Character output method and character output device |
| CN106297799A (en) * | 2016-08-09 | 2017-01-04 | 乐视控股（北京）有限公司 | Voice recognition processing method and device |
| CN107562907A (en) * | 2017-09-11 | 2018-01-09 | 武汉科技大学 | A kind of intelligent lawyer's expert system and case answering device |
| CN108268637A (en) * | 2018-01-17 | 2018-07-10 | 广东蔚海数问大数据科技有限公司 | An intelligent voice correction recognition method, device and user terminal |
| CN108984800A (en) * | 2018-08-22 | 2018-12-11 | 广东小天才科技有限公司 | A voice search method and terminal equipment |
| CN109213777A (en) * | 2017-06-29 | 2019-01-15 | 杭州九阳小家电有限公司 | A kind of voice-based recipe processing method and system |
| CN109326284A (en) * | 2018-08-22 | 2019-02-12 | 深圳创维数字技术有限公司 | Voice search method, device and storage medium |
| CN110008471A (en) * | 2019-03-26 | 2019-07-12 | 北京博瑞彤芸文化传播股份有限公司 | A kind of intelligent semantic matching process based on phonetic conversion |
| CN110264994A (en) * | 2019-07-02 | 2019-09-20 | 珠海格力电器股份有限公司 | Voice synthesis method, electronic equipment and intelligent home system |
| CN110619879A (en) * | 2019-08-29 | 2019-12-27 | 深圳市梦网科技发展有限公司 | Voice recognition method and device |
| CN111198936A (en) * | 2018-11-20 | 2020-05-26 | 北京嘀嘀无限科技发展有限公司 | Voice search method and device, electronic equipment and storage medium |
| CN111259170A (en) * | 2018-11-30 | 2020-06-09 | 北京嘀嘀无限科技发展有限公司 | Voice search method and device, electronic equipment and storage medium |
| CN111475619A (en) * | 2020-03-31 | 2020-07-31 | 北京三快在线科技有限公司 | Text information correction method and device, electronic equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1674091A (en) * | 2005-04-18 | 2005-09-28 | 南京师范大学 | Sound identifying method for geographic information and its application in navigation system |
| US20060018545A1 (en) * | 2004-07-23 | 2006-01-26 | Lu Zhang | User interface and database structure for Chinese phrasal stroke and phonetic text input |
| CN101415259A (en) * | 2007-10-18 | 2009-04-22 | 三星电子株式会社 | System and method for searching information of embedded equipment based on double-language voice enquiry |
| CN101441527A (en) * | 2008-12-24 | 2009-05-27 | 腾讯科技（深圳）有限公司 | Method and apparatus for prompting right pronunciation in phonetic input |
| CN101499277A (en) * | 2008-07-25 | 2009-08-05 | 中国科学院计算技术研究所 | Service intelligent navigation method and system |
| CN102221898A (en) * | 2010-04-13 | 2011-10-19 | 刘汉龙 | Fuzzy three-dimensional pinyin code chinese character input method |
| CN102750949A (en) * | 2012-07-16 | 2012-10-24 | 深圳市车音网科技有限公司 | Voice recognition method and device |
-
2013
- 2013-05-31 CN CN201310217287.9A patent/CN104216906A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060018545A1 (en) * | 2004-07-23 | 2006-01-26 | Lu Zhang | User interface and database structure for Chinese phrasal stroke and phonetic text input |
| CN1674091A (en) * | 2005-04-18 | 2005-09-28 | 南京师范大学 | Sound identifying method for geographic information and its application in navigation system |
| CN101415259A (en) * | 2007-10-18 | 2009-04-22 | 三星电子株式会社 | System and method for searching information of embedded equipment based on double-language voice enquiry |
| CN101499277A (en) * | 2008-07-25 | 2009-08-05 | 中国科学院计算技术研究所 | Service intelligent navigation method and system |
| CN101441527A (en) * | 2008-12-24 | 2009-05-27 | 腾讯科技（深圳）有限公司 | Method and apparatus for prompting right pronunciation in phonetic input |
| CN102221898A (en) * | 2010-04-13 | 2011-10-19 | 刘汉龙 | Fuzzy three-dimensional pinyin code chinese character input method |
| CN102750949A (en) * | 2012-07-16 | 2012-10-24 | 深圳市车音网科技有限公司 | Voice recognition method and device |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104819727A (en) * | 2015-04-13 | 2015-08-05 | 福建天晴数码有限公司 | Navigation method and system thereof |
| CN104794218A (en) * | 2015-04-28 | 2015-07-22 | 百度在线网络技术（北京）有限公司 | Voice searching method and device |
| CN104794218B (en) * | 2015-04-28 | 2019-07-05 | 百度在线网络技术（北京）有限公司 | Voice search method and device |
| CN105279227A (en) * | 2015-09-11 | 2016-01-27 | 百度在线网络技术（北京）有限公司 | Voice search processing method and device of homonym |
| CN105589573A (en) * | 2015-12-18 | 2016-05-18 | 魅族科技（中国）有限公司 | Character output method and character output device |
| CN106297799A (en) * | 2016-08-09 | 2017-01-04 | 乐视控股（北京）有限公司 | Voice recognition processing method and device |
| CN109213777A (en) * | 2017-06-29 | 2019-01-15 | 杭州九阳小家电有限公司 | A kind of voice-based recipe processing method and system |
| CN107562907A (en) * | 2017-09-11 | 2018-01-09 | 武汉科技大学 | A kind of intelligent lawyer's expert system and case answering device |
| CN107562907B (en) * | 2017-09-11 | 2020-10-02 | 武汉科技大学 | An intelligent lawyer expert case answering device |
| CN108268637A (en) * | 2018-01-17 | 2018-07-10 | 广东蔚海数问大数据科技有限公司 | An intelligent voice correction recognition method, device and user terminal |
| CN108984800A (en) * | 2018-08-22 | 2018-12-11 | 广东小天才科技有限公司 | A voice search method and terminal equipment |
| CN109326284A (en) * | 2018-08-22 | 2019-02-12 | 深圳创维数字技术有限公司 | Voice search method, device and storage medium |
| CN109326284B (en) * | 2018-08-22 | 2022-06-10 | 深圳创维数字技术有限公司 | Voice search method, device and storage medium |
| CN108984800B (en) * | 2018-08-22 | 2020-10-16 | 广东小天才科技有限公司 | Voice question searching method and terminal equipment |
| CN111198936A (en) * | 2018-11-20 | 2020-05-26 | 北京嘀嘀无限科技发展有限公司 | Voice search method and device, electronic equipment and storage medium |
| CN111198936B (en) * | 2018-11-20 | 2023-09-15 | 北京嘀嘀无限科技发展有限公司 | Voice search method and device, electronic equipment and storage medium |
| CN111259170A (en) * | 2018-11-30 | 2020-06-09 | 北京嘀嘀无限科技发展有限公司 | Voice search method and device, electronic equipment and storage medium |
| CN110008471A (en) * | 2019-03-26 | 2019-07-12 | 北京博瑞彤芸文化传播股份有限公司 | A kind of intelligent semantic matching process based on phonetic conversion |
| CN110264994A (en) * | 2019-07-02 | 2019-09-20 | 珠海格力电器股份有限公司 | Voice synthesis method, electronic equipment and intelligent home system |
| CN110619879A (en) * | 2019-08-29 | 2019-12-27 | 深圳市梦网科技发展有限公司 | Voice recognition method and device |
| CN111475619A (en) * | 2020-03-31 | 2020-07-31 | 北京三快在线科技有限公司 | Text information correction method and device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104216906A (en) | Voice searching method and device | |
| US9905228B2 (en) | System and method of performing automatic speech recognition using local private data | |
| CN104142915B (en) | A kind of method and system adding punctuate | |
| CN103971684B (en) | A kind of add punctuate method, system and language model method for building up, device | |
| US20080154604A1 (en) | System and method for providing context-based dynamic speech grammar generation for use in search applications | |
| CN104143331B (en) | A kind of method and system adding punctuate | |
| CN104731959A (en) | Video abstraction generating method, device and system based on text webpage content | |
| JP2012230670A (en) | System, method, and computer program for correcting incorrect recognition by return | |
| CN107203526B (en) | Query string semantic demand analysis method and device | |
| CN103187052A (en) | Method and device for establishing linguistic model for voice recognition | |
| CN112017642B (en) | Speech recognition method, device, equipment and computer-readable storage medium | |
| CN108304424B (en) | Text keyword extraction method and text keyword extraction device | |
| CN104485107A (en) | Name voice recognition method, name voice recognition system and name voice recognition equipment | |
| CN111198936B (en) | Voice search method and device, electronic equipment and storage medium | |
| CN111414757A (en) | Text recognition method and device | |
| US20200243074A1 (en) | Information processor, information processing method, and program | |
| CN104282301A (en) | Voice command processing method and system | |
| US20140067400A1 (en) | Phonetic information generating device, vehicle-mounted information device, and database generation method | |
| CN101477565B (en) | Method and apparatus for confirming correctness of input character string in search engine | |
| CN109033427A (en) | The screening technique and device of stock, computer equipment and readable storage medium storing program for executing | |
| CN101149271B (en) | Crossing point crossing searching device | |
| US8886668B2 (en) | Navigation system with search-term boundary detection mechanism and method of operation thereof | |
| CN103530298A (en) | Information searching method and device | |
| US20160047670A1 (en) | Method and apparatus for navigation | |
| US9336317B2 (en) | System and method for searching aliases associated with an entity |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20141217 |