CN107526826A - Phonetic search processing method, device and server - Google Patents
Phonetic search processing method, device and server Download PDFInfo
- Publication number
- CN107526826A CN107526826A CN201710773346.9A CN201710773346A CN107526826A CN 107526826 A CN107526826 A CN 107526826A CN 201710773346 A CN201710773346 A CN 201710773346A CN 107526826 A CN107526826 A CN 107526826A
- Authority
- CN
- China
- Prior art keywords
- search
- language
- search statement
- statement
- type
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3343—Query execution using phonetics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR 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
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/18—Speech classification or search using natural language modelling
- G10L15/183—Speech classification or search using natural language modelling using context dependencies, e.g. language models
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
- G10L25/54—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination for retrieval
Abstract
The present invention proposes a kind of phonetic search processing method, device and server, wherein, this method includes:Obtain phonetic search sentence;Respectively according to N kind language models, while the search statement is identified, the language form belonging to the search statement is judged, wherein every kind of language model corresponds to a type of language respectively, N is the positive integer more than 1;When it is determined that the search statement belongs to the language of the i-th type, recognition result corresponding to the language model of i-th type is obtained;Scanned for according to the recognition result.Hereby it is achieved that identification and search to phonetic search sentence, improve the efficiency of phonetic search processing, reduce the stand-by period of user, improve Consumer's Experience.
Description
Technical field
The present invention relates to field of computer technology, more particularly to a kind of phonetic search processing method, device and server.
Background technology
With the development of internet and information technology, increasing user passes through the various information of internet hunt.
The current search engine with multilingual function of search, when being scanned for for user, is getting search phrase
After sentence, generally first search statement is identified, then the accuracy of recognition result sentenced according to conventional category of language
It is disconnected, if accuracy is relatively low, after switching language form, search statement is re-recognized, until the recognition result determined
Accuracy is higher, is scanned for further according to the recognition result of determination.
This way of search, complicated to the judging process of the sound-type of search statement, time-consuming, and then causes at search
Time-consuming for reason process, and search efficiency is low, poor user experience.
The content of the invention
It is contemplated that at least solves one of technical problem in correlation technique to a certain extent.
Therefore, the present invention proposes a kind of phonetic search processing method, the identification and search to phonetic search sentence are realized,
The efficiency of phonetic search processing is improved, reduces the stand-by period of user, improves Consumer's Experience.
The present invention also proposes a kind of phonetic search processing unit.
The present invention also proposes a kind of server.
The present invention also proposes a kind of computer-readable recording medium.
First aspect present invention embodiment proposes a kind of phonetic search processing method, including:Obtain phonetic search sentence;
Respectively according to N kind language models, while the search statement is identified, to the language form belonging to the search statement
Judged, wherein every kind of language model corresponds to a type of language respectively, N is the positive integer more than 1;It is determined that described search
When rope sentence belongs to the language of the i-th type, recognition result corresponding to the language model of i-th type is obtained;According to institute
Recognition result is stated to scan for.
The phonetic search processing method of the embodiment of the present invention, phonetic search sentence is obtained first, then respectively according to N kind languages
Model is sayed, while search statement is identified, the language form belonging to search statement is judged, it is determined that search phrase
When sentence belongs to the language of the i-th type, recognition result corresponding to the language model of the i-th type is obtained, is finally tied according to identification
Fruit scans for.Hereby it is achieved that identification and search to phonetic search sentence, improve the efficiency of phonetic search processing, subtract
Lack the stand-by period of user, improve Consumer's Experience.
Second aspect of the present invention embodiment proposes a kind of phonetic search processing unit, including:First acquisition module, is used for
Obtain phonetic search sentence;Judge module, for according to N kind language models, the search statement being identified same respectively
When, the language form belonging to the search statement is judged, wherein every kind of language model corresponds to a type of language respectively
Speech, N are the positive integer more than 1;Second acquisition module, for when it is determined that the search statement belongs to the language of the i-th type,
Obtain recognition result corresponding to the language model of i-th type;Search module, for being searched according to the recognition result
Rope.
The phonetic search processing unit of the embodiment of the present invention, phonetic search sentence is obtained first, then respectively according to N kind languages
Model is sayed, while search statement is identified, the language form belonging to search statement is judged, it is determined that search phrase
When sentence belongs to the language of the i-th type, recognition result corresponding to the language model of the i-th type is obtained, is finally tied according to identification
Fruit scans for.Hereby it is achieved that identification and search to phonetic search sentence, improve the efficiency of phonetic search processing, subtract
Lack the stand-by period of user, improve Consumer's Experience.
Third aspect present invention embodiment proposes a kind of server, including:
Memory, processor and storage on a memory and the computer program that can run on a processor, when the place
Manage the phonetic search processing method realized when device performs described program as described in relation to the first aspect.
Fourth aspect present invention embodiment proposes a kind of computer-readable recording medium, is stored thereon with computer journey
Sequence, phonetic search processing method as described in relation to the first aspect is realized when the program is executed by processor.
Brief description of the drawings
Of the invention above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments
Substantially and it is readily appreciated that, wherein:
Fig. 1 is the flow chart of the phonetic search processing method of one embodiment of the invention;
Fig. 2 is the flow chart of the phonetic search processing method of another embodiment of the present invention;
Fig. 3 is the structural representation of the phonetic search processing unit of one embodiment of the invention;
Fig. 4 is the structural representation of the phonetic search processing unit of another embodiment of the present invention.
Embodiment
Embodiments of the invention are described below in detail, the example of the embodiment is shown in the drawings, wherein from beginning to end
Same or similar label represents same or similar element or the element with same or like function.Below with reference to attached
The embodiment of figure description is exemplary, it is intended to for explaining the present invention, and is not considered as limiting the invention.
The current search engine with multilingual function of search, when being scanned for for user, is getting search phrase
After sentence, generally first search statement is identified, then the accuracy of recognition result sentenced according to conventional category of language
It is disconnected, if accuracy is relatively low, after switching language form, search statement is re-recognized, until the recognition result determined
Accuracy is higher, is scanned for further according to the recognition result of determination.This way of search, to sentencing for the sound-type of search statement
Certainly process is complicated, and time-consuming, and then causes search process process time-consuming, and search efficiency is low, poor user experience.
Various embodiments of the present invention are getting phonetic search in view of the above-mentioned problems, propose a kind of phonetic search processing method
After voice, respectively according to multilingual model, while search voice is identified, to the language form belonging to search statement
Judged, wherein, every kind of language model corresponds to a type of language, it is determined that during language belonging to search statement, obtains
The recognition result of language model corresponding with the language form, so as to be scanned for according to recognition result.Hereby it is achieved that to language
The identification and search of sound search statement, the efficiency of phonetic search processing is improved, reduces the stand-by period of user, improves use
Experience at family.
Below with reference to the accompanying drawings phonetic search processing method, device and the server of the embodiment of the present invention are described.
Fig. 1 is the flow chart of the phonetic search processing method of one embodiment of the invention.
As shown in figure 1, the phonetic search processing method includes:
Step 101, phonetic search sentence is obtained.
Wherein, the executive agent of phonetic search processing method provided in an embodiment of the present invention is provided in an embodiment of the present invention
Phonetic search processing unit, the device can be configured in any server with function of search, with the voice to acquisition
Search statement scans for.
Specifically, the voice-input devices such as microphone can be pre-set in the terminal, so as to need to search for letter in user
During breath, terminal can obtain the phonetic search sentence of user's input by voice-input device, and search statement is sent into language
Sound search process device.
Step 102, respectively according to N kind language models, while search statement is identified, to belonging to search statement
Language form is judged, wherein every kind of language model corresponds to a type of language respectively, N is the positive integer more than 1.
Wherein, N kinds language model, can also may be used including language model corresponding to existing all types of language difference
Including the multilingual model determined as needed, not to be restricted herein.
It is understood that respectively according to N kind language models, before search statement is identified, it is also necessary to determine N kinds
Language model.Specifically, N kind language models can be determined according to various ways.
For example it can be determined according to historical search daily record, i.e. before step 102, can also include:According to going through
History searches for daily record, determines N kind language models.
Wherein, historical search daily record can be historical search record when user is scanned for using terminal or other go through
History search record, is not restricted herein.
Specifically, can be recorded according to historical search, when determining that the user belonging to terminal scans for, which is often used
The language of type scans for, so as to the search rate according to corresponding to the search statement of each language form difference, it is determined that will be to searching
The N kind language models that rope sentence is identified.
During specific implementation, N value can be pre-set, it is determined that the search statement of different language type is corresponding respectively searches
After rope frequency, language form can be sorted by search rate order from high to low, so as to which N kind class of languages above will be come
Language model corresponding to type difference, is defined as N kind language models search statement to be identified.
As an example it is assumed that N is 2, recorded according to the historical search in a period of time, determine that the user belonging to terminal utilizes
The frequency that Chinese scans for is 200, and the frequency scanned for using English is 300, and the frequency scanned for using Korean is
10.Then can according to Chinese, English, Korean respectively corresponding to search rate, by the language model of Chinese and English type, it is determined that
For 2 kinds of language models search statement to be identified.
Or search rate threshold value can be pre-set, so as to distinguish in the search statement that different language type is determined
, can be by search rate more than language model corresponding to the language form of predetermined threshold value after corresponding search rate, being defined as will
The language model that search statement is identified.
Furthermore it is possible to according to the history use information of terminal, it is determined that N kind language moulds search statement to be identified
Type.Wherein, history use information, can be use information of the user to each application in terminal, and terminal passes through within a period of time
Positional information where often, etc..
As an example it is assumed that the positional information according to where terminal, determines that user often travels to and fro between the U.S. and China, and it is beautiful
The conventional language form of user corresponding to state and China is English and Chinese, then can be by language corresponding to Chinese and English difference
Model, it is defined as language model search statement to be identified.
During specific implementation, the language form belonging to search statement can be sentenced by following steps 102a-102b
It is disconnected.
Step 102a, determine the characteristic vector of search statement.
Wherein, characteristic vector, for characterizing the feature of the phonetic search sentence got.
, can be by mel cepstrum coefficients, linear specifically, after phonetic search processing unit gets phonetic search sentence
Predict cepstrum coefficient, a variety of methods of Multimedia Content Description Interface etc., it is determined that the feature of the phonetic search sentence got to
Amount.
Step 102b, according to characteristic vector and the matching degree of default each language form model, determine belonging to search statement
Language form.
Specifically, it can be respectively trained to obtain each class of languages previously according to the history language material of substantial amounts of all kinds language
Pattern type, so as to which it is determined that after the characteristic vector of the phonetic search sentence obtained, characteristic vector can be inputted to each language form
Model carries out verification marking, and by the language form model of highest scoring, i.e. the matching degree highest class of languages with characteristic vector
Language form corresponding to pattern type, it is defined as the language form belonging to search statement.
Step 103, when it is determined that search statement belongs to the language of the i-th type, the language model pair of the i-th type is obtained
The recognition result answered.
Step 104, scanned for according to recognition result.
Specifically, the search statement of different language type can be pre-set, corresponding different resources bank, so as in basis
Language form belonging to search statement, after the recognition result for obtaining corresponding with language form language model, can with language
Scanned in resources bank corresponding to type.
Can be general search, i.e., it should be noted that when being scanned in resources bank corresponding to every kind of language form
Scanned in the unstructured resources storehouse of corresponding language form;Can also be vertical search, i.e., in corresponding language form
Unstructured resource scans in storehouse.
It is understood that phonetic search processing method provided in an embodiment of the present invention, by will be to belonging to search statement
The process that is judged of language form, with the process that search statement is identified while carrying out, phonetic search can be improved
The efficiency of processing, reduce the stand-by period of user.And by according to N kind language models, search statement being identified simultaneously,
After determining the language form belonging to search statement, the knowledge of language model corresponding with language form is obtained from a variety of recognition results
Other result, it is ensured that to the accuracy and reliability of the recognition result of phonetic search sentence.
In addition, in embodiments of the present invention, first the language form belonging to search statement can also be judged, Ran Hougen
It is judged that language model corresponding to the language form gone out, search statement is identified, to obtain recognition result, so as to according to
The recognition result of language model corresponding to language form scans for.
The phonetic search processing method of the embodiment of the present invention, phonetic search sentence is obtained first, then respectively according to N kind languages
Model is sayed, while search statement is identified, the language form belonging to search statement is judged, it is determined that search phrase
When sentence belongs to the language of the i-th type, recognition result corresponding to the language model of the i-th type is obtained, is finally tied according to identification
Fruit scans for.Hereby it is achieved that identification and search to phonetic search sentence, improve the efficiency of phonetic search processing, subtract
Lack the stand-by period of user, improve Consumer's Experience.
By above-mentioned analysis, after phonetic search voice is got, respectively according to multilingual model, to search phrase
While sound is identified, the language form belonging to search statement can be judged, it is determined that language belonging to search statement
Yan Shi, then the recognition result of language model corresponding with the language form is obtained, so as to be scanned for according to recognition result.In reality
During border uses, when judging the language form belonging to search statement, it can be carried out according only to the Partial Fragment of search statement
Judge, with reference to Fig. 2, be specifically described for the above situation.
Fig. 2 is the flow chart of the phonetic search processing method of another embodiment of the present invention.
As shown in Fig. 2 this method includes:
Step 201, phonetic search sentence is obtained.
Wherein, the specific implementation process and principle of above-mentioned steps 201, the detailed description of above-described embodiment is referred to, this
Place does not repeat.
Step 202, according to default rule, the fragment of interception preset length from search statement.
Wherein, default rule, for referring to the rule of regulation fragment of interception preset length from search statement.
Step 203, respectively according to N kind language models, while search statement is identified, according to the piece of preset length
Section, judges the language form belonging to search statement.
Wherein, every kind of language model corresponds to a type of language respectively, and N is the positive integer more than 1.
Preset length, it can arbitrarily set, be searched as long as the fragment of utility preset length can interpolate that out as needed
Language form belonging to rope sentence.Specifically, preset length could be arranged to regular length, such as 3 seconds (s), 4s;
It can specifically be set according to factors such as the length of search statement, such as be arranged to 1/3, etc. of search statement length, herein
It is not restricted.
Specifically, can be by following steps 203a-203b, according to the fragment of preset length, to belonging to search statement
Language form is judged.
Step 203a, according to the fragment of preset length, determine the characteristic vector of search statement.
Specifically, after phonetic search processing unit gets the fragment of preset length, mel cepstrum coefficients, line can be passed through
Property prediction cepstrum coefficient, a variety of methods of Multimedia Content Description Interface etc., determine the characteristic vector of phonetic search sentence.
Step 203b, according to characteristic vector and the matching degree of default each language form model, determine belonging to search statement
Language form.
Specifically, it can be respectively trained to obtain each class of languages previously according to the history language material of substantial amounts of all kinds language
Pattern type, so as to it is determined that after the characteristic vector of phonetic search sentence, characteristic vector be inputted to each language form model and entered
Row verification marking, and by the language form model of highest scoring, i.e. the matching degree highest language form model with characteristic vector
Corresponding language form, it is defined as the language form belonging to search statement.
It should be noted that the process that search statement is identified, it is referred to according to N kind language models respectively
The specific descriptions of embodiment are stated, here is omitted.In addition, step 202 and step 203 can be carried out simultaneously.
Step 204, when it is determined that search statement belongs to the language of the i-th type, the language model pair of the i-th type is obtained
The recognition result answered.
Step 205, scanned for according to recognition result.
Wherein, above-mentioned steps 204-205 specific implementation process and principle, it is referred to retouching in detail for above-described embodiment
State, do not repeat herein.
It is understood that due to according to language model, it is necessary to utilize complete language when search statement is identified
When being identified, and the language form belonging to search statement being judged, by intercepting preset length from search statement
Fragment judged, therefore, deterministic process it is time-consuming very short, before language model is to the end of identification of search statement, you can
Judge the language form belonging to search statement.So, in embodiments of the present invention, the language form belonging to search statement is determined
Afterwards, the identification process of language model corresponding to other Languages type can also be stopped.I.e., in step 204, search statement is determined
Belong to after the language of the i-th type, can also include:
Terminate according to other N-1 kinds language models, the process that search statement is identified.
Specifically, after determining the language form belonging to search statement, by stopping language mould corresponding to other Languages type
The identification process of type, it is possible to reduce the waste of resource.
The phonetic search processing method of the embodiment of the present invention, can be according to default rule after phonetic search sentence is obtained
Then, the fragment of preset length is intercepted from search statement, then according to N kind language models, search statement is being known respectively
It is other that according to the fragment of preset length, the language form belonging to search statement is judged simultaneously, so as to it is determined that search phrase
When sentence belongs to the language of the i-th type, recognition result corresponding to the language model of the i-th type can be obtained, with according to identification
As a result scan for.Hereby it is achieved that identification and search to phonetic search sentence, improve the efficiency that phonetic search is handled,
Reduce the stand-by period of user, improve Consumer's Experience.
Fig. 3 is the structural representation of the phonetic search processing unit of one embodiment of the invention.
As shown in figure 3, the phonetic search processing unit includes:
First acquisition module 31, for obtaining phonetic search sentence;
Judge module 32, for respectively according to N kind language models, while search statement is identified, to search phrase
Language form belonging to sentence is judged that, wherein every kind of language model corresponds to a type of language respectively, N is more than 1 just
Integer;
Second acquisition module 33, for when it is determined that search statement belongs to the language of the i-th type, obtaining the i-th type
Language model corresponding to recognition result;
Search module 34, for being scanned for according to recognition result.
Specifically, the phonetic search processing unit that the present embodiment provides, can be configured in arbitrarily with function of search
In server, for performing the phonetic search processing method as shown in above-mentioned embodiment, entered with the phonetic search sentence to acquisition
Row search.
In a kind of possible way of realization of the embodiment of the present application, above-mentioned judge module 32, including:
First determining unit, for determining the characteristic vector of search statement;
Second determining unit, for the matching degree according to characteristic vector and default each language form model, it is determined that search
Language form belonging to sentence.
In the alternatively possible way of realization of the embodiment of the present application, above-mentioned first determining unit, it is specifically used for:
According to default rule, the fragment of interception preset length from search statement;
According to the fragment of preset length, the characteristic vector of search statement is determined.
It should be noted that the foregoing explanation to phonetic search processing method embodiment is also applied for the embodiment
Phonetic search processing unit, here is omitted.
The phonetic search processing unit of the embodiment of the present invention, phonetic search sentence is obtained first, then respectively according to N kind languages
Model is sayed, while search statement is identified, the language form belonging to search statement is judged, it is determined that search phrase
When sentence belongs to the language of the i-th type, recognition result corresponding to the language model of the i-th type is obtained, is finally tied according to identification
Fruit scans for.Hereby it is achieved that identification and search to phonetic search sentence, improve the efficiency of phonetic search processing, subtract
Lack the stand-by period of user, improve Consumer's Experience.
Fig. 4 is the structural representation of the phonetic search processing unit of another embodiment of the present invention.
As shown in figure 4, on the basis of Fig. 3, the phonetic search processing unit, in addition to:
Determining module 41, for according to historical search daily record, determining N kind language models.
Terminate module 42, for terminating according to other N-1 kinds language models, the process that search statement is identified.
It should be noted that the foregoing explanation to phonetic search processing method embodiment is also applied for the embodiment
Phonetic search processing unit, here is omitted.
The phonetic search processing unit of the embodiment of the present invention, phonetic search sentence is obtained first, then respectively according to N kind languages
Model is sayed, while search statement is identified, the language form belonging to search statement is judged, it is determined that search phrase
When sentence belongs to the language of the i-th type, recognition result corresponding to the language model of the i-th type is obtained, is finally tied according to identification
Fruit scans for.Hereby it is achieved that identification and search to phonetic search sentence, improve the efficiency of phonetic search processing, subtract
Lack the stand-by period of user, improve Consumer's Experience.
Third aspect present invention embodiment proposes a kind of server, including:
Memory, processor and storage on a memory and the computer program that can run on a processor, when above-mentioned place
Manage when device performs described program and realize such as the phonetic search processing method in previous embodiment.
Fourth aspect present invention embodiment proposes a kind of computer-readable recording medium, is stored thereon with computer journey
Sequence, realized when the program is executed by processor such as the phonetic search processing method in previous embodiment.
Fifth aspect present invention embodiment proposes a kind of computer program product, when in the computer program product
When instruction is by computing device, perform such as the phonetic search processing method in previous embodiment.
In the description of this specification, reference term " one embodiment ", " some embodiments ", " example ", " specifically show
The description of example " or " some examples " etc. means specific features, structure, material or the spy for combining the embodiment or example description
Point is contained at least one embodiment or example of the present invention.In this manual, to the schematic representation of above-mentioned term not
Identical embodiment or example must be directed to.Moreover, specific features, structure, material or the feature of description can be with office
Combined in an appropriate manner in one or more embodiments or example.In addition, in the case of not conflicting, the skill of this area
Art personnel can be tied the different embodiments or example and the feature of different embodiments or example described in this specification
Close and combine.
In addition, term " first ", " second " are only used for describing purpose, and it is not intended that instruction or hint relative importance
Or the implicit quantity for indicating indicated technical characteristic.Thus, define " first ", the feature of " second " can be expressed or
Implicitly include at least one this feature.In the description of the invention, " multiple " are meant that at least two, such as two, three
It is individual etc., unless otherwise specifically defined.
Any process or method described otherwise above description in flow chart or herein is construed as, and represents to include
Module, fragment or the portion of the code of the executable instruction of one or more the step of being used to realize custom logic function or process
Point, and the scope of the preferred embodiment of the present invention includes other realization, wherein can not press shown or discuss suitable
Sequence, including according to involved function by it is basic simultaneously in the way of or in the opposite order, carry out perform function, this should be of the invention
Embodiment person of ordinary skill in the field understood.
Expression or logic and/or step described otherwise above herein in flow charts, for example, being considered use
In the order list for the executable instruction for realizing logic function, may be embodied in any computer-readable medium, for
Instruction execution system, device or equipment (such as computer based system including the system of processor or other can be held from instruction
The system of row system, device or equipment instruction fetch and execute instruction) use, or combine these instruction execution systems, device or set
It is standby and use.For the purpose of this specification, " computer-readable medium " can any can be included, store, communicate, propagate or pass
Defeated program is for instruction execution system, device or equipment or the dress used with reference to these instruction execution systems, device or equipment
Put.The more specifically example (non-exhaustive list) of computer-readable medium includes following:Electricity with one or more wiring
Connecting portion (electronic installation), portable computer diskette box (magnetic device), random access memory (RAM), read-only storage
(ROM), erasable edit read-only storage (EPROM or flash memory), fiber device, and portable optic disk is read-only deposits
Reservoir (CDROM).In addition, computer-readable medium, which can even is that, to print the paper of described program thereon or other are suitable
Medium, because can then enter edlin, interpretation or if necessary with it for example by carrying out optical scanner to paper or other media
His suitable method is handled electronically to obtain described program, is then stored in computer storage.
It should be appreciated that each several part of the present invention can be realized with hardware, software, firmware or combinations thereof.Above-mentioned
In embodiment, software that multiple steps or method can be performed in memory and by suitable instruction execution system with storage
Or firmware is realized.If, and in another embodiment, can be with well known in the art for example, realized with hardware
Any one of row technology or their combination are realized:With the logic gates for realizing logic function to data-signal
Discrete logic, have suitable combinational logic gate circuit application specific integrated circuit, programmable gate array (PGA), scene
Programmable gate array (FPGA) etc..
Those skilled in the art are appreciated that to realize all or part of step that above-described embodiment method carries
Suddenly it is that by program the hardware of correlation can be instructed to complete, described program can be stored in a kind of computer-readable storage medium
In matter, the program upon execution, including one or a combination set of the step of embodiment of the method.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing module, can also
That unit is individually physically present, can also two or more units be integrated in a module.Above-mentioned integrated mould
Block can both be realized in the form of hardware, can also be realized in the form of software function module.The integrated module is such as
Fruit is realized in the form of software function module and as independent production marketing or in use, can also be stored in a computer
In read/write memory medium.
Storage medium mentioned above can be read-only storage, disk or CD etc..Although have been shown and retouch above
Embodiments of the invention are stated, it is to be understood that above-described embodiment is exemplary, it is impossible to be interpreted as the limit to the present invention
System, one of ordinary skill in the art can be changed to above-described embodiment, change, replace and become within the scope of the invention
Type.
Claims (12)
- A kind of 1. phonetic search processing method, it is characterised in that including:Obtain phonetic search sentence;Respectively according to N kind language models, while the search statement is identified, to the language belonging to the search statement Type is judged, wherein every kind of language model corresponds to a type of language respectively, N is the positive integer more than 1;When it is determined that the search statement belongs to the language of the i-th type, corresponding to the language model of acquisition i-th type Recognition result;Scanned for according to the recognition result.
- 2. the method as described in claim 1, it is characterised in that it is described according to N kind language models, the search statement is carried out Before identification, in addition to:According to historical search daily record, the N kinds language model is determined.
- 3. the method as described in claim 1, it is characterised in that the language form to belonging to the search statement is sentenced It is disconnected, including:Determine the characteristic vector of the search statement;According to the characteristic vector and the matching degree of default each language form model, the language belonging to the search statement is determined Type.
- 4. method as claimed in claim 3, it is characterised in that the characteristic vector for determining the search statement, including:According to default rule, the fragment of interception preset length from the search statement;According to the fragment of the preset length, the characteristic vector of the search statement is determined.
- 5. the method as described in claim 1-4 is any, it is characterised in that described to determine that the search statement belongs to the i-th species After the language of type, in addition to:Terminate according to other N-1 kinds language models, the process that the search statement is identified.
- A kind of 6. phonetic search processing unit, it is characterised in that including:First acquisition module, for obtaining phonetic search sentence;Judge module, for respectively according to N kind language models, while the search statement is identified, to the search Language form belonging to sentence is judged, wherein every kind of language model corresponds to a type of language respectively, N is more than 1 Positive integer;Second acquisition module, for when it is determined that the search statement belongs to the language of the i-th type, obtaining i-th species Recognition result corresponding to the language model of type;Search module, for being scanned for according to the recognition result.
- 7. device as claimed in claim 6, it is characterised in that also include:Determining module, for according to historical search daily record, determining the N kinds language model.
- 8. device as claimed in claim 6, it is characterised in that the judge module, including:First determining unit, for determining the characteristic vector of the search statement;Second determining unit, for the matching degree according to the characteristic vector and default each language form model, it is determined that described Language form belonging to search statement.
- 9. device as claimed in claim 8, it is characterised in that first determining unit, be specifically used for:According to default rule, the fragment of interception preset length from the search statement;According to the fragment of the preset length, the characteristic vector of the search statement is determined.
- 10. the device as described in claim 6-9 is any, it is characterised in that also include:Terminate module, for terminating according to other N-1 kinds language models, the process that the search statement is identified.
- 11. a kind of server, including:Memory, processor and storage are on a memory and the computer program that can run on a processor, it is characterised in that institute The phonetic search processing method as described in any in claim 1-5 is realized when stating computing device described program.
- 12. a kind of computer-readable recording medium, is stored thereon with computer program, it is characterised in that the program is by processor The phonetic search processing method as described in any in claim 1-5 is realized during execution.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710773346.9A CN107526826B (en) | 2017-08-31 | 2017-08-31 | Voice search processing method and device and server |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710773346.9A CN107526826B (en) | 2017-08-31 | 2017-08-31 | Voice search processing method and device and server |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107526826A true CN107526826A (en) | 2017-12-29 |
CN107526826B CN107526826B (en) | 2021-09-17 |
Family
ID=60683057
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710773346.9A Active CN107526826B (en) | 2017-08-31 | 2017-08-31 | Voice search processing method and device and server |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107526826B (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108428446A (en) * | 2018-03-06 | 2018-08-21 | 北京百度网讯科技有限公司 | Audio recognition method and device |
CN108899035A (en) * | 2018-08-02 | 2018-11-27 | 科大讯飞股份有限公司 | Message treatment method and device |
CN110853647A (en) * | 2018-07-27 | 2020-02-28 | Tcl集团股份有限公司 | Video searching method, video playing terminal and storage medium |
CN111161706A (en) * | 2018-10-22 | 2020-05-15 | 阿里巴巴集团控股有限公司 | Interaction method, device, equipment and system |
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 |
CN111369978A (en) * | 2018-12-26 | 2020-07-03 | 北京搜狗科技发展有限公司 | Data processing method and device and data processing device |
CN111914070A (en) * | 2019-05-09 | 2020-11-10 | 上海触乐信息科技有限公司 | Intelligent information prompting assistant system, information prompting method and terminal equipment |
CN112133283A (en) * | 2019-06-24 | 2020-12-25 | 武汉慧人信息科技有限公司 | Voice response system design in multi-language environment |
CN112925889A (en) * | 2021-02-26 | 2021-06-08 | 北京声智科技有限公司 | Natural language processing method, device, electronic equipment and storage medium |
CN113380224A (en) * | 2021-06-04 | 2021-09-10 | 北京字跳网络技术有限公司 | Language determination method and device, electronic equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102043833A (en) * | 2010-11-25 | 2011-05-04 | 北京搜狗科技发展有限公司 | Search method and device based on query word |
CN104239459A (en) * | 2014-09-02 | 2014-12-24 | 百度在线网络技术(北京)有限公司 | Voice search method, voice search device and voice search system |
WO2015118645A1 (en) * | 2014-02-06 | 2015-08-13 | 三菱電機株式会社 | Speech search device and speech search method |
CN106407332A (en) * | 2016-09-05 | 2017-02-15 | 北京百度网讯科技有限公司 | Artificial intelligence-based search method and apparatus |
-
2017
- 2017-08-31 CN CN201710773346.9A patent/CN107526826B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102043833A (en) * | 2010-11-25 | 2011-05-04 | 北京搜狗科技发展有限公司 | Search method and device based on query word |
WO2015118645A1 (en) * | 2014-02-06 | 2015-08-13 | 三菱電機株式会社 | Speech search device and speech search method |
CN105981099A (en) * | 2014-02-06 | 2016-09-28 | 三菱电机株式会社 | Speech search device and speech search method |
CN104239459A (en) * | 2014-09-02 | 2014-12-24 | 百度在线网络技术(北京)有限公司 | Voice search method, voice search device and voice search system |
CN106407332A (en) * | 2016-09-05 | 2017-02-15 | 北京百度网讯科技有限公司 | Artificial intelligence-based search method and apparatus |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10978047B2 (en) | 2018-03-06 | 2021-04-13 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for recognizing speech |
CN108428446A (en) * | 2018-03-06 | 2018-08-21 | 北京百度网讯科技有限公司 | Audio recognition method and device |
CN108428446B (en) * | 2018-03-06 | 2020-12-25 | 北京百度网讯科技有限公司 | Speech recognition method and device |
CN110853647A (en) * | 2018-07-27 | 2020-02-28 | Tcl集团股份有限公司 | Video searching method, video playing terminal and storage medium |
CN108899035A (en) * | 2018-08-02 | 2018-11-27 | 科大讯飞股份有限公司 | Message treatment method and device |
CN111161706A (en) * | 2018-10-22 | 2020-05-15 | 阿里巴巴集团控股有限公司 | Interaction method, device, equipment and system |
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 |
CN111369978A (en) * | 2018-12-26 | 2020-07-03 | 北京搜狗科技发展有限公司 | Data processing method and device and data processing device |
CN111914070A (en) * | 2019-05-09 | 2020-11-10 | 上海触乐信息科技有限公司 | Intelligent information prompting assistant system, information prompting method and terminal equipment |
CN112133283A (en) * | 2019-06-24 | 2020-12-25 | 武汉慧人信息科技有限公司 | Voice response system design in multi-language environment |
CN112925889A (en) * | 2021-02-26 | 2021-06-08 | 北京声智科技有限公司 | Natural language processing method, device, electronic equipment and storage medium |
CN113380224A (en) * | 2021-06-04 | 2021-09-10 | 北京字跳网络技术有限公司 | Language determination method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN107526826B (en) | 2021-09-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107526826A (en) | Phonetic search processing method, device and server | |
CN107329967B (en) | Question answering system and method based on deep learning | |
CN108877778B (en) | Sound end detecting method and equipment | |
US10991366B2 (en) | Method of processing dialogue query priority based on dialog act information dependent on number of empty slots of the query | |
US11531818B2 (en) | Device and method for machine reading comprehension question and answer | |
CN110472224B (en) | Quality of service detection method, apparatus, computer device and storage medium | |
CN106528845A (en) | Artificial intelligence-based searching error correction method and apparatus | |
CN105336322A (en) | Polyphone model training method, and speech synthesis method and device | |
CN107679032A (en) | Voice changes error correction method and device | |
CN107679033A (en) | Text punctuate location recognition method and device | |
CN110197279B (en) | Transformation model training method, device, equipment and storage medium | |
CN104464751B (en) | The detection method and device for rhythm problem of pronouncing | |
CN106504768A (en) | Phone testing audio frequency classification method and device based on artificial intelligence | |
CN109616096A (en) | Construction method, device, server and the medium of multilingual tone decoding figure | |
CN107464566A (en) | Audio recognition method and device | |
CN106571139A (en) | Artificial intelligence based voice search result processing method and device | |
CN109976702A (en) | A kind of audio recognition method, device and terminal | |
CN108091324A (en) | Tone recognition methods, device, electronic equipment and computer readable storage medium | |
CN105845133A (en) | Voice signal processing method and apparatus | |
CN109448704A (en) | Construction method, device, server and the storage medium of tone decoding figure | |
Ando et al. | Hierarchical LSTMs with Joint Learning for Estimating Customer Satisfaction from Contact Center Calls. | |
CN105279227A (en) | Voice search processing method and device of homonym | |
CN106649250A (en) | Method and device for identifying emotional new words | |
CN107203265A (en) | Information interacting method and device | |
CN104750677A (en) | Speech translation apparatus, speech translation method and speech translation program |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |