CN107145509A - A kind of information search method and its equipment - Google Patents
A kind of information search method and its equipment Download PDFInfo
- Publication number
- CN107145509A CN107145509A CN201710194322.8A CN201710194322A CN107145509A CN 107145509 A CN107145509 A CN 107145509A CN 201710194322 A CN201710194322 A CN 201710194322A CN 107145509 A CN107145509 A CN 107145509A
- Authority
- CN
- China
- Prior art keywords
- field
- information
- lexicon
- voice messaging
- multimedia messages
- 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
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000011218 segmentation Effects 0.000 claims abstract description 35
- 238000003672 processing method Methods 0.000 description 4
- 238000003860 storage Methods 0.000 description 4
- 238000010801 machine learning Methods 0.000 description 3
- 238000001727 in vivo Methods 0.000 description 2
- 239000004575 stone Substances 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/43—Querying
- G06F16/438—Presentation of query results
- G06F16/4387—Presentation of query results by the use of playlists
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; 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
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Acoustics & Sound (AREA)
- Signal Processing (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention discloses a kind of information search method and its equipment, described information searching method includes：Inputted voice messaging is obtained, and the voice messaging is identified processing to generate the corresponding text message of the voice messaging；The text message is carried out word segmentation processing to obtain at least one field information in the text message；The critical field and destination field of the voice messaging are obtained at least one described field information using default lexicon；The corresponding multimedia messages list of the voice messaging is searched for using the critical field and the destination field, and exports the multimedia messages list.Using the present invention, natural-sounding can be converted into text message, the corresponding multimedia messages of voice messaging are searched for further according to the field information in text message, improve the program request success rate that terminal device recognizes VOD system based on natural-sounding.
Description
Technical field
The present invention relates to technical field of voice recognition, more particularly to a kind of information search method and its equipment.
Background technology
With scientific and technological constantly development and perfection, the terminal device such as smart mobile phone, wearable device and tablet personal computer is
Become a part indispensable in people's life.Application of the speech recognition technology in various terminal equipment also increasingly into
It is ripe, wherein, natural-sounding identification can recognize the voice that user inputs according to personal speech habits because of it, and application prospect is more
Extensively.In the prior art, the speech recognition system of terminal device can be with the voice of recognition command formula (for example：Recognize singer's name, song
Song name or movie and television play etc.), the program request to multimedia messages is realized, for example, requesting song or movies on demand etc., however, when with certainly
During right playing speech on demand multimedia messages, the speech recognition system in terminal device is due to the keyword being unable in accurate understanding voice
Information, reduces the success rate of program request.
The content of the invention
In view of this, the embodiment of the present invention provides a kind of information search method and its equipment, can convert natural-sounding
For text message, the corresponding multimedia messages of voice messaging are searched for further according to the field information in text message, terminal is improved
The VOD system program request success rate that equipment is recognized based on natural-sounding.
In order to solve the above-mentioned technical problem, the embodiments of the invention provide a kind of information search method, methods described includes：
Inputted voice messaging is obtained, and the voice messaging is identified processing to generate the voice messaging pair
The text message answered；
The text message is carried out word segmentation processing to obtain at least one field information in the text message；
The critical field and mesh of the voice messaging are obtained at least one described field information using default lexicon
Field；
The corresponding multimedia messages list of the voice messaging is searched for using the critical field and the destination field, and
Export the multimedia messages list.
Correspondingly, the embodiment of the present invention additionally provides a kind of information search equipment, and the equipment includes：
Text message generation unit, for obtaining inputted voice messaging, and is identified place to the voice messaging
Manage to generate the corresponding text message of the voice messaging；
Field information acquiring unit, for being obtained to text message progress word segmentation processing in the text message
At least one field information；
Field information taxon, for obtaining institute's predicate at least one described field information using default lexicon
The critical field and destination field of message breath；
Media information output unit, for searching for the voice messaging pair using the critical field and the destination field
The multimedia messages list answered, and export the multimedia messages list.
In embodiments of the present invention, by obtaining inputted voice messaging, voice messaging is identified processing with life
Carry out word segmentation processing to obtain at least one word in text message into the corresponding text message of voice messaging, and to text message
Segment information, the critical field and destination field of voice messaging is obtained using default lexicon, most at least one field information
Afterwards using critical field and the corresponding multimedia messages list of destination field search voice messaging, and export multimedia messages row
Table.By the way that natural-sounding is converted into text message, and language is obtained to text message progress word segmentation processing and terminology match processing
Message ceases corresponding critical field and destination field, then corresponding to matchmaker using critical field and destination field search voice messaging
Body information, improves the program request success rate that terminal device recognizes VOD system based on natural-sounding.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing needed to use in technology description to be briefly described, it should be apparent that, drawings in the following description are only this hair
Some bright embodiments, for those of ordinary skill in the art, on the premise of not paying creative work, can be with root
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is a kind of schematic flow sheet of information search method provided in an embodiment of the present invention；
Fig. 2 is the schematic flow sheet of another information search method provided in an embodiment of the present invention；
Fig. 3 is a kind of structural representation of information search equipment provided in an embodiment of the present invention；
Fig. 4 is the structural representation of another information search equipment provided in an embodiment of the present invention；
Fig. 5 is the structural representation of field information taxon provided in an embodiment of the present invention；
Fig. 6 is the structural representation of media information output unit provided in an embodiment of the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Based on this
Embodiment in invention, it is every other that those of ordinary skill in the art are obtained on the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
Information search method provided in an embodiment of the present invention can apply to terminal device and utilize natural-sounding identifying system
Free online multimedium information is (for example：By recognizing natural-sounding, requesting songs, movie and television play, variety show etc.) application scenarios in,
For example：By obtaining inputted voice messaging, voice messaging is identified processing to generate the corresponding text of voice messaging
Information, and word segmentation processing is carried out to obtain at least one field information in text message to text message, using default vocabulary
Storehouse obtains the critical field and destination field of voice messaging at least one field information, finally using critical field and purpose
The corresponding multimedia messages list of field searches voice messaging, and export multimedia messages list.By the way that natural-sounding is converted
For text message, and carry out word segmentation processing to text message and terminology match processing obtain the corresponding critical field of voice messaging and
Destination field, then it is corresponding to media information using critical field and destination field search voice messaging, improve terminal device
The program request success rate of the VOD system recognized based on natural-sounding.
Involved information search equipment can include tablet personal computer, smart mobile phone, wearable set in the embodiment of the present invention
Standby and mobile internet device (MID) etc. possesses the terminal device of natural-sounding identification function.
Below in conjunction with accompanying drawing 1 and accompanying drawing 2, information search method provided in an embodiment of the present invention is described in detail.
Fig. 1 is a kind of schematic flow sheet of information search method provided in an embodiment of the present invention.As shown in figure 1, of the invention
Method described in embodiment may comprise steps of S101- steps S104.
S101, obtains inputted voice messaging, and the voice messaging is identified processing to generate the voice
The corresponding text message of information.
Specifically, described information search equipment can obtain the voice messaging that user is inputted.It is understood that described
Voice messaging can be natural-sounding information, and the natural-sounding is different from instruction type voice, be used certainly by natural-sounding people
The hint expression that oneself usual tone, usual word will be expressed comes out (for example, ' playing the most popular folk rhyme of Zhang Lei '), and refers to
The formula voice of order is the order word (for example, turn off the light, turn on light, advance, retreat) that machine can be recognized.
Further, described information search equipment can be using default voice recognition processing method (for example, based on depth
Speech recognition algorithm, fuzzy recognition algorithm or the computer network instrument of neutral net (Deep Neural Networks, DNN)
Bag (Computational Network Toolkit, CNTK) etc.) voice messaging is identified processing, and generate institute
State the corresponding text message of voice messaging.
S102, word segmentation processing is carried out to the text message to obtain the letter of at least one field in the text message
Breath.
Specifically, described information search equipment can the text message corresponding to the voice messaging carry out at participle
Manage to obtain at least one field information in the text message.It is understood that the word segmentation processing can be based on
It is the segmenting method of text matches, the segmenting method (Conditional Random Field, CRF) based on condition random field, general
Rate statistic algorithm or machine learning method etc.., can be by it is understood that the text message is after the word segmentation processing
It is divided at least one field information.For example：The text message is " playing the most popular folk rhyme of Zhang Lei ", then by the participle
After processing, the field information that text information is divided into can be " play Zhang Lei most it is popular folk rhyme ".
S103, the critical field of the voice messaging is obtained using default lexicon at least one described field information
And destination field.
Specifically, described information search equipment can be obtained using default lexicon at least one described field information
The critical field and destination field of the voice messaging.It is understood that the destination field can be the voice messaging
In core vocabulary, the critical field can associated with the destination field be used to aid in the destination field to describe
The vocabulary of the voice messaging.For example, when at least one described field information that the voice messaging is obtained after treatment is
" play Zhang Lei most it is popular folk rhyme " when, described information search equipment is using the default lexicon in the field information
The destination field of acquisition can be " folk rhyme ", and critical field can be " Zhang Lei is most popular ".
It is understood that the default lexicon can described information search equipment inputted by obtaining user
Vocabulary is trained, and classification processing is carried out to the training vocabulary based on default syntactic category mode, it is finally raw after classification processing
Into lexicon.It is understood that the default syntactic category mode can be the syntactic category mode based on part of speech, example
Such as, it can be the mode classification based on nature of nouns, adjective attribute and adverbial word attribute, it is to be understood that described information is searched
Rope equipment can be classified the vocabulary in the training vocabulary according to vocabulary attribute.For example, the training vocabulary of nature of nouns
The lexicon of nature of nouns is classified to, the lexicon or adverbial word of training classified vocabulary to the adjective attribute of adjective attribute belong to
Property training classified vocabulary to the lexicon of adverbial word attribute etc..
S104, searches for the corresponding multimedia messages of the voice messaging using the critical field and the destination field and arranges
Table, and export the multimedia messages list.
Specifically, described information search equipment can be using the critical field and the destination field where server
Network side search out the corresponding multimedia messages list of the voice messaging.
Optionally, when only having destination field that critical field is not present at least one described field information, described information
The corresponding multimedia messages list of the voice messaging can be searched for using the destination field by searching for equipment.For example, working as voice
Information be " I wants to listen folk rhyme now " when, at least one corresponding field information can be " I it is present think listen the people
Ballad ", wherein there was only " folk rhyme " this destination field, not related to " folk rhyme " is used to do folk rhyme the pass further limited
Key field, then described information search equipment can be searched out pair from the network side where server only with destination field " folk rhyme "
All kinds of folk rhymes answered.
Optionally, when not only including destination field at least one described field information but also comprising critical field, the letter
Breath search equipment can search for the corresponding multimedia messages of the voice messaging using the destination field and the critical field
List.For example, when voice messaging is " I wants to listen cheerful and light-hearted folk rhyme now ", at least one corresponding field information can be with
Be " I it is present think listen it is cheerful and light-hearted folk rhyme ", wherein " folk rhyme " is destination field, " cheerful and light-hearted " is critical field, with folk rhyme
It is associated the type for further limiting folk rhyme, then described information search equipment can use destination field and critical field
" cheerful and light-hearted folk rhyme " searches out the folk rhyme of corresponding types from the network side where server.
Further, described information search equipment can export the multimedia messages list.It is understood that described
Information search equipment can be by the multimedia messages searched current display of the display output in terminal device in the form of a list
In the page.
Optionally, the multimedia messages that described information search equipment can be exported in the multimedia messages list are corresponding
Content of multimedia.For example, described information search equipment gets multimedia messages list (for example, in cheerful and light-hearted ballad series
Lied list) after, can direct all cheerful and light-hearted lieds searched of played in order audio content.
In embodiments of the present invention, by obtaining inputted voice messaging, voice messaging is identified processing with life
Carry out word segmentation processing to obtain at least one word in text message into the corresponding text message of voice messaging, and to text message
Segment information, the critical field and destination field of voice messaging is obtained using default lexicon, most at least one field information
Afterwards using critical field and the corresponding multimedia messages list of destination field search voice messaging, and export multimedia messages row
Table.By the way that natural-sounding is converted into text message, and language is obtained to text message progress word segmentation processing and terminology match processing
Message ceases corresponding critical field and destination field, then corresponding to matchmaker using critical field and destination field search voice messaging
Body information, improves the program request success rate that terminal device recognizes VOD system based on natural-sounding.
Fig. 2 is referred to, is that the embodiments of the invention provide the schematic flow sheet of another information search method.Such as Fig. 2 institutes
Show, the described information searching method in the present embodiment can include with step S201- steps S210.
S201, obtains inputted training vocabulary, and the training vocabulary is classified based on default syntactic category mode
Processing, and the default lexicon of generation after classification processing.
Specifically, described information search equipment can obtain the training vocabulary that user is inputted, it is possible to based on default language
Method mode classification carries out classification processing to the training vocabulary.It is understood that the default syntactic category mode can be
Syntactic category mode based on part of speech, for example, it may be the classification side based on nature of nouns, adjective attribute and adverbial word attribute
Formula, it is to be understood that described information search equipment can be divided the vocabulary in the training vocabulary according to vocabulary attribute
Class.For example, lexicon of the training classified vocabulary of nature of nouns to nature of nouns, the training classified vocabulary of adjective attribute to shape
Hold the lexicon of word attribute or the training classified vocabulary of adverbial word attribute to the lexicon of adverbial word attribute etc..
Further, described information search equipment can generate default vocabulary after carrying out classification processing to the training vocabulary
Storehouse.It is understood that the default lexicon can include the first lexicon, the second lexicon and the 3rd lexicon, it is described
First lexicon can be the lexicon of nature of nouns, and second lexicon can be the lexicon of adjective attribute, described
3rd lexicon can be the lexicon of adverbial word attribute.
S202, obtains inputted voice messaging, and the voice messaging is identified processing to generate the voice
The corresponding text message of information.
Specifically, described information search equipment can obtain inputted voice messaging, it is to be understood that the voice
Information can be natural-sounding information (for example, ' playing the most popular folk rhyme of Zhang Lei ').
Further, described information search equipment can be using default voice recognition processing method (for example, based on DNN
Speech recognition algorithm, fuzzy recognition algorithm or computer network kit CNTK etc.) place is identified to the voice messaging
Reason, and generate the corresponding text message of the voice messaging.
S203, word segmentation processing is carried out to the text message to obtain the letter of at least one field in the text message
Breath.
Specifically, described information search equipment can carry out word segmentation processing to obtain the text envelope to the text message
At least one field information in breath.It is understood that the word segmentation processing can be the segmenting method based on text matches,
Segmenting method based on CRF, probability statistics algorithm or machine learning method etc..It is understood that the text message passes through
After the word segmentation processing, at least one field information is divided into.For example：The text message is " broadcasting Zhang Lei is most popular
Folk rhyme ", then after the word segmentation processing, the field information that text information is divided into can be " play Zhang Lei
Most it is popular folk rhyme ".
S204, the first word that will be matched at least one described field information with the first vocabulary of first lexicon
Segment information is set to destination field.
Specifically, described information search equipment can by least one described field information with first lexicon
The first field information that first vocabulary matches is set to destination field.For example, when at least one described field information is " to broadcast
Put Zhang Lei most it is popular folk rhyme " when, " Zhang Lei " and " folk rhyme " can be that noun belongs to the vocabulary in first lexicon
Property vocabulary when matching, " Zhang Lei " and " folk rhyme " can be defined as the purpose word of the voice messaging by described information search equipment
Section.
S205, it is associated with the destination field when existing at least one described field information, and with described second
During the second field information that the second vocabulary of lexicon matches, second field information is defined as the first critical field.
Specifically, associated with the destination field when existing at least one described field information, and with described the
During the second field information that the second vocabulary of two lexicons matches, described information search equipment can believe second field
Breath is defined as the first critical field.For example, when at least one described field information for " play Zhang Lei most it is popular folk rhyme "
When, described information search equipment determines " Zhang Lei " and " folk rhyme " for after purpose field, it may be determined that associated with the destination field
Vocabulary is " broadcasting " and " popular ", and the i.e. part of speech matched with the vocabulary in second lexicon is adjective attribute
Vocabulary only has " popular ", so that described information search equipment can be determined in " popular " at least one field information for described in
The first critical field.
S206, when at least one described field information exist it is associated with first critical field, and with it is described
During three field information that the 3rd vocabulary of the 3rd lexicon matches, the 3rd field information is defined as the second keyword
Section.
Specifically, associated with first critical field when existing at least one described field information, and with institute
When stating three field information that the 3rd vocabulary of the 3rd lexicon matches, described information search equipment can be by the 3rd word
Segment information is defined as the second critical field.For example, when at least one described field information for " play Zhang Lei most it is popular the people
During ballad ", described information search equipment determine " Zhang Lei " and " folk rhyme " be purpose field, it is " popular " for after the first critical field,
The vocabulary associated with " popular " with " Zhang Lei ", " folk rhyme " can be determined for " most ", and " most " with the 3rd lexicon
Vocabulary match, i.e., match with part of speech for the vocabulary of adverbial word attribute, so that described information search equipment can be determined " most "
For the second critical field at least one described field information.
In embodiments of the present invention, the field information at least one described field information is classified using default lexicon
For purpose field, the first critical field and the second critical field, by the precise classification to field information, add by field
The accuracy of the corresponding multimedia messages list of information search voice messaging.
It should be noted that during the destination field described in the embodiment of the present invention can be at least one described field information
At least one vocabulary, first critical field and second critical field be intended merely to distinguish critical field, not
It is that ranking is carried out to critical field, it is to be understood that can also have the 3rd except the first critical field and the second critical field,
4th etc. critical field, other critical fielies be not described in detail in embodiments of the present invention.
S207, when there is the destination field at least one described field information and in the absence of first critical field
During with second critical field, corresponding first multimedia messages of the voice messaging are searched for using the destination field and arranged
Table, and export the first multimedia messages list.
Specifically, there is the destination field at least one field information described in and in the absence of first keyword
When section and second critical field, described information search equipment can use the destination field from the network where server
Side searches out the corresponding first multimedia messages list of the voice messaging, for example, when voice messaging is " I wants to listen the people now
During ballad ", at least one corresponding field information can be " I it is present think listen folk rhyme ", wherein only have " folk rhyme " this
Individual destination field, it is not associated with " folk rhyme " to be used to doing folk rhyme the first critical field further limited and second crucial
Field.Described information search equipment can search out the voice from the network side where server using the destination field and believe
Corresponding first multimedia messages list is ceased, it is for instance possible to use " I wants to listen the people " folk rhyme " this purpose field searches now
The information list of the corresponding lied of this voice messaging of ballad ".
Further, described information search equipment can export the first multimedia messages list.It is understood that
Described information search equipment can by the first multimedia messages searched in the form of a list display output in terminal device
In current display page.
Optionally, described information search equipment can export the first multimedia letter in the first multimedia messages list
Cease corresponding first content of multimedia.For example, described information search equipment gets the first multimedia messages list (for example, the people
Lied list in ballad series) after, can direct all lieds searched of played in order audio content.
S208, when there is the destination field and first critical field at least one described field information and do not deposit
In second critical field, the voice messaging is searched for using the destination field and first critical field corresponding
Second multimedia messages list, and export the second multimedia messages list.
Specifically, when there is the destination field and first critical field and not at least one described field information
When there is second critical field, described information search equipment can use the destination field and first critical field
The corresponding second multimedia messages list of the voice messaging is searched out from the network side where server, for example, when voice letter
When ceasing for " play Zhang Lei popular folk rhyme ", at least one corresponding field information can be " play Zhang Lei it is popular
Folk rhyme ", wherein " Zhang Lei " and " folk rhyme " is purpose field, " popular " is the first critical field, in the absence of the second critical field,
Described information search equipment can be " Zhang Lei popular folk rhyme " according to destination field and the first critical field from server institute
Network side search out the list of corresponding lied.
Further, described information search equipment can export the second multimedia messages list.It is understood that
Described information search equipment can by the second multimedia messages searched in the form of a list display output in terminal device
In current display page.
Optionally, described information search equipment can export the second multimedia letter in the second multimedia messages list
Cease corresponding second content of multimedia.For example, described information search equipment gets the second multimedia messages list (for example, opening
Build the lied list in popular ballad series) after, can direct all lieds searched of played in order sound
Frequency content.
S209, when there is the destination field, first critical field and described at least one described field information
During the second critical field, institute's predicate is searched for using the destination field, first critical field and second critical field
Message ceases corresponding multimedia messages list, and exports the multimedia messages list.
Specifically, there is the destination field, first critical field and institute at least one field information described in
When stating the second critical field, described information search equipment can be using the destination field, first critical field and described
Second critical field searches out the corresponding multimedia messages list of the voice messaging from the network side where server, for example,
When voice messaging is " plays Zhang Lei most popular folk rhyme ", at least one corresponding field information can be " play
Lei s prevalence folk rhymes ", wherein " Zhang Lei " and " folk rhyme " is purpose field, " popular " is the first critical field, and " most " is
Second critical field, described information search equipment can be " to open according to destination field, the first critical field and the second critical field
It is of heap of stone most it is popular folk rhyme " list of corresponding lied is searched out from the network side where server.
Further, described information search equipment can export the multimedia messages list.It is understood that described
Information search equipment can be by the multimedia messages searched current display of the display output in terminal device in the form of a list
In the page.
Optionally, the multimedia messages that described information search equipment can be exported in the multimedia messages list are corresponding
Content of multimedia.For example, described information search equipment gets multimedia messages list (for example, the most popular folk rhyme system of Zhang Lei
Lied list in row) after, can direct all lieds searched of played in order audio content.
S210, obtains the corresponding content of multimedia of multimedia messages in the multimedia messages list, and to described many
Media content is exported.
Specifically, described information search equipment can be obtained corresponding to the multimedia messages in the multimedia messages list
Content of multimedia, it is to be understood that the multimedia messages list can be multimedia messages (song title, singer,
Deliver time etc.) composition list, the content of multimedia can be the particular content (example indicated by the multimedia messages
Such as, song title A, singer, deliver the audio content of song A indicated by the song informations such as time).
Further, described information search equipment can be exported to content of multimedia, it is to be understood that the letter
Breath search equipment can directly play the content of multimedia (for example, song A audio content).
In embodiments of the present invention, by directly in the corresponding multimedia of multimedia messages in multimedia messages list
Appearance is exported, and reduces the operation complexity for playing content of multimedia.
In embodiments of the present invention, by obtaining inputted voice messaging, voice messaging is identified processing with life
Carry out word segmentation processing to obtain at least one word in text message into the corresponding text message of voice messaging, and to text message
Segment information, the critical field and destination field of voice messaging is obtained using default lexicon, most at least one field information
Afterwards using critical field and the corresponding multimedia messages list of destination field search voice messaging, and export multimedia messages row
Table.By the way that natural-sounding is converted into text message, and language is obtained to text message progress word segmentation processing and terminology match processing
Message ceases corresponding critical field and destination field, then corresponding to matchmaker using critical field and destination field search voice messaging
Body information, improves the program request success rate that terminal device recognizes VOD system based on natural-sounding；By at least one field
The precise classification of field in information, adds and searches for the accurate of the corresponding multimedia messages list of voice messaging by field information
Property；By directly being exported to the corresponding content of multimedia of multimedia messages in multimedia messages list, broadcasting is reduced
The operation complexity of content of multimedia.
Below in conjunction with accompanying drawing 3- accompanying drawings 6, information search equipment provided in an embodiment of the present invention is described in detail.Need
It is noted that the equipment shown in accompanying drawing 3- accompanying drawings 6, the method for performing Fig. 1 of the present invention and embodiment illustrated in fig. 2, in order to
It is easy to explanation, illustrate only the part related to the embodiment of the present invention, particular technique details is not disclosed, refer to figure of the present invention
Embodiment shown in 1 and Fig. 2.
Fig. 3 is referred to, is that the embodiments of the invention provide a kind of structural representation of information search equipment.As shown in figure 3,
The described information search equipment 1 of the embodiment of the present invention can include：Text message generation unit 11, field information acquiring unit
12nd, field information taxon 13 and media information output unit 14.
Text message generation unit 11, for obtaining inputted voice messaging, and the voice messaging is identified
Handle to generate the corresponding text message of the voice messaging.
In the specific implementation, the text message generation unit 11 can obtain the voice messaging that user is inputted.It can manage
Solution, the voice messaging can be natural-sounding information (for example, ' playing the most popular folk rhyme of Zhang Lei ').
Further, the text message generation unit 11 can be using default voice recognition processing method (for example, base
In DNN speech recognition algorithm, fuzzy recognition algorithm or computer network kit CNTK etc.) voice messaging is known
Other places are managed, and generate the corresponding text message of the voice messaging.
Field information acquiring unit 12, for being obtained to text message progress word segmentation processing in the text message
At least one field information.
In the specific implementation, the field information acquiring unit 12 can the text message corresponding to the voice messaging
Word segmentation processing is carried out to obtain at least one field information in the text message.It is understood that the word segmentation processing
Can be the segmenting method based on text matches, the segmenting method CRF based on condition random field, probability statistics algorithm or engineering
Learning method etc..It is understood that the text message is after the word segmentation processing, at least one field letter is divided into
Breath.For example：The text message is " playing the most popular folk rhyme of Zhang Lei ", then after the word segmentation processing, text information
The field information being divided into can be " play Zhang Lei most it is popular folk rhyme ".
Field information taxon 13, for obtaining described at least one described field information using default lexicon
The critical field and destination field of voice messaging.
In the specific implementation, the field information taxon 13 can be using default lexicon at least one described field
The critical field and destination field of the voice messaging are obtained in information.It is understood that the destination field can be institute
The core vocabulary in voice messaging is stated, the critical field can be associated with the destination field for aiding in the mesh
Field description described in voice messaging vocabulary.For example, when the voice messaging obtain after treatment it is described at least one
Field information for " play Zhang Lei most it is popular folk rhyme " when, the field information taxon 13 uses the default vocabulary
The destination field that storehouse is obtained in the field information can be " folk rhyme ", and critical field can be " Zhang Lei is most popular ".
Inputted it is understood that the default lexicon can be described information search equipment 1 by obtaining user
Training vocabulary, and classification processing is carried out to the training vocabulary based on default syntactic category mode, finally after classification processing
The lexicon of generation.It is understood that the default syntactic category mode can be the syntactic category mode based on part of speech, example
Such as, it can be the mode classification based on nature of nouns, adjective attribute and adverbial word attribute, it is to be understood that described information is searched
Rope equipment 1 can be classified the vocabulary in the training vocabulary according to vocabulary attribute.For example, the training vocabulary of nature of nouns
The lexicon of nature of nouns is classified to, the lexicon or adverbial word of training classified vocabulary to the adjective attribute of adjective attribute belong to
Property training classified vocabulary to the lexicon of adverbial word attribute etc..
Media information output unit 14, for searching for the voice messaging using the critical field and the destination field
Corresponding multimedia messages list, and export the multimedia messages list.
In the specific implementation, the media information output unit 14 can using the critical field and the destination field from
Network side where server searches out the corresponding multimedia messages list of the voice messaging.
Optionally, when only having destination field that critical field is not present at least one described field information, the media
Information output unit 14 can search for the corresponding multimedia messages list of the voice messaging using the destination field.For example,
When voice messaging be " I wants to listen folk rhyme now " when, at least one corresponding field information can be " I it is present think
Listen folk rhyme ", it is not related to " folk rhyme " to be used to do folk rhyme further limit wherein there was only " folk rhyme " this destination field
Fixed critical field, then the media information output unit 14 can be only with destination field " folk rhyme " from the net where server
Network side searches out corresponding all kinds of folk rhymes.
Optionally, when not only including destination field at least one described field information but also comprising critical field, the matchmaker
Body information output unit 14 can search for the corresponding many matchmakers of the voice messaging using the destination field and the critical field
Body information list.For example, when voice messaging is " I wants to listen cheerful and light-hearted folk rhyme now ", corresponding at least one field letter
Breath can be " I it is present think listen it is cheerful and light-hearted folk rhyme ", wherein " folk rhyme " is destination field, " cheerful and light-hearted " is critical field,
It is associated with folk rhyme for further limit folk rhyme type, then the media information output unit 14 can use destination field
The folk rhyme of corresponding types is searched out from the network side where server with critical field " cheerful and light-hearted folk rhyme ".
Further, the media information output unit 14 can export the multimedia messages list.It may be appreciated
It is that the media information output unit 14 can display output be set in terminal in the form of a list by the multimedia messages searched
In standby current display page.
Optionally, the media information output unit 14 can export the multimedia messages in the multimedia messages list
Corresponding content of multimedia.For example, the media information output unit 14 gets multimedia messages list (for example, cheerful and light-hearted
Lied list in ballad series) after, can direct all cheerful and light-hearted lieds searched of played in order audio
Content.
In embodiments of the present invention, by obtaining inputted voice messaging, voice messaging is identified processing with life
Carry out word segmentation processing to obtain at least one word in text message into the corresponding text message of voice messaging, and to text message
Segment information, the critical field and destination field of voice messaging is obtained using default lexicon, most at least one field information
Afterwards using critical field and the corresponding multimedia messages list of destination field search voice messaging, and export multimedia messages row
Table.By the way that natural-sounding is converted into text message, and language is obtained to text message progress word segmentation processing and terminology match processing
Message ceases corresponding critical field and destination field, then corresponding to matchmaker using critical field and destination field search voice messaging
Body information, improves the program request success rate that terminal device recognizes VOD system based on natural-sounding.
Fig. 4 is referred to, is that the embodiments of the invention provide the structural representation of another information search equipment.Such as Fig. 4 institutes
Show, the described information search equipment 1 of the embodiment of the present invention can include：Text message generation unit 11, field information obtain single
Member 12, field information taxon 13, media information output unit 14, lexicon generation unit 15 and media content output unit
16。
Lexicon generation unit 15, for obtaining inputted training vocabulary, based on default syntactic category mode to described
Vocabulary is trained to carry out classification processing, and the default lexicon of generation after classification processing.
In the specific implementation, the lexicon generation unit 15 can obtain the training vocabulary that user is inputted, it is possible to base
Classification processing is carried out to the training vocabulary in default syntactic category mode.It is understood that the default syntactic category side
Formula can be the syntactic category mode based on part of speech, for example, it may be based on nature of nouns, adjective attribute and adverbial word attribute
Mode classification, it is to be understood that described information search equipment can be by the vocabulary in the training vocabulary according to vocabulary attribute
Classified.For example, lexicon of the training classified vocabulary of nature of nouns to nature of nouns, the training vocabulary point of adjective attribute
Class is to the lexicon of adjective attribute or the training classified vocabulary of adverbial word attribute to the lexicon of adverbial word attribute etc..
Further, it can be generated after 15 pairs of the lexicon generation unit training vocabulary progress classification processing default
Lexicon.It is understood that the default lexicon can include the first lexicon, the second lexicon and the 3rd lexicon,
First lexicon can be the lexicon of nature of nouns, and second lexicon can be the lexicon of adjective attribute,
3rd lexicon can be the lexicon of adverbial word attribute.
Text message generation unit 11, for obtaining inputted voice messaging, and the voice messaging is identified
Handle to generate the corresponding text message of the voice messaging.
In the specific implementation, the text message generation unit 11 can obtain inputted voice messaging, it is possible to understand that
It is that the voice messaging can be natural-sounding information (for example, ' playing the most popular folk rhyme of Zhang Lei ').
Further, the text message generation unit 11 can be using default voice recognition processing method (for example, base
In DNN speech recognition algorithm, fuzzy recognition algorithm or computer network kit CNTK etc.) voice messaging is known
Other places are managed, and generate the corresponding text message of the voice messaging.
Field information acquiring unit 12, for being obtained to text message progress word segmentation processing in the text message
At least one field information.
In the specific implementation, the field information acquiring unit 12 can carry out word segmentation processing to obtain to the text message
At least one field information in the text message.It is understood that the word segmentation processing can be based on text matches
Segmenting method, the segmenting method based on CRF, probability statistics algorithm or machine learning method etc..It is understood that the text
This information is divided at least one field information after the word segmentation processing.For example：The text message is " broadcasting
The most popular folk rhyme of Zhang Lei ", then after the word segmentation processing, the field information that text information is divided into can be
" play Zhang Lei most it is popular folk rhyme ".
Field information taxon 13, for obtaining described at least one described field information using default lexicon
The critical field and destination field of voice messaging.
In the specific implementation, the field information taxon 13 can be using default lexicon at least one described field
The critical field and destination field of the voice messaging are obtained in information.It is understood that the destination field can be institute
The core vocabulary in voice messaging is stated, the critical field can be associated with the destination field for aiding in the mesh
Field description described in voice messaging vocabulary.
It is that the embodiments of the invention provide the structural representation of field information taxon 13 please also refer to Fig. 5.As schemed
Shown in 5, the field information taxon 13 can include：
Destination field set subelement 131, for by least one described field information with first lexicon
The first field information that first vocabulary matches is set to destination field.
In the specific implementation, the destination field set subelement 131 can will described at least one field information with institute
State the first field information that the first vocabulary of the first lexicon matches and be set to destination field.For example, when it is described at least one
Field information for " play Zhang Lei most it is popular folk rhyme " when, " Zhang Lei " and " folk rhyme " can with first lexicon
When vocabulary is that the vocabulary of nature of nouns matches, the destination field set subelement 131 can will " Zhang Lei " and " folk rhyme " it is true
It is set to the destination field of the voice messaging.
First field determination subelement 132, for existing and the destination field when at least one described field information
It is associated, and match with the second vocabulary of second lexicon the second field information when, second field is believed
Breath is defined as the first critical field.
In the specific implementation, associated with the destination field when existing at least one described field information, and with institute
When stating the second field information that the second vocabulary of the second lexicon matches, the first field determination subelement 132 can be by
Second field information is defined as the first critical field.For example, when at least one described field information for " play Zhang Lei
During prevalence folk rhymes ", the destination field sets subelement 131 to determine " Zhang Lei " and " folk rhyme " for after purpose field, institute
Stating the first field determination subelement 132 can determine that vocabulary associated with the destination field is " broadcasting " and " popular ", and with
The i.e. part of speech that vocabulary in second lexicon matches only has " popular " for the vocabulary of adjective attribute, so that described the
One field determination subelement 132 can determine " popular " the first critical field at least one described field information.
Second field determination subelement 133, when at least one described field information exist and first critical field
It is associated, and match with the 3rd vocabulary of the 3rd lexicon three field informations when, the 3rd field is believed
Breath is defined as the second critical field.
In the specific implementation, it is associated with first critical field when existing at least one described field information, and
During three field information matched with the 3rd vocabulary of the 3rd lexicon, the second field determination subelement 133 can
So that the 3rd field information is defined as into the second critical field.For example, when at least one described field information for " play
During Lei s prevalence folk rhymes ", the destination field sets subelement 131 to determine that " Zhang Lei " and " folk rhyme " is purpose field, institute
State the first field determination subelement 132 determine it is " popular " be the first critical field after, the second field determination subelement 133
The vocabulary associated with " popular " with " Zhang Lei ", " folk rhyme " can be determined for " most ", and " most " with the 3rd lexicon
Vocabulary match, i.e., match with part of speech for the vocabulary of adverbial word attribute, so that the second field determination subelement 133 can be with
It is determined that " most " is the second critical field at least one described field information.
In embodiments of the present invention, the field information at least one described field information is classified using default lexicon
For purpose field, the first critical field and the second critical field, by the precise classification to field information, add by field
The accuracy of the corresponding multimedia messages list of information search voice messaging.
It should be noted that during the destination field described in the embodiment of the present invention can be at least one described field information
At least one vocabulary, first critical field and second critical field be intended merely to distinguish critical field, not
It is that ranking is carried out to critical field, it is to be understood that can also have the 3rd except the first critical field and the second critical field,
4th etc. critical field, other critical fielies be not described in detail in embodiments of the present invention.
Media information output unit 14, for searching for the voice messaging using the critical field and the destination field
Corresponding multimedia messages list, and export the multimedia messages list.
In the specific implementation, the media information output unit 14 can using the critical field and the destination field from
Network side where server searches out the corresponding multimedia messages list of the voice messaging.
It is that the embodiments of the invention provide the structural representation of media information output unit 14 please also refer to Fig. 6.As schemed
Shown in 6, the media information output unit 14 can include：
The first information exports subelement 141, for when exist at least one described field information the destination field and
During in the absence of first critical field and second critical field, the voice messaging pair is searched for using the destination field
The the first multimedia messages list answered, and export the first multimedia messages list.
In the specific implementation, being closed when there is the destination field at least one described field information and be not present described first
When key field and second critical field, the first information output subelement 141 can use the destination field from clothes
Network side where business device searches out the corresponding first multimedia messages list of the voice messaging, for example, when voice messaging is
When " I wants to listen folk rhyme now ", at least one corresponding field information can be " I it is present think listen folk rhyme ", wherein
Only " folk rhyme " this destination field, it is not associated with " folk rhyme " to be used to doing folk rhyme first further limited crucial
Field and the second critical field.The first information output subelement 141 can be using the destination field where server
Network side search out the corresponding first multimedia messages list of the voice messaging, it is for instance possible to use " folk rhyme " this mesh
Field searches " I wants to listen folk rhyme now " corresponding lied of this voice messaging information list.
Further, the first information output subelement 141 can export the first multimedia messages list.Can be with
Understand, the first information output subelement 141 can show the first multimedia messages searched in the form of a list
Show output in the current display page of terminal device.
Optionally, first information output subelement 141 can export the in the first multimedia messages list
Corresponding first content of multimedia of one multimedia messages.For example, the first information output subelement 141 gets matchmaker more than first
, can direct all folk rhymes searched of played in order after body information list (for example, lied list in ballad series)
The audio content of song.
Second information exports subelement 142, for when exist at least one described field information the destination field and
First critical field and in the absence of second critical field when, using the destination field and first critical field
The corresponding second multimedia messages list of the voice messaging is searched for, and exports the second multimedia messages list.
In the specific implementation, there is the destination field and first critical field at least one field information described in
And during in the absence of second critical field, the second information output subelement 142 can use the destination field and institute
State the first critical field and search out the corresponding second multimedia messages list of the voice messaging from the network side where server,
For example, when voice messaging is " playing the popular folk rhyme of Zhang Lei ", at least one corresponding field information can " be broadcast
Fang Zhang Leis prevalence folk rhymes ", wherein " Zhang Lei " and " folk rhyme " is purpose field, " popular " is the first critical field, is not deposited
In the second critical field, the second information output subelement 142 can be " to open according to destination field and the first critical field
It is of heap of stone it is popular folk rhyme " list of corresponding lied is searched out from the network side where server.
Further, the second information output subelement 142 can export the second multimedia messages list.Can be with
Understand, the second information output subelement 142 can show the second multimedia messages searched in the form of a list
Show output in the current display page of terminal device.
Optionally, second information output subelement 142 can export the in the second multimedia messages list
Corresponding second content of multimedia of two multimedia messages.For example, the second information output subelement 142 gets matchmaker more than second
, can directly played in order be all searches after body information list (for example, lied list in the popular ballad series of Zhang Lei)
The audio content for the lied that rope is arrived.
Media information exports subelement 143, for there is the destination field, institute at least one field information described in
When stating the first critical field and second critical field, using the destination field, first critical field and described
Two critical fielies search for the corresponding multimedia messages list of the voice messaging, and export the multimedia messages list.
In the specific implementation, there is the destination field, first critical field at least one field information described in
During with second critical field, the media information output subelement 143 can be closed using the destination field, described first
Key field and second critical field search out the corresponding multimedia letter of the voice messaging from the network side where server
List is ceased, for example, when voice messaging is " playing the most popular folk rhyme of Zhang Lei ", at least one corresponding field information can
To be " playing Zhang Lei s prevalence folk rhymes ", wherein " Zhang Lei " and " folk rhyme " is purpose field, it is " popular " to be closed for first
Key field, " most " is the second critical field, and the media information output subelement 143 can be according to destination field, the first key
Field and the second critical field be " Zhang Lei most it is popular folk rhyme " search out corresponding folk rhyme from the network side where server
The list of song.
Further, the media information output subelement 143 can export the multimedia messages list.It is appreciated that
, subelement 143 can display output exists in the form of a list by the multimedia messages searched for the media information output
In the current display page of terminal device.
Optionally, the media information output subelement 143 can export the multimedia in the multimedia messages list
The corresponding content of multimedia of information.For example, media information output subelement 143 get multimedia messages list (for example,
Lied list in the most popular ballad series of Zhang Lei) after, can direct all lieds searched of played in order
Audio content.
Media content output unit 16, for obtaining the corresponding many matchmakers of multimedia messages in the multimedia messages list
Hold in vivo, and the content of multimedia is exported.
In the specific implementation, the media content output unit 16 can obtain the multimedia in the multimedia messages list
Content of multimedia corresponding to information, it is to be understood that the multimedia messages list can be multimedia messages (song title
Title, singer, deliver time etc.) list of composition, the content of multimedia can be the tool indicated by the multimedia messages
Hold in vivo (for example, song title A, singer, deliver the audio content of song A indicated by the song informations such as time).
Further, the media content output unit 16 can be exported to content of multimedia, it is to be understood that
Described information search equipment can directly play the content of multimedia (for example, song A audio content).
In embodiments of the present invention, by directly in the corresponding multimedia of multimedia messages in multimedia messages list
Appearance is exported, and reduces the operation complexity for playing content of multimedia.
In embodiments of the present invention, by obtaining inputted voice messaging, voice messaging is identified processing with life
Carry out word segmentation processing to obtain at least one word in text message into the corresponding text message of voice messaging, and to text message
Segment information, the critical field and destination field of voice messaging is obtained using default lexicon, most at least one field information
Afterwards using critical field and the corresponding multimedia messages list of destination field search voice messaging, and export multimedia messages row
Table.By the way that natural-sounding is converted into text message, and language is obtained to text message progress word segmentation processing and terminology match processing
Message ceases corresponding critical field and destination field, then corresponding to matchmaker using critical field and destination field search voice messaging
Body information, improves the program request success rate that terminal device recognizes VOD system based on natural-sounding；By at least one field
The precise classification of field in information, adds and searches for the accurate of the corresponding multimedia messages list of voice messaging by field information
Property；By directly being exported to the corresponding content of multimedia of multimedia messages in multimedia messages list, broadcasting is reduced
The operation complexity of content of multimedia.
It should be noted that for above each method embodiment, a series of actions is stated that in order to be briefly described
Combination, but those skilled in the art should know, the present invention is not limited by described sequence of movement, and some steps can be with
Serially or simultaneously carried out using other.Secondly, those skilled in the art should know, embodiment described in this description belongs to
In preferred embodiment, involved operation and the unit not necessarily present invention are necessary.And in the above-described embodiments, to each
The description of individual embodiment all emphasizes particularly on different fields, and does not have the part being described in detail in some embodiment, may refer to the correlation of other embodiment
Description.
In addition, each functional unit in each embodiment of the invention can be integrated in the unit of a processing, can also
It is that unit is individually physically present or two or more units are integrated in a unit.Above-mentioned integrated list
Member can both be realized in the form of hardware, it would however also be possible to employ the form of SFU software functional unit is realized.The integrated unit is such as
Fruit using in the form of SFU software functional unit realize and as independent production marketing or in use, a computer can be stored in can
Read in storage medium.Understood based on such, technical scheme substantially contributes to prior art in other words
Part or all or part of the technical scheme can be embodied in the form of software product, the computer software product
It is stored in a storage medium, including some instructions are to cause a computer equipment (can be personal computer, server
Or the network equipment etc.) perform all or part of step of each embodiment methods described of the invention.Wherein described storage is situated between
Matter includes：USB flash disk, read-only storage (Read-Only Memory, ROM), random access memory (Random Access
Memory, RAM), mobile hard disk, magnetic disc or CD etc. are various can be with the medium of store program codes.
Above disclosure is only preferred embodiment of present invention, can not limit the right model of the present invention with this certainly
Enclose, therefore the equivalent variations made according to the claims in the present invention, still belong to the scope that the present invention is covered.
Claims (10)
1. a kind of information search method, it is characterised in that including：
Inputted voice messaging is obtained, and it is corresponding to generate the voice messaging voice messaging to be identified processing
Text message；
The text message is carried out word segmentation processing to obtain at least one field information in the text message；
The critical field and purpose word of the voice messaging are obtained at least one described field information using default lexicon
Section；
The corresponding multimedia messages list of the voice messaging is searched for using the critical field and the destination field, and exported
The multimedia messages list.
2. the method as described in claim 1, it is characterised in that the voice messaging that the acquisition is inputted, and to the voice
Processing is identified with before generating the corresponding text message of the voice messaging in information, in addition to：
Inputted training vocabulary is obtained, based on default syntactic category mode to the training vocabulary progress classification processing, and
Default lexicon is generated after classification processing；
Wherein, the default lexicon includes the first lexicon, the second lexicon and the 3rd lexicon, and first lexicon is
The lexicon of nature of nouns, second lexicon is the lexicon of adjective attribute, and the 3rd lexicon is adverbial word attribute
Lexicon.
3. method as claimed in claim 2, it is characterised in that described to be believed using default lexicon at least one described field
The critical field and destination field of the voice messaging are obtained in breath, including：
The first field information matched at least one described field information with the first vocabulary of first lexicon is set
It is set to destination field；
It is associated with the destination field when existing at least one described field information, and with the of second lexicon
During the second field information that two vocabulary match, second field information is defined as the first critical field；
When at least one described field information exist it is associated with first critical field, and with the 3rd lexicon
The 3rd vocabulary match three field informations when, the 3rd field information is defined as the second critical field.
4. method as claimed in claim 3, it is characterised in that described to be searched for using the critical field and the destination field
The corresponding multimedia messages list of the voice messaging, and the multimedia messages list is exported, including：
When there is the destination field at least one described field information and in the absence of first critical field and described the
During two critical fielies, the corresponding first multimedia messages list of the voice messaging is searched for using the destination field, and export
The first multimedia messages list；
When there is the destination field and first critical field at least one described field information and in the absence of described the
During two critical fielies, the corresponding matchmaker more than second of the voice messaging is searched for using the destination field and first critical field
Body information list, and export the second multimedia messages list；
When there is the destination field, first critical field and second keyword at least one described field information
Duan Shi, searches for the voice messaging corresponding using the destination field, first critical field and second critical field
Multimedia messages list, and export the multimedia messages list.
5. the method as described in claim 1, it is characterised in that described to be searched for using the critical field and the destination field
After the corresponding multimedia messages of the voice messaging, in addition to：
The corresponding content of multimedia of multimedia messages in the multimedia messages list is obtained, and the content of multimedia is entered
Row output.
6. a kind of information search equipment, it is characterised in that including：
Text message generation unit, for obtaining inputted voice messaging, and the voice messaging is identified processing with
Generate the corresponding text message of the voice messaging；
Field information acquiring unit, for being obtained to text message progress word segmentation processing in the text message at least
One field information；
Field information taxon, for obtaining the voice letter at least one described field information using default lexicon
The critical field and destination field of breath；
Media information output unit, it is corresponding for searching for the voice messaging using the critical field and the destination field
Multimedia messages list, and export the multimedia messages list.
7. equipment as claimed in claim 6, it is characterised in that also include：
Lexicon generation unit, for obtaining inputted training vocabulary, based on default syntactic category mode to the training word
Converge and carry out classification processing, and the default lexicon of generation after classification processing；
Wherein, the default lexicon includes the first lexicon, the second lexicon and the 3rd lexicon, and first lexicon is
The lexicon of nature of nouns, second lexicon is the lexicon of adjective attribute, and the 3rd lexicon is adverbial word attribute
Lexicon.
8. equipment as claimed in claim 7, it is characterised in that the field information taxon includes：
Destination field sets subelement, for by the first vocabulary at least one described field information with first lexicon
The first field information matched is set to destination field；
First field determination subelement, for associated with the destination field when existing at least one described field information
, and match with the second vocabulary of second lexicon the second field information when, second field information is determined
For the first critical field；
Second field determination subelement, it is associated with first critical field when existing at least one described field information
, and match with the 3rd vocabulary of the 3rd lexicon three field informations when, the 3rd field information is determined
For the second critical field.
9. equipment as claimed in claim 8, it is characterised in that the media information output unit includes：
The first information exports subelement, for there is the destination field at least one field information described in and in the absence of institute
When stating the first critical field and second critical field, the voice messaging corresponding first is searched for using the destination field
Multimedia messages list, and export the first multimedia messages list；
Second information exports subelement, for there is the destination field and described first at least one field information described in
Critical field and during in the absence of second critical field, searches for described using the destination field and first critical field
The corresponding second multimedia messages list of voice messaging, and export the second multimedia messages list；
Media information exports subelement, for there is the destination field, described first at least one field information described in
When critical field and second critical field, using the destination field, first critical field and second key
The corresponding multimedia messages list of voice messaging described in field searches, and export the multimedia messages list.
10. equipment as claimed in claim 6, it is characterised in that also include：
Media content output unit, for obtaining in the corresponding multimedia of multimedia messages in the multimedia messages list
Hold, and the content of multimedia is exported.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710194322.8A CN107145509B (en) | 2017-03-28 | 2017-03-28 | Information searching method and equipment thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710194322.8A CN107145509B (en) | 2017-03-28 | 2017-03-28 | Information searching method and equipment thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107145509A true CN107145509A (en) | 2017-09-08 |
| CN107145509B CN107145509B (en) | 2020-11-13 |
Family
ID=59783582
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710194322.8A Active CN107145509B (en) | 2017-03-28 | 2017-03-28 | Information searching method and equipment thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107145509B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108334498A (en) * | 2018-02-07 | 2018-07-27 | 百度在线网络技术（北京）有限公司 | Method and apparatus for handling voice request |
| CN109147801A (en) * | 2018-09-30 | 2019-01-04 | 深圳市元征科技股份有限公司 | voice interactive method, system, terminal and storage medium |
| CN110083681A (en) * | 2019-04-12 | 2019-08-02 | 中国平安财产保险股份有限公司 | Searching method, device and terminal based on data analysis |
| CN110347901A (en) * | 2018-04-02 | 2019-10-18 | 和硕联合科技股份有限公司 | A kind of searching method and a kind of electronic device using this method |
| CN113658598A (en) * | 2021-08-12 | 2021-11-16 | 海信电子科技（深圳）有限公司 | Voice interaction method of display equipment and display equipment |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937433A (en) * | 2009-06-29 | 2011-01-05 | 天津一度搜索网络科技有限公司 | Real-time searching method of product |
| CN103810998A (en) * | 2013-12-05 | 2014-05-21 | 中国农业大学 | Method for off-line speech recognition based on mobile terminal device and achieving method |
| CN104462425A (en) * | 2014-12-12 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Method and device for displaying search suggestion |
| CN106250369A (en) * | 2016-07-28 | 2016-12-21 | 海信集团有限公司 | voice interactive method, device and terminal |
-
2017
- 2017-03-28 CN CN201710194322.8A patent/CN107145509B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101937433A (en) * | 2009-06-29 | 2011-01-05 | 天津一度搜索网络科技有限公司 | Real-time searching method of product |
| CN103810998A (en) * | 2013-12-05 | 2014-05-21 | 中国农业大学 | Method for off-line speech recognition based on mobile terminal device and achieving method |
| CN104462425A (en) * | 2014-12-12 | 2015-03-25 | 百度在线网络技术（北京）有限公司 | Method and device for displaying search suggestion |
| CN106250369A (en) * | 2016-07-28 | 2016-12-21 | 海信集团有限公司 | voice interactive method, device and terminal |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108334498A (en) * | 2018-02-07 | 2018-07-27 | 百度在线网络技术（北京）有限公司 | Method and apparatus for handling voice request |
| CN110347901A (en) * | 2018-04-02 | 2019-10-18 | 和硕联合科技股份有限公司 | A kind of searching method and a kind of electronic device using this method |
| CN109147801A (en) * | 2018-09-30 | 2019-01-04 | 深圳市元征科技股份有限公司 | voice interactive method, system, terminal and storage medium |
| CN110083681A (en) * | 2019-04-12 | 2019-08-02 | 中国平安财产保险股份有限公司 | Searching method, device and terminal based on data analysis |
| CN110083681B (en) * | 2019-04-12 | 2024-02-09 | 中国平安财产保险股份有限公司 | Searching method, device and terminal based on data analysis |
| CN113658598A (en) * | 2021-08-12 | 2021-11-16 | 海信电子科技（深圳）有限公司 | Voice interaction method of display equipment and display equipment |
| CN113658598B (en) * | 2021-08-12 | 2024-02-27 | Vidaa(荷兰)国际控股有限公司 | Voice interaction method of display equipment and display equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107145509B (en) | 2020-11-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108304375B (en) | Information identification method and equipment, storage medium and terminal thereof | |
| CN108288468B (en) | Audio recognition method and device | |
| CN107145509A (en) | A kind of information search method and its equipment | |
| US11823678B2 (en) | Proactive command framework | |
| US10811013B1 (en) | Intent-specific automatic speech recognition result generation | |
| CN105956053B (en) | A kind of searching method and device based on the network information | |
| EP3405912A1 (en) | Analyzing textual data | |
| CN107665188B (en) | Semantic understanding method and device | |
| US9922650B1 (en) | Intent-specific automatic speech recognition result generation | |
| CN107799119A (en) | Audio preparation method, apparatus and system | |
| CN107247769A (en) | Method for ordering song by voice, device, terminal and storage medium | |
| CN106021234A (en) | Label extraction method and system | |
| JP2013025648A (en) | Interaction device, interaction method and interaction program | |
| JP2008287210A (en) | Speech chat system, information processor, speech recognition method and program | |
| CN108920649A (en) | A kind of information recommendation method, device, equipment and medium | |
| CN100583086C (en) | Conversation control system and conversation control method | |
| CN109063182B (en) | Content recommendation method based on voice search questions and electronic equipment | |
| CN108345612A (en) | A kind of question processing method and device, a kind of device for issue handling | |
| CN107679196A (en) | A kind of multimedia recognition methods, electronic equipment and storage medium | |
| CN113505198A (en) | Keyword-driven generating type dialogue reply method and device and electronic equipment | |
| Bhattacharya et al. | A multimodal approach towards emotion recognition of music using audio and lyrical content | |
| CN111428487B (en) | Model training method, lyric generation method, device, electronic equipment and medium | |
| Gupta et al. | Songs recommendation using context-based semantic similarity between lyrics | |
| CN116343771A (en) | Music on-demand voice instruction recognition method and device based on knowledge graph | |
| CN112905835B (en) | Multi-mode music title generation method and device and storage medium |
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 |