CN109947955A - Voice search method, user equipment, storage medium and device - Google Patents
Voice search method, user equipment, storage medium and device Download PDFInfo
- Publication number
- CN109947955A CN109947955A CN201910219560.9A CN201910219560A CN109947955A CN 109947955 A CN109947955 A CN 109947955A CN 201910219560 A CN201910219560 A CN 201910219560A CN 109947955 A CN109947955 A CN 109947955A
- Authority
- CN
- China
- Prior art keywords
- phoneme
- group
- default
- current
- phoneme group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Landscapes
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses voice search method, user equipment, storage medium and devices.Speech recognition first can be carried out to voice messaging in the present invention, obtain corresponding current phoneme group；Phoneme replacement is carried out to each phoneme in current phoneme group, obtains alternative phoneme group；Alternative phoneme group default phoneme group corresponding with preset keyword is matched；In successful match, using the preset keyword of successful match as phonetic search result.Significantly, the present invention considers the voice messaging inputted, and there may be mistakes, so, phoneme replacement additionally will be carried out to current phoneme group, to radiate other more similar alternative phoneme groups, to improve a possibility that being successfully searched result, the accuracy of search result is improved, solves the lower technical problem of phonetic search accuracy.
Description
Technical field
The present invention relates to phonetic search technical field more particularly to voice search method, user equipment, storage medium and dresses
It sets.
Background technique
With the continuous development of search engine technique, in addition to can using text as search according to other than, can also by picture and
The other kinds of information such as voice obtains search result as search foundation.
For phonetic search, the search of all kinds of resources can be carried out by voice, for example, film is searched for, song search etc..
But it is different from other kinds of search foundation, voice messaging itself is easy to produce ambiguity and mistake there is biggish fallibility
Difference, for example, the accent of user itself and dialect are there is certain influence, the accuracy of search result is lower, it is more difficult to hit
Existing correct keyword in database.
, it can be said that the technical problem that phonetic search is lower there is accuracy.
Above content is only used to facilitate the understanding of the technical scheme, and is not represented and is recognized that above content is existing skill
Art.
Summary of the invention
The main purpose of the present invention is to provide voice search method, user equipment, storage medium and devices, it is intended to solve
The lower technical problem of phonetic search accuracy.
To achieve the above object, the present invention provides a kind of voice search method, and the voice search method includes following step
It is rapid:
Obtain voice messaging to be searched；
Speech recognition is carried out to the voice messaging, obtains current phoneme group corresponding with the voice messaging；
Phoneme replacement is carried out to each phoneme in the current phoneme group, obtains alternative phoneme group；
The alternative phoneme group default phoneme group corresponding with preset keyword is matched；
In successful match, using the preset keyword of successful match as phonetic search result.
Preferably, each phoneme in the current phoneme group carries out phoneme replacement, obtains alternative phoneme group, specifically
Include:
Current phoneme in the current phoneme group is matched with each fuzzy phoneme that default fuzzy phoneme is concentrated；
In successful match, corresponding close phoneme is inquired in default phoneme mapping relations according to the current phoneme；
Current phoneme in the current phoneme group is replaced with into the close phoneme, by the replaced current phoneme of phoneme
Organize alternately phoneme group.
Preferably, described in successful match, correspondence is inquired in default phoneme mapping relations according to the current phoneme
Close phoneme after, the voice search method further include:
Using the current phoneme of successful match as phoneme to be replaced, and count the initial phoneme number of the phoneme to be replaced
Amount；
When the initial phoneme quantity is greater than or equal to default phoneme threshold value, according to default Substitution Rules determine it is described to
The phoneme quantity to be operated currently being replaced in replacement phoneme；
The phoneme to be operated that quantity is the phoneme quantity to be operated is chosen from the phoneme to be replaced, it is described wait operate
Phoneme quantity is less than or equal to the initial phoneme quantity；
Correspondingly, the current phoneme by the current phoneme group replaces with the close phoneme, and phoneme is replaced
The alternately phoneme group of current phoneme group afterwards, specifically includes:
Phoneme to be operated described in the current phoneme group is replaced with into close sound corresponding with the phoneme to be operated
Element, by the replaced current phoneme group of phoneme alternately phoneme group.
Preferably, it is described the alternative phoneme group default phoneme group corresponding with preset keyword is matched after,
The voice search method further include:
When it fails to match, determines the new phoneme quantity to be operated currently being replaced in the phoneme to be replaced, return
The step of phoneme to be operated that quantity is the phoneme quantity to be operated is chosen described in receipt row from the phoneme to be replaced, directly
Extremely in successful match, using the preset keyword of successful match as phonetic search result.
Preferably, described to match the alternative phoneme group default phoneme group corresponding with preset keyword, specifically
Include:
By the current phoneme group and the alternative phoneme group, default phoneme group corresponding with preset keyword is carried out respectively
Matching；
Correspondingly, described in successful match, using the preset keyword of successful match as phonetic search as a result, specific packet
It includes:
It is matched into any phoneme group in the alternative phoneme group with the default phoneme group in the current phoneme group
When function, using the preset keyword of successful match as phonetic search result.
Preferably, described that the current phoneme group and the alternative phoneme group is corresponding with preset keyword default respectively
Phoneme group is matched, and is specifically included:
The current phoneme group is revised as default phoneme group corresponding with preset keyword by default edit mode, and
The first modification number that statistics is modified by the default edit mode；
The alternative phoneme group is revised as the default phoneme group by the default edit mode, and counts and passes through institute
State the second modification number that default edit mode is modified；
The first modification number is compared with default modification number respectively with the second modification number；
Correspondingly, any phoneme group in the current phoneme group and the alternative phoneme group and the default sound
When element group successful match, using the preset keyword of successful match as phonetic search as a result, specifically including:
Any modification number in the first modification number and the second modification number is less than the default modification
When number, the corresponding preset keyword of modification number of the default modification number will be less than as phonetic search result.
Preferably, described to carry out the first modification number and the second modification number with default modification number respectively
Compare, specifically include:
The first matching degree calculated by preset matching degree calculation formula according to the first modification number, according to described the
Two modification numbers calculate the second matching degree by the preset matching degree calculation formula；
First matching degree is compared with preset matching degree respectively with second matching degree；
Correspondingly, any modification number in the first modification number and the second modification number is less than institute
When stating default modification number, the corresponding preset keyword of modification number of the default modification number will be less than as phonetic search
As a result, specifically including:
It, will when any matching degree in first matching degree and second matching degree is spent greater than the preset matching
Greater than the preset matching degree the corresponding preset keyword of matching degree as phonetic search result.
In addition, to achieve the above object, the present invention also proposes a kind of user equipment, the user equipment include memory,
Processor and it is stored in the phonetic search program that can be run on the memory and on the processor, the phonetic search journey
Sequence is arranged for carrying out the step of voice search method as described above.
In addition, to achieve the above object, the present invention also proposes a kind of storage medium, voice is stored on the storage medium
The step of search program, the phonetic search program realizes voice search method as described above when being executed by processor.
In addition, to achieve the above object, the present invention also proposes a kind of voice searching device, the voice searching device packet
It includes:
Voice obtains module, for obtaining voice messaging to be searched；
Speech recognition module obtains corresponding with the voice messaging for carrying out speech recognition to the voice messaging
Current phoneme group；
Phoneme replacement module obtains alternative phoneme for carrying out phoneme replacement to each phoneme in the current phoneme group
Group；
Phoneme matching module, for will the alternative phoneme group default phoneme group progress corresponding with preset keyword
Match；
Voice search module is used in successful match, using the preset keyword of successful match as phonetic search result.
Speech recognition first can be carried out to voice messaging in the present invention, obtain corresponding current phoneme group；To current phoneme group
In each phoneme carry out phoneme replacement, obtain alternative phoneme group；By alternative phoneme group default phoneme corresponding with preset keyword
Group is matched；In successful match, using the preset keyword of successful match as phonetic search result.It is apparent that of the invention
In view of there may be mistakes for the voice messaging of input, so, phoneme replacement additionally will be carried out to current phoneme group, with radiation
Other more similar alternative phoneme groups improve the standard of search result to improve a possibility that being successfully searched result
Exactness solves the lower technical problem of phonetic search accuracy.
Detailed description of the invention
Fig. 1 is the user device architecture schematic diagram for the hardware running environment that the embodiment of the present invention is related to；
Fig. 2 is the flow diagram of voice search method first embodiment of the present invention；
Fig. 3 is the flow diagram of voice search method second embodiment of the present invention；
Fig. 4 is the flow diagram of voice search method 3rd embodiment of the present invention；
Fig. 5 is the structural block diagram of voice searching device first embodiment of the present invention.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
Referring to Fig.1, Fig. 1 is the user device architecture schematic diagram for the hardware running environment that the embodiment of the present invention is related to.
As shown in Figure 1, the user equipment may include: processor 1001, such as CPU, communication bus 1002, user interface
1003, network interface 1004, memory 1005.Wherein, communication bus 1002 is for realizing the connection communication between these components.
User interface 1003 may include display screen (Display), optional user interface 1003 can also include standard wireline interface,
Wireless interface, the wireline interface for user interface 1003 can be USB interface in the present invention.Network interface 1004 optionally may be used
To include standard wireline interface and wireless interface (such as WI-FI interface).Memory 1005 can be high speed RAM memory, can also
To be stable memory (non-volatile memory), such as magnetic disk storage.Memory 1005 optionally can also be
Independently of the storage device of aforementioned processor 1001.
It will be understood by those skilled in the art that structure shown in Fig. 1 does not constitute the restriction to user equipment, can wrap
It includes than illustrating more or fewer components, perhaps combines certain components or different component layouts.
As shown in Figure 1, as may include that operating system, network are logical in a kind of memory 1005 of computer storage medium
Believe module, Subscriber Interface Module SIM and phonetic search program.
In user equipment shown in Fig. 1, network interface 1004 is mainly used for connecting background server, takes with the backstage
Business device carries out data communication；User interface 1003 is mainly used for connecting peripheral hardware；The user equipment is called by processor 1001
The phonetic search program stored in memory 1005, and execute following operation:
Obtain voice messaging to be searched；
Speech recognition is carried out to the voice messaging, obtains current phoneme group corresponding with the voice messaging；
Phoneme replacement is carried out to each phoneme in the current phoneme group, obtains alternative phoneme group；
The alternative phoneme group default phoneme group corresponding with preset keyword is matched；
In successful match, using the preset keyword of successful match as phonetic search result.
Further, processor 1001 can call the phonetic search program stored in memory 1005, also execute following
Operation:
Current phoneme in the current phoneme group is matched with each fuzzy phoneme that default fuzzy phoneme is concentrated；
In successful match, corresponding close phoneme is inquired in default phoneme mapping relations according to the current phoneme；
Current phoneme in the current phoneme group is replaced with into the close phoneme, by the replaced current phoneme of phoneme
Organize alternately phoneme group.
Further, processor 1001 can call the phonetic search program stored in memory 1005, also execute following
Operation:
Using the current phoneme of successful match as phoneme to be replaced, and count the initial phoneme number of the phoneme to be replaced
Amount；
When the initial phoneme quantity is greater than or equal to default phoneme threshold value, according to default Substitution Rules determine it is described to
The phoneme quantity to be operated currently being replaced in replacement phoneme；
The phoneme to be operated that quantity is the phoneme quantity to be operated is chosen from the phoneme to be replaced, it is described wait operate
Phoneme quantity is less than or equal to the initial phoneme quantity；
Correspondingly, following operation is also executed:
Phoneme to be operated described in the current phoneme group is replaced with into close sound corresponding with the phoneme to be operated
Element, by the replaced current phoneme group of phoneme alternately phoneme group.
Further, processor 1001 can call the phonetic search program stored in memory 1005, also execute following
Operation:
When it fails to match, determines the new phoneme quantity to be operated currently being replaced in the phoneme to be replaced, return
The step of phoneme to be operated that quantity is the phoneme quantity to be operated is chosen described in receipt row from the phoneme to be replaced, directly
Extremely in successful match, using the preset keyword of successful match as phonetic search result.
Further, processor 1001 can call the phonetic search program stored in memory 1005, also execute following
Operation:
By the current phoneme group and the alternative phoneme group, default phoneme group corresponding with preset keyword is carried out respectively
Matching；
Correspondingly, following operation is also executed:
It is matched into any phoneme group in the alternative phoneme group with the default phoneme group in the current phoneme group
When function, using the preset keyword of successful match as phonetic search result.
Further, processor 1001 can call the phonetic search program stored in memory 1005, also execute following
Operation:
The current phoneme group is revised as default phoneme group corresponding with preset keyword by default edit mode, and
The first modification number that statistics is modified by the default edit mode；
The alternative phoneme group is revised as the default phoneme group by the default edit mode, and counts and passes through institute
State the second modification number that default edit mode is modified；
The first modification number is compared with default modification number respectively with the second modification number；
Correspondingly, following operation is also executed:
Any modification number in the first modification number and the second modification number is less than the default modification
When number, the corresponding preset keyword of modification number of the default modification number will be less than as phonetic search result.
Further, processor 1001 can call the phonetic search program stored in memory 1005, also execute following
Operation:
The first matching degree calculated by preset matching degree calculation formula according to the first modification number, according to described the
Two modification numbers calculate the second matching degree by the preset matching degree calculation formula；
First matching degree is compared with preset matching degree respectively with second matching degree；
Correspondingly, following operation is also executed:
It, will when any matching degree in first matching degree and second matching degree is spent greater than the preset matching
Greater than the preset matching degree the corresponding preset keyword of matching degree as phonetic search result.
Speech recognition first can be carried out to voice messaging in the present embodiment, obtain corresponding current phoneme group；To current phoneme
Each phoneme in group carries out phoneme replacement, obtains alternative phoneme group；By alternative phoneme group default sound corresponding with preset keyword
Plain group is matched；In successful match, using the preset keyword of successful match as phonetic search result.It is apparent that this reality
Apply example in view of input voice messaging there may be mistakes, so, will additionally to current phoneme group carry out phoneme replacement, with
It radiates other more similar alternative phoneme groups and improves search result to improve a possibility that being successfully searched result
Accuracy, solve the lower technical problem of phonetic search accuracy.
Based on above-mentioned hardware configuration, the embodiment of voice search method of the present invention is proposed.
It is the flow diagram of voice search method first embodiment of the present invention referring to Fig. 2, Fig. 2.
In the first embodiment, the voice search method the following steps are included:
Step S10: voice messaging to be searched is obtained.
It is understood that the executing subject of the present embodiment is user equipment, user equipment can be search server.If searching
Rope server is applicable in the technical solution of the present embodiment description, and user can be against the microphone of television set or cell phone oral account one
Section voice messaging, and this section of voice messaging is sent to search server, so that search server gets this section of voice letter
Breath.
Step S20: speech recognition is carried out to the voice messaging, obtains current phoneme corresponding with the voice messaging
Group.
It is understood that after getting one section of voice messaging as search foundation, it can be first by this section of voice messaging
Be converted to phoneme.
It should be noted that phoneme group refers to the set being made of one or more phonemes, phoneme includes initial consonant and rhythm
Mother etc., for example, including Chinese phonetic alphabet z, zh, c, ch, in and ing etc. in phoneme.
In the concrete realization, after the one section of voice messaging that will acquire carries out phoneme conversion, the current phoneme group of acquisition can
For " xiaomin ".
Step S30: phoneme replacement is carried out to each phoneme in the current phoneme group, obtains alternative phoneme group.
It should be understood that since there is certain fallibility, the voice messagings pair of user's input for voice messaging itself
The keyword answered in default resources bank and may be not present, for example, when searching for film, current phoneme group to be searched is
" xiaomin " presets in resources bank and " xiaomin " corresponding keyword is not present.Wherein, presetting in resources bank may include having
The preset keywords such as film name, Yan Yuanming, Dao Yanming, Ge Shouming, song title and album name.
In the concrete realization, it is contemplated that may and be not present completely corresponding with the phoneme of current phoneme group in default resources bank
Preset keyword, alternative phoneme group can be automatically generated and carry out phonetic search operation, to improve the hit rate of search result, rule
Carry out prediction of result to property.For example, including phoneme " x ", " i ", " ao ", " m " and " in " in current phoneme group " xiaomin ".
Phoneme replacement can be carried out to the phoneme in current phoneme group, for example, can be by phoneme " in " " replace with phoneme " ing ", then it obtains
Alternative phoneme group is " xiaoming ".
Step S40: the alternative phoneme group default phoneme group corresponding with preset keyword is matched.
It is understood that preset keyword is and performer's name if limit search type is performer's name search-type
The corresponding keyword of search-type, for example, performer's title.So can successively phoneme group corresponding with performer's title match.
In the concrete realization, the preset keyword in default resources bank first can be converted into default phoneme group, for example, default
Keyword may include " Xiao Ming ", and the default phoneme group of " Xiao Ming " is " xiaoming ".
Step S50: in successful match, using the preset keyword of successful match as phonetic search result.
It should be understood that being matched by phoneme group, it is possible to find exist in default resources bank corresponding with alternative crucial group
Preset keyword " Xiao Ming " can then be exported as phonetic search result, improve hit rate by preset keyword.In addition, if this
When search film, then the film and film performed by performer " Xiao Ming " can be further provided according to phonetic search result
The film etc. of entitled " Xiao Ming ".
Speech recognition first can be carried out to voice messaging in the present embodiment, obtain corresponding current phoneme group；To current phoneme
Each phoneme in group carries out phoneme replacement, obtains alternative phoneme group；By alternative phoneme group default sound corresponding with preset keyword
Plain group is matched；In successful match, using the preset keyword of successful match as phonetic search result.It is apparent that this reality
Apply example in view of input voice messaging there may be mistakes, so, will additionally to current phoneme group carry out phoneme replacement, with
It radiates other more similar alternative phoneme groups and improves search result to improve a possibility that being successfully searched result
Accuracy, solve the lower technical problem of phonetic search accuracy.
It is the flow diagram of voice search method second embodiment of the present invention referring to Fig. 3, Fig. 3, based on shown in above-mentioned Fig. 2
First embodiment, propose the second embodiment of voice search method of the present invention.
In second embodiment, further, the step S30 is specifically included:
Step S301: by the current phoneme group current phoneme and each fuzzy phoneme for concentrating of default fuzzy phoneme into
Row matching.
In the concrete realization, if current phoneme group be " xiaomin ", including current phoneme be " x ", " i ",
" ao ", " m " and " in " can respectively match this 5 current phonemes with fuzzy phoneme.Default fuzzy phone set is preparatory
Defined set of phonemes, the phoneme for including in the set of phonemes are denoted as fuzzy phoneme, and obscuring phoneme is the lower whole of discrimination
Syllable, including flat tongue consonant, cacuminal, pre-nasal sound and rear nasal sound etc..
Step S302: it in successful match, is inquired in default phoneme mapping relations according to the current phoneme corresponding
Close phoneme.
In the concrete realization, it is matched by phoneme, it is possible to find " in " in current phoneme group falls into default fuzzy phoneme
In the range of collection, then next step operation can be carried out using the current phoneme " in " of the successful match.
It is understood that in view of there is certain correlations each other for limited fuzzy phoneme, each other
More difficult differentiation, for example, between flat tongue consonant and cacuminal, between pre-nasal sound and rear nasal sound.It is reflected so default phoneme can be preset
Relationship is penetrated, is preset in phoneme mapping relations including the corresponding relationship between current phoneme and close phoneme.For example, can be by flat tongue consonant
It is corresponding with cacuminal, " z " correspondence " zh ", " c " correspondence " ch " and " s " correspondence " sh " etc.；Can be corresponding by pre-nasal sound and rear nasal sound,
" en " correspondence " eng " and " in " correspondence " ing "；Can be corresponding by the first default phoneme and the second default phoneme, " l " correspondence " r ".
It should be noted that fuzzy phoneme is whole qualified phonemes, fuzzy phoneme includes close phoneme.Close sound
Element is one group of relevant fuzzy phoneme, and close phoneme exists in groups in default phoneme mapping relations.
Step S303: the current phoneme in the current phoneme group is replaced with into the close phoneme, after phoneme is replaced
Current phoneme group alternately phoneme group.
In the concrete realization, due to remembering there is " in " correspondence " ing " in default phoneme mapping relations, then close phoneme is
Current phoneme " in " is replaced with close phoneme " ing " by " ing ", to obtain alternative phoneme group " xiaoming ".
Further, described in successful match, it is inquired in default phoneme mapping relations according to the current phoneme pair
After the close phoneme answered, the voice search method further include:
Using the current phoneme of successful match as phoneme to be replaced, and count the initial phoneme number of the phoneme to be replaced
Amount；
When the initial phoneme quantity is greater than or equal to default phoneme threshold value, according to default Substitution Rules determine it is described to
The phoneme quantity to be operated currently being replaced in replacement phoneme；
The phoneme to be operated that quantity is the phoneme quantity to be operated is chosen from the phoneme to be replaced, it is described wait operate
Phoneme quantity is less than or equal to the initial phoneme quantity；
Correspondingly, the current phoneme by the current phoneme group replaces with the close phoneme, and phoneme is replaced
The alternately phoneme group of current phoneme group afterwards, specifically includes:
Phoneme to be operated described in the current phoneme group is replaced with into close sound corresponding with the phoneme to be operated
Element, by the replaced current phoneme group of phoneme alternately phoneme group.
It is understood that in view of in current phoneme group may successful match current phoneme there are multiple, for example, if
Current phoneme group is " cenglong ", and the current phoneme of successful match may be " c ", " eng " and " l " totally 3 sounds to be replaced
Element, initial phoneme quantity are 3.Settable default phoneme threshold value is 2, it is seen then that initial phoneme quantity 3 at this time is greater than or equal to pre-
If phoneme threshold value 2.
It should be understood that may be simultaneously present operations that are multiple, and replacing in phoneme in view of the fuzzy phoneme being matched to
In be optionally replaced, to improve replacement efficiency.For example, phoneme totally 3 that need to replace, it first can only replace therein
One.
In the concrete realization, default Substitution Rules " can set phoneme quantity to be operated for first only replacing one of those
It is 1 ".Specifically, a phoneme to be operated first can be selected from phoneme to be replaced " c ", " eng " and " l ", for example, selecting
Phoneme to be operated can be " c ", be based on close phoneme corresponding with phoneme " c " to be operated known to default phoneme mapping relations
"ch".So, the current phoneme group " cenglong " of ratio, replaced alternative phoneme group are " chenglong ".
Certainly, the phoneme to be operated selected can also be " eng ", known based on default phoneme mapping relations and be somebody's turn to do sound to be operated
Plain " eng " corresponding close phoneme is " en ", then replaced alternative phoneme group is " cenlong "；The phoneme to be operated selected
It can also be " l " to be " r " based on close phoneme corresponding with phoneme " l " to be operated is somebody's turn to do known to default phoneme mapping relations, then replace
Alternative phoneme group afterwards is " cengrong "
Further, described to carry out the alternative phoneme group default phoneme group corresponding with preset keyword to match it
Afterwards, the voice search method further include:
When it fails to match, determines the new phoneme quantity to be operated currently being replaced in the phoneme to be replaced, return
The step of phoneme to be operated that quantity is the phoneme quantity to be operated is chosen described in receipt row from the phoneme to be replaced, directly
Extremely in successful match, using the preset keyword of successful match as phonetic search result.
In the concrete realization, the alternative phoneme group an of phoneme is substituted after being matched, possible it fails to match.That
, new phoneme quantity to be operated can be redefined, for example, new phoneme quantity to be operated can be set as 2, the sound to be operated selected
Element can be " c " and " eng ", and the current phoneme group " cenglong " of ratio, replaced alternative phoneme group is " chenlong "；Choosing
Phoneme to be operated out can be " c " and " l ", and the current phoneme group " cenglong " of ratio, replaced alternative phoneme group is
"chengrong".Similarly, new phoneme quantity to be operated is set as 2, replaced alternative phoneme group can for " chenlong ",
" chengrong " and " cenrong ".
Certainly, new phoneme quantity to be operated can also be set as 3, and replaced alternative phoneme group can be " chenrong ".
In short, there are 7 kinds for replaced alternative phoneme group if current phoneme group is " cenglong ".It can be by current phoneme
For group with alternative phonotactics at the array of one 8 members, which can be denoted as arrayA.
By the way that default Substitution Rules are arranged in the present embodiment, part regularly can be selected from multiple phonemes to be replaced
Phoneme to be operated carries out phoneme replacement operation, with final successful match.
It is the flow diagram of voice search method 3rd embodiment of the present invention referring to Fig. 4, Fig. 4, based on shown in above-mentioned Fig. 2
First embodiment, propose the 3rd embodiment of voice search method of the present invention.
In 3rd embodiment, the step S40 is specifically included:
Step S401: by the current phoneme group and the alternative phoneme group default sound corresponding with preset keyword respectively
Plain group is matched.
It is understood that the current phoneme group in view of user's input may just be existing pre- in default resources bank
If phoneme group, i.e., current phoneme group may be correct phoneme group, so, in the matching of phoneme group, current sound can be applied simultaneously
Element group and alternative phoneme group.
Correspondingly, the step S50, specifically includes:
Step S501: any phoneme group and the default phoneme in the current phoneme group and the alternative phoneme group
When group successful match, using the preset keyword of successful match as phonetic search result.
It should be understood that if current phoneme group is consistent with default phoneme group, using current phoneme group as phonetic search
As a result；If alternative phoneme group is consistent with default phoneme group, using alternative phoneme group as phonetic search result.
Further, described that the current phoneme group and the alternative phoneme group is corresponding with preset keyword pre- respectively
If phoneme group is matched, specifically include:
The current phoneme group is revised as default phoneme group corresponding with preset keyword by default edit mode, and
The first modification number that statistics is modified by the default edit mode；
The alternative phoneme group is revised as the default phoneme group by the default edit mode, and counts and passes through institute
State the second modification number that default edit mode is modified；
The first modification number is compared with default modification number respectively with the second modification number；
Correspondingly, any phoneme group in the current phoneme group and the alternative phoneme group and the default sound
When element group successful match, using the preset keyword of successful match as phonetic search as a result, specifically including:
Any modification number in the first modification number and the second modification number is less than the default modification
When number, the corresponding preset keyword of modification number of the default modification number will be less than as phonetic search result.
It is understood that judgement whether modification number under default edit mode can be used as successful match according to
According to.
It should be understood that default edit mode can be the default corresponding default edit mode of smallest edit distance algorithm,
Specifically, phoneme group can be revised as default resources bank by least editor's number by default smallest edit distance algorithm
In existing default phoneme group.Wherein, least editor's number can be denoted as minimal modifications number, minimal modifications number, which refers to, to be passed through
Increase and delete operation can achieve the duplicate minimum number of phonetic, it is primary to modify to increase a character, deletes one
A character is that modification is primary.For example, " 1234 " are become " 1245 " by deleting 3 and increasing by 5, minimal modifications number is 2 times.
In the concrete realization, for example, current phoneme group is if existing default phoneme group is " chenglong "
" cenglong ", it will be apparent that, a character " h " will be increased, minimal modifications number is 1 time.
It is understood that corresponding by the corresponding minimal modifications number of statistics current phoneme group, alternative phoneme group respectively
Minimal modifications number then the two minimal modifications numbers are compared with default modification number.If default modification time
Number is set as 1, then minimal modifications number is less than the corresponding preset keyword of phoneme group of default modification number 1 as phonetic search
As a result.For example, current phoneme group is " cenglong ", corresponding most light maintenance if existing default phoneme group is " chenglong "
Changing number is 1 time；Alternative phoneme group is " chenglong ", and corresponding minimal modifications number is 0 time.As it can be seen that alternative phoneme group pair
The minimal modifications number answered is less than default modification number 1, then can search using the corresponding preset keyword of alternative phoneme group as voice
Hitch fruit.
Further, it is described by it is described first modification number and it is described second modification number respectively with default modification number into
Row compares, and specifically includes:
The first matching degree calculated by preset matching degree calculation formula according to the first modification number, according to described the
Two modification numbers calculate the second matching degree by the preset matching degree calculation formula；
First matching degree is compared with preset matching degree respectively with second matching degree；
Correspondingly, any modification number in the first modification number and the second modification number is less than institute
When stating default modification number, the corresponding preset keyword of modification number of the default modification number will be less than as phonetic search
As a result, specifically including:
It, will when any matching degree in first matching degree and second matching degree is spent greater than the preset matching
Greater than the preset matching degree the corresponding preset keyword of matching degree as phonetic search result.
It is understood that matching degree can be measured based on minimal modifications number, and using matching degree as successful match with
No judgment basis.
In the concrete realization, preset matching degree calculation formula can be,
P=1-q/ (2*m)；
Wherein, p indicates matching degree, and q indicates that minimal modifications number, m indicate the total of current phoneme group or alternative phoneme group
Number of characters.For example, if q is 1, m 8, then calculated matching degree p=1-1/16=0.94.It is apparent that minimal modifications number
Numerical value is lower, indicates that matching is higher, minimal modifications number, which is 0, to be indicated to exactly match；The numerical value of matching degree is higher, expression
It is higher with property.
It should be understood that settable preset matching degree is 0.98, if corresponding first matching degree of the first modification number is small
In preset matching degree, corresponding second matching degree of the second modification number is greater than preset matching degree, then can be corresponding by alternative phoneme group
Preset keyword as phonetic search result.
In addition, the successful preset matching degree of tag match can be arranged according to the load capacity of search server, for example, in advance
If matching degree is 1, then it represents that exactly match, the fruiting quantities searched can seldom, and the operating pressure of server can be smaller；Instead
It, matching degree requirement is lower, and the quantity of search result can be more, and the operating pressure of server can be bigger.For example, if preset matching degree
It is 0.5, then it represents that phonetic search result can also be added by admitting of part phoneme mismatch.
Further, by adaptively adjusting preset matching degree, the default pass in phonetic search result finally obtained
Key word may not only 1, so, height sequence can be carried out according to the numerical value of matching degree.For example, phonetic search result can be fed back
It is shown to TV pusher side, and when showing phonetic search result, the high preset keyword of matching degree can be come to front.Its
In, it, can be corresponding by current phoneme group if high as the matching degree of current phoneme group matching degree corresponding with alternative phoneme group
Preset keyword comes front, and search terminates at this time.
Further, described that speech recognition is carried out to the voice messaging, it obtains corresponding with the voice messaging current
Phoneme group, specifically includes:
Speech recognition is carried out to the voice messaging, to obtain the corresponding current keyword of the voice messaging；
Phoneme conversion is carried out to the current keyword, to obtain current phoneme group corresponding with the current keyword.
It in the concrete realization, can be first by voice messaging speech recognition at corresponding text after getting one section of voice messaging
Word, for example, the text being identified as can be " Xiao Ming "；The text " Xiao Ming " is subjected to phonetic conversion again, it is " small to parse the text
It is bright " corresponding phonetic " xiaoming ".
In addition, it is contemplated that the case where there is unisonance difference words in Chinese, still, the present embodiment are completed based on phoneme group
Matching operation, rather than text itself the case where can evading falling the unisonance difference word, improves the accuracy of search.
Further, described to carry out the alternative phoneme group default phoneme group corresponding with preset keyword to match it
Before, the voice search method further include:
Obtain the preset keyword in default resources bank；
To the default Chinese keyword and preset number keyword progress phoneme conversion in the preset keyword, acquisition pair
The phoneme group to be modified answered；
Preset characters are removed from the phoneme group to be modified, the phoneme group after obtaining removal character, and character will be removed
Phoneme group afterwards is as default phoneme group corresponding with the preset keyword.
In the concrete realization, the preset keyword that largely can search for can be stored with by presetting in resources bank, for example, film
Name, Yan Yuanming, Dao Yanming, Ge Shouming, song title and album name etc..For example, preset keyword may be present, " see you tomorrow, small
It is bright！", then Chinese present in preset keyword and Arabic numerals can be converted into corresponding phonetic, are converted to
" mingtianjian, xiaoming！".Then, settable preset characters may include non-Chinese character in preset characters, it is non-in
Chinese character includes space and punctuation mark etc..From " mingtianjian, xiaoming！" in remove preset characters, can get
"mingtianjianxiaoming".And by " mingtianjianxiaoming " as " see you tomorrow, small with preset keyword
It is bright！" corresponding default phoneme group, it is used for subsequent matching operation.
In addition, may not include English character in preset characters, that is, retain the English character in default phoneme group.
It is understood that the preset keyword stored in conventional default resources bank will exist in the form of text, it is
The default phoneme group in phoneme dimension is added, need additionally to increase in default resources bank such as film name, director's name and is drilled
The phonetic dimension of member's name etc..It is apparent that this kind of processing mode increases data dimension, the data volume of database can be made to increase.
But the mode that the present embodiment is taken not will increase data volume, but before actual match, preset keyword is carried out real-time
Phoneme conversion operation, to obtain default phoneme group.
Search can be improved as search filter condition by setting minimal modifications number and matching degree in the present embodiment
As a result accuracy.After all, the current phoneme group of input may and imperfect i.e. user only have input incomplete title, or
Person, inherently there is the mistakes of part for the current phoneme group of input.In addition, by eliminating the predetermined word in preset keyword
Symbol so that the present embodiment can accurately be matched to the preset keyword containing preset characters, after all, the current phoneme group of input by
Speech recognition obtains, and will not carry the characters such as punctuation mark.
In addition, the embodiment of the present invention also proposes a kind of storage medium, phonetic search program is stored on the storage medium,
Following operation is realized when the phonetic search program is executed by processor:
Obtain voice messaging to be searched；
Speech recognition is carried out to the voice messaging, obtains current phoneme group corresponding with the voice messaging；
Phoneme replacement is carried out to each phoneme in the current phoneme group, obtains alternative phoneme group；
The alternative phoneme group default phoneme group corresponding with preset keyword is matched；
In successful match, using the preset keyword of successful match as phonetic search result.
Further, following operation is also realized when the phonetic search program is executed by processor:
Current phoneme in the current phoneme group is matched with each fuzzy phoneme that default fuzzy phoneme is concentrated；
In successful match, corresponding close phoneme is inquired in default phoneme mapping relations according to the current phoneme；
Current phoneme in the current phoneme group is replaced with into the close phoneme, by the replaced current phoneme of phoneme
Organize alternately phoneme group.
Further, following operation is also realized when the phonetic search program is executed by processor:
Using the current phoneme of successful match as phoneme to be replaced, and count the initial phoneme number of the phoneme to be replaced
Amount；
When the initial phoneme quantity is greater than or equal to default phoneme threshold value, according to default Substitution Rules determine it is described to
The phoneme quantity to be operated currently being replaced in replacement phoneme；
The phoneme to be operated that quantity is the phoneme quantity to be operated is chosen from the phoneme to be replaced, it is described wait operate
Phoneme quantity is less than or equal to the initial phoneme quantity；
Correspondingly, following operation is also realized:
Phoneme to be operated described in the current phoneme group is replaced with into close sound corresponding with the phoneme to be operated
Element, by the replaced current phoneme group of phoneme alternately phoneme group.
Further, following operation is also realized when the phonetic search program is executed by processor:
When it fails to match, determines the new phoneme quantity to be operated currently being replaced in the phoneme to be replaced, return
The step of phoneme to be operated that quantity is the phoneme quantity to be operated is chosen described in receipt row from the phoneme to be replaced, directly
Extremely in successful match, using the preset keyword of successful match as phonetic search result.
Further, following operation is also realized when the phonetic search program is executed by processor:
By the current phoneme group and the alternative phoneme group, default phoneme group corresponding with preset keyword is carried out respectively
Matching；
Correspondingly, following operation is also realized:
It is matched into any phoneme group in the alternative phoneme group with the default phoneme group in the current phoneme group
When function, using the preset keyword of successful match as phonetic search result.
Further, following operation is also realized when the phonetic search program is executed by processor:
The current phoneme group is revised as default phoneme group corresponding with preset keyword by default edit mode, and
The first modification number that statistics is modified by the default edit mode；
The alternative phoneme group is revised as the default phoneme group by the default edit mode, and counts and passes through institute
State the second modification number that default edit mode is modified；
The first modification number is compared with default modification number respectively with the second modification number；
Correspondingly, following operation is also realized:
Any modification number in the first modification number and the second modification number is less than the default modification
When number, the corresponding preset keyword of modification number of the default modification number will be less than as phonetic search result.
Further, following operation is also realized when the phonetic search program is executed by processor:
The first matching degree calculated by preset matching degree calculation formula according to the first modification number, according to described the
Two modification numbers calculate the second matching degree by the preset matching degree calculation formula；
First matching degree is compared with preset matching degree respectively with second matching degree；
Correspondingly, following operation is also realized:
It, will when any matching degree in first matching degree and second matching degree is spent greater than the preset matching
Greater than the preset matching degree the corresponding preset keyword of matching degree as phonetic search result.
Speech recognition first can be carried out to voice messaging in the present embodiment, obtain corresponding current phoneme group；To current phoneme
Each phoneme in group carries out phoneme replacement, obtains alternative phoneme group；By alternative phoneme group default sound corresponding with preset keyword
Plain group is matched；In successful match, using the preset keyword of successful match as phonetic search result.It is apparent that this reality
Apply example in view of input voice messaging there may be mistakes, so, will additionally to current phoneme group carry out phoneme replacement, with
It radiates other more similar alternative phoneme groups and improves search result to improve a possibility that being successfully searched result
Accuracy, solve the lower technical problem of phonetic search accuracy.
In addition, the embodiment of the present invention also proposes that a kind of voice searching device, the voice searching device include: referring to Fig. 5
Voice obtains module 10, for obtaining voice messaging to be searched；
Speech recognition module 20 obtains corresponding with the voice messaging for carrying out speech recognition to the voice messaging
Current phoneme group；
Phoneme replacement module 30 obtains alternative sound for carrying out phoneme replacement to each phoneme in the current phoneme group
Plain group；
Phoneme matching module 40, for will the alternative phoneme group default phoneme group progress corresponding with preset keyword
Match；
Voice search module 50 is used in successful match, using the preset keyword of successful match as phonetic search knot
Fruit.
Speech recognition first can be carried out to voice messaging in the present embodiment, obtain corresponding current phoneme group；To current phoneme
Each phoneme in group carries out phoneme replacement, obtains alternative phoneme group；By alternative phoneme group default sound corresponding with preset keyword
Plain group is matched；In successful match, using the preset keyword of successful match as phonetic search result.It is apparent that this reality
Apply example in view of input voice messaging there may be mistakes, so, will additionally to current phoneme group carry out phoneme replacement, with
It radiates other more similar alternative phoneme groups and improves search result to improve a possibility that being successfully searched result
Accuracy, solve the lower technical problem of phonetic search accuracy.
In one embodiment, the phoneme replacement module 30, be also used to by the current phoneme group current phoneme with
Each fuzzy phoneme that default fuzzy phoneme is concentrated is matched；In successful match, according to the current phoneme in default phoneme
Corresponding close phoneme is inquired in mapping relations；Current phoneme in the current phoneme group is replaced with into the close phoneme,
By the replaced current phoneme group of phoneme alternately phoneme group.
In one embodiment, the voice searching device further include:
Phoneme chooses module, for using the current phoneme of successful match as phoneme to be replaced, and counts described to be replaced
The initial phoneme quantity of phoneme；When the initial phoneme quantity is greater than or equal to default phoneme threshold value, advised according to default replacement
Then determine the phoneme quantity to be operated currently being replaced in the phoneme to be replaced；Quantity is chosen from the phoneme to be replaced
For the phoneme to be operated of the phoneme quantity to be operated, the phoneme quantity to be operated is less than or equal to the initial phoneme number
Amount；
The phoneme replacement module 30, is also used to replace with phoneme to be operated described in the current phoneme group and institute
The corresponding close phoneme of phoneme to be operated is stated, by the replaced current phoneme group of phoneme alternately phoneme group.
In one embodiment, the voice searching device further include:
Recycle matching module, for when it fails to match, determine currently be replaced in the phoneme to be replaced it is new
Phoneme quantity to be operated, return execute it is described chosen from the phoneme to be replaced quantity be the phoneme quantity to be operated to
The step of operating phoneme, until in successful match, using the preset keyword of successful match as phonetic search result.
In one embodiment, the phoneme matching module 40 is also used to the current phoneme group and the alternative phoneme
Default phoneme group corresponding with preset keyword is matched group respectively；
The voice search module 50, any phoneme being also used in the current phoneme group and the alternative phoneme group
When group is with the default phoneme group successful match, using the preset keyword of successful match as phonetic search result.
In one embodiment, the phoneme matching module 40 is also used to by presetting edit mode for the current phoneme
Group is revised as default phoneme group corresponding with preset keyword, and counts first to modify by the default edit mode
Modify number；The alternative phoneme group is revised as the default phoneme group by the default edit mode, and counts and passes through
The second modification number that the default edit mode is modified；By the first modification number and the second modification number point
It is not compared with default modification number；
The voice search module 50 is also used in the first modification number and any in the second modification number
When modifying number less than the default modification number, the corresponding default key of modification number of the default modification number will be less than
Word is as phonetic search result.
In one embodiment, the phoneme matching module 40 is also used to according to the first modification number by default
The first matching degree is calculated with degree calculation formula, the preset matching degree calculation formula meter is passed through according to the second modification number
Calculate the second matching degree；First matching degree is compared with preset matching degree respectively with second matching degree；
The voice search module 50, any matching being also used in first matching degree and second matching degree
When degree is spent greater than the preset matching, the corresponding preset keyword of matching degree that will be greater than the preset matching degree is searched as voice
Hitch fruit.
The other embodiments or specific implementation of voice searching device of the present invention can refer to above-mentioned each method and implement
Example, details are not described herein again.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the system that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or system.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.If listing equipment for drying
Unit claim in, several in these devices, which can be, to be embodied by the same item of hardware.Word first,
Second and the use of third etc. do not indicate any sequence, can be title by these word explanations.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in a storage medium
In (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that a terminal device (can be mobile phone, computer, clothes
Business device, air conditioner or the network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of voice search method, which is characterized in that the voice search method the following steps are included:
Obtain voice messaging to be searched；
Speech recognition is carried out to the voice messaging, obtains current phoneme group corresponding with the voice messaging；
Phoneme replacement is carried out to each phoneme in the current phoneme group, obtains alternative phoneme group；
The alternative phoneme group default phoneme group corresponding with preset keyword is matched；
In successful match, using the preset keyword of successful match as phonetic search result.
2. voice search method as described in claim 1, which is characterized in that each phoneme in the current phoneme group
Phoneme replacement is carried out, alternative phoneme group is obtained, specifically includes:
Current phoneme in the current phoneme group is matched with each fuzzy phoneme that default fuzzy phoneme is concentrated；
In successful match, corresponding close phoneme is inquired in default phoneme mapping relations according to the current phoneme；
Current phoneme in the current phoneme group is replaced with into the close phoneme, the replaced current phoneme group of phoneme is made
For alternative phoneme group.
3. voice search method as claimed in claim 2, which is characterized in that it is described in successful match, according to described current
After phoneme inquires corresponding close phoneme in default phoneme mapping relations, the voice search method further include:
Using the current phoneme of successful match as phoneme to be replaced, and count the initial phoneme quantity of the phoneme to be replaced；
When the initial phoneme quantity is greater than or equal to default phoneme threshold value, determined according to default Substitution Rules described to be replaced
The phoneme quantity to be operated currently being replaced in phoneme；
The phoneme to be operated that quantity is the phoneme quantity to be operated, the phoneme to be operated are chosen from the phoneme to be replaced
Quantity is less than or equal to the initial phoneme quantity；
Correspondingly, the current phoneme by the current phoneme group replaces with the close phoneme, and phoneme is replaced
Current phoneme group alternately phoneme group, specifically includes:
Phoneme to be operated described in the current phoneme group is replaced with into close phoneme corresponding with the phoneme to be operated, it will
The replaced current phoneme group of phoneme alternately phoneme group.
4. voice search method as claimed in claim 3, which is characterized in that described by the alternative phoneme group and default key
After the corresponding default phoneme group of word is matched, the voice search method further include:
When it fails to match, the new phoneme quantity to be operated currently being replaced in the phoneme to be replaced is determined, return is held
The step of phoneme to be operated that quantity is the phoneme quantity to be operated is chosen described in row from the phoneme to be replaced, until
When successful match, using the preset keyword of successful match as phonetic search result.
5. voice search method as described in claim 1, which is characterized in that described by the alternative phoneme group and default key
The corresponding default phoneme group of word is matched, and is specifically included:
By the current phoneme group and the alternative phoneme group, default phoneme group corresponding with preset keyword is matched respectively；
Correspondingly, described in successful match, using the preset keyword of successful match as phonetic search as a result, specifically including:
When any phoneme group and the default phoneme group successful match in the current phoneme group and the alternative phoneme group,
Using the preset keyword of successful match as phonetic search result.
6. voice search method as claimed in claim 5, which is characterized in that described by the current phoneme group and described alternative
Default phoneme group corresponding with preset keyword is matched phoneme group respectively, is specifically included:
The current phoneme group is revised as default phoneme group corresponding with preset keyword by default edit mode, and is counted
The the first modification number modified by the default edit mode；
The alternative phoneme group is revised as the default phoneme group by the default edit mode, and is counted by described pre-
If the second modification number that edit mode is modified；
The first modification number is compared with default modification number respectively with the second modification number；
Correspondingly, any phoneme group and the default phoneme group in the current phoneme group and the alternative phoneme group
When successful match, using the preset keyword of successful match as phonetic search as a result, specifically including:
Any modification number in the first modification number and the second modification number is less than the default modification number
When, the corresponding preset keyword of modification number of the default modification number will be less than as phonetic search result.
7. voice search method as claimed in claim 6, which is characterized in that described by the first modification number and described the
Two modification numbers are compared with default modification number respectively, are specifically included:
The first matching degree is calculated by preset matching degree calculation formula according to the first modification number, is repaired according to described second
Change number and the second matching degree is calculated by the preset matching degree calculation formula；
First matching degree is compared with preset matching degree respectively with second matching degree；
Correspondingly, any modification number modified in the first modification number with described second in number is less than described pre-
If modify number, the corresponding preset keyword of modification number of the default modification number will be less than as phonetic search knot
Fruit specifically includes:
When any matching degree in first matching degree and second matching degree is spent greater than the preset matching, it will be greater than
The corresponding preset keyword of matching degree of the preset matching degree is as phonetic search result.
8. a kind of user equipment, which is characterized in that the user equipment includes: memory, processor and is stored in the storage
Phonetic search program can be run on device and on the processor, the phonetic search program is realized when being executed by the processor
The step of voice search method as described in any one of claims 1 to 7.
9. a kind of storage medium, which is characterized in that be stored with phonetic search program, the phonetic search journey on the storage medium
The step of voice search method as described in any one of claims 1 to 7 is realized when sequence is executed by processor.
10. a kind of voice searching device, which is characterized in that the voice searching device includes:
Voice obtains module, for obtaining voice messaging to be searched；
Speech recognition module obtains corresponding with the voice messaging current for carrying out speech recognition to the voice messaging
Phoneme group；
Phoneme replacement module obtains alternative phoneme group for carrying out phoneme replacement to each phoneme in the current phoneme group；
Phoneme matching module, for matching the alternative phoneme group default phoneme group corresponding with preset keyword；
Voice search module is used in successful match, using the preset keyword of successful match as phonetic search result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910219560.9A CN109947955A (en) | 2019-03-21 | 2019-03-21 | Voice search method, user equipment, storage medium and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910219560.9A CN109947955A (en) | 2019-03-21 | 2019-03-21 | Voice search method, user equipment, storage medium and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109947955A true CN109947955A (en) | 2019-06-28 |
Family
ID=67011282
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910219560.9A Pending CN109947955A (en) | 2019-03-21 | 2019-03-21 | Voice search method, user equipment, storage medium and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109947955A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101236743A (en) * | 2007-01-30 | 2008-08-06 | 国际商业机器公司 | System and method for generating high quality speech |
| CN104238991A (en) * | 2013-06-21 | 2014-12-24 | 腾讯科技（深圳）有限公司 | Voice input matching method and voice input matching device |
| CN106570180A (en) * | 2016-11-10 | 2017-04-19 | 北京百度网讯科技有限公司 | Artificial intelligence based voice searching method and device |
| CN107633017A (en) * | 2017-08-23 | 2018-01-26 | 西安理工大学 | A kind of fuzzy set construction method of Chinese key |
| WO2018174968A1 (en) * | 2017-03-21 | 2018-09-27 | Rovi Guides, Inc. | Systems and methods for increasing language accessability of media content |
| CN108831473A (en) * | 2018-03-30 | 2018-11-16 | 联想(北京)有限公司 | A kind of audio-frequency processing method and device |
| CN109036419A (en) * | 2018-07-23 | 2018-12-18 | 努比亚技术有限公司 | A kind of speech recognition match method, terminal and computer readable storage medium |
-
2019
- 2019-03-21 CN CN201910219560.9A patent/CN109947955A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101236743A (en) * | 2007-01-30 | 2008-08-06 | 国际商业机器公司 | System and method for generating high quality speech |
| CN104238991A (en) * | 2013-06-21 | 2014-12-24 | 腾讯科技（深圳）有限公司 | Voice input matching method and voice input matching device |
| CN106570180A (en) * | 2016-11-10 | 2017-04-19 | 北京百度网讯科技有限公司 | Artificial intelligence based voice searching method and device |
| WO2018174968A1 (en) * | 2017-03-21 | 2018-09-27 | Rovi Guides, Inc. | Systems and methods for increasing language accessability of media content |
| CN107633017A (en) * | 2017-08-23 | 2018-01-26 | 西安理工大学 | A kind of fuzzy set construction method of Chinese key |
| CN108831473A (en) * | 2018-03-30 | 2018-11-16 | 联想(北京)有限公司 | A kind of audio-frequency processing method and device |
| CN109036419A (en) * | 2018-07-23 | 2018-12-18 | 努比亚技术有限公司 | A kind of speech recognition match method, terminal and computer readable storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 陆勇: "《面向信息检索的汉语同义词自动识别》", 31 December 2009 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108847241B (en) | Method for recognizing conference voice as text, electronic device and storage medium | |
| US11069335B2 (en) | Speech synthesis using one or more recurrent neural networks | |
| US7933774B1 (en) | System and method for automatic generation of a natural language understanding model | |
| CN110110041A (en) | Wrong word correcting method, device, computer installation and storage medium | |
| CN103365925B (en) | Obtain polyphone phonetic, method based on phonetic retrieval and related device thereof | |
| CN101593516B (en) | Method and system for speech synthesis | |
| EP2317507B1 (en) | Corpus compilation for language model generation | |
| CN110164435A (en) | Audio recognition method, device, equipment and computer readable storage medium | |
| CN111798868B (en) | Voice forced alignment model evaluation method and device, electronic equipment and storage medium | |
| CN108304375A (en) | A kind of information identifying method and its equipment, storage medium, terminal | |
| CN111310440B (en) | Text error correction method, device and system | |
| CN107229627B (en) | Text processing method and device and computing equipment | |
| CN109299245B (en) | Method and device for recalling knowledge points | |
| US8731930B2 (en) | Contextual voice query dilation to improve spoken web searching | |
| CN103559880B (en) | Voice entry system and method | |
| CN111767393A (en) | Text core content extraction method and device | |
| WO2021218027A1 (en) | Method and apparatus for extracting terminology in intelligent interview, device, and medium | |
| CN111930792A (en) | Data resource labeling method and device, storage medium and electronic equipment | |
| US20080002886A1 (en) | Adapting a neural network for individual style | |
| US8103506B1 (en) | Free text matching system and method | |
| CN110275938B (en) | Knowledge extraction method and system based on unstructured document | |
| CN111401012A (en) | Text error correction method, electronic device and computer readable storage medium | |
| CN112906376B (en) | Self-adaptive matching user English learning text pushing system and method | |
| CN107894882A (en) | A kind of pronunciation inputting method of mobile terminal | |
| CN106297766B (en) | Phoneme synthesizing method and system |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190628 |
|
| RJ01 | Rejection of invention patent application after publication |