CN104636323B - Handle the method and device of speech text - Google Patents
Handle the method and device of speech text Download PDFInfo
- Publication number
- CN104636323B CN104636323B CN201310554808.XA CN201310554808A CN104636323B CN 104636323 B CN104636323 B CN 104636323B CN 201310554808 A CN201310554808 A CN 201310554808A CN 104636323 B CN104636323 B CN 104636323B
- Authority
- CN
- China
- Prior art keywords
- processing
- disappear
- matched rule
- matched
- matching
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/683—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
- G06F16/685—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using automatically derived transcript of audio data, e.g. lyrics
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Library & Information Science (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses a kind of method and device for handling speech text, belong to technical field of information processing.Method includes：Entity mapping is named to speech text, obtains the first mapping result；Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；Second mapping result is matched with the preset rules including regularity, if obtaining a matched rule, speech text handled according to an obtained matched rule.The present invention is mapped by being named entity to speech text, obtains the first mapping result；Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；The second mapping result is matched with the preset rules including regularity afterwards, and speech text is handled according to an obtained matched rule, so as to which the configuration format of regularity and name entity rule is unified, the scope of application of processing speech text technology is expanded, so that the processing mode of speech text is optimized.
Description
Technical field
The present invention relates to technical field of information processing, more particularly to a kind of method and device for handling speech text.
Background technology
With the continuous development of the information processing technology, man-machine interaction is carried out with natural language and become a reality.Realize man-machine friendship
Mutual key is that the natural language for wanting accurate understanding user to send is instructed and operated accordingly.User is sending natural language
After instruction, the instruction is converted into speech text, how to handle speech text, turns into problem of concern.
Prior art has the mode of two kinds of processing speech texts, first way：Vocabulary mapping is carried out to speech text, obtained
To mapping result；Location parameter in mapping result is extracted by regular screens, obtains screens extraction result；Screens is extracted and tied
Fruit and preset rules are matched, and obtain a matched rule；According to an obtained matched rule to speech text at
Reason.Wherein, regular screens is the specified location obtained according to regularity.The second way：Reality is named to speech text
Body maps, and obtains the first mapping result；Vocabulary mapping is carried out to the result of name entity mapping, obtains the second mapping result；Will
Second mapping result and preset rules are matched, and obtain a matched rule；According to an obtained matched rule to voice
Text is handled.
During the present invention is realized, inventor has found that prior art at least has problems with：
Regularity underaction, actual language environment is not suitable for it.Entity rule is named for more fixed language
Speech environment and the scene for being not suitable for establishing entity name storehouse, it is impossible to play one's part to the full.Therefore, two kinds of prior art offer
The mode of processing speech text is respectively provided with certain limitation, causes the processing mode of speech text not enough to optimize.
The content of the invention
In order to solve problem of the prior art, the embodiments of the invention provide a kind of method and dress for handling speech text
Put.The technical scheme is as follows：
First aspect, there is provided a kind of method for handling speech text, methods described include：
Entity mapping is named to speech text, obtains the first mapping result；
Vocabulary mapping is carried out to first mapping result, obtains the second mapping result；
Second mapping result is matched with the preset rules including regularity, if obtaining a matching rule
Then, then the speech text is handled according to an obtained matched rule.
Second aspect, there is provided a kind of device for handling speech text, described device include：
First mapping block, for being named entity mapping to speech text, obtain the first mapping result；
Second mapping block, for carrying out vocabulary mapping to first mapping result, obtain the second mapping result；
Matching module, for second mapping result to be matched with the preset rules including regularity；
First processing module, for when obtaining a matched rule, according to an obtained matched rule to institute's predicate
Sound text is handled.
The third aspect, there is provided a kind of terminal, the terminal include processor and memory, be stored with the memory
One or more than one program, one or more than one program are loaded by the processor and performed to realize
The method for the processing speech text that first aspect is provided.
Fourth aspect, there is provided a kind of computer-readable recording medium, be stored with one or one in the storage medium
Program more than individual, one or more than one program are loaded by processor and performed to realize that first aspect is provided
Processing speech text method.
The beneficial effect that technical scheme provided in an embodiment of the present invention is brought is：
Mapped by being named entity to speech text, obtain the first mapping result；Word is carried out to the first mapping result
Film festival is penetrated, and obtains the second mapping result；The second mapping result is matched with the preset rules including regularity afterwards, and
A matched rule according to obtaining is handled speech text, so that the configuration lattice by regularity and name entity rule
Formula is unified, processing speech text technology is not only suitable for actual language environment, is applied to fixed language environment again, expands
The scope of application of speech text technology is handled, so that the processing mode of speech text is optimized.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, make required in being described below to embodiment
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is the method flow diagram for the processing speech text that the embodiment of the present invention one provides；
Fig. 2 is the method flow diagram for the processing speech text that the embodiment of the present invention two provides；
Fig. 3 is the apparatus structure schematic diagram for the first processing speech text that the embodiment of the present invention three provides；
Fig. 4 is the apparatus structure schematic diagram for second of processing speech text that the embodiment of the present invention three provides；
Fig. 5 is the apparatus structure schematic diagram for the third processing speech text that the embodiment of the present invention three provides；
Fig. 6 is the terminal structure schematic diagram for the processing speech text that the embodiment of the present invention four provides.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to embodiment party of the present invention
Formula is described in further detail.
Embodiment one
The embodiments of the invention provide a kind of method for handling speech text, and referring to Fig. 1, method flow includes following step
Suddenly：
101：Entity mapping is named to speech text, obtains the first mapping result；
102：Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；
Further, before carrying out vocabulary mapping to the first mapping result, including：
One or more name entity in first mapping result is expanded into corresponding speech text before mapping successively, obtained
At least two the 3rd mapping results；
Vocabulary mapping is carried out to the first mapping result, including：
The non-mapped speech text for name entity in each 3rd mapping result is subjected to vocabulary mapping, obtains second
Mapping result.
103：Second mapping result is matched with the preset rules including regularity, if obtaining a matching rule
Then, then speech text is handled according to an obtained matched rule.
Further, after the second mapping result is matched with the preset rules including regularity, in addition to：
If obtaining at least two matched rules, obtained all matched rules are carried out with match width successively and is disappeared Qi Chu
Reason, matching weighted value disappear qi processing, the hit quantity of qi processing, matching density that disappear disappear qi processing and name entity weights disappear Qi Chu
At least one of reason disappears qi processing, until obtain the matched rule after the qi that disappears；
Speech text is handled according to the matched rule after the obtained qi that disappears.
Further, match width disappear qi processing, including：
It is determined that the match width of the second mapping result corresponding to each matched rule, by the matched rule that match width is maximum
Matched rule as the qi processing that disappears by match width.
Further, matching weighted value disappear qi processing, including：
According to the vocabulary pre-set with naming the weighted value of entity to determine that each pending weighted value qi that disappears that match is handled
Matched rule weighted value；
Matched rule using the maximum matched rule of weighted value as the qi processing that disappears through overmatching weighted value；
Wherein, it is pending matching weighted value disappear qi processing matched rule be by match width disappear qi processing matching advise
Then.
Further, matching density disappear qi processing, including：
The matched rule of qi processing and the matching proportion of the second mapping result it is determined that each pending matching density disappears, general
Matched rule with the maximum matched rule of proportion as the qi processing that disappears through overmatching density；
Wherein, the disappear matched rule of qi processing of each pending matching density is the matching of qi processing of disappearing by match width
Rule.
Further, hit quantity disappear qi processing, including：
The parameter hit quantity of the matched rule of qi processing it is determined that each pending hit quantity disappears, parameter are real for name
One in body, vocabulary and location parameter；
Matched rule using the maximum matching result of parameter hit quantity as the qi processing that disappears by hit quantity；
Wherein, the disappear matched rule of qi processing of each pending hit quantity is the matching of qi processing of disappearing by match width
Rule.
Further, name entity weights disappear qi processing, including：
The weighted value of entity is named in the matched rule of qi processing it is determined that each pending name entity weights disappear, will be named
Matched rule of the maximum matched rule of the weighted value of entity as the qi processing that disappears by name entity weights；
Wherein, the disappear matched rule of qi processing of each pending name entity weights is to disappear qi processing by match width
Matched rule.
The method that the present embodiment provides, mapped by being named entity to speech text, obtain the first mapping result；It is right
First mapping result carries out vocabulary mapping, obtains the second mapping result；Afterwards by the second mapping result with including regularity
Preset rules are matched, and speech text is handled according to an obtained matched rule, so as to by regularity and
Name the configuration format of entity rule unified, processing speech text technology is not only suitable for actual language environment, be applied to again
Fixed language environment, the scope of application of processing speech text technology is expanded, so that the processing mode of speech text obtains
Optimization.
Embodiment two
The embodiments of the invention provide a kind of method for handling speech text, with reference to the content of above-described embodiment one, referring to
Fig. 2, method flow include：
201：Entity mapping is named to speech text, obtains the first mapping result；
Specifically, entity mapping is named to speech text, is included but is not limited to：Establish name entity storehouse；In voice
Being searched in text can be identified as naming voice this paper of the name entity in entity storehouse, and is replaced and found with name entity
Speech text.It should be noted that name entity is collected in substantial amounts of information from network, and name the identification of entity
It is the implementation using every field independence dictionary tree, i.e., can supports the feelings for naming entity all to overlap or partially overlap
All name entities are found out under condition.
In order to make it easy to understand, using speech text as " I wants to listen for the first time ", the Chinese braces of entity is named【】It is expressed as
Example illustrates.Speech text " I thinks " and " first time " all corresponding name entity in the name entity of foundation【Title of the song】；In language
Speech text " I thinks " and " first time " are found in sound text can be identified as naming in entity storehouse【Title of the song】, with life
Name entity【Title of the song】Replace the speech text " I thinks " found and the first mapping result is obtained after " first time ", i.e.,【Title of the song】Listen
【Title of the song】.
202：One or more name entity in first mapping result is expanded into corresponding speech text before mapping successively,
Obtain at least two the 3rd mapping results；
For the step, because name entity may be overlapping with vocabulary, then by one or more life in the first mapping result
Name entity expands into corresponding speech text before mapping successively, obtains at least two the 3rd mapping results, so as to add mapping
As a result, avoid accurately handling speech text in the case of naming entity and vocabulary overlapping.
If it should be noted that do not have to name the overlapping situation of entity part in the first mapping result, and first reflects
A shared n name entity in result is penetrated, then one or more name entity is expanded into corresponding speech text before mapping successively
After obtain 2^n the 3rd mapping results.
In order to make it easy to understand, still illustrated so that speech text is " I wants to listen for the first time " as an example.Wherein, speech text pair
The first mapping result answered is【Title of the song】Listen【Title of the song】.Will【Title of the song】Listen【Title of the song】In two name entities expand into reflect successively
Speech text corresponding to before penetrating, is obtained【Title of the song】Listen【Title of the song】, I want to listen【Title of the song】、【Title of the song】Listen for the first time, I want to listen first
It is secondary, this four the 3rd mapping results.
203:The non-mapped speech text for name entity in each 3rd mapping result is subjected to vocabulary mapping, obtained
Second mapping result；
For the step, the vocabulary carried out in vocabulary mapping is human configuration, identical meanings will be expressed in ordinary language
Speech text be mapped as identical vocabulary.Because the speech text of identical meanings may correspond to not in the language environment of reality
Same speech text, the non-mapped speech text for name entity is subjected to vocabulary mapping, it is possible to reduce the work of repeated matching
Measure.On the non-mapped speech text for name entity in each 3rd mapping result to be carried out to the mode of vocabulary mapping,
The present embodiment is not especially limited.
In order to make it easy to understand, illustrated by taking the 3rd mapping result in above-mentioned steps 202 as an example.Wherein, " I wants to listen "
It is mapped to vocabulary<Play>.The non-mapped speech text for name entity in each 3rd mapping result is carried out into vocabulary to reflect
Penetrate, obtaining the second mapping result is：【Title of the song】Listen【Title of the song】,<Play>【Title of the song】,【Title of the song】Listen for the first time,<Play>First
It is secondary.
204:Second mapping result is matched with the preset rules including regularity；
Specifically, preset rules include but is not limited to：Regularity and other default rules.Wherein, it is default other
Rule includes but is not limited to set the rule for meeting speech habits.The present embodiment does not make specific restriction to other default rules,
Other default rules can be arranged as required in practical application, the present embodiment is not especially limited to this.
Further, the second mapping result is matched with the preset rules including regularity, included but is not limited to：
Location parameter in second mapping result is extracted by regular screens, obtains screens extraction result；By screens extraction result and in advance
If other rules matched.Wherein, regular screens is the specified location obtained according to regularity；Carried by regular screens
The location parameter in the second mapping result is taken, i.e., is joined according to position of the regularity in specified location extracts the second mapping result
Number., can be directly by the second mapping result when the location parameter for not having to extract by regular screens in the second mapping result
Matched with other default rules, and then the rule matched.
It should be noted that after the second mapping result is matched with the preset rules including regularity, if
To a matched rule, then step 205 is performed；If obtaining at least two matched rules, step 206 is performed.
In order to make it easy to understand, with the second mapping result：【Title of the song】Listen【Title of the song】,<Play>【Title of the song】,【Title of the song】Listen first
It is secondary,<Play>For the first time, illustrated exemplified by.Due to the position for not having to extract by regular screens in the second mapping result
Parameter, then directly the second mapping result and other default rules are matched.When other default rules are<Play>【Song
Name】When, obtaining the second mapping result corresponding with the rule is<Play>【Title of the song】, and then obtain a matched rule, i.e.,<Broadcast
Put>【Title of the song】.
In another example speech text is " broadcasting Corner With Love ", then will be according to 201 to 203 pairs of speech texts of above-mentioned steps
Handled, obtained<Play>【Video name】,<Play>【Title of the song】Run into love,<Play>Corner With Love, these three second
Mapping result.Due to the location parameter for not having to extract by regular screens in these three second mapping results, then directly will
These three second mapping results and other default rules are matched.Because other default rules be both<Play>【Song
Name】, can also be<Play>【Video name】, then obtaining the second mapping result corresponding with the two rules is<Play>【Depending on
Frequency title】,<Play>【Title of the song】Run into love；And then obtain two matched rules, i.e.,<Play>【Title of the song】,<Play>【Video name
Claim】.
205:A matched rule according to obtaining is handled speech text；
For the step, due to obtaining a matched rule in above-mentioned steps 204, then according to an obtained matched rule
Speech text is handled.On the mode handled speech text, the present embodiment is not especially limited.
For example, still so that speech text is " broadcasting Corner With Love " as an example, if an obtained matched rule is<Play
>【Video name】, then the step 205 broadcasting is regarded when being handled according to an obtained matched rule speech text
The video of frequently entitled " Corner With Love ".
206:Obtained all matched rules are carried out successively match width disappear qi processing, matching weighted value disappear qi processing,
Matching density disappear qi processing, hit quantity disappear qi processing and name entity weights disappear qi processing at least one of the Qi Chu that disappears
Reason, until obtaining the matched rule after the qi that disappears；
For the step, due to obtaining at least two matched rules in above-mentioned steps 204, in order that the processing of speech text
As a result more accurate, the method that the present embodiment provides takes the mode for the qi processing that to all matched rules disappear.Wherein, disappear
The process of qi processing, includes but is not limited to：Obtained all matched rules are carried out successively match width disappear qi processing, matching plus
Weights disappear qi processing, matching density disappear qi processing, hit quantity disappear qi processing and name entity weights disappear qi processing in extremely
Few one kind disappears qi processing, until obtaining the matched rule after the qi that disappears.
On match width disappear qi processing mode, include but is not limited to：It is determined that second reflected corresponding to each matched rule
The match width of result is penetrated, the matched rule using the maximum matched rule of match width as the qi processing that disappears by match width.
The determination of match width is the original position of first parameter from the second mapping result to the end position of a last parameter.
Wherein, parameter includes but is not limited to：Name entity, vocabulary and location parameter.It should be noted that in order to avoid no actual meaning
Speech text of justice etc. influences the processing to speech text, it is determined that needing to set threshold value before match width, the threshold value is used for
The match width of the result of the second mapping is determined, i.e., thinks that the match width of the second mapping result is consistent in threshold range.If
The size for the threshold value put, the present embodiment are not especially limited, and arbitrary threshold value can be arranged as required in practical application.
In order to make it easy to understand, illustrated so that speech text is " broadcasting Corner With Love " as an example.Wherein, setting threshold value is
2 bytes.After being matched to speech text, obtain<Play>【Title of the song】,<Play>【Video name】, the two matching rule
Then；The two matched rules correspond to respectively<Play>【Title of the song】Run into love,<Play>【Video name】, the two second mapping knots
Fruit.Disappear qi processing to obtained all matched rules progress match width, determines matched rule<Play>【Title of the song】Corresponding
Two mapping results<Play>【Title of the song】Run into the match width of love：First parameter is beginning of the sentence<Play>, last parameter is
【Title of the song】；And because the threshold value of setting is 2 bytes, so " running into love " these three words can not be ignored, then matched rule<Broadcast
Put>【Title of the song】Match width be not from beginning of the sentence to sentence tail；Matched rule<Play>【Video name】Corresponding second mapping knot
Fruit<Play>【Video name】Match width：First parameter be<Play>, last parameter is【Video name】, then
With rule<Play>【Video name】Match width be from beginning of the sentence to sentence tail, compare matched rule<Play>【Title of the song】Matching it is wide
Degree is bigger；And then by the maximum matched rule of match width<Play>【Video name】Disappear what qi was handled as by match width
Matched rule, obtain the matched rule after the qi that disappears.
On matching weighted value disappear qi processing mode, include but is not limited to：It is real with name according to the vocabulary pre-set
The weighted value of body determine each pending matching weighted value disappear qi processing matched rule weighted value；
Matched rule using the maximum matched rule of weighted value as the qi processing that disappears through overmatching weighted value；
Wherein, it is pending matching weighted value disappear qi processing matched rule be by match width disappear qi processing matching advise
Then.
It should be noted that the weighted value size of the vocabulary pre-set and name entity, the present embodiment do not limit specifically
It is fixed.Because vocabulary is human configuration, and entity is named to collect from network in substantial amounts of information, then vocabulary under normal circumstances
Weight than name entity weight it is big.
In order to make it easy to understand, illustrated so that speech text is " asking for neighbouring restaurant " as an example.Speech text is entered
After row matching, obtain<Search><Dining room>,<Search>【Restaurant name】, the two matched rules；The two matched rules are right respectively
Should<Search>Neighbouring<Dining room>,<Search>Neighbouring【Restaurant name】, the two second mapping results.All matchings to obtaining
Rule carry out match width disappear qi processing, determine that the match width of the two matched rules is identical.Because match width disappears after qi
Matched rule it is more than one, then the matched rule after the qi that disappears to match width carry out match weighted value disappear qi processing.It is if pre-
The vocabulary first set and the weight ratio of the weighted value of name entity are 2：1, it is determined that matched rule<Search><Dining room>Weighting
Value is more than matched rule<Search>【Restaurant name】, i.e., the maximum matching of weighted value, which is advised, is<Search><Dining room>；And then will<Search><
Dining room>As the matched rule for the qi processing that disappears through overmatching weighted value, the matched rule after the qi that disappears is obtained.
On matching density disappear qi processing mode, include but is not limited to：The Qi Chu it is determined that each pending matching density disappears
The matching proportion of the matched rule of reason and the second mapping result, the maximum matched rule of proportion will be matched as through overmatching density
The matched rule of the qi that disappears processing；
Wherein, the disappear matched rule of qi processing of each pending matching density is the matching of qi processing of disappearing by match width
Rule.
It should be noted that matching density disappears, qi processing is to be directed to situation about including similar parameter in matched rule.For example,
All included in matched rule in vocabulary, or matched rule all comprising name entity.
In order to make it easy to understand, illustrated so that speech text is " playing the small epoch " as an example.Speech text is matched
Afterwards, obtain<Play>【Movie name】,<Play>【Title of the song】, the two matched rules；The two matched rules correspond to respectively<Play
>【Movie name】,<Play>It is small【Title of the song】, the two second mapping results.Match width is carried out to obtained all matched rules
Disappear qi, determines that the match width of the two matched rules is identical.The matched rule after qi that disappears due to match width is more than one, then
The matched rule after qi that disappears to match width match weighted value and disappeared qi processing, determines the weighted value phase of the two matched rules
Together.Because matching weighted value disappears, the matched rule after qi is more than one, then the matched rule after the qi that disappears to matching weighted value is carried out
Matching density disappears qi.It is determined that<Play>【Movie name】With<Play>【Movie name】Matching proportion be 100%,<Play>【Title of the song】
With<Play>It is small【Title of the song】Matching proportion be 80%, that is, matching the maximum matched rule of proportion is<Play>【Movie name】；Enter
And incite somebody to action<Play>【Movie name】As the matched rule for the qi processing that disappears through overmatching density, the rule of the matching after the qi that disappears are obtained
Then.
On hit quantity disappear qi processing mode, include but is not limited to：The Qi Chu it is determined that each pending hit quantity disappears
The parameter hit quantity of the matched rule of reason, parameter are one in name entity, vocabulary and location parameter；
Matched rule using the maximum matching result of parameter hit quantity as the qi processing that disappears by hit quantity；
Wherein, the disappear matched rule of qi processing of each pending hit quantity is the matching of qi processing of disappearing by match width
Rule.
In order to make it easy to understand, illustrated so that speech text is " playing song wind " as an example.Speech text is matched
Afterwards, obtain<Play><Song>%s1,<Play><Song>, the two matched rules；Wherein, %s1 is location parameter；This two
Individual matched rule is all corresponding<Play><Song>Wind, this second mapping result.Obtained all matched rules are matched
Width disappear qi processing, when set threshold value be 2 bytes when, determine that the match width of the two matched rules is identical.Due to matching
The width matched rule after qi that disappears is more than one, then the matched rule after the qi that disappears to match width match weighted value and disappeared Qi Chu
Reason, determines that the weighted value of the two matched rules is identical.It is because the matching weighted value matched rule after qi that disappears is more than one, then right
The matching weighted value matched rule after qi that disappears match density and disappeared qi.Due to<Play><Song>In do not include location parameter,
The matched rule after qi that can not then disappear to matching weighted value match density and disappeared qi processing, and then matching is weighted in sequence
The value matched rule after qi that disappears carries out hit quantity and disappeared qi processing.It is determined that<Play><Song>%s1 parameter hit quantity is 3,
<Play><Song>Parameter hit quantity be 2, i.e. the maximum matched rule of parameter hit quantity is<Play><Song>%s1；
And then will<Play><Song>Matched rules of the %s1 as the qi processing that disappears by hit quantity, obtains the matching after the qi that disappears
Rule.
On name entity weights disappear qi processing mode, include but is not limited to：It is determined that each pending name entity power
Value disappear qi processing matched rule in name the weighted value of entity, the maximum matched rule of the weighted value that will name entity is as passing through
Cross the matched rule for naming entity weights to disappear qi processing；
Wherein, the disappear matched rule of qi processing of each pending name entity weights is to disappear qi processing by match width
Matched rule.
In order to make it easy to understand, illustrated so that speech text is " play and cause the youth " as an example.Speech text is matched
Afterwards, obtain<Play>【Movie name】,<Play>【Title of the song】, the two matched rules correspond to respectively<Play>【Movie name】,<Play
>【Title of the song】, the two second mapping results.Obtained all matched rules are carried out with match width to disappear qi processing, determine this two
The match width of individual matched rule is identical.The matched rule after qi that disappears due to match width is more than one, then match width is disappeared
Matched rule after qi carry out match weighted value disappear qi processing, determine that the weighted value of the two matched rules is identical.Due to matching
The weighted value matched rule after qi that disappears is more than one, then the matched rule after the qi that disappears to matching weighted value match density and disappeared qi
Processing, determines that the matching density of the two matched rules is identical.Because the matching density matched rule after qi that disappears is more than one, then
The matched rule after qi that disappears to matching density carries out hit quantity and disappeared qi processing, determines the parameter hits of the two matched rules
Measure identical.The matched rule after qi that disappears due to hit quantity is more than one, then the matched rule after the qi that disappears to hit quantity is carried out
Name entity weights disappear qi processing.When it is determined that name entity【Movie name】Weighted value be more than name entity【Title of the song】Weighted value
When, it is determined that<Play>【Movie name】The weighted value of middle name entity is more than<Play>【Title of the song】The weighted value of middle name entity, i.e.,
Naming the matched rule of the weighted value maximum of entity is<Play>【Movie name】；And then will<Play>【Movie name】As by ordering
Name entity weights disappear qi processing matched rule, obtain the matched rule after the qi that disappears.
207：Speech text is handled according to the matched rule after the obtained qi that disappears.
For the step, on the side handled according to the matched rule after the obtained qi that disappears speech text
Formula, the present embodiment are not especially limited, and specific processing mode is identical with the processing mode principle in above-mentioned steps 205, specific detailed
See above-mentioned steps 205.
The method that the present embodiment provides, mapped by being named entity to speech text, obtain the first mapping result；It is right
First mapping result carries out vocabulary mapping, obtains the second mapping result；Afterwards by the second mapping result with including regularity
Preset rules are matched, and speech text is handled according to an obtained matched rule, so as to by regularity and
Name the configuration format of entity rule unified, processing speech text technology is not only suitable for actual language environment, be applied to again
Fixed language environment, the scope of application of processing speech text technology is expanded, so that the processing mode of speech text obtains
Optimization.
Embodiment three
Referring to Fig. 3, the embodiments of the invention provide a kind of device for handling speech text, the device includes：
First mapping block 301, for being named entity mapping to speech text, obtain the first mapping result；
Second mapping block 302, for carrying out vocabulary mapping to the first mapping result, obtain the second mapping result；
Matching module 303, for the second mapping result to be matched with the preset rules including regularity；
First processing module 304, for when obtaining a matched rule, according to an obtained matched rule to voice
Text is handled.
As a kind of preferred embodiment, referring to Fig. 4, the device also includes：
Deploy module 305, for by the first mapping result one or more name entity successively expand into mapping before it is right
The speech text answered, obtain at least two the 3rd mapping results；
Second mapping block 302, it is additionally operable to the non-mapped voice text for name entity in each 3rd mapping result
This progress vocabulary mapping, obtains the second mapping result.
As a kind of preferred embodiment, referring to Fig. 5, the device also includes：
Disappear qi module 306, for when obtaining at least two matched rules, being carried out successively to obtained all matched rules
Match width disappear qi processing, matching weighted value disappear qi processing, matching density disappear qi processing, hit quantity disappear qi processing and name
At least one of the qi processing that disappears of entity weights disappears qi processing, until obtain the matched rule after the qi that disappears；
Second processing module 307, for being handled according to the matched rule after the obtained qi that disappears speech text.
As a kind of preferred embodiment, disappear qi module 306, for determining the second mapping result corresponding to each matched rule
Match width, the maximum matched rule of match width is used as and disappeared the matched rule that qi handles by match width.
As a kind of preferred embodiment, disappear qi module 306, for the weighting according to the vocabulary pre-set and name entity
Value determine each pending matching weighted value disappear qi processing matched rule weighted value；The maximum matched rule of weighted value is made
For the matched rule for the qi processing that disappears through overmatching weighted value；
Wherein, it is pending matching weighted value disappear qi processing matched rule be by match width disappear qi processing matching advise
Then.
As a kind of preferred embodiment, disappear qi module 306, for determine it is each it is pending matching density disappear qi processing
With rule and the matching proportion of the second mapping result, the maximum matched rule of proportion will be matched and disappeared Qi Chu as through overmatching density
The matched rule of reason；
Wherein, the disappear matched rule of qi processing of each pending matching density is the matching of qi processing of disappearing by match width
Rule.
As a kind of preferred embodiment, disappear qi module 306, for determine each pending hit quantity disappear qi processing
Parameter hit quantity with rule, parameter are one in name entity, vocabulary and location parameter；Parameter hit quantity is maximum
Matching result as by hit quantity disappear qi processing matched rule；
Wherein, the disappear matched rule of qi processing of each pending hit quantity is the matching of qi processing of disappearing by match width
Rule.
As a kind of preferred embodiment, disappear qi module 306, for determine it is each it is pending name entity weights disappear qi processing
Matched rule in name the weighted value of entity, the maximum matched rule of the weighted value that will name entity is used as by name entity
Weights disappear qi processing matched rule；
Wherein, the disappear matched rule of qi processing of each pending name entity weights is to disappear qi processing by match width
Matched rule.
In summary, device provided in an embodiment of the present invention, mapped by being named entity to speech text, obtain the
One mapping result；Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；Afterwards by the second mapping result and bag
The preset rules for including regularity are matched, and speech text is handled according to an obtained matched rule, so as to
The configuration format of regularity and name entity rule is unified, processing speech text technology is not only suitable for actual language ring
Border, it is applied to fixed language environment again, the scope of application of processing speech text technology is expanded, so that the place of speech text
Reason mode is optimized.
Example IV
The embodiments of the invention provide a kind of terminal, Fig. 6 is refer to, it illustrates the terminal involved by the embodiment of the present invention
Structural representation, the terminal can be used for implement above-described embodiment in provide processing speech text method.Specifically：
Terminal 600 can include RF (Radio Frequency, radio frequency) circuit 110, include one or more meters
The memory 120 of calculation machine readable storage medium storing program for executing, input block 130, display unit 140, sensor 150, voicefrequency circuit 160,
WiFi (Wireless Fidelity, Wireless Fidelity) module 170, include one or the processing of more than one processing core
The part such as device 180 and power supply 190.It will be understood by those skilled in the art that the terminal structure shown in Fig. 6 is not formed pair
The restriction of terminal, it can include than illustrating more or less parts, either combine some parts or different part cloth
Put.Wherein：
RF circuits 110 can be used for receive and send messages or communication process in, the reception and transmission of signal, especially, by base station
After downlink information receives, transfer to one or more than one processor 180 is handled；In addition, it is sent to up data are related to
Base station.Generally, RF circuits 110 include but is not limited to antenna, at least one amplifier, tuner, one or more oscillators, use
Family identity module (SIM) card, transceiver, coupler, LNA (Low Noise Amplifier, low-noise amplifier), duplex
Device etc..In addition, RF circuits 110 can also be communicated by radio communication with network and other equipment.The radio communication can make
With any communication standard or agreement, and including but not limited to GSM (Global System of Mobile communication, entirely
Ball mobile communcations system), GPRS (General Packet Radio Service, general packet radio service), CDMA (Code
Division Multiple Access, CDMA), WCDMA (Wideband Code Division Multiple
Access, WCDMA), LTE (Long Term Evolution, Long Term Evolution), Email, SMS (Short
Messaging Service, Short Message Service) etc..
Memory 120 can be used for storage software program and module, and processor 180 is stored in memory 120 by operation
Software program and module, so as to perform various function application and data processing.Memory 120 can mainly include storage journey
Sequence area and storage data field, wherein, storing program area can storage program area, the application program (ratio needed at least one function
Such as sound-playing function, image player function) etc.；Storage data field can store uses created number according to terminal 600
According to (such as voice data, phone directory etc.) etc..In addition, memory 120 can include high-speed random access memory, can also wrap
Include nonvolatile memory, a for example, at least disk memory, flush memory device or other volatile solid-state parts.
Correspondingly, memory 120 can also include Memory Controller, to provide processor 180 and input block 130 to memory
120 access.
Input block 130 can be used for the numeral or character information for receiving input, and generation is set with user and function
Control relevant keyboard, mouse, action bars, optics or the input of trace ball signal.Specifically, input block 130 may include to touch
Sensitive surfaces 131 and other input equipments 132.Touch sensitive surface 131, also referred to as touch display screen or Trackpad, collect and use
Family on or near it touch operation (such as user using any suitable object or annex such as finger, stylus in touch-sensitive table
Operation on face 131 or near touch sensitive surface 131), and corresponding attachment means are driven according to formula set in advance.It is optional
, touch sensitive surface 131 may include both touch detecting apparatus and touch controller.Wherein, touch detecting apparatus detection is used
The touch orientation at family, and the signal that touch operation is brought is detected, transmit a signal to touch controller；Touch controller is from touch
Touch information is received in detection means, and is converted into contact coordinate, then gives processor 180, and can reception processing device 180
The order sent simultaneously is performed.Furthermore, it is possible to using polytypes such as resistance-type, condenser type, infrared ray and surface acoustic waves
Realize touch sensitive surface 131.Except touch sensitive surface 131, input block 130 can also include other input equipments 132.Specifically,
Other input equipments 132 can include but is not limited to physical keyboard, function key (such as volume control button, switch key etc.),
One or more in trace ball, mouse, action bars etc..
Display unit 140 can be used for display by the information of user's input or be supplied to the information and terminal 600 of user
Various graphical user interface, these graphical user interface can be made up of figure, text, icon, video and its any combination.
Display unit 140 may include display panel 141, optionally, can use LCD (Liquid Crystal Display, liquid crystal
Show device), the form such as OLED (Organic Light-Emitting Diode, Organic Light Emitting Diode) configure display panel
141.Further, touch sensitive surface 131 can cover display panel 141, when touch sensitive surface 131 detects touching on or near it
After touching operation, processor 180 is sent to determine the type of touch event, is followed by subsequent processing type of the device 180 according to touch event
Corresponding visual output is provided on display panel 141.Although in figure 6, touch sensitive surface 131 and display panel 141 are conducts
Two independent parts come realize input and input function, but in some embodiments it is possible to by touch sensitive surface 131 with display
Panel 141 is integrated and realizes input and output function.
Terminal 600 may also include at least one sensor 150, such as optical sensor, motion sensor and other sensings
Device.Specifically, optical sensor may include ambient light sensor and proximity transducer, wherein, ambient light sensor can be according to environment
The light and shade of light adjusts the brightness of display panel 141, and proximity transducer can close display when terminal 600 is moved in one's ear
Panel 141 and/or backlight.As one kind of motion sensor, gravity accelerometer can detect in all directions (generally
Three axles) acceleration size, size and the direction of gravity are can detect that when static, available for identification mobile phone posture application (ratio
Such as horizontal/vertical screen switching, dependent game, magnetometer pose calibrating), Vibration identification correlation function (such as pedometer, tap)；Extremely
The other sensors such as the gyroscope that can also configure in terminal 600, barometer, hygrometer, thermometer, infrared ray sensor, herein
Repeat no more.
Voicefrequency circuit 160, loudspeaker 161, microphone 162 can provide the COBBAIF between user and terminal 600.Audio
Electric signal after the voice data received conversion can be transferred to loudspeaker 161, sound is converted to by loudspeaker 161 by circuit 160
Sound signal exports；On the other hand, the voice signal of collection is converted to electric signal by microphone 162, after being received by voicefrequency circuit 160
Voice data is converted to, then after voice data output processor 180 is handled, through RF circuits 110 to be sent to such as another end
End, or voice data is exported to memory 120 further to handle.Voicefrequency circuit 160 is also possible that earphone jack,
To provide the communication of peripheral hardware earphone and terminal 600.
WiFi belongs to short range wireless transmission technology, and terminal 600 can help user's transceiver electronicses by WiFi module 170
Mail, browse webpage and access streaming video etc., it has provided the user wireless broadband internet and accessed.Although Fig. 6 is shown
WiFi module 170, but it is understood that, it is simultaneously not belonging to must be configured into for terminal 600, can exist as needed completely
Do not change in the essential scope of invention and omit.
Processor 180 is the control centre of terminal 600, utilizes various interfaces and each portion of connection whole mobile phone
Point, by running or performing the software program and/or module that are stored in memory 120, and call and be stored in memory 120
Interior data, the various functions and processing data of terminal 600 are performed, so as to carry out integral monitoring to mobile phone.Optionally, processor
180 may include one or more processing cores；Preferably, processor 180 can integrate application processor and modem processor,
Wherein, application processor mainly handles operating system, user interface and application program etc., and modem processor mainly handles nothing
Line communicates.It is understood that above-mentioned modem processor can not also be integrated into processor 180.
Terminal 600 also includes the power supply 190 (such as battery) to all parts power supply, it is preferred that power supply can pass through electricity
Management system and processor 180 are logically contiguous, so as to realize management charging, electric discharge and power consumption by power-supply management system
The functions such as management.Power supply 190 can also include one or more direct current or AC power, recharging system, power supply event
The random component such as barrier detection circuit, power supply changeover device or inverter, power supply status indicator.
Although being not shown, terminal 600 can also include camera, bluetooth module etc., will not be repeated here.Specifically in this reality
To apply in example, the display unit of terminal is touch-screen display, and terminal also includes memory, and one or more than one
Program, one of them or more than one program storage are configured to by one or more than one processing in memory
Device is performed, and one or more than one program bag, which contains, to be used to perform the instruction operated below：
Entity mapping is named to speech text, obtains the first mapping result；
Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；
Second mapping result is matched with the preset rules including regularity, if obtaining a matched rule,
Then speech text is handled according to an obtained matched rule.
Assuming that above-mentioned is the first possible embodiment, then provided based on the first possible embodiment
Second of possible embodiment in, in the memory of terminal, also include and be used to perform the instruction that operates below：
Before vocabulary mapping being carried out to the first mapping result, including：
One or more name entity in first mapping result is expanded into corresponding speech text before mapping successively, obtained
At least two the 3rd mapping results；
Vocabulary mapping is carried out to the first mapping result, including：
The non-mapped speech text for name entity in each 3rd mapping result is subjected to vocabulary mapping, obtains second
Mapping result.
In the third the possible embodiment provided based on the first or second of possible embodiment,
In the memory of terminal, also include and be used to perform the instruction operated below：
After second mapping result is matched with the preset rules including regularity, in addition to：
If obtaining at least two matched rules, obtained all matched rules are carried out with match width successively and is disappeared Qi Chu
Reason, matching weighted value disappear qi processing, the hit quantity of qi processing, matching density that disappear disappear qi processing and name entity weights disappear Qi Chu
At least one of reason disappears qi processing, until obtain the matched rule after the qi that disappears；
Speech text is handled according to the matched rule after the obtained qi that disappears.
In the 4th kind of possible embodiment provided based on the third possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Match width disappear qi processing, including：
It is determined that the match width of the second mapping result corresponding to each matched rule, by the matched rule that match width is maximum
Matched rule as the qi processing that disappears by match width.
In the 5th kind of possible embodiment provided based on the 4th kind of possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Matching weighted value disappear qi processing, including：
According to the vocabulary pre-set with naming the weighted value of entity to determine that each pending weighted value qi that disappears that match is handled
Matched rule weighted value；
Matched rule using the maximum matched rule of weighted value as the qi processing that disappears through overmatching weighted value；
Wherein, it is pending matching weighted value disappear qi processing matched rule be by match width disappear qi processing matching advise
Then.
In the 6th kind of possible embodiment provided based on the 5th kind of possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Matching density disappear qi processing, including：
The matched rule of qi processing and the matching proportion of the second mapping result it is determined that each pending matching density disappears, general
Matched rule with the maximum matched rule of proportion as the qi processing that disappears through overmatching density；
Wherein, the disappear matched rule of qi processing of each pending matching density is the matching of qi processing of disappearing by match width
Rule.
In the 7th kind of possible embodiment provided based on the 6th kind of possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Hit quantity disappear qi processing, including：
The parameter hit quantity of the matched rule of qi processing it is determined that each pending hit quantity disappears, parameter are real for name
One in body, vocabulary and location parameter；
Matched rule using the maximum matching result of parameter hit quantity as the qi processing that disappears by hit quantity；
Wherein, the disappear matched rule of qi processing of each pending hit quantity is the matching of qi processing of disappearing by match width
Rule.
In the 8th kind of possible embodiment provided based on the 7th kind of possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Name entity weights disappear qi processing, including：
The weighted value of entity is named in the matched rule of qi processing it is determined that each pending name entity weights disappear, will be named
Matched rule of the maximum matched rule of the weighted value of entity as the qi processing that disappears by name entity weights；
Wherein, the disappear matched rule of qi processing of each pending name entity weights is to disappear qi processing by match width
Matched rule.
In summary, terminal provided in an embodiment of the present invention, mapped by being named entity to speech text, obtain the
One mapping result；Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；Afterwards by the second mapping result and bag
The preset rules for including regularity are matched, and speech text is handled according to an obtained matched rule, so as to
The configuration format of regularity and name entity rule is unified, processing speech text technology is not only suitable for actual language ring
Border, it is applied to fixed language environment again, the scope of application of processing speech text technology is expanded, so that the place of speech text
Reason mode is optimized.
Embodiment five
The embodiment of the present invention additionally provides a kind of computer-readable recording medium, and the computer-readable recording medium can be
The computer-readable recording medium included in memory in above-described embodiment；Can also be individualism, without supplying eventually
Computer-readable recording medium in end.The computer-readable recording medium storage has one or more than one program, and this one
The method that individual either more than one program is used for performing a processing speech text by one or more than one processor, should
Method includes：
Entity mapping is named to speech text, obtains the first mapping result；
Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；
Second mapping result is matched with the preset rules including regularity, if obtaining a matched rule,
Then speech text is handled according to an obtained matched rule.
Assuming that above-mentioned is the first possible embodiment, then provided based on the first possible embodiment
Second of possible embodiment in, in the memory of terminal, also include and be used to perform the instruction that operates below：
Before vocabulary mapping being carried out to the first mapping result, including：
One or more name entity in first mapping result is expanded into corresponding speech text before mapping successively, obtained
At least two the 3rd mapping results；
Vocabulary mapping is carried out to the first mapping result, including：
The non-mapped speech text for name entity in each 3rd mapping result is subjected to vocabulary mapping, obtains second
Mapping result.
In the third the possible embodiment provided based on the first or second of possible embodiment,
In the memory of terminal, also include and be used to perform the instruction operated below：
After second mapping result is matched with the preset rules including regularity, in addition to：
If obtaining at least two matched rules, obtained all matched rules are carried out with match width successively and is disappeared Qi Chu
Reason, matching weighted value disappear qi processing, the hit quantity of qi processing, matching density that disappear disappear qi processing and name entity weights disappear Qi Chu
At least one of reason disappears qi processing, until obtain the matched rule after the qi that disappears；
Speech text is handled according to the matched rule after the obtained qi that disappears.
In the 4th kind of possible embodiment provided based on the third possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Match width disappear qi processing, including：
It is determined that the match width of the second mapping result corresponding to each matched rule, by the matched rule that match width is maximum
Matched rule as the qi processing that disappears by match width.
In the 5th kind of possible embodiment provided based on the 4th kind of possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Matching weighted value disappear qi processing, including：
According to the vocabulary pre-set with naming the weighted value of entity to determine that each pending weighted value qi that disappears that match is handled
Matched rule weighted value；
Matched rule using the maximum matched rule of weighted value as the qi processing that disappears through overmatching weighted value；
Wherein, it is pending matching weighted value disappear qi processing matched rule be by match width disappear qi processing matching advise
Then.
In the 6th kind of possible embodiment provided based on the 5th kind of possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Matching density disappear qi processing, including：
The matched rule of qi processing and the matching proportion of the second mapping result it is determined that each pending matching density disappears, general
Matched rule with the maximum matched rule of proportion as the qi processing that disappears through overmatching density；
Wherein, the disappear matched rule of qi processing of each pending matching density is the matching of qi processing of disappearing by match width
Rule.
In the 7th kind of possible embodiment provided based on the 6th kind of possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Hit quantity disappear qi processing, including：
The parameter hit quantity of the matched rule of qi processing it is determined that each pending hit quantity disappears, parameter are real for name
One in body, vocabulary and location parameter；
Matched rule using the maximum matching result of parameter hit quantity as the qi processing that disappears by hit quantity；
Wherein, the disappear matched rule of qi processing of each pending hit quantity is the matching of qi processing of disappearing by match width
Rule.
In the 8th kind of possible embodiment provided based on the 7th kind of possible embodiment, terminal is deposited
In reservoir, also include and be used to perform the instruction operated below：
Name entity weights disappear qi processing, including：
The weighted value of entity is named in the matched rule of qi processing it is determined that each pending name entity weights disappear, will be named
Matched rule of the maximum matched rule of the weighted value of entity as the qi processing that disappears by name entity weights；
Wherein, the disappear matched rule of qi processing of each pending name entity weights is to disappear qi processing by match width
Matched rule.
In summary, computer-readable recording medium provided in an embodiment of the present invention, by being named to speech text
Entity maps, and obtains the first mapping result；Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；Afterwards will
Second mapping result is matched with the preset rules including regularity, and according to an obtained matched rule to voice text
This is handled, thus the configuration format of regularity and name entity rule is unified, processing speech text technology was both fitted
For the language environment of reality, it is applied to fixed language environment again, expands the scope of application of processing speech text technology, from
And optimized the processing mode of speech text.
Embodiment six
A kind of graphical user interface is provided in the embodiment of the present invention, the graphical user interface is with terminal, the terminal
Including touch-screen display, memory and one or more than one place for performing an either more than one program
Manage device；The graphical user interface includes：
Entity mapping is named to speech text, obtains the first mapping result；
Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；
Second mapping result is matched with the preset rules including regularity, if obtaining a matched rule,
Then speech text is handled according to an obtained matched rule.
In summary, graphical user interface provided in an embodiment of the present invention is reflected by being named entity to speech text
Penetrate, obtain the first mapping result；Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result；Second is reflected afterwards
Penetrate result to be matched with the preset rules including regularity, and speech text is carried out according to an obtained matched rule
Processing, thus the configuration format of regularity and name entity rule is unified, processing speech text technology is not only suitable for reality
The language environment on border, it is applied to fixed language environment again, the scope of application of processing speech text technology is expanded, so that language
The processing mode of sound text is optimized.
It should be noted that：Above-described embodiment provide processing speech text device when handling speech text, only with
The division progress of above-mentioned each functional module, can be as needed and by above-mentioned function distribution by not for example, in practical application
Same functional module is completed, i.e., the internal structure of device is divided into different functional modules, to complete whole described above
Or partial function.In addition, the method for the device and processing speech text for the processing speech text that above-described embodiment provides is implemented
Example belongs to same design, and its specific implementation process refers to embodiment of the method, repeats no more here.
The embodiments of the present invention are for illustration only, do not represent the quality of embodiment.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment
To complete, by program the hardware of correlation can also be instructed to complete, described program can be stored in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
The foregoing is only presently preferred embodiments of the present invention, be not intended to limit the invention, it is all the present invention spirit and
Within principle, any modification, equivalent substitution and improvements made etc., it should be included in the scope of the protection.
Claims (16)
- A kind of 1. method for handling speech text, it is characterised in that methods described includes：Entity mapping is named to speech text, obtains the first mapping result；One or more name entity in first mapping result is expanded into corresponding speech text before mapping successively, obtained At least two the 3rd mapping results；The non-mapped speech text for name entity in each 3rd mapping result is subjected to vocabulary mapping, obtains the second mapping As a result；Second mapping result is matched with the preset rules including regularity, if obtaining a matched rule, Then the speech text is handled according to an obtained matched rule.
- 2. according to the method for claim 1, it is characterised in that it is described will second mapping result with include regularity Preset rules matched after, in addition to：If obtaining at least two matched rules, obtained all matched rules are carried out successively match width disappear qi processing, Matching weighted value disappear qi processing, matching density disappear qi processing, hit quantity disappear qi processing and name entity weights disappear qi processing At least one of the qi processing that disappears, until obtain the matched rule after the qi that disappears；The speech text is handled according to the matched rule after the obtained qi that disappears.
- 3. according to the method for claim 2, it is characterised in that the match width disappear qi processing, including：It is determined that the match width of the second mapping result corresponding to each matched rule, using the maximum matched rule of match width as By match width disappear qi processing matched rule.
- 4. according to the method for claim 3, it is characterised in that it is described matching weighted value disappear qi processing, including：According to the vocabulary pre-set with naming the weighted value of entity to determine that each pending weighted value that match disappears that qi handles Weighted value with rule；Matched rule using the maximum matched rule of weighted value as the qi processing that disappears through overmatching weighted value；Wherein, it is described it is pending matching weighted value disappear qi processing matched rule be by match width disappear qi processing matching advise Then.
- 5. according to the method for claim 4, it is characterised in that it is described matching density disappear qi processing, including：The matched rule of qi processing and the matching proportion of the second mapping result it is determined that each pending matching density disappears, will match ratio Matched rule of the maximum matched rule of weight as the qi processing that disappears through overmatching density；Wherein, the disappear matched rule of qi processing of each pending matching density is to be advised by the disappear matching of qi processing of match width Then.
- 6. according to the method for claim 5, it is characterised in that the hit quantity disappear qi processing, including：The parameter hit quantity of the matched rule of qi processing it is determined that each pending hit quantity disappears, the parameter are real for name One in body, vocabulary and location parameter；Matched rule using the maximum matching result of parameter hit quantity as the qi processing that disappears by hit quantity；Wherein, the disappear matched rule of qi processing of each pending hit quantity is to be advised by the disappear matching of qi processing of match width Then.
- 7. according to the method for claim 6, it is characterised in that it is described name entity weights disappear qi processing, including：The weighted value of entity is named in the matched rule of qi processing it is determined that each pending name entity weights disappear, entity will be named The maximum matched rule of weighted value be used as and disappear the matched rule that qi handles by name entity weights；Wherein, the disappear matched rule of qi processing of each pending name entity weights is the matching of qi processing of disappearing by match width Rule.
- 8. a kind of device for handling speech text, it is characterised in that described device includes：First mapping block, for being named entity mapping to speech text, obtain the first mapping result；Deploy module, for by first mapping result one or more name entity successively expand into mapping before corresponding to Speech text, obtain at least two the 3rd mapping results；Second mapping block, it is additionally operable to the non-mapped speech text for name entity in each 3rd mapping result carrying out word Film festival is penetrated, and obtains the second mapping result；Matching module, for second mapping result to be matched with the preset rules including regularity；First processing module, for when obtaining a matched rule, according to an obtained matched rule to voice text This is handled.
- 9. device according to claim 8, it is characterised in that described device, in addition to：Disappear qi module, for when obtaining at least two matched rules, it is wide to carry out matching successively to obtained all matched rules Degree disappear qi processing, matching weighted value disappear qi processing, matching density disappear qi processing, hit quantity disappear qi processing and name entity power Value at least one of the qi processing that disappears disappears qi processing, until obtain the matched rule after the qi that disappears；Second processing module, for being handled according to the matched rule after the obtained qi that disappears the speech text.
- 10. device according to claim 9, it is characterised in that the qi module that disappears, for determining each matched rule pair The match width for the second mapping result answered, disappear what qi was handled using the maximum matched rule of match width as by match width Matched rule.
- 11. device according to claim 10, it is characterised in that the qi module that disappears, for according to the word pre-set Converge and determine that each pending weighted value that matches disappears the weighted value of matched rule that qi handles with the weighted value of name entity；Will weighting Matched rule of the maximum matched rule of value as the qi processing that disappears through overmatching weighted value；Wherein, it is described it is pending matching weighted value disappear qi processing matched rule be by match width disappear qi processing matching advise Then.
- 12. device according to claim 11, it is characterised in that the qi module that disappears, for determining each pending The matching proportion for matched rule and the second mapping result of qi processing that disappear with density, will match the matched rule of proportion maximum as Through overmatching density disappear qi processing matched rule；Wherein, the disappear matched rule of qi processing of each pending matching density is to be advised by the disappear matching of qi processing of match width Then.
- 13. device according to claim 12, it is characterised in that the qi module that disappears, for determining each pending life Middle quantity disappear qi processing matched rule parameter hit quantity, the parameter is in name entity, vocabulary and location parameter One；Matched rule using the maximum matching result of parameter hit quantity as the qi processing that disappears by hit quantity；Wherein, the disappear matched rule of qi processing of each pending hit quantity is to be advised by the disappear matching of qi processing of match width Then.
- 14. device according to claim 13, it is characterised in that the qi module that disappears, for determining each pending life Name entity weights disappear qi processing matched rule in name the weighted value of entity, the maximum matching rule of the weighted value that entity will be named The then matched rule as the qi processing that disappears by name entity weights；Wherein, the disappear matched rule of qi processing of each pending name entity weights is the matching of qi processing of disappearing by match width Rule.
- 15. a kind of terminal, it is characterised in that the terminal includes processor and memory, and one is stored with the memory Or more than one program, one or more than one program are loaded by the processor and performed to realize as weighed The method that profit requires 1 to 7 any described processing speech text.
- A kind of 16. computer-readable recording medium, it is characterised in that be stored with the storage medium one or more than one Program, one or more than one program loaded by processor and performed with realize such as any institute of claim 1 to 7 The method for the processing speech text stated.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310554808.XA CN104636323B (en) | 2013-11-07 | 2013-11-07 | Handle the method and device of speech text |
| PCT/CN2014/088371 WO2015067116A1 (en) | 2013-11-07 | 2014-10-11 | Method and apparatus for processing speech texts |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310554808.XA CN104636323B (en) | 2013-11-07 | 2013-11-07 | Handle the method and device of speech text |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104636323A CN104636323A (en) | 2015-05-20 |
| CN104636323B true CN104636323B (en) | 2018-04-03 |
Family
ID=53040884
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310554808.XA Active CN104636323B (en) | 2013-11-07 | 2013-11-07 | Handle the method and device of speech text |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN104636323B (en) |
| WO (1) | WO2015067116A1 (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109885823A (en) * | 2017-12-01 | 2019-06-14 | 武汉楚鼎信息技术有限公司 | A kind of distributed semantic recognition methods of financial industry and system and device |
| US11232262B2 (en) | 2018-07-17 | 2022-01-25 | iT SpeeX LLC | Method, system, and computer program product for an intelligent industrial assistant |
| CN111079435B (en) * | 2019-12-09 | 2021-04-06 | 深圳追一科技有限公司 | Named entity disambiguation method, device, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101454750A (en) * | 2006-03-31 | 2009-06-10 | 谷歌公司 | Disambiguation of named entities |
| WO2009086312A1 (en) * | 2007-12-21 | 2009-07-09 | Kondadadi, Ravi, Kumar | Entity, event, and relationship extraction |
| CN101510221A (en) * | 2009-02-17 | 2009-08-19 | 北京大学 | Enquiry statement analytical method and system for information retrieval |
| CN102654866A (en) * | 2011-03-02 | 2012-09-05 | 北京百度网讯科技有限公司 | Method and device for establishing example sentence index and method and device for indexing example sentences |
| CN102831892A (en) * | 2012-09-07 | 2012-12-19 | 深圳市信利康电子有限公司 | Toy control method and system based on internet voice interaction |
| CN103021403A (en) * | 2012-12-31 | 2013-04-03 | 威盛电子股份有限公司 | Voice recognition based selecting method and mobile terminal device and information system thereof |
| JP5776539B2 (en) * | 2011-12-26 | 2015-09-09 | 富士通株式会社 | Extraction apparatus, extraction program, and extraction method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10198431B2 (en) * | 2010-09-28 | 2019-02-05 | Siemens Corporation | Information relation generation |
-
2013
- 2013-11-07 CN CN201310554808.XA patent/CN104636323B/en active Active
-
2014
- 2014-10-11 WO PCT/CN2014/088371 patent/WO2015067116A1/en active Application Filing
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101454750A (en) * | 2006-03-31 | 2009-06-10 | 谷歌公司 | Disambiguation of named entities |
| WO2009086312A1 (en) * | 2007-12-21 | 2009-07-09 | Kondadadi, Ravi, Kumar | Entity, event, and relationship extraction |
| CN101510221A (en) * | 2009-02-17 | 2009-08-19 | 北京大学 | Enquiry statement analytical method and system for information retrieval |
| CN102654866A (en) * | 2011-03-02 | 2012-09-05 | 北京百度网讯科技有限公司 | Method and device for establishing example sentence index and method and device for indexing example sentences |
| JP5776539B2 (en) * | 2011-12-26 | 2015-09-09 | 富士通株式会社 | Extraction apparatus, extraction program, and extraction method |
| CN102831892A (en) * | 2012-09-07 | 2012-12-19 | 深圳市信利康电子有限公司 | Toy control method and system based on internet voice interaction |
| CN103021403A (en) * | 2012-12-31 | 2013-04-03 | 威盛电子股份有限公司 | Voice recognition based selecting method and mobile terminal device and information system thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104636323A (en) | 2015-05-20 |
| WO2015067116A1 (en) | 2015-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103309673B (en) | A kind of conversation processing method based on gesture, device | |
| CN105788612B (en) | A kind of method and apparatus detecting sound quality | |
| CN103702297B (en) | Short message enhancement, apparatus and system | |
| CN105828145B (en) | Interactive approach and device | |
| CN107438200A (en) | The method and apparatus of direct broadcasting room present displaying | |
| CN104519404B (en) | The player method and device of graphic interchange format file | |
| CN104252341B (en) | The client device of the skin change method of application program, device and application program | |
| CN106126174B (en) | A kind of control method and electronic equipment of scene audio | |
| CN104519262B (en) | Obtain the method, apparatus and terminal of video data | |
| CN107273036A (en) | Mobile terminal and its split screen control method, computer-readable recording medium | |
| CN103745133B (en) | The method and terminal of a kind of information processing | |
| CN104618223B (en) | A kind of management method of information recommendation, device and system | |
| CN106488296B (en) | A kind of method and apparatus showing video barrage | |
| CN107168609A (en) | status bar display method, device and terminal | |
| CN103886198B (en) | Method, terminal, server and the system that a kind of data process | |
| CN107370670A (en) | Unread message extracts methods of exhibiting and device | |
| CN107566255A (en) | Unread message abstraction generating method and device | |
| CN104516624B (en) | A kind of method and device inputting account information | |
| CN106775267A (en) | A kind of method and apparatus of the function menu for showing game plug-in unit | |
| CN107277630A (en) | The display methods and device of information of voice prompt | |
| CN104699501B (en) | A kind of method and device for running application program | |
| CN106231575B (en) | Method for processing short messages and device | |
| CN106797336A (en) | The method and apparatus of history chat record displaying | |
| CN107436948A (en) | File search method, device and terminal | |
| CN107908770A (en) | A kind of photo searching method and mobile terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |