CN104636323A - Method and device for processing voice texts - Google Patents
Method and device for processing voice texts Download PDFInfo
- Publication number
- CN104636323A CN104636323A CN201310554808.XA CN201310554808A CN104636323A CN 104636323 A CN104636323 A CN 104636323A CN 201310554808 A CN201310554808 A CN 201310554808A CN 104636323 A CN104636323 A CN 104636323A
- Authority
- CN
- China
- Prior art keywords
- matched rule
- disappear
- mapping
- disappears
- named entity
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—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 method and device for processing voice texts, and belongs to the technical field of information processing. The method comprises the steps that named entity mapping is carried out on the voice texts to obtain a first mapping result; lexical mapping is carried out on the first mapping result to obtain a second mapping result; the second mapping result is matched with a preset rule comprising a regular rule, and if a matching rule is obtained, the voice texts are processed according to the obtained matching rule. According to the method and device, named entity mapping is carried out on the voice texts to obtain the first mapping result; lexical mapping is carried out on the first mapping result to obtain the second mapping result; the second mapping result is matched with the preset rule comprising the regular rule, and the voice texts are processed according to the obtained matching rule, so that the configuration format of the regular rule and the configuration format of a named entity rule are uniformed, the application range of the voice text processing technology is expanded, and the processing mode of the voice texts is optimized.
Description
Technical field
The present invention relates to technical field of information processing, particularly a kind of method of processed voice text and device.
Background technology
Along with the development of the information processing technology, carry out man-machine interaction with natural language and become a reality.The key realizing man-machine interaction is the natural language instruction of wanting accurate understanding user to send and operates accordingly.User is after sending natural language instruction, and this instruction is converted into speech text, how processed voice text, becomes the problem that people pay close attention to.
Prior art has the mode of two kinds of processed voice texts, first kind of way: carry out vocabulary mapping to speech text, obtain mapping result; Extract the location parameter in mapping result by regular screens, obtain screens and extract result; Result is extracted in screens and preset rules is mated, obtain a matched rule; A matched rule according to obtaining processes speech text.Wherein, regular screens is the assigned address obtained according to regularity.The second way: carry out named entity mapping to speech text, obtains the first mapping result; Vocabulary mapping is carried out to the result that named entity maps, obtains the second mapping result; Second mapping result and preset rules are mated, obtains a matched rule; A matched rule according to obtaining processes speech text.
Realizing in process of the present invention, inventor finds that prior art at least exists following problem:
Regularity underaction, is not suitable for actual language environment.Named entity rule is for relatively more fixing language environment and be not suitable for the scene setting up entity name storehouse, can not play one's part to the full.Therefore, the mode of two kinds of processed voice texts that prior art provides all has certain limitation, causes the processing mode of speech text to be optimized not.
Summary of the invention
In order to solve the problem of prior art, embodiments provide a kind of method and device of processed voice text.Described technical scheme is as follows:
On the one hand, provide a kind of method of processed voice text, described method comprises:
Named entity mapping is carried out to speech text, obtains the first mapping result;
Vocabulary mapping is carried out to described first mapping result, obtains the second mapping result;
Described second mapping result is mated with the preset rules comprising regularity, if obtain a matched rule, then according to the matched rule obtained, described speech text is processed.
On the other hand, provide a kind of device of processed voice text, described device comprises:
First mapping block, for carrying out named entity mapping to speech text, obtains the first mapping result;
Second mapping block, for carrying out vocabulary mapping to described first mapping result, obtains the second mapping result;
Matching module, for mating described second mapping result with the preset rules comprising regularity;
First processing module, for when obtaining a matched rule, a matched rule according to obtaining processes described speech text.
The beneficial effect that the technical scheme that the embodiment of the present invention provides is brought is:
By carrying out named entity mapping to speech text, obtain the first mapping result; Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result; Afterwards the second mapping result is mated with the preset rules comprising regularity, and according to the matched rule obtained, speech text is processed, thus by unified for the configuration format of regularity and named entity rule, processed voice text techniques is made both to be applicable to actual language environment, be applicable to again the language environment of fixing, expand the scope of application of processed voice text techniques, thus make the processing mode of speech text obtain optimization.
Accompanying drawing explanation
In order to be illustrated more clearly in the technical scheme in the embodiment of the present invention, below the accompanying drawing used required in describing embodiment is briefly described, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to these accompanying drawings.
Fig. 1 is the method flow diagram of the processed voice text that the embodiment of the present invention one provides;
Fig. 2 is the method flow diagram of the processed voice text that the embodiment of the present invention two provides;
Fig. 3 is the apparatus structure schematic diagram of the first processed voice text that the embodiment of the present invention three provides;
Fig. 4 is the apparatus structure schematic diagram of the second processed voice text that the embodiment of the present invention three provides;
Fig. 5 is the apparatus structure schematic diagram of the third processed voice text that the embodiment of the present invention three provides;
Fig. 6 is the terminal structure schematic diagram of the processed voice text that the embodiment of the present invention four provides.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, embodiment of the present invention is described further in detail.
Embodiment one
Embodiments provide a kind of method of processed voice text, see Fig. 1, method flow comprises the steps:
101: named entity mapping is carried out to speech text, obtain the first mapping result;
102: vocabulary mapping is carried out to the first mapping result, obtain the second mapping result;
Further, before vocabulary mapping is carried out to the first mapping result, comprising:
One or more named entity in first mapping result is expanded into successively speech text corresponding before mapping, obtain at least two the 3rd mapping result;
Vocabulary mapping is carried out to the first mapping result, comprising:
The speech text not being mapped as named entity in each 3rd mapping result is carried out vocabulary mapping, obtains the second mapping result.
103: the second mapping result is mated with the preset rules comprising regularity, if obtain a matched rule, then according to the matched rule obtained, speech text is processed.
Further, after the second mapping result is mated with the preset rules comprising regularity, also comprise:
If obtain at least two matched rules, then carry out match width qi process, coupling weighted value qi process, coupling density disappear qi process and the named entity weights at least one disappeared in qi process of qi process, hit quantity that disappear that disappear that disappear successively to all matched rules obtained to disappear qi process, until obtain the matched rule after the qi that disappears;
According to the matched rule after the qi that disappears obtained, speech text is processed.
Further, match width disappears qi process, comprising:
Determine the match width of the second mapping result that each matched rule is corresponding, using the matched rule of matched rule maximum for match width as the qi process that disappears through match width.
Further, coupling weighted value disappears qi process, comprising:
Determine that each pending weighted value of mating disappears the weighted value of matched rule of qi process according to the vocabulary pre-set and the weighted value of named entity;
Using the matched rule of matched rule maximum for weighted value as the qi process that disappears through overmatching weighted value;
Wherein, the disappear matched rule of qi process of pending coupling weighted value is the matched rule of qi process of disappearing through match width.
Further, coupling density disappears qi process, comprising:
Determine each pending coupling density disappear qi process matched rule and the second mapping result mate proportion, by the matched rule of matched rule maximum for coupling proportion as the qi process that disappears through overmatching density;
Wherein, the disappear matched rule of qi process of each pending coupling density is the matched rule of qi process of disappearing through match width.
Further, hit quantity disappears qi process, comprising:
Determine each pending hit quantity disappear qi process matched rule parameter hit quantity, parameter is one in named entity, vocabulary and location parameter;
Parameter is hit the matched rule of the maximum matching result of quantity as the qi process that disappears through hit quantity;
Wherein, the disappear matched rule of qi process of each pending hit quantity is the matched rule of qi process of disappearing through match width.
Further, named entity weights disappear qi process, comprising:
Determine each pending named entity weights disappear qi process matched rule in the weighted value of named entity, using the matched rule of matched rule maximum for the weighted value of named entity as the qi process that disappears through named entity weights;
Wherein, the disappear matched rule of qi process of each pending named entity weights is the matched rule of qi process of disappearing through match width.
The method that the present embodiment provides, by carrying out named entity mapping to speech text, obtains the first mapping result; Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result; Afterwards the second mapping result is mated with the preset rules comprising regularity, and according to the matched rule obtained, speech text is processed, thus by unified for the configuration format of regularity and named entity rule, processed voice text techniques is made both to be applicable to actual language environment, be applicable to again the language environment of fixing, expand the scope of application of processed voice text techniques, thus make the processing mode of speech text obtain optimization.
Embodiment two
Embodiments provide a kind of method of processed voice text, in conjunction with the content of above-described embodiment one, see Fig. 2, method flow comprises:
201: named entity mapping is carried out to speech text, obtain the first mapping result;
Particularly, named entity mapping is carried out to speech text, includes but not limited to: set up named entity storehouse; In speech text, search the voice of the named entity that can be identified as in named entity storehouse herein, and replace the speech text found with named entity.It should be noted that, named entity collects in a large amount of from network information, and the identification of named entity is the implementation using every field independence dictionary tree, finds out all named entities when namely can support that named entity all overlaps or partially overlaps.
For the ease of understanding, with speech text for " I wants to listen first time ", named entity is expressed as example with Chinese braces [] and is described.Speech text " I thinks " and " for the first time " all corresponding named entity [title of the song] in the named entity set up; Find in speech text [title of the song] that speech text " I thinks " and " for the first time " can both be identified as in named entity storehouse, obtain the first mapping result after replacing with named entity [title of the song] speech text " I thinks " and " for the first time " found, namely [title of the song] listens [title of the song].
202: one or more named entity in the first mapping result is expanded into successively speech text corresponding before mapping, obtain at least two the 3rd mapping result;
For this step, because named entity may be overlapping with vocabulary, then one or more named entity in the first mapping result is expanded into successively speech text corresponding before mapping, obtain at least two the 3rd mapping result, thus add mapping result, can not accurate processed voice text when avoiding named entity and vocabulary overlap.
It should be noted that, if there is no the partly overlapping situation of named entity in the first mapping result, and a total n named entity, then obtain 2^n the 3rd mapping result in the first mapping result after one or more named entity being expanded into successively speech text corresponding before mapping.
For the ease of understanding, be still described for " I want listen first time " for speech text.Wherein, the first mapping result that speech text is corresponding is listened [title of the song] for [title of the song].Speech text corresponding before [title of the song] being listened in [title of the song] two named entities to expand into successively to map, obtains that [title of the song] listens [title of the song], I want to listen [title of the song], [title of the song] listens first time, I want to listen first time, these four the 3rd mapping result.
203: the speech text not being mapped as named entity in each 3rd mapping result is carried out vocabulary mapping, obtains the second mapping result;
For this step, the vocabulary carried out in vocabulary mapping is human configuration, and the speech text of expressing identical meanings in ordinary language is mapped as identical vocabulary.Due to the speech text speech text that possibility is corresponding different in the language environment of reality of identical meanings, the speech text not being mapped as named entity is carried out vocabulary mapping, the workload of repeated matching can be reduced.About the mode of the speech text not being mapped as named entity in each 3rd mapping result being carried out vocabulary mapping, the present embodiment does not do concrete restriction.
For the ease of understanding, be described for the 3rd mapping result in above-mentioned steps 202.Wherein, " I wants to listen " is mapped to vocabulary < and plays >.The speech text not being mapped as named entity in each 3rd mapping result is carried out vocabulary mapping, obtaining the second mapping result is: [title of the song] listens [title of the song], < plays >[title of the song], [title of the song] listens first time, and < plays > first time.
204: the second mapping result is mated with the preset rules comprising regularity;
Particularly, preset rules includes but not limited to: regularity and other default rules.Wherein, other rules preset include but not limited to arrange the rule meeting speech habits.The present embodiment does not do concrete restriction to other rules preset, and can arrange other default rules in practical application as required, the present embodiment does not do concrete restriction to this.
Further, the second mapping result is mated with the preset rules comprising regularity, includes but not limited to: extract the location parameter in the second mapping result by regular screens, obtain screens and extract result; Screens is extracted result to mate with other rules preset.Wherein, regular screens is the assigned address obtained according to regularity; Extract the location parameter in the second mapping result by regular screens, namely to extract the location parameter in the second mapping result according to regularity at assigned address.When there is no the location parameter that can be extracted by regular screens in the second mapping result, directly the second mapping result and other default rules can be mated, and then obtaining the rule of coupling.
It should be noted that, after the second mapping result being mated with the preset rules comprising regularity, if obtain a matched rule, then perform step 205; If obtain at least two matched rules, then perform step 206.
For the ease of understanding, with the second mapping result: [title of the song] listens [title of the song], < plays >[title of the song], [title of the song] listens first time, and < plays > first time, for example is described.Owing to there is no the location parameter that can be extracted by regular screens in the second mapping result, then direct by the second mapping result and preset other rules mate.When other rules preset for < plays >[title of the song] time, obtaining second mapping result corresponding with this rule is that < plays >[title of the song], and then obtain a matched rule, namely < plays >[title of the song].
Again such as, speech text is " broadcasting Corner With Love ", then will process according to above-mentioned steps 201 to 203 pairs of speech texts, obtain < and play >[video name], < plays >[title of the song] run into love, < plays > Corner With Love, these three second mapping result.Owing to there is no the location parameter that can be extracted by regular screens in these three second mapping result, then direct by these three second mapping result and preset other rules mate.Because other rules preset are both for < plays >[title of the song], >[video name can also be play] for <, then obtaining corresponding second mapping result regular with these two is that < plays >[video name], < plays >[title of the song] run into love; And then obtain two matched rules, namely < plays >[title of the song], < plays >[video name].
205: a matched rule according to obtaining processes speech text;
For this step, owing to obtaining a matched rule in above-mentioned steps 204, then according to the matched rule obtained, speech text is processed.About the mode processed speech text, the present embodiment does not do concrete restriction.
Such as, still " broadcasting Corner With Love " for speech text, if obtain matched rule is that < plays >[video name], then this step 205 is when processing speech text according to the matched rule obtained, and will play the video of video name for " Corner With Love ".
206: match width qi process, coupling weighted value qi process, coupling density disappear qi process and the named entity weights at least one disappeared in qi process of qi process, hit quantity that disappear that disappear that disappear is carried out successively to all matched rules obtained and to disappear qi process, until obtain the matched rule after the qi that disappears;
For this step, owing to obtaining at least two matched rules in above-mentioned steps 204, in order to make the result of speech text more accurate, the method that the present embodiment provides takes the mode of the qi process that to disappear to all matched rules.Wherein, disappear the process of qi process, include but not limited to: match width qi process, coupling weighted value qi process, coupling density disappear qi process and the named entity weights at least one disappeared in qi process of qi process, hit quantity that disappear that disappear that disappear is carried out successively to all matched rules obtained and to disappear qi process, until obtain the matched rule after the qi that disappears.
Disappear about match width the mode of qi process, includes but not limited to: the match width determining the second mapping result that each matched rule is corresponding, using the matched rule of matched rule maximum for match width as the qi process that disappears through match width.The determination of match width is the end position of reference position to last parameter of first parameter from the second mapping result.Wherein, parameter includes but not limited to: named entity, vocabulary and location parameter.It should be noted that, in order to avoid there is no the process of the impacts such as the speech text of practical significance to speech text, needed to arrange threshold value before determining match width, this threshold value, for determining the match width of the result of the second mapping, namely thinks in threshold range that the match width of the second mapping result is consistent.The size of the threshold value arranged, the present embodiment does not do concrete restriction, can arrange arbitrary threshold value as required in practical application.
For the ease of understanding, be that " broadcasting Corner With Love " is described for speech text.Wherein, arranging threshold value is 2 bytes.After speech text is mated, obtain < and play >[title of the song], < plays >[video name], these two matched rules; These two matched rules respectively corresponding < play >[title of the song] run into love, < plays >[video name], these two second mapping result.Carry out match width to all matched rules obtained to disappear qi process, determine that matched rule < plays >[title of the song] the second corresponding mapping result < plays >[title of the song] running into the match width of love: first parameter is that beginning of the sentence < plays >, and last parameter is [title of the song], and because the threshold value arranged is 2 bytes, so " running into love " these three words can not be left in the basket, then matched rule < plays >[title of the song] match width be not from beginning of the sentence to sentence tail, matched rule < plays >[video name] the second corresponding mapping result < plays >[video name] match width: first parameter is that < plays >, last parameter is [video name], then matched rule < plays >[video name] match width be from beginning of the sentence to sentence tail, play >[title of the song than matched rule <] match width larger, and then matched rule < maximum for match width is play >[video name] as the matched rule of the qi process that disappears through match width, obtain the matched rule after the qi that disappears.
To disappear the mode of qi process about coupling weighted value, include but not limited to: determine that each pending weighted value of mating disappears the weighted value of matched rule of qi process according to the vocabulary pre-set and the weighted value of named entity;
Using the matched rule of matched rule maximum for weighted value as the qi process that disappears through overmatching weighted value;
Wherein, the disappear matched rule of qi process of pending coupling weighted value is the matched rule of qi process of disappearing through match width.
It should be noted that, the weighted value size of the vocabulary pre-set and named entity, the present embodiment does not do concrete restriction.Because vocabulary is human configuration, and named entity collects in a large amount of from network information, then the weight of vocabulary is larger than the weight of named entity under normal circumstances.
For the ease of understanding, be that " asking for neighbouring restaurant " is described for speech text.After mating speech text, obtain < and search >< dining room >, < searches >[restaurant name], these two matched rules; These two matched rules respectively corresponding < search the < dining room > near >, < searches [restaurant name] near >, these two second mapping result.Carry out match width to all matched rules obtained to disappear qi process, determine that the match width of these two matched rules is identical.Because the match width matched rule after qi that disappears is more than one, then the match width matched rule after qi that disappears is carried out coupling weighted value and to be disappeared qi process.If the weight ratio of the weighted value of the vocabulary pre-set and named entity is 2:1, then determine that weighted value that matched rule < searches >< dining room > is greater than matched rule < and searches >[restaurant name], the coupling rule that namely weighted value is maximum are that < searches >< dining room >; And then < is searched the matched rule of >< dining room > as the qi process that disappears through overmatching weighted value, obtain the matched rule after the qi that disappears.
To disappear the mode of qi process about coupling density, include but not limited to: determine each pending coupling density disappear qi process matched rule and the second mapping result mate proportion, by the matched rule of matched rule maximum for coupling proportion as the qi process that disappears through overmatching density;
Wherein, the disappear matched rule of qi process of each pending coupling density is the matched rule of qi process of disappearing through match width.
It should be noted that, the qi process that disappears of coupling density is situation for comprising similar parameter in matched rule.Such as, in matched rule, all comprise vocabulary, or all comprise named entity in matched rule.
For the ease of understanding, be that " playing the little epoch " is described for speech text.After speech text is mated, obtain < and play >[movie name], < plays >[title of the song], these two matched rules; These two matched rules respectively corresponding < play >[movie name], < plays > little [title of the song], these two second mapping result.Carry out match width to all matched rules obtained to disappear qi, determine that the match width of these two matched rules is identical.Because the match width matched rule after qi that disappears is more than one, then the match width matched rule after qi that disappears is carried out coupling weighted value and to be disappeared qi process, determine that the weighted value of these two matched rules is identical.Because the coupling weighted value matched rule after qi that disappears is more than one, then the coupling weighted value matched rule after qi that disappears is carried out coupling density and to be disappeared qi.Determine that < plays >[movie name] play >[movie name with <] the proportion that mates be 100%, < plays >[title of the song] proportion that mates of playing > little [title of the song] with < is 80%, the matched rule that namely coupling proportion is maximum is < broadcasting >[movie name]; And then < is play >[movie name] as the matched rule of the qi process that disappears through overmatching density, obtain the matched rule after the qi that disappears.
To disappear the mode of qi process about hit quantity, include but not limited to: determine that each pending hit quantity disappears the parameter hit quantity of matched rule of qi process, parameter is one in named entity, vocabulary and location parameter;
Parameter is hit the matched rule of the maximum matching result of quantity as the qi process that disappears through hit quantity;
Wherein, the disappear matched rule of qi process of each pending hit quantity is the matched rule of qi process of disappearing through match width.
For the ease of understanding, be that " played songs wind " is described for speech text.After speech text is mated, obtain < and play >< song >%s1, < plays >< song >, these two matched rules; Wherein, %s1 is location parameter; The corresponding < of these two matched rules plays >< song > wind, this second mapping result.Carry out match width to all matched rules obtained to disappear qi process, when arranging threshold value and being 2 bytes, determine that the match width of these two matched rules is identical.Because the match width matched rule after qi that disappears is more than one, then the match width matched rule after qi that disappears is carried out coupling weighted value and to be disappeared qi process, determine that the weighted value of these two matched rules is identical.Because the coupling weighted value matched rule after qi that disappears is more than one, then the coupling weighted value matched rule after qi that disappears is carried out coupling density and to be disappeared qi.Location parameter is not comprised because < plays in >< song >, then can not the coupling weighted value matched rule after qi that disappears be carried out coupling density and to be disappeared qi process, and then carry out hit quantity to disappear qi process to mating the weighted value matched rule after qi that disappears in order.Determining that the parameter of < broadcasting >< song >%s1 hits quantity is 3, the parameter hit quantity that < plays >< song > is 2, and the matched rule that namely parameter hit quantity is maximum is that < plays >< song >%s1; And then < is play the matched rule of >< song >%s1 as the qi process that disappears through hit quantity, obtain the matched rule after the qi that disappears.
Disappear about named entity weights the mode of qi process, include but not limited to: determine each pending named entity weights disappear qi process matched rule in the weighted value of named entity, using the matched rule of matched rule maximum for the weighted value of named entity as the qi process that disappears through named entity weights;
Wherein, the disappear matched rule of qi process of each pending named entity weights is the matched rule of qi process of disappearing through match width.
For the ease of understanding, be that " broadcasting causes the youth " is described for speech text.After speech text is mated, obtain < and play >[movie name], < plays >[title of the song], these two matched rules respectively corresponding < play >[movie name], < plays >[title of the song], these two second mapping result.Carry out match width to all matched rules obtained to disappear qi process, determine that the match width of these two matched rules is identical.Because the match width matched rule after qi that disappears is more than one, then the match width matched rule after qi that disappears is carried out coupling weighted value and to be disappeared qi process, determine that the weighted value of these two matched rules is identical.Because the coupling weighted value matched rule after qi that disappears is more than one, then the coupling weighted value matched rule after qi that disappears is carried out coupling density and to be disappeared qi process, determine that the coupling density of these two matched rules is identical.Because the coupling density matched rule after qi that disappears is more than one, then the coupling density matched rule after qi that disappears is carried out hit quantity and to be disappeared qi process, determine that the parameter of these two matched rules hits quantity identical.Because the hit quantity matched rule after qi that disappears is more than one, then the hit quantity matched rule after qi that disappears is carried out named entity weights and to be disappeared qi process.When determining that the weighted value of named entity [movie name] is greater than the weighted value of named entity [title of the song], determine that < plays >[movie name] in the weighted value of named entity be greater than < and play >[title of the song] in the weighted value of named entity, the matched rule that namely weighted value of named entity is maximum is that < plays >[movie name]; And then < is play >[movie name] as the matched rule of the qi process that disappears through named entity weights, obtain the matched rule after the qi that disappears.
207: according to the matched rule after the qi that disappears obtained, speech text is processed.
For this step, about the mode processed speech text according to the matched rule after the qi that disappears obtained, the present embodiment does not do concrete restriction, and concrete processing mode is identical with the processing mode principle in above-mentioned steps 205, specifically refers to above-mentioned steps 205.
The method that the present embodiment provides, by carrying out named entity mapping to speech text, obtains the first mapping result; Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result; Afterwards the second mapping result is mated with the preset rules comprising regularity, and according to the matched rule obtained, speech text is processed, thus by unified for the configuration format of regularity and named entity rule, processed voice text techniques is made both to be applicable to actual language environment, be applicable to again the language environment of fixing, expand the scope of application of processed voice text techniques, thus make the processing mode of speech text obtain optimization.
Embodiment three
See Fig. 3, embodiments provide a kind of device of processed voice text, this device comprises:
First mapping block 301, for carrying out named entity mapping to speech text, obtains the first mapping result;
Second mapping block 302, for carrying out vocabulary mapping to the first mapping result, obtains the second mapping result;
Matching module 303, for mating the second mapping result with the preset rules comprising regularity;
First processing module 304, for when obtaining a matched rule, a matched rule according to obtaining processes speech text.
As a kind of preferred embodiment, see Fig. 4, this device also comprises:
Launch module 305, mapping front corresponding speech text for being expanded into successively by one or more named entity in the first mapping result, obtaining at least two the 3rd mapping result;
Second mapping block 302, also for the speech text not being mapped as named entity in each 3rd mapping result is carried out vocabulary mapping, obtains the second mapping result.
As a kind of preferred embodiment, see Fig. 5, this device also comprises:
Disappear qi module 306, for when obtaining at least two matched rules, carry out match width qi process, coupling weighted value qi process, coupling density disappear qi process and the named entity weights at least one disappeared in qi process of qi process, hit quantity that disappear that disappear that disappear successively to all matched rules obtained to disappear qi process, until obtain the matched rule after the qi that disappears;
Second processing module 307, for processing speech text according to the matched rule after the qi that disappears obtained.
As a kind of preferred embodiment, disappear qi module 306, for determining the match width of the second mapping result that each matched rule is corresponding, using the matched rule of matched rule maximum for match width as the qi process that disappears through match width.
As a kind of preferred embodiment, disappear qi module 306, for determining that according to the vocabulary pre-set and the weighted value of named entity each pending weighted value of mating disappears the weighted value of matched rule of qi process; Using the matched rule of matched rule maximum for weighted value as the qi process that disappears through overmatching weighted value;
Wherein, the disappear matched rule of qi process of pending coupling weighted value is the matched rule of qi process of disappearing through match width.
As a kind of preferred embodiment, disappear qi module 306, for determine each pending coupling density disappear qi process matched rule and the second mapping result mate proportion, by the matched rule of matched rule maximum for coupling proportion as the qi process that disappears through overmatching density;
Wherein, the disappear matched rule of qi process of each pending coupling density is the matched rule of qi process of disappearing through match width.
As a kind of preferred embodiment, disappear qi module 306, for determine each pending hit quantity disappear qi process matched rule parameter hit quantity, parameter is one in named entity, vocabulary and location parameter; Parameter is hit the matched rule of the maximum matching result of quantity as the qi process that disappears through hit quantity;
Wherein, the disappear matched rule of qi process of each pending hit quantity is the matched rule of qi process of disappearing through match width.
As a kind of preferred embodiment, disappear qi module 306, for determine each pending named entity weights disappear qi process matched rule in the weighted value of named entity, using the matched rule of matched rule maximum for the weighted value of named entity as the qi process that disappears through named entity weights;
Wherein, the disappear matched rule of qi process of each pending named entity weights is the matched rule of qi process of disappearing through match width.
In sum, the device that the embodiment of the present invention provides, by carrying out named entity mapping to speech text, obtains the first mapping result; Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result; Afterwards the second mapping result is mated with the preset rules comprising regularity, and according to the matched rule obtained, speech text is processed, thus by unified for the configuration format of regularity and named entity rule, processed voice text techniques is made both to be applicable to actual language environment, be applicable to again the language environment of fixing, expand the scope of application of processed voice text techniques, thus make the processing mode of speech text obtain optimization.
Embodiment four
Embodiments provide a kind of terminal, please refer to Fig. 6, it illustrates the structural representation of the terminal involved by the embodiment of the present invention, this terminal may be used for the method implementing the processed voice text provided in above-described embodiment.Specifically:
Terminal 600 can comprise RF(Radio Frequency, radio frequency) circuit 110, the storer 120 including one or more computer-readable recording mediums, input block 130, display unit 140, sensor 150, voicefrequency circuit 160, WiFi(Wireless Fidelity, Wireless Fidelity) module 170, include the parts such as processor 180 and power supply 190 that more than or processes core.It will be understood by those skilled in the art that the restriction of the not structure paired terminal of the terminal structure shown in Fig. 6, the parts more more or less than diagram can be comprised, or combine some parts, or different parts are arranged.Wherein:
RF circuit 110 can be used for receiving and sending messages or in communication process, the reception of signal and transmission, especially, after being received by the downlink information of base station, transfer to more than one or one processor 180 to process; In addition, base station is sent to by relating to up data.Usually, RF circuit 110 includes but not limited to antenna, at least one amplifier, tuner, one or more oscillator, subscriber identity module (SIM) card, transceiver, coupling mechanism, LNA(Low Noise Amplifier, low noise amplifier), diplexer etc.In addition, RF circuit 110 can also by radio communication and network and other devices communicatings.Described radio communication can use arbitrary communication standard or agreement, include but not limited to GSM (Global System of Mobile communication, global system for mobile communications), GPRS (General Packet Radio Service, general packet radio service), CDMA (Code Division Multiple Access, CDMA), WCDMA (Wideband CodeDivision Multiple Access, Wideband Code Division Multiple Access (WCDMA)), LTE (Long Term Evolution, Long Term Evolution), Email, SMS (Short Messaging Service, Short Message Service) etc.
Storer 120 can be used for storing software program and module, and processor 180 is stored in software program and the module of storer 120 by running, thus performs the application of various function and data processing.Storer 120 mainly can comprise storage program district and store data field, and wherein, storage program district can store operating system, application program (such as sound-playing function, image player function etc.) etc. needed at least one function; Store data field and can store the data (such as voice data, phone directory etc.) etc. created according to the use of terminal 600.In addition, storer 120 can comprise high-speed random access memory, can also comprise nonvolatile memory, such as at least one disk memory, flush memory device or other volatile solid-state parts.Correspondingly, storer 120 can also comprise Memory Controller, to provide the access of processor 180 and input block 130 pairs of storeies 120.
Input block 130 can be used for the numeral or the character information that receive input, and produces and to arrange with user and function controls relevant keyboard, mouse, control lever, optics or trace ball signal and inputs.Particularly, input block 130 can comprise Touch sensitive surface 131 and other input equipments 132.Touch sensitive surface 131, also referred to as touch display screen or Trackpad, user can be collected or neighbouring touch operation (such as user uses any applicable object or the operations of annex on Touch sensitive surface 131 or near Touch sensitive surface 131 such as finger, stylus) thereon, and drive corresponding coupling arrangement according to the formula preset.Optionally, Touch sensitive surface 131 can comprise touch detecting apparatus and touch controller two parts.Wherein, touch detecting apparatus detects the touch orientation of user, and detects the signal that touch operation brings, and sends signal to touch controller; Touch controller receives touch information from touch detecting apparatus, and converts it to contact coordinate, then gives processor 180, and the order that energy receiving processor 180 is sent also is performed.In addition, the polytypes such as resistance-type, condenser type, infrared ray and surface acoustic wave can be adopted to realize Touch sensitive surface 131.Except Touch sensitive surface 131, input block 130 can also comprise other input equipments 132.Particularly, other input equipments 132 can include but not limited to one or more in physical keyboard, function key (such as volume control button, switch key etc.), trace ball, mouse, control lever etc.
Display unit 140 can be used for the various graphical user interface showing information or the information being supplied to user and the terminal 600 inputted by user, and these graphical user interface can be made up of figure, text, icon, video and its combination in any.Display unit 140 can comprise display panel 141, optionally, the form such as LCD (Liquid Crystal Display, liquid crystal display), OLED (Organic Light-Emitting Diode, Organic Light Emitting Diode) can be adopted to configure display panel 141.Further, Touch sensitive surface 131 can cover display panel 141, when Touch sensitive surface 131 detects thereon or after neighbouring touch operation, send processor 180 to determine the type of touch event, on display panel 141, provide corresponding vision to export with preprocessor 180 according to the type of touch event.Although in figure 6, Touch sensitive surface 131 and display panel 141 be as two independently parts realize input and input function, in certain embodiments, can by Touch sensitive surface 131 and display panel 141 integrated and realize input and output function.
Terminal 600 also can comprise at least one sensor 150, such as optical sensor, motion sensor and other sensors.Particularly, optical sensor can comprise ambient light sensor and proximity transducer, and wherein, ambient light sensor the light and shade of environmentally light can regulate the brightness of display panel 141, proximity transducer when terminal 600 moves in one's ear, can cut out display panel 141 and/or backlight.As the one of motion sensor, Gravity accelerometer can detect the size of all directions (are generally three axles) acceleration, size and the direction of gravity can be detected time static, can be used for identifying the application (such as horizontal/vertical screen switching, dependent game, magnetometer pose calibrating) of mobile phone attitude, Vibration identification correlation function (such as passometer, knock) etc.; As for terminal 600 also other sensors such as configurable gyroscope, barometer, hygrometer, thermometer, infrared ray sensor, do not repeat them here.
Voicefrequency circuit 160, loudspeaker 161, microphone 162 can provide the audio interface between user and terminal 600.Voicefrequency circuit 160 can by receive voice data conversion after electric signal, be transferred to loudspeaker 161, by loudspeaker 161 be converted to voice signal export; On the other hand, the voice signal of collection is converted to electric signal by microphone 162, voice data is converted to after being received by voicefrequency circuit 160, after again voice data output processor 180 being processed, through RF circuit 110 to send to such as another terminal, or export voice data to storer 120 to process further.Voicefrequency circuit 160 also may comprise 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 to send and receive e-mail by WiFi module 170, browse webpage and access streaming video etc., and its broadband internet wireless for user provides is accessed.Although Fig. 6 shows WiFi module 170, be understandable that, it does not belong to must forming of terminal 600, can omit in the scope of essence not changing invention as required completely.
Processor 180 is control centers of terminal 600, utilize the various piece of various interface and the whole mobile phone of connection, software program in storer 120 and/or module is stored in by running or performing, and call the data be stored in storer 120, perform various function and the process data of terminal 600, thus integral monitoring is carried out to mobile phone.Optionally, processor 180 can comprise one or more process core; Preferably, processor 180 accessible site application processor and modem processor, wherein, application processor mainly processes operating system, user interface and application program etc., and modem processor mainly processes radio communication.Be understandable that, above-mentioned modem processor also can not be integrated in processor 180.
Terminal 600 also comprises the power supply 190(such as battery of powering to all parts), preferably, power supply can be connected with processor 180 logic by power-supply management system, thus realizes the functions such as management charging, electric discharge and power managed by power-supply management system.Power supply 190 can also comprise one or more direct current or AC power, recharging system, power failure detection circuit, power supply changeover device or the random component such as inverter, power supply status indicator.
Although not shown, terminal 600 can also comprise camera, bluetooth module etc., does not repeat them here.Specifically in the present embodiment, the display unit of terminal is touch-screen display, terminal also includes storer, and one or more than one program, one of them or more than one program are stored in storer, and be configured to be performed by more than one or one processor, described more than one or one routine package is containing the instruction for performing following operation:
Named entity mapping is carried out 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 mated with the preset rules comprising regularity, if obtain a matched rule, then according to the matched rule obtained, speech text is processed.
Suppose that above-mentioned is the first possible embodiment, then, in the embodiment that the second provided based on the embodiment that the first is possible is possible, in the storer of terminal, also comprise the instruction for performing following operation:
Before vocabulary mapping is carried out to the first mapping result, comprising:
One or more named entity in first mapping result is expanded into successively speech text corresponding before mapping, obtain at least two the 3rd mapping result;
Vocabulary mapping is carried out to the first mapping result, comprising:
The speech text not being mapped as named entity in each 3rd mapping result is carried out vocabulary mapping, obtains the second mapping result.
In the third the possible embodiment provided, in the storer of terminal, also comprise the instruction for performing following operation based on the embodiment that the first or the second are possible:
After second mapping result is mated with the preset rules comprising regularity, also comprise:
If obtain at least two matched rules, then carry out match width qi process, coupling weighted value qi process, coupling density disappear qi process and the named entity weights at least one disappeared in qi process of qi process, hit quantity that disappear that disappear that disappear successively to all matched rules obtained to disappear qi process, until obtain the matched rule after the qi that disappears;
According to the matched rule after the qi that disappears obtained, speech text is processed.
In the 4th kind of possible embodiment provided based on the embodiment that the third is possible, in the storer of terminal, also comprise the instruction for performing following operation:
Match width disappears qi process, comprising:
Determine the match width of the second mapping result that each matched rule is corresponding, using the matched rule of matched rule maximum for match width as the qi process that disappears through match width.
In the 5th kind of possible embodiment provided based on the 4th kind of possible embodiment, in the storer of terminal, also comprise the instruction for performing following operation:
Coupling weighted value disappears qi process, comprising:
Determine that each pending weighted value of mating disappears the weighted value of matched rule of qi process according to the vocabulary pre-set and the weighted value of named entity;
Using the matched rule of matched rule maximum for weighted value as the qi process that disappears through overmatching weighted value;
Wherein, the disappear matched rule of qi process of pending coupling weighted value is the matched rule of qi process of disappearing through match width.
In the 6th kind of possible embodiment provided based on the 5th kind of possible embodiment, in the storer of terminal, also comprise the instruction for performing following operation:
Coupling density disappears qi process, comprising:
Determine each pending coupling density disappear qi process matched rule and the second mapping result mate proportion, by the matched rule of matched rule maximum for coupling proportion as the qi process that disappears through overmatching density;
Wherein, the disappear matched rule of qi process of each pending coupling density is the matched rule of qi process of disappearing through match width.
In the 7th kind of possible embodiment provided based on the 6th kind of possible embodiment, in the storer of terminal, also comprise the instruction for performing following operation:
Hit quantity disappears qi process, comprising:
Determine each pending hit quantity disappear qi process matched rule parameter hit quantity, parameter is one in named entity, vocabulary and location parameter;
Parameter is hit the matched rule of the maximum matching result of quantity as the qi process that disappears through hit quantity;
Wherein, the disappear matched rule of qi process of each pending hit quantity is the matched rule of qi process of disappearing through match width.
In the 8th kind of possible embodiment provided based on the 7th kind of possible embodiment, in the storer of terminal, also comprise the instruction for performing following operation:
Named entity weights disappear qi process, comprising:
Determine each pending named entity weights disappear qi process matched rule in the weighted value of named entity, using the matched rule of matched rule maximum for the weighted value of named entity as the qi process that disappears through named entity weights;
Wherein, the disappear matched rule of qi process of each pending named entity weights is the matched rule of qi process of disappearing through match width.
In sum, the terminal that the embodiment of the present invention provides, by carrying out named entity mapping to speech text, obtains the first mapping result; Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result; Afterwards the second mapping result is mated with the preset rules comprising regularity, and according to the matched rule obtained, speech text is processed, thus by unified for the configuration format of regularity and named entity rule, processed voice text techniques is made both to be applicable to actual language environment, be applicable to again the language environment of fixing, expand the scope of application of processed voice text techniques, thus make the processing mode of speech text obtain optimization.
Embodiment five
The embodiment of the present invention additionally provides a kind of computer-readable recording medium, and this computer-readable recording medium can be the computer-readable recording medium comprised in the storer in above-described embodiment; Also can be individualism, be unkitted the computer-readable recording medium allocated in terminal.This computer-readable recording medium stores more than one or one program, and this more than one or one program is used for the method for an execution processed voice text by one or more than one processor, the method comprises:
Named entity mapping is carried out 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 mated with the preset rules comprising regularity, if obtain a matched rule, then according to the matched rule obtained, speech text is processed.
Suppose that above-mentioned is the first possible embodiment, then, in the embodiment that the second provided based on the embodiment that the first is possible is possible, in the storer of terminal, also comprise the instruction for performing following operation:
Before vocabulary mapping is carried out to the first mapping result, comprising:
One or more named entity in first mapping result is expanded into successively speech text corresponding before mapping, obtain at least two the 3rd mapping result;
Vocabulary mapping is carried out to the first mapping result, comprising:
The speech text not being mapped as named entity in each 3rd mapping result is carried out vocabulary mapping, obtains the second mapping result.
In the third the possible embodiment provided, in the storer of terminal, also comprise the instruction for performing following operation based on the embodiment that the first or the second are possible:
After second mapping result is mated with the preset rules comprising regularity, also comprise:
If obtain at least two matched rules, then carry out match width qi process, coupling weighted value qi process, coupling density disappear qi process and the named entity weights at least one disappeared in qi process of qi process, hit quantity that disappear that disappear that disappear successively to all matched rules obtained to disappear qi process, until obtain the matched rule after the qi that disappears;
According to the matched rule after the qi that disappears obtained, speech text is processed.
In the 4th kind of possible embodiment provided based on the embodiment that the third is possible, in the storer of terminal, also comprise the instruction for performing following operation:
Match width disappears qi process, comprising:
Determine the match width of the second mapping result that each matched rule is corresponding, using the matched rule of matched rule maximum for match width as the qi process that disappears through match width.
In the 5th kind of possible embodiment provided based on the 4th kind of possible embodiment, in the storer of terminal, also comprise the instruction for performing following operation:
Coupling weighted value disappears qi process, comprising:
Determine that each pending weighted value of mating disappears the weighted value of matched rule of qi process according to the vocabulary pre-set and the weighted value of named entity;
Using the matched rule of matched rule maximum for weighted value as the qi process that disappears through overmatching weighted value;
Wherein, the disappear matched rule of qi process of pending coupling weighted value is the matched rule of qi process of disappearing through match width.
In the 6th kind of possible embodiment provided based on the 5th kind of possible embodiment, in the storer of terminal, also comprise the instruction for performing following operation:
Coupling density disappears qi process, comprising:
Determine each pending coupling density disappear qi process matched rule and the second mapping result mate proportion, by the matched rule of matched rule maximum for coupling proportion as the qi process that disappears through overmatching density;
Wherein, the disappear matched rule of qi process of each pending coupling density is the matched rule of qi process of disappearing through match width.
In the 7th kind of possible embodiment provided based on the 6th kind of possible embodiment, in the storer of terminal, also comprise the instruction for performing following operation:
Hit quantity disappears qi process, comprising:
Determine each pending hit quantity disappear qi process matched rule parameter hit quantity, parameter is one in named entity, vocabulary and location parameter;
Parameter is hit the matched rule of the maximum matching result of quantity as the qi process that disappears through hit quantity;
Wherein, the disappear matched rule of qi process of each pending hit quantity is the matched rule of qi process of disappearing through match width.
In the 8th kind of possible embodiment provided based on the 7th kind of possible embodiment, in the storer of terminal, also comprise the instruction for performing following operation:
Named entity weights disappear qi process, comprising:
Determine each pending named entity weights disappear qi process matched rule in the weighted value of named entity, using the matched rule of matched rule maximum for the weighted value of named entity as the qi process that disappears through named entity weights;
Wherein, the disappear matched rule of qi process of each pending named entity weights is the matched rule of qi process of disappearing through match width.
In sum, the computer-readable recording medium that the embodiment of the present invention provides, by carrying out named entity mapping to speech text, obtains the first mapping result; Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result; Afterwards the second mapping result is mated with the preset rules comprising regularity, and according to the matched rule obtained, speech text is processed, thus by unified for the configuration format of regularity and named entity rule, processed voice text techniques is made both to be applicable to actual language environment, be applicable to again the language environment of fixing, expand the scope of application of processed voice text techniques, thus make the processing mode of speech text obtain optimization.
Embodiment six
Provide a kind of graphical user interface in the embodiment of the present invention, this graphical user interface is used in terminal, and this terminal comprises touch-screen display, storer and one or more than one processor for performing one or more than one program; This graphical user interface comprises:
Named entity mapping is carried out 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 mated with the preset rules comprising regularity, if obtain a matched rule, then according to the matched rule obtained, speech text is processed.
In sum, the graphical user interface that the embodiment of the present invention provides, by carrying out named entity mapping to speech text, obtains the first mapping result; Vocabulary mapping is carried out to the first mapping result, obtains the second mapping result; Afterwards the second mapping result is mated with the preset rules comprising regularity, and according to the matched rule obtained, speech text is processed, thus by unified for the configuration format of regularity and named entity rule, processed voice text techniques is made both to be applicable to actual language environment, be applicable to again the language environment of fixing, expand the scope of application of processed voice text techniques, thus make the processing mode of speech text obtain optimization.
It should be noted that: the device of the processed voice text that above-described embodiment provides is when processed voice text, only be illustrated with the division of above-mentioned each functional module, in practical application, can distribute as required and by above-mentioned functions and be completed by different functional modules, inner structure by device is divided into different functional modules, to complete all or part of function described above.In addition, the device of the processed voice text that above-described embodiment provides and the embodiment of the method for processed voice text belong to same design, and its specific implementation process refers to embodiment of the method, repeats no more here.
The invention described above embodiment sequence number, just to describing, does not represent the quality of embodiment.
One of ordinary skill in the art will appreciate that all or part of step realizing above-described embodiment can have been come by hardware, the hardware that also can carry out instruction relevant by program completes, described program can be stored in a kind of computer-readable recording medium, the above-mentioned storage medium mentioned can be ROM (read-only memory), disk or CD etc.
The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.
Claims (16)
1. a method for processed voice text, is characterized in that, described method comprises:
Named entity mapping is carried out to speech text, obtains the first mapping result;
Vocabulary mapping is carried out to described first mapping result, obtains the second mapping result;
Described second mapping result is mated with the preset rules comprising regularity, if obtain a matched rule, then according to the matched rule obtained, described speech text is processed.
2. method according to claim 1, is characterized in that, described vocabulary mapping is carried out to described first mapping result before, comprising:
One or more named entity in described first mapping result is expanded into successively speech text corresponding before mapping, obtain at least two the 3rd mapping result;
Described vocabulary mapping is carried out to described first mapping result, comprising:
The speech text not being mapped as named entity in each 3rd mapping result is carried out vocabulary mapping, obtains the second mapping result.
3. method according to claim 1 and 2, is characterized in that, described described second mapping result is mated with the preset rules comprising regularity after, also comprise:
If obtain at least two matched rules, then carry out match width qi process, coupling weighted value qi process, coupling density disappear qi process and the named entity weights at least one disappeared in qi process of qi process, hit quantity that disappear that disappear that disappear successively to all matched rules obtained to disappear qi process, until obtain the matched rule after the qi that disappears;
According to the matched rule after the qi that disappears obtained, described speech text is processed.
4. method according to claim 3, is characterized in that, described match width disappears qi process, comprising:
Determine the match width of the second mapping result that each matched rule is corresponding, using the matched rule of matched rule maximum for match width as the qi process that disappears through match width.
5. method according to claim 4, is characterized in that, described coupling weighted value disappears qi process, comprising:
Determine that each pending weighted value of mating disappears the weighted value of matched rule of qi process according to the vocabulary pre-set and the weighted value of named entity;
Using the matched rule of matched rule maximum for weighted value as the qi process that disappears through overmatching weighted value;
Wherein, the disappear matched rule of qi process of described pending coupling weighted value is the matched rule of qi process of disappearing through match width.
6. method according to claim 5, is characterized in that, described coupling density disappears qi process, comprising:
Determine each pending coupling density disappear qi process matched rule and the second mapping result mate proportion, by the matched rule of matched rule maximum for coupling proportion as the qi process that disappears through overmatching density;
Wherein, the disappear matched rule of qi process of each pending coupling density is the matched rule of qi process of disappearing through match width.
7. method according to claim 6, is characterized in that, described hit quantity disappears qi process, comprising:
Determine each pending hit quantity disappear qi process matched rule parameter hit quantity, described parameter is one in named entity, vocabulary and location parameter;
Parameter is hit the matched rule of the maximum matching result of quantity as the qi process that disappears through hit quantity;
Wherein, the disappear matched rule of qi process of each pending hit quantity is the matched rule of qi process of disappearing through match width.
8. method according to claim 7, is characterized in that, described named entity weights disappear qi process, comprising:
Determine each pending named entity weights disappear qi process matched rule in the weighted value of named entity, using the matched rule of matched rule maximum for the weighted value of named entity as the qi process that disappears through named entity weights;
Wherein, the disappear matched rule of qi process of each pending named entity weights is the matched rule of qi process of disappearing through match width.
9. a device for processed voice text, is characterized in that, described device comprises:
First mapping block, for carrying out named entity mapping to speech text, obtains the first mapping result;
Second mapping block, for carrying out vocabulary mapping to described first mapping result, obtains the second mapping result;
Matching module, for mating described second mapping result with the preset rules comprising regularity;
First processing module, for when obtaining a matched rule, a matched rule according to obtaining processes described speech text.
10. device according to claim 9, is characterized in that, described device, also comprises:
Launch module, mapping front corresponding speech text for being expanded into successively by one or more named entity in described first mapping result, obtaining at least two the 3rd mapping result;
Described second mapping block, also for the speech text not being mapped as named entity in each 3rd mapping result is carried out vocabulary mapping, obtains the second mapping result.
11. devices according to claim 9 or 10, it is characterized in that, described device, also comprises:
Disappear qi module, for when obtaining at least two matched rules, carry out match width qi process, coupling weighted value qi process, coupling density disappear qi process and the named entity weights at least one disappeared in qi process of qi process, hit quantity that disappear that disappear that disappear successively to all matched rules obtained to disappear qi process, until obtain the matched rule after the qi that disappears;
Second processing module, for processing described speech text according to the matched rule after the qi that disappears obtained.
12. devices according to claim 11, it is characterized in that, the described qi module that disappears, for determining the match width of the second mapping result that each matched rule is corresponding, using the matched rule of matched rule maximum for match width as the qi process that disappears through match width.
13. devices according to claim 12, is characterized in that, described in disappear qi module, for determining that according to the vocabulary pre-set and the weighted value of named entity each pending weighted value of mating disappears the weighted value of matched rule of qi process; Using the matched rule of matched rule maximum for weighted value as the qi process that disappears through overmatching weighted value;
Wherein, the disappear matched rule of qi process of described pending coupling weighted value is the matched rule of qi process of disappearing through match width.
14. devices according to claim 13, it is characterized in that, the described qi module that disappears, for determine each pending coupling density disappear qi process matched rule and the second mapping result mate proportion, by the matched rule of matched rule maximum for coupling proportion as the qi process that disappears through overmatching density;
Wherein, the disappear matched rule of qi process of each pending coupling density is the matched rule of qi process of disappearing through match width.
15. devices according to claim 14, is characterized in that, described in disappear qi module, for determining that each pending hit quantity disappears the parameter hit quantity of matched rule of qi process, described parameter is one in named entity, vocabulary and location parameter; Parameter is hit the matched rule of the maximum matching result of quantity as the qi process that disappears through hit quantity;
Wherein, the disappear matched rule of qi process of each pending hit quantity is the matched rule of qi process of disappearing through match width.
16. devices according to claim 15, it is characterized in that, the described qi module that disappears, for determine each pending named entity weights disappear qi process matched rule in the weighted value of named entity, using the matched rule of matched rule maximum for the weighted value of named entity as the qi process that disappears through named entity weights;
Wherein, the disappear matched rule of qi process of each pending named entity weights is the matched rule of qi process of disappearing through match width.
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 true CN104636323A (en) | 2015-05-20 |
| CN104636323B 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) |
Cited By (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 |
| CN111079435A (en) * | 2019-12-09 | 2020-04-28 | 深圳追一科技有限公司 | Named entity disambiguation method, device, equipment and storage medium |
| TWI734142B (en) * | 2018-07-17 | 2021-07-21 | 美商艾提史畢克斯有限責任公司 | Method, system, and computer program product for an intelligent industrial assistant |
Citations (8)
| 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 |
| US20120078918A1 (en) * | 2010-09-28 | 2012-03-29 | Siemens Corporation | Information Relation Generation |
| 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 |
-
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 (8)
| 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 |
| US20120078918A1 (en) * | 2010-09-28 | 2012-03-29 | Siemens Corporation | Information Relation Generation |
| 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 |
Cited By (5)
| 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 |
| TWI734142B (en) * | 2018-07-17 | 2021-07-21 | 美商艾提史畢克斯有限責任公司 | Method, system, and computer program product for an intelligent industrial assistant |
| US11232262B2 (en) | 2018-07-17 | 2022-01-25 | iT SpeeX LLC | Method, system, and computer program product for an intelligent industrial assistant |
| CN111079435A (en) * | 2019-12-09 | 2020-04-28 | 深圳追一科技有限公司 | Named entity disambiguation method, device, equipment and storage medium |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104636323B (en) | 2018-04-03 |
| WO2015067116A1 (en) | 2015-05-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104252341B (en) | The client device of the skin change method of application program, device and application program | |
| CN104869468A (en) | Method and apparatus for displaying screen information | |
| CN104967896A (en) | Method for displaying bulletscreen comment information, and apparatus thereof | |
| CN104967910A (en) | Multimedia play progress control method and apparatus | |
| CN103559731B (en) | Method and terminal for displaying lyrics under screen locking state | |
| CN106331359B (en) | A kind of speech signal collection method, device and terminal | |
| CN104519404A (en) | Graphics interchange format file playing method and device | |
| CN105022616A (en) | Method and device for generating web page | |
| CN105094809A (en) | Combined picture layout modification method and device and terminal equipment | |
| CN104850406A (en) | Page switching method and device | |
| CN103475914A (en) | Video playing method, video playing device, terminal equipment and server | |
| CN105224556A (en) | Waterfall stream interface display packing and device | |
| CN104461597A (en) | Starting control method and device for application program | |
| CN104869465A (en) | Video playing control method and device | |
| CN103945241A (en) | Streaming data statistical method, system and related device | |
| CN105630846A (en) | Head portrait updating method and apparatus | |
| CN106170034A (en) | A kind of sound effect treatment method and mobile terminal | |
| CN104602135A (en) | Method and device for controlling full screen play | |
| CN104965642A (en) | Method and apparatus for generating a drop-down list | |
| CN105550316A (en) | Pushing method and device of audio list | |
| CN103399706A (en) | Page interaction method, device and terminal | |
| CN106210838B (en) | Caption presentation method and device | |
| CN104243394A (en) | Multimedia file playing method and device | |
| CN104598542A (en) | Display method and device for multimedia information | |
| CN104731806A (en) | Method and terminal for quickly finding user information in social network |
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 |