CN113553843B - Skill creation method and device - Google Patents
Skill creation method and device Download PDFInfo
- Publication number
- CN113553843B CN113553843B CN202110706140.0A CN202110706140A CN113553843B CN 113553843 B CN113553843 B CN 113553843B CN 202110706140 A CN202110706140 A CN 202110706140A CN 113553843 B CN113553843 B CN 113553843B
- Authority
- CN
- China
- Prior art keywords
- skill
- corpus
- skills
- original
- word
- 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
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000012360 testing method Methods 0.000 claims abstract description 93
- 238000011161 development Methods 0.000 claims abstract description 24
- 230000002618 waking effect Effects 0.000 claims abstract description 15
- 238000012508 change request Methods 0.000 claims description 14
- 238000012550 audit Methods 0.000 claims description 7
- 238000000605 extraction Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 abstract description 11
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000010586 diagram Methods 0.000 description 8
- 241000234295 Musa Species 0.000 description 7
- 235000018290 Musa x paradisiaca Nutrition 0.000 description 7
- 238000004590 computer program Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 4
- 238000003058 natural language processing Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000002994 raw material Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
- G06F40/35—Discourse or dialogue representation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Abstract
The invention discloses a skill creation method and a skill creation device. Wherein the method comprises the following steps: acquiring original skill corpus, wherein the original skill corpus is corpus for waking up non-called word skills, and the non-called word skills are skills which do not need to be woken up by using called words; identifying at least one slot of the skill corpus, and acquiring a replacement word of each slot; filling the groove into the original skill corpus by using the replacement words based on the groove position to obtain a target skill corpus; performing conflict test on the target skill corpus to obtain a first test result, and completing creation of non-calling word skills in the development platform under the condition that the first test result is that the target skill corpus is not in conflict. The invention solves the technical problem that the interaction steps are complicated due to the adoption of the call word awakening skill in the related technology, thereby reducing the user experience.
Description
Technical Field
The invention relates to the field of skill creation, in particular to a skill creation method and device.
Background
At present, the NPL (natural language processing ) system can only give a corpus to a service, if an open platform can add the corpus at will to point to its own service, the NLP logic is confused, and a large number of problems of robbing the platform corpus and classification errors occur.
However, at present, the corpus is restrained by the call words, the skills can be awakened only after the call words are input by the user, and the interaction steps are complicated, so that the user experience is reduced.
In view of the above problems, no effective solution has been proposed at present.
Disclosure of Invention
The embodiment of the invention provides a skill creation method and a skill creation device, which at least solve the technical problem that interaction steps are complicated due to the fact that calling words are adopted to wake up skills in the related technology, so that user experience is reduced.
According to an aspect of an embodiment of the present invention, there is provided a skill creation method including: acquiring original skill corpus, wherein the original skill corpus is corpus for waking up non-called word skills, and the non-called word skills are skills which do not need to be woken up by using called words; identifying at least one slot of the skill corpus, and acquiring a replacement word of each slot; filling the groove into the original skill corpus by using the replacement words based on the groove position to obtain a target skill corpus; performing conflict test on the target skill corpus to obtain a first test result, and completing creation of non-calling word skills in the development platform under the condition that the first test result is that the target skill corpus is not in conflict.
Optionally, before obtaining the original skill corpus, the method further comprises: starting to create non-call word skills according to skill creation instructions, wherein the skill creation instructions comprise at least one of the following: skill name, skill type; creating an intent that does not invoke word skills according to intent creation instructions, wherein the intent creation instructions include at least one of: intent name, user statement; the original skill corpus is obtained based on the intent.
Optionally, identifying at least one slot of the skill corpus and obtaining a replacement word for each slot includes: identifying at least one slot of the skill corpus; acquiring a dictionary corresponding to each slot; and extracting a preset number of replacement words from the dictionary corresponding to each slot.
Optionally, based on the slot position, filling the slot into the original skill corpus by using the replacement word to obtain the target skill corpus, including: and filling the groove into the original skill corpus by using a preset number of replacement words to obtain at least one target skill corpus.
Optionally, if the first test result is that the target skill corpus conflicts, feeding back a change request, wherein the change request is used for prompting a user to edit the original skill corpus; and receiving an editing completion instruction, and executing the step of acquiring the original skill corpus based on the editing completion instruction.
Optionally, in the case that the first test result is that the target skill corpus has no conflict, after the creation of the non-call word skills is completed in the development platform, the method further includes: performing conflict test on the target skill corpus in the application platform to obtain a second test result; under the condition that the second test result is that the target skill corpus has no conflict, performing standard auditing on the created non-call word skills; and under the condition that the standard audit is passed, the word skills are not invoked online in the application platform.
Optionally, under the condition that the second test result is that the target skill corpus has conflict, the online non-calling word skills in the application platform are forbidden.
According to another aspect of the embodiment of the present invention, there is also provided a skill creation apparatus, including: the first acquisition module is used for acquiring original skill corpus, wherein the original skill corpus is the corpus for waking up non-called word skills, and the non-called word skills are skills which do not need to be woken up by using called words; the recognition module is used for recognizing at least one slot position of the skill corpus and acquiring a replacement word of each slot position; the groove filling module is used for filling grooves into the original skill corpus by using the replacement words based on the groove positions to obtain the target skill corpus; the first test module is used for performing conflict test on the target skill corpus to obtain a first test result, and completing the creation of the non-calling word skills in the development platform under the condition that the first test result is that the target skill corpus is not in conflict.
Optionally, the apparatus further comprises: the first creating module is used for starting to create non-call word skills according to skill creating instructions, wherein the skill creating instructions comprise at least one of the following steps: skill name, skill type; a second creation module for creating an intent of the non-invoking word skill according to an intent creation instruction, wherein the intent creation instruction includes at least one of: intent name, user statement; and the second acquisition module is used for acquiring the original skill corpus based on the intention.
Optionally, the identification module includes: the recognition unit is used for recognizing at least one slot of the skill corpus; the acquisition unit is used for acquiring the dictionary corresponding to each slot position; the extraction unit is used for extracting a preset number of replacement words from the dictionary corresponding to each slot.
Optionally, the slot filling module includes: and the groove filling unit is used for filling grooves in the original skill corpus by using a preset number of replacement words to obtain at least one target skill corpus.
Optionally, the apparatus further comprises: the feedback module is used for feeding back a change request under the condition that the first test result is that the target skill corpus conflicts, wherein the change request is used for prompting a user to edit the original skill corpus; and receiving an editing completion instruction, and executing the step of acquiring the original skill corpus based on the editing completion instruction.
Optionally, the apparatus further comprises: the second testing module is used for performing conflict testing on the target skill corpus in the application platform to obtain a second testing result; the auditing module is used for carrying out standard auditing on the established non-call word skills under the condition that the second test result is that the target skill corpus has no conflict; and the online module is used for online non-calling word skills in the application platform under the condition that the standard audit is passed.
Optionally, the apparatus further comprises: and the prohibition module is used for prohibiting online non-call word skills in the application platform under the condition that the second test result is that the target skill corpus has conflict.
According to another aspect of the embodiment of the present invention, there is further provided a computer readable storage medium, where the computer readable storage medium includes a stored program, and when the program runs, the device on which the computer readable storage medium is controlled to execute the skill creation method described above.
According to another aspect of the embodiment of the present invention, there is also provided a processor, configured to run a program, where the program executes the skill creation method described above.
In the embodiment of the invention, the original skill corpus can be acquired first, wherein the original skill corpus is the corpus for waking up non-called word skills, and the non-called word skills are skills for waking up without using called words; then, at least one slot of the skill corpus is identified, and a replacement word of each slot is obtained; filling the groove into the original skill corpus by using the replacement words based on the groove position to obtain a target skill corpus; and finally, performing conflict test on the target skill corpus to obtain a first test result, and under the condition that the first test result is that the target skill corpus has no conflict, completing the creation of non-call word skills in the development platform, so that the creation of the non-call word skills is realized, a user can call the required skills without calling words, and the conflict between a plurality of corpora can be avoided through the conflict test, thereby causing logic confusion, further solving the technical problem that interaction steps are complicated due to the fact that calling words are adopted to wake up skills in the related technology, and further reducing user experience.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
FIG. 1 is a flow chart of a skill creation method according to an embodiment of the invention;
FIG. 2 is a flow chart of another skill creation method according to an embodiment of the invention;
FIG. 3 is a schematic diagram of an edited corpus in accordance with an embodiment of the invention;
FIG. 4 is a schematic diagram of a developer conflict test, according to an embodiment of the invention;
FIG. 5 is a schematic diagram of a skill creation apparatus according to an embodiment of the invention;
fig. 6 is a block diagram of a hardware structure of a computer terminal of a processing method of a data request according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Example 1
In accordance with an embodiment of the present invention, there is provided an embodiment of a skill creation method, it being noted that the steps illustrated in the flowchart of the figures may be performed in a computer system, such as a set of computer executable instructions, and that, although a logical order is illustrated in the flowchart, in some cases, the steps illustrated or described may be performed in an order other than that illustrated herein.
FIG. 1 is a flow chart of a skill creation method according to an embodiment of the invention, as shown in FIG. 1, comprising the steps of:
step S102, obtaining an original skill corpus.
The original skill corpus is the corpus for waking up non-called word skills, and the non-called word skills are skills which do not need to be woken up by using called words.
The original skill corpus can be corpus which can wake up skills and is input by a developer, for example, what garbage is in Beijing banana peel.
In an alternative embodiment, the original skill corpus may be invoked from the database according to the name, intent of the non-invoked word skill, and crawl from the web page may also be utilized.
The non-call word skills do not need to be awakened by using call words, and only the content input by the user is required to hit the skill corpus.
The skill creation method described above may be applied in a voice conversation platform or a voice interaction platform.
Step S104, at least one slot of the skill corpus is identified, and replacement words of each slot are obtained.
At least one slot in the skill corpus can be identified using a pre-trained slot identification model.
In an alternative embodiment, after identifying at least one slot of the skill corpus, the at least one slot may be marked, a dictionary associated with the slot may be determined based on terms in the slot, and at least one replacement term for the slot may be obtained from the dictionary.
Exemplary, the slots in the Beijing banana peel and what garbage are Beijing and banana peel, wherein the dictionary corresponding to Beijing is a place and the dictionary corresponding to banana peel is garbage, so that a plurality of replacement words can be selected from the dictionary at the place for filling slots, and a plurality of replacement words can be selected from the dictionary of garbage for filling slots. It should be noted that the substitution words of each slot can be freely combined.
And S106, filling the groove into the original skill corpus by using the replacement words based on the groove position to obtain the target skill corpus.
In an alternative embodiment, the original skill corpus may be filled with the replacement word according to the slot position, so as to obtain the target skill corpus containing the replacement word in the slot position.
In another alternative embodiment, a plurality of replacement words of each slot position can be freely combined, and the corresponding slot position in the original skill raw material is filled with slots according to the combined replacement word group, so that a plurality of target skill corpus is obtained.
And S108, performing conflict test on the target skill corpus to obtain a first test result, and completing the creation of the non-calling word skills in the development platform under the condition that the first test result is that the target skill corpus has no conflict.
The above conflict test refers to whether the target skill corpus conflicts with the skill corpus in other non-call word skills, if the target skill corpus conflicts with the skill corpus in other non-call word skills, the logic of the non-call word wake-up skills is wrong, for example, the same skill corpus may wake up two non-call skills at the same time.
In an optional embodiment, when the first test result is that the target skill corpus has no conflict, the target skill corpus is different from the skill corpus of other non-call word skills, and at this time, it can be determined that the target skill corpus can only wake up the non-call word skills, but cannot wake up other non-call word skills, so that the problem of logic occurrence is avoided, and therefore, the creation of the non-call word skills can be completed in the development platform.
Through the steps, the original skill corpus can be obtained, wherein the original skill corpus is the corpus for waking up non-called word skills, and the non-called word skills are skills which do not need to be woken up by using called words; then, at least one slot of the skill corpus is identified, and a replacement word of each slot is obtained; filling the groove into the original skill corpus by using the replacement words based on the groove position to obtain a target skill corpus; and finally, performing conflict test on the target skill corpus to obtain a first test result, and under the condition that the first test result is that the target skill corpus has no conflict, completing the creation of non-call word skills in the development platform, so that the creation of the non-call word skills is realized, a user can call the required skills without calling words, and the conflict between a plurality of corpora can be avoided through the conflict test, thereby causing logic confusion, further solving the technical problem that interaction steps are complicated due to the fact that calling words are adopted to wake up skills in the related technology, and further reducing user experience.
Optionally, before obtaining the original skill corpus, the method further comprises: starting to create non-call word skills according to skill creation instructions, wherein the skill creation instructions comprise at least one of the following: skill name, skill type; creating an intent that does not invoke word skills according to intent creation instructions, wherein the intent creation instructions include at least one of: intent name, user statement; the original skill corpus is obtained based on the intent.
In an alternative embodiment, before the original skill corpus is obtained, the user may click a button for skill creation in the skill development platform, input a skill name and a skill type of the non-call word skill to be created, click a determination button after completion of filling, and generate a skill creation instruction at this time, and start to create the non-call word skill according to the skill creation instruction.
Further, after starting to create the non-calling skills, a skill creation button may be displayed, at which time the user may click on the skill creation button and input the skill name and skill type of the non-calling word skills after clicking on the skill creation button, and after completion of the filling, a determination button may be clicked, at which time an intent creation instruction may be generated and the acquisition of the original skill corpus of the non-calling skills may be started according to the intent creation instruction.
In another alternative embodiment, the original skill corpus associated with the intent may be obtained from a database or web page based on the intent.
Optionally, identifying at least one slot of the skill corpus and obtaining a replacement word for each slot includes: identifying at least one slot of the skill corpus; acquiring a dictionary corresponding to each slot; and extracting a preset number of replacement words from the dictionary corresponding to each slot.
The dictionary may be edited by the user himself, wherein the dictionary may be provided by the development platform.
The predetermined number may be 10.
In an alternative embodiment, a plurality of slots of the skill corpus can be identified, a dictionary corresponding to each slot is obtained, then a preset number of replacement words are extracted from the dictionary corresponding to each slot, and the extracted replacement words in each slot are freely combined.
Optionally, based on the slot position, filling the slot into the original skill corpus by using the replacement word to obtain the target skill corpus, including: and filling the groove into the original skill corpus by using a preset number of replacement words to obtain at least one target skill corpus.
In an alternative embodiment, a preset number of replacement words in each slot position are freely combined to obtain a replacement phrase, and then the replacement words of the replacement phrase are utilized to fill slots in the corresponding slots in the original skill corpus to obtain at least one target skill corpus.
Optionally, if the first test result is that the target skill corpus conflicts, feeding back a change request, wherein the change request is used for prompting a user to edit the original skill corpus; and receiving an editing completion instruction, and executing the step of acquiring the original skill corpus based on the editing completion instruction.
In an alternative embodiment, when the target skill corpus collides, a change request may be fed back to the user, so that the user may edit the original skill corpus according to the change request, after the editing is completed, the user may click a completion button, an edit completion instruction may be generated at this time, the edit completion instruction is received, a step of obtaining the original skill corpus may be performed according to the edit completion instruction, and herein, a collision test may be performed on the edited original skill corpus until the original skill corpus does not collide.
Optionally, in the case that the first test result is that the target skill corpus has no conflict, after the creation of the non-call word skills is completed in the development platform, the method further includes: performing conflict test on the target skill corpus in the application platform to obtain a second test result; under the condition that the second test result is that the target skill corpus has no conflict, performing standard auditing on the created non-call word skills; and under the condition that the standard audit is passed, the word skills are not invoked online in the application platform.
In an optional embodiment, after the creation of the non-call word skills is completed in the development platform, a conflict test can be performed on the target skill corpus in the application platform, so that the original skill corpus and the skill corpus of other non-call skills are again guaranteed not to collide, logic errors are avoided, the created non-call skills can be subjected to standard auditing under the condition that the second test result is that the target skill corpus is not in conflict, whether the non-call skills conform to other standards of the platform is checked, and under the condition that the standard auditing is passed, the non-call skills can be online on the application platform, so that the non-call skills are disclosed, and the non-call skills are convenient for users to use.
Optionally, under the condition that the second test result is that the target skill corpus has conflict, the online non-calling word skills in the application platform are forbidden.
In an optional embodiment, when the second test result is that the target skill corpus has a conflict, it is stated that the original skill corpus exists in other non-calling skills, and at this time, the created non-calling skills need to be rejected in the application platform, so that logic confusion is avoided.
A preferred embodiment of the present invention will be described in detail with reference to fig. 2.
Fig. 2 is a flowchart of skill creation according to an embodiment of the present invention, where a developer first starts to create a skill in a development platform, then performs intent creation, after the intent creation is completed, the developer may input a skill corpus according to the skill to be created and the intent to create the skill, edit the skill corpus, perform a conflict test on the edited skill corpus when the skill corpus is edited, and re-edit the skill corpus when the conflict is generated until no conflict is generated in the conflict test, thereby completing the creation of the skill in the development platform. At the moment, the established skills can be submitted to an application platform side for auditing, the corpus of the skills is subjected to conflict test again at the platform side, and the established skills are directly rejected when the conflicts are generated; other audits of the skill may be performed when no conflict is generated, e.g., whether other aspects of the skill meet the platform specifications; in the event that the platform specification is met, the skill is brought on-line in the application platform.
Referring to fig. 3, a schematic diagram of editing a corpus according to an embodiment of the present invention is shown, a developer directly inputs a skill corpus, for example, "what garbage is in beijing banana peel", then marks the positions "beijing" and "banana peel" of the skill corpus, and obtains a dictionary associated with the positions, for example, a dictionary in which "beijing" can be associated with "place", and a dictionary in which "banana" can be associated with "garbage", so that "what garbage is in Shanghai" obtained by filling the slots in the skill corpus with "Shanghai" in the dictionary and "napkin" in the dictionary in "garbage" can hit the skill.
It should be noted that, the dictionary may be edited, and the development platform may be provided. For example, the "place" dictionary contains the provincial region of China.
Fig. 4 is a schematic diagram of a developer conflict test according to an embodiment of the present invention, firstly, 10 words may be randomly extracted from a slot according to the slot to combine a plurality of skill corpora, then, the combined skill corpora simulate a user voice test through an interface, and a test result is obtained, where the test result includes an actual hit skill, an NLP analysis result and a reply word, and when the actual hit skill is in an open range of a development platform, the actual hit skill is identified as non-conflict, and when the actual hit skill is in an open range of the development platform, the actual hit skill is identified as conflict. If conflict occurs, the developer is required to change the corpus by himself, if no conflict occurs, the developer can apply for online, wherein the developer can fill in the information of the skill when applying for online, and can apply for online submitting and auditing.
Further, when the application platform performs the conflict test, the manner of performing the conflict test is the same as that of the development platform, and will not be described herein.
Through the steps, after the corpus is edited by a developer, the corpus which can be said by a user can be combined in a way of extracting a slot dictionary, and when the conflict test is carried out, the combined skill corpus can be tested by using an interface, whether the conflict exists or not is marked according to the analysis result, the non-calling word skill is opened, and the risk of the non-calling word skill can be controlled, so that the non-calling word skill is realized on a skill platform.
Currently, the way to interact using the call word skills is as follows:
in the case of a user checking tax skills, the interaction of the call words is:
the user: 'opening differential tax'
A voice assistant: "Happy coming and coming differential tax"
The user: "what is there 9000 tax in Beijing wages? "
A voice assistant: 9000 tax in Beijing wages is 180 yuan'
The manner of interacting using non-calling word skills is as follows:
the user: "what is there 9000 tax in Beijing wages? "
A voice assistant: 9000 tax in Beijing wages is 180 yuan'
Therefore, the skill which can be created by the open platform can reduce the user interaction cost, the number of interaction rounds and the interaction time, and simultaneously, the memory cost of the user can be reduced, the user does not need to memorize the call word, the skill can be hit by directly speaking the natural corpus, and the interaction is completed.
Example 2
According to the embodiment of the present invention, a skill creation device is further provided, which can execute the skill creation method in the above embodiment, and the specific implementation manner and the preferred application scenario are the same as those in the above embodiment, and are not described herein.
FIG. 5 is a schematic diagram of a skill creation apparatus, as shown in FIG. 5, according to an embodiment of the invention, the apparatus comprising:
the first obtaining module 52 is configured to obtain an original skill corpus, where the original skill corpus is a corpus for waking up non-called word skills, and the non-called word skills are skills that do not need to be woken up by using a call word;
the recognition module 54 is configured to recognize at least one slot of the skill corpus, and obtain a replacement word of each slot;
the slot filling module 56 is configured to fill slots in the original skill corpus by using the replacement words based on the slot positions, so as to obtain a target skill corpus;
the first test module 58 is configured to perform a conflict test on the target skill corpus to obtain a first test result, and complete creation of the non-call word skill in the development platform if the first test result is that the target skill corpus has no conflict.
Optionally, the apparatus further comprises: the first creating module is used for starting to create non-call word skills according to skill creating instructions, wherein the skill creating instructions comprise at least one of the following steps: skill name, skill type; a second creation module for creating an intent of the non-invoking word skill according to an intent creation instruction, wherein the intent creation instruction includes at least one of: intent name, user statement; and the second acquisition module is used for acquiring the original skill corpus based on the intention.
Optionally, the identification module includes: the recognition unit is used for recognizing at least one slot of the skill corpus; the acquisition unit is used for acquiring the dictionary corresponding to each slot position; the extraction unit is used for extracting a preset number of replacement words from the dictionary corresponding to each slot.
Optionally, the slot filling module includes: and the groove filling unit is used for filling grooves in the original skill corpus by using a preset number of replacement words to obtain at least one target skill corpus.
Optionally, the apparatus further comprises: the feedback module is used for feeding back a change request under the condition that the first test result is that the target skill corpus conflicts, wherein the change request is used for prompting a user to edit the original skill corpus; and receiving an editing completion instruction, and executing the step of acquiring the original skill corpus based on the editing completion instruction.
Optionally, the apparatus further comprises: the second testing module is used for performing conflict testing on the target skill corpus in the application platform to obtain a second testing result; the auditing module is used for carrying out standard auditing on the established non-call word skills under the condition that the second test result is that the target skill corpus has no conflict; and the online module is used for online non-calling word skills in the application platform under the condition that the standard audit is passed.
Optionally, the apparatus further comprises: and the prohibition module is used for prohibiting online non-call word skills in the application platform under the condition that the second test result is that the target skill corpus has conflict.
An embodiment of the present invention also provides a storage medium including a stored program that, when executed, performs the method of any one of the above.
Alternatively, in the present embodiment, the above-described storage medium may be configured to store program code for performing the steps of:
s1, acquiring original skill corpus, wherein the original skill corpus is the corpus for waking up non-called word skills, and the non-called word skills are skills for waking up without using called words;
s2, identifying at least one slot of the skill corpus, and acquiring a replacement word of each slot;
s3, filling slots into the original skill corpus by using replacement words based on slots to obtain target skill corpus;
s4, performing conflict test on the target skill corpus to obtain a first test result, and completing creation of non-calling word skills in the development platform under the condition that the first test result is that the target skill corpus is not in conflict.
The present invention also provides an electronic device comprising a memory in which a computer program is stored and a processor arranged to run the computer program to perform the steps of any of the method embodiments described above.
Optionally, the electronic apparatus may further include a transmission device and an input/output device, where the transmission device is connected to the processor, and the input/output device is connected to the processor.
Alternatively, in the present embodiment, the above-described processor may be configured to execute the following steps by a computer program:
s1, acquiring original skill corpus, wherein the original skill corpus is the corpus for waking up non-called word skills, and the non-called word skills are skills for waking up without using called words;
s2, identifying at least one slot of the skill corpus, and acquiring a replacement word of each slot;
s3, filling slots into the original skill corpus by using replacement words based on slots to obtain target skill corpus;
s4, performing conflict test on the target skill corpus to obtain a first test result, and completing creation of non-calling word skills in the development platform under the condition that the first test result is that the target skill corpus is not in conflict.
Alternatively, in the present embodiment, the storage medium may include, but is not limited to: a U-disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
Alternatively, specific examples in this embodiment may refer to examples described in the foregoing embodiments and optional implementations, and this embodiment is not described herein.
The method embodiments provided by the embodiments of the present application may be performed in a computer terminal, or a similar computing device. Taking the example of running on a computer terminal, fig. 6 is a block diagram of the hardware structure of the computer terminal of a data request processing method according to an embodiment of the present invention. As shown in fig. 6, the computer terminal may include one or more processors 602 (only one shown in fig. 6) (the processor 602 may include, but is not limited to, a microprocessor MCU or a processing device such as a programmable logic device FPGA) and a memory 604 for storing data, and in one exemplary embodiment, may also include a transmission device 606 for communication functions and an input-output device 608. It will be appreciated by those skilled in the art that the configuration shown in fig. 6 is merely illustrative and is not intended to limit the configuration of the computer terminal described above. For example, the computer terminal may also include more or fewer components than shown in FIG. 6, or have a different configuration than the equivalent functions shown in FIG. 6 or more than the functions shown in FIG. 6.
The memory 604 may be used to store a computer program, for example, a software program of application software and a module, such as a computer program corresponding to a method for processing a data request in an embodiment of the present invention, and the processor 602 executes the computer program stored in the memory 604 to perform various functional applications and data processing, that is, implement the above-mentioned method. Memory 604 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid state memory. In some examples, memory 604 may further comprise memory located remotely from processor 602, which may be connected to the computer terminal via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The transmitting device 606 is used to receive or transmit data via a network. Specific examples of the network described above may include a wireless network provided by a communication provider of a computer terminal. In one example, the transmitting device 606 includes a network adapter (Network Interface Controller, simply referred to as NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 606 may be a Radio Frequency (RF) module, which is configured to communicate with the internet wirelessly.
It will be appreciated by those skilled in the art that the modules or steps of the invention described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may alternatively be implemented in program code executable by computing devices, so that they may be stored in a memory device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps within them may be fabricated into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the principle of the present invention should be included in the protection scope of the present invention.
Claims (12)
1. A skill creation method, comprising:
acquiring original skill corpus, wherein the original skill corpus is corpus for waking up non-called word skills, and the non-called word skills are skills for waking up without using called words;
identifying at least one slot of the skill corpus, and acquiring a replacement word of each slot;
based on the slot positions, filling slots into the original skill corpus by using the replacement words to obtain a target skill corpus;
performing conflict test on the target skill corpus to obtain a first test result, and completing creation of the non-calling word skills in a development platform under the condition that the first test result is that the target skill corpus is not in conflict;
when the first test result shows that the target skill corpus has conflict, a change request is fed back, wherein the change request is used for prompting a user to edit the original skill corpus;
receiving an editing completion instruction, acquiring an original skill corpus based on the editing completion instruction, and performing conflict test on the edited original skill corpus until the original skill corpus does not conflict;
wherein, discernment at least one slot of skill corpus to obtain the replacement word of every slot, include:
identifying at least one slot of the skill corpus;
acquiring a dictionary corresponding to each slot;
and extracting a preset number of replacement words from the dictionary corresponding to each slot.
2. The method of claim 1, wherein prior to obtaining the original skill corpus, the method further comprises:
starting to create the non-call word skills according to skill creation instructions, wherein the skill creation instructions comprise at least one of the following: skill name, skill type;
creating an intent of the non-invoking word skill according to an intent creation instruction, wherein the intent creation instruction comprises at least one of: intent name, user statement;
and acquiring the original skill corpus based on the intention.
3. The method of claim 1, wherein the filling the original skill corpus with the replacement word based on the slot position to obtain a target skill corpus comprises:
and filling the groove into the original skill corpus by using the preset number of the replacement words to obtain at least one target skill corpus.
4. The method of claim 1, wherein in the event that the first test result is that the target skill corpus does not conflict, after completion of creation of the non-invoking word skills in a development platform, the method further comprises:
performing conflict testing on the target skill corpus in an application platform to obtain a second testing result;
performing standard auditing on the created non-call word skills under the condition that the second test result is that the target skill corpus is not in conflict;
and if the specification audit is passed, the non-calling word skills are online in an application platform.
5. The method of claim 4, wherein the non-calling word skills are prohibited from being brought up in the application platform if the second test result is that there is a conflict with the target skill corpus.
6. A skill creation apparatus, comprising:
the first acquisition module is used for acquiring original skill corpus, wherein the original skill corpus is corpus for waking up non-called word skills, and the non-called word skills are skills which do not need to be woken up by using called words;
the recognition module is used for recognizing at least one slot position of the skill corpus and acquiring a replacement word of each slot position;
the groove filling module is used for filling grooves in the original skill corpus by using the replacement words based on the groove positions to obtain target skill corpus;
the first test module is used for performing conflict test on the target skill corpus to obtain a first test result, and completing creation of the non-calling word skills in a development platform under the condition that the first test result is that the target skill corpus is not in conflict;
wherein the apparatus further comprises: the feedback module is used for feeding back a change request when the first test result shows that the target skill corpus has conflict, wherein the change request is used for prompting a user to edit the original skill corpus; receiving an editing completion instruction, acquiring an original skill corpus based on the editing completion instruction, and performing conflict test on the edited original skill corpus until the original skill corpus does not conflict;
wherein, the identification module includes:
the recognition unit is used for recognizing at least one slot position of the skill corpus;
the acquisition unit is used for acquiring the dictionary corresponding to each slot position;
and the extraction unit is used for extracting a preset number of replacement words from the dictionary corresponding to each slot position.
7. The apparatus of claim 6, wherein the apparatus further comprises:
the first creating module is used for starting to create the non-call word skills according to skill creating instructions, wherein the skill creating instructions comprise at least one of the following steps: skill name, skill type;
a second creation module for creating an intent of the non-call word skill according to an intent creation instruction, wherein the intent creation instruction includes at least one of: intent name, user statement;
and the second acquisition module is used for acquiring the original skill corpus based on the intention.
8. The apparatus of claim 6, wherein the slot filling module comprises:
and the groove filling unit is used for filling grooves in the original skill corpus by utilizing the preset number of the replacement words to obtain at least one target skill corpus.
9. The apparatus of claim 6, wherein the apparatus further comprises:
the second testing module is used for performing conflict testing on the target skill corpus in the application platform to obtain a second testing result;
the auditing module is used for carrying out standard auditing on the established non-call word skills under the condition that the second test result is that the target skill corpus has no conflict;
and the online module is used for online the non-call word skills in the application platform under the condition that the standard audit is passed.
10. The apparatus of claim 9, wherein the apparatus further comprises:
and the prohibition module is used for prohibiting the non-calling word skills from being online in the application platform under the condition that the second test result shows that the target skill corpus has conflict.
11. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program, when run, controls a device in which the computer-readable storage medium is located to perform the skill creation method of any of claims 1 to 5.
12. A processor for running a program, wherein the program when run performs the skill creation method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110706140.0A CN113553843B (en) | 2021-06-24 | 2021-06-24 | Skill creation method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110706140.0A CN113553843B (en) | 2021-06-24 | 2021-06-24 | Skill creation method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113553843A CN113553843A (en) | 2021-10-26 |
| CN113553843B true CN113553843B (en) | 2023-12-19 |
Family
ID=78130896
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110706140.0A Active CN113553843B (en) | 2021-06-24 | 2021-06-24 | Skill creation method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113553843B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8250071B1 (en) * | 2010-06-30 | 2012-08-21 | Amazon Technologies, Inc. | Disambiguation of term meaning |
| CN108491220A (en) * | 2018-03-15 | 2018-09-04 | 苏州思必驰信息科技有限公司 | Method of skill training and system |
| CN109697979A (en) * | 2018-12-25 | 2019-04-30 | Oppo广东移动通信有限公司 | Voice assistant technical ability adding method, device, storage medium and server |
| CN110491394A (en) * | 2019-09-12 | 2019-11-22 | 北京百度网讯科技有限公司 | Wake up the acquisition methods and device of corpus |
| CN110674259A (en) * | 2019-09-27 | 2020-01-10 | 北京百度网讯科技有限公司 | Intention understanding method and device |
| CN111309876A (en) * | 2018-12-12 | 2020-06-19 | 北京嘀嘀无限科技发展有限公司 | Service request processing method and device, electronic equipment and storage medium |
| CN112052316A (en) * | 2020-08-12 | 2020-12-08 | 深圳市欢太科技有限公司 | Model evaluation method, model evaluation device, storage medium and electronic equipment |
| CN112541070A (en) * | 2020-12-25 | 2021-03-23 | 北京百度网讯科技有限公司 | Method and device for excavating slot position updating corpus, electronic equipment and storage medium |
| CN112905869A (en) * | 2021-03-26 | 2021-06-04 | 北京儒博科技有限公司 | Adaptive training method and device for language model, storage medium and equipment |
-
2021
- 2021-06-24 CN CN202110706140.0A patent/CN113553843B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8250071B1 (en) * | 2010-06-30 | 2012-08-21 | Amazon Technologies, Inc. | Disambiguation of term meaning |
| CN108491220A (en) * | 2018-03-15 | 2018-09-04 | 苏州思必驰信息科技有限公司 | Method of skill training and system |
| CN111309876A (en) * | 2018-12-12 | 2020-06-19 | 北京嘀嘀无限科技发展有限公司 | Service request processing method and device, electronic equipment and storage medium |
| CN109697979A (en) * | 2018-12-25 | 2019-04-30 | Oppo广东移动通信有限公司 | Voice assistant technical ability adding method, device, storage medium and server |
| CN110491394A (en) * | 2019-09-12 | 2019-11-22 | 北京百度网讯科技有限公司 | Wake up the acquisition methods and device of corpus |
| CN110674259A (en) * | 2019-09-27 | 2020-01-10 | 北京百度网讯科技有限公司 | Intention understanding method and device |
| CN112052316A (en) * | 2020-08-12 | 2020-12-08 | 深圳市欢太科技有限公司 | Model evaluation method, model evaluation device, storage medium and electronic equipment |
| CN112541070A (en) * | 2020-12-25 | 2021-03-23 | 北京百度网讯科技有限公司 | Method and device for excavating slot position updating corpus, electronic equipment and storage medium |
| CN112905869A (en) * | 2021-03-26 | 2021-06-04 | 北京儒博科技有限公司 | Adaptive training method and device for language model, storage medium and equipment |
Non-Patent Citations (2)
| Title |
|---|
| 人工智能技术在信息服务中的应用;赵西林;王占峰;马维峰;;电子技术与软件工程(05);200-202 * |
| 语音技能在智慧矿山中的应用;曾健;《煤炭技术》;第39卷(第10期);155-158 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113553843A (en) | 2021-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108509477B (en) | Method for recognizing semantics, electronic device and computer readable storage medium | |
| CN106528613B (en) | Intelligent answer method and device | |
| EP1677197A1 (en) | Test case automatic generation method for testing proactive GSM applications on SIM card | |
| CN109194689B (en) | Abnormal behavior recognition method, device, server and storage medium | |
| CN108664665A (en) | Data format method for transformation, device, equipment and readable storage medium storing program for executing | |
| CN111768781A (en) | Voice interruption processing method and device | |
| CN106909656A (en) | Obtain the method and device of Text Feature Extraction model | |
| CN110413753A (en) | The extended method and device of question and answer sample | |
| CN106933811A (en) | A kind of entry automatic generation method and device | |
| CN102982217B (en) | The treating method and apparatus of silk-screen | |
| CN110427470A (en) | Question and answer processing method, device and electronic equipment | |
| CN105959451B (en) | A kind of method and mobile terminal loading MBN | |
| CN113553843B (en) | Skill creation method and device | |
| CN110362826A (en) | Periodical submission method, equipment and readable storage medium storing program for executing based on artificial intelligence | |
| CN110990381B (en) | Processing method and device of server, storage medium and electronic device | |
| CN110070332A (en) | Interview method, apparatus, equipment and readable storage medium storing program for executing based on artificial intelligence | |
| CN111563037B (en) | Test optimization method and device based on test questions, storage medium and terminal | |
| CN113299294A (en) | Task type dialogue robot interaction method, device, equipment and storage medium | |
| CN104023124A (en) | Method and device for automatically identifying and extracting a name in short message | |
| CN116185853A (en) | Code verification method and device | |
| CN110516258A (en) | Data verification method and device, storage medium, electronic device | |
| CN114579767A (en) | Causal relationship extraction method and device, electronic equipment and readable storage medium | |
| CN112784024B (en) | Man-machine conversation method, device, equipment and storage medium | |
| CN115098665A (en) | Method, device and equipment for expanding session data | |
| CN108959049A (en) | The robustness of SMM and test method, device and the storage medium of stability |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |