CN108062955A - A kind of intelligence report-generating method, system and equipment - Google Patents
A kind of intelligence report-generating method, system and equipment Download PDFInfo
- Publication number
- CN108062955A CN108062955A CN201711318355.5A CN201711318355A CN108062955A CN 108062955 A CN108062955 A CN 108062955A CN 201711318355 A CN201711318355 A CN 201711318355A CN 108062955 A CN108062955 A CN 108062955A
- Authority
- CN
- China
- Prior art keywords
- talks
- text
- report
- content
- talking voice
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000013507 mapping Methods 0.000 claims description 15
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000004590 computer program Methods 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 5
- 238000012549 training Methods 0.000 claims description 5
- 230000000875 corresponding effect Effects 0.000 description 48
- 239000000463 material Substances 0.000 description 7
- 238000004321 preservation Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 230000006399 behavior Effects 0.000 description 4
- 238000003780 insertion Methods 0.000 description 4
- 230000037431 insertion Effects 0.000 description 4
- 230000001960 triggered effect Effects 0.000 description 3
- 239000000470 constituent Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 238000007792 addition Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000012217 deletion Methods 0.000 description 1
- 230000037430 deletion Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
-
- G—PHYSICS
- G11—INFORMATION STORAGE
- G11B—INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORD CARRIER AND TRANSDUCER
- G11B27/00—Editing; Indexing; Addressing; Timing or synchronising; Monitoring; Measuring tape travel
- G11B27/10—Indexing; Addressing; Timing or synchronising; Measuring tape travel
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Telephonic Communication Services (AREA)
- Machine Translation (AREA)
Abstract
The invention discloses a kind of intelligent report-generating method, system and equipment, wherein this method to include：It is recorded to obtain corresponding meeting talking voice during extraneous talks；Text resolution is carried out to meeting talking voice and obtains corresponding talks text；Report name is obtained, obtains the talks report of entitled report name, content including talks text.In above-mentioned technical proposal disclosed by the embodiments of the present invention, it is recorded to obtain corresponding meeting talking voice during talking in the external world, and then text resolution is carried out to the meeting talking voice and obtains corresponding talks text, it ultimately forms with the talks report of report name and content including talks text, without record personnel's manual record conversation content and the corresponding report of generation, but the record of conversation content is realized and to notifiable generation by recording and text resolution, the report thus generated avoids the appearance of situations such as loss conversation content occurred during manual record or record error, it can completely and accurately embody conversation content.
Description
Technical field
The present invention relates to speech recognition and text analysis technique field, more specifically to a kind of intelligent report generation
Method, system and equipment.
Background technology
During people carry out the important talks such as meeting or talk, it is often necessary to which conversation content is recorded simultaneously
Corresponding report is generated, for subsequent query and use.It is typically to arrange have special record personnel to be responsible for conversation content at present
Record and to notifiable generation, but since speech speed is above the writing speed of record personnel, so being easy to
Record personnel occur can not record and then lost part conversation content and that record is caused to occur is wrong due to conversation content of mishearing in time
Situation, the report for ultimately resulting in generation lack part conversation content and not accurate enough by mistake.
In conclusion how to provide a kind of technical side for the report generation that can completely and accurately embody conversation content
Case is a problem to be solved by those skilled in the art.
The content of the invention
The object of the present invention is to provide a kind of intelligent report-generating method, system and equipment so that the report of generation can
Embody completely and accurately conversation content.
To achieve these goals, the present invention provides following technical solution：
A kind of intelligence report-generating method, including：
It is recorded to obtain corresponding meeting talking voice during extraneous talks；
Text resolution is carried out to the meeting talking voice and obtains corresponding talks text；
Report name is obtained, obtains the entitled report name, content includes the talks report of the talks text.
Preferably, after obtaining the talks text, before obtaining the talks report, further include：
The meeting talking voice and the talks text are divided into more one's shares of expenses for a joint undertaking in units of prefixed time interval can talking voices
And son talks text, using crowdsourcing mode using manually based on the son can talking voice text of talking to corresponding son repair
Just, and by revised sub- talks text combination revised talks text is obtained.
Preferably, after obtaining revised talks text, further include：
Using the meeting talking voice and revised talks text as training corpus to the model of realization talks speech analysis
Incremental learning is carried out, to utilize the parsing for carrying out the model realization after incremental learning and talking voice.
Preferably, after obtaining the talks text, before obtaining the talks report, further include：
The talks text is directed into editable region；
It determines that there is associated related information with the talks text in the pre-stored information of database, and is touched in the external world
It gives and is inserted into the external world in the talks text by the content chosen in the related information.
Preferably, determine that there is associated related information, bag with the talks text in the pre-stored information of database
It includes：
Entity, attribute and event recognition are carried out to the talks text, obtain entity, the attribute for including the talks text
And the entity tuple of event；
It the entity tuple is configured to subgraph inputs carry out subgraph match into the pre-stored knowledge mapping of database,
If matching obtained similarity is more than default similarity threshold, it is determined that successful match, and to successful match in knowledge mapping
Information be ranked up and show according to similarity is descending as related information.
Preferably, the information of successful match in the knowledge mapping is shown, including：
Using floating, bullet box form shows the information of successful match in the knowledge mapping.
Preferably, under extraneous trigger by the external world by the related information in the content chosen be inserted into the talks text
In after, further include：
By between the talks text and the extraneous content chosen subgraph match pattern, the talks text and described outer
The content record that boundary chooses is to record set.
Preferably, after obtaining the talks report of the entitled report name, content including the talks text, also wrap
It includes：
The content included under extraneous trigger to the talks report carries out text editing processing；
Topic model is combined based on the expression study to the talks report and generates corresponding keyword, and by the keyword
It adds in into the talks report；
After the keyword is added in into the talks report, further include：
The talks report is saved as into template, is used for transferring the template as talks report template.
A kind of intelligence report preparing system, including：
Recording module is used for：It is recorded to obtain corresponding meeting talking voice during extraneous talks；
Parsing module is used for：Text resolution is carried out to the meeting talking voice and obtains corresponding talks text；
Report generation module, is used for：Report name is obtained, obtains the entitled report name, content includes the meeting
Talk the talks report of text.
A kind of intelligence report generation equipment, including：
Memory, for storing computer program；
Processor realizes the step of any one of the as above intelligent report-generating method during for performing the computer program
Suddenly.
The present invention provides a kind of intelligent report-generating method, system and equipment, wherein this method to include：In extraneous meeting
It is recorded to obtain corresponding meeting talking voice during talking；Text resolution is carried out to the meeting talking voice and obtains corresponding talks text
This；Report name is obtained, obtains the entitled report name, content includes the talks report of the talks text.The present invention
In above-mentioned technical proposal disclosed in embodiment, recorded to obtain corresponding meeting talking voice during talking in the external world, so it is right
The meeting talking voice carries out text resolution and obtains corresponding talks text, and ultimately form includes talks text with report name and content
Talks report, without record personnel's manual record conversation content and the corresponding report of generation, but pass through recording and text
Parsing realizes the record of conversation content and to notifiable generation, and the report thus generated avoids losing of occurring during manual record
The appearance for situations such as losing conversation content or record error, can completely and accurately embody conversation content.
Description of the drawings
It in order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention, for those of ordinary skill in the art, without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of flow chart of intelligent report-generating method provided in an embodiment of the present invention；
Fig. 2 is a kind of structure diagram of intelligent report preparing system provided in an embodiment of the present invention.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present invention, the technical solution in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, those of ordinary skill in the art are obtained every other without making creative work
Embodiment belongs to the scope of protection of the invention.
It, can be with referring to Fig. 1, it illustrates a kind of flow chart of intelligent report-generating method provided in an embodiment of the present invention
Including：
S11：It is recorded to obtain corresponding meeting talking voice during extraneous talks.
During extraneous talks, the voice that can be in the process generated to talks be recorded and then be obtained corresponding meeting
Talking voice, specifically, the corresponding record button of recording, recording can be provided with by realizing the system of the above method disclosed in the present application
Button original state is " starting to record ", and the external world proceeds by recording after clicking on " starting to record ", and the record button becomes " temporarily
Stop recording "；The external world clicks on " pause recording " and stops this time recording afterwards, and the record button becomes " starting to record ", and correspondence obtains
Recording be can talking voice.It may alternatively be arranged to voice and automatically save to attend the meeting talking voice afterwards automatically to the voice
It is preserved；It may be arranged as voice artificially to preserve, i.e., above system can be provided with " preservation " button, so as to start to record
The voice for obtaining recording is clicked on after sound and after the completion of having recorded by the external world to preserve.In addition to talks during
Recording can be real-time, as long as needing to record, then real-time implementation, which records and obtains current time correspondence, understands talking voice.
It should be noted that the corresponding waveform that can also will be recorded with time shaft in Recording Process is shown, for the external world
It obtains.
S12：Text resolution is carried out to meeting talking voice and obtains corresponding talks text.
Above method correspondence system provided by the invention is also provided with " speech analysis " button, so as to be clicked on when extraneous
Text resolution is carried out to the meeting talking voice that recording obtains during the button；Automatic real time parsing is may be arranged as, that is, is recorded
During to recording gained can talking voice carry out real-time text parsing, within protection scope of the present invention.In addition right
Can talking voice parsed during can also show the progress bar of parsing so that extraneous staff can by this into
Degree item knows the progress of speech analysis.It should be noted that as need to be only preserved meeting talking voice without carrying out any text in the application
This parsing can also click on pre-set " wouldn't parse, only preserve recording " button or above-mentioned " preservation " button by the external world
Meeting talking voice is preserved, and whole voice corresponding lists of preservation are updated accordingly, facilitates looking into for staff
Inquire about use.
In addition to that can also be divided into not according to the pause of meeting talking voice during talking voice progress real time parsing
Paragraph together makes correspondence markings, and the time which can talk and be kept to determine non-someone in Recording Process is not
Less than the pause of the default pause duration previously according to actual needs setting, i.e., the voice between two pauses corresponds to text can be with
The pre-set mark that pauses is added in for a paragraph or in each stall position；It can also be according to teller in meeting talking voice
The change division paragraph of object makes correspondence markings, you can tone color, frequency during with by talking to personage etc. judge
To distinguish different personages, so as to which the voice of same personage is corresponded to text as a paragraph or is replacing personage every time
Preset mark of substituting is added in during speech, adds in tone color, the frequency for substituting and marking and being specifically as follows when talking according to personage
The identification informations such as one number or title are set etc. the number of person for determining participation talks, and to each personage, thus at this
The identification information of the personage is added during a personage's speech before it corresponds to text.Pause and personage can certainly be considered more
It changes, text conduct will be corresponded to while substituting mark per voice between pausing twice by such as adding in correspond to when replacing personage's speech every time
One paragraph, etc. within protection scope of the present invention.
S13：Report name is obtained, obtains the talks report of entitled report name, content including talks text.
It should be noted that this report title can edit report for what is popped up during user's click " preservation " button for extraneous
The dialog box of title can be previously provided with Default Reports title, such as " XX companies "；Report name can also pass through intelligent association
By selecting the report name that there is the higher degree of association with text of currently talking in database report name stored in advance
For current report name, the wherein degree of association can be that the number that report name is appeared in talks text more at most proves to correspond to
The degree of association is more high.By the above-mentioned means, obtain that there is the talks of report name and report content report, for subsequently using and looking into
Ask etc..
In above-mentioned technical proposal disclosed by the embodiments of the present invention, recorded to obtain corresponding meeting during talking in the external world
Talking voice, and then text resolution is carried out to the meeting talking voice and obtains corresponding talks text is ultimately formed with report name and interior
The talks that holding includes talking text are reported, without record personnel's manual record conversation content and the corresponding report of generation, but
The record of conversation content is realized and to notifiable generation by recording and text resolution, the report thus generated avoids manually
The appearance of situations such as loss conversation content occurred during record or record error, can completely and accurately embody in talks
Hold.
A kind of intelligent report-generating method provided in an embodiment of the present invention, after obtaining talks text, obtains talks report
Before, can also include：
Will talking voice and talks text more one's shares of expenses for a joint undertaking are divided into units of prefixed time interval can talking voices and son talks
Text is modified corresponding sub- talks text based on sub- meeting talking voice using artificial using the mode of crowdsourcing, and will be after amendment
Son talks text combination obtain revised talks text.
Wherein prefixed time interval can be set according to actual needs, such as 10 minutes, it will talking voice and talks text
This is divided into more one's shares of expenses for a joint undertaking meeting talking voices and corresponding more one's share of expenses for a joint undertaking talks texts, neutron talks language in units of prefixed time interval
Sound is corresponded with son talks text and son meeting talking voice in correspondence with each other is identical with the period belonging to son talks text.It uses
The mode of crowdsourcing not only ensure that meeting after amendment using being manually modified based on son meeting talking voice to corresponding sub- talks text
The accuracy of text is talked, and important information leakage is also effectively prevented due to the use of the mode of crowdsourcing.Specifically, using crowd
The mode of bag is modified and can include to corresponding sub- talks text using son meeting talking voice is manually based on：The sub- meeting that will be obtained
Talking voice and corresponding son talks text are sent to different terminals, and (per one's share of expenses for a joint undertaking, meeting talking voice and corresponding son talks text correspond to one
Terminal), each terminal counterpart staff is carried out manually based on the son talks text that the son meeting talking voice docking received receives
Intervene and correct, such as by the content of son talks recording that manually listens back to receive and corresponding sub text of talking content whether phase
Together, and to different places the content that sub- talks are recorded of being subject to makes amendment, can retain on sub- talks text modified
Record directly records modified process in other memory space.
A kind of intelligent report-generating method provided in an embodiment of the present invention, after obtaining revised talks text, may be used also
To include：
Using meeting talking voice and revised talks text as training corpus to realizing that the model of talks speech analysis carries out
Incremental learning, with using carry out the model realization after incremental learning can talking voice parsing.Due to revised talks text phase
For more accurate before correcting, be more in line with it is corresponding understand talking voice, therefore can will talking voice and revised talks text
As training corpus to realizing that the model of speech analysis carries out incremental learning, to pass through increasing when subsequently needing and realizing speech analysis
Model realization after amount study, improves the accuracy rate that voice parses automatically.
A kind of intelligent report-generating method provided in an embodiment of the present invention, after obtaining talks text, obtains talks report
Before, can also include：
Talks text is directed into editable region；
Determine that there is associated related information in the pre-stored information of database with talks text, and under external world's triggering
The external world is inserted by the content chosen in related information in talks text.
The editable region as taken down notes column etc is also provided in above method correspondence system disclosed in the present application, it can
Corresponding text is directed into editable region during clicking on talks text entirety or wherein certain part in the external world (to be specifically as follows
Editable region end), the text is directed into editable region automatically after generation talks text can also be arranged to, at this
Within the protection domain of invention.And codified region is as a text edit tool, possess basic font, paragraph, color,
The text editings functions such as mark.Corresponding information is stored in pre-set database, selects the external world under being triggered in the external world
Information in the database taken is inserted into the extraneous position specified in text of talk, specifically can click on that " insertion is pressed by extraneous
Button " realizes selection and insertion of the information for needing to be inserted into etc.；It can select with talks text to have in database with Automatic sieve and close
The related information of connection, for the external world check or therefrom choose need be inserted into talks text information and the external world trigger under will
The information is inserted into talks text, such as one section of voice include work as predecessor company shareholder, then can using shareholder's corresponding informance as
Related information is chosen for extraneous.So as to accelerate accurately generating for report by this intelligent association prompt facility.
A kind of intelligent report-generating method provided in an embodiment of the present invention, determines participant in the pre-stored information of database
Talking text has associated related information, can include：
Entity, attribute and event recognition are carried out to talks text, obtain the entity, attribute and thing for including the talks text
The entity tuple of part；
It entity tuple is configured to subgraph inputs carry out subgraph match into the pre-stored knowledge mapping of database, if
It matches obtained similarity and is more than default similarity threshold, it is determined that successful match, and to the letter of successful match in knowledge mapping
Breath is ranked up and shows according to similarity is descending as related information.
To talks text carry out dynamic realtime parsing, including being segmented to it, part of speech identification, name Entity recognition and
Syntax dependency parsing, so as to obtain the entity included and the constituent structure in text, wherein constituent structure is included in talks text
Entity attributes and event, specifically, the definition of the entity in the application is self-existent financial instrument activity participant,
Including enterprise and investor etc.；And entity attribute is that the static state of entity elements is expressed, and is illustrated by taking enterprise as an example, enterprise
Attribute includes industrial and commercial information, industry attribute, affiliated enterprise, individual, critical behavior people and financial attribute etc., the wherein key of enterprise
Include actual controller during behavior is humanized, actual controller can include the shareholder of enterprise and be not enterprise shareholder but
It is the personnel that corporate behavior can be dominated by investment relation, agreement or other arrangements；And event is then the corresponding thing of entity
Part is illustrated by taking enterprise as an example, and the event of entity is the set of enterprise dynamic behavior, is draped over one's shoulders including financing, business activities, information
Dew etc., event also have corresponding attribute, such as the time of event, place, personage, the origin of an incident, process, result and influence in itself；
So as to by identifying the entity, attribute, concept and the event that are included in text of talking, obtain comprising entity, category in the talks text
Property and the tuple form of event, i.e. entity tuple；Entity tuple is constructed of the knowledge graph stored in subgraph input database
Subgraph match is carried out in spectrum, if matched similarity is more than the default similarity threshold previously according to actual needs setting,
Think to match content, and then the content that will match to is ranked up and shows according to similarity is descending as related information
Show, multiple subgraphs be wherein stored in knowledge mapping, by each subgraph in the subgraph and knowledge mapping of entity tuple successively into
Row compares, and the similarity compared is more than default similarity threshold and then illustrates successful match.So as to facilitate extraneous check
And it uses.It is further to note that it carries out generally can first carrying out the fuzzy matching of subgraph during subgraph match so as to generate
Multiple multiple candidate's subgraphs corresponding with the correspondings subgraph of talks text, it is then determined that text of talking corresponds to subgraph and each candidate's subgraph
In each node and attributes similarity, and determine that the similarity is highest and subgraph candidate's subgraph higher than default similarity threshold is
The subgraph of final successful match.
A kind of intelligent report-generating method provided in an embodiment of the present invention, carries out the information of successful match in knowledge mapping
It has been shown that, can include：
Using floating, bullet box form shows the information of successful match in knowledge mapping.
The display carried out in the application for the information of successful match can play box form realization using floating, specific next
Say, the entity part in text of talking can be highlighted after successful match is determined, and will need the information that shows with
The bullet box form that floats is shown, and the bullet frame that floats can be defaulted as close pattern, when the external world chooses highlighted entity (such as to move mouse
Move to highlighted entity) then show the floating bullet frame.The content for playing frame of floating is closed under being triggered after user browses in the external world,
The information is added in into talks text after any information can also be chosen in the external world.So as to be not only convenient for user to relevant information
Check and be inserted into, additionally it is possible to avoid excessively occupying page resource.
A kind of intelligent report-generating method provided in an embodiment of the present invention, by the external world by related information under being triggered in the external world
After the content of selection is inserted into talks text, it can also include：
By subgraph match pattern, talks text and the extraneous content chosen between text and the extraneous content chosen of talking
It records to record set.
After the external world, which selects, to be needed to add in the content of talks text, can will talk text, the extraneous content chosen
And subgraph match mode record is corresponded to record set, wherein the subgraph match pattern between the extraneous content chosen and talks text
It can include that the extraneous content chosen correspond to subgraph, text of talking corresponds to similarity between subgraph and two subgraphs；Thus when
It, can be by searching out subgraph match corresponding with the talks text in record set when needing to match again for certain talks text
If the subgraph match pattern searched out is multiple, it is outer to choose the highest subgraph match pattern correspondence of wherein similarity for pattern
The content that boundary chooses is for content corresponding with the talks text and by carrying out showing etc. that mode returns to extraneous user.Its neutron
The searching of figure match pattern can represent that study, similarity calculation and sorting technique etc. are real by relevant in deep learning
It is existing, it is not specifically limited herein.
A kind of intelligent report-generating method provided in an embodiment of the present invention, obtains entitled report name, content includes meeting
After the talks report for talking text, it can also include：
The content included under extraneous trigger to talks report carries out text editing processing；
Topic model is combined based on the expression study to talks report and generates corresponding keyword, and the keyword is added in
In being reported to talks.
After obtaining talks report, basic text editing capability is supported in talks report, can be carried out additions and deletions and be changed, adjusts respectively
The text editings processing such as kind form, the report form for needing its final adjustment for oneself so as to user.Furthermore it is also possible to pass through
The expression study of report name and content of report of talking combines topic model and generates corresponding keyword, as industry, industry and
The dimensions such as region, and these keywords are added in into talks report, user is facilitated to realize looking into for report based on these keywords
It looks for and classifies.
A kind of intelligent report-generating method provided in an embodiment of the present invention, after keyword is added in talks in reporting,
It can also include：
Talks report is saved as into template, is used for transferring the template as talks report template.
Talks report is saved as into template, system can also preserve the template of one or more typical standard in itself,
So as to call the template of any preservation when generation talks is needed to report, intelligence carries out content abundance, quickly forms new report.
The template of the talks report wherein preserved only includes the form of talks report, such as typesetting, bibliographic structure, does not include talks text
The substantive contents such as this form new talks report with the other content that is put into template.
It can in addition contain receive and preserve the notes of extraneous input or text, so as to need to be inserted into content to text of talking
This when, can realize insertion for the external world by choosing corresponding content in the notes preserved or text.
The application supports user to import notes or text, and as the candidate material of generation report, system is according to stored
Template in bibliographic structure and candidate material bibliographic structure, template in paragraph and candidate material paragraph similarity, establish mould
Correspondence in plate in catalogue and candidate material in catalogue, template in paragraph and candidate material between paragraph, by candidate's material
The catalogue and paragraph of material are added in into template at corresponding position, so as to carry out report generation, adjustment format automatically.
The embodiment of the present invention additionally provides a kind of intelligent report preparing system, as shown in Fig. 2, can include：
Recording module 11, is used for：It is recorded to obtain corresponding meeting talking voice during extraneous talks；
Parsing module 12, is used for：Text resolution is carried out to meeting talking voice and obtains corresponding talks text；
Report generation module 13, is used for：Report name is obtained, obtains entitled report name, content includes talks text
Talks report.
A kind of intelligent report preparing system provided in an embodiment of the present invention, can also include：
Correcting module is used for：Obtain talks text after, obtain talks report before, it will talking voice and talks text
More one's shares of expenses for a joint undertaking meeting talking voices and son talks text are divided into units of prefixed time interval, artificial base is utilized using the mode of crowdsourcing
In the son can talking voice corresponding son talks text is modified, and after revised sub- talks text combination is obtained amendment
Talks text.
A kind of intelligent report preparing system provided in an embodiment of the present invention, can also include：
Study module is used for：After obtaining revised talks text, made with meeting talking voice and revised talks text
Be training corpus to realizing that the model of talks speech analysis carries out incremental learning, with using carrying out the model realization after incremental learning
The parsing of meeting talking voice.
A kind of intelligent report preparing system provided in an embodiment of the present invention, can also include：
Module is inserted into, is used for：Obtain talks text after, obtain talks report before, it will talk text be directed into and can compile
Collect region；Determine that there is associated related information in the pre-stored information of database with talks text, and under external world's triggering
The external world is inserted by the content chosen in related information in talks text.
A kind of intelligent report preparing system provided in an embodiment of the present invention, insertion module can include：
Matching unit is used for：Entity, attribute and event recognition are carried out to talks text, obtain including the talks text
Entity, the entity tuple of attribute and event；Entity tuple is configured to subgraph to input to the pre-stored knowledge graph of database
Subgraph match is carried out in spectrum, if the similarity that matching obtains is more than default similarity threshold, it is determined that successful match, and to knowing
The information for knowing successful match in collection of illustrative plates is ranked up and shows according to similarity is descending as related information.
A kind of intelligent report preparing system provided in an embodiment of the present invention, matching unit can include：
It shows subelement, is used for：Using floating, bullet box form shows the information of successful match in knowledge mapping.
A kind of intelligent report preparing system provided in an embodiment of the present invention, can also include：
Logging modle is used for：Under extraneous trigger by the external world by related information in the content chosen be inserted into talks text
In after, it will talk subgraph match pattern, talks text and the extraneous content chosen between text and the extraneous content chosen
It records to record set.
A kind of intelligent report preparing system provided in an embodiment of the present invention, can also include：
Editor module is used for：After obtaining the talks report of entitled report name, content including talks text, outside
The content included under boundary's triggering to talks report carries out text editing processing；
Keyword-extraction module is used for：It obtains the talks of entitled report name, content including talks text and reports it
Afterwards, topic model is combined based on the expression study to talks report and generates corresponding keyword, and the keyword is added in meeting
It talks in report.
A kind of intelligent report preparing system provided in an embodiment of the present invention, can also include：
Preserving module is used for：After keyword is added in talks in reporting, it will talk report and save as template, for adjusting
The template is taken to be used as talks report template.
The embodiment of the present invention additionally provides a kind of intelligent report generation equipment, can include：
Memory, for storing computer program；
Processor realizes the step of any one of the as above intelligent report-generating method during for performing the computer program
Suddenly.
The embodiment of the present invention additionally provides a kind of computer readable storage medium, is deposited on the computer readable storage medium
Computer program is contained, the as above any one intelligent report generation can be realized when the computer program is executed by processor
The step of method.
It is related in a kind of intelligent report preparing system provided in an embodiment of the present invention, equipment and computer readable storage medium
Partial explanation refers to the detailed description of corresponding part in a kind of intelligent report-generating method provided in an embodiment of the present invention,
This is repeated no more.In addition in above-mentioned technical proposal provided in an embodiment of the present invention original is realized with corresponding to technical solution in the prior art
Consistent part and unspecified is managed, in order to avoid excessively repeat.
The foregoing description of the disclosed embodiments enables those skilled in the art to realize or use the present invention.To this
A variety of modifications of a little embodiments will be apparent for a person skilled in the art, and the general principles defined herein can
Without departing from the spirit or scope of the present invention, to realize in other embodiments.Therefore, the present invention will not be limited
The embodiments shown herein is formed on, and is to fit to consistent with the principles and novel features disclosed herein most wide
Scope.
Claims (10)
1. a kind of intelligence report-generating method, which is characterized in that including：
It is recorded to obtain corresponding meeting talking voice during extraneous talks；
Text resolution is carried out to the meeting talking voice and obtains corresponding talks text；
Report name is obtained, obtains the entitled report name, content includes the talks report of the talks text.
2. according to the method described in claim 1, it is characterized in that, after obtaining the talks text, the talks report is obtained
Before reporting to, further include：
The meeting talking voice and the talks text are divided into more one's shares of expenses for a joint undertaking in units of prefixed time interval can talking voices and son
Talks text is utilized using the mode of crowdsourcing and is manually modified based on the son meeting talking voice to corresponding to sub- talks text, and
Revised sub- talks text combination is obtained into revised talks text.
3. according to the method described in claim 2, it is characterized in that, after obtaining revised talks text, further include：
Using the meeting talking voice and revised talks text as training corpus to realizing that the model of talks speech analysis carries out
Incremental learning, with using carry out the model realization after incremental learning can talking voice parsing.
4. according to the method described in claim 1, it is characterized in that, after obtaining the talks text, the talks report is obtained
Before reporting to, further include：
The talks text is directed into editable region；
Determine that there is associated related information with the talks text in the pre-stored information of database, and under external world's triggering
The external world is inserted by the content chosen in the related information in the talks text.
5. according to the method described in claim 4, it is characterized in that, determine in database pre-stored information with the talks
Text has associated related information, including：
Entity, attribute and event recognition are carried out to the talks text, obtain the entity, attribute and thing for including the talks text
The entity tuple of part；
It the entity tuple is configured to subgraph inputs carry out subgraph match into the pre-stored knowledge mapping of database, if
It matches obtained similarity and is more than default similarity threshold, it is determined that successful match, and to the letter of successful match in knowledge mapping
Breath is ranked up and shows according to similarity is descending as related information.
6. according to the method described in claim 5, it is characterized in that, the information of successful match in the knowledge mapping is shown
Show, including：
Using floating, bullet box form shows the information of successful match in the knowledge mapping.
7. according to the method described in claim 5, it is characterized in that, the external world trigger under by the external world by the related information in select
After the content taken is inserted into the talks text, further include：
Subgraph match pattern, the talks text and the external world between the talks text and the extraneous content chosen is selected
The content record taken is to record set.
8. according to the method described in claim 1, it is characterized in that, the entitled report name, content are obtained including described
It talks after the talks report of text, further includes：
The content included under extraneous trigger to the talks report carries out text editing processing；
Topic model is combined based on the expression study to the talks report and generates corresponding keyword, and the keyword is added in
Into the talks report；
After the keyword is added in into the talks report, further include：
The talks report is saved as into template, is used for transferring the template as talks report template.
9. a kind of intelligence report preparing system, which is characterized in that including：
Recording module is used for：It is recorded to obtain corresponding meeting talking voice during extraneous talks；
Parsing module is used for：Text resolution is carried out to the meeting talking voice and obtains corresponding talks text；
Report generation module, is used for：Report name is obtained, obtains the entitled report name, content includes the talks text
This talks report.
10. a kind of intelligence report generation equipment, which is characterized in that including：
Memory, for storing computer program；
Processor realizes the intelligence report generation side as described in any one of claim 1 to 8 during for performing the computer program
The step of method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711318355.5A CN108062955B (en) | 2017-12-12 | 2017-12-12 | Intelligent report generation method, system and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711318355.5A CN108062955B (en) | 2017-12-12 | 2017-12-12 | Intelligent report generation method, system and equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108062955A true CN108062955A (en) | 2018-05-22 |
| CN108062955B CN108062955B (en) | 2020-12-11 |
Family
ID=62138309
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711318355.5A Active CN108062955B (en) | 2017-12-12 | 2017-12-12 | Intelligent report generation method, system and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108062955B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109766433A (en) * | 2018-12-06 | 2019-05-17 | 湖南科创信息技术股份有限公司 | Automated cataloging method and system, computer readable storage medium |
| CN109918628A (en) * | 2019-01-17 | 2019-06-21 | 平安科技（深圳）有限公司 | Method, device, device and storage medium for report generation based on speech recognition |
| CN110457424A (en) * | 2019-06-28 | 2019-11-15 | 谭浩 | Generate method, computer readable storage medium and the terminal device of interview report |
| CN110543559A (en) * | 2019-06-28 | 2019-12-06 | 谭浩 | Method for generating interview report, computer-readable storage medium and terminal device |
| CN110751950A (en) * | 2019-10-25 | 2020-02-04 | 武汉森哲地球空间信息技术有限公司 | Police conversation voice recognition method and system based on big data |
| CN111368515A (en) * | 2020-03-02 | 2020-07-03 | 中国农业科学院农业信息研究所 | Method and system for generating industry dynamic interactive report based on PDF document fragmentation |
| CN111507758A (en) * | 2020-04-09 | 2020-08-07 | 深圳传世智慧科技有限公司 | Semantic analysis-based investigation method, device, system and server |
| CN112862431A (en) * | 2021-01-19 | 2021-05-28 | 中国建设银行股份有限公司 | Method and device for generating conversation record, electronic equipment and readable storage medium |
| CN113421572A (en) * | 2021-06-23 | 2021-09-21 | 平安科技（深圳）有限公司 | Real-time audio conversation report generation method and device, electronic equipment and storage medium |
| CN114780733A (en) * | 2021-12-31 | 2022-07-22 | 海南大学 | DIKW atlas-based intelligent patent modification method, auxiliary response method and system |
| CN115186658A (en) * | 2022-06-20 | 2022-10-14 | 广东电网有限责任公司广州供电局 | Feasibility study report text processing method and device, and computer equipment |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105810207A (en) * | 2014-12-30 | 2016-07-27 | 富泰华工业（深圳）有限公司 | Meeting recording device and method thereof for automatically generating meeting record |
| CN105810206A (en) * | 2014-12-30 | 2016-07-27 | 富泰华工业（深圳）有限公司 | Meeting recording device and method thereof for automatically generating meeting record |
| CN105845129A (en) * | 2016-03-25 | 2016-08-10 | 乐视控股（北京）有限公司 | Method and system for dividing sentences in audio and automatic caption generation method and system for video files |
| CN106057193A (en) * | 2016-07-13 | 2016-10-26 | 深圳市沃特沃德股份有限公司 | Conference record generation method based on telephone conference and device |
| US20170004821A1 (en) * | 2014-10-30 | 2017-01-05 | Kabushiki Kaisha Toshiba | Voice synthesizer, voice synthesis method, and computer program product |
| US20170046331A1 (en) * | 2015-08-13 | 2017-02-16 | International Business Machines Corporation | Generating structured meeting reports through semantic correlation of unstructured voice and text data |
| CN106847256A (en) * | 2016-12-27 | 2017-06-13 | 苏州帷幄投资管理有限公司 | A kind of voice converts chat method |
| CN107147564A (en) * | 2017-05-09 | 2017-09-08 | 胡巨鹏 | Real-time speech recognition error correction system and identification error correction method based on cloud server |
-
2017
- 2017-12-12 CN CN201711318355.5A patent/CN108062955B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170004821A1 (en) * | 2014-10-30 | 2017-01-05 | Kabushiki Kaisha Toshiba | Voice synthesizer, voice synthesis method, and computer program product |
| CN105810207A (en) * | 2014-12-30 | 2016-07-27 | 富泰华工业（深圳）有限公司 | Meeting recording device and method thereof for automatically generating meeting record |
| CN105810206A (en) * | 2014-12-30 | 2016-07-27 | 富泰华工业（深圳）有限公司 | Meeting recording device and method thereof for automatically generating meeting record |
| US20170046331A1 (en) * | 2015-08-13 | 2017-02-16 | International Business Machines Corporation | Generating structured meeting reports through semantic correlation of unstructured voice and text data |
| CN105845129A (en) * | 2016-03-25 | 2016-08-10 | 乐视控股（北京）有限公司 | Method and system for dividing sentences in audio and automatic caption generation method and system for video files |
| CN106057193A (en) * | 2016-07-13 | 2016-10-26 | 深圳市沃特沃德股份有限公司 | Conference record generation method based on telephone conference and device |
| CN106847256A (en) * | 2016-12-27 | 2017-06-13 | 苏州帷幄投资管理有限公司 | A kind of voice converts chat method |
| CN107147564A (en) * | 2017-05-09 | 2017-09-08 | 胡巨鹏 | Real-time speech recognition error correction system and identification error correction method based on cloud server |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109766433B (en) * | 2018-12-06 | 2023-05-09 | 湖南科创信息技术股份有限公司 | Automatic cataloging method and system and computer readable storage medium |
| CN109766433A (en) * | 2018-12-06 | 2019-05-17 | 湖南科创信息技术股份有限公司 | Automated cataloging method and system, computer readable storage medium |
| CN109918628A (en) * | 2019-01-17 | 2019-06-21 | 平安科技（深圳）有限公司 | Method, device, device and storage medium for report generation based on speech recognition |
| CN110457424A (en) * | 2019-06-28 | 2019-11-15 | 谭浩 | Generate method, computer readable storage medium and the terminal device of interview report |
| CN110543559A (en) * | 2019-06-28 | 2019-12-06 | 谭浩 | Method for generating interview report, computer-readable storage medium and terminal device |
| CN110751950A (en) * | 2019-10-25 | 2020-02-04 | 武汉森哲地球空间信息技术有限公司 | Police conversation voice recognition method and system based on big data |
| CN111368515A (en) * | 2020-03-02 | 2020-07-03 | 中国农业科学院农业信息研究所 | Method and system for generating industry dynamic interactive report based on PDF document fragmentation |
| CN111507758A (en) * | 2020-04-09 | 2020-08-07 | 深圳传世智慧科技有限公司 | Semantic analysis-based investigation method, device, system and server |
| CN111507758B (en) * | 2020-04-09 | 2023-10-31 | 深圳传世智慧科技有限公司 | Investigation method, device, system and server based on semantic analysis |
| CN112862431A (en) * | 2021-01-19 | 2021-05-28 | 中国建设银行股份有限公司 | Method and device for generating conversation record, electronic equipment and readable storage medium |
| CN113421572A (en) * | 2021-06-23 | 2021-09-21 | 平安科技（深圳）有限公司 | Real-time audio conversation report generation method and device, electronic equipment and storage medium |
| CN113421572B (en) * | 2021-06-23 | 2024-02-02 | 平安科技（深圳）有限公司 | Real-time audio dialogue report generation method and device, electronic equipment and storage medium |
| CN114780733A (en) * | 2021-12-31 | 2022-07-22 | 海南大学 | DIKW atlas-based intelligent patent modification method, auxiliary response method and system |
| CN115186658A (en) * | 2022-06-20 | 2022-10-14 | 广东电网有限责任公司广州供电局 | Feasibility study report text processing method and device, and computer equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108062955B (en) | 2020-12-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108062955A (en) | A kind of intelligence report-generating method, system and equipment | |
| US7630892B2 (en) | Method and apparatus for transducer-based text normalization and inverse text normalization | |
| CN110232177B (en) | Bidding document generation system and method in government field | |
| US20230163988A1 (en) | Computer-implemented system and method for providing an artificial intelligence powered digital meeting assistant | |
| CN103530282A (en) | Corpus tagging method and equipment | |
| CN110147544B (en) | Instruction generation method and device based on natural language and related equipment | |
| CN109829052A (en) | A kind of open dialogue method and system based on human-computer interaction | |
| CN107798123B (en) | Knowledge base and establishing, modifying and intelligent question and answer methods, devices and equipment thereof | |
| CN110110039A (en) | A kind of method that the visualization of intelligent sound robot is built | |
| US12130849B2 (en) | Providing responses to queries of transcripts using multiple indexes | |
| JPWO2005010789A1 (en) | Capability evaluation apparatus, capability evaluation method, and capability evaluation program | |
| CN117217315B (en) | Method and device for generating high-quality question-answer data by using large language model | |
| CN118069797B (en) | A question-answering recommendation method and system for flood control dispatching rules based on large language model | |
| CN110738981A (en) | interaction method based on intelligent voice call answering | |
| CN109903122A (en) | House prosperity transaction information processing method, device, equipment and storage medium | |
| CN118503396A (en) | ERP system large model calling method, device and medium based on open prompt words | |
| CN117709359A (en) | Method, device, equipment and storage medium for acquiring insurance knowledge question-answer pair | |
| Parikh et al. | Auto-Query-A simple natural language to SQL query generator for an e-learning platform | |
| US20100250472A1 (en) | System and method of machine-aided information extraction rule development | |
| CN114462376A (en) | RPA and AI-based court trial record generation method, device, equipment and medium | |
| CN110502675A (en) | Voice calls user classification method and relevant device based on data analysis | |
| CN119226436A (en) | Memory-driven question-answering optimization method, device, equipment and storage medium | |
| CN116860957B (en) | Enterprise screening method, device and medium based on large language model | |
| CN115878818B (en) | Geographic knowledge graph construction method, device, terminal and storage medium | |
| CN104881403B (en) | Segmenting method and device |
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 |