CN105810207A - Meeting recording device and method thereof for automatically generating meeting record - Google Patents
Meeting recording device and method thereof for automatically generating meeting record Download PDFInfo
- Publication number
- CN105810207A CN105810207A CN201410839533.9A CN201410839533A CN105810207A CN 105810207 A CN105810207 A CN 105810207A CN 201410839533 A CN201410839533 A CN 201410839533A CN 105810207 A CN105810207 A CN 105810207A
- Authority
- CN
- China
- Prior art keywords
- minutes
- word
- user
- module
- 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.)
- Pending
Links
Landscapes
- Document Processing Apparatus (AREA)
Abstract
The invention provides a meeting recording device and a method thereof for automatically generating a meeting record. The method comprises: according to speech signals corresponding to a speech received on a meeting and a user speech feature table stored in a memory, identifying one or more users corresponding to the speech signals; converting the speech signals into characters including user names of the one or more users; and generating an original meeting record according to the characters obtained by converting and a meeting recording template. The meeting recording device and the method thereof for automatically generating a meeting record can automatically generate the meeting record according to the preset meeting recording template, and are thus more time-saving, convenient and humanized than the prior art.
Description
Technical field
The present invention relates to a kind of minutes device and the method automatically generating minutes thereof.
Background technology
The method of report and record in existing meeting, usually utilizes the equipment such as video camera, mike, recording pen that the speech of everyone in conference process is recorded and recorded a video.The personnel taken minutes after the meeting can check, play back recording and video recording to arrange minutes.But, by manually speech data being labeled and extracting, for user, time-consuming and extremely inconvenience.
Summary of the invention
In consideration of it, a kind of method being necessary minutes device to be provided and automatically generates minutes, it is possible to automatically generate minutes, to solve the problems referred to above.
The present invention provides a kind of minutes device, including memorizer and processor.That described minutes device also includes being controlled by described processor and be stored in described memorizer such as lower module: recognition module: as described in corresponding according to the voice received in a meeting voice signal and as described in the user vocal feature table of storage in memorizer, pick out one or more user that described voice signal is corresponding；Modular converter: for described voice signal being converted to the word of the user name comprising one or more user described；And generation module: for according to the original minutes of minutes template generation one of storage in the described word being converted to and described memorizer.
The present invention also provides for a kind of method automatically generating minutes, runs at least one device including memorizer and processor.Described method includes being controlled, by described processor, the following steps that in described memorizer, the module of storage performs: identification step: the user vocal feature table of storage in the voice signal corresponding according to the voice received in a meeting and described memorizer, picks out one or more user that described voice signal is corresponding；Switch process: described voice signal is converted to the word of the user name comprising one or more user described；And generation step: according to the original minutes of minutes template generation one of storage in the described word being converted to and described memorizer.
Minutes device of the present invention and the method automatically generating minutes thereof, can automatically generate minutes according to default minutes template, thus, more save time compared to existing mode, facilitate and hommization.
Accompanying drawing explanation
Fig. 1 is the applied environment schematic diagram of the minutes device of an embodiment of the present invention.
Fig. 2 is the functional block diagram of an embodiment of the minutes device shown in Fig. 1.
Fig. 3 is in an embodiment of the present invention, the schematic diagram of the minutes after the original minutes of generation and editor.
The flow chart of steps of Fig. 4-Fig. 7 respectively method automatically generating minutes of difference embodiment of the present invention.
Main element symbol description
| Minutes device | 100 |
| High in the clouds device | 200 |
| User | 1 |
| Original minutes | 310 |
| Minutes after editor | 320 |
| The method automatically generating minutes | 400、500、600、700 |
| Memorizer | 10 |
| Recording module | 11 |
| Modular converter | 12 |
| Recognition module | 13 |
| Judge module | 14 |
| Check and correction editor module | 15 |
| Generation module | 16 |
| Sending module | 17 |
| Segmentation module | 18 |
| Control module | 19 |
| Voice-input unit | 20 |
| Touch screen | 30 |
| Communication unit | 40 |
| Positioning module | 50 |
| Processor | 60 |
| Step | S401-S407、S501-S508、S601-S607、S701-S707 |
Following detailed description of the invention will further illustrate the present invention in conjunction with above-mentioned accompanying drawing.
Detailed description of the invention
Referring to Fig. 1, it is the applied environment schematic diagram of minutes device 100 of one embodiment of the present invention.In present embodiment, minutes device 100 can be connected with a high in the clouds device 200.Wherein, minutes device 100 is near each user 1, can receive each user 1 voice in meeting or report, the i.e. speech of user 1.Minutes device 100 and/or high in the clouds device 200 possess the voice according to minutes device 100 reception and automatically generate the function of minutes.User 1 is the participant of meeting or report.For convenience, below meeting or report are collectively referred to as meeting.
In one embodiment, minutes device 100 has the function automatically generating minutes, i.e. can generate minutes voluntarily.And minutes device 100 is independent of high in the clouds device 200, and voice according to its reception automatically generates minutes voluntarily.When multiple users 1 hold a meeting or report, minutes device 100 can record the voice of each user 1 automatically, and automatically will identify the voice of each user, and after the voice recognized is converted to word, automatically generate minutes according to default minutes template, and be automatically sent to related personnel according to default mode.Related personnel includes each user 1 and/or other meetings related personnel, for instance the personnel such as backlog director, relevant supervisor.Thus realizing record, generation automatically and sending the function of minutes.
For convenience of explanation, the word in the following bracket in this paragraph is the function declaration of the simplification of the word before it.Concrete refers to description below.Minutes device 100 can identification receives automatically each corresponding user (user in identification voice) of voice, then the voice of reception is converted to the word of the user name of user including picking out, or, the voice of reception is automatically converted to word (voice is converted to word), from word, then identifies the user name (user in recognition character) of each user 1.After word being carried out paragraph division (dividing paragraph according to word) according to the above-mentioned user name identified from voice and/or word afterwards, automatically generate minutes (generation minutes) further according to default minutes template.Minutes device 100 can also according to the automatic recognition of speech noiseless fragment therein (according to the noiseless fragment of speech recognition) received, according to the noiseless fragment identified, voice is divided into multiple sound bite (dividing paragraph according to voice), then respectively the plurality of sound bite is converted to the word (voice is converted to word) of correspondence, automatically generates minutes (generation minutes) further according to default minutes template.The words and phrases that minutes device 100 can also repeatedly repeat in identification voice and/or Word message automatically, and be stored in common-use words data base, thus in the process generating minutes, it is possible to automatically conventional words and phrases are proofreaded in the words and phrases in writing record.
In another embodiment, minutes device 100 can carry out data communication with high in the clouds device 200, thus the voice individually received according to minutes device 100 together with high in the clouds device 200 or by high in the clouds device 200 by minutes device 100 automatically generates minutes.nullThus，The present invention can also is that and by minutes device 100, meeting recorded，And the voice recorded is converted to voice signal，By conversion to voice signal and/or other data (word etc. being such as converted to according to voice signal) transmission to high in the clouds device 200，And by minutes device 100 and/or and high in the clouds device 200 perform all or part of in the following functions all performed by minutes device 100 in a upper embodiment respectively: voice is converted to word、User in identification voice and/or word、According to the noiseless fragment of voice and/or Text region、According to voice and/or word divide paragraph、Generate minutes、Conventional words and phrases in identification voice and/or word、The conventional words and phrases of storage are in common-use words data base，And according to conventional words and phrases automatic Proofreading/editor's word or minutes.
Referring to Fig. 2, it is an embodiment of the present invention.It should be noted that, the functional block diagram of the minutes device 100 in the shown only one embodiment of the present invention of Fig. 2, corresponding described above realizes the embodiments of the present invention, and minutes device 100 can also is that the functional unit/module only including the part shown in Fig. 2.High in the clouds device 200 then can include other functional unit/modules shown in Fig. 2.Such as, in the embodiment of function individually being automatically generated minutes by high in the clouds device 200 execution, minutes device 100 can include the voice-input unit 20 shown in Fig. 2, communication unit 40, processor 60, and high in the clouds device 200 can include the module 12-19 of storage in corresponding communication unit, processor and memorizer 10.Hereinafter corresponding description will be made when needed.
In present embodiment, minutes device 100 includes a memorizer 10, voice-input unit 20, touch screen 30, communication unit 40, positioning module 50 and processor 60.Memorizer 10, voice-input unit 20, touch screen 30, communication unit 40 are connected to processor 60 by holding wire and data wire.Minutes device 100 is a smart mobile phone, and in other embodiments, minutes device 100 can also is that the devices such as panel computer, notebook computer, desk computer and conference telephone.
In present embodiment, minutes device 100 can independently automatically generate minutes.The voice of reception, automatically according to the voice of the user 1 participating in meeting received by its voice-input unit 20, is converted to word, automatically generates minutes further according to default minutes template afterwards by minutes device 100.Concrete, the user in word after minutes device 100 can perform aforesaid the voice of reception to be automatically converted to word, automatic identification receives voice or conversion, according to the user name that picks out, word is carried out paragraph division, automatically generate minutes further according to default minutes template.Minutes device 100 can also according to the automatic recognition of speech noiseless fragment therein received, according to the noiseless fragment identified, voice is divided into multiple sound bite, then respectively the plurality of sound bite is converted to the word of correspondence, automatically generates minutes further according to default minutes template.The words and phrases that minutes device 100 can also repeatedly repeat in identification voice and/or Word message automatically, and be stored in common-use words data base, thus in the process generating minutes, it is possible to automatically conventional words and phrases are proofreaded in the words and phrases in writing record.The minutes of generation and/or backlog can also be automatically sent to the address of related personnel according to predetermined manner by minutes device 100.Wherein, what this predetermined manner included presetting sends form, default transmission time etc..The address of related personnel at least include following in one: e-mail address, telephone number, social account (such as QQ number, wechat account) etc..
Memorizer 10 stores a user vocal feature table, this phonetic feature table record one-to-one relationship of multiple user names and speech characteristic parameter thereof.In present embodiment, user name can be the Real Name of user, it is also possible to be the pet name or code name etc..This user vocal feature table can training in advance obtain, i.e. in the time before meeting/report starts, each user is carried out voice training, collection and obtain.Memorizer 10 can also store by the minutes template of user or systemic presupposition.Memorizer 10 can be also used for the voice text data storehouse etc. needed for the speech data of storage recording, language and characters conversion and common-use words data base.Wherein, common-use words data base performs in the process of its function automatically generating minutes at minutes device 100, and accumulation, screening store, it is also possible to download from a common-use words data base and store.
In present embodiment, voice-input unit 20 is for gathering the voice of each user during meeting, and the voice collected is converted to voice signal.Voice-input unit 20 is a mike.Communication unit 40 carries out data communication for the control of answer processor 60 with high in the clouds device 200.Positioning module 50 is for providing the real-time position information of minutes device 100, and it can be a GPS positioning module.
In one embodiment, minutes device 100 also includes a touch screen 30.
In the present embodiment, also stored for multiple functional module in memorizer 10, the plurality of functional module is configured to be performed by one or more processors (present embodiment is a processor 60), to complete the present invention.Such as, consult shown in Fig. 1, memorizer 10 stores recording module 11, modular converter 12, recognition module 13, judge module 14, check and correction editor module 15, generation module 16, sending module 17, segmentation module 18 and controls module 19.In other embodiments, in memorizer 10, the functional module of storage can also do corresponding change according to actual needs, such as, when voice is converted to the conventional words and phrases in word, automatic identification voice and/or word, the conventional words and phrases of storage in common-use words data base, and when being set to be performed by high in the clouds device 200 according to one or more function in the functions such as conventional words and phrases automatic Proofreading word, the memorizer 10 of minutes device 100 can not store the functional module performed needed for this function.Module alleged by the present invention has been the program segment of a specific function, is more suitable for describing software execution process in processor 60 than program.Function about each module will specifically describe in the flow chart of Fig. 4-Fig. 7.
It should be noted that, for convenience of explanation, below in relation in the introduction of method automatically generating minutes, it is all run in this way in a minutes device (such as minutes device 100) including corresponding unit and/or functional module to be introduced.Known according to introducing above, in the following method respectively automatically generating minutes, some step can also arrange the high in the clouds device (such as high in the clouds device 200) being connected with meeting recording equipment by and perform, therefore, accordingly, when needing, it is possible to increase minutes device in the step of the following method respectively automatically generating minutes by voice signal/data, lteral data and/or other data transmission to this high in the clouds device, and this high in the clouds device receives the step of signal/data.Because of those for those skilled in the art can some technological means of obtaining of the Content Implementation disclosed by this specification, therefore, for saving for the purpose of length, by specifically not detailed one by one in this manual description.
As shown in Figure 4, it is the flow chart of the method 400 automatically generating minutes of an embodiment of the present invention.The method 400 automatically generating minutes is after the minutes function of a meeting recording equipment (such as minutes device 100) and/or high in the clouds device (such as high in the clouds device 200) is unlocked, running on this minutes device and/or high in the clouds device, it may begin at step S401, step S402 or step S403.
Step S401, receiving step: voice-input unit 20 receives voice and the voice of reception is converted to corresponding voice signal.In present embodiment, minutes device 100 is located near the user 1 of meeting, and voice-input unit 20 is the mike being arranged in minutes device 100.
In another embodiment, can also this step S401 simultaneously or before perform following steps: control module 19 and control to open positioning module 50 to obtain the positional information of a meeting recording equipment 100 and current the time of meeting information, and positional information and the temporal information of acquisition are stored in memorizer 10.In other embodiments, minutes device 100 can also receive the relevant information of the active conference via touch screen 30 input and store, for instance, meeting date, time, place and people with a part in a conference person's name etc..
Step S402, step of recording: described voice signal recording is made speech data by recording module 11, and the speech data recorded is stored in memorizer 10.In one embodiment, the selection of response user, this step can also be omitted, and directly performs step S403.
Step S403, identification step: recognition module 13, according to the user vocal feature table of storage in described voice signal and memorizer 10, identifies one or more user that described voice signal is corresponding.In present embodiment, recognition module 13 obtains one or more phonetic feature according to described speech signal analysis, and from described phonetic feature table, inquire one or more user that identical/the most close phonetic feature is corresponding, thus obtain one or more user corresponding in speech data.When meeting or report carry out, in time having multiple user to make a speech/speak, recognition module 13 can identify, according to described voice signal and described phonetic feature table, the sound containing which user in described speech data.
In another embodiment, recognition module 13 returns the sound bite of different users plus different labels, and the sound bite of same user is plus identical label.
Step S404, switch process: described voice signal is converted to the word of the user name comprising one or more user described by modular converter 12.In present embodiment, modular converter 12 is according to the voice text data storehouse of storage in described voice signal and memorizer 10, described voice signal is converted to word, and a predeterminated position of the word being converted to corresponding to the voice signal of each user of one or more user recognized in recognition module 13 adds the user name of user of correspondence automatically, in present embodiment, predeterminated position be the word being converted to that the voice signal of each user is corresponding foremost.
In another embodiment, when recognition module 13 adds a little label to the sound bite of different users, the described word that modular converter 12 is converted to further comprises those labels.
Step S405, generation step: generation module 16 is according to the original minutes of minutes template generation one of storage in the described word being converted to and memorizer 10.Referring to shown in Fig. 3, it is shown with in an embodiment, the original minutes 310 that generation module 16 generates.
In one embodiment, positional information acquired for positioning module 50 and temporal information are also automatically added in the original minutes generated by generation module 16.Such as, temporal information is added in the field of the meeting date/time in minutes template, positional information is added in the field in the meeting-place in minutes template, etc..
Convention goer/attendant that user is inputted by touch screen 30 can also be automatically added in the field of the attendant/participant in minutes template by generation module 16.
In another embodiment, those user names are added in the field of the attendant/participant in minutes template by the user name of the user sending voice corresponding to described voice signal that generation module 16 can also obtain according to voice signal identification according to the user name comprised in the described word that recognition module 13 recognizes or recognition module 13 automatically.
Step S406, proofreads edit step: described original minutes are proofreaded and/or edited according to default check and correction editor's rule by check and correction editor module 15, to obtain minutes.
In present embodiment, described default check and correction editor's rule is that word is carried out paragraph division by each user name from described word.The user name of recognition module 13 also identification from the described word being converted to/identify user, described original minutes are carried out paragraph division by the user name comprised in the described word that check and correction editor module 15 then recognizes according to recognition module 13.Such as, check and correction editor module 15 divides paragraph with first of user name or the last character for boundary.When comprising user in described word and being called Wang Daming, check and correction editor module 15 is then first from the section using Wang Daming these three word as paragraph.It should be noted that in present embodiment, it is preferred that user name described herein is all the user name of the user obtained by identification voice by recognition module 13.In another embodiment, those user names can also is that the recognition module 13 user name according to storage original in memorizer 10, automatically identifies from described word.Referring to shown in Fig. 3, it is shown with the minutes 320 in an embodiment, after the editor that check and correction editor module 15 obtains after original minutes 310 are proofreaded and/or edited.
In another embodiment, described default check and correction editor's rule is the label added to the sound bite of different users according to recognition module 13, from the word corresponding to each sound bite section start, word paragraph is carried out cutting.
In a further embodiment, the described minutes after check and correction editor are also stored in described memorizer 10 by check and correction editor module 15.Or, sending module 17 controls to be sent the described minutes after check and correction editor to described high in the clouds device 200 by communication unit 40, to control to be stored in described minutes described high in the clouds device 200.
In other embodiments, minutes are edited by editor's signal that check and correction editor module 15 generates always according to touch screen 30.Such as, user can pass through touch screen 30 and input the content of edit to original minutes and/or edit operation, thus providing the function for the original minutes of user manual editing.Additionally, described default check and correction editor's rule also includes Intelligent Recognition check and correction word etc., specifically incorporated by reference to below according to Fig. 5 explanation carried out.
Step S407, forwarding step: the described minutes after check and correction and/or editor are automatically sent to the address of meeting related personnel according to the default rule that sends by sending module 17.In present embodiment, described default transmission rule can be send (namely immediately, namely minutes send after generating) to the address of meeting related personnel, it is also possible to it is that the Preset Time point after minutes generate sends the address to meeting related personnel.Described meeting related personnel can include one or more in following personnel: occurs in that the user (such as, the user of backlog) relating to/mentioning in the user of its user name, minutes, the supervisor of presetting, director, person liable etc. in meeting attendant, minutes.
In another embodiment, the described default regular default preset number of days expired a few days ago being additionally may included in backlog that sends sends the address of the personnel that the described minutes generated are correlated with to backlog, for example, it is possible to include the person directly responsible of backlog, relevant supervisor and other related personnel relevant to this backlog.
In other embodiments, it is also possible to be not provided with this step S407, and directly manually sent the minutes address to meeting related personnel by user；Or, when device 200 receives and store these minutes beyond the clouds, high in the clouds device 200 these minutes are sent to meeting related personnel.
As it is shown in figure 5, be the flow chart of the method 500 automatically generating minutes of an embodiment of the present invention.The method 500 automatically generating minutes is after the minutes function of a meeting recording equipment (such as minutes device 100) is unlocked, and runs on this minutes device.It should be noted that, in the step performed in the method 500 automatically generating minutes shown in Fig. 5 and the method 400 automatically generating minutes shown in Fig. 4, some is same or like, therefore, it is above-mentioned when the method 400 automatically generating minutes in Fig. 4 is described, step same or like in some replacements illustrated for certain step, the method 500 automatically generating minutes that other embodiments that can simultaneously perform also apply in Fig. 5, just no longer repeats one by one at this.The method 500 automatically generating minutes may begin at step S501.
Step S501, receiving step: voice-input unit 20 receives voice and the voice of reception is converted to corresponding voice signal.
Step S502, step of recording: described voice signal recording is made speech data by recording module 11, and the speech data recorded is stored in memorizer 10.In one embodiment, the selection of response user, this step can also be omitted, and directly performs step S503.
Step S503, identification step: recognition module 13 identifies the noiseless fragment in described speech data according to described voice signal.In present embodiment, described noiseless fragment be in described speech data for the fragment of quiet data, i.e. for described voice is quiet fragment.Such as, when the noiseless marginal value that the volume of the sound bite of speech data corresponding to certain part in described voice signal is preset less than, namely recognition module 13 identifies that this sound bite is noiseless fragment.Described speech data may includes multiple noiseless fragment.
In one embodiment, when not comprising step S502, in this step, recognition module 13 identifies the noiseless fragment in described voice according to described voice signal.
Step S504, it is judged that step: judge module 14 judges that whether time that described noiseless fragment goes through is more than a preset value, if it is, perform step S505, otherwise, flow process terminates.In one embodiment, described preset value is 3 seconds.
Step S505, segmentation step: described speech data is divided into multiple speech data fragment according to described noiseless fragment by segmentation module 18.In present embodiment, described speech data is split by segmentation module 18 from described noiseless fragment, when comprising multiple noiseless fragment that the time gone through is all higher than described preset value in described speech data, described speech data is divided into multiple speech data fragment according to multiple noiseless fragments by segmentation module 18.
Step S506, identification step: recognition module 13, according to splitting the user vocal feature table of storage in voice signal corresponding to the multiple speech data fragments obtained and memorizer 10, identifies one or more user corresponding in the plurality of speech data fragment.In one embodiment, the method 500 originally automatically generating minutes can also be without including this step.
Step S507, switch process: voice signal corresponding for multiple speech data fragments that segmentation obtains is converted to the word comprising multiple paragraph by modular converter 12.In present embodiment, in one or more user that modular converter 12 recognizes according to voice signal corresponding to the plurality of speech data fragment, recognition module 13 and memorizer 10, the voice text data storehouse of storage, is converted to voice signal corresponding for the plurality of speech data fragment and comprises and the word of each speech data fragment multiple paragraphs one to one.
Step S508, generation step: generation module 16 is according to the original minutes of minutes template generation one of storage in the word comprising multiple paragraph described in being converted to and memorizer 10.Mode concrete for this step S509 and the method 400 automatically generating minutes can be identical, just do not repeating at this.
In the present embodiment, after this step S508, can also carry out the step S406(in the method 400 automatically generating minutes proofread edit step) and step S407(forwarding step), just repeat no more at this.
As shown in Figure 6, it is the flow chart of the method 600 automatically generating minutes of an embodiment of the present invention.The method 600 automatically generating minutes is after the minutes function of a meeting recording equipment (such as minutes device 100) is unlocked, and runs on this minutes device.It should be noted that, in the step that the method 600 automatically generating minutes shown in Fig. 6 is performed with the method automatically generating minutes shown in Fig. 5 and Fig. 4, some is same or like, therefore, above-mentioned to the method 400 automatically generating minutes in Fig. 4 and when the method 500 automatically generating minutes in Fig. 5 is described, some replacements illustrated for certain step, other embodiments that can simultaneously perform also apply to step same or like in the method 600 automatically generating minutes in Fig. 6, also no longer repeat one by one at this.The method 600 automatically generating minutes may begin at step S601.
Step S601, receiving step: voice-input unit 20 receives voice and the voice of reception is converted to corresponding voice signal.
Step S602, step of recording: described voice signal recording is made the speech data comprising record length stamp by recording module 11, and the speech data recorded is stored in memorizer 10.In one embodiment, the selection of response user, this step can also be omitted, and directly performs step S603.
Step S603, identification step: recognition module 13, according to the user vocal feature table of storage in described voice signal and memorizer 10, identifies one or more user corresponding in described voice signal.In one embodiment, recognition module 13, according to the user vocal feature table of storage in the described speech data comprising record length stamp and memorizer 10, identifies one or more user that described voice signal is corresponding.In another embodiment, the method 600 automatically generating minutes can not also include this step.
Step S604, switch process: described voice signal is converted to the word of the user name comprising described record length stamp and one or more user described by modular converter 12.In present embodiment, described voice signal is converted to the word of the user name comprising described record length stamp and one or more user described by modular converter 12.Modular converter 12 contains the voice text data storehouse of storage in the speech data of record length stamp, one or more user of recognizing of recognition module 13 and memorizer 10 according to what described voice signal, recording module 11 were recorded, described voice signal is converted to the word containing described record length stamp the user name automatically adding corresponding user foremost of the word being converted at the voice signal of each user.In another embodiment, modular converter 12 according to described voice signal, recording module 11 record contain record length stamp speech data and memorizer 10 in storage voice text data storehouse, described voice signal is converted to contain described record length stamp word.
Step S605, it is judged that step: judge module 14 is according to the described word after conversion, it may be judged whether the interval described in record length stamp having adjacent word corresponding reaches a preset value, if it is, perform step S606, otherwise, flow process terminates.In one embodiment, described preset value is 3 seconds.The described adjacent word comprising described record length stamp is likely to include multiple interval and reaches this preset value.
Step S606, segmentation step: it is that boundary divides word paragraph that interval described for the record length stamp of described correspondence is reached the adjacent word of described preset value by segmentation module 18.In present embodiment, concrete, this adjacent word is divided into previous paragraph and adjacent later paragraph respectively, until each adjacent word that the interval described in the record length stamp of all of correspondence reaches described preset value is all divided into different paragraphs.
Step S607, generation step: generation module 16 is according to dividing the described word after paragraph and the original minutes of minutes template generation one of storage in memorizer 10.Mode concrete for this step S607 and the method 500 automatically generating minutes can be identical, just do not repeating at this.
As it is shown in fig. 7, be the flow chart of the method 700 automatically generating minutes of an embodiment of the present invention.The method 700 automatically generating minutes is after the minutes function of a meeting recording equipment (such as minutes device 100) is unlocked, and runs on this minutes device.It should be noted that, in the step that the method 700 automatically generating minutes shown in Fig. 7 is performed with the method automatically generating minutes shown in Fig. 5 and Fig. 4, some is same or like, therefore, above-mentioned to the method 400 automatically generating minutes in Fig. 4 and when the method 500 automatically generating minutes in Fig. 5 is described, some replacements illustrated for certain step, other embodiments that can simultaneously perform also apply to step same or like in the method 700 automatically generating minutes in Fig. 7, also no longer repeat one by one at this.Originally the method 700 automatically generating minutes may begin at step S701.
Step S701, builds storehouse step: controls module 19 and sets up a common-use words data base comprising common-use words and calibration object thereof, and by described common-use words database purchase in memorizer 10.In present embodiment, it is possible to be when minutes device 100 is use the function automatically generating minutes first, control module 19 and automatically set up described common-use words data base.Comprising the corresponding relation of at least one common-use words and calibration object thereof in described common-use words data base, each common-use words are at least corresponding with a calibration object.Described common-use words include following in one or more: commonly used word, everyday words, common sentence etc., it is also possible to be speech data or lteral data.The calibration object of each common-use words can be accumulated in user manual editing, amendment minutes process, record.Calibration object include in following speech data and/or lteral data following in one or more: word, word, sentence etc..
In another embodiment, the method 700 originally automatically generating minutes can also be without including this step S701.But in this minutes device, it is previously stored with a common-use words data base, common-use words data base performs in the process of its function automatically generating minutes at minutes device 100, accumulation, screening store, it is also possible to download from a common-use words data base and store.
Step S702, receiving step: voice-input unit 20 receives voice and the voice of reception is converted to corresponding voice signal.
Step S703, switch process: described voice signal is converted to word by modular converter 12.In one embodiment, it is also possible to include automatically generating the receiving step that comprises in either method in the method 400,500 and 600 of minutes to other steps between switch process.I.e., it is possible to comprise the step converting voice signals into word of various embodiment described above.
Step S704, identify storage everyday words step: the described words and phrases repeating this preset times, when identification judges to comprise, in described speech data and/or described word, the words and phrases repeating a preset times, are stored in described common-use words data base by judge module 14 as common-use words.In present embodiment, the words and phrases repeating this preset times can be speech data and/or the lteral datas such as word, word, sentence.In one embodiment, this step S704 can also omit.Described preset times is 20 times.
Step S705, it is judged that step: judge module 14 judges that the described word after changing is no and comprises a calibration object, if it is, perform step S706, otherwise, flow process terminates.
Step S706, proofreads step: the calibration object that described word comprises automatically is corrected to the common-use words of correspondence by check and correction editor module 15 according to described common-use words data base.In an embodiment party, this step S706 can also perform after step S707.
Step S707, generation step: generation module 16 according to correction after described word and memorizer 10 in storage the original minutes of minutes template generation one.Mode concrete for this step S707 and the method 500 automatically generating minutes can be identical, just do not repeating at this.
Above-mentioned minutes device 100 provided by the invention and the method automatically generating minutes thereof, can automatically generate minutes according to default minutes template, and minutes can carry out the language and characters identification of intelligence, content format and editor 's proofreading.Moreover, it is also possible to minutes are sent to related personnel according to default rule.Thus, more save time compared to existing mode, facilitate and hommization.
Finally it should be noted that, above example is only in order to illustrate technical scheme and unrestricted, although the present invention being described in detail with reference to preferred embodiment, it will be understood by those within the art that, technical scheme can be modified or equivalent replacement, without deviating from the spirit and scope of technical solution of the present invention.
Claims (18)
1. the method automatically generating minutes, runs at least one device including memorizer and processor, it is characterised in that described method includes being controlled, by described processor, the following steps that in described memorizer, the module of storage performs:
Identification step: the user vocal feature table of storage in the voice signal corresponding according to the voice received in a meeting and described memorizer, picks out one or more user that described voice signal is corresponding；
Switch process: described voice signal is converted to the word of the user name comprising one or more user described；And
Generation step: according to the original minutes of minutes template generation one of storage in the described word being converted to and described memorizer.
2. the method for claim 1, it is characterised in that also include an edit step: according to default check and correction editor's rule, described original minutes are edited, to obtain minutes.
3. the method for claim 1, it is characterised in that described switch process includes:
According to the voice text data storehouse of storage in described voice signal and described memorizer, described voice signal is converted to word；And
A predetermined position at word corresponding to the voice signal of one or more user described adds the user name of the user of correspondence automatically.
4. the method for claim 1, it is characterised in that described generation step also includes:
The positional information and the temporal informations that obtain a positioning module of described device are automatically added in the minutes generated；
The user name comprised in word described in identification, or the user name of the user of the voice of correspondence is sent according to described voice signal identification；And
Automatically the user name recognized is added in the field of the attendant in minutes template.
5. method as claimed in claim 2, it is characterised in that: described check and correction editor's rule is that word is carried out paragraph division by each user name from described word.
6. the method as described in any one of claim 2-4, it is characterised in that also include:
Record step: described voice signal recording is made speech data, and the speech data recorded is stored in memorizer；And
Forwarding step: the described minutes after compiled are automatically sent to the address of meeting related personnel according to the default rule that sends.
7. the method as described in any one of claim 1-4, it is characterised in that also include:
Judge step: judge that described word is no and comprise a calibration object；And
Check and correction step: when described word comprises a calibration object, is automatically corrected to the common-use words of correspondence according to a common-use words data base by the calibration object that described word comprises.
8. method as claimed in claim 7, it is characterized in that, also include an identification storage everyday words step: when identification judges to comprise, in described voice and/or described word, the words and phrases repeating a preset times, be stored in described common-use words data base as common-use words by the described words and phrases repeating this preset times.
9. method as claimed in claim 1 or 2, it is characterised in that described default check and correction editor's rule is divide word paragraph according to the plurality of speech data fragment, and described method also includes:
Identify the noiseless fragment in described voice；And
When judging that time that described noiseless fragment is gone through is more than a preset value, according to described noiseless fragment, speech data corresponding for described voice is divided into multiple speech data fragment.
10. a minutes device, including memorizer and processor, it is characterised in that also include being controlled by described processor and be stored in described memorizer such as lower module:
Recognition module: the user vocal feature table of storage in the described voice signal corresponding according to the voice received in a meeting and described memorizer, picks out one or more user that described voice signal is corresponding；
Modular converter: for described voice signal being converted to the word of the user name comprising one or more user described；And
Generation module: for according to the original minutes of minutes template generation one of storage in the described word being converted to and described memorizer.
11. minutes device as claimed in claim 10, it is characterised in that also include a check and correction editor module, for editing described original minutes according to the check and correction editor's rule preset, to obtain minutes.
12. minutes device as claimed in claim 10, it is characterised in that described modular converter is additionally operable to:
According to the voice text data storehouse of storage in described voice signal and described memorizer, described voice signal is converted to word；And
A predetermined position at word corresponding to the voice signal of one or more user described adds the user name of the user of correspondence automatically.
13. minutes device as claimed in claim 10, it is characterised in that described generation module also includes:
The positional information and the temporal informations that obtain a positioning module of described device are automatically added in the minutes generated；
The user name comprised in word described in identification, or the user name of the user of the voice of correspondence is sent according to described voice signal identification；And
Automatically the user name recognized is added in the field of the attendant in minutes template.
14. minutes device as claimed in claim 11, it is characterised in that: described check and correction editor's rule is that word is carried out paragraph division by each user name from described word.
15. the minutes device as described in any one of claim 11-13, it is characterised in that also include:
Recording module: for described voice signal recording being made speech data, and the speech data recorded is stored in memorizer；And
Sending module: for the described minutes after compiled being automatically sent to the address of meeting related personnel according to the rule that sends preset.
16. the minutes device as described in any one of claim 10-13, it is characterised in that also include a judge module, it is used for judging that described word is no and comprises a calibration object；Described check and correction editor module is additionally operable to when described word comprises a calibration object, and the calibration object that described word comprises automatically is corrected to the common-use words of correspondence according to a common-use words data base.
17. minutes device as claimed in claim 16, it is characterized in that, described judge module is additionally operable to, when identification judges to comprise, in described voice and/or described word, the words and phrases repeating a preset times, be stored in described common-use words data base as common-use words by the described words and phrases repeating this preset times.
18. the minutes device as described in claim 10 or 11, it is characterised in that:
Described default check and correction editor's rule is divide word paragraph according to the plurality of speech data fragment；
Described minutes device also includes a segmentation module；
Described recognition module is additionally operable to the noiseless fragment identifying in described voice；
Described judge module is additionally operable to judge that whether time that described noiseless fragment goes through is more than a preset value；And
Described segmentation module for being divided into multiple speech data fragment according to described noiseless fragment by speech data corresponding for described voice.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410839533.9A CN105810207A (en) | 2014-12-30 | 2014-12-30 | Meeting recording device and method thereof for automatically generating meeting record |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410839533.9A CN105810207A (en) | 2014-12-30 | 2014-12-30 | Meeting recording device and method thereof for automatically generating meeting record |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105810207A true CN105810207A (en) | 2016-07-27 |
Family
ID=56980993
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410839533.9A Pending CN105810207A (en) | 2014-12-30 | 2014-12-30 | Meeting recording device and method thereof for automatically generating meeting record |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105810207A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108062955A (en) * | 2017-12-12 | 2018-05-22 | 深圳证券信息有限公司 | A kind of intelligence report-generating method, system and equipment |
| CN108231065A (en) * | 2016-12-12 | 2018-06-29 | 株式会社速录抓吧 | Multi-lingual person's speech recognition correction system |
| CN108416565A (en) * | 2018-01-25 | 2018-08-17 | 北京云知声信息技术有限公司 | Minutes method |
| CN108447502A (en) * | 2018-03-09 | 2018-08-24 | 福州米鱼信息科技有限公司 | A kind of memo method and terminal based on voice messaging |
| CN108538299A (en) * | 2018-04-11 | 2018-09-14 | 深圳市声菲特科技技术有限公司 | A kind of automatic conference recording method |
| CN109285548A (en) * | 2017-07-19 | 2019-01-29 | 阿里巴巴集团控股有限公司 | Information processing method, system, electronic equipment and computer storage medium |
| CN109473103A (en) * | 2018-11-16 | 2019-03-15 | 上海玖悦数码科技有限公司 | A kind of meeting summary generation method |
| WO2019062587A1 (en) * | 2017-09-27 | 2019-04-04 | 阿里巴巴集团控股有限公司 | Task generation method and device |
| CN109660368A (en) * | 2018-12-24 | 2019-04-19 | 广州维速信息科技有限公司 | A kind of cloud conference system and method |
| CN111048093A (en) * | 2018-10-12 | 2020-04-21 | 深圳海翼智新科技有限公司 | Conference sound box, conference recording method, device, system and computer storage medium |
| WO2020147407A1 (en) * | 2019-01-16 | 2020-07-23 | 平安科技（深圳）有限公司 | Conference record generation method and apparatus, storage medium and computer device |
| CN112786045A (en) * | 2021-01-04 | 2021-05-11 | 上海明略人工智能(集团)有限公司 | Device, server, method and system for conference recording |
| CN113885741A (en) * | 2021-06-08 | 2022-01-04 | 北京字跳网络技术有限公司 | Multimedia processing method, device, equipment and medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040064322A1 (en) * | 2002-09-30 | 2004-04-01 | Intel Corporation | Automatic consolidation of voice enabled multi-user meeting minutes |
| CN1773536A (en) * | 2004-11-11 | 2006-05-17 | 国际商业机器公司 | Method, equipment and system for generating speech summary |
| TW201207838A (en) * | 2010-08-04 | 2012-02-16 | Hon Hai Prec Ind Co Ltd | Electronic recording apparatus and method thereof |
| CN102436812A (en) * | 2011-11-01 | 2012-05-02 | 展讯通信（上海）有限公司 | Conference recording device and method for recording conference by using same |
| CN102842306A (en) * | 2012-08-31 | 2012-12-26 | 深圳Tcl新技术有限公司 | Voice control method and device as well as voice response method and device |
| CN102982800A (en) * | 2012-11-08 | 2013-03-20 | 鸿富锦精密工业（深圳）有限公司 | Electronic device with audio video file video processing function and audio video file processing method |
| CN103165131A (en) * | 2011-12-17 | 2013-06-19 | 富泰华工业（深圳）有限公司 | Voice processing system and voice processing method |
| CN103247289A (en) * | 2012-02-01 | 2013-08-14 | 鸿富锦精密工业（深圳）有限公司 | Recording system, recording method, sound inputting device, voice recording device and voice recording method |
| CN103714048A (en) * | 2012-09-29 | 2014-04-09 | 国际商业机器公司 | Method and system used for revising text |
-
2014
- 2014-12-30 CN CN201410839533.9A patent/CN105810207A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040064322A1 (en) * | 2002-09-30 | 2004-04-01 | Intel Corporation | Automatic consolidation of voice enabled multi-user meeting minutes |
| CN1773536A (en) * | 2004-11-11 | 2006-05-17 | 国际商业机器公司 | Method, equipment and system for generating speech summary |
| TW201207838A (en) * | 2010-08-04 | 2012-02-16 | Hon Hai Prec Ind Co Ltd | Electronic recording apparatus and method thereof |
| CN102436812A (en) * | 2011-11-01 | 2012-05-02 | 展讯通信（上海）有限公司 | Conference recording device and method for recording conference by using same |
| CN103165131A (en) * | 2011-12-17 | 2013-06-19 | 富泰华工业（深圳）有限公司 | Voice processing system and voice processing method |
| CN103247289A (en) * | 2012-02-01 | 2013-08-14 | 鸿富锦精密工业（深圳）有限公司 | Recording system, recording method, sound inputting device, voice recording device and voice recording method |
| CN102842306A (en) * | 2012-08-31 | 2012-12-26 | 深圳Tcl新技术有限公司 | Voice control method and device as well as voice response method and device |
| CN103714048A (en) * | 2012-09-29 | 2014-04-09 | 国际商业机器公司 | Method and system used for revising text |
| CN102982800A (en) * | 2012-11-08 | 2013-03-20 | 鸿富锦精密工业（深圳）有限公司 | Electronic device with audio video file video processing function and audio video file processing method |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108231065A (en) * | 2016-12-12 | 2018-06-29 | 株式会社速录抓吧 | Multi-lingual person's speech recognition correction system |
| CN109285548A (en) * | 2017-07-19 | 2019-01-29 | 阿里巴巴集团控股有限公司 | Information processing method, system, electronic equipment and computer storage medium |
| US11664030B2 (en) | 2017-07-19 | 2023-05-30 | Alibaba Group Holding Limited | Information processing method, system, electronic device, and computer storage medium |
| WO2019062587A1 (en) * | 2017-09-27 | 2019-04-04 | 阿里巴巴集团控股有限公司 | Task generation method and device |
| CN108062955A (en) * | 2017-12-12 | 2018-05-22 | 深圳证券信息有限公司 | A kind of intelligence report-generating method, system and equipment |
| CN108416565A (en) * | 2018-01-25 | 2018-08-17 | 北京云知声信息技术有限公司 | Minutes method |
| CN108447502B (en) * | 2018-03-09 | 2020-09-22 | 福州米鱼信息科技有限公司 | Memorandum method and terminal based on voice information |
| CN108447502A (en) * | 2018-03-09 | 2018-08-24 | 福州米鱼信息科技有限公司 | A kind of memo method and terminal based on voice messaging |
| CN108538299A (en) * | 2018-04-11 | 2018-09-14 | 深圳市声菲特科技技术有限公司 | A kind of automatic conference recording method |
| CN111048093A (en) * | 2018-10-12 | 2020-04-21 | 深圳海翼智新科技有限公司 | Conference sound box, conference recording method, device, system and computer storage medium |
| CN109473103A (en) * | 2018-11-16 | 2019-03-15 | 上海玖悦数码科技有限公司 | A kind of meeting summary generation method |
| CN109660368A (en) * | 2018-12-24 | 2019-04-19 | 广州维速信息科技有限公司 | A kind of cloud conference system and method |
| WO2020147407A1 (en) * | 2019-01-16 | 2020-07-23 | 平安科技（深圳）有限公司 | Conference record generation method and apparatus, storage medium and computer device |
| CN112786045A (en) * | 2021-01-04 | 2021-05-11 | 上海明略人工智能(集团)有限公司 | Device, server, method and system for conference recording |
| CN112786045B (en) * | 2021-01-04 | 2024-03-12 | 上海明略人工智能(集团)有限公司 | Device, server, method and system for conference recording |
| CN113885741A (en) * | 2021-06-08 | 2022-01-04 | 北京字跳网络技术有限公司 | Multimedia processing method, device, equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105810207A (en) | Meeting recording device and method thereof for automatically generating meeting record | |
| TWI619115B (en) | Meeting minutes device and method thereof for automatically creating meeting minutes | |
| TWI590240B (en) | Meeting minutes device and method thereof for automatically creating meeting minutes | |
| CN105810206A (en) | Meeting recording device and method thereof for automatically generating meeting record | |
| TW201624467A (en) | Meeting minutes device and method thereof for automatically creating meeting minutes | |
| Morrison et al. | Protocol for the collection of databases of recordings for forensic-voice-comparison research and practice | |
| US20040064322A1 (en) | Automatic consolidation of voice enabled multi-user meeting minutes | |
| US6895257B2 (en) | Personalized agent for portable devices and cellular phone | |
| US6687671B2 (en) | Method and apparatus for automatic collection and summarization of meeting information | |
| US10613825B2 (en) | Providing electronic text recommendations to a user based on what is discussed during a meeting | |
| CN108922538A (en) | Conferencing information recording method, device, computer equipment and storage medium | |
| US20120022865A1 (en) | System and Method for Efficiently Reducing Transcription Error Using Hybrid Voice Transcription | |
| CA3147813A1 (en) | Method and system of generating and transmitting a transcript of verbal communication | |
| KR102357620B1 (en) | Chatbot integration agent platform system and service method thereof | |
| CN102982800A (en) | Electronic device with audio video file video processing function and audio video file processing method | |
| CN105810208A (en) | Meeting recording device and method thereof for automatically generating meeting record | |
| CN1639681A (en) | System and method for concurrent multimodal communication using concurrent multimodal tags | |
| WO2018188936A1 (en) | Electronic communication platform | |
| CN106713111B (en) | Processing method for adding friends, terminal and server | |
| KR20170126667A (en) | Method for generating conference record automatically and apparatus thereof | |
| CN104820662A (en) | Service server device | |
| CN101211615A (en) | Method, system and apparatus for automatic recording for specific human voice | |
| CN107147564A (en) | Real-time speech recognition error correction system and identification error correction method based on cloud server | |
| CN111583932A (en) | Sound separation method, device and equipment based on human voice model | |
| CN112997207A (en) | Report automatic generation system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160727 |
|
| RJ01 | Rejection of invention patent application after publication |