CN107862071A - The method and apparatus for generating minutes - Google Patents
The method and apparatus for generating minutes Download PDFInfo
- Publication number
- CN107862071A CN107862071A CN201711174526.1A CN201711174526A CN107862071A CN 107862071 A CN107862071 A CN 107862071A CN 201711174526 A CN201711174526 A CN 201711174526A CN 107862071 A CN107862071 A CN 107862071A
- Authority
- CN
- China
- Prior art keywords
- electronic equipment
- identity
- characteristic parameter
- speech
- speech characteristic
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/60—Information retrieval; Database structures therefor; File system structures therefor of audio data
- G06F16/68—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/686—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using information manually generated, e.g. tags, keywords, comments, title or artist information, time, location or usage information, user ratings
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/166—Editing, e.g. inserting or deleting
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
Abstract
A kind of method and apparatus for generating minutes are provided, methods described includes：Speech characteristic parameter corresponding with the identity of the user of electronic equipment is obtained from each electronic equipment in multiple electronic equipments, wherein, the electronic equipment that the multiple electronic equipment is each held for multiple users of participation meeting；Gather the speech data in meeting；The identity of the spokesman of the speech data of collection is identified according to the speech characteristic parameter obtained from the multiple electronic equipment；The speech data of collection is converted into text data, wherein, the label of the identity of the spokesman of the insertion instruction speech data in text data；Minutes are generated based on the text data for inserting label.Using the method and apparatus of the generation minutes of exemplary embodiment of the present, pass through obtain all users for participating in meeting speech characteristic parameter conference speech content is corresponding with object of specifically making a speech, so as to improve the intelligent of minutes.
Description
Technical field
The present invention relates to electronic technology field, more particularly, is related to a kind of method and apparatus for generating minutes.
Background technology
At present, existing minutes method, typically using the equipment such as recording pen, video camera in conference process each one
The speech of member is recorded and recorded a video, and the personnel to take minutes after the meeting can check, play back recording and record a video to arrange meeting
Record.Special minutes personnel can also be specified to be recorded in conference process, but minutes personnel are typically only capable to
Concentrate one's energy on record, it is difficult to participate in session discussing.Minutes are carried out manually, frequently can lead to very big wave
Expense, high cost simultaneously easily make a mistake.
The content of the invention
The purpose of the exemplary embodiment of the present invention is to provide a kind of method and apparatus for generating minutes.
One side according to an exemplary embodiment of the present invention, there is provided a kind of method for generating minutes, methods described bag
Include：(A) phonetic feature corresponding with the identity of the user of electronic equipment is obtained from each electronic equipment in multiple electronic equipments
Parameter, wherein, the electronic equipment that the multiple electronic equipment is each held for multiple users of participation meeting；(B) meeting is gathered
In speech data；(C) speech data of collection is identified according to the speech characteristic parameter obtained from the multiple electronic equipment
Spokesman identity；(D) speech data of collection is converted into text data, wherein, the insertion instruction institute in text data
State the label of the identity of the spokesman of speech data；(E) minutes are generated based on the text data for inserting label.
Alternatively, the multiple electronic equipment may include described in the predetermined electronic equipment for being arranged to WIFI hot spot and access
At least one other electronic equipment of WIFI hot spot, wherein, step (A) may include：From at least one other electronic equipment
Obtain speech characteristic parameter corresponding with the identity of the user of electronic equipment.
Alternatively, the multiple electronic equipment may include predetermined electronic equipment for performing methods described and except described pre-
Determine at least one other electronic equipment outside electronic equipment, wherein, step (A) may include：When from described at least one other
When electronic equipment receives predetermined message, the identity with the user of electronic equipment is obtained from least one other electronic equipment
Corresponding speech characteristic parameter, wherein, the predetermined message may indicate that the language stored at least one other electronic equipment
Sound characteristic parameter is arranged to shared file for the predetermined electronic equipment.
Alternatively, step (A) can also include：The user with the predetermined electronic equipment is obtained from the predetermined electronic equipment
The corresponding speech characteristic parameter of identity.
Alternatively, step (C) may include：(C1) speech characteristic parameter of the speech data of extraction collection；(C2) from acquisition
Speech characteristic parameter in search the speech characteristic parameter that matches of speech characteristic parameter with extraction, and by the voice spy with extraction
The identity of user is defined as the identity of the spokesman of the speech data corresponding to the speech characteristic parameter of sign parameter matching.
Alternatively, step (C2) may include：Phonetic feature list is established according to the speech characteristic parameter of acquisition, wherein, institute
Predicate sound feature list includes speech characteristic parameter and the identity of user corresponding with speech characteristic parameter；Use the language of extraction
Sound characteristic parameter searches the identity of user corresponding with the speech characteristic parameter extracted in phonetic feature list, by what is found
Identity of the identity of user as the spokesman of the speech data.
Alternatively, methods described can also include：When there is new user to add meeting, the electronics held from the new user
Equipment obtains speech characteristic parameter corresponding with the identity of the new user, wherein, step (C) also includes：According to acquisition with
The corresponding speech characteristic parameter of identity of the new user identifies the identity of the spokesman of the speech data of collection.
Alternatively, methods described can also include：Established according to the speech characteristic parameter obtained from the multiple electronic equipment
Phonetic feature list；By the speech characteristic parameter corresponding with the identity of the new user obtained from the new electronic equipment and institute
The phonetic feature list is arrived in the identity storage for stating new user, to be updated to the phonetic feature list.
Another aspect according to an exemplary embodiment of the present invention, there is provided a kind of device for generating minutes, described device
Including：Parameter acquiring unit, the identity phase with the user of electronic equipment is obtained from each electronic equipment in multiple electronic equipments
The speech characteristic parameter answered, wherein, the electronic equipment that the multiple electronic equipment is each held for multiple users of participation meeting；
Data acquisition unit, gather the speech data in meeting；Recognition unit, it is special according to the voice obtained from the multiple electronic equipment
Parameter is levied to identify the identity of the spokesman of the speech data of collection；Converting unit, the speech data of collection is converted into text
Data, wherein, the label of the identity of the spokesman of the insertion instruction speech data in text data；Generation unit, it is based on
Insert the text data generation minutes of label.
Alternatively, the multiple electronic equipment may include described in the predetermined electronic equipment for being arranged to WIFI hot spot and access
At least one other electronic equipment of WIFI hot spot, wherein, parameter acquiring unit can be from least one other electronic equipment
Obtain speech characteristic parameter corresponding with the identity of the user of electronic equipment.
Alternatively, the multiple electronic equipment may include the predetermined electronic equipment comprising described device and except the predetermined electricity
At least one other electronic equipment outside sub- equipment, wherein, parameter acquiring unit can be worked as from least one other electronics
When equipment receives predetermined message, obtained from least one other electronic equipment corresponding to the identity of the user of electronic equipment
Speech characteristic parameter, wherein, the predetermined message may indicate that the voice that is stored at least one other electronic equipment is special
Sign parameter is arranged to shared file for the predetermined electronic equipment.
Alternatively, parameter acquiring unit can also obtain the user with the predetermined electronic equipment from the predetermined electronic equipment
The corresponding speech characteristic parameter of identity.
Alternatively, recognition unit may include：Subelement is extracted, extracts the speech characteristic parameter of the speech data of collection；Body
Part identification subelement, the phonetic feature matched with the speech characteristic parameter extracted is searched from the speech characteristic parameter of acquisition and is joined
Number, and the identity of user corresponding with the speech characteristic parameter of the speech characteristic parameter of extraction matching is defined as the voice number
According to spokesman identity.
Alternatively, identification subelement may include：List Generating Module, language is established according to the speech characteristic parameter of acquisition
Sound feature list, wherein, the phonetic feature list includes speech characteristic parameter and user corresponding with speech characteristic parameter
Identity；Searching modul, searched in phonetic feature list using the speech characteristic parameter of extraction and joined with the phonetic feature of extraction
The identity of user corresponding to number, the identity using the identity of the user found as the spokesman of the speech data.
Alternatively, parameter acquiring unit can be when there is new user to add meeting, and the electronics held from the new user is set
It is standby to obtain speech characteristic parameter corresponding with the identity of the new user, wherein, recognition unit can according to acquisition with it is described newly
The corresponding speech characteristic parameter of identity of user identifies the identity of the spokesman of the speech data of collection.
Alternatively, described device can also include：List update unit, will be obtained from the new electronic equipment with it is described new
Phonetic feature list is arrived in the corresponding speech characteristic parameter of identity of user and the identity storage of the new user, with to the voice
Feature list is updated, wherein, the phonetic feature list can be according to the phonetic feature obtained from the multiple electronic equipment
Parameter and establish.
Another further aspect according to an exemplary embodiment of the present invention, there is provided a kind of electronic equipment, it is characterised in that the electronics
Equipment includes：Processor；Memory, computer program is stored with, when the computer program is executed by processor, in realization
The method in electronic equipment generation minutes stated.
Another further aspect according to an exemplary embodiment of the present invention, there is provided a kind of to be stored with the computer-readable of computer program
Storage medium, realized when the computer program is being executed by processor above-mentioned in the side of electronic equipment generation minutes
Method.
Using the method and electronic equipment for generating minutes described in exemplary embodiment of the present in electronic equipment, energy
The speech characteristic parameter that is stored in enough electronic equipments held based on the user for participating in meeting carries out speech recognition, with will
View speech content is corresponding with specific speech object, automatically forms meeting text, improves the intelligent of minutes.
Brief description of the drawings
Pass through the detailed description carried out with reference to the accompanying drawing for being exemplarily illustrated embodiment, exemplary embodiment of the present
Above and other objects, features and advantages will become apparent, wherein：
Fig. 1 shows the flow chart of the method for generation minutes according to an exemplary embodiment of the present invention；
Fig. 2 shows the step of identity of the spokesman of the speech data of identification collection according to an exemplary embodiment of the present invention
Flow chart；
Fig. 3 shows the structure chart of the device of generation minutes according to an exemplary embodiment of the present invention；
Fig. 4 shows the structure chart of recognition unit according to an exemplary embodiment of the present invention；
Fig. 5 shows the structure chart of identification subelement according to an exemplary embodiment of the present invention.
Embodiment
Now, different example embodiments is more fully described with reference to the accompanying drawings, wherein, some exemplary embodiments are attached
Shown in figure.
Fig. 1 shows the flow chart of the method for generation minutes according to an exemplary embodiment of the present invention.
Reference picture 1, in step slo, the use with electronic equipment is obtained from each electronic equipment in multiple electronic equipments
The corresponding speech characteristic parameter of identity at family.Here, the multiple electronic equipment is each held to participate in multiple users of meeting
Electronic equipment.As an example, speech characteristic parameter can refer to vocal print feature.
Preferably, the corresponding speech characteristic parameter of the identity of the user with electronic equipment in each electronic equipment can pass through
Study to the speech data of the user of electronic equipment and obtain, to improve the identification of the voice to the user of the electronic equipment
Precision.
As an example, the multiple electronic equipment can be located within a predetermined range, for example, the preset range can refer to make
The multiple electronic equipment meets the scope (as in a meeting room) of closely connection request, and before a conference begins from described
Each electronic equipment in multiple electronic equipments obtains speech characteristic parameter corresponding with the identity of the user of electronic equipment.
It should be understood that the multiple electronic equipment can be the stationary electronic devices and/or mobile electron that can be recorded
Equipment, as an example, mobile electronic device may include notebook, smart mobile phone, Intelligent worn device, tablet personal computer, individual number
Word assistant, digital multimedia player etc..
A kind of situation, the multiple electronic equipment may include described in the predetermined electronic equipment for being arranged to WIFI hot spot and access
At least one other electronic equipment of WIFI hot spot.For example, predetermined electronic equipment can be arranged to WIFI hot spot, should by access
The electronic equipment of WIFI hot spot is defined as at least one other electronic equipment.
In the case, can be obtained in step slo from least one other electronic equipment for being linked into WIFI hot spot
Take speech characteristic parameter corresponding with the identity of the user of electronic equipment.Preferably, can also be obtained from the predetermined electronic equipment
Speech characteristic parameter corresponding with the identity of the user of the predetermined electronic equipment, now, what step S10 was obtained is predetermined electronics
Speech characteristic parameter in equipment and at least one other electronic equipment.
Preferably, at least one other electronic equipment and the predetermined electronic equipment (that is, can exist before a conference begins
Before step S10) the corresponding speech characteristic parameter of the identity of the user with electronic equipment each stored is arranged to shared text
Part, so that the equipment for performing the method for above-mentioned generation minutes can be from least one other electronic equipment and/or described
Predetermined electronic equipment obtains speech characteristic parameter.
It should be understood that now the method for the generation minutes shown in Fig. 1 can perform in the predetermined electronic equipment, also may be used
Performed in specified electronic equipment at least one other electronic equipment, can also access the WIFI hot spot except described
Performed in conference terminal outside multiple electronic equipments.
Another situation, the multiple electronic equipment may include the predetermined of the method for performing above-mentioned generation minutes
Electronic equipment and at least one other electronic equipment in addition to the predetermined electronic equipment.
In the case, at least one other electronic equipment can incite somebody to action (that is, before step S10) before a conference begins
Speech characteristic parameter corresponding with electronic equipment user identity that is each storing is arranged to shared for predetermined electronic equipment
File, and at least one other electronic equipment is sent in advance to the predetermined electronic equipment after above-mentioned shared setting is completed
Determine message.
In step slo when the predetermined electronic equipment receives described make a reservation for from least one other electronic equipment
During message, phonetic feature ginseng corresponding with the identity of the user of electronic equipment is obtained from least one other electronic equipment
Number.Here, the predetermined message may indicate that the speech characteristic parameter stored at least one other electronic equipment is directed to institute
State predetermined electronic equipment and be arranged to shared file.That is, at least one other electronic equipment is set to predetermined electronics
The standby equipment that have sent predetermined message.
Preferably, can also be obtained from the predetermined electronic equipment corresponding with the identity of the user of the predetermined electronic equipment
Speech characteristic parameter, now, what step S10 was obtained is the language in predetermined electronic equipment and at least one other electronic equipment
Sound characteristic parameter.As an example, use that can be using the filename of the device name of electronic equipment or shared file as electronic equipment
The identity at family, for example, the file of device name and shared file can be defined with the name or code name of the user of electronic equipment
Name.Now the device name of electronic equipment or the filename of shared file are also obtained while speech characteristic parameter is obtained.
Closely connected for example, can be established between predetermined electronic equipment and at least one other electronic equipment, make a reservation for electricity
Sub- equipment is closely connected to be obtained from least one other electronic equipment with the user's of electronic equipment by foundation
The corresponding speech characteristic parameter of identity.As an example, closely connection may include that bluetooth, WIFI, WIFI are direct-connected.
In step S20, the speech data in meeting is gathered.For example, using audio collecting device (such as microphone, Mike
Wind) gather the speech data in meeting.
In step s 30, the voice of collection is identified according to the speech characteristic parameter obtained from the multiple electronic equipment
The identity of the spokesman of data.
The step of referring to Fig. 2 to introduce the identity of the spokesman of the speech data of identification collection.
Fig. 2 shows the step of identity of the spokesman of the speech data of identification collection according to an exemplary embodiment of the present invention
Flow chart.
Reference picture 2, in step S301, extract the speech characteristic parameter of the speech data of collection.Here, using existing
Various modes speech characteristic parameter is extracted from speech data.
After speech characteristic parameter is extracted, it can be searched in the speech characteristic parameter obtained from multiple electronic equipments
The speech characteristic parameter matched with the speech characteristic parameter of extraction, and the phonetic feature that the speech characteristic parameter with extraction is matched
The identity of user corresponding to parameter is defined as the identity of the spokesman of the speech data of collection.
Particularly, in step s 302, can be according to the voice obtained from each electronic equipment in multiple electronic equipments
Characteristic parameter establishes phonetic feature list.Here, the phonetic feature list may include speech characteristic parameter and special with voice
Levy the identity of user corresponding to parameter.
For example, can be based on the phonetic feature ginseng obtained from least one other electronic equipment in the multiple electronic equipment
Number establish phonetic feature list, or may be based on from the predetermined electronic equipment in the multiple electronic equipment and it is at least one its
Speech characteristic parameter that his electronic equipment obtains establishes phonetic feature list.
In step S303, searched and the voice of extraction spy in phonetic feature list using the speech characteristic parameter of extraction
The identity of user corresponding to parameter is levied, the identity using the identity of the user found as the spokesman of the speech data of collection.
It should be understood that the step of identity of the spokesman for the speech data that the identification shown in Fig. 2 gathers merely illustrative, this area
Technical staff can also identify the language of collection based on the speech characteristic parameter obtained from multiple electronic equipments by other means
The identity of the spokesman of sound data.
Fig. 1 is returned, in step s 40, the speech data of collection is converted into text data.Here, can be in text data
The label of the identity of the spokesman of the speech data of middle insertion instruction collection.
In step s 50, minutes are generated based on the text data for inserting label.That is, show in the present invention
It is to establish speech content with the identity of spokesman to associate, so as to generate the body of speech content and spokesman in example property embodiment
The minutes of text data corresponding to part.
Preferably, the method for generation minutes according to an exemplary embodiment of the present invention can also include：Generation is provided
The step of minutes.
As an example, provide generation minutes the step of may include it is at least one in following steps：By generation
Minutes are sent at least one other electronic equipment, the minutes of generation are uploaded to cloud server, will give birth to
Into minutes be sent to designated terminal outside the multiple electronic equipment.
Preferably, the method for generation minutes according to an exemplary embodiment of the present invention can also include：When there is new user
When adding meeting, the electronic equipment held from the new user obtains phonetic feature ginseng corresponding with the identity of the new user
Number.Now, can be held in step s 30 according to the speech characteristic parameter obtained from multiple electronic equipments and from new user
Corresponding with the identity of the new user speech characteristic parameter that obtains of electronic equipment identify the hair of the speech data of collection
Say the identity of people.Here, it can determine there is new user to add meeting when it is determined that there is the electronic equipment of the WIFI hot spot of new access setting
The electronic equipment of new access WIFI hot spot, can now be defined as the electronic equipment that the new user is held by view, or also can be
When receiving new predetermined message, it is determined that there is new user to add meeting, the electronics for sending the new predetermined message can now be set
It is standby to be defined as the electronic equipment that the new user is held.
In the case, the method for generation minutes according to an exemplary embodiment of the present invention can also include：To establishing
Phonetic feature list the step of being updated.Particularly, can be by the obtain and new user from the new electronic equipment
The corresponding speech characteristic parameter of identity and the new user identity storage arrive the phonetic feature list, with to the voice
Feature list is updated.So, phonetic feature list can dynamically be updated according to the variation for the user for participating in meeting, even if
There is new user to add meeting in a meeting, the new user for adding meeting can be also identified, it is ensured that wrapped in phonetic feature list
Integrality, the real-time of the speech characteristic parameter contained so that the identification to the identity of spokesman is more accurate, effectively prevent to new
The problem of user's identification probability is low.
In addition, the method for generation minutes according to an exemplary embodiment of the present invention can also include：In response to meeting knot
Beam (for example, terminating collection speech data), delete the phonetic feature list of foundation.That is, after meeting adjourned, ginseng is deleted
Add the speech characteristic parameter of all users of meeting, avoid causing the leakage of speech characteristic parameter, improve security.It is same with this
When, it is preferable that in response to meeting adjourned, at least one other electronic equipment can be cancelled at least one other electronic equipment
The speech characteristic parameter of middle storage is shared.
Fig. 3 shows the structure chart of the device of generation minutes according to an exemplary embodiment of the present invention.
As shown in figure 3, the device of generation minutes according to an exemplary embodiment of the present invention includes：Parameter acquiring unit
10th, data acquisition unit 20, recognition unit 30, converting unit 40 and generation unit 50.
Particularly, parameter acquiring unit 10 obtains and electronic equipment from each electronic equipment in multiple electronic equipments
The corresponding speech characteristic parameter of identity of user.Here, to participate in, multiple users of meeting are each to control oneself the multiple electronic equipment
Some electronic equipments.As an example, speech characteristic parameter can refer to vocal print feature.
A kind of situation, the multiple electronic equipment may include described in the predetermined electronic equipment for being arranged to WIFI hot spot and access
At least one other electronic equipment of WIFI hot spot.For example, predetermined electronic equipment can be arranged to WIFI hot spot, should by access
The electronic equipment of WIFI hot spot is defined as at least one other electronic equipment.
In the case, parameter acquiring unit 10 can be from least one other electronic equipment for being linked into WIFI hot spot
Obtain speech characteristic parameter corresponding with the identity of the user of electronic equipment.Preferably, can also be obtained from the predetermined electronic equipment
Speech characteristic parameter corresponding with the identity of the user of the predetermined electronic equipment is taken, now, what parameter acquiring unit 10 obtained
It is the speech characteristic parameter in predetermined electronic equipment and at least one other electronic equipment.
Preferably, at least one other electronic equipment and the predetermined electronic equipment can be before a conference begins by each
The corresponding speech characteristic parameter of identity of the user with electronic equipment of storage is arranged to shared file, so that parameter acquiring unit
10 can obtain speech characteristic parameter from least one other electronic equipment and/or the predetermined electronic equipment.
It should be understood that now the device of the generation minutes shown in Fig. 3 may be provided in the predetermined electronic equipment, also may be used
It is arranged in the specified electronic equipment at least one other electronic equipment, can also be provided at and access removing for the WIFI hot spot
In conference terminal outside the multiple electronic equipment.
Another situation, the multiple electronic equipment may include the predetermined electronic equipment comprising described device and except described pre-
Determine at least one other electronic equipment outside electronic equipment.
In the case, at least one other electronic equipment will can each be stored and set with electronics before a conference begins
The corresponding speech characteristic parameter of identity of standby user is arranged to shared file for predetermined electronic equipment, and above-mentioned common in completion
At least one other electronic equipment sends predetermined message to the predetermined electronic equipment after enjoying setting.
When parameter acquiring unit 10 receives the predetermined message from least one other electronic equipment, parameter obtains
Unit 10 is taken to obtain phonetic feature ginseng corresponding with the identity of the user of electronic equipment from least one other electronic equipment
Number.Here, the predetermined message may indicate that the speech characteristic parameter stored at least one other electronic equipment is directed to institute
State predetermined electronic equipment and be arranged to shared file.That is, at least one other electronic equipment is set to predetermined electronics
The equipment that standby (that is, parameter acquiring unit 10) have sent predetermined message.
Preferably, parameter acquiring unit 10 can also obtain the use with the predetermined electronic equipment from the predetermined electronic equipment
The corresponding speech characteristic parameter of identity at family, now, parameter acquiring unit 10 obtain be predetermined electronic equipment and it is described at least
Speech characteristic parameter in one other electronic equipment.As an example, can be by the device name of electronic equipment or shared file
Identity of the filename as the user of electronic equipment, for example, can be set with the name or code name of the user of electronic equipment to define
The filename of standby title and shared file.Now parameter acquiring unit 10 also obtains electronics while speech characteristic parameter is obtained
The device name of equipment or the filename of shared file.
For example, the device of generation minutes according to an exemplary embodiment of the present invention can also include communication connection unit,
Closely connected for being established between predetermined electronic equipment and at least one other electronic equipment, parameter acquiring unit 10 passes through
The closely connection established comes corresponding with the identity of the user of electronic equipment from least one other electronic equipment acquisition
Speech characteristic parameter.As an example, closely connection may include that bluetooth, WIFI, WIFI are direct-connected.
Data acquisition unit 20 gathers the speech data in meeting.
Recognition unit 30 identifies the voice number of collection according to the speech characteristic parameter obtained from the multiple electronic equipment
According to spokesman identity.
The process of the identity of the spokesman of the speech data of the identification collection of recognition unit 30 is introduced referring to Fig. 4.
Fig. 4 shows the structure chart of recognition unit according to an exemplary embodiment of the present invention.
As shown in figure 4, recognition unit according to an exemplary embodiment of the present invention may include that extracting subelement 31 and identity knows
Small pin for the case unit 32.
Particularly, the speech characteristic parameter of the speech data of the extraction collection of subelement 31 is extracted.Here, subelement is extracted
31 can be the existing various devices that speech characteristic parameter can be extracted from speech data.
After speech characteristic parameter is extracted, identification subelement 32 can be searched from the speech characteristic parameter of acquisition
The speech characteristic parameter matched with the speech characteristic parameter of extraction, and the phonetic feature that the speech characteristic parameter with extraction is matched
The identity of user corresponding to parameter is defined as the identity of the spokesman of the speech data of collection.
Referring to the mistake of the identity of the spokesman of Fig. 5 speech datas gathered to introduce identification subelement 32 to determine
Journey.
Fig. 5 shows the structure chart of identification subelement according to an exemplary embodiment of the present invention.
As shown in figure 5, identification subelement 32 according to an exemplary embodiment of the present invention may include List Generating Module
321 and searching modul 322.
Particularly, generation module 321 can be special according to the voice obtained from each electronic equipment in multiple electronic equipments
Sign parameter establishes phonetic feature list.Here, the phonetic feature list may include speech characteristic parameter and and phonetic feature
The identity of user corresponding to parameter.
For example, generation module 321 can based on from the multiple electronic equipment at least one other electronic equipment obtain
Speech characteristic parameter establish phonetic feature list, or may be based on from the predetermined electronic equipment in the multiple electronic equipment
The speech characteristic parameter obtained with least one other electronic equipment establishes phonetic feature list.
Searching modul 322 searches the phonetic feature with extraction using the speech characteristic parameter of extraction in phonetic feature list
The identity of user corresponding to parameter, the identity using the identity of the user found as the spokesman of the speech data.
It should be understood that the mistake of the identity of the spokesman for the speech data that the identification of identification subelement 32 shown in Fig. 5 gathers
Journey is merely illustrative, and those skilled in the art can also come based on the phonetic feature ginseng obtained from multiple electronic equipments by other means
Count to identify the identity of the spokesman of the speech data of collection.
Fig. 3 is returned to, the speech data of collection is converted to text data by converting unit 40.Here, can be in text data
The label of the identity of the spokesman of the speech data of insertion instruction collection.
Generation unit 50 generates minutes based on the text data for inserting label.That is, in example of the present invention
Property embodiment in, converting unit 40 is to establish the identity of make a speech content and spokesman associate, so as to which the generation of generation unit 50 is sent out
Say the minutes of content text data corresponding with the identity of spokesman.
Preferably, the device of generation minutes according to an exemplary embodiment of the present invention can also include：Transmitting element, use
In the minutes for providing generation.
As an example, transmitting element can in the following manner at least one provide the minutes of generation：Will be raw
Into minutes be sent at least one other electronic equipment, the minutes of generation uploaded to cloud server,
The minutes of generation are sent to the designated terminal outside the multiple electronic equipment.
Preferably, the device of generation minutes according to an exemplary embodiment of the present invention can also include：Determining unit, use
In it is determined that whether there is new user to add meeting in conference process.For example, determining unit can be it is determined that have what new access was set
Determine there is new user to add meeting during the electronic equipment of WIFI hot spot, the electronic equipment of new access WIFI hot spot can now be determined
The electronic equipment held for the new user.Or determining unit also can be when receiving new predetermined message, it is determined that there is new use
Family adds meeting, and the electronic equipment for sending the new predetermined message can be now defined as to the electronics that the new user is held and set
It is standby.
When the electricity that determining unit is determined when having new user to add meeting, and parameter acquiring unit 10 is held from the new user
Sub- equipment obtains speech characteristic parameter corresponding with the identity of the new user.Now, can be according to from multiple in recognition unit 30
The speech characteristic parameter that is obtained in electronic equipment and the electronic equipment held from new user obtain with the new user's
Identity corresponding speech characteristic parameter identifies the identity of the spokesman of the speech data of collection.
In the case, the device of generation minutes according to an exemplary embodiment of the present invention can also include：List is more
New unit, for being updated to the phonetic feature list of foundation.Particularly, will can be obtained from the new electronic equipment with
The phonetic feature list is arrived in the corresponding speech characteristic parameter of identity of the new user and the identity storage of the new user, with
The phonetic feature list is updated.
In addition, the device of generation minutes according to an exemplary embodiment of the present invention can also include：Delete unit, response
In meeting adjourned (for example, terminate collection speech data), the phonetic feature list of foundation is deleted.That is, meeting adjourned
Afterwards, the speech characteristic parameter that element deletion participates in all users of meeting is deleted, avoids causing the leakage of speech characteristic parameter, carries
High security.At the same time, it is preferable that in response to meeting adjourned, at least one other electronic equipment can cancel to it is described at least
The speech characteristic parameter stored in one other electronic equipment is shared.
A kind of computing device is also provided according to the exemplary embodiment of the present invention.The computing device includes processor and storage
Device.Memory is used to store computer program.The computer program is executed by processor so that computing device is as described above
Generation minutes method computer program.
A kind of computer-readable storage medium for being stored with computer program is also provided according to the exemplary embodiment of the present invention
Matter.The computer-readable recording medium storage has causes the above-mentioned generation minutes of computing device when being executed by a processor
The computer program of method.The computer readable recording medium storing program for performing is the arbitrary data that can store the data read by computer system
Storage device.The example of computer readable recording medium storing program for performing includes：Read-only storage, random access memory, read-only optical disc, magnetic
Band, floppy disk, optical data storage devices and carrier wave (data transfer for such as passing through internet through wired or wireless transmission path).
Using the method and apparatus of the generation minutes of exemplary embodiment of the present, can be based on participating in meeting
The electronic equipment held of user in the speech characteristic parameter that stores carry out speech recognition, by conference speech content and tool
Body speech object is corresponding, automatically forms meeting text, improves the intelligent of minutes.
In addition, the method and apparatus of the generation minutes using exemplary embodiment of the present, additionally it is possible to when meeting
When having the new user to add meeting in view, the speech characteristic parameter stored in the electronic equipment held based on new user is come dynamically more
New speech feature list, to identify the speech of new user.
In addition, the method and apparatus of the generation minutes using exemplary embodiment of the present, pass through in meeting
Before beginning, multiple respective speech characteristic parameters of electronic equipment are shared so that conference speech content and specific speech object are accurate
It is corresponding, the quality of minutes is improved, and phonetic feature list, the cancellation of multiple electronic equipments are deleted to it after meeting adjourned
Sharing for the speech characteristic parameter of storage, avoids the leakage of the speech characteristic parameter for the user for participating in meeting, improves safety
Property.
Although the present invention, those skilled in the art are particularly shown and described with reference to its exemplary embodiment
It should be understood that in the case where not departing from the spirit and scope of the present invention that claim is limited, form can be carried out to it
With the various changes in details.
Claims (11)
1. a kind of method for generating minutes, methods described include：
(A) phonetic feature corresponding with the identity of the user of electronic equipment is obtained from each electronic equipment in multiple electronic equipments
Parameter, wherein, the electronic equipment that the multiple electronic equipment is each held for multiple users of participation meeting；
(B) speech data in meeting is gathered；
(C) identify the spokesman's of the speech data of collection according to the speech characteristic parameter obtained from the multiple electronic equipment
Identity；
(D) speech data of collection is converted into text data, wherein, the insertion instruction speech data in text data
The label of the identity of spokesman；
(E) minutes are generated based on the text data for inserting label.
2. according to the method for claim 1, it is characterised in that the multiple electronic equipment includes being arranged to WIFI hot spot
Predetermined electronic equipment and at least one other electronic equipment for accessing the WIFI hot spot,
Wherein, step (A) includes：
Speech characteristic parameter corresponding with the identity of the user of electronic equipment is obtained from least one other electronic equipment.
3. according to the method for claim 1, it is characterised in that the multiple electronic equipment includes being used to perform methods described
Predetermined electronic equipment and at least one other electronic equipment in addition to the predetermined electronic equipment,
Wherein, step (A) includes：
When receiving predetermined message from least one other electronic equipment, obtained from least one other electronic equipment
Speech characteristic parameter corresponding with the identity of the user of electronic equipment is taken, wherein, the predetermined message instruction is described at least one
The speech characteristic parameter stored in other electronic equipments is arranged to shared file for the predetermined electronic equipment.
4. according to the method in claim 2 or 3, it is characterised in that step (A) also includes：From the predetermined electronic equipment
Obtain speech characteristic parameter corresponding with the identity of the user of the predetermined electronic equipment.
5. according to the method for claim 1, it is characterised in that step (C) includes：
(C1) speech characteristic parameter of the speech data of extraction collection；
(C2) speech characteristic parameter matched with the speech characteristic parameter extracted is searched from the speech characteristic parameter of acquisition, and will
The identity of user corresponding with the speech characteristic parameter that the speech characteristic parameter of extraction matches is defined as the hair of the speech data
Say the identity of people.
6. according to the method for claim 5, it is characterised in that step (C2) includes：
Phonetic feature list is established according to the speech characteristic parameter of acquisition, wherein, the phonetic feature list includes phonetic feature
Parameter and the identity of user corresponding with speech characteristic parameter；
User corresponding with the speech characteristic parameter extracted is searched in phonetic feature list using the speech characteristic parameter of extraction
Identity, the identity using the identity of the user found as the spokesman of the speech data.
7. according to the method for claim 1, it is characterised in that methods described also includes：
When there is new user to add meeting, the electronic equipment held from the new user obtains the identity phase with the new user
The speech characteristic parameter answered,
Wherein, step (C) also includes：
The speech of the speech data of collection is identified according to the speech characteristic parameter corresponding with the identity of the new user of acquisition
The identity of people.
8. according to the method for claim 7, it is characterised in that methods described also includes：
Speech characteristic parameter according to being obtained from the multiple electronic equipment establishes phonetic feature list；
By the speech characteristic parameter corresponding with the identity of the new user obtained from the new electronic equipment and the new user
Identity storage arrive the phonetic feature list, to be updated to the phonetic feature list.
9. a kind of device for generating minutes, described device include：
Parameter acquiring unit, obtained from each electronic equipment in multiple electronic equipments corresponding to the identity of the user of electronic equipment
Speech characteristic parameter, wherein, the multiple electronic equipment is participates in the electronic equipment each held of multiple users of meeting；
Data acquisition unit, gather the speech data in meeting；
Recognition unit, the hair of the speech data of collection is identified according to the speech characteristic parameter obtained from the multiple electronic equipment
Say the identity of people；
Converting unit, the speech data of collection is converted into text data, wherein, the insertion instruction voice in text data
The label of the identity of the spokesman of data；
Generation unit, minutes are generated based on the text data for inserting label.
10. a kind of computing device, it is characterised in that the computing device includes：
Processor；
Memory, computer program is stored with, when the computer program is executed by processor, realized in claim 1 to 8
Any one described in method.
A kind of 11. computer-readable recording medium for being stored with computer program, when the computer program is held by processor
The method described in any one in claim 1 to 8 is realized during row.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711174526.1A CN107862071A (en) | 2017-11-22 | 2017-11-22 | The method and apparatus for generating minutes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711174526.1A CN107862071A (en) | 2017-11-22 | 2017-11-22 | The method and apparatus for generating minutes |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107862071A true CN107862071A (en) | 2018-03-30 |
Family
ID=61703318
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711174526.1A Pending CN107862071A (en) | 2017-11-22 | 2017-11-22 | The method and apparatus for generating minutes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107862071A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108399923A (en) * | 2018-02-01 | 2018-08-14 | 深圳市鹰硕技术有限公司 | More human hairs call the turn spokesman's recognition methods and device |
| CN108986826A (en) * | 2018-08-14 | 2018-12-11 | 中国平安人寿保险股份有限公司 | Automatically generate method, electronic device and the readable storage medium storing program for executing of minutes |
| CN109741754A (en) * | 2018-12-10 | 2019-05-10 | 上海思创华信信息技术有限公司 | A kind of conference voice recognition methods and system, storage medium and terminal |
| CN110767235A (en) * | 2019-11-14 | 2020-02-07 | 北京中电慧声科技有限公司 | Voice transcription processing device with role separation function and control method |
| CN111667837A (en) * | 2019-02-21 | 2020-09-15 | 奇酷互联网络科技（深圳）有限公司 | Conference record acquisition method, intelligent terminal and device with storage function |
| CN112970061A (en) * | 2018-11-14 | 2021-06-15 | 惠普发展公司，有限责任合伙企业 | Policy license based content |
| CN113517002A (en) * | 2020-03-25 | 2021-10-19 | 钉钉控股（开曼）有限公司 | Information processing method, device and system, conference terminal and server |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102572372A (en) * | 2011-12-28 | 2012-07-11 | 中兴通讯股份有限公司 | Extraction method and device for conference summary |
| CN103000211A (en) * | 2011-09-17 | 2013-03-27 | 闫天时 | High-definition audio playing system capable of realizing multi-resource sharing |
| CN203492206U (en) * | 2013-10-12 | 2014-03-19 | 网易（杭州）网络有限公司 | Electronic device |
| CN104243586A (en) * | 2014-09-18 | 2014-12-24 | 北京奇虎科技有限公司 | Content sharing method, device and equipment |
| CN105389099A (en) * | 2014-08-29 | 2016-03-09 | 三星电子株式会社 | method and apparatus for voice recording and playback |
| CN106453481A (en) * | 2015-08-04 | 2017-02-22 | 三星电子株式会社 | Electronic device and method for providing service in the electronic device |
| US10129392B1 (en) * | 2017-08-25 | 2018-11-13 | Global Tel*Link Corporation | Systems and methods for detecting inmate to inmate conference calls |
-
2017
- 2017-11-22 CN CN201711174526.1A patent/CN107862071A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103000211A (en) * | 2011-09-17 | 2013-03-27 | 闫天时 | High-definition audio playing system capable of realizing multi-resource sharing |
| CN102572372A (en) * | 2011-12-28 | 2012-07-11 | 中兴通讯股份有限公司 | Extraction method and device for conference summary |
| CN203492206U (en) * | 2013-10-12 | 2014-03-19 | 网易（杭州）网络有限公司 | Electronic device |
| CN105389099A (en) * | 2014-08-29 | 2016-03-09 | 三星电子株式会社 | method and apparatus for voice recording and playback |
| CN104243586A (en) * | 2014-09-18 | 2014-12-24 | 北京奇虎科技有限公司 | Content sharing method, device and equipment |
| CN106453481A (en) * | 2015-08-04 | 2017-02-22 | 三星电子株式会社 | Electronic device and method for providing service in the electronic device |
| US10129392B1 (en) * | 2017-08-25 | 2018-11-13 | Global Tel*Link Corporation | Systems and methods for detecting inmate to inmate conference calls |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108399923A (en) * | 2018-02-01 | 2018-08-14 | 深圳市鹰硕技术有限公司 | More human hairs call the turn spokesman's recognition methods and device |
| CN108986826A (en) * | 2018-08-14 | 2018-12-11 | 中国平安人寿保险股份有限公司 | Automatically generate method, electronic device and the readable storage medium storing program for executing of minutes |
| CN112970061A (en) * | 2018-11-14 | 2021-06-15 | 惠普发展公司，有限责任合伙企业 | Policy license based content |
| CN109741754A (en) * | 2018-12-10 | 2019-05-10 | 上海思创华信信息技术有限公司 | A kind of conference voice recognition methods and system, storage medium and terminal |
| CN111667837A (en) * | 2019-02-21 | 2020-09-15 | 奇酷互联网络科技（深圳）有限公司 | Conference record acquisition method, intelligent terminal and device with storage function |
| CN110767235A (en) * | 2019-11-14 | 2020-02-07 | 北京中电慧声科技有限公司 | Voice transcription processing device with role separation function and control method |
| CN113517002A (en) * | 2020-03-25 | 2021-10-19 | 钉钉控股（开曼）有限公司 | Information processing method, device and system, conference terminal and server |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107862071A (en) | The method and apparatus for generating minutes | |
| US20190088262A1 (en) | Method and apparatus for pushing information | |
| CN107562723A (en) | Meeting processing method, medium, device and computing device | |
| CN106897372B (en) | Voice query method and device | |
| US20200066254A1 (en) | Spoken dialog system, spoken dialog device, user terminal, and spoken dialog method | |
| CN109951743A (en) | Barrage information processing method, system and computer equipment | |
| CN109842712B (en) | Call record generation method and device, computer equipment and storage medium | |
| CN106789543A (en) | The method and apparatus that facial expression image sends are realized in session | |
| CN110536166B (en) | Interactive triggering method, device and equipment of live application program and storage medium | |
| CN107463636B (en) | Voice interaction data configuration method and device and computer readable storage medium | |
| US20230268073A1 (en) | Inquiry information processing method and apparatus, and medium | |
| CN107071554B (en) | Method for recognizing semantics and device | |
| CN107943914A (en) | Voice information processing method and device | |
| CN108334498A (en) | Method and apparatus for handling voice request | |
| CN109493866A (en) | Intelligent sound box and its operating method | |
| CN108629241B (en) | Data processing method and data processing equipment | |
| CN109428859A (en) | A kind of synchronized communication method, terminal and server | |
| CN112653902A (en) | Speaker recognition method and device and electronic equipment | |
| CN109271503A (en) | Intelligent answer method, apparatus, equipment and storage medium | |
| CN109684978A (en) | Employees'Emotions monitoring method, device, computer equipment and storage medium | |
| JPWO2019073669A1 (en) | Information processing equipment, information processing methods, and programs | |
| CN106407287A (en) | Multimedia resource pushing method and system | |
| KR20150041592A (en) | Method for updating contact information in callee electronic device, and the electronic device | |
| CN110427099A (en) | Information recording method, device, system, electronic equipment and information acquisition method | |
| CN111460085A (en) | Knowledge point recommendation method and device based on graph structure |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180330 |
|
| RJ01 | Rejection of invention patent application after publication |