KR102492008B1 - Apparatus for managing minutes and method thereof - Google Patents

Abstract

A meeting minutes management method and apparatus in which metadata is mapped to words corresponding to minimum recognition units are provided. A method for managing meeting minutes according to an embodiment of the present invention, performed by a computing device, includes obtaining speech data including speech text and mapping meta data to key words of the speech text. can

Description

Meeting minutes management method and device {APPARATUS FOR MANAGING MINUTES AND METHOD THEREOF}

The present invention relates to a method and apparatus for managing meeting minutes. More specifically, it relates to a method and apparatus for mapping meta data to words corresponding to a minimum recognition unit.

The proceedings and texts of remarks of the meetings held by the National Assembly, metropolitan councils or local councils are recorded by the secretary. And, it is common that the minutes recorded in this way are disclosed to the general public.

The conventional method of recording meeting minutes is written based on information heard by the clerk, so there may be errors, and there are disadvantages in that excessive costs and manpower are consumed. In addition, the conventional meeting minutes management method has a problem in that it is difficult for ordinary people with relatively low legal knowledge to intuitively recognize the speaker's intention to speak from the meeting minutes.

Korean Patent Registration No. 10-1618084

A technical problem to be solved by the present invention is to provide a method and apparatus capable of managing meeting minutes by mapping metadata to words corresponding to a minimum recognition unit.

Another technical problem to be solved by the present invention is to provide a method and apparatus for managing meeting minutes capable of preparing meeting minutes in an automated manner and easily modifying the meeting minutes.

Another technical problem to be solved by the present invention is to provide a method and apparatus for managing meeting minutes that provide an interface that can be more intuitively recognized by the general public who reads meeting minutes.

The technical problems of the present invention are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

In order to solve the above technical problem, a method for managing meeting minutes according to an embodiment of the present invention is a method performed by a computing device, comprising the steps of obtaining speech data including speech text structured to enable identification of key words. , mapping meta data to the key words, and generating meeting minutes data using the meta data.

In one embodiment, the obtaining of the speech data includes converting speech data into speech text by STT (Speech-To-Text) processing, and the meta data includes the STT for the main word. It may contain data related to processing.

In one embodiment, the obtaining of the speech data includes converting voice data in which speech of the conference is recorded into the speech data and obtaining a conversion accuracy predictive rate of the main word included in the speech data; The meta data may include the obtained conversion accuracy prediction rate, and generating the meeting record data may include highlighting and displaying words to be adjusted with the conversion accuracy prediction rate equal to or less than a reference value.

In an embodiment, the metadata of the word to be adjusted includes recorded voice data corresponding to the main word, and the generating of the meeting minutes data includes the voice included in the metadata of the word to be adjusted. It may include reproducing the data.

In an embodiment, the reproducing of the voice data may include reproducing the voice data of the target word to be adjusted and obtaining a corrected text for the target word to be adjusted repeatedly for each word to be adjusted. can include

In an embodiment, the reproducing of the voice data may include: reproducing the voice data of the adjustment target word selected by the user from among the adjustment target words; and corrected text for the adjustment target word. It may include a step of performing obtaining.

In an embodiment, the meta data includes a pronunciation required time for a utterance corresponding to the main word, and the generating of the meeting record data includes highlighting and displaying an emphasized word having a pronunciation required time equal to or greater than a reference value. can include

In one embodiment, the generating of the meeting minutes data includes receiving a user input and modifying the speech data using the user input, wherein the modifying the speech data comprises: a first word and mapping meta data of a second word, which is a key word closest to the first word, to the first word when being added to the speech text.

In one embodiment, generating the meeting minutes data may include visualizing the speech data according to a template referring to the meta data.

In one embodiment, the visualizing of the speech data may include retrieving a first word satisfying meta data defined in a first item of the template, and using the first word as a word corresponding to the first item. The step of determining and displaying the first word in an area corresponding to the first item may be included.

In one embodiment, a meeting minutes search step of receiving a first keyword through a search user interface and outputting together a search result of the first keyword and a search result of a second keyword mapped to an abbreviation of the first keyword And, the first keyword may be an official name of the second keyword.

In one embodiment, a meeting minutes search step of receiving a second keyword through a search user interface and outputting a search result of the second keyword and a search result of the first keyword mapped to an official name of the second keyword together and the second keyword may be an abbreviation of the first keyword.

In one embodiment, the meeting minutes management apparatus includes a processor, a network interface, a memory, and a computer program loaded into the memory and executed by the processor, wherein the computer program is structured to enable identification of key words. It may include an instruction for obtaining speech data including spoken speech text, an instruction for mapping meta data to the main words, and an instruction for generating meeting minutes data using the meta data.

1 is a diagram for explaining a meeting record management device according to an embodiment of the present invention.
2 is a flowchart of a method for managing meeting minutes according to an embodiment of the present invention.
3 is a diagram for explaining meta data mapping that may be referred to in some embodiments of the present invention.
FIG. 4 is a flowchart for explaining in detail some operations of the meeting minutes management method described with reference to FIG. 2 .
FIG. 5 is a flowchart for explaining in detail the operation of the meeting minutes management method described with reference to FIG. 4 .
FIG. 6 is an exemplary diagram illustrating the method for managing meeting minutes described with reference to FIG. 5 .
FIG. 7 is a flowchart for explaining in detail the operation of the method for managing meeting minutes described with reference to FIG. 2 .
FIG. 8 is a flowchart for explaining in detail the operation of the method for managing meeting minutes described with reference to FIG. 2 .
9 is an exemplary diagram illustrating the method for managing meeting minutes described with reference to FIG. 8 .
FIG. 10 is a flowchart for explaining in detail the operation of the method for managing meeting minutes described with reference to FIG. 2 .
FIG. 11 is a flowchart for explaining in detail the operation of the method for managing meeting minutes described with reference to FIG. 2 .
12 is a diagram for explaining an exemplary computing device capable of implementing a meeting record management device according to an embodiment of the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Advantages and features of the present invention, and methods of achieving them, will become clear with reference to the detailed description of the following embodiments taken in conjunction with the accompanying drawings. However, the technical idea of the present invention is not limited to the following embodiments and can be implemented in various different forms, only the following embodiments complete the technical idea of the present invention, and in the technical field to which the present invention belongs It is provided to fully inform those skilled in the art of the scope of the present invention, and the technical spirit of the present invention is only defined by the scope of the claims.

In adding reference numerals to components of each drawing, it should be noted that the same components have the same numerals as much as possible even if they are displayed on different drawings. In addition, in describing the present invention, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present invention, the detailed description will be omitted.

Unless otherwise defined, all terms (including technical and scientific terms) used in this specification may be used in a meaning commonly understood by those of ordinary skill in the art to which the present invention belongs. In addition, terms defined in commonly used dictionaries are not interpreted ideally or excessively unless explicitly specifically defined. Terminology used herein is for describing the embodiments and is not intended to limit the present invention. In this specification, singular forms also include plural forms unless specifically stated otherwise in a phrase.

In addition, in describing the components of the present invention, terms such as first, second, A, B, (a), and (b) may be used. These terms are only used to distinguish the component from other components, and the nature, order, or order of the corresponding component is not limited by the term. When an element is described as being “connected,” “coupled to,” or “connected” to another element, that element is directly connected or connectable to the other element, but there is another element between the elements. It will be understood that elements may be “connected”, “coupled” or “connected”.

As used herein, "comprises" and/or "comprising" means that a stated component, step, operation, and/or element is the presence of one or more other components, steps, operations, and/or elements. or do not rule out additions.

Prior to the description of this specification, some terms used in this specification will be clarified.

In this specification, a council is a procedure in which two or more participants exchange opinions and information on one or more proposed agendas and proceed with decision-making. Meetings according to an embodiment of the present invention may include, but are not limited to, plenary sessions, legislation and judicial committees, relevant committees, special committees, standing committees, administrative affairs audits and investigations, and governmental audits and investigations. In addition, participants in a meeting according to an embodiment of the present invention may include a chairperson or a chairperson, a member of the council or a member, but are not limited thereto.

Minutes are documents in which the progress of the meeting, texts of remarks by meeting participants, or meeting results are recorded. The meeting minutes according to an embodiment of the present invention may include, but are not limited to, the progress of the meeting, the utterance text, or the meeting result as text.

Hereinafter, some embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram for explaining a meeting record management apparatus 300 according to an embodiment of the present invention. Each component of the meeting minutes management system disclosed in FIG. 1 represents functionally differentiated functional elements, and any one or more components may be integrated and implemented in an actual physical environment.

Referring to FIG. 1 , the meeting minutes management system according to an embodiment of the present invention includes a meeting recording device 100, a log management server 200, a meeting minutes management device 300, and a meeting minutes reading device 400a, 400b, 400c or less, 400) may be included for convenience of explanation. Hereinafter, each component will be described in detail.

The meeting recording device 100 may generate and transmit data related to the meeting minutes to the meeting minutes managing device 300 . More specifically, the meeting recording device 100 may generate speech data. Here, the speech data refers to the first recorded or converted meeting minutes, and is string data including text-based speech text. The conference recording device 100 may generate speech data based on text data stenographed by the secretary of the conference. However, the meeting recorder 100 is not limited thereto, and the meeting recorder 100 may generate speech data transmitted by the log management server 200 by performing Speech-To-Text (STT) conversion on voice data in which the speech text of the meeting is recorded. . Also, the meeting recorder 100 may transmit the generated speech data to the meeting record management apparatus 300 .

In one embodiment, the conference recording device 100 may generate speech data by performing STT conversion on the audio data of the conference transmitted by the log management server 200 in real time.

Next, the log management server 200 may generate and transmit log data generated according to the progress of the conference to the meeting record management device 300 . More specifically, the log management server 200 may generate log data. Here, the log data is character string data in which a log of an event generated in the course of a conference is recorded. Such log data may include identifiers (IDs) of the plurality of microphones 201, 202, and 20n installed in the conference hall and activation times of the respective microphones 201, 202, and 20n, but are not limited thereto.

The log management server 200 may generate voice data recorded from the plurality of microphones 201 , 202 , and 20n installed in the conference hall and transmit it to the conference record management device 300 . However, it is not limited thereto and may be transmitted to the conference recording device 100 .

The plurality of microphones 201, 202, and 20n installed in the conference hall may be activated when a member of the assembly or member participating in the conference presses a button of the microphones 201, 202, and 20n to speak. In addition, the microphone may be automatically activated when a voice higher than a standard value is recognized. However, it is not limited thereto. When the microphones 201, 202, and 20n of members or members participating in the meeting are activated, the log management server 200 displays the identifiers of the activated microphones 201, 202, and 20n and the corresponding microphones 201, 202, and 20n. Log data may be generated by cumulatively recording the activation time. The identifier of the microphone included in the log data may be composed of a character string having a predetermined size, but is not limited thereto. And, the activation time of the microphone may have a Universal Time Coordinated (UTC) format, but is not limited thereto.

Next, the meeting record managing device 300 may manage meeting minute data based on the log data and the speech data received from the meeting recording device 100 and the log management server 200 . In addition, the meeting minutes managing device 300 may generate a document for viewing the meeting minutes based on the managed meeting minutes data, and transmit the document to the meeting minutes viewing device 400 . Such a meeting record management apparatus 300 according to some embodiments of the present invention will be described in detail later with reference to FIGS. 2 to 12 .

Next, the meeting minutes reading device 400 may receive and output a document for reading meeting minutes from the meeting minutes management device 300 . More specifically, the meeting minutes reading device 400 transmits a meeting minutes reading request to the meeting minutes management device 300 according to a user's input. Then, the meeting minutes reading device 400 receives a reading document from the meeting minutes management device 300 .

Here, the reading document is a document generated by the meeting minutes management device 300 at the request of the meeting minutes reading device 400 . Such a reading document may include the progress of the meeting, the speaking text, the speaking time, and the meeting result. In particular, unlike conventional meeting minutes, a reading document according to an embodiment of the present invention may additionally include speaker information, related keyword information, and the like. Such a view document may be a web document in a format such as HTML (HyperText Markup Language) or XML (eXtensible Markup Language), but is not limited thereto, and may be a PDF (Portable Document Format) document. In addition, an interface for keyword search may be provided in the browsing document according to an embodiment.

Then, the meeting minutes reading device 400 may output the received reading document on the screen. The meeting minutes reading device 400 may have a web browser or a dedicated application installed to output the reading document.

The meeting minutes reading device 400 according to an embodiment of the present invention may be any device capable of outputting a reading document received from the meeting minutes management device 300 through a network. For example, the meeting minutes reading device 400 according to an embodiment of the present invention includes a desktop, a workstation, a server, a laptop, a tablet, and a smart phone. ) or a phablet, but is not limited thereto, and is not limited thereto, such as a portable multimedia player (PMP), a personal digital assistant (PDA), or an e-book reader (E-Book Reader). ) and the like.

Referring to FIGS. 2 to 11, a method for managing meeting minutes according to an embodiment of the present invention will be described in more detail. Hereinafter, steps to be described in some flowcharts may be understood to be performed by the meeting record management device 300 unless otherwise specified.

2 is a flowchart of a method for managing meeting minutes according to an embodiment of the present invention.

In step S100, speech data including speech text is obtained. As described with reference to FIG. 1 , speech data may be generated and transmitted by the conference recording device 100 . However, it is not limited thereto. A detailed method related to generation of speech data will be described in detail with reference to FIG. 4 later.

In step S200, meta data is mapped to key words of speech text included in speech data. In one embodiment, the main word may mean all words included in the speech text. Also, in one embodiment, the main word may mean a noun included in the speech text. Also, in one embodiment, the main word may mean the beginning of a sentence. Also, in one embodiment, the main word may mean the ending of a sentence. However, it is not limited thereto, and all criteria for selecting words with a certain rule on the speech text may be included.

For a detailed description of mapped meta data, refer to FIG. 3 . 3 is a diagram for explaining meta data mapping that may be referred to in some embodiments of the present invention.

Referring to FIG. 3 , in the exemplified speech data, a string including speech text is described. The speech text includes the main word 30 to which at least one meta data is mapped.

In one embodiment, meta data may be mapped to at least one word among words of speech text. By selectively mapping meta data to some of words, which are minimum recognition units, it is possible to reduce the computing load while having the effects of some embodiments described below.

In one embodiment, meta data may be mapped to every word in the speech text. By mapping meta data to all words, which are the minimum recognition units, it is possible to provide a meeting minutes writer with an interface that is easy to modify and search, and a document that can more easily recognize the contents of the minutes.

3 illustrates key words 30 to which metadata is mapped. In one embodiment, the mapped metadata may include the transformation accuracy prediction rate 31 . A specific operation related to this will be described in detail with reference to FIGS. 4 to 6 to be described later. In one embodiment, mapped metadata may include voice data 32 . A specific operation related to this will be described in detail in FIG. 5 to be described later. In one embodiment, the mapped meta data may include a pronunciation duration 33 . A specific operation related to this will be described in detail with reference to FIGS. 8 and 9 to be described later. However, it is not limited thereto, and meta data may include speaker, meeting name, agenda, speech type, speech start time, and original file information. In one embodiment, meta data may include data related to STT processing for key words. For example, the conversion accuracy prediction rate of key words corresponds to this. However, it is not limited thereto, and all information for user convenience of the meeting minutes management device 300 may be included.

Referring back to FIG. 2 , meeting minutes data is generated in step S300. By mapping the meta data to key words, the work efficiency of the meeting minutes writer can be improved and the usability of the meeting minutes viewer can be improved. Step S300 is embodied in some embodiments to be described later.

The meeting minutes management method according to an embodiment of the present invention described with reference to FIGS. 2 and 3 maps meta data to words corresponding to a minimum recognition unit, thereby providing an interface that is easy to modify and search for meeting minutes writers, and meeting minutes A document that can more easily recognize the contents of the meeting minutes can be provided to the reader.

The method for managing meeting minutes according to an embodiment of the present invention maps meta data to key words, so that, in some embodiments to be described later, an environment in which an STT-converted speech text can be modified word by word can be provided to a meeting minutes writer. It is possible to provide meeting minutes that are easier to recognize by providing the reader with highlighted words, templates, keyword search, and the like.

Hereinafter, with reference to FIGS. 4 to 6, an embodiment according to the present invention using transformation accuracy prediction rate as meta data will be described in more detail. FIG. 4 is a flowchart for explaining in detail some operations of the meeting minutes management method described with reference to FIG. 2 .

Referring to FIG. 4 , the step of obtaining speech data including speech text (S100) is embodied. In step S110, voice data in which the conference speech is recorded is converted. The operation of step S110 may be performed by the meeting recorder 100 and transmitted to the meeting record management device 300 as described above with reference to FIG. 1 . However, it is not limited thereto, and step S110 may be performed by the meeting record management device 300 . The meeting minutes management device 300 may generate speech data transmitted by the log management server 200 by performing Speech-To-Text (STT) conversion on voice data in which speech text of a meeting is recorded. In one embodiment, the meeting minutes management device 300 may be generated by performing STT conversion on audio data of a conference transmitted by the log management server 200 in real time.

In step S130, a conversion accuracy prediction rate is obtained. The conversion accuracy prediction rate is an expression predicting how accurately the character string data converted by the STT conversion is converted from the original voice data. The STT transform and the algorithm for calculating the STT transform accuracy predictive rate are obvious to those skilled in the art and will be omitted.

In one embodiment, speech data may be generated automatically without shorthand by a scribe by generating speech data based on recorded voice data. Speech data written by the scribe's shorthand may have errors, excessive cost, and consumption of manpower, so these disadvantages can be overcome.

FIG. 5 is a flowchart for explaining in detail the operation of the meeting minutes management method described with reference to FIG. 4 .

Referring to FIG. 5 , step S130 is embodied in step S131. In step S131, a conversion accuracy prediction rate of the word is obtained. The transformation accuracy prediction rate of the STT transformation described above with reference to FIG. 4 is obtained for each word. In step S210, step S200 is embodied. In step S210, conversion accuracy prediction rates of key words are mapped.

When the mapped conversion accuracy prediction rate is less than or equal to the reference value (S311), the word to be adjusted is highlighted and displayed in step S313. By highlighting words with a low conversion accuracy prediction rate, the meeting minutes writer can easily recognize the part requiring correction.

Output for displaying the words to be adjusted may be performed through an output means separately provided in the meeting record management device 300 . Also, it can be performed through the meeting recording device 100 . However, it is not limited thereto.

In one embodiment, the reference value of the conversion accuracy prediction rate may be set to a plurality of sections. For example, if the conversion accuracy prediction rate corresponds to 80% or more, black is assigned, if it corresponds to 50% or more and less than 80%, orange is assigned, and if it corresponds to less than 50%, red is assigned. . As it is set to a plurality of sections, the meeting minutes writer can easily recognize the part requiring further correction (red color in this example). However, it is not limited to this example.

In one embodiment, conversion accuracy prediction rates may be obtained for only some words of speech text of speech data. As described above, since the computational load can be reduced by mapping meta data to some words, the computational load can be reduced by obtaining conversion accuracy prediction rates only for some words.

In one embodiment, a transformation accuracy prediction rate may be obtained for every word in the speech text. In the case of a document requiring high accuracy in disclosure, conversion accuracy prediction rates are obtained for all words, and a region marked with a low conversion accuracy prediction rate may be corrected.

In one embodiment, it may operate together with an embodiment in which voice data corresponding to a word to be described with reference to FIG. 7 is output. A detailed description of this will be described later with reference to FIG. 7 .

FIG. 6 is an exemplary diagram illustrating the method for managing meeting minutes described with reference to FIG. 5 .

Referring to FIG. 6 , words 61 and 63 having a low conversion accuracy prediction rate are displayed in a first color unlike other words in the speech text. As described above, by converting the colors of the words 61 and 63 having a low conversion accuracy prediction rate into the first color, the meeting minutes writer can easily recognize the part requiring correction.

In one embodiment, a word whose conversion accuracy prediction rate is equal to or less than a reference value may be displayed after being converted into a character format. For example, the font, thickness, degree of italics, etc. may be converted. However, it is not limited thereto, and all expressions displaying transformations that a meeting minutes writer can recognize may be included in the highlighting.

FIG. 7 is a flowchart for explaining in detail the operation of the method for managing meeting minutes described with reference to FIG. 2 .

Referring to FIG. 7 , step S200 is embodied in step S230. In step S230, voice data corresponding to words is mapped. That is, voice data transmitted by the log management server 200 is divided, and a portion of voice data corresponding to a specific word is mapped.

For example, referring to FIG. 6, the source of voice data corresponding to the text of the speech "Establishment of foundations such as railroads and roads for expansion of inter-Korean exchange and continental politics??" is divided, Voice data is mapped to the word 'North and South', and voice data corresponding to 'interchange' is mapped to the word 'interchange'.

Referring back to FIG. 7 , in step S331, an input for selecting a first word is obtained. An input for selecting the first word may be obtained through an input means separately provided in the meeting record management device 300 . In addition, it may be performed through the meeting minutes reading device 400 . However, it is not limited thereto.

In response to the input for selecting the first word in step S333, first voice data mapped to the first word is output. Voice data may be reproduced through an output unit provided separately in the meeting record management device 300 . Also, it can be reproduced through the meeting minutes reading device 400 . However, it is not limited thereto. By mapping meta data to words, there is an effect of reproducing voice data corresponding to the words.

In one embodiment, when the conversion accuracy prediction rate of the word described above with reference to FIGS. 4 to 6 corresponds to a reference value or less, in response to an input for selecting a word in addition to an embodiment of highlighting a word to be adjusted, the selected word An embodiment in which voice data mapped to may be reproduced may be used together. A word with a low conversion accuracy prediction rate may be selected, audio data corresponding to the word may be heard, and the meeting minutes writer may correct the word. This embodiment can provide a management interface that is easy to modify to a meeting minutes writer by specifying a word requiring correction and reproducing voice data corresponding to the word.

In an embodiment, reproducing voice data for each of a plurality of target words to be adjusted and repeatedly performing obtaining a corrected text for the target word to be adjusted may be included. It is possible to efficiently correct meeting minutes by continuously reproducing and correcting a plurality of words to be adjusted.

FIG. 8 is a flowchart for explaining in detail the operation of the method for managing meeting minutes described with reference to FIG. 2 .

Referring to FIG. 8 , step S200 is embodied in step S250. In step S250, the required pronunciation time for the utterance corresponding to the word is mapped. The pronunciation required time means the time required to pronounce a word in voice data. For example, referring to FIG. 9, the time at which 'South and North (91)' is pronounced is 1.22 seconds. The time at which 'AC (92)' is pronounced is 1.19 seconds.

Referring back to FIG. 8 , the meeting start time is obtained in step S351. The meeting start time may be obtained through an input unit provided separately in the meeting record management device 300 . In addition, an input may be received through the meeting record device 100 and transmitted to the meeting record management device 300 . However, it is not limited thereto.

In step S353, an input for selecting the first word is obtained. An input for selecting the first word may be obtained through an input means separately provided in the meeting record management device 300 . In addition, an input may be received through the meeting record device 100 and transmitted to the meeting record management device 300 . However, it is not limited thereto.

In step S355-1, step S355 of outputting the start time of the first word utterance is embodied. In step S355-1, the sum of pronunciation required times mapped to all words existing before the first word in the speech text is obtained.

In step S355-3, the sum of the required pronunciation times is added to the conference start time. If the pronunciation time required is mapped to all words in the speech text, the sum of the pronunciation time required for all words existing before the selected first word is obtained and the meeting start time is added to automatically obtain the first word utterance start time. there is.

In this embodiment, the start time of the first word utterance can be output using the meeting start time and the pronunciation required time obtained by processing the voice data without additionally inputting the utterance start time itself.

See FIG. 9 for a specific example related to this. 9 is an exemplary diagram illustrating the method for managing meeting minutes described with reference to FIG. 8 .

Referring to FIG. 9 , words 91 , 92 , 93 , and 94 to which word speaking times are mapped are shown. For simplicity of description, only some of the words to which word utterance times are mapped are shown. In an example to be described with reference to FIG. 9 , a speech start time is displayed for each paragraph. The speech start time 95 of the first paragraph shown is 15:08:01. The speech start time 97 of the illustrated second paragraph is 15:09:21. The speech start time 97 of the second paragraph is obtained by adding the speech start time 95 of the first paragraph to the sum of the pronunciation required times of all words described in the first paragraph. That is, according to the table, the total pronunciation required time for all words in the first paragraph corresponds to 80.02 seconds, so the speech start time 97 of the second paragraph is obtained by adding to the speech start time 95 of the first paragraph.

In one embodiment, speaker data is additionally included in meta data mapped to words, so that when the speaker data is changed, the changed speaker's speaking start time may be output. The time at which the speaking was performed for each speaker can be automatically obtained. In the automatically formatable meeting minutes to be provided according to some embodiments to be described later, the time at which a speech was made is displayed for each speaker without any additional input, so that the meeting minutes viewer can more easily recognize the meeting minutes.

In one embodiment, when the agenda data is changed by additionally including the agenda data in meta data mapped to words, the speech start time of the changed agenda may be output. The time at which a speech was performed for each agenda item can be automatically obtained. In the automatically formatable meeting minutes to be provided according to some embodiments to be described later, the time at which remarks were made for each agenda is displayed without any additional input, so that the meeting minutes viewer can more easily recognize the meeting minutes.

In one embodiment, the meta data may include a pronunciation required time for a utterance corresponding to a main word, and a step of highlighting and displaying an emphasized word having a pronunciation required time equal to or greater than a reference value.

Output for displaying the highlighted words may be performed through an output means separately provided in the meeting record management device 300 . Also, it may be performed through the meeting minutes management device 400 . However, it is not limited thereto.

Referring to FIG. 9 , for example, a stressed word 93 having a pronunciation time of 2 seconds or more, which is a reference value, is highlighted and displayed. In one embodiment, a word whose pronunciation time is equal to or greater than a reference value may be displayed after being converted into a character format. For example, the font, thickness, degree of italics, etc. may be converted. However, it is not limited thereto, and all expressions displaying transformations that can be recognized by a reader of meeting minutes may be included in the highlighting.

By highlighting words whose pronunciation time is higher than the standard value, the usability of the meeting minutes reader is improved. A viewer of the meeting minutes can understand the context of the entire meeting minutes based on the highlighted words without having to concentrate on reading all the words.

FIG. 10 is a flowchart for explaining in detail the operation of the method for managing meeting minutes described with reference to FIG. 2 .

Referring to FIG. 10 , speech data is modified in step S371. An input for correcting the speech data may be obtained through an input means separately provided in the meeting record management device 300 . It may also be obtained by the meeting recording device 100 . However, it is not limited thereto. Referring to some of the embodiments described above, the meeting minutes writer can modify the speech data.

Specifically, when a specific word is deleted from the speech text of the speech data, the mapped metadata of the deleted specific word is also deleted (S373).

When a specific word in the speech text of the speech data is substituted with another word, the meta data mapped to the substituted specific word is mapped to the substituted specific word and maintained (S375).

When a specific word is added in the speech text of the speech data, meta data of a key word closest to the added specific word is mapped (S377). An adjacent word may be the immediately preceding word. Also, it may be the word immediately following it. However, it is not limited thereto, and meta data mapped to the preceding or following words is mapped to the specific word to be added according to the position, length, etc. of the specific word to be added.

FIG. 11 is a flowchart for explaining in detail the operation of the method for managing meeting minutes described with reference to FIG. 2 .

Referring to FIG. 11 , keywords are extracted from speech text in step S391. A keyword means a word that can be used as an access point when searching speech text. In one embodiment, the keyword may correspond to the agenda name. However, it is not limited thereto.

In step S393, keywords are output according to the standard rule. Keywords may be output through an output unit provided separately in the meeting record management device 300 . In an embodiment, the reference rule may be an order of frequency of occurrence of keywords included in speech text. However, it is not limited thereto, and may include several rules capable of extracting keywords. The meeting minutes viewer can more easily recognize the meeting minutes by looking at the outputted keywords and reading the meeting minutes by concentrating on words with high importance.

In step S395, an interface for searching for keywords is provided. The interface may be provided through an output unit provided separately in the meeting record management device 300 . Also, it may be provided in the meeting minutes reading device 400 . However, it is not limited thereto. The input of the meeting minutes reader may be obtained through an input unit provided separately in the meeting minutes management device 300 . It can also be obtained by the meeting minutes browsing device 400 . However, it is not limited thereto.

In one embodiment, specific utterances including keyword portions in utterance text may be retrieved in order. In one embodiment, a specific utterance including a keyword portion may be searched for among utterance texts in related other conferences. In one embodiment, a second keyword associated with the first keyword may be output. In one embodiment, specific utterances containing a second keyword associated with a first keyword may be retrieved in order.

That is, various outputs and search interfaces may be provided through speech data obtained in a specific meeting and keywords obtained by extracting speech text included in the speech data. Keywords with high importance can be listed in order, utterances including the listed keywords can be searched for, and utterances included in other conferences associated with a specific conference can also be searched for. In addition, other keywords associated with a specific keyword may be output.

In step S397, the search result is output. The search result may be provided through an output unit provided separately in the meeting record management device 300 . Also, it may be provided by the meeting minutes reading device 400 . However, it is not limited thereto.

In one embodiment, a search result of a first keyword and a search result of a second keyword mapped to an abbreviation of the first keyword may be output together. Also, in one embodiment, a search result of the second keyword and a search result of the first keyword mapped to the official name of the second keyword may be output together. In one embodiment, the search result may be a collection of utterances containing the keyword. In one embodiment, the search result may be a collection of utterances for which the keyword is an agenda.

The meeting minutes management method according to an embodiment of the present invention may further include outputting a result value in which at least one or more of the mapped metadata is displayed. By displaying the mapped data in the speech data, a result value including speaker data, speech start time data, and agenda data can be output.

A viewer of meeting minutes can view a document automatically formatted with appropriately applied metadata, rather than meeting minutes in which the contents of a meeting are simply listed as strings. Therefore, the meeting minutes viewer can more easily recognize the meeting minutes.

In one embodiment, meeting minutes data may be generated by visualizing speech data according to a template referring to meta data. In one embodiment, a first word satisfying meta data defined in the first item of the template may be searched, and the first word determined as a word corresponding to the first item may be displayed. By arranging the first word according to the format of the template with reference to the meta data, the meeting minutes reader can easily recognize the contents of the meeting minutes.

In one embodiment, a document may be formatted by aggregating the statements of a particular member. In one embodiment, a document may be formatted by aggregating remarks about a particular agenda. In one embodiment, a document may be formatted by synthesizing a query utterance and a corresponding response utterance. However, it is not limited to this, and all methods that can be formatted based on the content included in the speech text are included.

So far, with reference to FIGS. 1 to 11 , the meeting minutes management method and application fields thereof according to an embodiment of the present invention have been described. Hereinafter, an exemplary computing device 1500 capable of implementing the meeting minutes management device 300 according to an embodiment of the present invention will be described.

12 is a hardware configuration diagram illustrating an exemplary computing device 1500 capable of implementing the meeting minutes management device 300 according to an embodiment of the present invention.

As shown in FIG. 12, the computing device 1500 loads one or more processors 1510, a bus 1550, a communication interface 1570, and a computer program 1591 executed by the processor 1510. It may include a memory 1530 for storing and a storage 1590 for storing the computer program 1591 . However, only components related to the embodiment of the present invention are shown in FIG. 12 . Accordingly, those skilled in the art to which the present invention pertains can know that other general-purpose components may be further included in addition to the components shown in FIG. 12 .

The processor 1510 controls the overall operation of each component of the computing device 1500 . The processor 1510 includes a Central Processing Unit (CPU), a Micro Processor Unit (MPU), a Micro Controller Unit (MCU), a Graphic Processing Unit (GPU), or any type of processor well known in the art. It can be. Also, the processor 1510 may perform an operation for at least one application or program for executing a method according to embodiments of the present invention. Computing device 1500 may include one or more processors.

Memory 1530 stores various data, commands and/or information. Memory 1530 may load one or more programs 1591 from storage 1590 to execute a method according to embodiments of the present invention. The memory 1530 may be implemented as a volatile memory such as RAM, but the technical scope of the present invention is not limited thereto.

The bus 1550 provides a communication function between components of the computing device 1500 . The bus 1550 may be implemented as various types of buses such as an address bus, a data bus, and a control bus.

The communication interface 1570 supports wired and wireless Internet communication of the computing device 1500 . Also, the communication interface 1570 may support various communication methods other than internet communication. To this end, the communication interface 1570 may include a communication module well known in the art.

According to some embodiments, communication interface 1570 may be omitted.

The storage 1590 may non-temporarily store the one or more programs 1591 and various data.

The storage 1590 may be a non-volatile memory such as read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, or the like, a hard disk, a removable disk, or a device well known in the art. It may be configured to include any known type of computer-readable recording medium.

Computer program 1591 may include one or more instructions that when loaded into memory 1530 cause processor 1510 to perform methods/operations in accordance with various embodiments of the invention. That is, the processor 1510 may perform methods/operations according to various embodiments of the present disclosure by executing the one or more instructions.

In the above case, the meeting minutes management device 300 according to an embodiment of the present invention may be implemented through the computing device 1500 .

So far, various embodiments of the present invention and effects according to the embodiments have been described with reference to FIGS. 1 to 12 . Effects according to the technical idea of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

The technical idea of the present invention described with reference to FIGS. 1 to 12 so far can be implemented as computer readable code on a computer readable medium. The computer-readable recording medium may be, for example, a removable recording medium (CD, DVD, Blu-ray disc, USB storage device, removable hard disk) or a fixed recording medium (ROM, RAM, computer-equipped hard disk). can The computer program recorded on the computer-readable recording medium may be transmitted to another computing device through a network such as the Internet, installed in the other computing device, and thus used in the other computing device.

In the above, even though all the components constituting the embodiment of the present invention have been described as being combined or operated as one, the technical spirit of the present invention is not necessarily limited to these embodiments. That is, within the scope of the object of the present invention, all of the components may be selectively combined with one or more to operate.

Although actions are shown in a particular order in the drawings, it should not be understood that the actions must be performed in the specific order shown or in a sequential order, or that all shown actions must be performed to obtain a desired result. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of the various components in the embodiments described above should not be understood as requiring such separation, and the described program components and systems may generally be integrated together into a single software product or packaged into multiple software products. It should be understood that there is

Although the embodiments of the present invention have been described with reference to the accompanying drawings, those skilled in the art can implement the present invention in other specific forms without changing the technical spirit or essential features. can understand that there is Therefore, the embodiments described above should be understood as illustrative in all respects and not limiting. The protection scope of the present invention should be construed according to the claims below, and all technical ideas within the equivalent range should be construed as being included in the scope of the technical ideas defined by the present invention.

Claims (13)

In a method performed by a computing device,
Obtaining speech data including speech text structured to enable identification of key words;
mapping meta data including a required pronunciation time for a utterance corresponding to the main word to the main word; and
Generating meeting minutes data using the metadata,
Generating the meeting minutes data,
Determining a word whose STT (Speech-To-Text) conversion accuracy prediction rate is less than a reference value among the key words as an adjustment target word;
reproducing recorded voice data corresponding to the word to be adjusted included in meta data of the word to be adjusted in response to a user's correction input;
If a first word is added to the speech text, mapping meta data of the first word based on meta data of a second word that is a key word closest to the position where the first word is added; and
Highlighting and displaying an emphasized word whose pronunciation time is equal to or greater than a reference value,
The displaying step is
In response to a selection input for a third word, automatically calculating a speech start time of the third word using pronunciation required times of all words existing before the third word; and
Including the step of displaying the utterance start time of the third word,
How to manage meeting minutes. According to claim 1,
Obtaining the speech data,
Converting speech data into speech text by STT (Speech-To-Text) processing;
The meta data includes data related to the STT processing for the main word,
How to manage meeting minutes. According to claim 1,
Obtaining the speech data,
converting voice data in which the speech of the conference is recorded into the speech data; and
Obtaining a conversion accuracy prediction rate of the main word included in the speech data,
The meta data is
Including the obtained conversion accuracy prediction rate,
Generating the meeting minutes data,
Highlighting and displaying a word to be adjusted for which the conversion accuracy prediction rate is less than or equal to a reference value,
How to manage meeting minutes. delete According to claim 1,
Reproducing the voice data,
Reproducing the audio data of the target word for each adjustment target word and obtaining a corrected text for the adjustment target word repeatedly.
How to manage meeting minutes. According to claim 1,
Reproducing the voice data,
Reproducing the voice data of the adjustment target word selected by a user from among the adjustment target words and obtaining a corrected text for the adjustment target word,
How to manage meeting minutes. delete delete According to claim 1,
Generating the meeting minutes data,
Including visualizing the speech data according to a template referring to the meta data,
How to manage meeting minutes. According to claim 9,
The step of visualizing the speech data,
Retrieving a first word that satisfies meta data defined in a first item of the template;
determining the first word as a word corresponding to the first item; and
Displaying the first word in an area corresponding to the first item,
How to manage meeting minutes. According to claim 1,
Further comprising a meeting minutes search step of receiving a first keyword through a search user interface and outputting together a search result of the first keyword and a search result of a second keyword mapped to an abbreviation of the first keyword,
The first keyword is an official name of the second keyword,
How to manage meeting minutes. According to claim 1,
Further comprising a meeting minutes search step of receiving a second keyword through a search user interface and outputting a search result of the second keyword and a search result of the first keyword mapped to an official name of the second keyword together;
The second keyword is an abbreviation of the first keyword,
How to manage meeting minutes. processor;
network interface;
Memory; and
A computer program loaded into the memory and executed by the processor,
The computer program,
instructions for obtaining speech data including speech text structured to enable identification of key words;
an instruction for mapping meta data including a required pronunciation time for a utterance corresponding to the main word to the main word; and
Including an instruction for generating meeting minutes data using the metadata,
The instruction for generating the meeting minutes data,
An instruction for determining a word whose speech-to-text (STT) conversion accuracy prediction rate is less than a reference value among the main words as an adjustment target word,
In response to a user's correction input, the method further includes an instruction for reproducing recorded voice data corresponding to the adjustment target word included in meta data of the adjustment target word;
when a first word is added to the speech text, mapping metadata of the first word based on metadata of a second word that is a key word closest to the location where the first word is added; and
Includes instructions for highlighting and displaying emphasized words whose pronunciation time is equal to or greater than a reference value;
The instruction to display,
In response to a selection input for a third word, an instruction for automatically calculating a speech start time of a third word using pronunciation required times of all words existing before the third word; and
Including instructions for displaying the utterance start time of the third word,
Meeting minutes management device.

Images