Terminals and methods

Abstract

The purpose of the present disclosure is to provide a terminal and a method that can suppress registered word correction errors during error correction of texts including registered words. A terminal 10 according to one aspect of the present disclosure comprises: an acquisition unit 13 that acquires a first text including a registered word which has been registered in advance; a replacement unit 14 that outputs a second text resulting from replacing the registered word with a dummy character string in the first text; a correction unit 15 that outputs a third text resulting from performing error correction on the second text; and a restoration unit 16 that outputs a fourth text resulting from replacing the dummy character string with the registered word in the third text.

Description

Technical Field

[0001] The present invention relates to a terminal and a method.

Background Art

[0002] Techniques for performing error correction on speech recognition results and the like are known. Patent Document 1 discloses a technique for performing error correction on an error portion using a speech recognition result sequence to be error-corrected and a correction network. The technique described in Patent Document 1 is a technique for improving the accuracy of error correction. Patent Document 2 discloses a technique for extracting a proper noun section, which is an error correction candidate, from related information for a speech recognition result. The technique described in Patent Document 2 is a technique for determining which part of the speech recognition result is a proper noun.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] For example, in speech recognition, in order to correctly recognize proper nouns and the like, word registration for pre-registering proper nouns and the like may be performed. The registered proper nouns and the like are called registered words. When error correction is performed on a text (for example, a speech recognition result) including registered words, the registered words may be erroneously corrected. Therefore, it is desired to suppress correction errors of registered words.

[0005] An object of the present disclosure is to provide a terminal and a method capable of suppressing correction errors of registered words in error correction of a text including registered words.

Means for Solving the Problems

[0006] A terminal relating to one aspect of this disclosure includes: an acquisition unit that acquires a first text containing pre-registered registered words; a replacement unit that outputs a second text in which the registered words in the first text are replaced with dummy strings; a correction unit that outputs a third text in which errors have been corrected on the second text; and a restoration unit that outputs a fourth text in which the dummy strings in the third text are replaced with registered words.

[0007] In one aspect of this disclosure, the terminal replaces the registered word with a dummy string, performs error correction, and then replaces the dummy string with the registered word. In other words, error correction is performed with the registered word masked by the dummy string. As a result, errors in correcting text containing the registered word can be suppressed. [Effects of the Invention]

[0008] According to this disclosure, errors in correcting registered words in text containing registered words can be suppressed. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 is a block diagram showing an example of a terminal configuration. [Figure 2] Figure 2 shows an example of registered words. [Figure 3] Figure 3 is an example illustrating the overview of error correction. [Figure 4] Figure 4 shows an example of training data for an error correction model. [Figure 5] Figure 5 is a flowchart showing an example of terminal operation. [Figure 6] Figure 6 shows an example of the hardware configuration of a terminal. [Modes for carrying out the invention]

[0010] Embodiments of this disclosure will be described with reference to the attached drawings. Where possible, the same parts will be denoted by the same reference numerals, and redundant descriptions will be omitted.

[0011] Figure 1 is a block diagram showing an example configuration of terminal 10. Terminal 10 is a computer used by the user. The type of terminal 10 is not limited. For example, terminal 10 may be a personal computer, a high-function mobile phone (smartphone), a tablet device, or a wearable device. Terminal 10 is connected to the conference system 20 via a network so as to be able to communicate.

[0012] The conference system 20 is a system that provides online meetings. An online meeting is a meeting conducted via multiple terminals connected to a network, and is also called a web conference or network meeting. The conference system 20 may extract proper nouns from the online meeting and send the extracted proper nouns to the terminal 10. For example, the conference system 20 may obtain the names of the online meeting participants as proper nouns and send them to the terminal 10. For example, in an online meeting, documents and other materials can be shared on the screen. The conference system 20 may recognize the characters displayed on the screen and extract company names and other names from among them as proper nouns.

[0013] The terminal 10 includes, as functional elements, a storage unit 11, a recognition unit 12, an acquisition unit 13, a replacement unit 14, a correction unit 15, a restoration unit 16, and an addition unit 17.

[0014] The memory unit 11 is a non-temporary storage medium or storage device that stores information about registered words. Registered words (specific words) are proper nouns such as people's names, company names, product names, or service names. The memory unit 11 may also store attribute information of registered words. Attribute information may be information indicating categories or labels such as people's names, company names, product names, or service names.

[0015] The recognition unit 12 performs speech recognition on the input audio. The recognition unit 12 performs so-called word registration speech recognition using the audio and the registered words stored in the memory unit 11. The recognition unit 12 outputs the speech recognition result.

[0016] The acquisition unit 13 acquires a first text containing pre-registered words. For example, the acquisition unit 13 acquires a speech recognition result containing registered words as the first text. The first text is not limited to a speech recognition result. The acquisition unit 13 may acquire the first text via user input or another system.

[0017] The replacement unit 14 outputs a second text in which the registered words in the first text have been replaced with dummy strings (also called tokens or mask data). The replacement unit 14 uses the registered words stored in the storage unit 11 to replace the registered words in the first text with dummy strings. The dummy string may be any string enclosed by identification symbols such as "<" and ">". The identification symbols are not limited to these. In one example, the dummy string is: <tokenn>It can be expressed as. The Token is a character string, and N may be, for example, an integer.

[0018] The correction unit 15 outputs a third text in which error correction has been performed on the second text. The correction unit 15 may perform error correction using, for example, an error correction model. The error correction model is a model obtained by performing machine learning using pair data of a text including errors and a text in which the errors have been corrected. When the second text is input, the correction model corrects the errors in the second text and outputs the third text.

[0019] The restoration unit 16 outputs a fourth text in which the dummy character string in the third text has been replaced with a registered word. That is, the restoration unit 16 performs replacement to return the dummy character string to the registered word. The restoration unit 16 may display the fourth text on a display device or transmit it to another device or the like.

[0020] The addition unit 17 acquires proper nouns extracted from an online meeting in which the terminal 10 participates and adds the proper nouns to the registered words. The addition unit 从会议系统20获取固有名词。例如,固有名词是在线会议的参与者姓名和从在线会议屏幕上提取的文字中的至少一个。固有名词不限于此。添加部17将固有名词存储在存储部11中。添加部17可以将参与者姓名的属性信息作为人名存储在存储部11中。

[0021] FIG. 2 is a diagram showing an example of registered words. For example, the storage unit 11 stores the character string of the registered word and the attribute information of the registered word in association with an identifier (ID) that uniquely identifies the registered word. The storage unit 11 may store other information.

[0022] In one example, the memory unit 11 stores the registered word "Company X" and the attribute information "Company Name" in association with ID "1". In another example, the memory unit 11 stores the registered word "YY" and the attribute information "Product Name" in association with ID "2". In yet another example, the memory unit 11 stores the registered word "Mr. Z" and the attribute information "Person's Name" in association with ID "3".

[0023] Figure 3 illustrates an overview of error correction. The memory unit 11 has pre-stored registered words R. The registered words include "Company X", "YY", and "Mr. Z". The user speaks voice V. Voice V is "What are the results of Mr. Z's research on Company X's new product, YY?".

[0024] The recognition unit 12 performs word registration speech recognition using the speech V and the registered word R. The recognition unit 12 outputs the speech recognition result. The acquisition unit 13 acquires the speech recognition result as the first text T1. The first text T1 is "What are the results of Mr. Z's research on uh Company X's known product, YY?". The difference between the speech V and the speech recognition result T1 is the speech recognition error. The first text T1 includes "uh" and "knew" as speech recognition errors. The first text T1 also includes the registered word R.

[0025] The replacement unit 14 replaces the registered word R in the first text T1 with a dummy string. In the first text T1, the registered word R appears as "Mr. Z", "Company X", and "YY" in that order. For example, the replacement unit 14 replaces "Mr. Z", "Company X", and "YY" with dummy strings. <token1> 、 <token2>, and <token3>The replacement unit 14 outputs a second text T2 in which the registered word R has been replaced with a dummy string. The second text T2 is "What are the results of <token1>’s research on uh <token2>’s knew product, <token3>"?"

[0026] The correction unit 15 performs error correction on the second text T2. The correction unit 15 may perform error correction using, for example, an error correction model. The correction unit 15 inputs the second text T2 into the error correction model and outputs the third text T3 in which the errors in the second text T2 have been corrected. The third text T3 is "What are the results of <token1>’s research on <token2>’s new product, <token3>The third text T3 corrects the speech recognition errors "uh" and "knew" in the second text T2. Specifically, in the third text T3, "uh" is deleted and "knew" is corrected to "new".

[0027] The restoration unit 16 replaces the dummy string in the third text T3 with the registered word R. The restoration unit 16 replaces the dummy string with the registered word R using the correspondence between the registered word R replaced by the replacement unit 14 and the dummy string. For example, the restoration unit 16, <token1> 、 <token2>, and <token3>These are replaced with "Mr. Z", "Company X", and "YY", respectively. The restoration unit 16 outputs a fourth text T4 in which the dummy string has been replaced with the registered word R. The fourth text T4 is "What are the results of Mr. Z's research on Company X's new product, YY?".

[0028] Figure 4 shows an example of training data for an error correction model. The error correction model is trained using supervised learning with paired data of speech-recognized text (sentences containing errors) as input and speech-recognized error-corrected text as output. The speech-recognized error-corrected text can be considered the correct answer data (training data).

[0029] Figure 4 shows pair data P1 to P5 as examples of paired data. Pair data P1 is an example of paired data where the speech recognition text and the speech recognition error correction text do not contain dummy strings. Pair data P2 is an example where the speech recognition text and the speech recognition error correction text do not contain dummy strings. <token1>This is an example of paired data that includes the following: Paired data P3 contains speech recognition text and speech recognition error correction text, each with a dummy string. <token1>and <token2>This is an example of paired data that includes the following: Paired data P4 contains speech recognition text and speech recognition error correction text, respectively.<Token1_Name> This is an example of paired data containing the following: Here, Name is an example of attribute information indicating a person's name. Paired data P5 contains speech recognition text and speech recognition error correction text, respectively.<Token1_Product> and<Token2_Product> This is an example of paired data containing the following: Here, Product is an example of attribute information indicating the product name.

[0030] Referring to Figure 5, the operation of terminal 10 will be explained, along with the method relating to this disclosure. Figure 5 is a flowchart showing an example of the operation of terminal 10.

[0031] In step S1, terminal 10 registers a word. For example, terminal 10 may accept input from the user regarding the registered word and attribute information. Storage unit 11 may store the registered word and attribute information in association with an ID (see, for example, Figure 2).

[0032] The additional unit 17 may obtain proper nouns extracted from online meetings in which terminal 10 is participating and add the proper nouns to the registered words. The additional unit 17 may obtain proper nouns from the meeting system 20. The additional unit 17 may store the proper nouns in the storage unit 11.

[0033] In step S2, the recognition unit 12 performs speech recognition on the input audio. The recognition unit 12 performs word registration speech recognition using the audio and the registered words stored in the memory unit 11. The recognition unit 12 outputs the speech recognition result.

[0034] In step S3, the acquisition unit 13 acquires a first text containing pre-registered registered words. For example, the acquisition unit 13 acquires a speech recognition result containing registered words as the first text. The first text is not limited to a speech recognition result. The acquisition unit 13 may acquire the first text via user input or another system.

[0035] In step S4, the replacement unit 14 outputs a second text in which the registered words in the first text have been replaced with dummy strings. The replacement unit 14 uses the registered words stored in the memory unit 11 to replace the registered words in the first text with dummy strings.

[0036] In step S5, the correction unit 15 outputs a third text in which the second text has been error-corrected. The correction unit 15 may perform error correction using, for example, an error correction model.

[0037] In step S6, the restoration unit 16 outputs a fourth text in which the dummy strings in the third text have been replaced with registered words. The restoration unit 16 may display the fourth text on a display device or transmit it to another device.

[0038] As described above, the terminal 10 relating to one aspect of this disclosure includes an acquisition unit 13 that acquires a first text containing pre-registered registered words, a replacement unit 14 that outputs a second text in which the registered words in the first text are replaced with dummy strings, a correction unit 15 that outputs a third text in which errors have been corrected on the second text, and a restoration unit 16 that outputs a fourth text in which the dummy strings in the third text are replaced with registered words.

[0039] A method relating to one aspect of this disclosure includes the steps of: obtaining a first text containing pre-registered registered words; outputting a second text in which the registered words from the first text are replaced with dummy strings; outputting a third text in which the second text has been error-corrected; and outputting a fourth text in which the dummy strings from the third text are replaced with registered words.

[0040] In the terminal 10 and method relating to one aspect of this disclosure, the registered word is replaced with a dummy string, error correction is performed, and the dummy string is replaced with the registered word. That is, error correction is performed with the registered word masked by the dummy string. As a result, errors in correcting text containing the registered word can be suppressed.

[0041] The correction unit 15 performs error correction using an error correction model that has been machine-trained using training data containing dummy strings. By using training data containing dummy strings in the machine learning process, it is possible to suppress the erroneous correction of dummy strings. As a result, errors in correcting registered words can be further suppressed.

[0042] The dummy string contains attribute information of the registered word. The attribute information is replaced in the dummy string along with the registered word. In addition, the attribute information is reflected in the training of the error correction model. By performing error correction that takes attribute information into account, the accuracy of error correction around the dummy string is improved.

[0043] The system includes an addition unit 17 that retrieves proper nouns extracted from online meetings in which terminal 10 is participating and adds these proper nouns to registered words. Registered words are dynamically added based on online meetings. As a result, correction errors can be further suppressed.

[0044] Proper nouns are at least one of the names of online meeting participants and characters extracted from the online meeting screen. Adding such proper nouns to the registered words can further reduce correction errors.

[0045] This disclosure may have the following configuration: [1] An acquisition unit that acquires a first text containing pre-registered words, A replacement unit outputs a second text in which the registered words in the first text are replaced with dummy strings, A correction unit outputs a third text in which errors have been corrected against the second text, A restoration unit outputs a fourth text in which the dummy string in the third text is replaced with the registered word, A terminal equipped with the following features. [2] The terminal described in [1] performs the error correction using an error correction model that has been machine-trained using training data including the dummy string. [3] The dummy string is the terminal described in [2], which includes attribute information of the registered word. [4] A terminal according to any one of [1] to [3], comprising an additional unit that obtains proper nouns extracted from an online meeting in which the terminal is participating and adds the proper nouns to the registered words. [5] The terminal described in [4], wherein the proper noun is at least one of the names of participants in the online meeting and characters extracted from the screen of the online meeting. [6] The steps include obtaining a first text containing pre-registered words, The steps include outputting a second text in which the registered words in the first text are replaced with dummy strings, The steps include: outputting a third text in which errors have been corrected against the second text; The steps include outputting a fourth text in which the dummy string from the third text is replaced with the registered word, A method that includes [a certain feature].

[0046] The block diagrams used in the description of the above embodiments show functional units. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method of realizing each functional block is not particularly limited. That is, each functional block may be realized using one device that is physically or logically coupled, or it may be realized using two or more physically or logically separated devices that are directly or indirectly connected (for example, using wired or wireless connections). A functional block may also be realized by combining the above one device or the above multiple devices with software.

[0047] Functions include, but are not limited to, judgment, decision, judgment, calculation, calculation, processing, derivation, investigation, exploration, confirmation, reception, transmission, output, access, resolution, selection, selection, establishment, comparison, assumption, expectation, assumption, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating (mapping), and assigning. For example, a functional block (configuration part) that enables transmission is called a transmitting unit or transmitter. As mentioned above, the method of implementation is not particularly limited.

[0048] For example, terminal 10 in one embodiment of the present disclosure may function as a computer that performs information processing according to the present disclosure. Figure 6 is a diagram showing an example of the hardware configuration of terminal 10 according to one embodiment of the present disclosure. The above-described terminal 10 may be physically configured as a computer device including a processor 1001, memory 1002, storage 1003, communication device 1004, input device 1005, output device 1006, bus 1007, etc.

[0049] In the following explanation, the term "device" can be replaced with "circuit," "device," "unit," etc. The hardware configuration of terminal 10 may include one or more of the devices shown in the figure, or it may be configured to omit some of the devices.

[0050] Each function in terminal 10 is realized by loading predetermined software (programs) onto hardware such as the processor 1001 and memory 1002, which causes the processor 1001 to perform calculations, control communication by the communication device 1004, and control at least one of data reading and writing in the memory 1002 and storage 1003.

[0051] The processor 1001 controls the entire computer, for example, by running the operating system. The processor 1001 may consist of a central processing unit (CPU) that includes interfaces with peripheral devices, control units, arithmetic units, registers, etc. For example, each of the functions in the terminal 10 described above may be implemented by the processor 1001.

[0052] Furthermore, the processor 1001 reads programs (program code), software modules, data, etc., from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes accordingly. The program used is one that causes the computer to execute at least a part of the operations described in the above embodiment. For example, each function of the terminal 10 may be implemented by a control program stored in the memory 1002 and operated on the processor 1001. Although the above processes have been described as being executed by one processor 1001, they may be executed simultaneously or sequentially by two or more processors 1001. The processor 1001 may be implemented by one or more chips. The program may also be transmitted from the network via a telecommunications line.

[0053] Memory 1002 is a computer-readable recording medium and may consist of at least one of the following: ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), RAM (Random Access Memory), etc. Memory 1002 may also be called a register, cache, main memory, etc. Memory 1002 can store executable programs (program code), software modules, etc., for carrying out information processing according to one embodiment of this disclosure.

[0054] The storage 1003 is a computer-readable recording medium and may consist of at least one of the following: an optical disc such as a CD-ROM (Compact Disc ROM), a hard disk drive, a flexible disk, a magneto-optical disk (e.g., a compact disc, a digital multipurpose disc, a Blu-ray® disc), a smart card, flash memory (e.g., a card, a stick, a key drive), a floppy® disk, a magnetic strip, etc. The storage 1003 may also be called an auxiliary storage device. The storage medium provided by the terminal 10 may be, for example, a database, server, or other suitable medium including at least one of the memory 1002 and the storage 1003.

[0055] The communication device 1004 is hardware (transceiver / receiver device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as a network device, network controller, network card, communication module, etc.

[0056] The input device 1005 is an input device that accepts input from an external source (e.g., a keyboard, mouse, microphone, switch, button, sensor, etc.). The output device 1006 is an output device that outputs to an external source (e.g., a display, speaker, LED lamp, etc.). The input device 1005 and the output device 1006 may be configured as an integrated unit (e.g., a touch panel).

[0057] Furthermore, each device, such as the processor 1001 and memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or different buses may be configured for each device.

[0058] Furthermore, terminal 10 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array), and some or all of each functional block may be realized by such hardware. For example, processor 1001 may be implemented using at least one of these hardware components.

[0059] The processing procedures, sequences, flowcharts, etc., of each aspect / embodiment described herein may be reordered, provided they are consistent with each other. For example, the methods described herein present various step elements in an exemplary order and are not limited to that specific order.

[0060] Input and output information may be stored in a specific location (e.g., memory) or managed using a management table. Input and output information may be overwritten, updated, or appended to. Output information may be deleted. Input information may be transmitted to other devices.

[0061] The determination may be made by a value represented by 1 bit (0 or 1), by a boolean value (true or false), or by a numerical comparison (for example, a comparison with a predetermined value).

[0062] Each aspect / embodiment described herein may be used individually, in combination, or switched between as needed during implementation. Furthermore, notification of specific information (e.g., notification that "X is") is not limited to explicit notification, but may also be implicit (e.g., by not providing such notification).

[0063] Although the present disclosure has been described in detail above, it will be clear to those skilled in the art that the present disclosure is not limited to the embodiments described herein. The present disclosure can be implemented in modified and altered forms without departing from the intent and scope of the present disclosure as defined by the claims. Therefore, the descriptions in the present disclosure are illustrative and not intended to be restrictive in any way.

[0064] Software should be broadly interpreted to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, execution threads, procedures, functions, and so on, whether they are called software, firmware, middleware, microcode, hardware description languages, or by any other name.

[0065] Furthermore, software, instructions, information, etc., may be transmitted and received via a transmission medium. For example, if software is transmitted from a website, server, or other remote source using at least one of wired technology (such as coaxial cable, fiber optic cable, twisted pair, or digital subscriber line (DSL)) and wireless technology (such as infrared or microwave), then at least one of these wired and wireless technologies is included in the definition of a transmission medium.

[0066] The terms “system” and “network” as used in this disclosure are interchangeable.

[0067] Furthermore, the information, parameters, etc., described in this disclosure may be expressed using absolute values, relative values ​​from a predetermined value, or corresponding other information.

[0068] As used in this disclosure, the terms “determining” and “determining” may encompass a wide variety of actions. “Determining” may include, for example, judging, calculating, computing, processing, deriving, investigating, looking up, searching, inquiry (e.g., searching in a table, database, or other data structure), and ascertaining. “Determining” may also include, for example, receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, and accessing (e.g., accessing data in memory). Furthermore, "judgment" and "decision" can include considering something as having been "judged" or "decided" after resolving, selecting, choosing, establishing, comparing, etc. In other words, "judgment" and "decision" can include considering something as having been "judged" or "decided" after some action. Also, "judgment (decision)" can be reinterpreted as "assuming," "expecting," or "considering."

[0069] The terms “connected,” “coupled,” or any variation thereof, mean any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are “connected” or “coupled” with each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, “connection” may be reinterpreted as “access.” As used in this disclosure, two elements may be considered to be “connected” or “coupled” with each other using at least one of one or more wires, cables, and printed electrical connections, and, in some non-limiting and non-exclusive examples, electromagnetic energy having wavelengths in the radio frequency domain, microwave domain, and optical (both visible and invisible) domain.

[0070] In this disclosure, the phrase "based on" does not mean "based solely on" unless otherwise specified. In other words, the phrase "based on" means both "based solely on" and "based at least on."

[0071] Any reference to elements using the designations “first,” “second,” etc., as used in this disclosure does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient way to distinguish between two or more elements. Accordingly, references to the first and second elements do not imply that only two elements may be employed, or that the first element must precede the second element in any way.

[0072] Where the terms “include,” “including,” and variations thereof are used in this disclosure, these terms are intended to be inclusive, as is the term “comprising.” Furthermore, the term “or” as used in this disclosure is not intended to mean exclusive OR.

[0073] In this disclosure, if articles are added through translation, such as a, an, and the in English, this disclosure may include the fact that the noun following these articles is plural.

[0074] In this disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "combine" may be interpreted similarly to "different." [Explanation of Symbols]

[0075] 10...Terminal, 11...Storage unit, 12...Recognition unit, 13...Acquisition unit, 14...Replacement unit, 15...Correction unit, 16...Restoration unit, 17...Addition unit, 20...Conference system. < / token1> < / token1> < / tokenn>

Claims

1. An acquisition unit that acquires a first text containing pre-registered words, A replacement unit outputs a second text in which the registered words in the first text are replaced with dummy strings, A correction unit outputs a third text in which errors have been corrected against the second text, A restoration unit outputs a fourth text in which the dummy string in the third text is replaced with the registered word, A terminal equipped with the following features.

2. The terminal according to claim 1, wherein the correction unit performs the error correction using an error correction model that has been machine-trained using training data including the dummy string.

3. The terminal according to claim 2, wherein the dummy string includes attribute information of the registered word.

4. The terminal according to claim 1, further comprising an addition unit that obtains proper nouns extracted from an online meeting in which the terminal is participating and adds the proper nouns to the registered words.

5. The terminal according to claim 4, wherein the proper noun is at least one of the names of participants in the online meeting and characters extracted from the screen of the online meeting.

6. The steps include obtaining a first text containing pre-registered words, The steps include outputting a second text in which the registered words in the first text are replaced with dummy strings, The steps include: outputting a third text in which errors have been corrected against the second text; The steps include outputting a fourth text in which the dummy string from the third text is replaced with the registered word, A method that includes [a certain feature].

Images