WO2022215120A1 - Dispositif, procédé et programme de traitement d'informations - Google Patents

Dispositif, procédé et programme de traitement d'informations Download PDF

Info

Publication number
WO2022215120A1
WO2022215120A1 PCT/JP2021/014513 JP2021014513W WO2022215120A1 WO 2022215120 A1 WO2022215120 A1 WO 2022215120A1 JP 2021014513 W JP2021014513 W JP 2021014513W WO 2022215120 A1 WO2022215120 A1 WO 2022215120A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
input
unit
target person
presenting
Prior art date
Application number
PCT/JP2021/014513
Other languages
English (en)
Japanese (ja)
Inventor
公之 茶谷
直樹 千葉
Original Assignee
株式会社KPMG Ignition Tokyo
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 株式会社KPMG Ignition Tokyo filed Critical 株式会社KPMG Ignition Tokyo
Priority to PCT/JP2021/014513 priority Critical patent/WO2022215120A1/fr
Publication of WO2022215120A1 publication Critical patent/WO2022215120A1/fr

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16—Sound input; Sound output
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00—Handling natural language data
    • G06F40/10—Text processing
    • G06F40/12—Use of codes for handling textual entities
    • G06F40/151—Transformation
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L13/00—Speech synthesis; Text to speech systems
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/08—Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L13/00—Speech synthesis; Text to speech systems
    • G10L13/08—Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
    • G10L13/10—Prosody rules derived from text; Stress or intonation

Definitions

  • the present invention relates to information processing technology.
  • chatbots and virtual assistants (hereinafter collectively referred to as AI assistants) is underway with the aim of supporting users using artificial intelligence (AI).
  • AI assistants are widely used for online customer service, and automatically select the appropriate response from a group of pre-prepared responses to inquiries entered in text by customers visiting websites. respond.
  • Virtual assistants are implemented in smart devices such as smart speakers, smartphones, and smart watches, as well as in mobile applications. They automatically respond to user inquiries by voice, etc., using the vast amount of information available on the Internet. .
  • the present invention has been made in view of this situation, and its purpose is to provide an information processing apparatus that presents information appropriately according to the user.
  • An information processing apparatus includes an information input unit for inputting information, a reading designation unit for designating how to read the input information by referring to background information about a person to whom information is to be presented, an information presentation unit for presenting the input information to the information presentation target person in accordance with the designated reading.
  • An information processing apparatus includes an information input unit for inputting information, and a presentation mode specification unit for specifying a presentation mode of the input information by referring to background information about an information presentation target person. and an information presentation unit that presents the input information to the information presentation target person in accordance with the designated presentation mode.
  • An information processing apparatus includes an information input unit into which information on minutes is input, and background information about an information presentation target person, so that the input information on the minutes is input as an information presentation target. and an information presenting unit for presenting the information of the processed minutes to the information presentation target person.
  • An information processing apparatus includes an information input unit for inputting information, a privacy determination unit for determining whether or not privacy protection is necessary with reference to background information about a person to whom information is to be presented, a privacy An encryption processing unit that performs encryption processing on at least part of the input information when it is determined that protection is necessary, and an information presentation unit that presents the information subjected to the encryption processing to an information presentation target person.
  • information can be presented appropriately according to the user.
  • FIG. 1 is a functional block diagram of an information processing device according to a first embodiment
  • FIG. It is a figure which shows the registration example of a phrase in a phrase database.
  • 4 is a flowchart showing processing of the information processing apparatus according to the first embodiment
  • FIG. 7 is a functional block diagram of an information processing device according to a second embodiment
  • FIG. 4 is a diagram showing an example of registration of presentation modes in a presentation mode database
  • 9 is a flowchart showing processing of an information processing apparatus according to the second embodiment
  • FIG. 11 is a functional block diagram of an information processing device according to a third embodiment
  • FIG. 10 is a flowchart showing processing of an information processing apparatus according to the third embodiment
  • FIG. 11 is a functional block diagram of an information processing device according to a fourth embodiment
  • FIG. 14 is a flowchart showing processing of an information processing apparatus according to a fourth embodiment
  • FIG. 10 is a flowchart showing processing of an information processing apparatus according to the third embodiment
  • FIG. 11 is a functional block diagram of an information processing device according to a fourth embodiment
  • FIG. 14 is a flowchart showing processing of an information processing apparatus according to a fourth embodiment
  • FIG. 1 schematically shows an example in which the information processing apparatus 1 of this embodiment is configured in a client-server model.
  • the information processing apparatus 1 includes a server 11 that centrally handles information processing, a group of user devices 12 in which at least one functions as a client for the user 10, and a network 13 that interconnects the server 11 and the group of user devices 12. be.
  • the information input unit 110 is provided on the server 11 side, but the information input unit 110 may be provided on the client side.
  • the functions of the information input unit 110 may be implemented in one or more user devices belonging to the user device group 12 .
  • the user device functioning as the information input unit 110 generates various types of information to be presented to the user 10 based on the operation of the user 10 and situation detection described later.
  • Information generated by the user device is sent to the server 11 via the network 13 and subjected to information processing, which will be described later.
  • the information processed by the server 11 is sent again to the user device group 12 via the network 13 and presented to the user 10 .
  • the function of the information input unit 110 may be implemented in the server 11.
  • the server 11 functioning as the information input unit 110 generates information to be presented to the user 10 based on various information held therein or acquired via the network 13 .
  • the information generated by the server 11 and to be presented to the user 10 is information generated by another user 10 ′ by operating the server 11 itself or another user device 12 ′ connected to the server 11 via the network 13 . and includes, but is not limited to, information intended to be communicated to user 10 by other user 10'.
  • the server 11 performs information processing on the information input from the information input unit 110 according to background information such as attributes and circumstances of the user 10 . Attributes of the user 10 include age, gender, hometown, nationality, professional qualifications, occupation, and affiliated organization. These attribute information may be stored in advance in the background information database 120 accessible by the server 11, or may be stored in the user device group 12 in real time via the network 13 during information processing by the server 11. It may be read out. As for the situation of the user 10, the server 11 refers to situation detection information detected from the user 10 in real time by the user device group 12 as main information and situation history information held in the background information database 120 as secondary information. Interpret the user 10 situation.
  • the user device group 12 is one or more arbitrary devices that perform at least one of inputting and/or outputting information regarding the user 10 .
  • the user device is not limited to a device owned and used by the user 10, but may be a device installed in a place where the user 10 is temporarily or permanently present, or a device used by a third party around the user 10.
  • User devices are broadly classified into those that both input and output information about the user 10, those that only input information, and those that only output information.
  • Wearable devices such as a personal computer 12A, a smart phone 12B, a tablet, a smart speaker, and a smart watch are exemplified as input/output type user devices of the first type.
  • the second type of input-type user device includes a camera 12C, a watch 12D having a function of measuring biological signals, etc., a microphone for acquiring the speech of the user 10 and sounds around the user 10, and the situation of the user 10 himself/herself.
  • Examples include various sensors capable of measuring the status of the user 10 or the situation or environment around the user 10 (hereinafter also referred to as the situation of the user 10 or the situation in which the user 10 is placed) with or without contact with the user 10 .
  • "input information about the user” means information input by the user himself/herself by operating the user device (mainly the first type), including both the information that measured the
  • a display 12E as a display unit and a speaker 12F as an audio output unit are exemplified as the output type user device of the third type.
  • the various user devices as described above may be connected by wire or wirelessly so as to be able to communicate with each other.
  • the client-server model described above is merely one configuration example of the information processing apparatus 1, and a standalone configuration in which each function of the information processing apparatus 1 is implemented in the user device group 12 locally used by the user 10 may be employed.
  • FIG. 2 is a functional block diagram of the information processing device 1 according to the first embodiment.
  • the information processing apparatus 1 includes an information input unit 110, a phrase extraction unit 131, a phrase database 132, a reading designation unit 141, a background information database 120, a background information acquisition unit 150, an information presentation unit 161, and an explanation acquisition unit. It has a section 162 and a reading checking section 170 .
  • These functional blocks are realized through cooperation between hardware resources such as the computer's central processing unit, memory, input device, output device, and peripheral devices connected to the computer, and software executed using them. . Regardless of the type of computer or installation location, each of the above functional blocks may be implemented using the hardware resources of a single computer, or may be implemented by combining hardware resources distributed among multiple computers. .
  • Information to be presented to the information presentation target person 10 is input to the information input unit 110 .
  • the type of information to be input is not particularly limited as long as it includes linguistic information from which words can be extracted by the word/phrase extraction unit 131 in the subsequent stage. Examples include character information (text information), image information (image information), voice information (audio information), and image information (video information).
  • non-text information such as image information, audio information, and video information is input to the information input unit 110
  • the linguistic information included in these information is extracted for phrase extraction using image recognition technology, voice recognition technology, or the like. Convert to text information.
  • the phrase extraction unit 131 extracts phrases from the linguistic information input to the information input unit 110 . Specifically, the word/phrase extraction unit 131 searches the word/phrase database 132 for each word/phrase included in the language information input to the information input unit 110 . Phrases including multiple readings are registered in advance in the phrase database 132, and the phrase extraction unit 131 extracts phrases hit by the search and provides them to the reading designation unit 141 in the subsequent stage. Specific examples of the phrases registered in the phrase database 132 will be described later. Phrases are registered in the phrase database 132 .
  • words of parts of speech other than nouns such as verbs, adjectives, adverbs, pronouns, auxiliary verbs, conjunctions, articles, and interjections may be registered in the word database 132. good.
  • the reading designation unit 141 refers to the background information about the information presentation target person 10 and designates the reading of the phrase extracted by the phrase extraction unit 131 .
  • a specific processing example of the reading designation unit 141 will be described later with reference to FIG.
  • the background information about the information presentation target person 10 referred to by the reading designation unit 141 is acquired by the background information acquisition unit 150 .
  • the background information acquisition section 150 includes an attribute acquisition section 151 and a situation detection section 152 .
  • the attribute acquisition unit 151 acquires the attribute information of the information presentation target person 10, such as age, gender, hometown, nationality, professional qualifications, occupation, and affiliated organization, as background information. These pieces of attribute information may be stored in advance in the attribute information database 121 included in the background information database 120 accessible by the background information acquisition unit 150, or may be stored in the user device group 12 (Fig. 1) may be stored in advance. Note that if there is a discrepancy between the attribute information held by the attribute information database 121 and the attribute information held by the user device group 12, the attribute information with the latest storage date or the attribute information held by the majority of the devices is the most reliable. It is provided to the reading designation unit 141 as attribute information.
  • the situation detection unit 152 detects the situation in which the information target person 10 is placed as background information. Specifically, the situation detection unit 152 selects from the user device group 12 ( FIG. 1 ) used by the information presentation target person 10 or located in the information presentation target person 10 Obtain various types of status information that directly or indirectly indicate status.
  • the situation detection unit 152 analyzes the conversation of the information presentation target person 10 and detects the situation.
  • a smart speaker, a smart phone, or the like on which a virtual assistant having a voice recognition function is implemented functions as a user device that listens to conversation as situation information of the information presentation target person 10 .
  • the conversation heard by the user device is analyzed by the user device itself or an analysis engine mounted in the situation detection unit 152, and the situation in which the information presentation target person 10 is placed is detected.
  • the atmosphere of whether the information presentation target person 10 is in a formal or casual setting can be detected from the content, speed, tone, tone, etc. of the information presentation target person 10 or the conversation partner.
  • the conversation including the information presentation target person 10 itself It is possible to infer with high accuracy the attributes of the participants and the situation in which the conversation is taking place.
  • the situation in which the information presentation target person 10 is placed may be detected based on other situation information that can be acquired from the user device group 12 .
  • situation information information input by the information presentation target person 10 by operating the personal computer 12A, the smartphone 12B, a tablet, or the like is highly valuable as situation information, similar to what the information presentation target person 10 says in conversation.
  • the facial expression, posture, gesture, clothing, etc. of the information presentation target person 10 can , furniture, surrounding congestion, and other useful situational information.
  • the location of the information presentation target person 10 can also be detected from a positioning sensor such as a GPS (Global Positioning System) sensor built into a smartphone or the like.
  • GPS Global Positioning System
  • the information presentation target person 10 wears a wearable device such as a smart watch capable of measuring biosignals such as heartbeat, body temperature, blood pressure, respiration, and perspiration
  • the biometric information obtained therefrom is also an information presentation target. It can be utilized as situation information suggesting the situation in which the person 10 is placed.
  • the user device group 12 includes a sensor that measures information related to the environment of the information presentation target person 10, such as temperature, humidity, and luminance, such environmental measurement information can also be used as situation information. Date and time information that can be obtained from a clock can also be referred to as background information in a broad sense.
  • Each of these biometric information, environmental measurement information, and date and time information is merely information that indirectly suggests the situation in which the information presentation target person 10 is placed.
  • the multifaceted situation of the information presentation target person 10 including the physical and mental conditions of the target person 10 such as physical condition and mood can be detected.
  • the situation information that can be obtained from the user device group 12 illustrated above is roughly classified into four types: “conversation”, “operation input”, “image”, and “measurement information”. In the following, to simplify the explanation, these types will be used as appropriate.
  • the situation detection unit 152 is configured by artificial intelligence so as to make an accurate situation judgment by synthesizing such various types of situation information.
  • a situation history database 122 is provided in the background information database 120 for the purpose of supporting situation determination by the situation detection unit 152 .
  • situation information acquired by various user devices and descriptions of situations in which the user was placed at that time are associated as past history data of the information presentation target person 10 or a third party. A lot of reference data is kept.
  • a situation history database in which a description of the user's situation, such as “a few people in suits are having a conversation in a private room” detected by a camera and a description of the user's situation, such as "a user consults with a lawyer about inheritance,” is associated with each other as reference data. 122.
  • the information presentation target person 10 works at a specific place B such as a coffee shop during time zone A on weekdays
  • a specific place B such as a coffee shop during time zone A on weekdays
  • "time zone A” measured by a clock built in a smartphone or the like and "place A” measured by a GPS sensor B” is stored in the situation history database 122 as reference data in which a description relating to the situation of the information presentation target person 10 is associated with “at work”.
  • the situation detection unit 152 searches the situation history database 122 based on the group of situation information acquired from the group of user devices 12, and finds reference data containing a group of similar situation information. Since there is a high possibility that the “description regarding the situation in which the user is placed” included in the found reference data represents the current situation in which the information presentation target person 10 is placed, the situation detection unit 152 stores the information in the situation history database. By referring to 122, the situation of the information presentation target person 10 can be detected with high accuracy. It should be noted that the situation detection unit 152 may be composed of artificial intelligence capable of machine learning, and the reference data held in the situation history database 122 may be machine-learned in advance as training data. In this case, the machine-learned situation detection unit 152 can quickly determine the situation without referring to the situation history database 122 .
  • the information presentation unit 161 presents the information input by the information input unit 110 to the information presentation target person 10 according to the reading specified by the reading specification unit 141 . Specifically, when the information presenting unit 161 presents the information, the reading specified by the reading specifying unit 141 is assigned to the word/phrase extracted by the word/phrase extracting unit 131 . When the information presentation unit 161 presents information to the information presentation target person 10 through the speaker 12F (which may be a speaker built into the smartphone 12B or the like) as an audio output unit, the phrase extracted by the phrase extraction unit 131 is input to the pronunciation designation unit 141. The speaker 12F is made to output the voice read aloud according to the reading specified in .
  • the information presentation unit 161 presents information to the information presentation target person 10 on the display 12E (which may be a display built into the smartphone 12B or the like) as a display unit
  • the words extracted by the phrase extraction unit 131 are read by the reading designation unit.
  • the reading information specified by 141 is displayed on the display 12E. For example, if the pronunciation of "igon" is specified for the word "will” in the example of FIG. However, it is possible to display the reading in parentheses after the kanji like ⁇ igon'', or display only the hiragana (reading) ⁇ igon'' instead of the kanji for ⁇ will''. You may
  • the description acquisition unit 162 acquires a description of the information input by the information input unit 110.
  • the explanation of the word/phrase extracted by the word/phrase extraction unit 131 is obtained from the word/phrase database 132 .
  • the explanation acquisition unit 162 acquires the explanation that the business can be carried out freely without doing anything.
  • the information presentation unit 161 presents the explanation acquired by the explanation acquisition unit 162 to the information presentation target person 10 together with the information input by the information input unit 110 .
  • the information presenting unit 161 causes the speaker 12F to read out "freedom to operate” and then read out the explanation "you can freely do business without infringing on the intellectual property rights of a third party". .
  • the information presenting unit 161 causes the word/phrase extracting unit 131 to display a message such as “FTO (freedom to operate: free to do business without infringing on the intellectual property rights of a third party)” on the display 12E.
  • the extracted word/phrase, the reading specified by the reading specification unit 141, and the explanation acquired by the explanation acquisition unit 162 may be collectively displayed. It is also possible to display words and phrases and how to read them, and to display explanations outside the main text as footnotes, such as "* You can freely do business without infringing on the intellectual property rights of third parties.”
  • the reading confirmation unit 170 includes an inquiry unit 171 and a response reception unit 172. If there are multiple reading candidates for the word extracted by the word extracting unit 131 and the reading specifying unit 141 cannot determine the reading even by referring to the background information acquired by the background information acquiring unit 150, the inquiry unit 171 obtains information The presentation target person 10 is asked how to read. The inquiry to the information presentation target person 10 may be made by voice through the speaker 12F, or may be made by display through the display 12E.
  • the reply receiving section 172 receives the information presentation target person 10's reply to the inquiry from the inquiry section 171 .
  • the information presentation target person 10 can use any input means of the user device group 12 to reply with the correct reading of the phrase related to the inquiry.
  • the information presentation target person 10 may respond by voice to a user device having a voice recognition function, may input the correct reading as text on the screen of the smartphone 12B or the like, or may select reading candidates. If the number is limited, the correct reading may be selected on the screen of the smartphone 12B or the like.
  • the reading designation unit 141 designates the reading of the word/phrase extracted by the word/phrase extraction unit 131 in response to the reply from the information presentation candidate 10 received by the reply reception unit 172 . Henceforth, unless the background information acquired by the background information acquisition unit 150 changes significantly, the reading specified here will be consistently used. Therefore, it is possible to prevent redundant inquiries to the information presentation target person 10 regarding the same phrase.
  • FIG. 3 shows an example of word/phrase registration in the word/phrase database 132 .
  • Multiple readings are registered for each word, and for each reading, a type, a typical usage scene, and an explanation are registered.
  • the type is a type such as a field in which the reading is used.
  • a typical usage scene is a typical scene in which the reading is used, and corresponds to the attribute and the situation of the information presentation target person 10 (or the conversation partner) acquired by the background information acquisition unit 150 .
  • the description as an optional item is a description of a word or phrase in that reading, and the information presentation section 161 presents the information presentation target person 10 with the description acquired by the description acquisition section 162 .
  • Each term will be described below.
  • the reading designation unit 141 designates the reading “yuigon” for the word “will”.
  • the type may be estimated from the attributes of the information presentation target person 10 or the conversation partner acquired by the background information acquisition unit 150 or the situation, and the reading of the type that matches the estimated type may be adopted.
  • the phrase "FTO” has two readings: “Freedom to Operate” and "FTO".
  • "Freedom to operate” is a reading used in the field of "intellectual property.”
  • the attributes and circumstances of the information presentation target person 10 or the conversation partner acquired by the background information acquisition unit 150 match or resemble a typical usage scene "examination of third party's intellectual property rights in the business development area”.
  • the reading designation unit 141 designates the reading “freedom to operate” for the word “FTO”.
  • the explanation acquiring unit 162 acquires from the word/phrase database 132 an explanation that ⁇ you can freely do business without infringing on the intellectual property rights of a third party,'' and causes the information presenting unit 161 to present it together.
  • “Ftio” is the reading of each alphabet as it is.
  • the attributes and the situation of the information presentation target person 10 or the conversation partner acquired by the background information acquisition unit 150 can be used in a typical usage scene of the reading "freedom to operate” as "a third party's intellectual property in the business development area”. If it is not similar to "Ken no shinsei", the reading designation unit 141 designates the reading "ef-to-oh” for the word "FTO”.
  • the word “0" that appears in the middle digit of the number may be read as “fly” or nothing.
  • ⁇ 1010 yen'' may be read as ⁇ sen ⁇ tonde'' juen'' or as ⁇ senjuen''.
  • "Tonde” is a reading used in the field of "finance”. If the attributes and situation of the information presentation target person 10 or the conversation partner acquired by the background information acquisition unit 150 match or resemble a typical usage scene “reading out amounts in the financial field”, the reading specification unit 141 selects the phrase Specify the reading "tonde” for "0".
  • the reading designation unit 141 designates that the word "0" is read as nothing.
  • the phrase "KYC” has two readings: “know your customer” and “keyy sea”. "Know Your Customer” is a reading used in the field of "commerce”. If the attributes and the situation of the information presentation target person 10 or the conversation partner acquired by the background information acquisition unit 150 match or resemble a typical usage scene “personal identification at the time of opening a bank account, etc.”, the reading designation unit 141 specifies the reading "know your customer” for the phrase “KYC”. At this time, the explanation acquiring unit 162 acquires the explanation “customer confirmation required when starting a commercial transaction with a customer” from the word/phrase database 132 and causes the information presenting unit 161 to present it together. "Kyysee” is the reading of each alphabet as it is.
  • the attributes and circumstances of the information presentation target person 10 or the conversation partner acquired by the background information acquisition unit 150 are used in a typical usage scene of "identity verification when opening a bank account, etc.” If they are not similar, the reading specifying unit 141 specifies the reading "Kywaisi" for the word "KYC”.
  • the phrase "Mita” has two readings: "Mita” and "Sanda.” "Mita” is a name often used in the Kanto region of Japan, and "Sanda” is a name often used in the Kansai region of Japan. Even if you refer to the background information, it is highly likely that you will not be able to determine how to read it. Therefore, a flag of "inquiry required” is entered in the "type” column so that the reading checking unit 170 inquires of the information presentation target person 10 about the reading. If this flag is present, the reading confirming unit 170 inquires the information presentation target person 10 about the reading, except in exceptional cases where the reading specifying unit 141 can determine the reading from the background information. The reading designation unit 141 designates the reading according to the response from the information presentation target person 10 to the inquiry.
  • FIG. 4 is a flowchart showing processing of the information processing device 1 according to the first embodiment.
  • “S” in the flow chart means “step”.
  • information to be presented to the information presentation target person 10 is input to the information input section 110 .
  • the word/phrase extraction unit 131 searches the word/phrase database 132 for each word/phrase included in the information input in S1.
  • the word/phrase extraction unit 131 determines whether or not the word/phrase searched in S2 is found in the word/phrase database 132 (hit). If no hits are found in the search, the information input in S1 does not contain a word or phrase that should be read carefully. presented to If the search hits, the word/phrase extraction unit 131 extracts the word/phrase and proceeds to S4.
  • the background information acquisition unit 150 acquires background information (attributes and/or situations) of the information presentation target person 10 or the conversation partner.
  • the reading designation unit 141 determines whether it is necessary to inquire of the information presentation target person 10 how to read the words extracted in S3. If the reading specification unit 141 can specify the reading based on the background information acquired in S4, the process proceeds to S8 without inquiring the information presentation target person 10. FIG. If the reading cannot be specified even with reference to the background information acquired in S4, or if there is a flag of "inquiry required" as in the example of "Mita” in FIG. 10 is determined to be required, and the process proceeds to S6.
  • the inquiry unit 171 inquires of the information presentation target person 10 how to read the words determined to require inquiry in S5.
  • the response receiving unit 172 receives the information presentation target person 10's response to the inquiry in S6.
  • the reading specifying unit 141 refers to the background information acquired in S4 and specifies how to read the words extracted in S3.
  • the reading specification unit 141 specifies the readings according to the information presentation target person 10's reply received in S7.
  • the explanation acquisition unit 162 acquires the explanation registered in the phrase database 132 for the phrase whose pronunciation is specified in S8.
  • the information presentation unit 161 presents the information input in S1 to the information presentation target person 10 according to the reading specified in S8. It should be noted that the explanation of the phrase whose explanation is acquired in S9 is presented to the information presentation target person 10 together with the information inputted in S1.
  • FIG. 5 is a functional block diagram of the information processing device 1 according to the second embodiment.
  • the same reference numerals are given to the same constituent elements as in the above-described embodiment, and the description thereof is omitted.
  • the presentation mode specification unit 142 refers to the background information about the information presentation target person 10 acquired by the background information acquisition unit 150 and specifies the presentation mode of the information input by the information input unit 110 .
  • the presentation modes registered in the presentation mode database 133 accessible by the presentation mode designating unit 142 include the speed of presenting the input information, the amount of presenting the input information, the input It includes at least one of a tone for presenting the input information and a voice for reading out the input information.
  • the information presentation unit 161 presents the information input by the information input unit 110 to the information presentation target person 10 in accordance with the presentation mode specified by the presentation mode specification unit 142 .
  • FIG. 6 shows an example of presentation mode registration in the presentation mode database 133 .
  • the presentation manner database 133 associates the background information (attribute/situation) of the information presentation target person 10 acquired by the background information acquisition unit 150 with the information presentation manner (speed/information amount/tone/volume/voice). configured as a table.
  • the presentation mode designating unit 142 may be configured by machine-learnable artificial intelligence, and a table held in the presentation mode database 133 may be machine-learned in advance as training data.
  • the presentation mode specifying unit 142 can not only quickly process cases that match the table, but also flexibly process cases that do not match the table while performing autonomous machine learning.
  • the information presentation speed is "slow” and the information presentation amount is “small”.
  • the reading speed is slowed down when the information is presented on the speaker 12F
  • the display speed is slowed down when the information is presented on the display 12E.
  • Edit processing such as deletion is performed.
  • the attribute of the information presentation target person 10 is "minor"
  • the tone of information presentation "gentle” when the information is presented on the speaker 12F, the reading tone is softened, and when the information is presented on the display 12E, hiragana characters are used more than kanji characters. Soften the expression by using ⁇ desu-masu-cho'' instead of ⁇ dearu-cho''.
  • the attribute of the information presentation target person 10 When the attribute of the information presentation target person 10 is "elderly”, it is considered that basically the same information presentation mode as “minor” is suitable. The volume of information presented by the speaker 12F is increased. When information is presented on the display 12E, the display size of characters is “larger.” In addition, if the attribute of the information presentation target person 10 is "person whose native language is not the presentation language”, it is assumed that the information processing ability in the presentation language is inferior, so the same as “minor” and “elderly” to "slow” the information presentation speed and "decrease” the amount of information presented. When the attribute of the information presentation target person 10 is "person from a specific region", the information may be presented by interweaving the accent or dialect of the hometown.
  • the information presentation speed is set to “slow” and the information presentation amount is set to "low” so as not to interfere with the relaxed atmosphere or cause noise to the family and neighbors.
  • the information presentation target person 10 is in a situation of "working/studying”, there is no problem even if the information presentation amount is "large” because the information processing ability is enhanced by active brain activity. Keep the volume low so as not to disturb your concentration.
  • the concentration of the information presentation target person 10 is particularly high, by temporarily refraining from presenting information other than information that is highly important or urgent for the information presentation target person 10, the information presentation target person 10 can concentrate.
  • the information presentation target person 10 When the information presentation target person 10 is placed in an "emergency" situation, it is necessary to concisely and reliably convey to the information presentation target person 10 only the information necessary to deal with the emergency, so the information presentation speed is reduced.
  • the amount of information presented is “faster”
  • the amount of information presented is “smaller”
  • the tone of information presentation is “severe”
  • the sound volume is “larger”.
  • the information presentation speed is set to "slow” so that even the information presentation target person 10 whose information processing ability is reduced can understand.
  • the tone of information presentation is set to be gentle and the sound volume is set to low so as not to disturb the rest of the information presentation target person 10.
  • the information presentation mode may be determined for each place or event. For example, information may be presented to the information presentation target person 10 using terms, expressions, accents, dialects, character voices, event information, etc. specific to a place or event.
  • the weather around the information presentation target person 10 can be predicted from the temperature and humidity information, or the weather forecast for that place on that date and time can be obtained from a predetermined weather forecast server according to the date and time information and the place information. Therefore, it is possible to present information according to those weather conditions. For example, if it is raining, a soft voice is provided, and if it is fine weather, a dry voice is provided. Provides tense audio if in a typhoon.
  • FIG. 7 is a flowchart showing processing of the information processing device 1 according to the second embodiment.
  • information to be presented to the information presentation target person 10 is input to the information input unit 110 .
  • the background information acquisition unit 150 acquires background information (attributes and/or situations) of the information presentation target person 10 or the conversation partner.
  • the presentation mode specifying unit 142 refers to the background information acquired in S12 and the presentation mode database 133, and specifies the presentation mode of the information input in S11.
  • the information presentation unit 161 presents the information input in S11 to the information presentation target person 10 in accordance with the presentation mode specified in S13.
  • information can be presented in an appropriate manner that matches the information presentation target person 10's attributes and circumstances.
  • FIG. 8 is a functional block diagram of the information processing device 1 according to the third embodiment.
  • the same reference numerals are given to the same constituent elements as in the above-described embodiment, and the description thereof is omitted.
  • the information input unit 110 receives information on minutes to be presented to the information presentation target person 10 .
  • the phrase extraction unit 131 extracts phrases from the minutes information input to the information input unit 110 .
  • the word/phrase extraction unit 131 searches the word/phrase database 132 for each word/phrase included in the information of the minutes input to the information input unit 110 .
  • phrases whose presentation mode should be changed according to the background information or attributes of the information presentation target person 10 are registered in advance. 143.
  • the minutes processing unit 143 refers to the background information about the information presentation target person 10 held in the information presentation target person database 153, and transmits the information of the minutes input to the information input unit 110 to the information presentation target person 10. processed together. Specifically, the minutes processing unit 143 changes the presentation mode of the phrases extracted by the phrase extraction unit 131 in accordance with the background information of the information presentation target person 10 .
  • the information presentation unit 161 presents the information of the minutes processed by the minutes processing unit 143 to the information presentation target person 10 .
  • an electronic file of the minutes processed for each information presentation target person 10 or an e-mail in which the minutes processed for each information presentation target person 10 are written in the text are sent to each information presentation target person 10. sent separately to
  • FIG. 9 shows an example of word/phrase registration in the word/phrase database 132 .
  • a type for each of the words “AAA” to “GGG”, a type, a disclosing range, a well-known range, and an explanation to be added when disclosing outside the well-known range are registered.
  • the type is a pattern of each word, and examples include "company name”, “business name”, “product name”, “service name”, “project name”, “technical name”, and “organization name”.
  • the disclosing range is the range in which each word can be disclosed, and is specified by the organization such as company or division, title such as president or manager, membership in project or task force, or the like.
  • the minutes processing unit 143 deletes the words and phrases, and paraphrases them into words and phrases that can be disclosed. etc. processing or confidentiality processing.
  • the well-known range is the range in which each word is well-known, and is specified by affiliated organization, title, membership, etc., similar to the disclosing range.
  • the explanation added when disclosing outside the known range is a supplementary explanation added by the explanation acquisition unit 162 when providing the minutes to the information presentation target person 10 who is within the disclosing possible range and outside the known range, Typically, it is information about the outline of each word. These summaries are not included in the minutes because they are known to the information presentation target persons 10 within the known range, but are useful for understanding unknown words and phrases for the information presentation target persons 10 outside the known range. included in the record.
  • FIG. 10 shows an example of registration of background information in the information presentation target person database 153.
  • Each information presentation target person 10 is listed as a person to whom minutes are to be sent, and an external flag, an organization to which the person belongs, and a title are registered for each person to whom minutes are to be sent.
  • the outside flag is a flag indicating that the person to whom the minutes are to be sent is outside the company.
  • the illustrated example is the information presentation target person database 153 constructed by "AAA Corporation", and the outside flag is set for the minutes delivery target person "F" who belongs to "XXX Corporation” outside "AAA Corporation”. be done.
  • each confidential word registered in the word/phrase database 132 each word whose disclosure range is limited to "AAA Corporation" ) must be carefully processed (or concealed). These processes are performed by the minutes processing unit 143.
  • the minutes processing unit 143 or the information presenting unit 161 can determine whether or not to forward the minutes to the person in charge of final confirmation according to the presence or absence of the external flag.
  • the minutes processing unit 143 does not process any of the minutes delivery target persons "A" to "F".
  • these words and phrases are included in the minutes to be sent to "A” to "E” who belong to "AAA Corporation”.
  • No supplementary explanation is added.
  • supplementary explanations summary of AAA corporation or outline of BBB business for these words are added to the minutes to be sent to the minutes recipient "F" who belongs to "XXX Corporation”.
  • the minutes processing unit 143 is subjected to confidentiality processing such as deletion of the word "DDD".
  • the word "DDD” is not well known, the minutes sent to the participants "A” to “C” within the scope of possible disclosure include a supplementary explanation of the word “DDD” (for the DDD service). summary) is added.
  • the minutes processing unit 143 uses the word “GGG ” will be processed, such as rephrasing it into other words that can be disclosed.
  • the range of the word “GGG” that has been known is only “AAA Co., Ltd. / GGG Office”, and in the minutes to be sent to the target persons "A” to "E” who are not in the known range , Supplementary explanation about the word “GGG” (outline of the GGG room) is added.
  • FIG. 11 is a flowchart showing processing of the information processing device 1 according to the third embodiment.
  • the information of the minutes to be presented to the information presentation target person 10 is input to the information input section 110 .
  • the word/phrase extraction unit 131 searches the word/phrase database 132 for each word/phrase included in the information of the minutes input in S15.
  • the word/phrase extraction unit 131 determines whether or not the word/phrase searched in S16 is found in the word/phrase database 132 (hit). If no hits are found in the search, the information input in S15 does not include words that should be noted when sending the minutes to each information presentation target person 10. The minutes input in step 2 are sent to the information presentation target person 10 as they are. If the search hits, the word/phrase extraction unit 131 extracts the word/phrase and proceeds to S18.
  • the minutes processing unit 143 designates one of the minutes delivery target persons "A" to "F" registered in the information presentation target person database 153.
  • the minutes processing unit 143 acquires from the information presentation target person database 153 the background information such as the "outside company flag", "organization”, and "title” of the person to whom the minutes are to be sent specified in S18.
  • the minutes processing unit 143 searches the word/phrase database 132 based on the background information acquired in S19, and determines whether or not there is a word/phrase outside the disclosing range. If there is a word/phrase outside the disclosing range, the process proceeds to S21, and the minutes processing unit 143 deletes or corrects the word/phrase.
  • the minutes processing unit 143 searches the word/phrase database 132 based on the background information acquired in S19, and determines whether or not there is a word/phrase outside the known range. If there is a word/phrase outside the known range, the process proceeds to S23, and the explanation acquisition unit 162 adds a supplementary explanation to the word/phrase.
  • the minutes processing unit 143 determines whether or not all the minutes delivery target persons "A" to "F" have been specified in S18. If there is an unspecified person to whom the minutes are to be sent, the process returns to S18, a new person to whom the minutes are to be sent is specified, and the processes of S19 to S24 are repeated. When the processing of S18 to S23 is completed for all the minutes delivery target persons "A" to "F", the process proceeds to S25, and the information presentation unit 161 displays the processed minutes for each minutes delivery target person. to the person.
  • the minutes can be appropriately processed according to the information presentation target person 10's attributes and circumstances.
  • the technical idea of the third embodiment may be applied to the first embodiment or the second embodiment. That is, when presenting the input information to the information presentation target person in voice or text, referring to the phrase database, it is determined whether it is within the range that can be disclosed and whether it is within the known range, The voice or text to be output may be changed according to the determination result.
  • FIG. 12 is a functional block diagram of the information processing device 1 according to the fourth embodiment.
  • the same reference numerals are given to the same constituent elements as in the above-described embodiment, and the description thereof is omitted.
  • the privacy determination unit 144 refers to background information about the information presentation target person 10 acquired by the background information acquisition unit 150 and determines whether or not privacy protection is necessary.
  • the criterion for determining whether privacy protection is necessary can be arbitrarily set based on the attribute of the information presentation target person 10 acquired by the attribute acquisition unit 151 and the information presentation target person 10 situation detected by the situation detection unit 152 .
  • AAA Corporation's privacy protection means that confidential information related to AAA Corporation's business, products, services, projects, technology, organization, etc. is kept confidential so that it is not leaked outside the company, as in the third embodiment above. means that
  • the anonymization processing unit 145 subjects at least part of the information input by the information input unit 110 to an encryption process.
  • the anonymization processing unit 145 deletes the personal information by which the information presentation target person 10 can be identified, Confidentiality processing such as replacement with the information of For example, when calling the information presentation target person 10 by automatic voice from the speaker 12F functioning as the information presentation unit 161 in a hospital, government office, etc., normally, "Mr. AAA, please come to the counter" etc.
  • the anonymization processing unit 145 includes the information presentation target person 10 himself, the affiliation AAA Corporation, the AAA Corporation's business, Confidentiality processing is performed, such as deleting information that can identify products, services, projects, technologies, organizations, etc., or replacing it with other information.
  • the information presentation unit 161 presents the information subjected to the security processing by the security processing unit 145 to the information presentation target person 10 .
  • FIG. 13 is a flowchart showing processing of the information processing device 1 according to the fourth embodiment.
  • S ⁇ b>26 information to be presented to the information presentation target person 10 is input to the information input unit 110 .
  • the background information acquisition unit 150 acquires background information (attributes and/or situations) of the information presentation target person 10 .
  • the privacy determination unit 144 determines the necessity of privacy protection by referring to the background information about the information presentation target person 10 acquired in S27.
  • the anonymization processing unit 145 performs an anonymization process on at least part of the information input in S26.
  • the information presentation unit 161 presents the information subjected to the confidentiality processing in S29 to the information presentation target person 10.
  • FIG. 1 information to be presented to the information presentation target person 10 is input to the information input unit 110 .
  • the privacy can be appropriately protected according to the information presentation target person 10's attributes and circumstances.
  • each device described in the embodiments can be realized by hardware resources or software resources, or by cooperation between hardware resources and software resources.
  • Processors, ROMs, RAMs, and other LSIs can be used as hardware resources.
  • Programs such as operating systems and applications can be used as software resources.
  • the present invention relates to information processing technology.
  • 1 information processing device 10 information presentation target person, 12 user device group, 110 information input unit, 120 background information database, 121 attribute information database, 122 situation history database, 12E display, 12F speaker, 131 phrase extraction unit, 132 phrase database , 133 presentation mode database, 141 reading specification unit, 142 presentation mode specification unit, 143 minutes processing unit, 144 privacy determination unit, 145 confidentiality processing unit, 150 background information acquisition unit, 151 attribute acquisition unit, 152 situation detection unit, 153 Information presentation target person database, 161 information presentation unit, 162 description acquisition unit, 170 reading check unit, 171 inquiry unit, 172 reply reception unit.

Landscapes

  • Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computational Linguistics (AREA)
  • Human Computer Interaction (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • General Health & Medical Sciences (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Artificial Intelligence (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

Un dispositif de traitement d'informations (1) comprend : une unité d'entrée d'informations (110) dans laquelle des informations à présenter à un sujet (10) de présentation d'informations sont entrées ; une unité d'extraction de phrase (131) qui extrait une phrase enregistrée dans une base de données de phrases (132) à partir des informations entrées dans l'unité d'entrée d'informations (110) ; une unité de désignation de lecture phonétique (141) qui se réfère à des informations d'arrière-plan relatives au sujet (10) de présentation d'informations acquises par une unité d'acquisition d'informations d'arrière-plan (150) et désigne une lecture phonétique relative à la phrase extraite par l'unité d'extraction de phrase (131) ; et une unité de présentation d'informations (161) qui présente les informations entrées dans l'unité d'entrée d'informations (110) au sujet (10) de présentation d'informations conformément à la lecture phonétique désignée par l'unité de désignation de lecture phonétique (141).
PCT/JP2021/014513 2021-04-05 2021-04-05 Dispositif, procédé et programme de traitement d'informations WO2022215120A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/014513 WO2022215120A1 (fr) 2021-04-05 2021-04-05 Dispositif, procédé et programme de traitement d'informations

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/JP2021/014513 WO2022215120A1 (fr) 2021-04-05 2021-04-05 Dispositif, procédé et programme de traitement d'informations

Publications (1)

Publication Number Publication Date
WO2022215120A1 true WO2022215120A1 (fr) 2022-10-13

Family

ID=83546295

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2021/014513 WO2022215120A1 (fr) 2021-04-05 2021-04-05 Dispositif, procédé et programme de traitement d'informations

Country Status (1)

Country Link
WO (1) WO2022215120A1 (fr)

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002207728A (ja) * 2001-01-12 2002-07-26 Fujitsu Ltd 表音文字生成装置及びそれを実現するためのプログラムを記録した記録媒体
JP2008139969A (ja) * 2006-11-30 2008-06-19 Fuji Xerox Co Ltd 議事録作成装置、会議情報管理システム及びプログラム
JP2011102910A (ja) * 2009-11-11 2011-05-26 Nippon Telegr & Teleph Corp <Ntt> 聴覚特性を反映した音声読み上げ方法、その装置及びプログラム
JP2013065284A (ja) * 2011-08-11 2013-04-11 Apple Inc 言語変換において複数の読み方の曖昧性を除去する方法
JP2016122183A (ja) * 2014-12-09 2016-07-07 アップル インコーポレイテッド 音声合成における同綴異音異義語の曖昧さの解消
JP2019008477A (ja) * 2017-06-22 2019-01-17 富士通株式会社 判別プログラム、判別装置及び判別方法
WO2020110744A1 (fr) * 2018-11-28 2020-06-04 ソニー株式会社 Dispositif de traitement d'informations, procédé de traitement d'informations et programme
JP2020149628A (ja) * 2019-03-15 2020-09-17 エヌ・ティ・ティ・コミュニケーションズ株式会社 情報処理装置、情報処理方法およびプログラム

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002207728A (ja) * 2001-01-12 2002-07-26 Fujitsu Ltd 表音文字生成装置及びそれを実現するためのプログラムを記録した記録媒体
JP2008139969A (ja) * 2006-11-30 2008-06-19 Fuji Xerox Co Ltd 議事録作成装置、会議情報管理システム及びプログラム
JP2011102910A (ja) * 2009-11-11 2011-05-26 Nippon Telegr & Teleph Corp <Ntt> 聴覚特性を反映した音声読み上げ方法、その装置及びプログラム
JP2013065284A (ja) * 2011-08-11 2013-04-11 Apple Inc 言語変換において複数の読み方の曖昧性を除去する方法
JP2016122183A (ja) * 2014-12-09 2016-07-07 アップル インコーポレイテッド 音声合成における同綴異音異義語の曖昧さの解消
JP2019008477A (ja) * 2017-06-22 2019-01-17 富士通株式会社 判別プログラム、判別装置及び判別方法
WO2020110744A1 (fr) * 2018-11-28 2020-06-04 ソニー株式会社 Dispositif de traitement d'informations, procédé de traitement d'informations et programme
JP2020149628A (ja) * 2019-03-15 2020-09-17 エヌ・ティ・ティ・コミュニケーションズ株式会社 情報処理装置、情報処理方法およびプログラム

Similar Documents

Publication Publication Date Title
US10679005B2 (en) Speech recognition and summarization
US10540521B2 (en) Selective enforcement of privacy and confidentiality for optimization of voice applications
US10777206B2 (en) Voiceprint update method, client, and electronic device
US11849256B2 (en) Systems and methods for dynamically concealing sensitive information
US11475344B2 (en) User identification with voiceprints on online social networks
JP2022551788A (ja) 補助システムのためのプロアクティブコンテンツを生成すること
US20080240379A1 (en) Automatic retrieval and presentation of information relevant to the context of a user&#39;s conversation
US11562744B1 (en) Stylizing text-to-speech (TTS) voice response for assistant systems
WO2020086343A1 (fr) Sensibilisation à la confidentialité pour des communications d&#39;assistant personnel
US20160171109A1 (en) Web content filtering
US11321675B2 (en) Cognitive scribe and meeting moderator assistant
US9564124B2 (en) Displaying relevant information on wearable computing devices
US20160065539A1 (en) Method of sending information about a user
JP2023531346A (ja) 補助システムにおけるマルチパーソンコーリングのための単一の要求の使用
Schulze et al. Conversational context helps improve mobile notification management
KR20150041592A (ko) 피호출자의 전자 디바이스에서 연락처 정보를 업데이트하는 방법 및 전자 디바이스
US20220035840A1 (en) Data management device, data management method, and program
WO2022215120A1 (fr) Dispositif, procédé et programme de traitement d&#39;informations
US12003575B2 (en) Routing of sensitive-information utterances through secure channels in interactive voice sessions
US12056448B2 (en) Systems and methods for proactively extracting data from complex documents
US11782986B2 (en) Interactive query based network communication through a media device
Yeasmin Privacy analysis of voice user interfaces
JP2022018724A (ja) 情報処理装置、情報処理方法、及び情報処理プログラム
US20240038222A1 (en) System and method for consent detection and validation
WO2023013062A1 (fr) Système, dispositif et procédé de traitement d&#39;informations ainsi que support d&#39;enregistrement

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 21935928

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 21935928

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: JP