JP2020144676A - Information linking system, information linking server, voice i/o device, information linking method, and program - Google Patents

Abstract

To provide a technique that allows for recording necessary information through natural dialogue without having to perform special manipulation.SOLUTION: An information linking system is provided, comprising: a plurality of voice input/output devices for receiving utterance voice; a symptom information storage unit for storing medical record information including biological information of patients; a voice recognition unit for converting entered utterance voice into text; a symptom information generation unit configured to extract at least biological information of a patient from the text and store the extracted information in the symptom information storage unit; and an output generation unit configured to send the biological information to the voice input/output device having key information associated with each other to let the voice input/output devices output the biological information.SELECTED DRAWING: Figure 1

Description

The present invention relates to an information linkage system, an information linkage server, a voice input / output device, an information linkage method and a program.

Patent Document 1 describes "an input support device that adds a voice input function to application software that displays an input area for inputting information on an input screen, and the display position of the input area and the input. The screen configuration table storage unit that stores the screen configuration table associated with the first information that specifies the reading that represents the area, and the second information that represents the result of the voice recognition processing for the spoken voice are acquired and acquired. With reference to the screen configuration table, the display position of the input area corresponding to the second information is specified, and the operation information for selecting the specified display position is input to the application software with the input information generation unit. , Is described as "input support device having."

Japanese Unexamined Patent Publication No. 2013-061793

In the above technique, the input area is operated by utterance, and the item name can be input by uttering the input content after uttering, but it cannot be properly input from a natural dialogue.

An object of the present invention is to record necessary information by natural dialogue without any special operation.

The present application includes a plurality of means for solving at least a part of the above problems, and examples thereof are as follows. The information linkage system according to one aspect of the present invention converts a plurality of voice input / output devices that receive spoken voice, a symptom information storage unit that stores chart information including biological information of a patient, and input spoken voice into text. A voice recognition unit, a symptom information creation unit that extracts at least the biological information of the patient from the text and stores it in the symptom information storage unit, and the voice input / output device having key information associated with each other. It has an output creation unit for transmitting and outputting to.

Further, the information linkage system according to another aspect of the present invention includes a plurality of voice input / output devices and an information linkage server device that is communicably connected to the voice input / output device via a network, and includes the voice input / output device. Each device includes a control unit that converts the input speech voice into text and transmits it to the information cooperation server device together with predetermined key information, and the information cooperation server device stores information including biometric information of the patient. The symptom information storage unit, the symptom information creation unit that extracts at least the biological information of the patient from the text and stores the biological information in the symptom information storage unit, and the voice having the key information associated with each other. It has an output creation unit for transmitting and outputting to an input / output device.

Further, in the above-mentioned information cooperation system, at least one of the above-mentioned voice input / output devices may be provided in the emergency vehicle, and the above-mentioned patient may be a patient housed in the above-mentioned emergency vehicle.

Further, in the above-mentioned information linkage system, at least one of the above-mentioned voice input / output devices is provided in an emergency room where a medical worker performs medical treatment, and the above-mentioned voice input / output device displays at least the above-mentioned biological information. It may be characterized by including a display device.

Further, in the above-mentioned information cooperation system, the control unit of the above-mentioned voice input / output device performs a voice recognition process for recognizing the above-mentioned uttered voice in the process of converting the uttered voice into text, and in the process, the above-mentioned patient's The predetermined terms related to the biometric information may be characterized in that they are not converted into similar terms.

Further, in the above-mentioned information cooperation system, the above-mentioned information cooperation server device includes a time-series conversation processing unit that transmits the above-mentioned texts to the above-mentioned voice input / output device having the above-mentioned key information common to each other and shares them in time series. When an utterance about a new patient is made from one of the voice input / output devices, the key information is issued, and the key information is transmitted to at least one of the other voice input / output devices different from the voice input / output device. It may be characterized by associating.

Further, in the above-mentioned information cooperation system, the above-mentioned output creation unit creates the above-mentioned screen information for displaying the above-mentioned text in time series together with the above-mentioned biometric information, and has the above-mentioned key information associated with each other. It may be characterized in that it is output to.

Further, the above-mentioned information linkage system is provided with a log collecting unit that collects log information regarding the operating state of a predetermined device, and the output creating unit arranges the operating state and the time of utterance in chronological order. It may be characterized in that it generates information.

Further, the information linkage server device according to another aspect of the invention includes a communication unit that receives textualized information of spoken voice input from a plurality of voice input / output devices and key information via a network, and a patient. The symptom information storage unit that stores information including the biological information of the above, the symptom information creation unit that extracts at least the biological information of the patient from the text and stores the biological information in the symptom information storage unit, and the biological information are associated with each other. It also has an output creation unit that transmits and outputs to the audio input / output device having the key information.

Further, the above-mentioned information linkage server device is characterized in that at least one of the above-mentioned voice input / output devices is provided in an emergency vehicle, and the above-mentioned patient is a patient housed in the above-mentioned emergency vehicle. You may.

Further, in the above-mentioned information linkage server device, at least one of the above-mentioned voice input / output devices is provided in an emergency room where a medical worker performs medical treatment, and the above-mentioned voice input / output device displays at least the above-mentioned biometric information. It may be characterized in that it is provided with a display device.

Further, the above-mentioned information linkage server device includes a time-series conversation processing unit that transmits the above-mentioned texts to the above-mentioned voice input / output device having the above-mentioned key information associated with each other and shares them in a time-series manner. When a new patient is spoken from one of the devices, the key information is issued to associate the key information with at least one of other voice input / output devices different from the voice input / output device. It may be.

Further, in the above-mentioned information linkage server device, the above-mentioned output creation unit creates the above-mentioned screen information for displaying the above-mentioned text in chronological order together with the above-mentioned biometric information, and has the above-mentioned key information associated with each other. It may be characterized in that it is transmitted to an apparatus and output.

Further, the information linkage server device according to another aspect of the invention receives input spoken voice information and key information from a plurality of voice input / output devices via a network, a communication unit, and a patient's biological information. A symptom information storage unit that stores information including the information, a voice recognition unit that converts the spoken voice information into text, and a symptom information creation unit that extracts at least the biological information of the patient from the text and stores it in the symptom information storage unit. And an output creation unit that transmits and outputs the biometric information to the voice input / output device having the key information associated with each other.

Further, the information linkage system according to another aspect of the present invention includes a plurality of voice input / output devices and an information linkage server device that is communicably connected to the voice input / output device via a network. Each output device includes a control unit that receives the input spoken voice and transmits it to the information linkage server device together with predetermined key information, and the information linkage server device stores information including biometric information of the patient. A symptom information storage unit, a voice recognition unit that receives the spoken voice and converts it into a text, a symptom information creation unit that extracts at least the biological information of the patient from the text and stores it in the symptom information storage unit. It has an output creation unit that transmits and outputs the biometric information to the voice input / output device having the key information associated with each other.

Further, the voice input / output device according to another aspect of the invention includes a communication unit that connects to a predetermined information linkage server device so as to be communicable via a network, and the input spoken voice is converted into text and together with predetermined key information. The control unit includes a control unit for transmitting to the information cooperation server device, and the control unit outputs the biometric information when the biometric information is received from the information cooperation server device.

Further, the above-mentioned voice input / output device may be characterized by including at least a display device for displaying the above-mentioned biological information.

Further, in the above-mentioned voice input / output device, the above-mentioned control unit performs a voice recognition process for recognizing the above-mentioned utterance sound in a process of converting the above-mentioned utterance sound into text, and in the process, a predetermined term relating to biological information is used. May be characterized in that it does not convert to an approximate word.

Further, the information linkage method according to another aspect of the invention is an information linkage method by an information linkage system, in which the information linkage system includes a plurality of voice input / output devices for receiving spoken voice and information including biometric information of a patient. The control unit includes a symptom information storage unit and a control unit that memorize the above, and the control unit extracts at least the biometric information of the patient from the text and a voice recognition step for converting the input spoken voice into text. A symptom information creation step to be stored in the symptom information storage unit and an output creation step to transmit and output the biological information to the voice input / output device having key information associated with each other are performed.

Further, the program according to another aspect of the invention is a program for causing a computer to link information, and the computer is provided with information obtained by converting spoken voice input from a plurality of voice input / output devices into text and key information. , A communication unit that receives the above information via a network, a control unit, and a symptom information storage unit that stores information including the biometric information of the patient, and the control unit receives at least the biometric information of the patient from the text. A symptom information creation step of extracting and storing the symptom information storage unit and an output creation step of transmitting the biological information to the voice input / output device having the key information associated with each other and outputting the information are performed. It is characterized by.

According to the present invention, it is possible to provide a technique for recording necessary information through natural dialogue without any special operation. Issues, configurations and effects other than those described above will be clarified by the description of the following embodiments.

It is a block diagram of an emergency information cooperation system. It is a figure which shows the example of the data structure of a term storage part. It is a figure which shows the data structure example of the user voiceprint storage part. It is a figure which shows the example of the data structure of the belonging room storage part. It is a figure which shows the example of the data structure of the symptom information storage part. It is a figure which shows the example of the data structure of the utterance history storage part. It is a figure which shows the hardware configuration example of a smart speaker. It is a figure which shows the hardware configuration example of the emergency information cooperation server apparatus. It is a figure which shows the example of the flow of the process at the time of emergency transportation (until the start of a room). It is a figure which shows the example of the flow of the emergency transport processing (addition of symptom information). It is a figure which shows the example of the output screen. It is a block diagram of an emergency information cooperation system. It is a figure which shows the example of the data structure of the peripheral device operation log storage part. It is a figure which shows the example of the output screen. It is a block diagram of an emergency information cooperation system. It is a figure which shows the example of the flow of the process at the time of emergency transportation (until the start of a room). It is a figure which shows the example of the flow of the emergency transport processing (addition of symptom information).

Hereinafter, an emergency information linkage system to which an embodiment of the present invention is applied will be described with reference to the drawings. In the following embodiments, when necessary for convenience, the description will be divided into a plurality of sections or embodiments, but unless otherwise specified, they are not unrelated to each other, and one is the other. It is related to some or all of the modified examples, details, supplementary explanations, etc.

In addition, in the following embodiments, when the number of elements (including the number, numerical value, quantity, range, etc.) is referred to, when it is specified in particular, or when it is clearly limited to a specific number in principle, etc. Except, the number is not limited to the specific number, and may be more than or less than the specific number.

Furthermore, in the following embodiments, the components (including element steps, etc.) are not necessarily essential unless otherwise specified or clearly considered to be essential in principle. Needless to say.

Similarly, in the following embodiments, when referring to the shape, positional relationship, etc. of a component or the like, the shape is substantially the same except when it is clearly stated or when it is considered that it is not so clearly in principle. Etc., etc. shall be included. This also applies to the above numerical values and ranges.

Further, in all the drawings for explaining the embodiment, the same members are, in principle, given the same reference numerals, and the repeated description thereof will be omitted.

FIG. 1 is a block diagram of an emergency information linkage system according to this embodiment. The emergency information cooperation system 1 includes an emergency information cooperation server device 200, and a plurality of smart speakers 100 and 300 that are communicably connected to the emergency information cooperation server device 200 via a network 50. A plurality of users have a dialogue using the smart speakers 100 and 300, detect biological information included in the dialogue (voice), and configure symptom information. In addition, the emergency information linkage system 1 converts the history of time-series remarks (voices) in dialogues using the smart speakers 100 and 300 into texts (characters), and stores them so that they can be easily referred to later. Let me. Each of the smart speakers 100 and 300 is used by connecting to the emergency information cooperation server device 200.

When connecting from the smart speakers 100 and 300 to the emergency information linkage server device 200, a LAN (Local Area Network), WAN (Wide Area Network), the Internet, a mobile phone network, or a communication network in which these are combined is used. It is connected via a certain network 50. The network 50 may be a VPN (Virtual Private Network) or the like on a wireless communication network such as a mobile phone communication network.

The smart speakers 100 and 300 are typically arranged in an emergency vehicle such as an ambulance and an emergency room in a hospital, and receive voice input by a staff member who works at each location. For example, emergency personnel (emergency paramedics, etc.) and firefighters are assigned to the emergency vehicle to perform lifesaving work, and doctors, nurses, and other staff are assigned to the emergency room to perform emergency response work. The smart speakers 100 and 300 receive the spoken voices of these staff members as inputs and transmit them to the emergency information cooperation server device 200.

Here, when a patient is brought into an ambulance or the like, an ambulance or the like makes an emergency request to an emergency counter or the like of a nearby hospital, and the patient's condition and treatment status are coordinated through a voice dialogue of each staff member. In addition, biometric information such as patient observation findings is entered in documents in a predetermined format such as pre-hospital rescue records and pre-hospital trauma transfer forms, and is sent from the ambulance to the hospital.

In this cooperation, there are many cases where information is not sufficiently shared due to the possibility that the dialogue and the document are inconsistent, the cooperation in the dialogue is unclear, or the description in the document is unclear. In addition, emergency personnel in emergency vehicles need to prepare documents such as rescue records promptly and accurately in parallel with the lifesaving action, and it may be difficult to be compatible with the lifesaving action.

When the smart speakers 100 and 300 receive utterances, the emergency information linkage system 1 performs voice recognition processing for converting the input utterance voice into text, extracts at least the patient's biological information from the text, and is symptom information that can be said to be chart information. Is performed, and biometric information is transmitted to smart speakers 100 and 300 having key information associated with each other for output.

Specifically, when the smart speakers 100 and 300 receive utterances by emergency personnel or medical staff, the input utterance voice is converted into text and the network 50 is connected to the emergency information cooperation server device 200 together with predetermined key information. Send via. Since the predetermined key information is not held for the first time, empty information is transmitted.

Further, the smart speakers 100 and 300 store the voiceprint information of the user in advance, determine the voiceprint from the spoken voice, and identify the speaker.

When the emergency information linkage server device 200 receives the text of the utterance voice, it issues an ID that identifies the virtual room shared with any of the smart speakers 300 installed in the emergency room, and thereafter uses that ID as key information. It will be sent to both smart speakers 100 and 300. When the text is received together with the predetermined key information, the ID for identifying the virtual room is not issued, and the predetermined key information is used as the ID for identifying the virtual room.

The ID that identifies the virtual room may be given an effective time in advance. For example, the time during which the addition of remarks is valid (for example, within 30 minutes from the issuance of the ID) and the time during which only the reference is valid (for example, within 3 hours after the issuance of the ID) are given separately. May be good. In that case, if the ID is received together with the text information within the time when the addition of the remark becomes effective (for example, within 30 minutes from the issuance of the ID), the emergency information linkage server device 200 speaks to the virtual room. Is added, and information is added to the biometric information. If the ID is accepted together with the text information within the time when the addition of remarks becomes effective but only the reference becomes effective (for example, within 3 hours after 30 minutes have passed from the issuance of the ID), emergency information linkage The server device 200 transmits the utterance history and biometric information of the virtual room to the smart speaker of the transmission source.

When the ID is accepted together with the text information within the time when the addition of the remark becomes effective and not within the time when only the reference becomes effective (for example, 3 hours after the issuance of the ID), or only the text information When the message is received, the emergency information linkage server device 200 issues an ID for identifying the virtual room shared with the smart speaker 300 to the source smart speaker 100 and sends it to both the smart speaker 100 and 300.

By setting the time limit activated in the virtual room in this way, the information for one emergency dispatch (patient) can be organized without hassle and can be referred to from the hospital to which the patient was transported.

When the emergency information linkage server device 200 receives the text of the spoken voice, at least the patient's biological information (for example, quantified information such as pulse and blood pressure, presence / absence of bleeding, presence / absence of pain, position of pain) is received from the text. , Awareness level, coping status, etc., or multiple pieces of information) to create symptom information that can be called medical record information. Then, an output creation process of transmitting the information to the smart speakers 100 and 300 having key information associated with each other and outputting the information is performed.

Here, in order to extract the biological information of the patient, the emergency information cooperation server device 200 analyzes the utterance. For example, a trained model constructed by performing machine learning related to a language can be used for this analysis, but the analysis is not limited to this, and analysis is performed by various arithmetic processes generally called artificial intelligence. It is also possible to process it by arithmetic processing using a rule-based engine.

Further, the emergency information linkage server device 200 receives the utterance voice together with the text of the utterance voice, and stores the utterance person and the utterance content in chronological order for each virtual room.

As shown in FIG. 1, the smart speaker 100 includes a storage unit 110, a control unit 120, a communication unit 130, a voice input unit 140, and a voice output unit 150. The storage unit 110 includes a term storage unit 111, a user voiceprint storage unit 112, and a belonging room storage unit 113.

FIG. 2 is a diagram showing an example of a data structure of a term storage unit. The term storage unit 111 includes a term identifier 111A, a term 111B, pronunciation information 111C, and a part of speech 111D. Term identifier 111A is an identifier that can be distinguished from other terms. Term 111B is a medical term, various terms, abbreviations, dialects, etc. in emergency lifesaving work. The pronunciation information 111C is information indicating the word specified by the term 111B by a predetermined phonetic symbol or the like, or voice information itself. Part of speech 111D is information that identifies the part of speech of the word specified by term 111B.

FIG. 3 is a diagram showing an example of a data structure of the user voiceprint storage unit. The user voiceprint storage unit 112 includes a user ID 112A, a name 112B, a date of birth 112C, a gender 112D, and voiceprint information 112E. The user ID 112A is an identifier of the user, that is, a medical person. The name 112B is information that identifies the name of the user specified by the user ID 112A. The date of birth 112C is information that identifies the date of birth of the user specified by the user ID 112A. The gender 112D is information that identifies the gender of the user specified by the user ID 112A. The voiceprint information 112E is information that identifies the voiceprint of the user specified by the user ID 112A.

FIG. 4 is a diagram showing an example of a data structure of the belonging room storage unit. The affiliation room storage unit 113 includes the affiliation room 113A, and stores a virtual room ID that identifies the virtual room to which the smart speaker 100 belongs at a certain point in time.

The control unit 120 includes a voice recognition unit 121, a term identification unit 122, a user identification unit 123, and a room designation unit 124. The voice recognition unit 121 converts the input spoken voice into text. The voice recognition unit 121 performs various language-dependent processes such as phoneme analysis, morphological analysis, parsing, and semantic analysis, identifies the utterance content, and converts it into text. Alternatively, the voice recognition unit 121 transmits the input spoken voice information in order to delegate the voice recognition process to a predetermined external voice recognition server that can be connected via the network, and the text information of the recognized result. To get.

The term specifying unit 122 estimates the utterance that the voice recognition unit 121 could not recognize from the pronunciation information or the commonality of the terms for the unrecognized part, and recognizes the term as it is without any interpretation. For example, if there is a word that the voice recognition unit 121 fails to recognize, if it is stored in the term storage unit 111 as a term, the voice recognition unit 121 is told to treat the term as it is as a notation or meaning.

The user identification unit 123 extracts a voiceprint from the phonological characteristics of the spoken voice, and identifies the user by estimating from the degree of coincidence of the voiceprints with reference to the user voiceprint storage unit 112.

When transmitting information to the emergency information linkage server device 200, the room designation unit 124 transmits the information together with the virtual room ID as key information if it belongs to any of the virtual rooms. If the room designation unit 124 does not belong to any of the virtual rooms, the room designation unit 124 transmits the empty information as the ID of the virtual room.

The communication unit 130 communicates with the emergency information cooperation server device 200, which is another device, via the network 50.

The voice input unit 140 receives voice input from the user. The voice output unit 150 outputs voice when there is a voice command from the emergency information cooperation server device 200 to the user. The output information is various output information such as biological information and utterance history.

The smart speaker 300 has substantially the same configuration as the smart speaker 100.

The emergency information cooperation server device 200 includes a storage unit 210, a control unit 220, and a communication unit 230.

The storage unit 210 includes a symptom information storage unit 211 and an utterance history storage unit 212.

FIG. 5 is a diagram showing an example of a data structure of the symptom information storage unit. The symptom information storage unit 211 includes a patient name 211A, a patient gender 211B, a patient age 211C, a pulse 211D, a chief complaint 211E, a treatment 211F, and a patient room 211G that identify a patient to be symptomatic.

The patient name 211A is the patient's name, the patient gender 211B is the patient's gender, and the patient age 211C is the patient's age. By combining these pieces of information, it can be said that the information can identify a specific individual. The pulse 211D is a measured value of the pulse, which is one of the biometric information of the patient. Chief complaint 211E is the main symptom complained by the patient. Treatment 211F is information for explaining the treatment performed on the patient in natural language. The patient room 211G is key information for identifying the virtual room at the time of emergency dispatch of the patient.

FIG. 6 is a diagram showing an example of a data structure of the utterance history storage unit. The utterance history storage unit 212 includes a patient room 212A, a speaker 212B, an utterance time 212C, and an utterance content 212D.

The patient room 212A is key information for identifying the virtual room. Speaker 212B is information on a user ID that identifies the speaker who made a statement. Normally, an ambulance crew or a hospital medical worker whose voiceprint is registered in advance is the speaker, but the voiceprint is registered. Even if it is not, a temporary user ID may be assigned as an anonymous speaker.

The utterance time 212C is information that identifies the time when the speech was made. For example, the utterance time 212C is time information indicated in the YYYY / MM / DD hh: mm format. The utterance content 212D is a record of the utterance voice converted into text.

The control unit 220 includes a symptom information creation unit 221, a time-series conversation processing unit 222, and an output creation unit 223.

The symptom information creation unit 221 organizes information on the patient's symptom into predetermined items and stores it in the symptom information storage unit 211 using the information included in the utterance content. This symptom information creation unit 221 can be realized by a learned model constructed by performing machine learning or the like particularly related to medical terms such as medical records. However, the present invention is not limited to this, and it is possible to analyze by various arithmetic processes generally called artificial intelligence, and it is also possible to process by arithmetic processing using a rule-based engine.

The time-series conversation processing unit 222 stores the utterance content converted into text in the utterance history storage unit 212 in association with the virtual room in which the utterance was made, the speaker, and the utterance time.

The output creation unit 223 reads the utterance content converted into text for each virtual room from the utterance history storage unit 212 in association with the speaker and the utterance time, and configures output screen information of a predetermined design. It is transmitted to the smart speaker 100 or 300 or the like. However, this does not apply when the smart speaker 100 or 300 or the like is not provided with a display.

The communication unit 230 communicates with another device, the smart speaker 100 or 300, via the network 50.

The smart speaker 300 basically has the same configuration as the smart speaker 100. That is, the smart speaker 300 includes a storage unit 310, a control unit 320, a communication unit 330, an audio input unit 340, and an audio output unit 350. Further, the storage unit 310 includes a term storage unit 311, a user voiceprint storage unit 312, and an affiliation room storage unit 313, and the control unit 320 includes a voice recognition unit 321 and a term identification unit 322. , A user identification unit 323 and a room designation unit 324 are included.

FIG. 7 is a diagram showing a hardware configuration example of the smart speaker. The smart speakers 100 and 300 include a hardware configuration realized by a speaker device having so-called processing power, a personal computer, a smartphone, or a tablet terminal housing. The smart speakers 100 and 300 connect the arithmetic unit 101, the main storage device 102, the auxiliary storage device 103, the voice input device 104, the voice output device 105, the display device 106, the communication device 107, and the respective devices. It includes a bus 108. In addition, the smart speakers 100 and 300 may be provided with input / output devices such as a touch panel and a display.

The arithmetic unit 101 is, for example, an arithmetic unit such as a CPU (Central Processing Unit).

The main storage device 102 is, for example, a memory device such as a RAM (Random Access Memory).

The auxiliary storage device 103 is a non-volatile storage device such as a so-called hard disk (Hard Disk Drive), SSD (Solid State Drive), or flash memory that can store digital information.

The voice input device 104 is a sound collecting device such as a microphone. The voice output device 105 is a sounding device such as a speaker. The display device 106 is a display device such as a liquid crystal display or an organic EL (Electro-Luminence). The communication device 107 is a network interface card (NIC) or the like.

The voice recognition unit 121, the term identification unit 122, the user identification unit 123, and the room designation unit 124 of the smart speakers 100 and 300 described above are realized by a program that causes the arithmetic unit 101 to perform processing. This program is stored in the main storage device 102, the auxiliary storage device 103, or a ROM device (not shown), is loaded on the main storage device 102 for execution, and is executed by the arithmetic unit 101.

Further, the storage unit 110 of the smart speakers 100 and 300 is realized by the main storage device 102 and the auxiliary storage device 103. Further, the communication unit 130, the voice input unit 140, and the voice output unit 150 are realized by the communication device 107 voice input device 104 and the voice output device 105, respectively. The above is a hardware configuration example of the smart speakers 100 and 300.

The configurations of the smart speakers 100 and 300 can be further classified into more components depending on the processing content. It can also be categorized so that one component performs more processing.

Further, each control unit (voice recognition unit 121, term identification unit 122, user identification unit 123, room designation unit 124) is constructed by dedicated hardware (ASIC, GPU, etc.) that realizes each function. May be done. Further, the processing of each control unit may be executed by one hardware or may be executed by a plurality of hardware.

FIG. 8 is a diagram showing a hardware configuration example of the emergency information linkage server device 200. The emergency information linkage server device 200 includes a hardware configuration realized by a housing of a so-called workstation, personal computer, smartphone or tablet terminal. The emergency information cooperation server device 200 includes an arithmetic unit 201, a main storage device 202, an auxiliary storage device 203, a communication device 204, and a bus 205 connecting each device. The communication device 204 communicates with other devices, the smart speakers 100 and 300, via a network.

The arithmetic unit 201 is an arithmetic unit such as a CPU, and the main storage device 202 is a memory device such as a RAM. Further, the auxiliary storage device 203 is a non-volatile storage device such as a so-called hard disk, SSD, or flash memory that can store digital information. The communication device 204 is a network interface card (NIC) or the like. The above is a hardware configuration example of the emergency information linkage server device 200.

The configuration of the emergency information linkage server device 200 can be further classified into more components according to the processing content. It can also be categorized so that one component performs more processing.

Next, the operation of the emergency information linkage system 1 in this embodiment will be described.

FIG. 9 is a diagram showing an example of a flow of processing during emergency transportation (until the start of the room). The emergency transport processing (until the start of the room) is activated in response to a start instruction from the smart speaker 100 mounted on the emergency vehicle.

First, the voice recognition unit 121 receives the utterance (step S001). Then, the user identification unit 123 identifies the user by using the user voiceprint storage unit 112. Then, the user is transmitted to the emergency information linkage server device 200 as the speaker (step S002).

The time-series conversation processing unit 222 that has received the speaker information prepares the virtual room (step S003). Specifically, the time-series conversation processing unit 222 newly allocates a virtual room ID as key information, and notifies the smart speaker 100 of the assigned virtual room ID as a room ID.

Then, the voice recognition unit 121 recognizes the voice and converts it into text (step S004). Specifically, the voice recognition unit 121 performs voice recognition using the trained model. Alternatively, the voice recognition unit 121 transmits the input spoken voice information in order to delegate the voice recognition process to a predetermined external voice recognition server that can be connected via the network, and the text information of the recognized result. To get.

Then, the term identification unit 122 specifies the term (step S005). Specifically, the term identification unit 122 uses the trained model and the term storage unit 111 to target the text obtained in step S004 by a predetermined algorithm, and uses medical terms, abbreviations, common names, common names, dialects, and the like. The term is specified in consideration of the fluctuation of the expression of, the term is adopted as it is, and the meaning is interpreted without conversion processing. Then, the room designation unit 124 transmits the room ID and the utterance text to the emergency information cooperation server device 200.

Then, the time-series conversation processing unit 222 records the conversation in a time-series manner (step S006). Specifically, the time-series conversation processing unit 222 utters the room ID and the utterance text received in step S005 and the utterance information received in step S002 with the patient room 212A of the utterance history storage unit 212, respectively. The content 212D and the speaker 212B are stored in association with each other. Further, the time information for specifying the processing time point is stored in the utterance time 212C.

Then, the symptom information creation unit 221 creates symptom information (step S007). Specifically, the symptom information creation unit 221 analyzes the utterance text recorded in step S006 using the learned model, specifies the value for each item set in the symptom information storage unit 211, and causes the symptom. It is stored in the information storage unit 211. Specifically, the symptom information creation unit 221 extracts information such as the patient name, patient gender, patient age, pulse, chief complaint, and treatment from the utterance text and stores it in the symptom information storage unit 211, and also stores it in the patient room 211G. Store the room ID.

Then, the time-series conversation processing unit 222 invites the smart speaker 300 to the room (step S008). Specifically, the time-series conversation processing unit 222 allocates a hospital emergency room by a predetermined algorithm (for example, searches for a hospital in which the emergency department is open) to the smart speaker 300 existing in the emergency room. And send the room ID along with the invitation information.

Then, the voice recognition unit 321 of the smart speaker 300 receives the utterance of the medical staff in the emergency room (step S009).

Then, the user identification unit 323 identifies the user by using the user voiceprint storage unit 312. Then, the user is transmitted to the emergency information linkage server device 200 as the speaker (step S010). In response to this, the time-series conversation processing unit 222 of the emergency information linkage server device 200 adds a speaker to the room ID (step S011).

Then, the voice recognition unit 321 recognizes the voice and converts it into text (step S012). Specifically, the voice recognition unit 321 performs voice recognition using the trained model. Alternatively, the voice recognition unit 321 transmits the input spoken voice information in order to delegate the voice recognition process to a predetermined external voice recognition server that can be connected via the network, and the text information of the recognized result. To get.

Then, the term specifying unit 322 specifies the term (step S013). Specifically, the term identification unit 322 uses the trained model and the term storage unit 311 to target the text obtained in step S011 by a predetermined algorithm, and uses medical terms, abbreviations, common names, common names, dialects, and the like. The term is specified in consideration of the fluctuation of the expression of, the term is adopted as it is, and the meaning is interpreted without conversion processing. Then, the room designation unit 324 transmits the room ID and the utterance text to the emergency information cooperation server device 200.

Then, the time-series conversation processing unit 222 records the conversation in a time-series manner (step S014). Specifically, the time-series conversation processing unit 222 utters the room ID and the utterance text received in step S012 and the utterance information received in step S010 to the patient room 212A of the utterance history storage unit 212 and the utterance, respectively. The content 212D and the speaker 212B are stored in association with each other. Further, the time information for specifying the processing time point is stored in the utterance time 212C.

The above is an example of the flow of processing during emergency transportation (until the start of the room). According to the flow of emergency transport processing (until the room starts), the smart speaker accepts voice input, creates a virtual room on the emergency information linkage server device 200, and shares the key information of the virtual room with other smart speakers. can do. In addition, symptom information can be automatically created from the utterance content. As a result, smart speakers having shared key information can share utterance text information and patient symptom information in a virtual room.

FIG. 10 is a diagram showing an example of a flow of processing during emergency transportation (addition of symptom information). The emergency transport process (addition of symptom information) is activated after the emergency transport process (until the room starts).

First, the voice recognition unit 321 of the smart speaker 300 receives the utterance of the medical staff in the emergency room (step S101).

Then, the user identification unit 323 identifies the user by using the user voiceprint storage unit 312 (step S102).

Then, the voice recognition unit 321 recognizes the voice and converts it into text (step S103). Specifically, the voice recognition unit 321 performs voice recognition using the trained model. Alternatively, the voice recognition unit 321 transmits the input spoken voice information in order to delegate the voice recognition process to an external predetermined voice recognition server that can be connected via the network, and the text information of the recognized result. To get.

Then, the term specifying unit 322 specifies the term (step S104). Specifically, the term identification unit 322 uses the trained model and the term storage unit 311 to target the text obtained in step S103 by a predetermined algorithm, and targets medical terms, abbreviations, common names, common names, dialects, and the like. The term is specified in consideration of the fluctuation of the expression of, the term is adopted as it is, and the meaning is interpreted without conversion processing. Then, the room designation unit 324 transmits the room ID and the utterance text to the emergency information cooperation server device 200.

Then, the time-series conversation processing unit 222 records the conversation in a time-series manner (step S105). Specifically, the time-series conversation processing unit 222 uses the room ID, the utterance text, and the utterance information received in step S104 as the patient room 212A of the utterance history storage unit 212, the utterance content 212D, and the utterance. It is stored in association with the person 212B. Further, the time information for specifying the processing time point is stored in the utterance time 212C.

Then, the symptom information creation unit 221 specifies a symptom-related item (step S106). Specifically, the symptom information creation unit 221 analyzes the utterance text recorded in step S105 using the learned model, and asks a question about any of the items set in the symptom information storage unit 211. Judge whether or not. If the question is not about any of the items set in the symptom information storage unit 211, the symptom information creation unit 221 ends the flow of emergency transport processing (addition of symptom information).

When the question is about any of the items set in the symptom information storage unit 211, the utterance text is transmitted to the smart speaker 100 as a question command and as a specified symptom-related item for that item (step S107). ..

The voice recognition unit 121 of the smart speaker 100 outputs a question sentence in response to the question command (step S108).

Then, the voice recognition unit 121 of the smart speaker 100 receives the utterance of the emergency personnel of the emergency vehicle (step S109).

Then, the user identification unit 123 identifies the user by using the user voiceprint storage unit 112 (step S110).

Then, the voice recognition unit 121 recognizes the voice and converts it into text (step S111). Specifically, the voice recognition unit 121 performs voice recognition using the trained model. Alternatively, the voice recognition unit 121 transmits the input spoken voice information in order to delegate the voice recognition process to a predetermined external voice recognition server that can be connected via the network, and the text information of the recognized result. To get.

Then, the term identification unit 122 specifies the term (step S112). Specifically, the term identification unit 122 uses the trained model and the term storage unit 111 to target the text obtained in step S111 by a predetermined algorithm, and targets medical terms, abbreviations, common names, common names, dialects, and the like. The term is specified in consideration of the fluctuation of the expression of, the term is adopted as it is, and the meaning is interpreted without conversion processing. Then, the room designation unit 124 transmits the room ID, the utterance text, and the information of the speaker to the emergency information cooperation server device 200.

Then, the time-series conversation processing unit 222 records the conversation in a time-series manner (step S113). Specifically, the time-series conversation processing unit 222 uses the room ID, the utterance text, and the utterance information received in step S112 in the patient room 212A of the utterance history storage unit 212, the utterance content 212D, and the utterance. It is stored in association with the person 212B. Further, the time information for specifying the processing time point is stored in the utterance time 212C.

Then, the symptom information creation unit 221 creates symptom information (step S114). Specifically, the symptom information creation unit 221 analyzes the utterance text recorded in step S113 using the learned model, specifies the value for each item set in the symptom information storage unit 211, and causes the symptom. It is stored in the information storage unit 211. Specifically, the symptom information creation unit 221 extracts information such as patient name, patient gender, patient age, pulse, chief complaint, and treatment from the spoken text, and stores the information in the patient room 211G of the symptom information storage unit 211. It is stored as information on patients whose IDs match.

The above is an example of the flow of emergency transportation processing (addition of symptom information). According to the flow of emergency transport processing (addition of symptom information), the smart speaker accepts voice input and requests symptom information from other smart speakers that share the key information of the virtual room, and automatically starts from the utterance content. It is possible to create symptom information.

FIG. 11 is a diagram showing an example of an output screen. When the information in the utterance history storage unit 212 is updated, the output screen 400 is transmitted from the emergency information linkage server device 200 to all of the smart speakers 100 and 300 belonging to the updated virtual room. However, this does not apply to smart speakers that do not have a display device.

On the output screen 400, the icon 401 indicating the speaker, the text information 402 of the utterance content, and the utterance time 403 are displayed side by side in the time series of the utterance time 403 in the vertical direction. That is, the higher the screen, the older the utterance time is displayed. As a result, the process of utterance can be referred to using a terminal having a common virtual room ID, and it becomes possible to know the history of who said what and in what order and process. This can be objective information, for example, to confirm whether the health care worker has given appropriate instructions, or whether the ambulance crew has properly communicated the necessary information to the health care worker.

Further, on the output screen 400, important information may be highlighted or highlighted according to a predetermined standard, such as by increasing the highlight or size. The output creation unit 223 performs the selection and emphasis processing of the emphasis target. For example, the characters related to biometric information may be written in bold and the instructions from the doctor may be highlighted. In addition to displaying the text, the automatic voice reading process or the like may be used to output voice for confirmation, instruction, report, etc., not limited to the question and answer session.

Further, the output screen 400 may output patient information having the same room ID stored in the patient room 211G of the symptom information storage unit 211 according to a predetermined format. The format is often different depending on the jurisdiction of the emergency vehicle and the jurisdiction of the hospital, and various formats may be selectable.

The above is the emergency information cooperation system 1 according to the first embodiment. According to this system, necessary information can be recorded by natural dialogue without any special operation.

The present invention is not limited to the above embodiments. The above embodiments can be variously modified within the scope of the technical idea of the present invention. For example, in the above embodiment, only conversation information is shared, but the present invention is not limited to this. For example, the life-saving device mounted on the emergency vehicle (for example, AED, but not limited to this, various devices such as an electrocardiogram monitor may be used) obtains information that can specify the operating status and operates. It may indicate the situation. If you do this, if you are unconscious and ambulance transported, or if you are found to have a serious illness, what kind of treatment was taken, whether there were any deficiencies in the process, etc. Information can be confirmed objectively.

Such an example will be described with reference to FIGS. 12 to 14 as the second embodiment. The emergency information cooperation system 1 ′ according to the second embodiment is basically the same as the emergency information cooperation system 1 according to the first embodiment, but is partially different. This difference will be mainly described.

FIG. 12 is a configuration diagram of an emergency information cooperation system. In the emergency information linkage system 1', the storage unit 110'of the smart speaker 100' includes the peripheral device operation log storage unit 114. Further, the control unit 120'includes a peripheral device control unit 125 that controls the peripheral device. A peripheral device 160 such as an AED (Automated External Defibrillator) is connected to the smart speaker 100'. The smart speaker 300'has the same configuration.

FIG. 13 is a diagram showing an example of a data structure of the peripheral device operation log storage unit. The peripheral device operation log storage units 114 and 314 include a peripheral device ID 114A, a device name 114B, a time 114C, and an operation content 114D. Peripheral device ID 114A is information for identifying an individual of peripheral device 160 or 360. The device name 114B is information that identifies the model of the peripheral device 160 or 360. The time 114C is information for specifying the time when the operation event of the peripheral device 160 or 360 occurs. The operation content 114D is information for identifying an event of the peripheral device 160 or 360. That is, the operation logs of the peripheral devices 160 or 360 are accumulated in the peripheral device operation log storage unit 114 or 314.

Peripheral device control units 125 and 325 control the operation of the connected peripheral device 160 or 360. For example, the peripheral device control units 125 and 325 control various operations such as start-up start, preparation, diagnosis start, diagnosis control, diagnosis result determination, and execution. In addition, the peripheral device control units 125 and 325 automatically record the operation time, and store the records in the peripheral device operation log storage unit 114 or 314. Further, the peripheral device control units 125 and 325 are connected to the peripheral device operation log storage unit 114 or 314 at a predetermined timing to the emergency information linkage server device 200'via the room designation unit 124 while connected to the virtual room. The stored operation log is transmitted in association with the room ID.

Further, the storage unit 210'of the emergency information cooperation server device 200' includes the equipment operation information storage unit 213. The control unit 220'includes an interlocking information creation unit 224.

The equipment operation information storage unit 213 stores data having the same data structure as the peripheral equipment operation log storage unit 114 or 314. When the interlocking information creation unit 224 receives the operation log transmitted by the peripheral device control units 125 and 325, the interlocking information creation unit 224 stores each room ID in the device operation information storage unit 213.

In addition, the interlocking information creation unit 224 outputs and creates the operation log information of the equipment (peripheral equipment) associated with the room ID of the virtual room at the corresponding time zone in the process of the output creation unit 223 creating the output screen. Hand over to department 223. The output creation unit 223 configures an output screen so as to output utterances and operation log information in chronological order.

FIG. 14 is a diagram showing an example of an output screen. The output screen 400'is basically the same as the output screen 400, but is different in that the log information 404, which is the operation log of the peripheral device, is displayed. In the example of FIG. 14, as the log information 404, "12:04 emergency vehicle AED has started", "12:04 emergency vehicle AED is ready", "12:05 emergency vehicle AED". Information such as "Diagnosis has started" is displayed as text.

The above is the emergency information linkage system 1'related to the second embodiment. In the emergency information linkage system 1'according to the second embodiment, the above-mentioned output screen 400' is displayed, so that the medical staff in the emergency room surely confirms that the AED is used for the patient. be able to. In addition, if you are unconscious and are transported by ambulance, or if you are found to have a serious illness, you can check more detailed information such as when and what kind of treatment was taken, and whether there were any deficiencies in the process. Can be done objectively.

Further, for example, in the emergency information cooperation system according to the first embodiment and the second embodiment described above, the smart speaker receives the utterance and converts it into text, but the text is converted by the emergency information cooperation server device 200 ″. You may do so.

Such an example will be described with reference to FIGS. 15 to 17 as the third embodiment. The emergency information cooperation system 1 ″ according to the third embodiment is basically the same as the emergency information cooperation system 1 according to the first embodiment, but is partially different. This difference will be mainly described.

FIG. 15 is a configuration diagram of an emergency information cooperation system. In the emergency information linkage system 1 ″, the storage unit 110 ″ of the smart speaker 100 ″ includes the belonging room storage unit 113. The term storage unit 111 and the user voiceprint storage unit 112 included in the storage unit 110 according to the first embodiment are the term storage unit 215 and the term storage unit 215 in the storage unit 210 ″ of the emergency information linkage server device 200 ″. It is included as a user voiceprint storage unit 216.

Further, the control unit 120 ″ of the smart speaker 100 ″ includes a room designation unit 124. The voice recognition unit 121, the term identification unit 122, and the user identification unit 123 included in the control unit 120 according to the first embodiment are combined with the control unit 220 ″ of the emergency information cooperation server device 200 ″. , Voice recognition unit 227, term identification unit 228, and user identification unit 229.

The smart speaker 300 ″ is the same as the smart speaker 100 ″.

FIG. 16 is a diagram showing an example of a flow of emergency transport processing (until the start of the room) in the third embodiment. The emergency transport processing (until the start of the room) is activated in response to a start instruction from the smart speaker 100 ″ mounted on the emergency vehicle.

First, the voice input unit 140 of the smart speaker 100 ″ receives the utterance. Then, the control unit 120 ″ transmits the utterance voice information to the emergency information cooperation server device 200 ″ (step S001 ″).

Then, the user identification unit 229 identifies the user by using the user voiceprint storage unit 216 (step S002').

The time-series conversation processing unit 222 that identifies the speaker prepares the virtual room (step S003). Specifically, the time-series conversation processing unit 222 newly allocates a virtual room ID.

Then, the voice recognition unit 227 recognizes the voice and converts it into a text (step S004'). Specifically, the voice recognition unit 227 performs voice recognition using the trained model. Alternatively, the voice recognition unit 227 transmits the received utterance voice information in order to delegate the voice recognition process to an external predetermined voice recognition server that can be connected via the network, and transmits the text information of the recognized result. get.

Then, the term identification unit 228 identifies the term (step S005'). Specifically, the term identification unit 228 uses the trained model and the term storage unit 215 to target the text obtained in step S004'by a predetermined algorithm, and uses medical terms, abbreviations, common names, common names, and dialects. The terms are specified in consideration of the fluctuation of expressions such as, etc., the terms are adopted as they are, and the meaning is interpreted without conversion processing.

Then, the time-series conversation processing unit 222 records the conversation in a time-series manner (step S006). Specifically, the time-series conversation processing unit 222 utters the room ID assigned in step S003, the utterance text obtained in step S004', and the information of the speaker specified in step S002', respectively. The patient room 212A of the history storage unit 212, the utterance content 212D, and the speaker 212B are stored in association with each other. Further, the time information for specifying the processing time point is stored in the utterance time 212C.

Then, the symptom information creation unit 221 creates symptom information (step S007'). Specifically, the symptom information creation unit 221 analyzes the utterance text recorded in step S006'using the learned model, and specifies the value for each item set in the symptom information storage unit 211. It is stored in the symptom information storage unit 211. Specifically, the symptom information creation unit 221 extracts information such as the patient name, patient gender, patient age, pulse, chief complaint, and treatment from the utterance text and stores it in the symptom information storage unit 211, and also stores it in the patient room 211G. Store the room ID.

Then, the time-series conversation processing unit 222 invites the smart speakers 100 ″ and 300 ″ to the room (step S008 ′). Specifically, the time-series conversation processing unit 222 allocates a hospital emergency room by a predetermined algorithm (for example, searches for a hospital in which the emergency department is open), and the smart speaker 300 ″ 300 ″ existing in the emergency room. The room ID is sent together with the invitation information. Further, the time-series conversation processing unit 222 notifies the smart speaker 100 ″ of the assigned virtual room ID as the room ID.

Then, the voice input unit 140 of the smart speaker 300 ″ receives the utterance of the medical worker. Then, the utterance voice information is associated with the room ID via the room designation unit 324 and transmitted to the emergency information cooperation server device 200 ″ by the control unit 320 ″ (step S009 ″).

Then, the user identification unit 229 adds one speaker to the virtual room (step S020). Then, the user identification unit 229 identifies the user by using the user voiceprint storage unit 216. (Step S010').

Then, the voice recognition unit 227 recognizes the voice and converts it into text (step S012'). Specifically, the voice recognition unit 227 performs voice recognition using the trained model. Alternatively, the voice recognition unit 227 transmits the received utterance voice information in order to delegate the voice recognition process to an external predetermined voice recognition server that can be connected via the network, and transmits the text information of the recognized result. get.

Then, the term identification unit 228 identifies the term (step S013'). Specifically, the term identification unit 228 uses the trained model and the term storage unit 215 to target the text obtained in step S011'by a predetermined algorithm, and uses medical terms, abbreviations, common names, common names, and dialects. The terms are specified in consideration of the fluctuation of expressions such as, etc., the terms are adopted as they are, and the meaning is interpreted without conversion processing.

Then, the time-series conversation processing unit 222 records the conversation in a time-series manner (step S014). Specifically, the time-series conversation processing unit 222 transfers the room ID and the utterance text received in step S009'and the information of the speaker identified in step S010' to the patient room 212A of the utterance history storage unit 212, respectively. , The utterance content 212D and the speaker 212B are stored in association with each other. Further, the time information for specifying the processing time point is stored in the utterance time 212C.

The above is an example of the flow of emergency transportation processing (until the start of the room) according to the third embodiment. According to the flow of processing during emergency transportation (until the room starts), the smart speaker accepts voice input and converts it into text on the emergency information linkage server device, so that the processing capacity required for the smart speaker can be suppressed. Therefore, the smart speaker can be configured in a simple manner.

FIG. 17 is a diagram showing an example of a flow of emergency transport processing (addition of symptom information) according to the third embodiment. The emergency transport process (addition of symptom information) is activated after the emergency transport process (until the room starts).

First, the voice input unit 340 of the smart speaker 300 ″ receives the utterance of the medical staff in the emergency room. Then, the utterance voice information is associated with the room ID via the room designation unit 324 and transmitted to the emergency information cooperation server device 200 ″ by the control unit 320 ″ (step S101 ″).

Then, the user identification unit 229 identifies the user by using the user voiceprint storage unit 216 (step S102').

Then, the voice recognition unit 227 recognizes the voice and converts it into text (step S103'). Specifically, the voice recognition unit 227 performs voice recognition using the trained model. Alternatively, the voice recognition unit 227 transmits the received utterance voice information in order to delegate the voice recognition process to an external predetermined voice recognition server that can be connected via the network, and transmits the text information of the recognized result. get.

Then, the term identification unit 228 identifies the term (step S104'). Specifically, the term identification unit 228 uses the trained model and the term storage unit 215 to target the text obtained in step S103'by a predetermined algorithm, and uses medical terms, abbreviations, common names, common names, and dialects. The terms are specified in consideration of the fluctuation of expressions such as, etc., the terms are adopted as they are, and the meaning is interpreted without conversion processing.

Then, the time-series conversation processing unit 222 records the conversation in a time-series manner (step S105). Specifically, the time-series conversation processing unit 222 uses the room ID, the utterance text, and the speaker's information received in step S101'in the patient room 212A of the utterance history storage unit 212, the utterance content 212D, and the utterance content 212D, respectively. It is stored in association with the speaker 212B. Further, the time information for specifying the processing time point is stored in the utterance time 212C.

Then, the symptom information creation unit 221 specifies a symptom-related item (step S106). Specifically, the symptom information creation unit 221 analyzes the utterance text recorded in step S105 using the learned model, and asks a question about any of the items set in the symptom information storage unit 211. Judge whether or not. If the question is not about any of the items set in the symptom information storage unit 211, the symptom information creation unit 221 ends the flow of emergency transport processing (addition of symptom information).

If the question is about any of the items set in the symptom information storage unit 211, the utterance text is converted into a voice as a question command and as a specified symptom-related item and transmitted to the smart speaker 100 ″. (Step S107').

The smart speaker 100 ″ outputs a question command and symptom-related items by voice (step S108 ′).

Then, the voice input device of the smart speaker 100 ″ receives the utterance of the emergency personnel of the emergency vehicle. Then, the utterance voice information is associated with the room ID via the room designation unit 124 and transmitted to the emergency information cooperation server device 200 ″ (step S109 ″).

Then, the user identification unit 229 identifies the user by using the user voiceprint storage unit 216 (step S110').

Then, the voice recognition unit 227 recognizes the voice and converts it into text (step S111'). Specifically, the voice recognition unit 227 performs voice recognition using the trained model. Alternatively, the voice recognition unit 227 transmits the received utterance voice information in order to delegate the voice recognition process to an external predetermined voice recognition server that can be connected via the network, and transmits the text information of the recognized result. get.

Then, the term identification unit 228 identifies the term (step S112'). Specifically, the term identification unit 228 uses the trained model and the term storage unit 215 to target the text obtained in step S111'by a predetermined algorithm, and uses medical terms, abbreviations, common names, common names, and dialects. The terms are specified in consideration of the fluctuation of expressions such as, etc., the terms are adopted as they are, and the meaning is interpreted without conversion processing.

Then, the time-series conversation processing unit 222 records the conversation in a time-series manner (step S113). Specifically, the time-series conversation processing unit 222 uses the room ID, the utterance text, and the speaker's information received in step S109'in the patient room 212A of the utterance history storage unit 212, the utterance content 212D, and the utterance content 212D, respectively. It is stored in association with the speaker 212B. Further, the time information for specifying the processing time point is stored in the utterance time 212C.

Then, the symptom information creation unit 221 creates symptom information (step S114). Specifically, the symptom information creation unit 221 analyzes the utterance text recorded in step S113 using the learned model, specifies the value for each item set in the symptom information storage unit 211, and causes the symptom. It is stored in the information storage unit 211. Specifically, the symptom information creation unit 221 extracts information such as patient name, patient gender, patient age, pulse, chief complaint, and treatment from the spoken text, and stores the information in the patient room 211G of the symptom information storage unit 211. It is stored as information on patients whose IDs match.

The above is an example of the flow of emergency transport processing (addition of symptom information) according to the third embodiment. According to the flow of emergency transport processing (addition of symptom information), the smart speaker accepts voice input and requests symptom information from other smart speakers that share the key information of the virtual room, and automatically starts from the utterance content. It is possible to create symptom information.

Further, the technical elements of the above-described embodiment may be applied independently, or may be applied separately in a plurality of parts such as a program component and a hardware component.

Further, the trained model may be applied to all the analysis of the above-mentioned utterances, the likelihood may be estimated by calculating the likelihood according to the degree of matching of the keywords, or simply included. The corresponding item may be specified based on the rule base of the word. The present invention has been described above with a focus on embodiments.

1 ... Emergency information linkage system, 50 ... Network, 100, 300 ... Smart speaker, 110, 310 ... Storage unit, 111, 311 ... Term storage unit, 112, 312 ... Use Person voice pattern storage unit, 113,313 ... Affiliation room storage unit, 114,314 ... Peripheral device operation log storage unit, 120,320 ... Control unit, 121,321 ... Voice recognition unit, 122, 322 ... Term identification unit, 123, 323 ... User identification unit, 124,324 ... Room designation unit, 125,325 ... Peripheral device control unit, 130, 330 ... Communication unit, 140 , 340 ... Voice input unit, 150, 350 ... Voice output unit, 200 ... Emergency information linkage server device, 210 ... Storage unit, 211 ... Symptom information storage unit, 212 ... Speaking History storage unit, 213 ... Equipment operation information storage unit, 215 ... Term storage unit, 216 ... User voice print storage unit, 220 ... Control unit, 221 ... Symptom information creation unit, 222. .. Time-series conversation processing unit, 223 ... Output creation unit, 227 ... Voice recognition unit, 228 ... Term identification unit, 229 ... User identification unit, 230 ... Communication unit.

Claims (20)

Multiple voice input / output devices that accept spoken voice,
A symptom information storage unit that stores medical record information including the patient's biological information,
A voice recognition unit that converts the input utterance voice into text,
A symptom information creation unit that extracts at least the biological information of the patient from the text and stores it in the symptom information storage unit.
An output creation unit that transmits and outputs the biometric information to the audio input / output device having key information associated with each other.
An information linkage system characterized by having. With multiple audio input / output devices
The voice input / output device is provided with an information linkage server device that is communicably connected via a network.
Each of the voice input / output devices includes a control unit that converts the input spoken voice into text and transmits it to the information linkage server device together with predetermined key information.
The information linkage server device is
A symptom information storage unit that stores information including the patient's biological information,
A symptom information creation unit that extracts at least the biological information of the patient from the text and stores it in the symptom information storage unit.
An output creation unit that transmits and outputs the biometric information to the voice input / output device having the key information associated with each other.
An information linkage system characterized by having. The information linkage system according to claim 1 or 2.
At least one of the voice input / output devices is provided in an emergency vehicle.
The patient is a patient housed in the emergency vehicle,
An information linkage system characterized by this. The information linkage system according to any one of claims 1 to 3.
At least one of the voice input / output devices is installed in an emergency room where medical staff perform medical treatment.
The audio input / output device includes at least a display device that displays the biological information.
An information linkage system characterized by this. The information linkage system according to any one of claims 1 to 4.
The control unit of the voice input / output device performs a voice recognition process for recognizing the spoken voice in the process of converting the spoken voice into text, and in the process, the predetermined terms related to the biological information of the patient are approximated. Do not convert to words,
An information linkage system characterized by this. The information linkage system according to claim 2.
The information linkage server device is
A time-series conversation processing unit is provided, which transmits the text to the voice input / output device having the key information common to each other and shares the text in a time series.
When one of the voice input / output devices makes an utterance about a new patient, the key information is issued to associate the key information with at least one of other voice input / output devices different from the voice input / output device. ,
An information linkage system characterized by this. The information linkage system according to any one of claims 1 to 6.
The output creation unit creates screen information for displaying the text in time series together with the biometric information, and outputs the screen information to the voice input / output device having the key information associated with each other.
An information linkage system characterized by this. The information linkage system according to any one of claims 1 to 7.
Equipped with a log collection unit that collects log information related to the operating status of a given device
The output creation unit generates information in which the operating state and the time of the utterance are arranged in chronological order.
An information linkage system characterized by this. A communication unit that receives textualized voice information and key information input from multiple voice input / output devices via a network.
A symptom information storage unit that stores information including the patient's biological information,
A symptom information creation unit that extracts at least the biological information of the patient from the text and stores it in the symptom information storage unit.
An output creation unit that transmits and outputs the biometric information to the voice input / output device having the key information associated with each other.
An information linkage server device characterized by having. The information linkage server device according to claim 9.
At least one of the voice input / output devices is provided in an emergency vehicle.
The patient is a patient housed in the emergency vehicle,
An information linkage server device characterized by this. The information linkage server device according to claim 9 or 10.
At least one of the voice input / output devices is installed in an emergency room where medical staff perform medical treatment.
The audio input / output device includes at least a display device that displays the biological information.
An information linkage server device characterized by this. The information linkage server device according to any one of claims 9 to 11.
A time-series conversation processing unit is provided, which transmits the text to the voice input / output device having the key information associated with each other and shares the text in a time series.
When one of the voice input / output devices makes an utterance about a new patient, the key information is issued to associate the key information with at least one of other voice input / output devices different from the voice input / output device. ,
An information linkage server device characterized by this. The information linkage server device according to any one of claims 9 to 12.
The output creation unit creates screen information for displaying the text in time series together with the biometric information, and transmits the screen information to the audio input / output device having the key information associated with each other for output.
An information linkage server device characterized by this. A communication unit that receives input voice information and key information from multiple voice input / output devices via a network.
A symptom information storage unit that stores information including the patient's biological information,
A voice recognition unit that converts the spoken voice information into text,
A symptom information creation unit that extracts at least the biological information of the patient from the text and stores it in the symptom information storage unit.
An output creation unit that transmits and outputs the biometric information to the voice input / output device having the key information associated with each other.
An information linkage server device characterized by having. With multiple audio input / output devices
The voice input / output device is provided with an information linkage server device that is communicably connected via a network.
Each of the voice input / output devices includes a control unit that receives the input voice and transmits it to the information linkage server device together with predetermined key information.
The information linkage server device is
A symptom information storage unit that stores information including the patient's biological information,
A voice recognition unit that receives the spoken voice and converts it into text,
A symptom information creation unit that extracts at least the biological information of the patient from the text and stores it in the symptom information storage unit.
An output creation unit that transmits and outputs the biometric information to the voice input / output device having the key information associated with each other.
An information linkage system characterized by having. A communication unit that connects to a predetermined information linkage server device so that it can communicate via a network,
It is equipped with a control unit that converts the input utterance voice into text and transmits it to the information linkage server device together with predetermined key information.
When the control unit receives the biometric information from the information linkage server device, the control unit outputs the biometric information.
An audio input / output device characterized by this. The audio input / output device according to claim 16.
A display device for displaying at least the biometric information is provided.
An audio input / output device characterized by this. The audio input / output device according to claim 16 or 17.
The control unit performs a voice recognition process for recognizing the uttered voice in the process of converting the uttered voice into text, and does not convert a predetermined word related to biological information into an approximate word in the process.
An audio input / output device characterized by this. It is an information linkage method using an information linkage system.
The information linkage system is
Multiple voice input / output devices that accept spoken voice,
It is equipped with a symptom information storage unit that stores information including biological information of the patient and a control unit.
The control unit
A voice recognition step that converts the input utterance voice into text,
A symptom information creation step of extracting at least the biological information of the patient from the text and storing it in the symptom information storage unit.
An output creation step of transmitting and outputting the biometric information to the voice input / output device having key information associated with each other.
An information linkage method characterized by carrying out. A program that lets a computer link information
Information including a communication unit, a control unit, and a patient's biological information for receiving textualized information and key information of spoken voice input from a plurality of voice input / output devices on the computer via a network. To function as a symptom information storage unit to memorize
In the control unit
A symptom information creation step of extracting at least the biological information of the patient from the text and storing it in the symptom information storage unit.
An output creation step of transmitting and outputting the biometric information to the voice input / output device having the key information associated with each other.
A program characterized by having the program implemented.

Images