JP4005089B2 - Communication recording system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a communication recording system which increases the evidential value of a recorded content, making always calls and always recordings possible at all times. <P>SOLUTION: A head set is provided with a microphone 17 to detect a sound of a wearer of the head set and to generate a first sound signal, and a transmission means 11 to transmit the first sound signal to equipment 70 by a short-range radio communication which complies with the Bluetooth (R) Standard. The equipment 70 is provided with a receiving means 11 to receive the first sound signal from the head set by the short-range radio communication which complies with the Bluetooth standard, a sound input means 87 to detect a sound of other person other than the wearer of the head set and to generate a second sound signal, a time information obtaining means 33 to obtain a generated time of the first and the second sound signals, an information making means 85 to make the information to be stored, and a storing means 81 to store the information prepared. <P>COPYRIGHT: (C)2005,JPO&amp;NCIPI

Description

  The present invention relates to a communication recording system that realizes confidentiality of communication contents and simplification of search / playback while enabling continuous communication and continuous recording by connecting ubiquitous devices by short-range wireless communication.

  A short-range wireless communication system called Bluetooth has been developed, and it is now possible to connect and control devices such as PCs, printers, mobile phones, and digital cameras in the room or office without using a cord. It has become.

  Under such circumstances, it has been proposed to use a headset with a voice input function in order to promote the application of the system to a mobile environment and further enhance the convenience and smoothness of communication.

  Currently, voice communication using a mobile phone, mail and chat using a mobile terminal are frequently performed as wireless communication. However, in the case of a mobile phone, it is necessary to identify and connect to the other party each time communication is performed. In addition, the conversations exchanged are limited on the spot, and if the mobile phone is turned off, the conversation contents up to that point will be lost. On the other hand, in the case of mail and chat by wireless communication, the communication content is stored as characters, but all communication content must be manually input when used.

  Therefore, there is a demand for a system capable of recording and storing the contents of wireless communication by voice in real time and effectively using it later. Up to now, the recording of voice has been usually performed only by using a recording device such as a tape recorder and the direct sound propagating as a sound wave within a necessary range. The operation of continuous recording is costly and laborious and has not been realized except for special cases such as observing human behavior for academic purposes. In recent years, in the field of wearable computers, body-mounted computers and their application research have been conducted, but how to process various types of information (environmental information and biological information) always sensed, It is still in the research stage.

  In addition, a technology has been proposed for recording sound around a personal computer so that the surrounding sound can be recorded continuously. However, this technology is only a first-person technology, and it records direct sound propagating from the outside as sound waves. To do. The recorded contents are first stored in a memory in the PC, and the detected sound cannot be transmitted to a third party in real time.

  In addition, there is a conference record system that records the content of a conference, and reproduces and retrieves it at a later date, but in a fixed situation, a discussion on a predetermined issue is only recorded for a certain period of time.

  Under such circumstances, a system is proposed in which a user wearing a headset with a wireless communication function can communicate and record digitized audio in real time, and use the recorded data meaningfully at a later date. . In the patent application filed on the same day by the same applicant as this case, a user wearing a headset with wireless communication function controls the device with voice commands in real time, or communicates and records voice data such as dictation in real time Has been proposed.

  Since such a headset system is wireless, the user can perform communication, recording, and device control in a free situation without being restricted by actions. In addition, since the user's voice is collected by a close-talking microphone, digitized, transmitted, and recorded, clear voice with little noise superposition can be communicated and stored. The recorded data has little deterioration and is suitable for storage, copying, direct texting, searching, etc.

  However, at the same time as these merits, it is necessary to deal with new problems such as unauthorized reproduction, unauthorized use, disclosure, and falsification. In particular, when the above-described headset system is introduced to a highly confidential conference or medical place, privacy protection and confidentiality of recorded data must be ensured.

  Further, since the recorded data is not misused or the headset itself is not illegally used, a headset system with high security is required.

  Further, in order to effectively use the recorded data at a later date, a technique capable of simplifying search and reproduction while ensuring the confidentiality of the recorded contents is required.

  In addition, when the recorded contents are used as evidence at a later date, it is preferable to have evidence value comparable to that of literature or the like.

  Furthermore, assuming constant communication and constant recording, it is naturally expected that there are cases where recording is not desired, recording is not desired, or recording is performed but playback is not desired, depending on the situation. . Therefore, a technique for suppressing recording and reproduction in such a case is also required.

  In addition, depending on the mental state of the headset wearer, it is necessary to consider emotional factors such as wanting to hear, not wanting to hear, may ask, not wanting to listen to the other, or wanting to hear.

  Accordingly, a first object of the present invention is to provide a headset with a wireless communication function (hereinafter referred to as a case where the recorded content can be enhanced while enhancing the evidence value of the recorded content while allowing continuous communication and recording in any free situation. And a communication recording system using the same.

  A second object of the present invention is to provide a headset with a short-range wireless communication function that can reliably protect the privacy of conversation contents by setting various conditions for recording and playback processing. There is.

  A third object of the present invention is to provide a headset with a wireless function capable of easily, reliably searching, reproducing and using recorded information, and a communication recording system using the headset.

  A fourth object of the present invention is to provide a headset with a wireless function in which security is ensured so that recorded data is not illegally used or the headset itself is not illegally used, and a communication recording system using the headset. It is to provide.

  The fifth object of the present invention is to consider human emotional aspects such as wanting to listen to communication contents, not wanting to hear it, listening to it, or not wanting the other party to hear it, or wanting to hear it, on the premise of constant communication. However, it is an object of the present invention to provide a headset system that can control communication, recording, and reproduction according to the state of the wearer of the headset.

  In a first aspect of the present invention, a communication recording system including one headset and one device is provided. This system includes a headset with a wireless communication function and a device capable of wireless communication with the headset. The headset includes a microphone that detects a wearer's voice and generates a first voice signal, and a transmission unit that transmits the first voice signal to the device by wireless communication. The device includes a receiving unit that receives a first audio signal from the headset, a voice input unit that detects a voice other than a person wearing the headset and generates a second audio signal, and first and second Time information acquisition means for acquiring the generation time of the audio signal, an information generation section for generating information to be stored by associating the time information with the first and second audio signals, and storage means for storing the generated information With.

  In this system, a user wearing a headset can interact with a person having voice input means such as a wireless microphone. The voice input means is connected to a mobile terminal such as a stationary Bluetooth device or a PDA. This system is suitable for interview-type communication and recording. Since there is sufficient capacity on the device side, the content of the conversation can be recorded in real time together with the time information without hindering the actions of the parties involved in the conversation. Since it has time information stored in relation to the recorded contents, the evidence ability of the recording is improved.

  In a second aspect of the present invention, a communication recording system comprising two headsets and one device is provided. That is, the system includes a first headset with a wireless communication function, a second headset capable of wireless communication with the first headset, and a device capable of transmitting and receiving wirelessly with the first and second headsets. Including. Each of the first and second headsets detects a wearer's voice and generates a first voice signal, and transmits the first voice signal to the other headset via the device. Transmission / reception means for receiving the second audio signal transmitted from the headset via the device. The device includes a time information acquisition unit that acquires the generation times of the first and second audio signals, and an information generation unit that generates information to be stored in association with the first and second audio signals and the time information. And storage means for storing the created information.

  In the communication recording system described in the first and second aspects, the time information acquisition unit, the information creation unit, and the storage unit are provided in a device other than the headset, so that the structure of the headset itself is simplified and the size is reduced. It is possible to reduce the weight. Since the conversation contents between the headsets or between the headset and the microphone attached to the device are stored in the device in association with the time when the headset is performed, the recording evidence capability is enhanced. External devices usually have sufficient capacity, so they are suitable for long-term recording.

  Furthermore, by providing the device with a voice / text conversion function, the conversation content can be stored as characters or reproduced as characters.

  Preferably, the device includes a digital watermark adding unit that adds a digital watermark (watermark) when transmitting the first audio signal to the other headset. When the digital watermark is added, the content received by the headset is recorded (recorded) as it is, but the digital watermark is detected during reproduction. Therefore, fraudulent acts such as later modification or forgery of records are effectively prevented. As a result, the reliability of the system security is improved.

  It is possible to increase the proof value of the recorded contents while allowing continuous calls and recordings in any free situation.

  By setting various conditions for recording and playback processing, the privacy of the conversation content is reliably protected.

  The recorded information can be searched, played back and used easily and reliably while maintaining confidentiality.

  It is possible to effectively prevent unauthorized use of recorded data and unauthorized use of the headset itself.

  Communication, recording, and playback can be controlled in accordance with the state of the wearer of the headset while taking into account human emotional aspects.

  Embodiments of the present invention will be described below with reference to the drawings.

(First embodiment)
FIG. 1 shows an appearance of a headset with a wireless communication function according to the first embodiment of the present invention. FIG. 1A shows a binaural type headset 10, and FIG. 1B shows a single ear type headset 20. Both types of headsets transmit a microphone 17 that detects a wearer's (user's) voice and generates a first voice signal, and transmits the first voice signal by short-range wireless communication. The wireless communication module 11 that receives the second audio signal returned from another headset wearer in response to the audio signal by the short-range wireless communication, the CPU board 13, and outputs the received audio and the reproduced audio A speaker 15 is provided.

  The headset also has a function selection switch 19 that can switch and select between various modes to dictate. The ear pads 23 are connected by a flexible frame and supported by the user's head. An arm 18 extends from one ear pad 23 and a microphone 17 is attached to the tip. The microphone 17 is located almost at the user's mouth when the user wears the headset 10 (or 20), and detects sound with little superposition of ambient noise.

  A speaker 15, a CPU board 13, a wireless communication module 11, and a battery 21 are built in the earpiece 23. Although not shown, each element is connected by a cable or the like as necessary.

  Such headsets 10 and 20 with a wireless communication function perform short-range wireless communication, for example, wireless communication based on the Bluetooth standard. In the case of Bluetooth, it is possible to transmit data at 1 Mbit per second using radio waves in the 2.4 GHz band. The connection of data between the same Bluetooth devices is performed automatically, so it is possible to connect the devices with Bluetooth function by simply placing them nearby without having to perform the “connect” operation. There is. In addition, since an authentication function is provided, there is no possibility of network connection with an unknown device. A headset using Bluetooth can transmit and receive an audio signal having a sampling frequency such as 8 kHz, 16 kHz, 22.05 kHz, or 44.1 kHz depending on settings. However, the present invention is not limited to the Bluetooth standard headset, and intends a headset having a wireless communication function.

  FIG. 2 shows a system configuration of a communication recording system including the headset shown in FIG. This system is based on the premise that basically recording is always performed, that is, recording is automatically performed while a user wears a headset and communicates. Communication may be started by turning on the power switch when the headset is attached, or may be automatically started when the headset is attached by a temperature sensor or the like.

  The CPU board 13 decodes the audio detector 36 that converts the first audio signal generated by the microphone 17 into a digital signal, and the second audio signal that is transmitted from the second headset 10B and received by the wireless communication module 11. Decoding means 39 for performing, and clock means 33 for obtaining the generation times of the first and second audio signals. The clock means 33 uses, for example, a standard clock function.

  The first audio signal digitized by the audio detector 36 and the received and combined second audio signal are input to the information creating unit 35. The information creation unit 35 associates the time information obtained by the clock means 33 with the first and second audio signals, and creates record information. At this time, the information creation unit 35 extracts keywords from the first and second audio signals. Although not shown in FIG. 2, the headset further has a keyword storage unit, and keywords corresponding to the use of the headset are stored in advance. The information creation unit 35 refers to the keyword storage unit, and extracts keywords when they are included in the first or second audio signal. By extracting the keyword, when the voice recorded later is reproduced, the search is facilitated by inputting the keyword by voice or text.

  The storage means 31 stores the first and second audio signals associated with the time information. In the example of FIG. 2, since the dialogue content exchanged between the headsets 10A and 10B is stored in the storage means 31 as a digital audio signal, it can sufficiently cope with long-time recording. In addition, the audio data may be compressed and recorded using an appropriate compression technique. The storage means 31 may be built in the headset or may be removable.

  The second audio signal from the headset 10 </ b> B is also input to the audio player 37. The received second audio signal is converted into an analog signal by the audio player 37 and output from the speaker 15. Therefore, the user wearing the headset 10A has a conversation with the user of the headset 10B via wireless communication, and at the same time, the conversation content is recorded in the memory (storage means) 31 in the headset together with the time information ( Recorded).

  The recorded content can be reproduced at a later date, for example, by operating the function selection switch of FIG. Further, when the storage means 31 is a removable memory card (for example, an SD card), it can be played back by another device. Since the conversation content is recorded in association with the time information (including the date), the search at the time of reproduction becomes easy, and the evidence ability as a record is high. In addition, since the keyword during the dialogue is extracted, the search can be facilitated by inputting the keyword by voice, for example.

  Although not shown, an encryption / decryption unit may be installed on the CPU board 13. In this case, it is possible to further enhance the reliability of the storage and use of recorded information and to protect privacy.

  When the first audio signal is transmitted from the headsets 10A to 10B, the signal digitally converted by the audio detector 36 is encoded by the encoding means 38 and transmitted wirelessly from the wireless communication module 11. The converting means 38 and the decoding means 39 are usually built in the wireless communication module 11.

  In the example of FIG. 2, only the headset 10A of the two headsets has the clock means 33, the information creation unit 35, and the storage means 31, but at least one of them only needs to have these components. Of course, both headsets 10A, 10B may have these elements. The clock means 33 may be a radio wave clock using a standard radio wave with time information as described later. In this case, a standard radio wave receiving antenna is required separately. When a radio clock is used, certainty of time is ensured, and unauthorized use of recorded contents due to illegally changing the time is prevented.

  FIG. 3 is a modification of the communication recording system using the headset of FIG. In the system of FIG. 3, both the headsets 40A and 40B with wireless communication function have clock means, and compare and adjust either one of the two times to improve the reliability of the time information.

  The second audio signal generated by the microphone 17 of the second headset 40B is encoded together with the second time information obtained by the clock means 33 of the second headset 40B, and wirelessly transmitted to the first headset 40A. The Also in the first headset 40A, the first time information is uniquely obtained by the clock means 33. The first time information and the second time information are input to the time adjustment unit 45 of the first headset 40A. The clock adjustment unit 45 compares the first time information and the second time information, and if the same or within the allowable error range, supplies the first and second time information as correct information to the record creation unit 35. . If the two pieces of time information do not match, there is a method of urging correction by, for example, an alarm.

  In the example of FIG. 3, only two headsets are shown. However, a communication recording system using a headset with a clock adjustment function is particularly effective when discussing by a large number of people such as a conference. In this case, the clock adjustment unit 45 compares the time information input from all the headsets, and when most of the time information matches, supplies it to the record creation unit 35 as appropriate time information. . Thereby, the evidence ability of the recorded contents is further improved.

  FIG. 4 shows yet another variation of the headset and system shown in FIG.

In addition to the clock means 33, the wireless function headset 50A includes a position specifying unit 47 that uses GPS, for example, to acquire position information. GPS is a system developed by the United States for navigation support of aircraft, ships, and the like. This system consists of 24 GPS satellites (4 each on 6 orbital planes) orbiting about 20,000 km above, a control station for tracking and controlling GPS satellites, and a user receiver for positioning. It is configured. Similar to aircraft, ships, etc., this BTHS determines its own position etc. by knowing the distances from four or more GPS satellites simultaneously. The distance from the GPS satellite is obtained from the time required for the radio wave transmitted from the GPS satellite to reach the receiver.

  The position information obtained by the position specifying unit 47 is input to the information creating unit 35 together with the time information. The information creating unit 35 associates the time information and the position information with the first voice of the user detected by the headset 50A and the second voice of another user transmitted from the headset 50B and stores the information. Create The storage means 31 stores an audio signal to which time information and position information are attached.

  In such a configuration, since the time and place can be specified together with the content of the conversation, the ability to prove the conversation content is further improved.

  The communication recording system of the first embodiment is effective for group inspections and report creation where the range of action of each person is widened. It is also expected to be applied to medical education and group education in the field. For example, in the case of home care, if both the dispatched caregiver and the patient wear headsets and take care records, the caregiver's momentum is large, so the patient wears a headset that allows free use of both hands. It is conceivable to automatically record conversations with the. Since time information is added to the recorded content, the evidence as a record is high and the reliability of care is improved. In addition, the capacity of evidence can be further increased by adding a number of headsets and interposing a third person such as a nurse or an attendant.

(Second Embodiment)
FIG. 5 shows a communication recording system using a headset with a wireless function according to the second embodiment of the present invention. In the second embodiment, a system is configured by combining a headset with a wireless function and an external device. In the example of FIG. 5, a communication recording system is configured by one wireless function headset 60 and one external device 70. The device 70 is, for example, a fixed device with a BT function (hereinafter referred to as “BT device”).

  The headset 60 detects the voice of the wearer (user) and generates a first audio signal, and wirelessly transmits the first audio signal to the BT device 70 and is transmitted from the BT device. The communication module 11 that receives the audio signal and the speaker 15 that outputs the second audio signal as audio.

  The BT device 70 has a microphone 77 as an audio input unit. In the example of FIG. 5, the microphone 77 is connected to the BT device 70 by a cable, but may be a wireless microphone. In the case of a wireless microphone, the user's freedom of action increases and convenience increases.

  The BT device 70 also includes a communication module 11 that performs wireless transmission and reception with the headset, a sound detection unit 87 that digitally converts the second sound signal generated by the microphone 77, and a clock unit 83. The first audio signal received from the headset and the second audio signal generated by the microphone 77 are input to the information creation unit 85. The time information acquired by the clock means 33 and the device ID of the headset are also input to the information creation unit 85. The information creation unit 85 creates information to be stored by associating the time information and the device ID with the first and second audio signals. The created information is stored in the storage means 81.

  The clock means 83 is a radio clock using standard radio waves as shown in FIG. 6, for example. The radio timepiece receives a standard radio wave carrying time information with a built-in ultra-high performance antenna (radio wave receiving means) 71, and a time extraction unit 72 extracts and corrects the time. Standard radio waves are transmitted from the “Otakadoya Yama standard radio wave transmitter” located near the summit of Mt. The atomic clock at the transmitting station is linked by remote control with the Japan Standard Time created by the atomic clock in the Japan Radio Standards Group, Japan Standard Time Group (Koganei City, Tokyo).

  The time extraction unit amplifies and decodes the time signal received by the antenna 71. By using the radio wave clock as the clock means 83, the clock function of the BT device 70 itself and the headset 60 can be corrected as necessary. In addition, since the accurate time is supplied to the information creating unit 85, the ability to prove the recorded content is improved. Further, as described in the first embodiment, when communicating and recording between a plurality of headsets, the clock means of any one of the headsets may be a radio clock. In this case, time information of other headsets can be corrected and unified to an accurate time. As a result, time management is accurately performed in the entire system, and the reliability (proof ability) of recorded contents is further improved.

  Returning to FIG. 5, by making the BT device 70 stationary, the position of the headset 60 is automatically specified and recorded by connection with the BT device 70. That is, although not shown in the drawing, in addition to time information, position information is also input to the information creation unit 85, and a recording record with higher evidence capability is created.

  Further, the BT device 70 may be a mobile terminal such as a PDA instead of the stationary type. In this case, position information can be acquired by providing the mobile terminal with a GPS position specifying unit.

  Further, the headset 60 may be provided with a storage unit so that a record created on the BT device 70 side is wirelessly received and stored in a memory in the headset 60.

  The headset system shown in FIG. 5 can be effectively applied to, for example, medical and educational fields. Considering the medical case, the patient is given a B headset along with an insurance card, or a personal headset is lent at the reception when he goes to the hospital. In a case where the headset has a memory and can record conversation contents at all times, it is desirable that the patient and the headset have a one-to-one correspondence. When there is no memory on the headset side and conversation contents are stored in the BT device on the hospital side, it is possible to lend the headset to any patient.

  The doctor explains the current symptoms and precautions to the patient wearing the unique headset 60 using, for example, the microphone 77. The microphone 77 is used to answer the patient's questions and to ask the patient questions. These communication contents are automatically stored in the storage means 81 together with the device ID and time / position information of the patient's headset. When the patient and the headset are not in a one-to-one correspondence, the patient ID may be generated from the patient's voiceprint, stored in advance, and the patient ID based on the voiceprint may be used instead of the device ID for each dialogue.

  When this system is used, the doctor can obtain proof of the fact that a correct diagnosis has been made by the recorded recording as long as the explanation and answer are given correctly.

Such a recorded record can be used as an implementation record for medical malpractice lawsuits and informed consent. Also, by using this system, the doctor's response can be improved and the patient can be trusted. In addition, an insurance company can introduce a business that lowers the insurance premium rate of compensation insurance against medical malpractice lawsuits for hospitals that employ such a communication recording system.

  On the other hand, the merit for the patient is that the recorded contents can be reproduced and confirmed at a later date even if important matters such as explanations of symptom from the doctor and precautions are forgotten. Since time information and position information are attached to the recorded contents, it is easy to search. Even when an unspecified number of lending headsets are used instead of the patient-specific headsets, the recorded contents recorded by the stationary BT device in the hospital can be stored in an MD or the like and passed to the patient. In addition, there is an advantage that the waiting situation and the call can be confirmed via the headset. Since information on how many people are currently being examined can be obtained, the waiting time can be used effectively. Even if you are away from the waiting room, you can call it through the headset. Further, in the waiting room, for example, it is possible to listen to favorite music provided by many BT devices installed in the hospital.

  In the application to education, it is possible to give each student a unique headset at the same time as admission. In this application example, conversation contents such as education guidance, career guidance, and two-party interview can be stored as digital data in association with the headset ID of the student and time / position information. The record contents thus stored have high evidence ability because the time and place, or the conversation partner (student) is specified.

  FIG. 5 shows an example in which the patient wears a headset and the doctor uses a microphone. However, the doctor wears the headset, passes the microphone to the patient to be visited, and the conversation content is transferred to the BT device 70. It is good also as a structure to record. Since the configuration of the headset itself is simplified, the headset can be reduced in size and weight. On the other hand, since the BT device has a sufficient capacity, various records can be stored.

  If there is sufficient capacity on the BT device side, a camera function can be added to one or both of the headset and the BT device. In this case, the conversation content and video information exchanged by wireless communication are stored in association with the time / position information.

  FIG. 7 shows a modification of the system configuration of FIG. The communication recording system shown in FIG. 7 includes two headsets with a wireless function and one external device.

Specific examples include a first headset 90A, a second headset 90B, and a BT device 110 based on, for example, the Bluetooth standard, which can transmit and receive with these headsets by wireless communication. The first and second headsets 90 </ b> A and 90 </ b> B respectively detect a voice spoken by the wearer (user) and generate a first audio signal via the microphone 97, and the first audio signal via the device 110. Transmitting and receiving units 108 and 109 for receiving the second audio signal transmitted from the second headset via the device 110. The device 110 includes a clock unit 113 that acquires the first and second sound generation times, an information generation unit 115 that generates information to be stored in association with the first and second sound signals, Storage means 111 for storing the created information.

  In the system of FIG. 7, the second headset 97 and the BT device 110 are used instead of the BT device 70 with the attached microphone 77 used in the system of FIG. 5. Thus, for example, in the medical case, both doctors and patients wear headsets. The device 110 may be a stationary type or a mobile type. The conversation exchanged between the two headsets is stored in the storage unit 11 with time information attached by the information creation unit 115. The system of FIG. 5 is suitable for interview-type dialog recording, but the system of FIG. 7 expands the number of wearers and the action range. For example, in the rehabilitation field, in addition to the conversation exchanged between the physical therapist and the patient, all the words of the doctor giving instructions in the separate room are suitable for real-time recording on the stationary device 110. When the recorded content is also recorded in association with the device ID of the headset, the conversation partner can be specified in addition to the time and place, and the proof ability of the recorded content is further enhanced.

  Moreover, in the place of care, a caregiver and a patient can wear a headset, and conversation records can be stored in real time by wireless communication in a device 110 stationary at the patient's home instead of the memory in the headset. . Or the caregiver brings a mobile terminal and records it in the mobile terminal. In the case of a stationary device, as described above, the location information is automatically obtained when the headset is connected to the device. Further, as in the system of FIG. 5, a camera function may be added to any of the headsets. In this case, audio and video may be recorded using a compression technique as appropriate.

  For example, when a caregiver helps a caregiver to change clothes, the change of clothes can be recorded as a video together with a voice saying "I will change Yamada-san from now on". When searching for “changing clothes” as a keyword at a later date, for example, in addition to the voice recognition result around “changing clothes” that is the search result, it is possible to reproduce the sound and video during recording.

(Third embodiment)
FIG. 8 shows an outline of a communication recording system according to the third embodiment of the present invention. This system is obtained by adding a text conversion function to the system of the second embodiment. In FIG. 8, a system configuration composed of two headsets and one device shown in FIG. 7 is adopted, but as shown in FIG. 5, a device with one headset and one voice input means is used. It may be a system configuration.

  In the communication recording system of FIG. 8, the device 120 receives the first and second sounds wirelessly transmitted from the first and second headsets 90A and 90B, and transmits the received sounds to the other headset. Transmitters 118 and 119 are included. The received first and second sounds are sent to the other headset and simultaneously input to the information creation unit 115 of the device 120. The first and second sounds are stored in the storage unit 111 in association with the time information obtained by the clock unit 113.

  The device 120 further includes a voice recognition unit 121 and a text information storage unit 123. The voice recognition unit 121 converts the voice data created by the information creation unit 115 or the voice data stored in the storage unit 111 into text (character string). The text information storage unit 123 stores the text-converted information. Therefore, in the communication recording system of FIG. 8, in the device 120, the conversation content is recorded as voice and at the same time is recorded as character information in real time.

  The device 120 also has a keyword recognizing unit 125, and extracts a keyword from the recorded voice and the recorded text. When a keyword is input by voice or text from the keyword input unit 127 at a later date, the keyword recognition unit 125 searches the storage unit 111 and the text information storage unit 123 when the keyword is an extracted keyword, Specify the desired recording position easily and quickly. The voice or text at the specified position is reproduced and displayed on the display screen of the device 120 or output from the speaker.

  FIG. 9 is a schematic block diagram of the speech recognition unit 121 of FIG. The voice analysis unit 132 and the feature extraction unit 133 create feature parameters from the input digital voice data. The input digital audio data is digitally converted with a predetermined sampling frequency and quantization bits. The voice analysis unit 132 performs 512-point FFT analysis with a frame length of 24 ms (Hamming window) and a frame shift of 8 ms, and the feature extraction unit 133 calculates 32 channel BPF group output parameters x (t, f), f = 1. create.

  The voice quantizing unit 543 outputs a voice segment time series by quantizing a time-frequency pattern in a range of 48 ms while shifting by 8 ms. In the third embodiment, 235 types of speech segments are defined. The speech segment time series can be expressed as a series of word HMMs composed of phoneme HMMs. The HMM here is a left-to-right type SMQ / HMM, and triphones classified in consideration of the phoneme context are used for the phoneme HMM. The acoustic model 136 is an HMM dictionary for examining the acoustic similarity of input speech. On the other hand, the language model 137 is a dictionary for examining similarity as a language, and stores information on the ease of connecting consecutive words using word double chains (bigram) and word triple chains (trigram). . In addition to the acoustic model 136 and the language model 137, the text conversion unit 135 converts the acoustic parameter into a Japanese text candidate series using a word dictionary 138 that stores 105,000 words.

  The keyword extraction unit 139 extracts, for example, nouns and proper nouns as keywords for search based on the part of speech. For example, if a caregiver wearing a headset 90A utters “closes to Yamada” from the close-up microphone 97, this voice is automatically recognized, “Yamada”, “Change of clothes”. Keywords such as are extracted.

  The extracted keyword is sent to the keyword recognizing unit 125, and the keyword is registered as voice data as necessary. By using the keyword extracted in this way, the search speed is increased. Items related to nursing care requests such as “changing clothes” are automatically registered and described in the invoice. Of course, the caregiver checks the final confirmation. On the other hand, when you want to reconfirm the situation when you have changed your clothes, you can search by entering “Change of clothes” by voice or text from the keyword input section 127 and search for the phrase “Change of clothes”. Will be immediately searched and played. The keyword recognition unit 125 searches the storage unit 111 and the text information storage unit 123 according to the method in which the keyword is input, that is, character input or voice input.

  FIG. 10 shows an example of a character string stored in the text information storage unit. The conversation content exchanged by the caregiver wearing the headset 90A (device ID 400001) and the cared person wearing the headset B (device ID 80001) is converted into text and stored together with the time information. The waveform corresponding to the text string converted into text is a waveform detected as a sound wave by the microphone 97 of the headset 90 and is not stored in the text information storage unit, but for convenience of showing the correspondence, FIG. Is shown.

In the system shown in FIGS. 8 to 9, a voice recognition unit is used at the time of recording, and the inputted voice is also stored as characters. However, the recording may be stored only with the voice data, and the voice recognition unit may be operated during reproduction to reproduce the character string.
In such a communication recording system, the conversation content exchanged by wireless communication is recorded in real time as voice data and at the same time as characters, so that the utility value as a record is high.

  In addition, the search is facilitated by extracting the keywords.

(Fourth embodiment)
11 and 12 are schematic diagrams of a headset with a wireless communication function according to the fourth embodiment of the present invention and a communication recording system using the headset. In the fourth embodiment, a description will be given of a configuration in which, when recording conversation contents, recording is performed only when there is an agreement or permission for recording. FIG. 11 shows an example in which the person to be recorded (user wearing the headset 140B) has the right to determine recording permission. In FIG. 12, the person to record (user wearing the headset 160A) has the right to determine recording permission. Here is an example.

  In the present invention, it is assumed that the communication content (interaction content) is automatically recorded when the headset is worn and communication is started. However, depending on the situation and the content of the interaction, recording may be undesirable. It is good if there is a record agreement in advance between the users who have interacted with each other, but if this is not the case, for example, a privacy violation problem will arise.

  One way to obtain a record agreement is to consider it as agreed by wearing a specific headset. For example, if the patient is wearing a headset, it is a method that pre-determines that there is no objection even if the voice that was spoken and input at that time is recorded. In this case, in the example shown in FIG. 11, the recording display unit of the doctor's headset 140A is activated when the patient wears the headset 140B. The patient sees this, and if he / she agrees with the record, he / she can just interact through the headset. If you do not agree, ask them to stop recording.

  The second method is a method in which the recording (recording) side obtains consent to record in advance for a specific partner, place, and time and registers them. FIG. 11 and FIG. 12 show a system configuration for realizing the second method.

  Taking medical application as an example, in FIG. 11, the headset 140 </ b> B worn by the patient has a permission information storage unit 153 and a recording permission determination unit 151. The permission information storage unit 153 stores, for example, recording permission information illustrated in FIG. It is assumed that recording permission information can be appropriately written into the permission information storage unit 153 via an input unit (not shown). The recording permission determination unit 151 refers to the permission information recording unit to determine whether or not the recording of the conversation content should be permitted, and automatically wirelessly transmits the determination result to the doctor's headset 140A.

  For example, in the example of FIG. 13A, a condition for permitting recording only for the conversation with the doctor B is set in the patient's headset 140B regardless of time and place. Here, an asterisk “***” indicates that there is no particular condition. In order to specify that the other party is a doctor B, the connection status with the headset of the doctor B may be used. That is, when the patient's headset 140B starts communication with the headset 140A whose device ID worn by the doctor B is 100001, it is automatically determined that recording of the communication content is permitted, and the determination result is the doctor's head. Wirelessly transmitted to the set 140A. Upon receiving the recording permission determination, the doctor's headset 140A automatically starts recording the conversation content. At this time, the patient-side headset 140B displays that recording is permitted, for example, with a green lamp. On the other hand, the doctor's headset 140A displays that recording is in progress.

  When the same patient is diagnosed by a doctor C other than the doctor B, the recording permission determination unit 151 refers to the permission information storage unit 153, and recording of this dialogue is permitted based on the device ID of the doctor C's headset. Automatically determine that it should not. In this case, the permitting display unit 146 of the patient's headset 140B does not operate, for example, the lamp does not light up. The determination result is wirelessly transmitted to the C doctor's headset, and the C doctor's headset does not record the interaction with the patient.

  As shown in FIG. 13 (b), it is possible to register so as to permit recording of conversations with all physicians in hospital A regardless of time and place. In this case as well, based on the connection with the device ID of the headset each worn by the medical doctor, it is automatically determined whether or not recording is permitted, the determination result is sent to the doctor's headset, and a permission indication is displayed. The display state of the unit 146 is controlled.

  Similarly, in the example of FIG. 13C, registration is performed so as to permit recording only when entering the medical examination room of Hospital A, regardless of the time or the conversation partner. When a stationary BT device (FIG. 8) as described in the second embodiment is installed in the medical examination room of Hospital A, the location information is obtained by connecting this device and the patient's headset 140B. It is automatically detected, and the recording permission determination unit determines recording permission. This determination result is wirelessly communicated to the doctor's headset 140A, while the permission display 146 of the patient's headset 140B is lit.

  In the example of FIG. 13D, recording is permitted when entering Hospital A. In this case, when connection is established with a stationary BT device installed in Hospital A or two or more headsets worn by hospital personnel, location information is detected and recording permission is determined. Is issued.

  In the example of FIG. 13 (e), a condition for permitting recording of dialogue contents for three months from December 20 is set in the medical examination room of hospital A. Further, as shown in FIG. 13 (f), the period until the end of treatment is added to the time condition under the AND condition between December 20 and 3 months, or from December 20 to 3 as shown in FIG. 13 (g). The period until the end of treatment can be added to the time condition under the OR condition for a month.

  Thus, by recording based on consent, the credibility and evidence ability of the recorded content are significantly increased.

  Of course, it is possible to permit recording of a specific act, for example by law, without confirming consent or disagreement. In this case, the recording display unit 144 of the doctor's headset 140A is lit red, for example, but when the patient speaks without concern, the content of the conversation is recorded.

  In the example of FIG. 12, the recording-side headset 160A includes a permission information storage unit 173 and a recording permission determination unit. That is, the headset 160A transmits the voice signal to the second headset 160B through short-range wireless communication, and the microphone 167 that detects the wearer's voice and generates a first voice signal. With reference to the transmission / reception means 178 and 179 for receiving the second audio signal returned from the set 160B, the permission information storage means 173 for storing the conditions for permitting recording, and the permission information storage means 173, the first and first A recording permission determination unit 171 that determines whether or not to record the second audio signal, and information to be stored based on the first and second audio signals when the recording permission determination unit 171 permits the recording And an information creation unit 175 that performs the processing.

  In the system shown in FIG. 12, instead of obtaining the permission of the conversation partner, the recording side is permitted only in a specific situation (specific partner, place, time, etc.) on the recording side.

For example, when a group in a project interacts with members, there are many topics about the project, so recording is automatically allowed when talking to members. Also in this case, the permission information shown in FIG. 13 is stored in advance as in the case of FIG.

  In the headset with the wireless communication function according to the third embodiment and the communication recording system using the headset, it is assumed that the communication is always performed and the recording is always performed. Determine whether recording is allowed. This protects the privacy of the interaction between the headsets and improves the convenience of the system.

(Fifth embodiment)
FIG. 14 shows a headset with a wireless communication function according to a fourth embodiment of the present invention and a communication recording system using the headset. In the third embodiment, conditions for permitting recording such as a conversation partner, a place, and time are set in advance, and it is automatically determined whether or not to record conversation contents. The fourth embodiment provides a configuration in which an off-record instruction is set with the intention of the wearer (user) of the headset when content that should be concealed suddenly occurs during the conversation.

  For example, when multiple doctors are wearing conversation headsets with wireless communication function headsets that are supposed to record at all times, it is necessary to exchange information regarding patient privacy. is there. Therefore, the headset 160C according to the fourth embodiment further includes an off-record setting unit in addition to the basic configuration shown in the first embodiment. In the example shown in FIG. 14, the off-record setting means is realized by an off-record switch 172 provided in the headset 160C. When a topic requiring confidentiality occurs, the user of the headset 160C presses the off-record switch 172 to speak in the off-record mode. The off-record mode continues until the off-record switch is pressed again to cancel the mode.

  The information creation unit 175 creates information to be stored in association with the audio data, with the off-record mode instruction and the off-record mode start time and end time. For the off-record mode instruction, for example, off-record tags are assigned before and after the section in the audio data where off-record is instructed. In the off-record mode, the sound is recorded as it is, but since the off-record tag is added, the off-record tag is detected during reproduction. When an off-the-record tag is detected, it imposes severe playback restrictions such as prohibiting playback or not playing without the permission of the party that interacted.

  In the example of FIG. 14, the off-record switch 172 is provided only for the first headset 160C having a recording (sound recording) function, but a similar off-record switch may be provided for the second headset 160D. In this case, the second user who is the conversation partner can also request the setting of the off-record instruction. The off-record instruction request by the second user is input to the information creation unit 175 of the first headset 160C via wireless communication, and an off-record tag is added to the audio data. The off-record tag is preferably assigned in association with the device ID of the user who has set off-record support. Such an off-record mode setting basically does not require the consent of the other party and can be set by the intention of the speaker.

  In the example of FIG. 14, the off-record setting means is the off-record switch 172 provided in the headset, but it is also possible to instruct the off-record mode by a voice command. In this case, the headset 160C further includes the voice recognition unit (FIG. 8) shown in the third embodiment, and registers commands such as “off record”, “off record from here”, and “off record release” as standard commands. Keep it. After the headset wearer utters a specific keyword for shifting to the voice recognition mode and shifts to the voice recognition mode, when the above-described off-record command is uttered, the off-record command is recognized and the corresponding processing (off-record start, Off-record recording is executed. Also in this case, it is desirable that the off-record setting is performed in association with the device ID of the user's headset that has issued the off-record instruction or the voice print of the user.

  When the system of FIG. 14 is provided with a voice recognition unit, voice recognition processing is performed by a headset. However, as shown in FIG. 8, voice recognition may be performed by a stationary BT device. In any case, the voice uttered is recognized, and off record mode setting / canceling processing is executed based on the result.

  According to the headset with the wireless communication function and the communication recording system using the same according to the fifth embodiment, when confidential information that is not expected in advance is unexpectedly conversed, the headset wearer intends to It becomes possible to. This protects the privacy of the conversation and improves the reliability of the headset and the overall communication recording system. Moreover, since the off-record mode can be set by the user's intention, it is excellent in convenience.

(Sixth embodiment)
FIG. 15 shows a headset with a wireless communication function according to the sixth embodiment of the present invention and a communication recording system using the headset.

  The sixth embodiment provides a headset and a system that can cope with a situation in which recording is performed but playback is to be suppressed while assuming a headset for continuous recording.

  As shown in FIG. 15, the headset 180 </ b> A with a wireless communication function has a reproduction determination function in addition to the basic configuration shown in FIG. 2 in association with the first embodiment. That is, the reproduction permission information storage unit stores in advance the conditions under which reproduction of the audio data stored in the storage unit 191 is permitted. When there is a reproduction request, the reproduction determination unit 203 refers to the reproduction permission information storage unit 201 and determines whether or not audio data should be reproduced. If reproduction is permitted as a result of the determination, the audio data stored in the storage unit 191 is supplied to the audio player 197. As described in the first embodiment, when the audio data is encrypted, the data is decrypted by the decrypting means (not shown) and then supplied to the audio player 197.

  FIG. 16 shows an example of reproduction permission information, and FIG. 17 shows a mode in which recording is performed and the reproduction is performed. In FIG. 16, a playback permission location, permission contents, a person in charge of playback, and the like are set as playback permission conditions. For example, in FIG. 16A, no limitation is imposed on any of the conditions. The reproduction determination unit 203 permits reproduction only when these conditions are met. As a playback mode, the content recorded with the headset may be played back with the same headset (playback method 1 in FIG. 17B), or in the case of a removable memory card, the content recorded with the headset. May be reproduced by a device such as a PC (reproduction method 2 in FIG. 17B). There is also a mode in which recording itself is performed in advance on a PC and played back on the PC (playback method 2).

  FIG. 17 (a) shows an example in which the conversation contents are recorded in the device in the system using the two headsets described in the second embodiment and one external device as the recording mode in FIG. However, a configuration in which recording is performed on one of two headsets, a configuration in which a system is configured with one headset and a device with one voice input function, and recording on the device is naturally possible.

  FIG. 18 shows the reproduction permission determination when the person in charge of reproduction is set as shown in FIGS. 16 (b) to 16 (d). The reproduction permission condition storage unit 201 stores a reproduction person in charge ID designated as a permission condition. The information creating unit 195 associates the person in charge of reproduction, which is a reproduction permission condition, with the audio information and stores it in the storage unit 191.

  When there is a reproduction request, the reproduction determination unit 203 authenticates whether the person who made the reproduction request and the authorized person in charge of reproduction are the same. As shown in FIG. 16B, when the person in charge of reproduction is “doctor in hospital A”, the reproduction person in charge input unit 205 may be prompted to input the doctor ID in hospital A. As shown in FIGS. 16 (c) and 16 (d), when “Doctor B” and “Conversation partner” are designated as individuals, personal authentication is performed using a password and biometric information (voice print, face, retina, etc.). Also good.

  As a result of the determination, if it is determined that they are the same person, the audio data is supplied from the storage unit 191 to the audio player 197 and output from the speaker 185.

  As shown in FIG. 16 (e), when a reproduction permission place is designated, the place can be specified by connection with a fixed BT device provided, for example. Further, when the headset is based on the Bluetooth standard, it is possible to determine whether to permit reproduction by using a Bluetooth positioning function, otherwise, by using a position specifying function (see FIG. 4) incorporated in the headset.

  FIG. 19 shows a playback permission determination in the case where it is set as a condition that only a specific device is used for playback in relation to the specification of the playback location. The reproduction permission information storage unit 201 stores a reproduction device ID specified as a permission condition. The information creating unit 195 associates the playback device information, which is a playback permission condition, with the audio information, and stores it in the storage unit 191. The reproduction determination unit 203 determines whether or not the reproduction-permitted device information associated with the audio information stored in the storage unit is the same as the device ID used for reproduction. If they are the same, the audio data stored in the storage unit 191 is supplied to the audio player 197 and output from the speaker 185.

  FIG. 20 shows a reproduction permission determination when dialogue content, for example, “diagnosis content” is set as the reproduction permission condition, as shown in FIG. The information creation unit 195 refers to the keyword stored in the reproduction permission information storage unit 201 and creates information in which the keyword information is associated with the voice data. The storage unit 191 stores voice data together with designated keyword information.

  If the recorded content is set as a condition for permitting reproduction, the speech recognition unit 221 performs speech recognition processing on the speech data stored in the storage unit 191 as shown in FIG. As described above, keywords are extracted. In this case, the playback determination unit 203 permits playback of audio in a certain section including the keyword when the extracted keyword matches the specified playback content condition.

  For example, if the keyword “change of clothes” in nursing care is set in the reproduction permission condition, the section determination means 223 can be searched by searching for “change of clothes” as a key in order to reconfirm the situation when changing clothes. Then, a phrase “Yamada-san will be changed from now on” is specified as a certain section including this keyword. The rule synthesizing means 225 reproduces the sound of the specified section and outputs it through the speaker 185.

  When recording, if a speaker intentionally inputs a keyword that specifies the content, for example, words such as “From here on the medical condition explanation”, “End of medical condition explanation”, the keyword can be easily extracted and specified. Only the sound of the section corresponding to the permitted content can be reproduced correctly. Depending on the speaker, if a keyword is intentionally input and a correct reproduction is performed later, it can be used more effectively.

  FIG. 22 shows an example in which the recorded content is reproduced with synthesized sound as another example of the reproduction restriction.

  The amount of information transmitted is greatly different between a recorded voice and an artificially synthesized voice from text. For example, the voice of the boss and the voice of a colleague are exactly the same in terms of language, but the psychological influence on the listener is different. In the case of performing reproduction, if the reproduction condition specified by the reproduction determination unit 203 is met, as described above, in addition to the method of reproducing the sound as it is sent to the sound reproducer 197 and outputting from the speaker 185, It is preferable to provide means for synthesizing and outputting speech.

  In this case, only the linguistic information is extracted by recognizing the voice data that is the subject of the reproduction request by the voice recognition unit 211 and converting it into text. Next, this text is supplied to the synthesis determination unit 213 and further to the rule synthesis unit 215, and re-converted into a different voice by rule synthesis. The synthesized voice is output from the speaker 185. With this configuration, it is possible to output a voice that excludes emotions included in the real voice and personality. For example, it is possible to use a situation where emotional remarks at a meeting or the like are heard later and judgment is made, and it is considered that a calm judgment cannot be made by listening to the real voice. On the other hand, in order to realistically reproduce the personality and the situation on the spot, it may be better to listen to the recorded voice as it is.

  A configuration using such rule composition means may be used in combination with the above-described reproduction permission information. For example, when the utterance content is designated as the reproduction permission information, as shown in FIG. 21, the text of a certain section including the keyword extracted by the keyword spotting is sent to the rule synthesizing means, and the synthesized sound is generated and output. You can also Note that whether to play with synthesized sound or to play with real voice is determined inside the playback determination means, but this instruction is either the instruction by the speaker at the time of recording or the instruction of the side to play and listen May be based on

  However, if any information that can be reproduced only by the real voice is given to the specified voice section of the recorded voice by the speaker, the request may be made even if the listener specifies to listen to the synthesized voice. It is also possible to perform control such that the voice is rejected and the voice is forcibly reproduced. In any case, it is important to have a mechanism that allows you to listen to both synthetic sounds and real voices, and the operation can be used with various modifications.

  Further, since the recorded information is a record of voices of a plurality of speakers, for example, a caregiver and a cared person, it is possible to adopt a method that is limited to a case where both are agreed to reproduce. Further, in the case of doctors and patients, doctors authorized by law or lawyers, administrative officers, etc. may be allowed to restore without consent.

  FIG. 23 shows, as another example of the playback restriction, a configuration in which playback of a certain portion is prohibited instead of allowing playback under a certain condition. In this configuration example, the headset includes a voice recognition unit 221, a reproduction suppression determination unit 233, and a reproduction suppression unit 231 in addition to the reproduction determination unit. The person in charge of reproduction, the location, the content of reproduction, and the like listed as the reproduction permission conditions in FIG. 16 can be used as the reproduction prohibition conditions in FIG.

  In the example of FIG. 23, an example is shown in which a reproduction person in charge is designated to prohibit reproduction, and a case where certain keywords are prohibited from being reproduced. When the person in charge of reproduction is prohibited, the information creation unit 195 creates storage information by associating the information on the person in charge of reproduction prohibited with the audio data. When reproduction of a certain keyword is prohibited, stored information is created by associating the restricted data with audio data.

  When the person in charge of reproduction is prohibited, the reproduction determination unit 203 determines whether the information on the person in charge of reproduction prohibited from the audio data stored in the storage unit 191 is the same as the person in charge of reproduction. Judge whether or not. In the case of being the same, the audio data is not reproduced by the reproduction inhibiting means 231. That is, even if the person in charge who is prohibited from reproducing makes a reproduction request, the sound is not reproduced.

  For a predetermined keyword, first, the voice stored in the storage unit 191 is recognized by the voice recognition unit 221, and the reproduction suppression determination unit 233 determines whether the recognized content includes a restricted keyword. When the restriction keyword is included, reproduction is inhibited by the reproduction inhibition unit 231. More specifically, the playback sound is controlled so as not to be heard by converting the playback sound with an interfering sound or skipping the restriction keyword itself. Similarly, when a reproduction prohibited place is set, if the condition is met, control is performed so that the audio portion corresponding to the prohibited place is not reproduced.

  Furthermore, as described in the fifth embodiment, when an off-record tag is assigned to a predetermined section, the playback determination unit 203 detects the off-record tag and suppresses playback for the section specified by the tag. The reproduction determination unit 203 refers to the speaker information (user information) recorded by the information creation unit 195 linked to the voice. If it is determined that the permission of the person who has set the off-record mode is required, it may be a serious problem if disclosed to a third party such as privacy issues, confidential matters, etc., so the authentication unit (not shown) Authenticate with That is, in order to collate a person (headset wearer) who wants to play the off-recording section and a person recorded as the off-record mode setter, for example, prompt the playback requester to enter a password, or Encourage utterance and perform authentication using voice information. When the reproduction of the off-recording section is not permitted, a process such as outputting white noise or a beep sound for the section designated by the off-record tag or skipping this section is performed.

  Although not shown, an authentication unit is provided in the headset to authenticate a supervisor who can reproduce the off-record portion without permission from the person when it is determined that the recorded content is requested as evidence in a trial or the like. Functions can also be set. The supervisor corresponds to, for example, a judge who is socially recognized and can make a valid judgment as a third party.

  In the headset with the wireless communication function of the sixth embodiment and the communication storage system using the headset, the reproduction of the conversation contents that are not reproduced is suppressed, so that the privacy of the conversation contents is protected.

(Seventh embodiment)
24 and 25 show a communication recording system according to the seventh embodiment of the present invention. The seventh embodiment provides a configuration that can improve the security of the headset itself and the entire communication recording system. In the example of FIG. 24, a configuration in which an electronic watermark called a watermark is added to an audio signal transmitted and received between headsets, and in FIG. 25, it is authenticated that the wearer of the headset is an appropriate user. The configuration is shown.

  In the communication system of FIG. 24, an electronic watermark is used as a countermeasure when the sound reproduced and output from the speaker is recorded in another device and used illegally. Headsets 240 </ b> A and 240 </ b> B interact through device 250. For example, the first audio signal generated by the microphone 242 of the headset 240 </ b> A is first wirelessly transmitted from the encoding / transmission means 248 to the device 240. The device 250 supplies the received first audio signal to the information creating unit 255 and also supplies it to the digital watermark adding unit 265. The digital watermark adding unit 265 adds a digital watermark to the first audio signal, and wirelessly transmits the first audio signal from the encoding / transmission unit 268 to the second headset 240B.

  Similarly, the second headset 240B generates a second audio signal from the voice uttered by the user in response to the content of the first audio signal received and output from the speaker 243, and the second audio signal is generated. The signal is wirelessly transmitted from the encoding / transmission means 148 to the device 250.

  The device 250 supplies the received second audio signal to the information creation unit 255 and also supplies it to the digital watermark adding unit 265 to add the digital watermark. The second audio signal to which the digital watermark is added is transmitted to the first headset 240A via wireless communication.

  The audio signal to which the electronic watermark is added is recorded on another recording device after being output from a speaker, for example. However, when trying to reproduce it, a digital watermark is always detected. Therefore, even if it is recorded without permission by a malicious headset user and the contents are abused, if the watermark is detected from the recorded voice, the voice information is obtained through an unauthorized route. Determined. For example, it is possible to determine whether or not the information is obtained by an unauthorized operation by letting a person who misuses the recording contents as evidence material in a trial to submit the original recording voice and confirm the presence or absence of the watermark.

  In the example of FIG. 24, the system is composed of two headsets 240B and the BT device 250. However, as shown in FIG. 8, a BT device with one headset and one voice input means. Needless to say, digital watermarking is also applied when the system is configured as described above.

  The headset 270A with a wireless communication function in FIG. 25 includes an authentication unit 293 to authenticate that the wearer of the headset is an appropriate user. In the case of communication between headsets, it is necessary to illuminate the communication partner (device) that the information is transmitted by an authorized user of the headset. Since speech is used on a daily basis and does not burden the user, authentication systems based on speaker authentication have been known for a long time. However, in conventional speaker authentication systems, (1) a speaker is registered in advance. (2) Since it is vulnerable to changes over time in the voice quality of the speaker, it is often necessary to re-register the speech. Otherwise, it is necessary to reduce the verification accuracy to facilitate authentication (3 ) There was a problem that it was necessary to set or wear a microphone each time it was registered.

  For such a problem, in the headset 270A with the wireless communication function of FIG. 25, the storage means 281 stores the audio signal generated by the microphone 277 of the headset 270A in addition to the conversation contents to be stored. Is updated every time. The authentication unit 293 compares the voice signal generated by the microphone 277 this time with the updated latest voice signal stored in the storage unit 281, and the voice collected this time is continuously collected for a certain period of time in the past. It is determined whether or not the voice of the speaker is the same. If they match, the match result is output to the encoding / transmission means 288 together with the device ID of the headset 270A. The encoding / transmission means 288 wirelessly transmits an audio signal from the headset 270A to the second headset 240B together with the authentication result. The second headset 240 can confirm that the call originates from a legitimate user of the headset 270A by referring to the authentication result.

  According to such a configuration, even if a third party improperly uses the headset 270A, if a voice is not input for a certain time or more, it is not determined that the user is a valid user. In this case, the first sound is transmitted to the second headset 270B without being authenticated. By determining the presence or absence of authentication by the receiving means of the second headset 270B, it is determined whether or not the headset 270A is being used illegally.

  Even if a legitimate user of the headset 270A has voice modulation due to a cold or the like, the recorded voice feature is updated each time the user utters, so as long as the voice is input to the microphone 277 for a certain period of time. Is judged correctly.

  Since the headset 270 basically assumes constant communication and constant recording, it is always worn on the user's head, and the burden on the user for authentication is very small. By providing such an authentication function, the security of the headset itself and of the communication recording system using the headset is remarkably improved.

(Eighth embodiment)
FIG. 26 is a diagram showing a voice communication control system according to the eighth embodiment of the present invention, and FIG. 27 is a diagram of a headset with a wireless communication function and a communication recording system to which such a voice communication control system is applied. It is.

  In the eighth embodiment, the transmitting side and the receiving side take into account the psychological state of the transmitting side that they want to hear, can listen to, or do not want to hear, and the psychological state of the receiving side that they want to hear, want to hear, or do not want to hear. A configuration for determining a control method of voice communication by looking at a combination of states on the side is provided.

  That is, as shown in FIG. 26, the user's state that he / she wants to hear is sent to the voice communication control method determination unit 321 via the voice communication state input unit (selection unit) 30 of the headset with wireless communication function. Entered. Environment information, headset user relationship information, communication control method instruction information, and the like are also input to the control method determination unit 321.

  The control method determination unit 321 determines a control method for communication performed between the headsets with reference to the state selected by the user and various types of information. The voice communication execution unit 323 executes the presence / absence of voice input / output, volume / sound quality level control, recording / playback control, and the like based on the determined control method.

  Among the above-mentioned information, the environmental information is information on the environment (state) of the unintended user such as the place where the user wearing the headset is acting, the time, the activity state, the temperature, the working state, and the busyness. It is. For example, when the user of the headset is a doctor and a patient, referring to the environmental information, recording and reproduction of voice information is permitted when in the hospital. In addition, it is determined that members of a project in an organization can record and play back only when the pro-eject members are together, and record and play back if they are inside the company / organization they belong to. . By taking such environmental information into consideration, two birds with one stone can achieve the enhancement of human memory function and the enhancement of privacy and security functions.

  Headset user relationship information refers to the human relationship between headset users, such as doctors and patients, doctors and nurses, teachers and students, supervisors and subordinates of the organization, police officers and suspects, dealers and customers, It is human relationship information such as parents and children. This information is used to appropriately control privacy and security controls according to human relationships. In this way, by considering human and socio-cultural aspects, it becomes possible for IT technology to cope with new values, security, safety, PL issues, and the like.

  The control instruction information is instruction information related to specific communication, recording, and the like. For example, at a certain time, the patient's heart is monitored by wireless communication, the value of an instrument is checked, and those values are captured.

  When these pieces of information are used, for example, when the wearer of the headset is a doctor and a patient, the doctor's advice and instructions for medication can be given based on the time and monitor information. As described above, more detailed communication control can be performed by using at least one of the user relationship information, the environment information, and the control instruction information together with the communication state of the headset or independently.

  FIG. 27 is a schematic configuration diagram of a headset system using headsets 300A and 300B with a wireless communication function to which the communication control method of FIG. 26 is applied.

  This headset system includes a first headset 300A with a wireless communication function and a second headset 300A capable of wireless communication with the first headset.

Each of the first and second headsets detects a wearer's voice and generates a first voice signal, and transmits the first voice signal to the other headset and sends it from the other headset. Transmission / reception means 318 and 319 for receiving the second audio signal, and a state selection unit 301 for selecting a desired communication state by the wearer.

  The audio signal detected by the microphone 307 of the first headset 300 </ b> A is input to the audio communication execution unit 323. The audio signal transmitted by the headset 300 </ b> B is also input to the audio communication execution unit 323. On the other hand, the user of headset 300 </ b> A inputs his / her current desired communication state via state selection (input) unit 301. This first state is input to the control method determination unit 321. The user of the headset 300B similarly selects a state, and this second state is input to the control method determination unit 321 via wireless transmission. The control method determination unit 321 determines a control method for voice communication based on these input states.

  The control method determination unit 321 is also input with the environment information, the user relationship information, and the control instruction information as described above, and determines a control method in consideration of some or all of them as necessary.

  For example, in the case of reproduction, the voice communication execution unit 323 converts the sound volume and sound quality to a predetermined volume and then outputs the sound to the voice player 317. When recording, the information is output to the information creation unit 315.

  FIG. 28 is an example of a voice communication state table storing various states. The feature of the headset with short-range wireless function provided by the present invention is not only to make a voice call like a telephone without a code, but also to communication of voice and audio signals (communication) and various dialogs using voice ( Interaction) is possible between humans and information devices and between humans and humans while acting anytime, anywhere and with anyone. It is an ideal ubiquitous and wearable device and the most familiar information device for humans. New value is created by fusing such a headset with speech recognition / understanding, speech synthesis, and speech dialogue technology as described in the above-described embodiment.

  However, at the cost of the convenience, information that increases due to the widespread use of the Internet and intranets, as well as digital broadcasting, databases, and mails will become more flooded, and the problem of information flooding will become more serious.

  Therefore, in the communication and interaction between humans and human / machine systems, the situation at the time of information acquisition that you want to hear (I want to know), I can hear (you can know), or I don't want to hear (I don't want to know) Or, a short-range wireless headset system that takes into account human emotions and cognitive aspects that you want to hear (want to know), you can hear (you can inform), or you do not want to hear (don't want to hear) To introduce. In particular, auditory (audio) information and spoken language information are different in nature from other sensory information such as text, vision, and touch, and play an important role in a short-range wireless headset system.

  The headsets 300A and 300B store the states “I want to hear”, “I want to hear”, and “I don't want to hear” at the time of transmission, together with identifiers t1, t2, and t3, respectively. Furthermore, “I want to hear”, “I can hear”, and “I don't want to hear” are stored together with identifiers r1, r2, and r3 as states when receiving. The control method decision unit refers to the state selected by the user from among these, and also refers to environmental information, etc., if necessary, forcing voice message (PUSH) transmission to strengthen privacy protection To determine the actual operation.

FIG. 29 is a diagram for explaining the state table in more detail. For example, it is assumed that the state of the headset A is “I want to hear” or “I don't want to hear”, and the state of the headset B is “I can hear” or “I can hear”. In this case, according to FIG. 29A, the sound of the headset B is output from the communication execution unit, but the sound of the headset A is not output.
Although not shown, states such as “I want to record”, “I don't want to record”, “I want to record”, “I don't want to play” can be stored. By selecting such a state, for example, when you do not want to hear the other party's story, even in a situation where it is difficult to say "I do not want to listen" to the other person in the face-to-face conversation, just select the state with the headset And you can set the situation that you do not hear without being noticed by the other party. In other words, when the headset with the wireless communication function of the eighth embodiment is used, new voice communication having an effect different from that of direct dialogue between humans becomes possible.

  In each headset, voice communication state information is prepared in advance at the time of transmission / reception. For example, when using a normal (default) headset, it is set in a state where free communication and recording (recording) such as “I want to hear” and “I want to hear” are possible. In this state, the human communication range is expanded and the storage capacity of the headset is enhanced. Conversely, in situations where privacy and security protection are important, the headset's normal state (at the time of default) is a state where the audio listening level is low, not heard at all, distorted, or disabled for recording or playback. Etc. can be set.

  As an application example of such a headset system, for example, when an information weak person such as an elderly person wears the headset A, on the receiving side, priority is given to listening (want to know). Also, under the conditions for reproduction, reproduction by the elderly person and his / her family (guardian) is possible when voices related to diagnosis and diagnosis of a doctor who is in contact with the elderly person as a patient are recorded. In this case, it is convenient for the family to use the headset. In addition, as control instruction information, information that is to be recorded and played back to the elderly person is specified in advance, and the necessary notes in daily life must be played back by specifying the time, place, meal time, etc. Is also possible. In addition to audio information recorded for the headset, text and video information can be effectively provided together with appropriate timing (control instructions) and environmental conditions. Furthermore, control of voice communication from a doctor to an elderly patient's family that can be told to the family and guardians can be easily realized. It goes without saying that the security and copyright protection system described in the above embodiment or the conventional communication control technology can be used as appropriate.

  In the eighth embodiment, an example in which voice communication is performed between two users has been described. However, even when there are three or more users, the voice communication state table and the voice communication control method can be easily modified. With such a headset system, it is possible to pursue and realize an ideal way of human-to-human voice communication. It also qualitatively changes the new face-to-face communication that allows each other's faces to be seen.

  As described above, according to the headset and system with a wireless communication function of the present invention, the conversation, intention, and intention between the users of the headset are shared while maintaining privacy. For example, in the case of doctors and teachers, information related to the privacy of patients and students is also included in the conversation. Or, the status of medical practice in the examination room or operating room, which is a closed space where doctors take the initiative, and the implementation status of education in classrooms where teachers take the initiative in relation to students are recorded. , Parents can share information with peace of mind. This enables effective cooperation with home treatment and home education. For doctors, nurses and teachers. As the openness increases, the burden in the positive sense increases, but the effect obtained is great.

  Information recorded by the communication recording system or the headset system according to the present invention has high value as evidence. For example, in the event of a medical accident or serious bullying problem, reprocessing at a third party or an assessment / institution will be possible if consent is obtained from the hospital director, school principal, court, or parties. Is also possible as appropriate.

  The system of the present invention also has a privacy and security mechanism, and it is a great advantage that the user can have an off-the-record conversation with peace of mind.

Unlike actual virtual-na Internet and intranet emails and bulletin board discussions, which are conscious of audiences and records, real-world “places” such as meeting rooms, eating and drinking, talking, cars, airplanes, and smoking areas You can take advantage of the relaxed and transient voice conversation. In other words, recording and playback are possible if the other party's approval is obtained, playback is possible only when the other party is present, possible if the other party is allowed, playback is possible only at the same location as recording, etc. This creates new effects such as memory assistance, respect for originality, and comfortable and comfortable conversation.

FIG. 1 is an external view of a headset with a wireless function according to the first embodiment of the present invention. FIG. 2 is a schematic diagram of a headset with a wireless function and a communication recording system using the same according to the first embodiment of the present invention. It is a modification of the headset and communication recording system shown in FIG. It is another modification of the headset and communication recording system which are shown in FIG. It is the schematic of the communication recording system which concerns on 2nd Embodiment of this invention. FIG. 6 is a schematic block diagram of a radio timepiece used in the system of FIG. 5. It is a modification of the communication recording system which concerns on 2nd Embodiment. It is the schematic of the communication recording system which concerns on 3rd Embodiment of this invention. It is a block diagram which shows schematic structure of the speech recognition part used with the system of FIG. An example of the text information recorded by the system of FIG. 8 is shown. It is the schematic of the headset with a wireless communication function which concerns on 4th Embodiment of this invention, and a communication recording system using the same. It is a modification of the system shown in FIG. It is a figure which shows the example of recording permission information used with the system of FIG. It is the schematic of the headset with a radio | wireless communication function which concerns on 5th Embodiment of this invention, and a communication recording system using the same. It is the schematic of the headset with a wireless communication function which concerns on 6th Embodiment of this invention, and a communication recording system using the same. It is a figure which shows the example of the reproduction | regeneration permission information used with the system of FIG. It is a figure which shows the aspect of dialog recording and reproduction | regeneration. It is the schematic which shows the example in which reproduction | regeneration is permitted according to a reproduction person in charge. It is the schematic which shows the example in which reproduction | regeneration is permitted according to the kind of apparatus reproduced | regenerated. It is the schematic which shows the example in which reproduction | regeneration is permitted by recording content. It is the schematic which shows the example which makes a reproduction | regeneration sound a synthetic | combination sound when reproduction | regeneration is permitted. It is the schematic which shows the example by which only the one part containing a predetermined keyword is reproduced | regenerated. It is the schematic which shows the example which does not reproduce about the part by which reproduction | regeneration is suppressed. It is a schematic block diagram of the communication recording system which concerns on 6th Embodiment of this invention. It is the schematic which shows another structural example in 6th Embodiment. It is a figure which shows the control system of the voice communication which concerns on 7th Embodiment of this invention. It is a schematic block diagram of the headset system with a radio | wireless communication function to which the audio | voice communication control system of FIG. 26 is applied. 28 is an example of a voice communication state table used in the system of FIG. FIG. 29 is a diagram for further explaining the voice communication state table of FIG. 28.

Explanation of symbols

10, 40, 50, 90, 140, 160, 180, 240, 270, 300 Headset with wireless communication function 11 Wireless communication module 13 CPU board 15 Speaker 17 Microphone 31 Storage means 33 Clock means (time acquisition means)
35, 85, 115, 155 Information creation unit 45 Clock adjustment unit 47 Position specifying unit 121 Voice recognition unit 171 Recording permission determination unit 172 Off-recording switch (off-record setting unit)
173 Permission information storage unit 201 Playback permission information storage unit 203 Playback determination unit 321 Control method determination unit 323 Voice communication execution unit

Claims (5)

A headset with wireless communication function;
A device capable of wireless communication with the headset,
The headset is
A microphone that detects the voice of the wearer of the headset and generates a first audio signal;
Transmitting means for transmitting the first audio signal to the device by short-range wireless communication compliant with a Bluetooth standard;
The equipment is
Receiving means for receiving the first audio signal from the headset by near field communication conforming to a Bluetooth standard;
Voice input means for detecting the voice of the user of the device and generating a second voice signal;
Time information acquisition means for acquiring time information indicating generation times of the first and second audio signals;
A keyword included in the first and second audio signals is extracted, the keyword, the first and second audio signals, the time information, and an identifier for identifying the headset or the wearer. Information creating means for creating information to be stored in association with each other;
Storage means for storing the created information,
The headset and the device are communication recording systems used in a situation where a wearer of the headset and a user of the device face each other . A first headset with a wireless communication function;
A second headset capable of communicating with the first headset;
The first and second headsets and a device capable of transmitting and receiving by short-range wireless communication conforming to the Bluetooth standard,
Each of the first and second headsets includes:
A microphone that detects the wearer's voice and generates a first voice signal;
The first audio signal is transmitted to the other headset via the device by short-range wireless communication conforming to the Bluetooth standard, and the second audio signal transmitted from the other headset is transmitted to the device. Via a short-range wireless communication conforming to the Bluetooth standard via,
The equipment is
Time information acquisition means for acquiring time information indicating generation times of the first and second audio signals;
A keyword included in the first and second audio signals is extracted, the keyword, the first and second audio signals, the time information, and an identifier for identifying the headset or the wearer. Information creating means for creating information to be stored in association with each other;
Storage means for storing the created information,
The first headset and the second headset are communication recording systems used in a situation where a wearer of the first headset and a wearer of the second headset face each other .   At least one of the headset and the device further includes a video acquisition unit, and the information creation unit corresponds to the time information on the first and second audio signals and the video signal acquired by the video acquisition unit. In addition, the communication recording system according to claim 1 or 2, wherein information to be stored is created.   The device further includes position information acquisition means for specifying a location where the first and second sounds are generated, and the information creation means includes the time information and position information in the first and second sound signals. The communication recording system according to claim 1 or 2, wherein information to be stored is created in association with each other. The equipment is
A voice recognition unit that recognizes the first and second voice signals and converts them into character strings;
The communication recording system according to claim 1, further comprising a text information storage unit that stores the character string.

Images