WO2016158267A1 - 情報処理装置、情報処理方法、およびプログラム - Google Patents
情報処理装置、情報処理方法、およびプログラム Download PDFInfo
- Publication number
- WO2016158267A1 WO2016158267A1 PCT/JP2016/057392 JP2016057392W WO2016158267A1 WO 2016158267 A1 WO2016158267 A1 WO 2016158267A1 JP 2016057392 W JP2016057392 W JP 2016057392W WO 2016158267 A1 WO2016158267 A1 WO 2016158267A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- user
- information processing
- users
- processing apparatus
- information
- Prior art date
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/107—Computer-aided management of electronic mailing [e-mailing]
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/20—Speech recognition techniques specially adapted for robustness in adverse environments, e.g. in noise, of stress induced speech
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
- G06T11/206—Drawing of charts or graphs
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/24—Speech recognition using non-acoustical features
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/48—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use
- G10L25/51—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 specially adapted for particular use for comparison or discrimination
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S19/00—Satellite radio beacon positioning systems; Determining position, velocity or attitude using signals transmitted by such systems
- G01S19/01—Satellite radio beacon positioning systems transmitting time-stamped messages, e.g. GPS [Global Positioning System], GLONASS [Global Orbiting Navigation Satellite System] or GALILEO
- G01S19/13—Receivers
- G01S19/34—Power consumption
Definitions
- the present disclosure relates to an information processing apparatus, an information processing method, and a program.
- Patent Document 1 proposes a technique for extracting a conversation group based on the similarity of speech feature amounts such as frequency components extracted from speech information transmitted from each user's terminal device. Yes. This makes it possible to analyze conversations that occur unspecified between unspecified persons.
- a new and improved information processing apparatus information processing, which makes it possible to detect conversations between users more robustly in various aspects while using feature amounts extracted from audio data. Propose methods and programs.
- communication occurs between the plurality of users based on a feature amount indicating an interaction between the plurality of users including the user extracted from voice data including at least the user's speech.
- an information processing apparatus including a communication determination unit that determines whether or not the communication is performed.
- the processor is based on a feature amount indicating an interaction between a plurality of users including the user, which is extracted from voice data including at least the user's speech.
- An information processing method including determining whether or not communication is occurring is provided.
- communication between the plurality of users is performed based on a feature amount indicating an interaction between the plurality of users including the user, which is extracted from voice data including at least the user's speech.
- a program for causing a computer to realize a function of determining whether or not it has occurred is provided.
- conversations between users can be detected more robustly in various aspects while using feature amounts extracted from audio data.
- 1st embodiment of this indication it is a figure which shows the example at the time of expressing the state of the conversation which has generate
- 1st embodiment of this indication it is a figure which shows the example at the time of expressing the state of the conversation which has generate
- FIG. 3 is a block diagram illustrating a hardware configuration example of an information processing apparatus according to an embodiment of the present disclosure.
- FIG. 1 is a diagram for conceptually explaining conversation detection according to an embodiment of the present disclosure.
- FIGS. 1A to 1C show a process of identifying a user having a conversation from a plurality of users in the present embodiment as (a) to (c).
- users other than the target user are divided into candidate users and other users.
- the candidate user is a user who is indicated to be close to the target user by GNSS or Wi-Fi positioning, for example.
- sensor data is acquired for each of the target user and the candidate user. More specifically, the sensor data includes, for example, audio data acquired by a microphone (sound sensor) and sensor data such as acceleration indicating a user's motion. As shown in (c), it is determined whether or not a conversation is occurring between the target user and the candidate user based on the feature amount indicating the interaction between users extracted from these sensor data. .
- the target user can be classified into a common conversation group together with candidate users who are determined to have conversations.
- FIG. 2 is a diagram illustrating a configuration example of a system according to an embodiment of the present disclosure.
- system 10 includes wearable terminal 100 (eyewear 100 a and listware 100 b), mobile terminal 200, and server 300.
- the devices constituting the system 10 can be realized by, for example, a hardware configuration of an information processing device described later.
- Wearable terminal 100 is worn by each user.
- Wearable terminal 100 has a microphone (sound sensor), for example, and acquires voice data including a user's speech.
- Wearable terminal 100 has other sensors, such as an acceleration sensor and a gyro sensor, and may acquire sensor data, such as acceleration which shows a user's motion, for example.
- the eyewear 100a may be able to acquire sensor data indicating acceleration or angular velocity corresponding to the user's nodding.
- the listware 100b it may be possible to acquire sensor data indicating a biological index such as an acceleration or angular velocity corresponding to a user's hand movement or a pulse.
- the wearable terminal 100 may be used to present information generated by information processing according to the present embodiment as described later to the user. More specifically, the wearable terminal 100 may include an output device such as a display or a speaker, and information may be presented to the user from these output devices by an image or sound. In the illustrated example, wearable terminal 100 and mobile terminal 200 are shown separately, but in other examples, functions of wearable terminal 100 may be included in mobile terminal 200. In this case, the mobile terminal 200 acquires sensor data using a microphone, an acceleration sensor, a gyro sensor, or the like, and presents information generated by information processing to the user.
- the mobile terminal 200 is carried by each user.
- the mobile terminal 200 relays communication between the wearable terminal 100 and the server 300. More specifically, for example, communication between the wearable terminal 100 and the mobile terminal 200 is performed by wireless communication such as Bluetooth (registered trademark), and communication between the mobile terminal 200 and the server 300 is performed by network communication such as the Internet.
- the mobile terminal 200 may transmit the information received from the wearable terminal 100 to the server 300 after processing the information as necessary.
- the mobile terminal 200 may analyze sensor data including audio data received from the wearable terminal 100 and extract an intermediate feature amount.
- the mobile terminal 200 may transfer the sensor data received from the wearable terminal 100 to the server 300 as it is.
- the system 10 may not necessarily include the mobile terminal 200.
- the mobile terminal 200 may be used to present information generated by information processing according to the present embodiment as described later to the user instead of or along with the wearable terminal 100.
- the server 300 is realized by one or a plurality of information processing apparatuses on the network, and provides a service to each user. For example, the server 300 extracts feature quantities from sensor data collected from each user's wearable terminal 100 via the mobile terminal 200, and determines whether or not a conversation between users is occurring based on the feature quantities. judge. For example, the server 300 may generate information that expresses the state of occurrence of conversation between users based on the determination result. This information may be used for, for example, a screen display for a user (which may be a user who is not participating in a conversation or a user who is not a target for conversation detection) to grasp the occurrence state in real time. , May be accumulated as a log.
- the information accumulated as a log may be referred to by a user as described above afterwards, or a graph structure that expresses the relationship among a plurality of users is specified based on the information accumulated as a log. May be.
- these processes may be executed by the mobile terminal 200 that acts as a host between the wearable terminal 100 and the mobile terminal 200 of each user, for example.
- the system 10 does not necessarily include the server 300.
- FIG. 3 is a diagram illustrating a functional configuration example of a system according to an embodiment of the present disclosure.
- the system 10 includes a sensing unit 11, an action detection unit 12, a candidate selection unit 13, a feature amount extraction unit 14, a conversation determination unit 15, a score calculation unit 16, as functional configurations. And a grouping unit 17.
- the sensing unit 11 is mounted by a sensor such as a microphone (sound sensor), an acceleration sensor, and / or a gyro sensor included in the wearable terminal 100 described with reference to FIG.
- the action detection unit 12, the candidate selection unit 13, the feature amount extraction unit 14, the conversation determination unit 15, the score calculation unit 16, and the grouping unit 17 are servers 300, mobile terminals 200, and / or a processor such as a CPU in the server 300.
- the functional configuration implemented by the above processor may be integrated and implemented in any one of the information processing apparatuses included in the system 10, or may be distributed and implemented in a plurality of information processing apparatuses.
- each functional configuration will be further described.
- the sensing unit 11 includes, as inputs of the system 10, for example, a microphone (sound sensor) that acquires voice data, a sensor such as an acceleration sensor or a gyro sensor that acquires sensor data such as acceleration indicating user motion. Furthermore, the sensing unit 11 includes a wireless communication device such as a GNSS receiver or Wi-Fi that acquires user location information.
- the sensing unit 11 is implemented in the wearable terminal 100 such as the eyewear 100a and the listware 100b illustrated in FIG. 2, for example. As described above, the function of the wearable terminal 100 is included in the mobile terminal 200. In this case, the sensing unit 11 is mounted in the mobile terminal 200.
- the action detection unit 12 detects the action of each user who provided the sensor data from the sensor data (which may include voice data) acquired by the sensing unit 11. More specifically, for example, the action detection unit 12 detects an action of a user's utterance from voice data.
- the action detection unit 12 does not necessarily have to detect the voice feature or the content of the utterance in the utterance. That is, the action detection unit 12 may simply detect whether or not the user has spoken at a certain time.
- the action detection unit 12 may detect the voice characteristics and the content of the utterance when they can be detected.
- the action detection unit 12 detects an action such as a user's nodding or hand movement (gesture) from sensor data of acceleration and angular velocity.
- the action detection unit 12 may detect a user's psychological action from sensor data of a biological index such as a user's pulse.
- the candidate selection unit 13 detects the positional relationship of each user who provided the sensor data from the sensor data acquired by the sensing unit 11, and selects a user whose positional relationship satisfies a predetermined condition as a candidate for a user who forms a conversation group Elected as. More specifically, the candidate selection unit 13 selects other users who are shown to be close to the target user by GNSS, Wi-Fi positioning, or the like as candidate users. In addition, when the candidate selection part 13 selects a candidate user, the positional information on each user may not necessarily be available. For example, the candidate selection unit 13 allows each user terminal device (for example, the wearable terminal 100 or the mobile terminal 200) to directly communicate with each other by wireless communication such as Bluetooth (registered trademark). May be recognized as close to each other.
- wireless communication such as Bluetooth (registered trademark). May be recognized as close to each other.
- the candidate selection part 13 may select a candidate user based on each user's action information. More specifically, for example, the candidate selection unit 13 acquires a user's action recognition result (for example, work in a office, meeting, etc.) associated with the position, and the action recognition result common to the target user is acquired. Other users may be selected as candidate users. In addition, for example, the candidate selection unit 13 acquires a user's schedule associated with the position (for example, work, meeting, etc. in the office, similar to the action recognition result), and the schedule common to the target user is acquired. Other users may be selected as candidate users.
- a user's action recognition result for example, work in a office, meeting, etc.
- Other users may be selected as candidate users.
- the feature amount extraction unit 14 extracts, for each of the candidate users extracted by the candidate selection unit 13, a feature amount indicating an interrelationship of actions detected by the action detection unit 12, that is, an interaction with the target user.
- such feature amounts are extracted based on the temporal relationship between actions.
- the feature amount extraction unit 14 extracts a feature amount indicating an interaction between a plurality of users including the user from voice data including the user's uttered voice. More specifically, the plurality of users include a first user and a second user, and the feature amount extraction unit 14 determines the utterance voice of the first user (which may be the target user) and the second user (candidate). The feature amount is extracted based on the temporal relationship with the uttered voice of the user. This feature may indicate that an utterance exchange has occurred between the first user and the second user. For example, when the first user and the second user are having a conversation, it is unlikely that the utterance sections of the first user and the second user will be largely overlapped, and the utterance sections of the respective users are substantially alternating. Should have occurred.
- the audio data acquired by the sensing unit 11 includes first audio data including the first user's utterance audio and second audio data including the second user's utterance audio, respectively. It may be included separately. Or the audio
- the feature amount extraction unit 14 provides the utterance voice of the user included in the voice data provided by each user between the first user and the second user, as well as from each user.
- the feature amount may be extracted based on the temporal relationship with the motion or the biological index indicated by the sensor data. That is, for example, the feature amount extraction unit 14 performs the motion or biological index indicated by the first user's speech voice included in the voice data provided by the first user and the sensor data provided by the second user.
- the feature amount may be extracted based on the relationship between Further, the feature quantity extraction unit 14 may extract not only the feature quantity between the target user and the candidate user but also the feature quantity between the candidate users.
- the conversation determination unit 15 determines whether or not a conversation is occurring between a plurality of users based on the feature amount extracted by the feature amount extraction unit 14.
- the conversation determination part 15 is a conversation between the users selected from the plurality of users (all users to be processed) based on the positional relationship of each user. It is determined whether or not this has occurred.
- the user to be determined may be selected based on the behavior information of each user. More specifically, for example, when the conversation determination unit 15 exceeds the predetermined threshold, the conversation occurrence probability calculated based on the feature amount extracted between the first user and the second user. In addition, it is determined that a conversation is occurring between the first user and the second user.
- the conversation determination unit 15 calculates the occurrence probability based on the feature amount extracted by the feature amount extraction unit 14 between the target user and the candidate user, so that the candidate user who has conversation with the target user Can be specified. Furthermore, the conversation determination unit 15 can specify the conversation that has occurred between the candidate users by calculating the occurrence probability based on the feature amount extracted by the feature amount extraction unit 14 between the candidate users. . By specifying a conversation that occurs not only between the target user and the candidate user but also between the candidate users, it is possible to grasp the situation of the conversation that has occurred around the target user.
- the score calculation unit 16 calculates a score among each of a plurality of users based on the conversation occurrence history based on the determination of the conversation determination unit 15. For example, the score calculation unit 16 may calculate the score by integrating the time during which conversation has occurred between the users in a predetermined period. Alternatively, the score calculation unit 16 may calculate the score based on the frequency at which a conversation of a predetermined time or more has occurred between each user during a predetermined period. In addition, for example, the score calculation unit 16 refers to the conversation occurrence probability calculated by the conversation determination unit 15 when it is determined that a conversation has occurred between users, and determines that the conversation has occurred with a higher occurrence probability. A higher score may be calculated among the users who have been.
- the score calculation unit 16 estimates the degree of conversation excitement based on these, and the excitement is larger. Higher scores may be calculated among users who have conversations.
- the grouping unit 17 groups users based on the score calculated by the score calculation unit 16. Grouping expressions can vary. For example, the grouping unit 17 classifies a plurality of users whose mutual scores exceed a threshold value into a common group. Further, the grouping unit 17 may specify a graph structure that expresses the relationship between users. The graph structure may be defined separately from the group, or the group may be defined according to the presence / absence or strength of the graph structure link. Note that the generation of information based on the determination result of the conversation determination unit 15 in the present embodiment is not limited to the grouping unit 17 and can be implemented in various forms. Such other examples will be described later.
- FIG. 4 is a diagram for describing action detection according to an embodiment of the present disclosure.
- the wearable terminal 100c includes a headset 110 and a motion sensor 120 for detecting the user's nodding.
- the headset 110 includes a microphone 112 and acquires audio data.
- the motion sensor 120 includes a gyro sensor 122 and an acceleration sensor 124, and acquires sensor data of angular velocity and acceleration.
- the action detection unit 12 in the system 10 described above uses the start condition that the energy extracted from the audio data exceeds the threshold value, and the end condition that the state where the energy is below the threshold value continues for a predetermined time or more. Can be detected.
- the action detection unit 12 can detect a section where the user's nod is generated by removing a section where the dispersion of acceleration is large from a section where the predetermined frequency of the angular velocity is large.
- the feature amount extraction unit 14 in the system 10 calculates a feature amount indicating an interaction between the first user and the second user.
- the feature amount extraction unit 14 extracts a positive feature amount with respect to the interaction between users based on, for example, the following events. That is, when the occurrence frequency of the following events is high, the feature amount indicating the interaction between users can be high.
- the feature amount extraction unit 14 calculates a negative feature amount for the interaction between users based on, for example, the following events. That is, when the occurrence frequency of the following events is high, the feature amount indicating the interaction between users can be low.
- ⁇ Speaking section coincidence (the first user's utterance and the second user's utterance occur simultaneously) ⁇ No non-speaker reaction to utterance
- the feature amount extraction unit 14 performs the calculation of the feature amount based on the above event at a predetermined cycle (100 Hz as an example).
- the conversation determination unit 15 may calculate the calculated feature value for a predetermined period (feature value calculation period may be longer. For example, 0.2 Hz. In this case, the feature value may be treated as an average value every 30 s.
- the determiner is, for example, a binary determiner, and determines whether the first user and the second user are likely to have a conversation or not.
- Such a determiner is generated by machine learning, for example.
- a machine learning technique for example, SVM (Support Vector Machine) can be used, but not limited to this example, various known techniques can be used.
- the determination device in this embodiment should just be a determination device which can perform determination which is demonstrated below by the output. More specifically, the determiner may be a binary determiner or a determiner that outputs a probability. Further, the determiner is not necessarily generated by machine learning.
- FIG. 5 is a diagram for describing the occurrence determination of the conversation according to an embodiment of the present disclosure.
- the conversation determination unit 15 of the system 10 calculates the occurrence probability according to the output of the determiner, and determines whether or not conversation is occurring based on the occurrence probability. More specifically, the conversation determining unit 15 increases the probability of occurrence when the above-described determiner is a binary determiner and the output of the determiner is positive (conversation seems to have occurred) If the output is negative (no conversation appears to have occurred), the probability of occurrence is lowered.
- the conversation determination unit 15 may change the score of the occurrence probability according to the magnitude of the probability output from the determiner.
- the output of the determiner is implemented at 0.2 Hz, and the occurrence probability is updated every 5 s.
- the conversation determination part 15 determines with the conversation having generate
- FIG. 6 to 8 are diagrams illustrating an example in a case where the state of conversation occurring between users is expressed in time series in the embodiment of the present disclosure.
- the conversation state between users based on the conversation occurrence determination by the conversation determination unit 15 as shown in FIG. May be output in time series.
- Such an output may be presented to the user himself / herself who is involved in the conversation by the wearable terminal 100 or the mobile terminal 200 shown in the example of FIG.
- it may be presented to other users who do not use the system 10 but have viewing authority.
- a display control unit that displays a screen for presenting detected conversations in time series can be implemented by the processor of wearable terminal 100, mobile terminal 200, or server 300.
- FIG. 6 is an example of expressing the state of conversation occurring between two users.
- the occurrence state of the conversation between the user U1 and the user U2 is indicated by the link L1 and the link L2.
- the widths of the link L1 and the link L2 change corresponding to the occurrence probability of the conversation between the user U1 and the user U2 calculated as in the example shown in FIG. That is, in the illustrated example, the link L1 indicates that the probability of occurrence of a conversation between the user U1 and the user U2 is low. For example, when the occurrence probability of the conversation is greater than 0 but does not reach the predetermined threshold, the link L1 can be displayed with the narrowest width. On the other hand, the link L2 having a wider width can be displayed as the probability of occurrence of conversation rises above the threshold.
- the probability of occurrence of conversation is used to determine whether or not conversation is occurring, for example, by determination using a threshold as shown in FIG.
- the probability of occurrence of conversation is calculated based on the feature amount as exemplified above, the high probability of occurrence of conversation means that the exchange of utterances and the nodding of non-speakers occur frequently. Can represent. Accordingly, it is possible to interpret the probability of occurrence of conversation as, for example, a continuous value indicating the conversation activity and use it for the change in display form as described above.
- FIG. 7 is an example of expressing the state of conversation occurring between three users.
- feature quantities are extracted for a pair of users (first user and second user) included in a plurality of target users, and further based on the feature quantities. It is determined whether or not a conversation is occurring according to the calculated conversation occurrence probability.
- a wide link L2 is displayed between the user U1 and the user U2 and between the user U2 and the user U3.
- the link L2 represents that a conversation is actively occurring between these users, as in the example shown in FIG.
- a link L3 having a medium width is displayed between the user U3 and the user U1.
- the link L3 expresses that the conversation is occurring between the user U3 and the user U1, but is not so active. From such a display, it is estimated that the conversation between the three users U1 to U3 is progressing with the user U2 as a hub.
- the user U4 may be treated as a candidate user because the position of the user U4 is close to the users U1 to U3.
- the feature quantity indicating the interaction between users extracted by the feature quantity extraction unit 14 does not become positive with respect to the occurrence of conversation, and therefore the probability of occurrence of conversation calculated by the conversation determination unit 15 is also high. It will not be high. Therefore, for example, a narrow link L1 may be displayed between the user U1 and the user U4 or between the user U3 and the user U4 as shown in FIG. Since the link does not exceed the width of the displayed link, the link L1 disappears when the user U4 moves away.
- FIG. 8 is an example of expressing the state of conversation occurring between four users. Also in this example, a feature amount is extracted for a pair of users included in a plurality of target users, and it is further determined whether or not conversation is occurring according to the conversation occurrence probability calculated based on the feature amount.
- a wide link L2 is displayed between the user U1 and the user U2 and between the user U3 and the user U4.
- the link L2 represents that a conversation is actively occurring between these users, as in the example described above.
- a narrow link L1 is displayed between the remaining combinations of the four users.
- the link L1 expresses that almost no conversation has occurred between these users.
- the feature amount extraction unit 14 extracts feature amounts indicating the interaction between users, it is possible to accurately specify a combination of users who are actually talking. is there.
- FIG. 9 is a diagram for explaining optimization of a conversation graph structure according to an embodiment of the present disclosure.
- the graph structure is optimized according to preset rules. More specifically, in the illustrated example, in the original graph structure including the users U1 to U7, the link between the user U2 and the user U5 is cut by optimization, and the conversation group including the users U1 to U4 The conversation group including the users U5 to U7 is separated.
- the conversation determination unit 15 performs the optimization as described above by minimizing the energy of the generated graph structure (rule that minimizes energy).
- the conversation determination unit 15 optimizes the graph structure according to a rule based on common sense such that the user who becomes a conversation hub is one person in the conversation group like the user U2 in the example of FIG. Also good.
- the conversation determination unit 15 may use the speed of another user's reaction to a certain user's utterance to determine whether or not to maintain the link (speaker) And links between non-speakers who had a quick reaction to utterances are easier to maintain).
- the wearable terminal 100 can detect the face orientation of each user, the conversation determination unit 15 may use the face orientation of each user to determine whether or not to maintain the link. Good (the link between the speaker and the speaker whose face is facing the speaker is more easily maintained).
- FIG. 10 is a diagram for describing feature amount expansion according to an embodiment of the present disclosure.
- the user's utterance and nodding are used as an action for extracting a feature amount for detecting a conversation between users.
- feature amounts may be extracted based on various actions.
- FIG. 10 includes a microphone 112, a motion sensor 120 (which may include the gyro sensor 122 and the acceleration sensor 124 illustrated in FIG. 4), a geomagnetic sensor 126, and sensors that can be used to extract such feature quantities.
- a biosensor 128 is illustrated. These sensors are included in the wearable terminal 100 or the mobile terminal 200, for example.
- the illustrated actions 130 to 140 will be described.
- the nod 130 is detected based on sensor data acquired by the motion sensor 120 mounted on the terminal device such as the eyewear 100a mounted on the user's head. Is done. Further, the nod 130 may be detected using the sensor data of the geomagnetic sensor 126 similarly mounted on the terminal device.
- the utterance section 132 is detected based on voice data including the user's utterance voice acquired by the microphone 112.
- the speech content 134 may be detectable from the voice data. In this case, for example, detecting a conversation occurring between users based on the commonality of conversation topics in addition to the state of interaction between users indicated by the temporal relationship of the utterance section 132. Can do.
- the body direction 136 is detected using sensor data acquired by the geomagnetic sensor 126, for example. As described above with reference to FIG. 9, for example, when a non-speaker is facing a speaker or a speaker is facing a non-speaker, the occurrence of a conversation is detected there. Can be a positive element to do.
- the gesture 138 is detected using sensor data acquired by the motion sensor 120 or the geomagnetic sensor 126, for example.
- the gesture 138 can be used as an element indicating an interaction between users by specifying a temporal relationship with the utterance interval, like the nod in the example described with reference to FIG. 4, for example.
- the pulse 140 is detected using the biosensor 128, for example. For example, when the user is actively talking, if the pulse 140 also shows an upward trend, estimating the consistency between the pulse state and the user's conversation state, and whether the user is talking (E.g., the user is not actually speaking if the pulse 140 is not elevated, even though other actions or features indicate that it is actively speaking) there is a possibility).
- the conversation determining unit 15 may consider the context of the user's behavior and the usage method of the specified conversation group when classifying the user into the conversation group. For example, when trying to share a user's private image between specified conversation groups, by setting a high threshold for determining that a conversation is occurring between users, It is possible to prevent the image from being shared between them. In addition, for example, in a party venue where there is a high possibility that participants are having a conversation in a wide range, by setting the threshold value low, it is possible to classify the partner with whom the user has a conversation into a conversation group.
- the user may be in a place with few people, such as a house, while setting a high threshold to prevent false detections during daytime hours when the user is often in a crowd such as the city. You may set a low threshold value at the time zone of many nights.
- FIG. 11 is a diagram for describing a usage example of information obtained by detecting conversation in an embodiment of the present disclosure.
- the determination result of the conversation occurrence by the conversation determination unit 15 is used by the score calculation unit 16 and the grouping unit 17. It is not limited to such an example, and various other uses are possible.
- FIG. 11 shows a UI generation unit 171 to a topic recommendation unit 183 as functional configurations for such use.
- These functional configurations are implemented in software in the terminal device or server using, for example, a processor and memory or storage.
- information provided by these functional configurations will be further described.
- the UI generation unit 171 may provide a user interface that graphically displays the state of conversation between users in a time series as described above with reference to FIGS. 6 to 8, for example.
- the UI generation unit 171 may provide a user interface that displays a list of real-time conversation states as described above.
- the UI generation unit 171 may enable the real-time conversation state displayed as described above to be linked to another application.
- the UI generation unit 171 cooperates with the social media link function 172 to enable sharing of data such as images among users belonging to a common conversation group or to play a game at the same time. May be supported.
- the conversation state between the users detected as described above is used for various purposes, the conversation close to the actual situation of the conversation by ad hoc conversation group recognition between the terminal devices as shown in FIG. Groups can be recognized.
- terminal devices 100x and 100y (the terminal devices used by each user may be used, for example, the wearable terminal 100 in the example of Fig. 2 or the mobile terminal 200.
- Each of them includes a sensing unit 11, an action detection unit 12, a candidate selection unit 13, a feature amount extraction unit 14, a conversation determination unit 15, a communication unit 31, a display unit 32, and a UI generation unit 171.
- the communication unit 31 is implemented by a communication device such as Bluetooth (registered trademark) included in each of the terminal devices 100x and 100y.
- the display unit 32 is implemented by a display such as an LCD provided in each of the terminal devices 100x and 100y.
- the corresponding functional configurations of the terminal device 100x and the terminal device 100y cooperate to specify that users of each other belong to a common conversation group.
- the candidate selection unit 13 selects a candidate user based on the position information acquired by the sensing unit 11 and the position information acquired by the partner sensing unit 11. At this time, the users of the terminal devices 100x and 100y are each selected as a candidate user.
- the action detection unit 12 identifies a section in which an action such as utterance or nodding occurs based on the sensor data acquired by the sensing unit 11.
- the feature amount extraction unit 14 shares information such as the section specified by the action detection unit 12 of each terminal device via the communication unit 31, and the feature amount indicating the interaction between the users of the terminal devices 100x and 100y. To extract.
- the conversation determination unit 15 determines whether a conversation is occurring between the users of the terminal devices 100x and 100y based on the extracted feature amount.
- the UI generation unit 171 generates the above-described graph or list user interface according to the determination result, and presents it to each user via the display unit 32.
- FIGS. 13 and 14 are diagrams showing examples of user interfaces provided in the first example described above.
- a user interface a screen is displayed on the display 210 of the mobile terminal 200 (for example, a similar screen may be displayed on the wearable terminal 100).
- conversation groups between users are displayed as graphs on the screens 2100a and 2100b.
- the user can transition between the screen 2100a and the screen 2100b by a zoom-in / out operation.
- On the screen 2100a for example, only other users who are more reliably estimated to belong to the same conversation group are displayed based on, for example, a relatively high occurrence probability threshold.
- the user can correct the recognition result of the conversation group by performing an operation on another user's icon displayed on the screens 2100a and 2100b via a touch panel or the like.
- the user can exclude other users indicated by the icons from the conversation group by moving the icons of the other users out of the screens 2100a and 2100b by a swipe operation.
- conversation groups between users are displayed as a list on the screen 2100c.
- the display order of the list may be the order in which the conversation occurrence probability calculated by the conversation determination unit 15 is high. Therefore, the display order of the list can change dynamically.
- the maximum number of users included in the conversation group may be limited.
- the history of the conversation partner is output on the timeline by the log output unit 175 and the link function 172 to social media.
- FIG. 15 shows a functional configuration for such output (note that in the illustrated example, the estimation of the conversation is executed by the terminal device, but may be executed by the server).
- the terminal device 100 z includes a sensing unit 11, an action detection unit 12, a candidate selection unit 13, a feature amount extraction unit 14, a conversation determination unit 15, a communication unit 31, a log output unit 175, and social A function 172 for linking with media is included.
- the log output unit 175 is based on the conversation occurrence history based on the determination of the conversation determination unit 15, or information on a partner with whom at least one user included in a plurality of users (all users to be processed) has conversation, or conversation A log containing at least one of conversation information with the other party is output.
- the log generated by the log output unit 175 is a timeline of social media presented to the at least one user via the function 172 for linking with social media (in another example, a time not related to social media). Line (which may be a line).
- the voice recognition unit 34 and the topic recognition unit 35 are implemented in software.
- another user who exchanged a certain amount of conversation (which can be determined based on conversation time or high conversation probability) can be recommended as a friend on social media.
- information such as a conversation topic recognized by the processing of the voice recognition unit 34 and the topic recognition unit 35, information on a place where the conversation has occurred, an image, and the like may be added to the log.
- the conversation log is filtered and displayed by topic or conversation partner, it is useful as a user's memory assist tool or memory recording means.
- FIG. 16 is a diagram illustrating an example of a user interface provided in the second example described above.
- a screen is displayed on the display 210 of the mobile terminal 200 as a user interface (for example, a similar screen may be displayed on the wearable terminal 100).
- a date 2101 and a list on the timeline indicating other users with whom the user has a conversation on that day are displayed on the screen 2100 d.
- the list may include, for example, other user names 2103 and conversation time 2105.
- the user can display the conversation history screen 2100e of the user by selecting one of the users displayed as a list on the screen 2100d.
- the screen 2100e includes a name 2103 and a conversation time 2105 of another user, an icon 2107 for becoming a friend with the user on social media, and a past conversation history 2109 with the user.
- the conversation topic 2111 may be displayed in addition to the date and conversation time when the conversation occurred.
- the feature amount extraction unit 14 not only extracts a feature amount based on the relationship between the utterance speeches of a plurality of users, but also the utterance speech of one user, It is possible to extract a feature amount based on a temporal relationship with an action (for example, motion or biological index) other than the user's utterance.
- an action for example, motion or biological index
- the person recognition unit 173 who has spoken in a single conversation is not limited to other users in which the user exchanges utterances, but also other users who have performed some action on the user's utterances, or It is possible to recognize other users who have made utterances to which user actions are directed and display them on the timeline provided by the log output unit 175. Based on this, the user can perform an action 174 on the cloud for the above-mentioned other users (often not acquaintances). At this time, in the action on the cloud, for protection of privacy, for example, the personal information does not necessarily have to be exchanged to the extent that the other party's avatar can be seen.
- FIG. 17 is a diagram showing a functional configuration for output as described above.
- the terminal device 100w includes a sensing unit 11, an action detection unit 12, a candidate selection unit 13, a feature amount extraction unit 14, a conversation determination unit 15, a communication unit 31, a log output unit 175, a post process. Part 36.
- the post-processing unit 36 corresponds to the above-described single recognition person recognition unit 173 and action 174 on the cloud, and is implemented in software by, for example, a processor included in the terminal device 100w operating according to a program.
- the log output unit 175 outputs a conversation group generation result as a log.
- the post-processing unit 36 identifies another user in the log in which a communication including a conversation of a predetermined time or less or an utterance of only one user is detected. Further, the post-processing unit 36 extracts other users who have had a temporary encounter by excluding users who are already friends on social media from among the identified other users. It is possible to execute an action on the cloud for such a user.
- the topic recommendation unit 183 shown in FIG. 11 supports user communication by providing a topic.
- the topic recommendation unit 183 estimates the degree of conversation excitement from the conversation tempo indicated by the feature amount extracted by the feature amount extraction unit 14, and recommends the topic accordingly.
- the topic recommendation unit 183 has a tendency that the conversation occurrence probability calculated by the conversation determination unit 15 tends to decrease, or a separately estimated conversation excitement level (for example, a user's utterance tempo or voice If the user is in a downward trend (estimated by size), it may be determined that the user needs a new topic and a topic different from that may be recommended.
- the topic recommendation unit 183 determines that the current topic is continued when the occurrence probability of the conversation is increasing or the degree of excitement of the conversation matches with the increasing tendency, Information about the topic may be provided to the user.
- the topic recommendation unit 183 may provide a topic to the user according to the log output by the log output unit 175 or the familiarity calculated by the familiarity graph generation unit 177 described later. Good. More specifically, for example, when the topic recommending unit 183 is conversing with a partner with whom the user normally has a conversation (a partner with a lot of conversation logs remaining) or a partner with a high degree of familiarity, Since the conversation should be exciting, when it is estimated that the conversation is inactive as described above, it may be determined to provide a new topic.
- the topic recommending unit 183 may have a case where conversation is not particularly necessary when the user is talking to a partner who does not usually talk much (a partner who does not have a lot of conversation logs), or a partner with a low familiarity. Therefore, provision of a new topic may be suppressed even when it is estimated that the conversation is inactive as described above.
- the familiarity graph generation unit 177 illustrated in FIG. 11 generates a graph indicating the familiarity between users based on the log output by the log output unit 175.
- the familiarity graph generated here can be said to be a graph structure that expresses the relationship between a plurality of users based on the conversation history. Therefore, it can be said that the familiarity graph generation unit 177 is a relationship graph specifying unit that specifies such a graph structure.
- the closeness graph generation unit 177 generates a strong link in the closeness graph between users whose conversation frequency is high or the accumulated time of the conversation is long due to the log of the log output unit 175.
- the familiarity graph generation unit 177 estimates the familiarity based on the number and types of reactions extracted by the action detection unit 12 and the feature amount extraction unit 14, and establishes a strong link between users with high familiarity. It may be generated. More specifically, the closeness graph generation unit 177 determines the strength of the link between users in the closeness graph depending on whether a user often speaks or nods in a conversation with another user. You may change the height. In addition, the closeness graph generation unit 177 adds a label (for example, a parent, a brother, a boss, a colleague, a friend, a lover, etc.) to a link between users based on information acquired from the outside (such as a user's profile). May be given. Similarly, the familiarity graph generation unit 177 may give a label (for example, family, company, synchronization, circle) to a group formed on the familiarity graph.
- a label for example, family, company, synchronization, circle
- the familiarity graph generation unit 177 may calculate the familiarity C with other users using a mathematical expression such as the following Expression 1.
- the index i is given each of the conversations that occurred with other users, t now is the current time, t Past_i the time conversation with i th other users occurs (therefore, in equation 1, the more conversations that occurred in the past, the effect on decrease familiarity), duration i is the cumulative time of the i-th conversation, the speak i time that was spoken in the i-th conversation, nod i is i
- the time spent nodding in the second conversation (so, in Formula 1, the longer the time spoken compared to the time nodded, the greater the intimacy), positive i and negative i are based on biological information etc.
- Estimated emotions to the other user during the i-th conversation (positive and negative. In formula 1, if positive emotions are stronger, Degree is a plus, indicating the degree of intimacy is negative if stronger is more of negative emotions).
- the sharing desire graph generation unit 179 illustrated in FIG. 11 applies the filter generated by the adaptive graph generation unit 181 to the closeness graph generated by the closeness graph generation unit 177, thereby allowing the user to Set the range for sharing content.
- the shared desire graph generation unit 179 is shared in a graph structure that expresses the relationship between users in an aspect in which at least one user included in a plurality of users (all users to be processed) shares information. It is an example of the shared user specific
- FIG. 18 is a diagram for schematically explaining generation of a shared desire graph in the present embodiment.
- FIG. 18 shows that the sharing desire graph G2 can be obtained by applying the filter F to the group familiarity graph G1.
- the group closeness graph G1 is generated by the group closeness graph generation unit 178 shown in FIG. 11, for example.
- the group closeness graph G1 is generated by, for example, integrating the closeness graph for each user generated by the closeness graph generation unit 177 for an arbitrary user group (including a plurality of users).
- the filter F corresponds to the adaptive graph generated by the adaptive graph generation unit 181 and a filter related to the shared information is selected.
- a graph of interest is selected from graphs such as place, interest, and affiliation, and a filter F corresponding to the selected graph is applied.
- the adaptive graph generation unit 181 includes a context at the time of content sharing recognized by the context recognition unit 180 (what kind of content is shared in what situation), a user An adaptive graph that provides an appropriate filter F is generated based on the current profile 182.
- the sharing desire graph G2 the positional relationship of other users constituting the graph is changed compared to the group familiarity graph G1.
- One user is strengthened by applying filter F, and another user is weakened by applying filter F (in the example shown, the strength of the link is represented by the distance from the center of the graph). ).
- the group familiarity graph G1 is simply used. Therefore, it is possible to set an appropriate sharing destination or its candidate according to the type of content and the context in which sharing has occurred, rather than determining the sharing destination and its candidate.
- an adaptive graph corresponding to the location attribute may be selected, and links with other users in the user's current location (travel destination) may be strengthened (included in the graph structure) Filter configured based on the user's positional relationship).
- an adaptive graph corresponding to the job may be selected, and a link with another user (for example, a colleague) having a work relationship may be strengthened (in the graph structure). Filter configured based on the user's affiliation).
- an adaptive graph corresponding to the interest may be selected, and a link with another user who is interested in the sport may be strengthened (graph structure) Filter configured based on the user's interests included in the).
- graph structure Filter configured based on user behavior information included in the.
- the adaptive graph may be selected so that the link with other users who have knowledge is strengthened ( Filter configured based on knowledge of users included in graph structure).
- the adaptive graph generation unit 181 automatically selects an adaptive graph (for example, on a rule basis) based on a recognized context, a user profile, and the like. It may be possible to select the adaptation graph according to the user's selection after presenting it to the user.
- the adaptive graph is selected according to the user's selection, the selection criterion of the adaptive graph (based on the context of the user's situation, the type of content to be shared, etc.) is learned based on the user's selection result, Eventually, the adaptive graph generation unit 181 may be configured to automatically select the adaptive graph.
- FIG. 19 is a diagram for explaining dynamic correction of the familiarity graph in the present embodiment.
- the group familiarity graph G3 for the user A and the user B is corrected by the position of the user A's remark that "C should have come" in the conversation between the user A and the user B.
- the familiarity graph G4 in which the link between the user C and the users A and B is strengthened (the position of the user C is brought closer to the center of the graph) is generated.
- the links with the user C are strengthened, and other users (users D to F) Link is weakened (the position moves away from the center of the graph).
- the link with the user C is strengthened because the name of the user C is mentioned in the actual utterance of the user A, but the same processing is performed, for example, when the user A and the user B chat online. This is also possible when the name of the user C is included in the text input by the user A (or the user B).
- This may be an example of temporarily correcting when the name of the user C is included in the content transmitted by the user A in the case of a virtual conversation such as an online chat.
- the group familiarity graph generation unit 178 temporarily strengthens the relationship between the user A and the user C in the group familiarity graph.
- the group closeness graph generation unit 178 may temporarily strengthen the link in the closeness graph for other users whose users have turned their eyes.
- the sharing desire graph (G3 shown in FIG. 18) is presented to the user as an image.
- a shared range (range R indicated by G3 in FIG. 18) may be superimposed on the shared desire graph.
- the sharing range R is finally determined after the automatically generated candidates are first presented to the user, and the user corrects the candidates. For example, the user enlarges / reduces a figure such as a circle indicating the shared range R, includes other users displayed as icons in the shared range R, or excludes them from the shared range R.
- the shared range R can be corrected.
- the sharing desire graph generation unit 179 learns the user's intention regarding the sharing range from the correction result of the sharing range R, and eventually the sharing desire graph generation unit 179 automatically sets an appropriate sharing range R. May be.
- a user who receives content sharing can selectively share content of other users who are really intimate or content of interest. , Higher satisfaction in content sharing can be obtained.
- content that a user is experiencing in real time (such as live sports watching) can be shared in real time with other users at remote locations, thereby sharing the experience.
- an information processing apparatus for example, an information processing apparatus, a system, an information processing method executed by the information processing apparatus or system, a program for causing the information processing apparatus to function, and a program are recorded. It may include tangible media that is not temporary.
- the conversation is not necessarily limited to all of the related users, for example, only some users speak, and other users nods according to the utterance. It is possible to detect the case where the above action is taken. Therefore, it can be said that the occurrence of communication between users can be detected in the embodiment of the present disclosure including the case where such a case is detected separately from the conversation in other embodiments (the conversation is a communication A kind). Therefore, it can be said that the conversation determination unit is an example of a communication determination unit.
- Second embodiment> (2-1. Overview and system configuration)
- the embodiment has been described in which it is determined whether or not a conversation is occurring between the target user and the candidate user based on the feature amount indicating the interaction between the users.
- 2nd embodiment which is an application example of 1st embodiment mentioned above is described.
- a system in which positioning information is exchanged between a plurality of users will be described.
- GNSS positioning consumes a large amount of power, and it is desirable that GNSS positioning can be performed with low power in a terminal having a small battery such as the mobile terminal 100 or the wearable terminal 200. Therefore, an embodiment in which positioning information is exchanged among a plurality of users will be described below.
- FIG. 20 is a diagram showing an outline of the GNSS positioning method in the second embodiment.
- a group is formed among a plurality of users by using a wearable terminal 100 such as a pendant type, a spectacle type, or a wristwatch type.
- FIG. 20 illustrates a case where a group is formed by three users U1 to U3.
- the plurality of users include at least one accompanying person who acts together with the specific user.
- the relationship between a specific user and a companion is not important, and the other user (others in the case of a plurality of users) viewed from one user is treated as a companion.
- a method for recognizing a plurality of accompanying persons who form a group will be described later.
- a GNSS positioning right (conceptually shown by a key figure in FIG. 20B) for performing GNSS positioning among a plurality of users of the formed group is exchanged.
- the wearable terminal 100 of the user U1 having the GNSS positioning right performs GNSS positioning
- the wearable terminals 100 of the other users (U2 and U3) not having the GNSS positioning right are GNSS positioning.
- Positioning information is received from the wearable terminal 100 of the user U1 who performs and shares positioning information.
- the received positioning information is used as a group representative value (latitude, longitude information, etc.).
- the above-mentioned GNSS positioning right may be exchanged at predetermined time intervals. Further, when the remaining battery level of each wearable terminal 100 is recognized, the GNSS positioning right may be passed to the wearable terminal 100 having a large remaining battery level. As described above, the GNSS positioning is performed by the wearable terminal 100 having a large remaining battery level, whereby the remaining battery level of the terminals in the group can be smoothed. Further, as shown in FIG. 21, the GNSS positioning right may be exchanged through an operation screen displayed on the user's mobile terminal 200.
- FIG. 21 three users are recognized in the group. And it is shown that “Mr. BBB” currently has the GNSS positioning right. That is, on the display screen, a mark indicating that GNSS positioning right is possessed is displayed in the check box next to Mr. BBB's name.
- the GNSS positioning rights are passed to other users by marking the checkbox that appears next to the name of the user that the user wants to pass GNSS positioning rights. .
- FIG. 22 is a diagram showing a system configuration of a system that performs the above-described operation.
- the system of the present embodiment includes a server 300m and wearable terminals 100m and 100n. Note that the number of wearable terminals 100 is not limited to two.
- the server 300m includes a communication unit 37, a companion recognition unit 38, and a GNSS positioning determination unit 39.
- the communication unit 37 communicates with each wearable terminal 100m, 100n.
- the companion recognition unit 38 groups a plurality of companions based on information sent from the wearable terminals 100m and 100n. Further, the GNSS positioning determination unit 39 determines to which user in the group recognized by the companion recognition unit 38 the GNSS positioning right is given.
- the wearable terminals 100m and 100n include a communication unit 31, a display unit 32, a sensing unit 11, a companion recognition unit 40, a GNSS positioning unit 41, a GNSS control unit 42, and a virtual GNSS positioning unit 43.
- the communication unit 31 communicates with the server 300m.
- the display part 32 displays information, such as information regarding a plurality of users belonging to a group, for example.
- the communication unit 31 is implemented by a communication device such as Bluetooth (registered trademark) or Wi-Fi provided in each of the wearable terminals 100m and 100n.
- the sensing unit 11 includes a microphone, an acceleration sensor, and / or a gyro sensor as described above, and may further include an imaging unit such as a camera.
- the companion recognition unit 40 receives information from the sensing unit 11 and the communication unit 31, and transmits the received information to the companion identification unit 38 of the server 300m via the communication unit 31.
- the accompanying person recognition part 40 receives the information of the accompanying person recognized by the accompanying person recognition part 38 of the server 300m.
- the companion information may be displayed on the display unit 32, and the displayed companion information may be corrected by the user.
- the GNSS positioning unit 41 receives a GNSS signal from a GNSS satellite and performs positioning.
- the virtual GNSS positioning unit 43 determines the position of the own terminal using the positioning information received from another terminal.
- the GNSS control unit 42 switches between turning on the GNSS positioning unit 41 or turning on the virtual GNSS positioning unit 43 based on the GNSS positioning right generated by the GNSS positioning determination unit 39 of the server 300m. Further, as described above with reference to FIG. 21, when the GNSS positioning right is manually changed, the GNSS control unit 42 recognizes the manual change of the GNSS positioning right and turns on the GNSS positioning unit 41 or virtually Switches whether to turn on the GNSS positioning unit 43.
- the companion recognition unit 40 of the wearable terminals 100m and 100n receives the following information from the sensing unit 11, the GNSS control unit 42, or the communication unit 31.
- Positioning information generated by the GNSS positioning unit 41 or the virtual GNSS positioning unit 43 (2) Bluetooth (registered trademark) of other terminals to be received or Wi-Fi terminal identification information (ID) (3) Sound received by the microphone (4) Information of the captured image captured by the camera
- the companion recognition unit 40 of the wearable terminals 100m and 100n transmits the information (1) to (4) described above to the companion recognition unit 38 of the server 300m. Then, the companion recognition unit 38 of the server 300m that has received the information determines the distance between each wearable terminal 100 from the positioning information of (1), for example, and if the distance is equal to or less than a predetermined distance, the user having the wearable terminal 100 is determined. You may recognize as a companion.
- the companion recognition unit 38 of the server 300m may recognize, as the companion, a user who has the wearable terminal 100 of the terminal identification information observed in the long term with respect to the terminal identification information (2). That is, when wearable terminal 100 having terminal identification information A observes wearable terminal 100 having terminal identification information B for a long time, a user having wearable terminal 100 having terminal identification information B is identified as a companion.
- the companion recognition unit 38 of the server 300m may perform environmental sound matching based on the audio information (3) and recognize a wearable terminal user having similar audio information as an accompanying person. Further, the companion recognition unit 38 of the server 300m may recognize a person recognized in the captured image for a predetermined period as an accompanying person based on the image information (4). At this time, person data (for example, face image data) used for image recognition may be stored in each of the wearable terminals 100m and 100n, and the companion recognition unit 40 of the wearable terminals 100m and 100n may store the person data. May be transmitted to the server 300m.
- person data for example, face image data
- the companion recognition unit 38 of the server 300m described above includes a plurality of users (that is, companions) based on actions such as the user's nodding and hand movements (gestures) described in the first embodiment, or speech between the users. ) May be recognized on the basis of the feature amount indicating the interaction.
- the companion may be recognized by integrating the information of (1) to (4) and the interaction between users.
- a recognition method corresponding to the situation of the wearable terminals 100m and 100n is selected. For example, when the camera is activated, information on the captured image of the camera may be used for the recognition of the accompanying person. Also, when the microphone is activated, audio information may be used for the recognition of the accompanying person.
- a companion can be identified with higher accuracy.
- the information on (1) to (4) and the interaction between users can be considered as an example of companion recognition information used for recognizing a companion.
- FIG. 23 is a diagram illustrating a configuration of an example in which a companion is recognized in each wearable terminal 100m and 100n.
- the companion identification unit 40 in FIG. 23 has the function of the companion recognition unit 38 of the server 300m in FIG.
- the GNSS control unit 42 in FIG. 23 has the function of the GNSS positioning determination unit 39 of the server in FIG.
- a similar system may be formed by the mobile terminal 200 having the same configuration.
- FIG. 24 is a block diagram showing the wearable terminal 100 using the vehicle 400 having the GNSS positioning unit 45 and the positioning information measured by the vehicle 400.
- the wearable terminal 100 and the vehicle 400 shown in FIG. 24 are associated by communication by the communication units 31 and 44. This may be done by pairing such as Bluetooth (registered trademark).
- the GNSS control unit 42 of the wearable terminal 100 associated with the vehicle 400 turns off the power of the GNSS positioning unit 41. Then, the GNSS control unit 42 acquires the positioning information measured by the GNSS positioning unit 45 of the vehicle 400 via the communication unit 31. The GNSS control unit 42 turns on the virtual GNSS positioning unit 43 and recognizes the position of the terminal itself using the acquired control information. When the association between wearable terminal 100 and vehicle 400 is released, wearable terminal 100 turns on GNSS positioning unit 41 of wearable terminal 100 and performs positioning by itself.
- the wearable terminal 100 when the wearable terminal 100 is associated with a device having sufficient power, such as the vehicle 400, the wearable terminal 100 uses positioning information measured by a device having sufficient power. As a result, power consumption of the wearable terminal 100 is suppressed.
- FIG. 25 is a flowchart showing the operation of the application example.
- the wearable terminal 100 scans for a nearby terminal using the communication unit 31 such as Bluetooth (registered trademark).
- the communication unit 31 such as Bluetooth (registered trademark).
- near field communication such as Bluetooth Low Energy
- a terminal having a radius of several meters is detected by the communication unit 31 of the wearable terminal 100.
- the wearable terminal 100 determines the number of neighboring terminals scanned in S100.
- the wearable terminal 100 performs intermittent positioning, which will be described in detail later, based on the number of neighboring terminals determined in S102.
- wearable terminal 100 determines whether or not to receive positioning information from another terminal in S108.
- positioning information cannot be obtained from another terminal
- the process proceeds to S112, and the wearable terminal 100 performs GNSS positioning by itself.
- the wearable terminal receives the positioning information from the other terminal in S108
- the process proceeds to S110, and the wearable terminal recognizes the position of the own terminal using the positioning information received from the other terminal. And a process returns to S100 and the process mentioned above is repeated.
- the intermittent rate may be changed according to the number of neighboring terminals determined in S102. For example, it is assumed that the number of proximity terminals determined in S102 is 10, and each of them performs positioning at an intermittent rate of 90%.
- the intermittent rate of 90% means that, for example, the GNSS positioning unit 41 is turned on for 1 second every 10 seconds.
- wearable terminal 100 has a very high probability of receiving positioning information from other terminals at intervals of at least about 3 seconds. Therefore, in the system described above, wearable terminal 100 can obtain positioning information with sufficient accuracy while maintaining an intermittent rate of 90%.
- the wearable terminal 199 can increase the intermittent rate if the number of detected proximity terminals is large, and the wearable terminal 100 decreases the intermittent rate if the number of detected adjacent terminals is small. There must be.
- the GNSS positioning unit 41 is intermittently operated, so that power saving is achieved in the wearable terminal 100.
- the GNSS positioning method that performs intermittent positioning the GNSS positioning may be complemented and executed using past positioning information. At this time, if the past positioning information is too old, there is a possibility that complementation cannot be performed.
- positioning information can be obtained from another terminal even if the intermittent rate is increased, so that the positioning information is appropriately complemented.
- an information processing apparatus for example, an information processing apparatus, a system, an information processing method executed by the information processing apparatus or system, a program for causing the information processing apparatus to function, and a program are recorded. It may include tangible media that is not temporary.
- the companion recognition may be performed using a dedicated application that registers the user as a companion in advance.
- the companion recognition may be performed using a group function in an existing SNS (social network service).
- FIG. 26 is a block diagram illustrating a hardware configuration example of the information processing apparatus according to the embodiment of the present disclosure.
- the information processing apparatus 900 includes a CPU (Central Processing unit) 901, a ROM (Read Only Memory) 903, and a RAM (Random Access Memory) 905.
- the information processing apparatus 900 may include a host bus 907, a bridge 909, an external bus 911, an interface 913, an input device 915, an output device 917, a storage device 919, a drive 921, a connection port 923, and a communication device 925.
- the information processing apparatus 900 may include an imaging device 933 and a sensor 935 as necessary.
- the information processing apparatus 900 may include a processing circuit such as a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array) instead of or in addition to the CPU 901.
- DSP Digital Signal Processor
- ASIC Application Specific Integrated Circuit
- FPGA Field-Programmable Gate Array
- the CPU 901 functions as an arithmetic processing device and a control device, and controls all or a part of the operation in the information processing device 900 according to various programs recorded in the ROM 903, the RAM 905, the storage device 919, or the removable recording medium 927.
- the ROM 903 stores programs and calculation parameters used by the CPU 901.
- the RAM 905 primarily stores programs used in the execution of the CPU 901, parameters that change as appropriate during the execution, and the like.
- the CPU 901, the ROM 903, and the RAM 905 are connected to each other by a host bus 907 configured by an internal bus such as a CPU bus. Further, the host bus 907 is connected to an external bus 911 such as a PCI (Peripheral Component Interconnect / Interface) bus via a bridge 909.
- PCI Peripheral Component Interconnect / Interface
- the input device 915 is a device operated by the user, such as a mouse, a keyboard, a touch panel, a button, a switch, and a lever.
- the input device 915 may be, for example, a remote control device that uses infrared rays or other radio waves, or may be an external connection device 929 such as a mobile phone that supports the operation of the information processing device 900.
- the input device 915 includes an input control circuit that generates an input signal based on information input by the user and outputs the input signal to the CPU 901. The user operates the input device 915 to input various data and instruct processing operations to the information processing device 900.
- the output device 917 is configured by a device capable of notifying the acquired information to the user using a sense such as vision, hearing, or touch.
- the output device 917 can be, for example, a display device such as an LCD (Liquid Crystal Display) or an organic EL (Electro-Luminescence) display, an audio output device such as a speaker or headphones, or a vibrator.
- the output device 917 outputs the result obtained by the processing of the information processing device 900 as video such as text or image, sound such as sound or sound, or vibration.
- the storage device 919 is a data storage device configured as an example of a storage unit of the information processing device 900.
- the storage device 919 includes, for example, a magnetic storage device such as an HDD (Hard Disk Drive), a semiconductor storage device, an optical storage device, or a magneto-optical storage device.
- the storage device 919 stores, for example, programs executed by the CPU 901 and various data, and various data acquired from the outside.
- the drive 921 is a reader / writer for a removable recording medium 927 such as a magnetic disk, an optical disk, a magneto-optical disk, or a semiconductor memory, and is built in or externally attached to the information processing apparatus 900.
- the drive 921 reads information recorded on the attached removable recording medium 927 and outputs the information to the RAM 905.
- the drive 921 writes a record in the attached removable recording medium 927.
- the connection port 923 is a port for connecting a device to the information processing apparatus 900.
- the connection port 923 can be, for example, a USB (Universal Serial Bus) port, an IEEE 1394 port, a SCSI (Small Computer System Interface) port, or the like.
- the connection port 923 may be an RS-232C port, an optical audio terminal, an HDMI (registered trademark) (High-Definition Multimedia Interface) port, or the like.
- the communication device 925 is a communication interface configured with, for example, a communication device for connecting to the communication network 931.
- the communication device 925 can be, for example, a communication card for LAN (Local Area Network), Bluetooth (registered trademark), Wi-Fi, or WUSB (Wireless USB).
- the communication device 925 may be a router for optical communication, a router for ADSL (Asymmetric Digital Subscriber Line), or a modem for various communication.
- the communication device 925 transmits and receives signals and the like using a predetermined protocol such as TCP / IP with the Internet and other communication devices, for example.
- the communication network 931 connected to the communication device 925 is a network connected by wire or wireless, and may include, for example, the Internet, a home LAN, infrared communication, radio wave communication, satellite communication, or the like.
- the imaging device 933 uses various members such as an imaging element such as a CMOS (Complementary Metal Oxide Semiconductor) or a CCD (Charge Coupled Device), and a lens for controlling the formation of a subject image on the imaging element. It is an apparatus that images a real space and generates a captured image.
- the imaging device 933 may capture a still image or may capture a moving image.
- the sensor 935 is various sensors such as an acceleration sensor, an angular velocity sensor, a geomagnetic sensor, an illuminance sensor, a temperature sensor, an atmospheric pressure sensor, or a sound sensor (microphone).
- the sensor 935 acquires information about the state of the information processing apparatus 900 itself, such as the posture of the information processing apparatus 900, and information about the surrounding environment of the information processing apparatus 900, such as brightness and noise around the information processing apparatus 900, for example. To do.
- the sensor 935 may include a GPS receiver that receives a GPS (Global Positioning System) signal and measures the latitude, longitude, and altitude of the device.
- GPS Global Positioning System
- Each component described above may be configured using a general-purpose member, or may be configured by hardware specialized for the function of each component. Such a configuration can be appropriately changed according to the technical level at the time of implementation.
- the following configurations also belong to the technical scope of the present disclosure.
- An information processing apparatus including a communication determination unit that determines whether or not.
- the plurality of users include a first user and a second user, The information according to (1), wherein the feature amount is extracted based on a temporal relationship between the utterance voice of the first user and the utterance voice of the second user included in the voice data. Processing equipment.
- (3) The voice data includes the first voice data including the voice of the first user and the second voice data including the voice of the second user.
- Information processing device includes the first voice data including the voice of the first user and the second voice data including the voice of the second user.
- the information processing apparatus includes single voice data including a voice of the first user and a voice of the second user.
- the information processing apparatus according to any one of (1) to (4), further including a feature amount extraction unit that extracts the feature amount from the audio data.
- the communication determination unit determines whether or not the communication is occurring between users selected from the plurality of users based on the positional relationship between the users, (1) to (5) The information processing apparatus according to any one of the above.
- the communication determination unit determines whether or not the communication is occurring between users selected from the plurality of users based on the behavior information of each user, (1) to (6) The information processing apparatus according to any one of the above.
- the information processing apparatus according to any one of (1) to (7), wherein the feature amount is further extracted from sensor data indicating the motions or biological indices of the plurality of users.
- the plurality of users include a third user and a fourth user, The feature amount is extracted based on a relationship between the speech voice of the third user included in the voice data and the motion or biological index of the fourth user indicated by the sensor data. ).
- the information processing apparatus according to any one of (1) to (9), further including a display control unit configured to display a screen for presenting the communication in time series.
- (11) The information processing apparatus according to (10), wherein the communication is presented on the screen in a form corresponding to the occurrence probability of the communication calculated based on the feature amount.
- a log that outputs a log including at least one of information of a partner communicated by at least one user included in the plurality of users or conversation information with the communicated partner
- the information processing apparatus according to any one of (1) to (11), further including an output unit.
- (14) In any one of (1) to (13), further comprising a relationship graph specifying unit that specifies a graph structure expressing the relationship among the plurality of users based on the occurrence history of the communication. The information processing apparatus described.
- the information processing apparatus further including a shared user specifying unit for specifying.
- the filter includes a positional relationship of users included in the graph structure, a user's affiliation included in the graph structure, a target of interest of the user included in the graph structure, and an action of the user included in the graph structure.
- the information processing apparatus configured based on information or knowledge of a user included in the graph structure.
- the relationship graph specifying unit has generated the graph structure specified based on the occurrence history of the communication in the first period in the second period that is shorter than the first period.
- the information processing apparatus according to any one of (14) to (16), wherein the information is temporarily corrected according to a content of communication.
- the plurality of users include a fifth user and a sixth user;
- the relationship graph specifying unit includes the fifth graph in the graph structure when the name of the sixth user is included in the content transmitted by the fifth user in the communication generated in the second period.
- the information processing apparatus according to (17), wherein the relationship between the user and the sixth user is temporarily strengthened.
- Communication occurs between the plurality of users based on a feature amount indicating an interaction between the plurality of users including the user, which is extracted from voice data including at least the user's speech. An information processing method including determining whether or not the information is being processed.
- the companion recognition information is a feature amount indicating an interaction between the user and the companion, image information captured by a first information processing apparatus possessed by the user, or the first information processing apparatus and the Including any one of the information regarding the distance to the second information processing device or the terminal identification information issued by the first information processing device or the second information processing device or a combination thereof (21 ).
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Computational Linguistics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- General Health & Medical Sciences (AREA)
- Signal Processing (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Computing Systems (AREA)
- Computer Hardware Design (AREA)
- User Interface Of Digital Computer (AREA)
- Telephonic Communication Services (AREA)
Abstract
Description
1.第一の実施形態
1-1.概要およびシステム構成
1-2.会話検出のための処理の例
1-3.応用的な情報の生成例
1-4.第一の実施形態に係る補足
2.第二の実施形態
2-1.概要およびシステム構成
2-2.応用例
2-3.第二の実施形態に係る補足
3.ハードウェア構成
(1-1.概要およびシステム構成)
図1は、本開示の一実施形態における会話の検出について概念的に説明するための図である。図1には、(a)~(c)として、本実施形態において複数のユーザの中から会話が発生しているユーザが特定される過程が示されている。まず、(a)に示すように、対象ユーザとの間で会話が発生しているか否かを判定するために、対象ユーザ以外のユーザが、候補ユーザと他のユーザとに分けられる。候補ユーザは、例えば、GNSSやWi-Fi測位などによって対象ユーザに近い位置にいることが示されるユーザである。他のユーザ、すなわち物理的な制約から会話が発生している可能性が低いと推定されるユーザを以降の検出処理の対象から除外することによって、処理量を削減し、かつ検出の精度を向上させることができる。
図4は、本開示の一実施形態におけるアクションの検出について説明するための図である。図4に示された例では、ウェアラブル端末100cに、ヘッドセット110と、ユーザのうなずきを検出するためのモーションセンサ120とが含まれている。ヘッドセット110は、マイクロフォン112を含み、音声データを取得する。モーションセンサ120は、ジャイロセンサ122および加速度センサ124を含み、角速度および加速度のセンサデータを取得する。ここで、上述したシステム10におけるアクション検出部12は、音声データ抽出されるエネルギーが閾値を超えることを開始条件、エネルギーが閾値を下回った状態が所定の時間以上継続することを終了条件として、ユーザの発話区間を検出することができる。一方、アクション検出部12は、角速度の所定の周波数が大きい区間から加速度の分散の大きい区間を除去することによって、ユーザのうなずきが発生した区間を検出することができる。
・発話の交換(第1のユーザの発話と第2のユーザの発話とが交互に発生)
・発話中の非発話者のうなずき
・発話の短い間における非発話者のうなずき
・両者同時のうなずき
・相手の発話の間における発話+うなずきの応答
・発話区間の一致(第1のユーザの発話と第2のユーザの発話とが同時に発生)
・発話に対する非発話者のリアクションがない
・発話者の発話終了時における非発話者のジェスチャによる反応
・発話に含まれる単語に共通性がある
・発話内容に共通性があり、受け答えが整合している
・発話者と非発話者との身体方向が交差している
・歩行や食事などの動作が共通している
・両者の脈拍の変化に相関性がある
図11は、本開示の一実施形態における、会話の検出によって得られた情報の利用例について説明するための図である。例えば、上記で図3を参照して説明した例では、会話判定部15による会話発生の判定結果が、スコア算出部16およびグルーピング部17によって利用されたが、本実施形態における情報の利用例はこのような例には限られず、他にも様々な利用が可能である。図11には、そのような利用のための機能構成として、UI生成部171~話題推薦部183を示した。これらの機能構成は、端末装置またはサーバにおいて、例えばプロセッサおよびメモリまたはストレージを用いてソフトウェア的に実装される。以下、これらの機能構成によって提供される情報について、さらに説明する。
UI生成部171は、例えば、上記で図6~図8を参照して説明したような、ユーザ間の会話の状態を時系列でグラフ表示するユーザインターフェースを提供してもよい。また、例えば、UI生成部171は、上記のようなリアルタイムの会話の状態をリストで表示するユーザインターフェースを提供してもよい。また、UI生成部171は、そのようにして表示されたリアルタイムの会話の状態を、他のアプリケーションに連携することを可能にしてもよい。例えば、UI生成部171は、ソーシャルメディアへのリンク機能172とも協働して、共通の会話グループに属するユーザの間で画像などのデータの共有を可能にしたり、ゲームを同時にプレイしたりすることをサポートしてもよい。
第2の例では、ログ出力部175、およびソーシャルメディアへのリンク機能172によって、会話した相手の履歴がタイムライン上に出力される。図15には、そのような出力のための機能構成が示されている(なお、図示された例では会話の推定は端末装置で実行されているが、サーバで実行されてもよい)。図15に示された例では、端末装置100zが、センシング部11、アクション検出部12、候補選出部13、特徴量抽出部14、会話判定部15、通信部31、ログ出力部175、およびソーシャルメディアとリンクする機能172を有する。ログ出力部175は、会話判定部15の判定に基づく会話の発生履歴に基づいて、複数のユーザ(処理対象になるすべてのユーザ)に含まれる少なくとも1のユーザが会話した相手の情報、または会話した相手との会話情報の少なくともいずれかを含むログを出力する。ログ出力部175によって生成されたログは、ソーシャルメディアとリンクする機能172を介して、上記少なくとも1のユーザに提示されるソーシャルメディアのタイムライン(他の例では、ソーシャルメディアとは関係のないタイムラインであってもよい)上に出力される。また、サーバ300zでは、音声認識部34およびトピック認識部35がソフトウェア的に実装される。
第3の例では、上記の第2の例において、必ずしも会話を交換していない相手について、例えばソーシャルメディア上でのアクションが可能になる。上述のように、本実施形態において、特徴量抽出部14は、複数のユーザのそれぞれの発話音声同士に関係に基づいて特徴量を抽出するだけではなく、一方のユーザの発話音声と、他方のユーザの発話以外のアクション(例えばモーションまたは生体指標)との時間的な関係に基づいて特徴量を抽出することが可能である。これを利用して、例えば、単発的に会話した人認識部173は、ユーザが発話を交換して会話した他のユーザだけではなく、ユーザの発話に対して何らかのアクションをした他のユーザ、または、ユーザのアクションが向けられた発話をした他のユーザを認識して、ログ出力部175によって提供されるタイムライン上に表示することが可能である。これに基づいて、ユーザは、上記の他のユーザ(多くの場合、知り合いではない)に対して、クラウド上でのアクション174を実施することが可能である。このとき、クラウド上のアクションでは、プライバシーの保護のため、例えば相手のアバターが見える程度で、必ずしも個人情報はやりとりされなくてもよい。
第4の例では、図11に示された話題推薦部183が、話題を提供することによってユーザのコミュニケーションをサポートする。例えば、話題推薦部183は、特徴量抽出部14によって抽出された特徴量によって示される会話のテンポから、会話の盛り上がりの程度を推定し、それに応じて話題を推薦する。より具体的には、話題推薦部183は、会話判定部15によって算出される会話の発生確率が低下傾向であったり、別途推定される会話の盛り上がり度合い(例えばユーザの発話のテンポや、声の大きさによって推定される)が低下傾向であったりする場合に、ユーザが新たな話題を必要としていると判断して、それまでとは異なる話題を推薦してもよい。また、例えば、話題推薦部183は、会話の発生確率が上昇傾向であったり、会話の盛り上がり度合いが上昇傾向で合ったりする場合には、現在の話題が継続されると判断して、現在の話題に関する情報をユーザに提供してもよい。
第5の例では、図11に示された親密度グラフ生成部177が、ログ出力部175によって出力されたログに基づいてユーザ間の親密度を示すグラフを生成する。ここで生成される親密度グラフは、会話の発生履歴に基づいて、複数のユーザの関係性を表現するグラフ構造ともいえる。従って、親密度グラフ生成部177は、そのようなグラフ構造を特定する関係性グラフ特定部ともいえる。親密度グラフ生成部177は、例えば、ログ出力部175のログによって会話の頻度が高かったり、会話の累計時間が長かったりするユーザ同士の間で、親密度グラフにおける強いリンクを生成する。また、親密度グラフ生成部177は、アクション検出部12や特徴量抽出部14によって抽出されたリアクションの多さや種類に基づいて親密度を推定し、親密度が高いユーザ同士の間に強いリンクを生成してもよい。より具体的には、親密度グラフ生成部177は、あるユーザが他のユーザとの会話の中で、よく発話するのか、うなずいているだけなのかによって、親密度グラフにおけるユーザ間のリンクの強さを変えてもよい。また、親密度グラフ生成部177は、例えば外部(ユーザのプロフィールなど)から取得される情報に基づいて、ユーザ間のリンクにラベル(例えば、親、兄弟、上司、同僚、友人、恋人など)を与えてもよい。同様に、親密度グラフ生成部177は、親密度グラフ上で形成されたグループにラベル(例えば、家族、会社、同期、サークル)などを与えてもよい。
第6の例では、図11に示した共有欲求グラフ生成部179が、親密度グラフ生成部177によって生成された親密度グラフに適応グラフ生成部181によって生成されたフィルタを適用することによって、ユーザがコンテンツを共有する範囲を設定する。共有欲求グラフ生成部179は、複数のユーザ(処理対象になるすべてのユーザ)に含まれる少なくとも1のユーザが情報を共有する局面において、ユーザ同士の関係性を表現するグラフ構造に、共有される情報に関連するフィルタを適用することによって情報を共有する他のユーザを特定する共有ユーザ特定部の一例である。近年、ユーザ個人の生活に密着した画像や音声、資料、ライフログなどのコンテンツが大量に取得されているため、そのようなコンテンツの共有範囲を自動的に設定できる、または共有範囲の候補を自動的に絞り込むことができることは有用でありうる。
本開示の実施形態は、例えば、上記で説明したような情報処理装置、システム、情報処理装置またはシステムで実行される情報処理方法、情報処理装置を機能させるためのプログラム、およびプログラムが記録された一時的でない有形の媒体を含みうる。
(2-1.概要およびシステム構成)
以上では、ユーザ間のインタラクションを示す特徴量に基づいて、対象ユーザと候補ユーザとの間で会話が発生しているか否かが判定される実施形態について説明された。以下では、上述された第一の実施形態の応用例である第二の実施形態について説明される。第二の実施形態では、複数のユーザ間において測位情報が授受されるシステムについて説明される。
(1)GNSS測位部41または仮想GNSS測位部43が生成した測位情報
(2)受信される他の端末のBluetooth(登録商標)またはWi-Fiの端末識別情報(ID)
(3)マイクロフォンが受音する音声
(4)カメラが撮像する撮像画像の情報
以上では他の装置が測位した測位情報を用いるシステムの例が説明された。以下では、当該システムの応用例が説明される。当該応用例では、近接に位置する複数の端末間で測位情報が共有される。この応用例は、例えばショッピングモールなどの限られた範囲に多数の端末が密集するような状況で有効である。
次に、図26を参照して、本開示の実施形態に係る情報処理装置のハードウェア構成について説明する。図26は、本開示の実施形態に係る情報処理装置のハードウェア構成例を示すブロック図である。
(1)少なくともユーザの発話音声を含む音声データから抽出される、前記ユーザを含む複数のユーザの間のインタラクションを示す特徴量に基づいて、前記複数のユーザの間でコミュニケーションが発生しているか否かを判定するコミュニケーション判定部を備える、情報処理装置。
(2)前記複数のユーザは、第1のユーザおよび第2のユーザを含み、
前記特徴量は、前記音声データに含まれる前記第1のユーザの発話音声と前記第2のユーザの発話音声との時間的な関係性に基づいて抽出される、前記(1)に記載の情報処理装置。
(3)前記音声データは、前記第1のユーザの発話音声を含む第1の音声データと、前記第2のユーザの発話音声を含む第2の音声データとを含む、前記(2)に記載の情報処理装置。
(4)前記音声データは、前記第1のユーザの発話音声と前記第2のユーザの発話音声とを含む単一の音声データを含む、前記(2)に記載の情報処理装置。
(5)前記音声データから前記特徴量を抽出する特徴量抽出部をさらに備える、前記(1)~(4)のいずれか1項に記載の情報処理装置。
(6)前記コミュニケーション判定部は、前記複数のユーザから各ユーザの位置関係に基づいて選出されたユーザの間で前記コミュニケーションが発生しているか否かを判定する、前記(1)~(5)のいずれか1項に記載の情報処理装置。
(7)前記コミュニケーション判定部は、前記複数のユーザから各ユーザの行動情報に基づいて選出されたユーザの間で前記コミュニケーションが発生しているか否かを判定する、前記(1)~(6)のいずれか1項に記載の情報処理装置。
(8)前記特徴量は、さらに、前記複数のユーザのモーションまたは生体指標を示すセンサデータから抽出される、前記(1)~(7)のいずれか1項に記載の情報処理装置。
(9)前記複数のユーザは、第3のユーザおよび第4のユーザを含み、
前記特徴量は、前記音声データに含まれる前記第3のユーザの発話音声と、前記センサデータによって示される前記第4のユーザのモーションまたは生体指標との関係に基づいて抽出される、前記(8)に記載の情報処理装置。
(10)前記コミュニケーションを時系列で提示する画面を表示させる表示制御部をさらに備える、前記(1)~(9)のいずれか1項に記載の情報処理装置。
(11)前記画面において、前記コミュニケーションは、前記特徴量に基づいて算出される前記コミュニケーションの発生確率に対応した形態で提示される、前記(10)に記載の情報処理装置。
(12)前記コミュニケーションの発生履歴に基づいて、前記複数のユーザに含まれる少なくとも1のユーザがコミュニケートした相手の情報、または前記コミュニケートした相手との会話情報の少なくともいずれかを含むログを出力するログ出力部をさらに備える、前記(1)~(11)のいずれか1項に記載の情報処理装置。
(13)前記ログ出力部は、前記ログを前記少なくとも1のユーザに提示されるタイムライン上に出力する、前記(12)に記載の情報処理装置。
(14)前記コミュニケーションの発生履歴に基づいて、前記複数のユーザの関係性を表現するグラフ構造を特定する関係性グラフ特定部をさらに備える、前記(1)~(13)のいずれか1項に記載の情報処理装置。
(15)前記複数のユーザに含まれる少なくとも1のユーザが情報を共有する局面において、前記グラフ構造に、前記共有される情報に関連するフィルタを適用することによって前記情報を共有する他のユーザを特定する共有ユーザ特定部をさらに備える、前記(14)に記載の情報処理装置。
(16)前記フィルタは、前記グラフ構造に含まれるユーザの位置関係、前記グラフ構造に含まれるユーザの所属先、前記グラフ構造に含まれるユーザの興味の対象、前記グラフ構造に含まれるユーザの行動情報、または前記グラフ構造に含まれるユーザの有するナレッジに基づいて構成される、前記(15)に記載の情報処理装置。
(17)前記関係性グラフ特定部は、第1の期間における前記コミュニケーションの発生履歴に基づいて特定された前記グラフ構造を、前記第1の期間よりも短い直近の第2の期間において発生した前記コミュニケーションの内容に応じて一時的に補正する、前記(14)~(16)のいずれか1項に記載の情報処理装置。
(18)前記複数のユーザは、第5のユーザおよび第6のユーザを含み、
前記関係性グラフ特定部は、前記第2の期間において発生した前記コミュニケーションにおいて前記第5のユーザが発信した内容に前記第6のユーザの名称が含まれる場合に、前記グラフ構造において前記第5のユーザと前記第6のユーザとの関係性を一時的に強化する、前記(17)に記載の情報処理装置。
(19)プロセッサが、少なくともユーザの発話音声を含む音声データから抽出される、前記ユーザを含む複数のユーザの間のインタラクションを示す特徴量に基づいて、前記複数のユーザの間でコミュニケーションが発生しているか否かを判定することを含む、情報処理方法。
(20)少なくともユーザの発話音声を含む音声データから抽出される、前記ユーザを含む複数のユーザの間のインタラクションを示す特徴量に基づいて、前記複数のユーザの間でコミュニケーションが発生しているか否かを判定する機能をコンピュータに実現させるためのプログラム。
(21)
前記ユーザの同伴者を認識するための同伴者認識情報に基づいて前記同伴者を認識する同伴者認識部と、
GNSS測位を行うためのGNSS測位権を前記ユーザが有する第1の情報処理装置に与えるか前記同伴者が有する第2の情報処理装置に与えるかを判定するGNSS測位決定部と、を備える、前記(1)に記載の情報処理装置。
(22)
前記同伴者認識情報は、前記ユーザと前記同伴者との間のインタラクションを示す特徴量、または前記ユーザが有する第1の情報処理装置が撮像する画像情報、または前記第1の情報処理装置と前記第2の情報処理装置との間の距離に関する情報、または前記第1の情報処理装置または前記第2の情報処理装置が発する端末識別情報のいずれか1つまたはこれらの組み合わせを含む、前記(21)に記載の情報処理装置。
(23)
前記第1の情報処理装置および前記第2の情報処理装置の電池残量を認識し、前記電池残量に基づいて前記GNSS測位権を与える情報処理装置を決定する、前記(21)または前記(22)に記載の情報処理装置。
(24)
前記第1の情報処理装置に近接するGNSS測位を行うことが可能な車両が認識された場合、前記車両から測位情報を取得する、前記(21)から前記(23)のいずれか1項に記載の情報処理装置。
(25)
通信部をさらに備え、前記通信部が認識する近接端末の数に応じて、間欠的にGNSS測位を行う頻度を変更する、前記(21)から前記(24)のいずれか1項に記載の情報処理装置。
11 センシング部
12 アクション検出部
13 候補選出部
14 特徴量抽出部
15 会話判定部
16 スコア算出部
17 グルーピング部
38、40 同伴者認識部
39 GNSS測位決定部
100a アイウェア
100b リストウェア
200 モバイル端末
300 サーバ
Claims (25)
- 少なくともユーザの発話音声を含む音声データから抽出される、前記ユーザを含む複数のユーザの間のインタラクションを示す特徴量に基づいて、前記複数のユーザの間でコミュニケーションが発生しているか否かを判定するコミュニケーション判定部を備える、情報処理装置。
- 前記複数のユーザは、第1のユーザおよび第2のユーザを含み、
前記特徴量は、前記音声データに含まれる前記第1のユーザの発話音声と前記第2のユーザの発話音声との時間的な関係性に基づいて抽出される、請求項1に記載の情報処理装置。 - 前記音声データは、前記第1のユーザの発話音声を含む第1の音声データと、前記第2のユーザの発話音声を含む第2の音声データとを含む、請求項2に記載の情報処理装置。
- 前記音声データは、前記第1のユーザの発話音声と前記第2のユーザの発話音声とを含む単一の音声データを含む、請求項2に記載の情報処理装置。
- 前記音声データから前記特徴量を抽出する特徴量抽出部をさらに備える、請求項1に記載の情報処理装置。
- 前記コミュニケーション判定部は、前記複数のユーザから各ユーザの位置関係に基づいて選出されたユーザの間で前記コミュニケーションが発生しているか否かを判定する、請求項1に記載の情報処理装置。
- 前記コミュニケーション判定部は、前記複数のユーザから各ユーザの行動情報に基づいて選出されたユーザの間で前記コミュニケーションが発生しているか否かを判定する、請求項1に記載の情報処理装置。
- 前記特徴量は、さらに、前記複数のユーザのモーションまたは生体指標を示すセンサデータから抽出される、請求項1に記載の情報処理装置。
- 前記複数のユーザは、第3のユーザおよび第4のユーザを含み、
前記特徴量は、前記音声データに含まれる前記第3のユーザの発話音声と、前記センサデータによって示される前記第4のユーザのモーションまたは生体指標との関係に基づいて抽出される、請求項8に記載の情報処理装置。 - 前記コミュニケーションを時系列で提示する画面を表示させる表示制御部をさらに備える、請求項1に記載の情報処理装置。
- 前記画面において、前記コミュニケーションは、前記特徴量に基づいて算出される前記コミュニケーションの発生確率に対応した形態で提示される、請求項10に記載の情報処理装置。
- 前記コミュニケーションの発生履歴に基づいて、前記複数のユーザに含まれる少なくとも1のユーザがコミュニケートした相手の情報、または前記コミュニケートした相手との会話情報の少なくともいずれかを含むログを出力するログ出力部をさらに備える、請求項1に記載の情報処理装置。
- 前記ログ出力部は、前記ログを前記少なくとも1のユーザに提示されるタイムライン上に出力する、請求項12に記載の情報処理装置。
- 前記コミュニケーションの発生履歴に基づいて、前記複数のユーザの関係性を表現するグラフ構造を特定する関係性グラフ特定部をさらに備える、請求項1に記載の情報処理装置。
- 前記複数のユーザに含まれる少なくとも1のユーザが情報を共有する局面において、前記グラフ構造に、前記共有される情報に関連するフィルタを適用することによって前記情報を共有する他のユーザを特定する共有ユーザ特定部をさらに備える、請求項14に記載の情報処理装置。
- 前記フィルタは、前記グラフ構造に含まれるユーザの位置関係、前記グラフ構造に含まれるユーザの所属先、前記グラフ構造に含まれるユーザの興味の対象、前記グラフ構造に含まれるユーザの行動情報、または前記グラフ構造に含まれるユーザの有するナレッジに基づいて構成される、請求項15に記載の情報処理装置。
- 前記関係性グラフ特定部は、第1の期間における前記コミュニケーションの発生履歴に基づいて特定された前記グラフ構造を、前記第1の期間よりも短い直近の第2の期間において発生した前記コミュニケーションの内容に応じて一時的に補正する、請求項14に記載の情報処理装置。
- 前記複数のユーザは、第5のユーザおよび第6のユーザを含み、
前記関係性グラフ特定部は、前記第2の期間において発生した前記コミュニケーションにおいて前記第5のユーザが発信した内容に前記第6のユーザの名称が含まれる場合に、前記グラフ構造において前記第5のユーザと前記第6のユーザとの関係性を一時的に強化する、請求項17に記載の情報処理装置。 - プロセッサが、少なくともユーザの発話音声を含む音声データから抽出される、前記ユーザを含む複数のユーザの間のインタラクションを示す特徴量に基づいて、前記複数のユーザの間でコミュニケーションが発生しているか否かを判定することを含む、情報処理方法。
- 少なくともユーザの発話音声を含む音声データから抽出される、前記ユーザを含む複数のユーザの間のインタラクションを示す特徴量に基づいて、前記複数のユーザの間でコミュニケーションが発生しているか否かを判定する機能をコンピュータに実現させるためのプログラム。
- 前記ユーザの同伴者を認識するための同伴者認識情報に基づいて前記同伴者を認識する同伴者認識部と、
GNSS測位を行うためのGNSS測位権を前記ユーザが有する第1の情報処理装置に与えるか前記同伴者が有する第2の情報処理装置に与えるかを判定するGNSS測位決定部と、を備える、請求項1に記載の情報処理装置。 - 前記同伴者認識情報は、前記ユーザと前記同伴者との間のインタラクションを示す特徴量、または前記ユーザが有する第1の情報処理装置が撮像する画像情報、または前記第1の情報処理装置と前記第2の情報処理装置との間の距離に関する情報、または前記第1の情報処理装置または前記第2の情報処理装置が発する端末識別情報のいずれか1つまたはこれらの組み合わせを含む、請求項21に記載の情報処理装置。
- 前記第1の情報処理装置および前記第2の情報処理装置の電池残量を認識し、前記電池残量に基づいて前記GNSS測位権を与える情報処理装置を決定する、請求項21に記載の情報処理装置。
- 前記第1の情報処理装置に近接するGNSS測位を行うことが可能な車両が認識された場合、前記車両から測位情報を取得する、請求項21に記載の情報処理装置。
- 通信部をさらに備え、前記通信部が認識する近接端末の数に応じて、間欠的にGNSS測位を行う頻度を変更する、請求項21に記載の情報処理装置。
Priority Applications (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201680016943.0A CN107408396A (zh) | 2015-03-27 | 2016-03-09 | 信息处理装置、信息处理方法和程序 |
US15/554,328 US20180040317A1 (en) | 2015-03-27 | 2016-03-09 | Information processing device, information processing method, and program |
EP16772151.3A EP3276622A4 (en) | 2015-03-27 | 2016-03-09 | Information processing device, information processing method, and program |
JP2017509478A JP6756328B2 (ja) | 2015-03-27 | 2016-03-09 | 情報処理装置、情報処理方法、およびプログラム |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2015066901 | 2015-03-27 | ||
JP2015-066901 | 2015-03-27 | ||
PCT/JP2015/085187 WO2016157642A1 (ja) | 2015-03-27 | 2015-12-16 | 情報処理装置、情報処理方法、およびプログラム |
JPPCT/JP2015/085187 | 2015-12-16 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016158267A1 true WO2016158267A1 (ja) | 2016-10-06 |
Family
ID=57004909
Family Applications (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/085187 WO2016157642A1 (ja) | 2015-03-27 | 2015-12-16 | 情報処理装置、情報処理方法、およびプログラム |
PCT/JP2016/057392 WO2016158267A1 (ja) | 2015-03-27 | 2016-03-09 | 情報処理装置、情報処理方法、およびプログラム |
Family Applications Before (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/085187 WO2016157642A1 (ja) | 2015-03-27 | 2015-12-16 | 情報処理装置、情報処理方法、およびプログラム |
Country Status (5)
Country | Link |
---|---|
US (1) | US20180040317A1 (ja) |
EP (1) | EP3276622A4 (ja) |
JP (1) | JP6756328B2 (ja) |
CN (1) | CN107408396A (ja) |
WO (2) | WO2016157642A1 (ja) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2019095609A (ja) * | 2017-11-22 | 2019-06-20 | 株式会社ぐるなび | 情報処理装置、情報処理方法及びプログラム |
JP2019138937A (ja) * | 2018-02-06 | 2019-08-22 | Kddi株式会社 | 情報処理装置、情報処理方法、及びプログラム |
WO2020095714A1 (ja) * | 2018-11-09 | 2020-05-14 | ソニー株式会社 | 情報処理装置および方法、並びに、プログラム |
JP2020086656A (ja) * | 2018-11-19 | 2020-06-04 | トヨタ自動車株式会社 | 情報処理システム、プログラム、及び情報処理方法 |
JP2020091622A (ja) * | 2018-12-05 | 2020-06-11 | Kddi株式会社 | グループ推定装置及びグループ推定方法 |
JP2020166360A (ja) * | 2019-03-28 | 2020-10-08 | 株式会社日本総合研究所 | プログラム及び情報処理装置 |
JP2021064011A (ja) * | 2016-12-30 | 2021-04-22 | グーグル エルエルシーGoogle LLC | 選択的センサポーリング |
JP2022116607A (ja) * | 2021-01-29 | 2022-08-10 | エヌ・ティ・ティ・コミュニケーションズ株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
JP7434451B2 (ja) | 2022-07-28 | 2024-02-20 | エヌ・ティ・ティ・コミュニケーションズ株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
JP7465040B1 (ja) | 2022-12-27 | 2024-04-10 | 株式会社I’mbesideyou | コミュニケーション可視化システム |
WO2024100703A1 (ja) * | 2022-11-07 | 2024-05-16 | マクセル株式会社 | 映像表示装置、映像表示システム及び映像表示装置の制御方法 |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10337962B2 (en) * | 2013-03-15 | 2019-07-02 | Fluke Corporation | Visible audiovisual annotation of infrared images using a separate wireless mobile device |
JP6468258B2 (ja) * | 2016-08-01 | 2019-02-13 | トヨタ自動車株式会社 | 音声対話装置および音声対話方法 |
JP2019106054A (ja) * | 2017-12-13 | 2019-06-27 | 株式会社東芝 | 対話システム |
EP3859568A4 (en) * | 2018-09-28 | 2021-09-29 | Fujitsu Limited | DIALOGUE DEVICE, DIALOGUE PROCESS AND DIALOGUE PROGRAM |
JP7528638B2 (ja) | 2020-08-26 | 2024-08-06 | トヨタ自動車株式会社 | コミュニケーションシステム |
US20220189503A1 (en) * | 2020-12-14 | 2022-06-16 | Liine, LLC | Methods, systems, and computer program products for determining when two people are talking in an audio recording |
WO2022137507A1 (en) | 2020-12-25 | 2022-06-30 | Nec Corporation | Speaker identificantion apparatus, method, and program |
JP6924912B1 (ja) * | 2021-01-29 | 2021-08-25 | Kddi株式会社 | 情報処理装置及び情報処理方法 |
KR102616058B1 (ko) * | 2022-04-06 | 2023-12-21 | 네이버 주식회사 | 음성 기록을 시각화하여 재연하는 방법, 컴퓨터 장치, 및 컴퓨터 프로그램 |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003228449A (ja) * | 2002-02-01 | 2003-08-15 | Sharp Corp | 対話装置及び対話処理プログラムを記録した記録媒体 |
WO2007105436A1 (ja) * | 2006-02-28 | 2007-09-20 | Matsushita Electric Industrial Co., Ltd. | ウェアラブル端末 |
JP2008242318A (ja) * | 2007-03-28 | 2008-10-09 | Toshiba Corp | 対話を検出する装置、方法およびプログラム |
JP2012133735A (ja) * | 2010-12-24 | 2012-07-12 | Kddi Corp | ソーシャルグラフ更新システム、ソーシャルグラフ更新方法、およびプログラム |
JP2012530321A (ja) * | 2009-06-16 | 2012-11-29 | フェイスブック,インク. | ソーシャルネットワークにおける選択的コンテンツアクセス性 |
JP2013058221A (ja) * | 2012-10-18 | 2013-03-28 | Hitachi Ltd | 会議分析システム |
JP2013140534A (ja) * | 2012-01-06 | 2013-07-18 | Fuji Xerox Co Ltd | 音声解析装置、音声解析システムおよびプログラム |
JP2013164468A (ja) * | 2012-02-09 | 2013-08-22 | Fuji Xerox Co Ltd | 音声解析装置、音声解析システムおよびプログラム |
Family Cites Families (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
DE4227826C2 (de) * | 1991-08-23 | 1999-07-22 | Hitachi Ltd | Digitales Verarbeitungsgerät für akustische Signale |
JP2002073072A (ja) * | 2000-08-31 | 2002-03-12 | Sony Corp | モデル適応装置およびモデル適応方法、記録媒体、並びにパターン認識装置 |
JP2002108396A (ja) * | 2000-09-29 | 2002-04-10 | Matsushita Electric Ind Co Ltd | 音声通信方法 |
US6782363B2 (en) * | 2001-05-04 | 2004-08-24 | Lucent Technologies Inc. | Method and apparatus for performing real-time endpoint detection in automatic speech recognition |
US20030158735A1 (en) * | 2002-02-15 | 2003-08-21 | Canon Kabushiki Kaisha | Information processing apparatus and method with speech synthesis function |
WO2004021601A1 (en) * | 2002-08-27 | 2004-03-11 | Closer Communications Llc | Wireless information retrieval and content dissemination system and method |
US7428000B2 (en) * | 2003-06-26 | 2008-09-23 | Microsoft Corp. | System and method for distributed meetings |
US7933226B2 (en) * | 2003-10-22 | 2011-04-26 | Palo Alto Research Center Incorporated | System and method for providing communication channels that each comprise at least one property dynamically changeable during social interactions |
WO2005114576A1 (ja) * | 2004-05-21 | 2005-12-01 | Asahi Kasei Kabushiki Kaisha | 動作内容判定装置 |
US7730411B2 (en) * | 2007-02-01 | 2010-06-01 | Cisco Technology, Inc. | Re-creating meeting context |
JP2008262046A (ja) * | 2007-04-12 | 2008-10-30 | Hitachi Ltd | 会議可視化システム、会議可視化方法、及び集計処理サーバ |
US9491573B2 (en) * | 2008-11-06 | 2016-11-08 | Texas Instruments Incorporated | Communication device for providing value-added information based upon content and/or context information |
JP5949550B2 (ja) * | 2010-09-17 | 2016-07-06 | 日本電気株式会社 | 音声認識装置、音声認識方法、及びプログラム |
JP2012155374A (ja) * | 2011-01-21 | 2012-08-16 | Sony Corp | 情報処理装置、情報処理方法、プログラムおよび情報処理システム |
US8990416B2 (en) * | 2011-05-06 | 2015-03-24 | Oracle International Corporation | Support for a new insert stream (ISTREAM) operation in complex event processing (CEP) |
JP5772448B2 (ja) * | 2011-09-27 | 2015-09-02 | 富士ゼロックス株式会社 | 音声解析システムおよび音声解析装置 |
JP2013135325A (ja) * | 2011-12-26 | 2013-07-08 | Fuji Xerox Co Ltd | 音声解析装置 |
CN103888423B (zh) * | 2012-12-20 | 2019-01-15 | 联想(北京)有限公司 | 信息处理方法和信息处理设备 |
US20130290427A1 (en) * | 2013-03-04 | 2013-10-31 | Hello Inc. | Wearable device with unique user ID and telemetry system in communication with one or more social networks |
JP6464449B2 (ja) * | 2014-08-29 | 2019-02-06 | 本田技研工業株式会社 | 音源分離装置、及び音源分離方法 |
US10120907B2 (en) * | 2014-09-24 | 2018-11-06 | Oracle International Corporation | Scaling event processing using distributed flows and map-reduce operations |
-
2015
- 2015-12-16 WO PCT/JP2015/085187 patent/WO2016157642A1/ja active Application Filing
-
2016
- 2016-03-09 JP JP2017509478A patent/JP6756328B2/ja not_active Expired - Fee Related
- 2016-03-09 EP EP16772151.3A patent/EP3276622A4/en not_active Ceased
- 2016-03-09 WO PCT/JP2016/057392 patent/WO2016158267A1/ja active Application Filing
- 2016-03-09 US US15/554,328 patent/US20180040317A1/en not_active Abandoned
- 2016-03-09 CN CN201680016943.0A patent/CN107408396A/zh active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2003228449A (ja) * | 2002-02-01 | 2003-08-15 | Sharp Corp | 対話装置及び対話処理プログラムを記録した記録媒体 |
WO2007105436A1 (ja) * | 2006-02-28 | 2007-09-20 | Matsushita Electric Industrial Co., Ltd. | ウェアラブル端末 |
JP2008242318A (ja) * | 2007-03-28 | 2008-10-09 | Toshiba Corp | 対話を検出する装置、方法およびプログラム |
JP2012530321A (ja) * | 2009-06-16 | 2012-11-29 | フェイスブック,インク. | ソーシャルネットワークにおける選択的コンテンツアクセス性 |
JP2012133735A (ja) * | 2010-12-24 | 2012-07-12 | Kddi Corp | ソーシャルグラフ更新システム、ソーシャルグラフ更新方法、およびプログラム |
JP2013140534A (ja) * | 2012-01-06 | 2013-07-18 | Fuji Xerox Co Ltd | 音声解析装置、音声解析システムおよびプログラム |
JP2013164468A (ja) * | 2012-02-09 | 2013-08-22 | Fuji Xerox Co Ltd | 音声解析装置、音声解析システムおよびプログラム |
JP2013058221A (ja) * | 2012-10-18 | 2013-03-28 | Hitachi Ltd | 会議分析システム |
Non-Patent Citations (2)
Title |
---|
MASAYUKI OKAMOTO ET AL.: "Adhoc-Meeting Detection by Cross Correlation of Terminal Voice", THE INSTITUTE OF ELECTRONICS, INFORMATION AND COMMUNICATION ENGINEERS DAI 19 KAI PROCEEDINGS OF DATA ENGINEERING WORKSHOP, 7 April 2008 (2008-04-07), pages 1 - 8, XP009506481 * |
See also references of EP3276622A4 * |
Cited By (21)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP7136941B2 (ja) | 2016-12-30 | 2022-09-13 | グーグル エルエルシー | 選択的センサポーリング |
JP2021064011A (ja) * | 2016-12-30 | 2021-04-22 | グーグル エルエルシーGoogle LLC | 選択的センサポーリング |
JP2019095609A (ja) * | 2017-11-22 | 2019-06-20 | 株式会社ぐるなび | 情報処理装置、情報処理方法及びプログラム |
JP7006176B2 (ja) | 2017-11-22 | 2022-01-24 | 株式会社ぐるなび | 情報処理装置、情報処理方法及びプログラム |
JP2019138937A (ja) * | 2018-02-06 | 2019-08-22 | Kddi株式会社 | 情報処理装置、情報処理方法、及びプログラム |
WO2020095714A1 (ja) * | 2018-11-09 | 2020-05-14 | ソニー株式会社 | 情報処理装置および方法、並びに、プログラム |
US11553158B2 (en) | 2018-11-09 | 2023-01-10 | Sony Group Corporation | Information processing apparatus, information processing method, and program |
JP7415940B2 (ja) | 2018-11-09 | 2024-01-17 | ソニーグループ株式会社 | 情報処理装置および方法、並びに、プログラム |
JPWO2020095714A1 (ja) * | 2018-11-09 | 2021-10-07 | ソニーグループ株式会社 | 情報処理装置および方法、並びに、プログラム |
JP2020086656A (ja) * | 2018-11-19 | 2020-06-04 | トヨタ自動車株式会社 | 情報処理システム、プログラム、及び情報処理方法 |
JP7155927B2 (ja) | 2018-11-19 | 2022-10-19 | トヨタ自動車株式会社 | 情報処理システム、プログラム、及び情報処理方法 |
JP7041046B2 (ja) | 2018-12-05 | 2022-03-23 | Kddi株式会社 | グループ推定装置及びグループ推定方法 |
JP2020091622A (ja) * | 2018-12-05 | 2020-06-11 | Kddi株式会社 | グループ推定装置及びグループ推定方法 |
JP7290977B2 (ja) | 2019-03-28 | 2023-06-14 | 株式会社日本総合研究所 | プログラム及び情報処理装置 |
JP2020166360A (ja) * | 2019-03-28 | 2020-10-08 | 株式会社日本総合研究所 | プログラム及び情報処理装置 |
JP7225276B2 (ja) | 2021-01-29 | 2023-02-20 | エヌ・ティ・ティ・コミュニケーションズ株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
JP2022116607A (ja) * | 2021-01-29 | 2022-08-10 | エヌ・ティ・ティ・コミュニケーションズ株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
JP7434451B2 (ja) | 2022-07-28 | 2024-02-20 | エヌ・ティ・ティ・コミュニケーションズ株式会社 | 情報処理装置、情報処理方法及び情報処理プログラム |
WO2024100703A1 (ja) * | 2022-11-07 | 2024-05-16 | マクセル株式会社 | 映像表示装置、映像表示システム及び映像表示装置の制御方法 |
JP7465040B1 (ja) | 2022-12-27 | 2024-04-10 | 株式会社I’mbesideyou | コミュニケーション可視化システム |
WO2024142291A1 (ja) * | 2022-12-27 | 2024-07-04 | 株式会社I’mbesideyou | コミュニケーション可視化システム |
Also Published As
Publication number | Publication date |
---|---|
US20180040317A1 (en) | 2018-02-08 |
CN107408396A (zh) | 2017-11-28 |
JP6756328B2 (ja) | 2020-09-16 |
EP3276622A1 (en) | 2018-01-31 |
JPWO2016158267A1 (ja) | 2018-01-25 |
EP3276622A4 (en) | 2018-10-03 |
WO2016157642A1 (ja) | 2016-10-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2016158267A1 (ja) | 情報処理装置、情報処理方法、およびプログラム | |
US20220377467A1 (en) | Hearing aid systems and mehods | |
US10853650B2 (en) | Information processing apparatus, information processing method, and program | |
CN107850443B (zh) | 信息处理装置、信息处理方法以及程序 | |
JP6729571B2 (ja) | 情報処理装置、情報処理方法及びプログラム | |
EP2727104B1 (en) | Identifying people that are proximate to a mobile device user via social graphs, speech models, and user context | |
EP3418967A1 (en) | Method relating to presence granularity with augmented reality | |
CN111433766B (zh) | 用于对时间序列数据进行分类的方法和系统 | |
JP6311478B2 (ja) | 情報処理装置、情報処理方法およびプログラム | |
JP2016006611A (ja) | 情報処理装置、情報処理方法およびプログラム | |
JPWO2019116658A1 (ja) | 情報処理装置、情報処理方法、およびプログラム | |
US9977510B1 (en) | Gesture-driven introduction system | |
WO2016136104A1 (ja) | 情報処理装置、情報処理方法及びプログラム | |
WO2015190141A1 (ja) | 情報処理装置、情報処理方法、およびプログラム | |
EP4080388A1 (en) | Multimodal, dynamic, privacy preserving age and attribute estimation and learning methods and systems | |
JP6733662B2 (ja) | 情報処理装置、情報処理方法及びコンピュータプログラム | |
US11217253B2 (en) | Information processing apparatus, information processing method, and information processing system | |
WO2016189909A1 (ja) | 情報処理装置、情報処理方法、及びプログラム | |
WO2015194270A1 (ja) | 情報処理装置、情報処理方法およびプログラム | |
Yu et al. | Group Behavior Recognition | |
WO2016147693A1 (ja) | 情報処理装置、情報処理方法およびプログラム | |
WO2015198672A1 (ja) | 情報処理装置、情報処理方法およびプログラム |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16772151 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2017509478 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15554328 Country of ref document: US |
|
REEP | Request for entry into the european phase |
Ref document number: 2016772151 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |