JP2019101805A - Dialogue system - Google Patents

Dialogue system Download PDF

Info

Publication number
JP2019101805A
JP2019101805A JP2017232649A JP2017232649A JP2019101805A JP 2019101805 A JP2019101805 A JP 2019101805A JP 2017232649 A JP2017232649 A JP 2017232649A JP 2017232649 A JP2017232649 A JP 2017232649A JP 2019101805 A JP2019101805 A JP 2019101805A
Authority
JP
Japan
Prior art keywords
passenger
topic
category
determination unit
utterance
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP2017232649A
Other languages
Japanese (ja)
Inventor
滋樹 青島
Shigeki Aoshima
滋樹 青島
慧悟 堀
Keigo Hori
慧悟 堀
恭平 増井
Kyohei Masui
恭平 増井
Original Assignee
トヨタ自動車株式会社
Toyota Motor Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by トヨタ自動車株式会社, Toyota Motor Corp filed Critical トヨタ自動車株式会社
Priority to JP2017232649A priority Critical patent/JP2019101805A/en
Publication of JP2019101805A publication Critical patent/JP2019101805A/en
Pending legal-status Critical Current

Links

Images

Abstract

An interactive system with improved service quality mounted on a vehicle or the like is provided. A dialogue system includes a passenger determination unit that determines whether there is a passenger who is mounted on a vehicle and is seated in a seat other than a driver's seat, and topic information for determining utterance contents. When the topic storage unit that classifies and stores at least the first category that is permitted to be used and the second category that is not permitted to be used and the passenger determination unit determine that there is no passenger, the topic storage unit When the topic information belonging to the first category or the second category is acquired, the utterance content is determined using the acquired topic information, and the passenger determination unit determines that there is a passenger, the first item is stored in the topic storage unit. An utterance content determination unit that acquires topic information belonging to a category and determines the utterance content using the acquired topic information. [Selection] Figure 1

Description

  The present invention relates to a dialog system mounted on a vehicle or the like.
  Various dialogue systems have been proposed which interact with the user in response to the user's voice. For example, Patent Document 1 realizes natural dialogue by estimating emotions and tones and texts from the user's voice, estimating emotions from the user's facial expressions captured by a camera, and generating a response voice based on these. Dialog system is disclosed. Further, Patent Document 2 discloses a dialogue system for activating communication by outputting an empathic utterance or an egocentric utterance based on a predetermined dialogue control rule and an utterance generation rule.
JP, 2006-71936, A JP, 2010-186237, A
  By mounting a dialog system having advanced uttered content generation capability on a vehicle, it can be expected to improve service to users by using it for vehicle operation or imparting an entertainment property to the dialog content. With the improvement of services, it is possible that information related to privacy and confidentiality, such as personal information of the driver who is the user, will be included in the dialogue contents. However, there may be other passengers in the vehicle, and it is not safe for the dialogue system to utter content including privacy and confidential information, which may impair security or affect the mood of the user or the passenger. There is a risk that the service quality may be undesirable.
  The present invention is made in view of the above-mentioned subject, and an object of the present invention is to provide a dialogue system which is carried in vehicles etc. and which raised service quality.
  In order to solve the above-mentioned subject, one aspect of the present invention is a dialogue system carried in vehicles. The dialogue system permits use of the passenger determination unit that determines the presence or absence of a passenger who is a passenger seated in a seat other than the driver's seat, and topic information for determining the content of the utterance when the passenger is present If the topic storage unit that classifies and stores at least one category and the second category for which use is not permitted, and the passenger determination unit determines that there is no passenger, the topic storage unit belongs to the first category or the second category The topic information is acquired, the utterance content is determined using the acquired topic information, and when the passenger determination unit determines that a passenger is present, the topic information belonging to the first category is acquired from the topic storage unit, And an utterance content determination unit that determines the utterance content using the acquired topic information.
  As a result, when there is a passenger, the topic category is limited, and for example, it is possible to suppress the utterance of content relating to privacy and confidentiality, and to improve the service quality.
  As described above, according to the present invention, when there is a passenger, it is possible to suppress the utterance of content relating to privacy and confidentiality, and to provide a dialogue system with an improved quality of service mounted on a vehicle or the like. .
Functional block diagram of dialogue system according to one embodiment of the present invention Flowchart of utterance content determination processing according to an embodiment of the present invention
(Overview)
The dialogue system according to the present invention classifies and stores topic information at least into a first category which is permitted to be used when a passenger is present, and a second category which is not permitted to be used, for example, in relation to privacy and confidentiality. If it is determined that a passenger is present, the speech content is determined using the topic information of the first category, so that topics related to privacy and confidentiality are suppressed.
(Embodiment)
Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.
<Configuration>
FIG. 1 shows a functional block diagram of the dialogue system 1 in the present embodiment. The dialogue system 1 is mounted on a vehicle, and includes a voice input unit 10, a voice recognition unit 20, a passenger determination unit 30, an utterance content determination unit 40, a topic storage unit 50, a voice synthesis unit 60, and a voice output unit 70.
  The voice input unit 10 receives voice input from the user by collecting it with a microphone. The speech recognition unit 20 performs meaning estimation by sequentially analyzing the phonemes, words, and sentence levels of the speech input received by the speech input unit 10.
  The passenger determination unit 30 acquires the weight of the passenger detected by the seat sensor of the vehicle, the photographed image of the passenger detected by the camera, and the like, and detects the presence or absence of the passenger. The method is not limited as long as the presence or absence of a passenger can be detected.
  The utterance content determination unit 40 determines the utterance content based on the meaning estimation result by the speech recognition unit 20 and the detection result of the presence or absence of the passenger by the passenger determination unit 30. Further, topic information is stored in the topic storage unit 50, and the utterance content determination unit 40 refers to the topic information stored in the topic storage unit 50, and determines the utterance content also based on the topic information. Basically, such speech content determination methods can adopt algorithms of various known dialogue systems. The topic information includes, for example, a pattern dictionary such as a combination of general nouns collected and generated from the Internet and the like, frequently associated modifiers and predicate dependencies. In this embodiment, in addition to this, the user's statement history, operation history of the car navigation system, driving history, other vehicle devices linked with the dialogue system 1, billing information, mail, schedule, etc. And a user-specific pattern dictionary extracted and learned from them. The topic information is classified into at least a first category that may be used when there is a passenger and a second category that is prohibited to use when there is a passenger, and the topic storage unit 50 It is memorized.
  The speech synthesis unit 60 generates a speech based on the utterance content determined by the utterance content determination unit 40. The voice output unit 70 outputs the uttered voice generated by the voice synthesis unit 60 from the speaker.
(processing)
In the present embodiment, the utterance content determination unit 40 is characterized in the usage method of the topic information stored in the topic storage unit 50. The utterance content determination process will be described below focusing on this point. The following process may be performed, for example, when the speech input unit 10 receives a speech input of the user and the speech recognition unit 20 performs the meaning estimation, or any process without the speech input of the user. The utterance content determination unit 40 may perform at the timing.
  Step S101: The utterance content determination unit 40 acquires the determination result by the passenger determination unit 30.
  Step S102: If the determination result of the passenger determination unit 30 is "with passenger", the process proceeds to step S103. If it is "with no passenger", the process proceeds to step S104.
  Step S103: The utterance content determination unit 40 selects and acquires a topic from the topic information of the first category from the topic storage unit 50, and determines the utterance content based on this. Thereafter, the process proceeds to step S105.
  Step S104: The utterance content determination unit 40 selects and acquires a topic from the topic information of the first category and the second category from the topic storage unit 50, and determines the utterance content based on this. Thereafter, the process proceeds to step S105.
  Step S105: The voice synthesis unit 60 generates a voice based on the utterance content determined by the utterance content determination unit 40, and the voice output unit 70 outputs the generated voice. As described above, one utterance content determination and utterance processing are ended, but by repeating this processing as needed, it is possible to interact with the user.
  By the above processing, for example, when the user is alone in a vehicle, the dialogue system 1 utters "It is her birthday soon. What would you like to give it?", And if there is a passenger, Processing such as saying "Where are you going from now?" Is possible.
  The topic information stored in the topic storage unit 50 may be classified into a first category that may be used when a passenger is present and a second category that is prohibited to use when a passenger is present. Can be specified by voice input such as "A's a lover, but it's a secret" while the authenticated user is in conversation. Alternatively, the user-specific patterns that have been learned are classified into the second category in the initial state, and voice input from an authenticated user such as "A's a lover, I can tell everyone" is the first It may be classified into categories.
  In addition, the utterance content determination unit 40 acquires information such as engine status, vehicle position, input content to the car navigation system, route information, time, weather, schedule, etc. from inside and outside the vehicle while the vehicle is stopped or traveling. The timing at which these are updated may be used as a trigger for the utterance content, or the situation may be estimated based on these and may be used to determine the utterance content. The topic storage unit 50 may also classify topic information into categories subdivided according to the information. This makes it possible to speak according to the driving situation.
  Further, the passenger determination unit 30 may be able to learn and specify the passenger based on not only the determination of the presence or absence of the passenger but also voice authentication, face authentication, a weight value, and the like. Further, the topic storage unit 50 may generate and categorize topic information based on an action pattern such as a passenger's boarding history. In this case, only the topic related to the passenger may be topic information belonging to the second category, which may be used to determine the content of the utterance. As a result, for example, the passenger gets his / her schedule and the home position, and the utterance is "two people are on a business trip from now, thank you very much!" It becomes possible.
  As described above, according to the dialogue system according to the present invention, when there is a passenger in a vehicle or the like, it is possible to suppress voice related to privacy and confidentiality and to improve service quality. Alternatively, it is possible to perform various dialogues corresponding to the driving situation and the passenger in detail.
  The present invention is useful for a dialog system mounted on a vehicle or the like.
DESCRIPTION OF SYMBOLS 1 dialogue system 10 speech input unit 20 speech recognition unit 30 companion determination unit 40 utterance content determination unit 50 topic storage unit 60 speech synthesis unit 70 speech output unit

Claims (1)

  1. A dialog system mounted on a vehicle,
    A passenger determination unit that determines the presence or absence of a passenger who is a passenger seated in a seat other than the driver's seat;
    A topic storage unit that classifies and stores topic information for determining the content of an utterance at least into a first category that is permitted to be used when a passenger is present and a second category that is not permitted to be used;
    When the passenger determination unit determines that no passenger is present, the topic information belonging to the first category or the second category is acquired from the topic storage unit, and the utterance content is acquired using the acquired topic information Decide
    When the passenger determination unit determines that there is a passenger, the topic content belonging to the first category is acquired from the topic storage unit, and the utterance content determination to determine the utterance content using the acquired topic information A dialogue system comprising a department.
JP2017232649A 2017-12-04 2017-12-04 Dialogue system Pending JP2019101805A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2017232649A JP2019101805A (en) 2017-12-04 2017-12-04 Dialogue system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2017232649A JP2019101805A (en) 2017-12-04 2017-12-04 Dialogue system

Publications (1)

Publication Number Publication Date
JP2019101805A true JP2019101805A (en) 2019-06-24

Family

ID=66973800

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2017232649A Pending JP2019101805A (en) 2017-12-04 2017-12-04 Dialogue system

Country Status (1)

Country Link
JP (1) JP2019101805A (en)

Similar Documents

Publication Publication Date Title
US20160379643A1 (en) Group Status Determining Device and Group Status Determining Method
US10176806B2 (en) Motor vehicle operating device with a correction strategy for voice recognition
JP2010128099A (en) In-vehicle voice information providing system
JP6466385B2 (en) Service providing apparatus, service providing method, and service providing program
KR100201256B1 (en) Starting control system using voice for a vehicle
JP2019191828A (en) Information provision system and information provision method
JP2017009825A (en) Conversation state analyzing device and conversation state analyzing method
US10592997B2 (en) Decision making support device and decision making support method
US20190355352A1 (en) Voice and conversation recognition system
US20190279629A1 (en) Speech system
CN108780644A (en) The system and method for means of transport, speech pause length for adjusting permission in voice input range
US8452599B2 (en) Method and system for extracting messages
CN103390406B (en) Speaker validation method, the preparation method of speaker verification and electronic installation
JP2019101805A (en) Dialogue system
US11107464B2 (en) Feeling estimation device, feeling estimation method, and storage medium
DE102017213260A1 (en) Method, device, mobile user device, computer program for controlling an audio system of a vehicle
JP2019101472A (en) Emotion estimation device
JP2019123259A (en) Vehicle control device, vehicle control method, and vehicle control program
JP6332072B2 (en) Dialogue device
JP2018027731A (en) On-vehicle device, control method of on-vehicle device, and content providing system
JP2018055155A (en) Voice interactive device and voice interactive method
JP2019053785A (en) Service providing device
JP4425718B2 (en) Voice recognition device for vehicles
Kono et al. Prototype of conversation support system for activating group conversation in the vehicle
Lee et al. Understanding and designing for deaf or hard of hearing drivers on Uber