Intelligent device, control method of intelligent device and storage medium

By detecting and correcting the occupants' catchphrases and dialects, the agent device solves the problem of insufficient service development in the existing technology, and improves the effectiveness of the agent service and the occupant experience.

CN112241628BInactive Publication Date: 2024-11-05HONDA MOTOR CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202010683589.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-07-18
Filing Date
2020-07-15
Publication Date
2024-11-05
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

When existing agent technology provides driving support and dialogue services, it is difficult to effectively correct the occupants' mantras and dialects, which limits the development of the service.

Method used

The agent device detects the occupant's speech method and dialect, registers and corrects the mantras and dialects with high frequency, provides correction information, or guides the dialect according to the occupant's designated guidance, to achieve a more developed utilization plan.

Benefits of technology

By correcting the occupants' catchphrases and dialects, the effectiveness of the agent service and the occupant experience are improved, and a more developed utilization solution is provided.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112241628B_ABST
    Figure CN112241628B_ABST
Patent Text Reader

Abstract

An intelligent body device, a control method of the intelligent body device, and a storage medium that can provide a more developmental utilization solution are provided. An intelligent body device provides a service including a response based on sound according to a user's speech, wherein the intelligent body device comprises: a detection unit that detects the user's speaking manner when speaking; and an information providing unit that, when the detection unit detects that the user's speech includes a predetermined speaking manner, provides the user with information for correcting the predetermined speaking manner.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to an intelligent body device, a control method of the intelligent body device and a storage medium. Background Art

[0002] In the past, technologies related to intelligent body functions that communicate with vehicle occupants while providing driving support-related information corresponding to the occupants' requests, vehicle control, other applications, etc. have been disclosed (for example, refer to Patent Document 1 (Japanese Patent Publication No. 2006-335231)). Summary of the invention

[0003] Problems to be solved by the invention

[0004] However, in the conventional technology, the schemes used are limited.

[0005] One of the objects of the present invention is to provide an intelligent body device, a control method of the intelligent body device, and a storage medium that can provide a more expansive utilization solution.

[0006] Means for solving problems

[0007] The intelligent body device, the control method of the intelligent body device and the storage medium of the present invention adopt the following structure.

[0008] (1): An intelligent body device of one embodiment of the present invention provides a service including a sound-based response based on a user's speech, wherein the intelligent body device comprises: a detection unit that detects the user's speaking manner when speaking; and an information providing unit that provides the user with information for correcting the prescribed speaking manner when the detection unit detects that the user's speech includes a prescribed speaking manner.

[0009] (2): In the scheme of (1) above, the intelligent body device also includes a catchphrase registration unit for registering catchphrases of the user detected by the detection unit when speaking, and when the detection unit detects that the frequency of the catchphrases registered by the catchphrase registration unit is above a threshold, the information providing unit provides the user with information for correcting the catchphrases of the user detected at a frequency above the threshold as the prescribed speaking manner.

[0010] (3): In the above-mentioned scheme (1) or (2), the intelligent body device also has a dialect registration unit for registering the dialect of the vehicle occupant when speaking detected by the detection unit, and the information providing unit provides the user with information for correcting the prescribed dialect as the prescribed way of speaking when the detection unit detects the prescribed dialect registered by the dialect registration unit.

[0011] (4): Another embodiment of the present invention is an intelligent body device that provides a service including a sound-based response based on the speech of a user, wherein the intelligent body device comprises: a dialect designation receiving unit that receives the user's indication of the designation of a dialect; and an information providing unit that provides the user with information for guiding the user to speak in a manner that makes the user's speaking style close to the dialect received by the dialect designation receiving unit.

[0012] (5): In another embodiment of the present invention, in a control method of an intelligent body device, a computer is caused to perform the following processing: providing a service including a sound-based response based on a user's speech; detecting the user's speaking manner when speaking; and when it is detected that a prescribed speaking manner is included in the user's speech, providing the user with information for correcting the prescribed speaking manner.

[0013] (6): A storage medium of another embodiment of the present invention stores a program that causes a computer to perform the following processing: providing a service including a sound-based response based on a user's speech; detecting the user's speaking manner when speaking; and when it is detected that a specified speaking manner is included in the user's speech, providing the user with information for correcting the specified speaking manner.

[0014] Effects of the Invention

[0015] According to (1) to (6), a more promising utilization scheme can be provided. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a diagram showing a structure of an agent system 1 including an agent device 100.

[0017] Figure 2 1 is a diagram showing the structure of the intelligent body device 100 and devices mounted on the vehicle M according to the first embodiment.

[0018] Figure 3 This is a diagram showing an example of the content of data registered in the speaking pattern DB 205 .

[0019] Figure 4 2 is a diagram showing a part of the structure of the agent server 200 and the structure of the agent device 100 .

[0020] Figure 5 This is a flowchart for explaining the flow of a series of processes of the intelligent body device 100 according to the first embodiment.

[0021] Figure 6 It is a diagram for explaining the operation of the intelligent body device 100 according to the first embodiment.

[0022] Figure 71 is a diagram showing the structure of the intelligent body device 100 and devices mounted on the vehicle M according to the second embodiment.

[0023] Figure 8 This is a flowchart for explaining the flow of a series of processes of the intelligent body device 100 according to the second embodiment.

[0024] Fig. 9 It is a diagram for explaining the operation of the intelligent body device 100 according to the second embodiment.

[0025] Fig.10 1 is a diagram showing the structure of an intelligent body device 100 and devices mounted on a vehicle M according to a third embodiment.

[0026] Fig.11 This is a flowchart for explaining the flow of a series of processes of the intelligent body device 100 according to the third embodiment.

[0027] Fig.12 It is a diagram for explaining the operation of the intelligent body device 100 according to the third embodiment.

[0028] Description of reference numerals:

[0029] 10…microphone, 20…display / operation device, 30…speaker unit, 40…navigation device, 50…vehicle equipment, 60…communication device, 100…intelligent body device, 110…management unit, 112…sound processing unit, 114…wake-up determination unit, 116…display control unit, 118…sound control unit, 150…intelligent body function unit, 152…detection unit, 154…catchphrase registration unit, 156…information provision unit, 200…intelligent body server. DETAILED DESCRIPTION

[0030] <First Embodiment>

[0031] Hereinafter, a first embodiment of an intelligent body device, an intelligent body device control method, and a storage medium according to the present invention will be described with reference to the accompanying drawings.

[0032] An intelligent body device is a device that implements part or all of an intelligent body system. Below, as an example of an intelligent body device, an intelligent body device installed in a vehicle (hereinafter referred to as vehicle M) and having an intelligent body function is described. An intelligent body function is, for example, a function that at least partially utilizes a network connection using a wireless communication device to communicate with the occupants of the vehicle M while providing various information based on the requirements (instructions) contained in the occupants' speech or mediating network services. Among the intelligent body functions, there may be an intelligent body function that has the function of controlling equipment in the vehicle (such as equipment involved in driving control and body control).

[0033] The intelligent agent functions, for example, in addition to the voice recognition function for recognizing the voice of the occupant (the function of converting the voice into text), are also realized by comprehensively utilizing the natural language processing function (the function of understanding the structure and meaning of the text), the dialogue management function, the network search function of searching other devices via the network or searching the specified database held by the device, etc. Some or all of these functions can be realized by AI (Artificial Intelligence) technology. In addition, part of the structure for performing these functions (especially the voice recognition function and the natural language processing interpretation function) can also be mounted on an intelligent agent server (external device) that can communicate with the communication device of the vehicle M. In the following description, it is assumed that part of the structure is mounted on the intelligent agent server and the intelligent agent device and the intelligent agent server cooperate to realize the intelligent agent system. In addition, the service provider (service entity) that hypothetically appears due to the cooperation of the intelligent agent device and the intelligent agent server is called an intelligent agent.

[0034] <Overall Structure>

[0035] Figure 1 The figure is a structural diagram of an intelligent body system 1 including an intelligent body device 100. The intelligent body system 1 includes, for example, the intelligent body device 100 and an intelligent body server 200. The intelligent body server 200 is a server operated by a provider of the intelligent body system 1. Examples of the provider include automobile manufacturers, network service operators, e-commerce operators, sellers and manufacturers of portable terminals, and any subject (legal person, group, individual, etc.) can become a provider of the intelligent body system 1.

[0036] The intelligent body device 100 communicates with the intelligent body server 200 via the network NW. The network NW includes, for example, part or all of the Internet, cellular network, Wi-Fi network, WAN (Wide Area Network), LAN (Local Area Network), public lines, telephone lines, wireless base stations, etc. Various web servers 300 are connected to the network NW, and the intelligent body server 200 or the intelligent body device 100 can obtain web pages from various web servers 300 via the network NW.

[0037] The agent device 100 communicates with the occupant of the vehicle M, transmits the voice of the occupant to the agent server 200, and presents the answer obtained from the agent server 200 to the occupant in the form of voice output and image display.

[0038] The intelligent server 200, for example, has a speaking style DB 205. Information related to the speaking style of the occupants of the vehicle M is registered in the speaking style DB 205. The information related to the speaking style is information obtained through daily conversations between the occupants of the vehicle M and the intelligent device 100. The information related to the speaking style includes, for example, the catchphrases of the occupants of the vehicle M. The catchphrases of the occupants of the vehicle M are phrases that have become habitual for the occupants of the vehicle M, such as short sentences that the occupants of the vehicle M frequently use when speaking. The intelligent device 100 may also include the speaking style DB 205 instead of or in addition to the intelligent server 200.

[0039] [vehicle]

[0040] Figure 2 1 is a diagram showing the structure of the intelligent body device 100 of the first embodiment and the equipment mounted on the vehicle M. The vehicle M is equipped with, for example, one or more microphones 10, a display and operation device 20, a speaker unit 30, a navigation device 40, a vehicle equipment 50, a communication device 60, and the intelligent body device 100. These devices are connected to each other via a multi-channel communication line such as a CAN (Controller Area Network) communication line, a serial communication line, a wireless communication network, etc. It should be noted that Figure 2 The structure shown is just an example, and part of the structure may be omitted or another structure may be further added.

[0041] The microphone 10 is a sound receiving unit that collects sounds emitted in the vehicle cabin. The display / operation device 20 is a device (or a group of devices) that can display images and accept input operations. The display / operation device 20 includes, for example, a display device configured as a touch panel. The display / operation device 20 may further include a HUD (Head Up Display) and a mechanical input device. The speaker unit 30 includes, for example, a plurality of speakers (sound output units) arranged at different positions in the vehicle cabin. The display / operation device 20 may also be shared by the intelligent body device 100 and the navigation device 40.

[0042] The navigation device 40 includes a navigation HMI (Human machine Interface), a position measuring device such as a GPS (Global Positioning System), a storage device storing map information, and a control device (navigation controller) for performing route search, etc. A part or all of the microphone 10, the display and operation device 20, and the speaker unit 30 can also be used as a navigation HMI. The navigation device 40 searches for a path (navigation path) for moving from the position of the vehicle M determined by the position measuring device to the destination input by the occupant, and uses the navigation HMI to output guidance information in a manner so that the vehicle M can travel along the path. The route search function can also exist in a navigation server that can be accessed via the network NW. In this case, the navigation device 40 obtains the path from the navigation server and outputs the guidance information.

[0043] The vehicle equipment 50 includes, for example, an engine, a driving motor and other driving force output devices, an engine starter motor, a door locking device, a door opening and closing device, windows, a window opening and closing device and a window opening and closing control device, seats, a seat position control device, an interior rearview mirror and an angle position control device, lighting devices inside and outside the vehicle and their control devices, wipers, defoggers and their respective control devices, turn signal lights and their control devices, an air conditioning device, a vehicle information device such as driving distance, tire air pressure information, and fuel remaining information.

[0044] The communication device 60 can access the network NW using, for example, a cellular network or a Wi-Fi network. The communication device 60 may be an in-vehicle communication device or a general-purpose communication device such as a smart phone brought into the vehicle.

[0045] [Intelligent Device]

[0046] return Figure 2 The body device 100 includes a management unit 110 and an body function unit 150. The management unit 110 includes, for example, a sound processing unit 112, a wake-up determination unit 114, a display control unit 116, and a sound control unit 118. Figure 2 The software configuration shown is simply shown for the purpose of explanation, and can actually be arbitrarily changed, for example, so that the management unit 110 can be interposed between the agent function unit 150 and the communication device 60 .

[0047] Each component of the intelligent device 100 is implemented by executing a program (software) through a hardware processor such as a CPU (Central Processing Unit). Some or all of these components may also be implemented by hardware (including circuitry) such as LSI (Large Scale Integration), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array), GPU (Graphics Processing Unit), etc., or may be implemented through the coordination of software and hardware. The program may be pre-stored in a storage device such as a HDD (Hard Disk Drive) or a flash memory (a storage device having a non-temporary storage medium), or may be stored in a removable storage medium such as a DVD or CD-ROM (a non-temporary storage medium), and installed by assembling the storage medium into a drive device.

[0048] The management unit 110 functions by executing programs such as an OS (Operating System) and middleware.

[0049] The sound processing unit 112 performs sound processing on the input sound in a manner that makes it suitable for recognizing the wake-up word preset for the agent.

[0050] The wake-up determination unit 114 recognizes the wake-up word preset for the agent from the sound (sound stream) after the sound processing. First, the wake-up determination unit 114 detects the sound interval based on the amplitude and zero crossing of the sound waveform in the sound stream. The wake-up determination unit 114 can also perform interval detection based on the sound recognition and non-sound recognition of the frame unit based on the Gaussian mixture model (GMM).

[0051] Next, the wake-up determination unit 114 converts the sound in the detected sound interval into text and sets it as text information. Then, the wake-up determination unit 114 determines whether the textualized text information matches the wake-up word. When it is determined to be a wake-up word, the wake-up determination unit 114 starts the intelligent body function unit 150. It should be noted that the function equivalent to the wake-up determination unit 114 can also be mounted on the intelligent body server 200. In this case, the management unit 110 sends the sound stream that has been sound-processed by the sound processing unit 112 to the intelligent body server 200. When the intelligent body server 200 determines that it is a wake-up word, the intelligent body function unit 150 starts according to the instruction from the intelligent body server 200. It should be noted that the intelligent body function unit 150 can also be in the startup state at all times and make the determination of the wake-up word by itself. In this case, the management unit 110 does not need to have the wake-up determination unit 114.

[0052] The agent function unit 150 includes, for example, a detection unit 152, a catchphrase registration unit 154, and an information provision unit 156. The agent function unit 150 cooperates with the agent server 200 to make an agent appear, and provides services including voice-based responses according to the speech of the occupant of the vehicle M. The agent function unit 150 is given the authority to control the vehicle equipment 50. In addition, the agent function unit 150 communicates with the agent server 200 via the communication device 60.

[0053] The detection unit 152 detects the catchphrases of the occupants of the vehicle M when speaking by analyzing the sound processed by the sound processing unit 112. The catchphrases are an example of the way the occupants of the vehicle M speak. The catchphrases include positive catchphrases that are likely to give a good impression to the person speaking to, and negative catchphrases that are likely to give a bad impression to the person speaking to. Examples of positive catchphrases include short sentences such as "happy", "happy", "excited", and "interesting". Examples of negative catchphrases include short sentences such as "damn it", "but", "having said that", "anyway", "forget it", "no time", "no money", "busy", "tired", and "troublesome".

[0054] The catchphrase registering unit 154 registers the catchphrase of the passenger of the vehicle M when speaking detected by the detecting unit 152. For example, when the catchphrase of the passenger of the vehicle M when speaking is detected by the detecting unit 152, the catchphrase registering unit 154 transmits information related to the detected catchphrase to the agent server 200 through the communication device 60. The agent server 200 registers the information related to the catchphrase received from the catchphrase registering unit 154 in the speaking method DB 205. For example, when the catchphrase of the passenger of the vehicle M when speaking is detected by the detecting unit 152, if the detected catchphrase is already registered in the speaking method DB 205 of the agent server 200, the catchphrase registering unit 154 adds the frequency of the corresponding catchphrase to update the information related to the catchphrase registered in the speaking method DB 205.

[0055] Figure 3 : is a diagram showing an example of data content of the speaking style DB205. Information related to the speaking style of each passenger of the vehicle M is registered in the speaking style DB205. In the illustrated example, in the speaking style DB205, for example, a correspondence is established between the content of a catchphrase and the frequency of the catchphrase and the passenger ID. The passenger ID is identification information for specifying the passenger of the vehicle M. The content of the catchphrase is a short sentence detected as the catchphrase of the passenger of the vehicle M. In this example, for example, negative catchphrases such as "damn it", "having said that", and "anyway" are registered as the catchphrases of the passenger of the vehicle M. The frequency of the catchphrase is the frequency with which the catchphrases of the passenger of the vehicle M are detected.

[0056] The information providing unit 156 provides information for correcting the catchphrases of the occupant of the vehicle M to the occupant of the vehicle M. When the catchphrase is detected by the detection unit 152, the information providing unit 156 refers to the speech pattern DB 205 to determine whether the frequency of the catchphrases registered by the catchphrase registration unit 154 is detected to be above the threshold. When the frequency of the caughtphrases detected is above the threshold, the information providing unit 156 provides information for correcting the caughtphrases detected with a frequency above the threshold to the occupant of the vehicle M. For example, when the frequency of negative caughtphrases detected is above the threshold, the information providing unit 156 provides information for correcting the negative caughtphrases detected with a frequency above the threshold to the occupant of the vehicle M. For example, when the occupant of the vehicle M speaks a catchphrase including an object, the information providing unit 156 corrects the catchphrases of the occupant of the vehicle M by outputting a warning for visualizing the catchphrases including the object in the speech of the occupant of the vehicle M from the intelligent device 100 to the occupant of the vehicle M. In addition, for example, when the occupant of the vehicle M makes a catchphrase including an object, the information providing unit 156 may correct the catchphrase of the occupant of the vehicle M by outputting the catchphrase not including the object from the intelligent body device 100 to the occupant of the vehicle M.

[0057] The display control unit 116 causes the display / operation device 20 to display an image according to instructions from the intelligent agent function unit 150. The display control unit 116 generates an image of an anthropomorphic intelligent agent (hereinafter referred to as an intelligent agent image) that communicates with passengers in the vehicle cabin through the control of a part of the intelligent agent function unit 150, and causes the display / operation device 20 to display the generated intelligent agent image. The intelligent agent image is, for example, an image of a form of talking to passengers. The intelligent agent image may include, for example, a facial image to the extent that the expression and facial orientation are at least recognized by the viewer (passenger). For example, the intelligent agent image may indicate parts that simulate eyes and noses in the facial area, and recognize the expression and facial orientation based on the position of the parts in the facial area. In addition, the intelligent agent image may also be perceived three-dimensionally, by including a head image in a three-dimensional space so that the viewer can recognize the facial orientation of the intelligent agent, or by including an image of the main body (torso, hands and feet) so that the action, behavior, posture, etc. of the intelligent agent can be recognized. In addition, the intelligent agent image may also be an animated image.

[0058] The sound control unit 118 causes part or all of the speakers included in the speaker unit 30 to output sound according to instructions from the intelligent body function unit 150. The sound control unit 118 may also use a plurality of speaker units 30 to control the sound image of the intelligent body sound to be localized at a position corresponding to the display position of the intelligent body image. The position corresponding to the display position of the intelligent body image is, for example, a position where the occupant is predicted to feel that the intelligent body image is speaking the intelligent body sound, specifically, a position near the display position of the intelligent body image. In addition, sound image localization refers to, for example, determining the spatial position of the sound source felt by the occupant by adjusting the volume of the sound transmitted to the left and right ears of the occupant.

[0059] [Agent Server]

[0060] Figure 4 1 is a diagram showing a part of the structure of the agent server 200 and the structure of the agent device 100. The following describes the operation of the agent function unit 150 and the like together with the structure of the agent server 200. Here, the description of the physical communication from the agent device 100 to the network NW is omitted.

[0061] The intelligent body server 200 has a communication unit 210. The communication unit 210 is, for example, a network interface such as a NIC (Network Interface Card). In addition, the intelligent body server 200 has, for example, a voice recognition unit 220, a natural language processing unit 222, a dialogue management unit 224, a network search unit 226, and a response text generation unit 228. These components are implemented by executing a program (software) through a hardware processor such as a CPU. Some or all of these components can also be implemented by hardware such as LSI, ASIC, FPGA, GPU (including circuitry), or can be implemented through the coordination of software and hardware. The program can be pre-stored in a storage device such as an HDD, a flash memory (a storage device having a non-temporary storage medium), or can be stored in a removable storage medium such as a DVD, a CD-ROM (a non-temporary storage medium), and installed by assembling the storage medium to the drive device.

[0062] The agent server 200 includes a storage unit 250. The storage unit 250 is implemented by the various storage devices described above. In addition to the speech pattern DB 205, the storage unit 250 stores data and programs such as a personal profile 252, a dictionary DB (database) 254, a knowledge base DB 256, and a response rule DB 258.

[0063] In the intelligent body device 100, the intelligent body function unit 150 sends the sound stream or the sound stream after compression, encoding, etc. to the intelligent body server 200. When the intelligent body function unit 150 recognizes a sound instruction that can be processed locally (not processed by the intelligent body server 200), it can perform the processing required by the sound instruction. The sound instruction that can be processed locally is a sound instruction that can be answered by referring to the storage unit (not shown) of the intelligent body device 100, or a sound instruction to control the vehicle equipment 50 (for example, an instruction to turn on the air conditioner, etc.). Therefore, the intelligent body function unit 150 can also have a part of the functions of the intelligent body server 200.

[0064] After acquiring the sound stream, the sound recognition unit 220 performs sound recognition and outputs textual information, and the natural language processing unit 222 interprets the meaning of the text information while referring to the dictionary DB254. The dictionary DB254 is a database that establishes a corresponding relationship between abstract meaning information and text information. The dictionary DB254 may also include a list of synonyms and antonyms. The processing of the sound recognition unit 220 and the processing of the natural language processing unit 222 are not clearly divided into stages, and can be carried out in such a way that the sound recognition unit 220 receives the processing results of the natural language processing unit 222 and corrects the recognition results, etc.

[0065] For example, when the natural language processing unit 222 recognizes the meaning of "Today's weather?" or "How's the weather?" as a recognition result, it generates a command that is replaced with the standard text information "Today's weather". As a result, even if there is a difference in the expression of the request voice, it is easy to have a conversation that meets the requirements. In addition, the natural language processing unit 222 can also use artificial intelligence processing such as machine learning processing using probability to recognize the meaning of the text information and generate a command based on the recognition result.

[0066] The dialogue management unit 224 determines the content of the speech to the occupants of the vehicle M based on the processing result (command) of the natural language processing unit 222, while referring to the personal profile 252, the knowledge base DB 256, and the response rule DB 258. The personal profile 252 includes the personal information, hobbies, and past conversation history of the occupants stored for each occupant. The knowledge base DB 256 is information that specifies the relationship between things. The response rule DB 258 is information that specifies the action (answer, content of device control, etc.) that the agent should perform in response to the command.

[0067] In addition, the dialogue management unit 224 may also identify the occupant by comparing the characteristic information obtained from the sound stream with the personal profile 252. In this case, in the personal profile 252, for example, a correspondence is established between the personal information and the characteristic information of the sound. The characteristic information of the sound is, for example, information related to the characteristics of the speaking style such as the pitch, intonation, rhythm (pattern of the high and low pitch of the sound), and the characteristic quantity based on the Mel Frequency Cepstrum Coefficients. The characteristic information of the sound is, for example, information obtained by having the occupant speak a predetermined word, sentence, etc. and recognizing the sound produced when the occupant is initially registered.

[0068] When the session management unit 224 instructs the network search unit 226 to search for information that can be searched via the network NW, the network search unit 226 accesses various web servers 300 via the network NW to obtain the desired information. The “information that can be searched via the network NW” is, for example, the evaluation results of ordinary users of restaurants located around the vehicle M, or the weather forecast corresponding to the location of the vehicle M on the day.

[0069] The response text generation unit 228 generates a response text in a manner that conveys the content of the speech determined by the dialogue management unit 224 to the occupant of the vehicle M, and transmits the response text to the intelligent device 100. When the response text generation unit 228 determines that the occupant is a occupant registered in the personal profile, the response text generation unit 228 may generate a response text that calls out the occupant's name or sets a speech method that imitates the occupant's speech method.

[0070] After obtaining the response text, the agent function unit 150 instructs the voice control unit 118 to synthesize and output the voice. In addition, the agent function unit 150 instructs the display control unit 116 to display the image of the agent in conjunction with the voice output. In this way, the agent function of the virtual agent responding to the occupant of the vehicle M is realized.

[0071] [Processing flow of intelligent device]

[0072] Hereinafter, the flow of a series of processes of the intelligent body device 100 according to the first embodiment will be described using a flowchart. Figure 5 The process of the flowchart shown can be executed when, for example, a speech of the occupant of the vehicle M is input.

[0073] First, the detection unit 152 detects the catchphrases of the occupants of the vehicle M when speaking by analyzing the speech input from the occupants of the vehicle M (step S10). The catchphrase registration unit 154 establishes a correspondence between the catchphrases detected by the detection unit 152 and the occupant ID of the occupants of the vehicle M and registers them in the speaking method DB 205 (step S12). Next, the information providing unit 156 refers to the speaking method DB 205 to determine whether the detected catchphrases with a frequency above the threshold are included (step S14). If the information providing unit 156 determines that the detected catchphrases with a frequency above the threshold are included, the information providing unit 156 provides the occupants of the vehicle M with information for correcting the detected catchphrases with a frequency above the threshold (step S16). Thus, the processing of this flowchart ends. On the other hand, if the information providing unit 156 determines that the detected catchphrases with a frequency above the threshold are not included, the information for correcting the catchphrases is not provided to the occupants of the vehicle M, and the processing of this flowchart ends.

[0074] Figure 6 1 is a diagram for explaining the operation of the intelligent body device 100 according to the first embodiment. In the example shown in the diagram, when the catchphrases of the passengers of the vehicle M include negative catchphrases, information for correcting the catchphrases is provided to the passengers of the vehicle M.

[0075] The intelligent body device 100 refers to the speech pattern DB 205 to detect the catchphrases of the passengers of the vehicle M whose detected frequency is above the threshold. In the illustrated example, the intelligent body device 100 detects the negative catchphrase "Hell" as the catchphrase of the passengers of the vehicle M whose detected frequency is above the threshold. In this case, the intelligent body device 100 outputs a warning to the passengers of the vehicle M for visualizing the negative catchphrase "Hell".

[0076] The occupant of the vehicle M inputs a speech requesting correction of the warned catchphrase to the agent device 100. In the illustrated example, the occupant of the vehicle M inputs a speech requesting correction of the negative catchphrase "damn" to the positive catchphrase "bad".

[0077] When a speech including a negative catchphrase is input from a passenger of the vehicle M after receiving a request for correction of a catchphrase, the intelligent body device 100 provides information for correcting the negative catchphrase to the passenger of the vehicle M. In the example shown in the figure, a speech including a negative catchphrase "Hell" is input to the intelligent body device 100 from the passenger of the vehicle M. Therefore, the intelligent body device 100 outputs a speech using a positive catchphrase "Not good" instead of the negative catchphrase "Hell" as a response to the speech from the passenger of the vehicle M.

[0078] According to the intelligent body device 100 of the first embodiment described above, the catchphrases of the occupants of the vehicle M can be corrected with a more developmental utilization scheme. The catchphrases of the occupants of the vehicle M are information obtained from daily conversations with the occupants of the vehicle M, and it is sometimes difficult to set up an opportunity to detect the catchphrases of the occupants of the vehicle M. Therefore, in the intelligent body device 100 of the first embodiment, the catchphrases of the occupants of the vehicle M are detected from the conversations between the occupants of the vehicle M and the intelligent body device 100, and information for correcting the detected catchphrases is provided to the occupants of the vehicle M. As a result, the catchphrases of the occupants of the vehicle M can be corrected with a more developmental utilization scheme.

[0079] <Second Embodiment>

[0080] The second embodiment will be described below. The second embodiment differs from the first embodiment in that the processing content is to provide information for correcting the dialect of the occupant of the vehicle M. The following description will focus on this difference.

[0081] Figure 7 1 is a diagram showing the configuration of the body device 100 according to the second embodiment and devices mounted on the vehicle M. The body function unit 150A of the body device 100 according to the second embodiment includes, for example, a detection unit 152 , a dialect registration unit 154A, and an information provision unit 156 .

[0082] The detection unit 152 detects the dialect of the occupant of the vehicle M when speaking by analyzing the sound processed by the sound processing unit 112. The dialect is an example of the way the occupant of the vehicle M speaks. The dialect means a language system for each region, including, for example, Osaka dialect and Kyoto dialect. The dialect is defined by, for example, vocabulary, grammar, intonation, accent, and the like.

[0083] The dialect registration unit 154A registers the dialect of the occupant of the vehicle M detected by the detection unit 152. For example, when the dialect of the occupant of the vehicle M is detected by the detection unit 152, the dialect registration unit 154A transmits information related to the detected dialect to the agent server 200 via the communication device 60. The agent server 200 registers the information related to the dialect received from the dialect registration unit 154A in the speaking method DB 205.

[0084] The information providing unit 156 provides the occupants of the vehicle M with information for correcting the dialect of the occupants of the vehicle M. When the detection unit 152 detects a dialect, the information providing unit 156 refers to the speaking style DB 205 to determine whether the detected dialect is a prescribed dialect registered by the dialect registration unit 154A. When the information providing unit 156 determines that the dialect detected by the detection unit 152 is a prescribed dialect, the information providing unit 156 provides the occupants of the vehicle M with information for correcting the detected dialect. For example, the information providing unit 156 may pre-register features of the detected dialect such as intonation and words that the occupants of the vehicle M are concerned about, and provide the occupants of the vehicle M with information for correcting the features of the pre-registered dialect.

[0085] Hereinafter, the flow of a series of processes of the intelligent body device 100 according to the second embodiment will be described using a flowchart. Figure 8 The process of the flowchart shown can be executed when, for example, a speech of the occupant of the vehicle M is input.

[0086] The detection unit 152 detects the dialect of the occupant of the vehicle M when speaking by analyzing the speech input from the occupant of the vehicle M (step S20). The detection unit 152 detects the dialect of the occupant of the vehicle M by analyzing the vocabulary, grammar, phonetics, accent, etc. of the speech of the occupant of the vehicle M, for example. In addition, the detection unit 152 establishes a correspondence between the dialect detected in step S10 and the occupant ID of the occupant of the vehicle M and registers it in the speaking method DB205 (step S22). Next, the information providing unit 156 refers to the speaking method DB205 to determine whether the prescribed dialect is corresponding to the occupant of the vehicle M and registered in the speaking method DB205 (step S24). When the information providing unit 156 determines that the prescribed dialect is corresponding to the occupant of the vehicle M and registered in the speaking method DB205, the information providing unit 156 provides the occupant of the vehicle M with information for correcting the prescribed dialect (step S26). Thus, the processing of this flowchart ends. On the other hand, when the prescribed dialect is not associated with the occupant of the vehicle M and registered in the speaking pattern DB 205 , the information providing unit 156 does not provide the occupant of the vehicle M with information for correcting the prescribed dialect, and the processing of this flowchart ends.

[0087] Fig. 91 is a diagram for explaining the operation of the intelligent body device 100 according to the second embodiment. In the example shown in the diagram, when the speech of the occupant of the vehicle M includes a predetermined dialect, information for correcting the predetermined dialect is provided to the occupant of the vehicle M.

[0088] The agent device 100 receives a speech input from an occupant of the vehicle M. In the illustrated example, the agent device 100 receives a speech input from an occupant of the vehicle M including a topic related to the weather.

[0089] The intelligent body device 100 analyzes the received speech to detect the dialect of the occupant of the vehicle M. In the example shown in the figure, the intelligent body device 100 detects that the dialect of the occupant of the vehicle M is "Osaka dialect". In this case, the intelligent body device 100 outputs a warning to the occupant of the vehicle M to visualize that the dialect of the occupant of the vehicle M is "Osaka dialect".

[0090] The occupant of the vehicle M inputs a speech requesting correction of the warned dialect to the agent device 100. In the example shown in the figure, the occupant of the vehicle M inputs a speech requesting correction of "Osaka dialect" to "Tokyo dialect" to the agent device 100.

[0091] When a speech including a dialect to be corrected is input from a passenger of the vehicle M after receiving a request for correction of the dialect, the intelligent body device 100 provides information for correcting the dialect to the passenger of the vehicle M. In the example shown in the figure, a speech including "Osaka dialect" is input to the intelligent body device 100 from the passenger of the vehicle M. Therefore, the intelligent body device 100 outputs a speech using "Tokyo dialect" instead of "Osaka dialect" as a response to the speech from the passenger of the vehicle M.

[0092] According to the intelligent body device 100 of the second embodiment described above, in addition to the effects of the intelligent body device 100 of the first embodiment, the dialect of the occupants of the vehicle M can be corrected in a more developmental utilization scheme. The dialect of the occupants of the vehicle M is information obtained from daily conversations with the occupants of the vehicle M, and sometimes it is difficult to detect the dialect of the occupants of the vehicle M. Therefore, in the intelligent body device 100 of the second embodiment, the dialect of the occupants of the vehicle M is detected from the conversation between the occupants of the vehicle M and the intelligent body device 100 during riding, and information for correcting the detected dialect is provided to the occupants of the vehicle M. As a result, the dialect of the occupants of the vehicle M can be corrected in a more developmental utilization scheme.

[0093] <Third Embodiment>

[0094] The third embodiment will be described below. The third embodiment differs from the first embodiment in that the processing content is to provide information for guiding in a dialect close to that designated by the occupant of the vehicle M. The following description will focus on this difference.

[0095] Fig.10 1 is a diagram showing the configuration of the intelligent body device 100 according to the third embodiment and devices mounted on the vehicle M. The intelligent body function unit 150B of the intelligent body device 100 according to the third embodiment includes, for example, a dialect designation accepting unit 154B and an information providing unit 156 .

[0096] The dialect designation receiving unit 154B receives an instruction to designate a dialect from an occupant of the vehicle M. For example, when an occupant of the vehicle M operates the display / operation device 20 to designate a dialect, the dialect designation receiving unit 154B receives the instruction to designate a dialect based on an operation signal output from the display / operation device 20. The designated dialect is not limited to Japanese, and may be a local dialect such as English, or a local dialect used in a specific region such as the Oxford-Cambridge accent.

[0097] The information providing unit 156 provides information for guiding the occupant of the vehicle M so that the dialect of the occupant of the vehicle M approaches the dialect received by the dialect designation accepting unit 154B. For example, when the dialect designation accepting unit 154B receives an instruction to designate a dialect, the information providing unit 156 guides the dialect of the occupant of the vehicle M by outputting a speech including the received dialect from the intelligent device 100 to the occupant of the vehicle M.

[0098] Hereinafter, the flow of a series of processes of the intelligent body device 100 according to the third embodiment will be described using a flowchart. Fig.11 The process of the flowchart shown can be executed when, for example, a speech of the occupant of the vehicle M is input.

[0099] The dialect designation accepting unit 154B determines whether the dialect is designated by the occupant of the vehicle M (step S30). When the information providing unit 156 determines that the dialect is designated by the dialect designation accepting unit 154B, it provides the occupant of the vehicle M with information for guiding in a manner close to the designated dialect (step S32). Thus, the processing of this flowchart ends. On the other hand, when the information providing unit 156 determines that the dialect is not designated by the dialect designation accepting unit 154B, it does not guide the occupant of the vehicle M in the dialect, and the processing of this flowchart ends.

[0100] Fig.121 is a diagram for explaining the operation of the intelligent body device 100 according to the third embodiment. In the example shown in the figure, when the occupant of the vehicle M specifies a predetermined dialect, information for guiding in a manner close to the specified dialect is provided to the occupant of the vehicle M.

[0101] The agent device 100 receives an utterance requesting dialect correction input from the occupant of the vehicle M. In the example shown in the figure, the agent device 100 requests guidance to correct the dialect of the occupant of the vehicle M to be close to the "Tokyo dialect".

[0102] When a speech including a prescribed dialect is input from a passenger of the vehicle M after receiving a request for guidance in a dialect, the agent device 100 provides information for guidance in a manner close to the prescribed dialect to the passenger of the vehicle M. In the example shown in the figure, a speech including "Osaka dialect" is input from the passenger of the vehicle M to the agent device 100. Therefore, the agent device 100 outputs a response using "Tokyo dialect" to the speech from the passenger of the vehicle M using "Osaka dialect".

[0103] According to the intelligent body device 100 of the third embodiment described above, in addition to the effects of the intelligent body device 100 of the first or second embodiment, it is also possible to guide the dialect of the occupants of the vehicle M in accordance with the intention of the occupants of the vehicle M. The dialect of the occupants of the vehicle M is habitually spoken, and its guidance is sometimes accompanied by difficulties. Therefore, in the intelligent body device 100 of the third embodiment, information for guiding in a manner close to the dialect specified by the occupants of the vehicle M is provided to the occupants of the vehicle M. As a result, the dialect of the occupants of the vehicle M can be guided in accordance with the intention of the occupants of the vehicle M.

[0104] [Modification of the embodiment]

[0105] In the first or second embodiment, when the speech of the occupant of the vehicle M includes catchphrases or dialects, the agent device 100 may ignore the speech of the occupant without responding to the speech of the occupant, thereby urging the occupant to correct the speech.

[0106] In the above-mentioned embodiments, the agent device 100 may start correcting the speech of the occupants of the vehicle M by using, for example, news of inappropriate remarks by a politician as a scene that easily arouses emotions of the occupants of the vehicle M as a trigger condition.

[0107] In the above-mentioned embodiments, the intelligent body device 100 may estimate the number of passengers and the relationship between passengers by, for example, talking with passengers of the vehicle M or taking an image of the interior of the vehicle, and determine whether to start correcting the speech of the passenger based on the result of the estimation. The intelligent body device 100 may start correcting speech when the passenger is alone in the vehicle, or may start correcting speech when the passenger consists of only family members.

[0108] In each of the above-mentioned embodiments, the intelligent body device 100 may be provided in a portable information terminal, for example. In this case, the portable information terminal may correct the user's speaking style during the conversation with the user.

[0109] Although specific embodiments of the present invention have been described above using the embodiments, the present invention is not limited to such embodiments at all, and various modifications and substitutions can be made without departing from the gist of the present invention.

Claims

1. An intelligent agent device that provides a service including a voice-based response based on a user's speech, wherein: The intelligent body device includes a detection unit and an information providing unit that operate according to the request made by the user through speech, The detection unit detects the user's speaking style when speaking, The information providing unit provides the user with information for correcting the predetermined speaking manner when the detection unit detects that the speech of the user includes the predetermined speaking manner. The information providing unit starts correcting the speech of the user using a scene that easily causes the user's emotions to become excited as a trigger.

2. The intelligent device according to claim 1, wherein: The intelligent body device further includes a catchphrase registering unit for registering the catchphrase detected by the detection unit when the user speaks. When the detection unit detects that the frequency of the catchphrase registered by the catchphrase registration unit is greater than or equal to a threshold, the information providing unit provides the user with information for correcting the catchphrase of the user detected at a frequency greater than or equal to the threshold as the predetermined speaking manner.

3. The intelligent agent device according to claim 1 or 2, wherein: The intelligent body device further includes a dialect registration unit for registering the dialect of the user when speaking detected by the detection unit. The information providing unit provides the user with information for correcting the predetermined dialect as the predetermined speaking style when the predetermined dialect registered by the dialect registration unit is detected by the detection unit.

4. A control method for an intelligent device, wherein: The control method of the intelligent device enables the computer to perform the following processing: Providing services including voice-based responses based on speech of the user; as well as detecting the user's speaking manner when speaking according to the user's request through speech, and providing the user with information for correcting the prescribed speaking manner when detecting that the user's speech includes a prescribed speaking manner, Correction of the user's speech is started using a scene that tends to make the user's emotions rise as a trigger.

5. A storage medium storing a program, wherein: The program causes the computer to execute the following processing: Providing services including voice-based responses based on speech of the user; as well as detecting the user's speaking manner when speaking according to the user's request through speech, and providing the user with information for correcting the prescribed speaking manner when detecting that the user's speech includes a prescribed speaking manner, Correction of the user's speech is started using a scene that tends to make the user's emotions rise as a trigger.