Information processing device, information processing method, and information processing program

Abstract

To output contents properly, effectively utilizing a period during which a driving load is estimated to be low.SOLUTION: An information processing device according to the present application is configured to suspend output of contents when a degree of safety of driving drops. The information processing device has: an acquisition unit that acquires section information indicative of a section where a driving load of a user is inferred to be low based on the degree of safety; a calculation unit that calculates, on the basis of speed of a vehicle driven by a target user being a target of output of contents, a required time for the vehicle to pass through the section as an output allowed time during which the output of the contents is allowed; a prediction unit that predicts an output required time for completing the output of the content on the basis of prescribed information on the content for each of contents of the output candidates; and a selection unit that selects, as an output target content, a content inferred to be completely outputted within the output allowed time from among the contents of the output candidates, on the basis of the output allowed time and the output required time.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and an information processing program.

Background Art

[0002] Conventionally, technologies have been proposed for more suitably providing information to a user riding in a vehicle.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] For example, the above-described conventional technology determines the magnitude of the burden on the driver based on the driver's state and provides information to the driver according to the result. As an example, when driving of the vehicle is easy, the above-described conventional technology provides appropriate information corresponding to the magnitude of the burden at an appropriate timing according to the magnitude of the burden at this time.

[0005] Here, due to the relationship of operating a moving body such as a vehicle, it is considered that there is a certain period during which the driver has a margin in driving the vehicle (that is, the burden is small). However, the above-described conventional technology only provides information at a specific timing according to the magnitude of the burden, and it is hard to say that the period with a margin in driving is effectively utilized.

[0006] The present invention has been made in view of the above, and provides an information processing apparatus, an information processing method, and an information processing program capable of appropriately outputting content by effectively utilizing a period during which the driving load is estimated to be low.

Means for Solving the Problems

[0007] The information processing device described in claim 1 is an information processing device that stops outputting content when the safety level of driving decreases, and is characterized by comprising: an acquisition unit that acquires section information indicating a section in which the user's driving load is estimated to be low based on the safety level; a calculation unit that calculates the time required for the vehicle to pass through the section as an output time in which content can be output, based on the speed of the vehicle being driven by the target user to whom the content is to be output; a prediction unit that predicts the output time required to complete the output of each candidate content based on predetermined information relating to the content; and a selection unit that selects, from among the candidate content to be output, content that is estimated to be completed within the output time, based on the output time and the output time, as content to be output.

[0008] The information processing method described in claim 13 is an information processing method executed by an information processing device that stops outputting content when the level of driving safety decreases, and is characterized by including: an acquisition step of acquiring section information indicating a section in which the user's driving load is estimated to be low based on the level of safety; a calculation step of calculating the time required for the vehicle to pass through the section as an output-capable time in which the content can be output, based on the speed of the vehicle being driven by the target user to whom the content is to be output; a prediction step of predicting the output time required to complete the output of each candidate content based on predetermined information relating to the content; and a selection step of selecting, based on the output-capable time and the output time, content from among the candidate content to be output that is estimated to be completed within the output-capable time as content to be output.

[0009] The information processing program described in claim 14 is an information processing program executed by an information processing device that stops outputting content when the level of driving safety decreases, and is characterized in that it causes the information processing device to execute: an acquisition procedure for acquiring section information indicating a section in which the user's driving load is estimated to be low based on the level of safety; a calculation procedure for calculating the time required for a vehicle to pass through the section as an output-capable time in which content can be output, based on the speed of the vehicle being driven by a target user for whom content is to be output; a prediction procedure for predicting the output time required to complete the output of each candidate content based on predetermined information relating to the content; and a selection procedure for selecting, from among the candidate content, content that is estimated to be output within the output-capable time, as content to be output, based on the output-capable time and the output time. [Brief explanation of the drawing]

[0010] [Figure 1] Figure 1 shows an example of an information processing system according to an embodiment. [Figure 2] Figure 2 is an explanatory diagram (1) illustrating the overall structure of the information processing according to this embodiment. [Figure 3] Figure 3 is an explanatory diagram (2) illustrating the overall structure of the information processing according to this embodiment. [Figure 4] Figure 4 is a diagram (1) showing a specific example of the prediction process according to the embodiment. [Figure 5] Figure 5 is Figure (2) showing a specific example of the prediction process according to the embodiment. [Figure 6] Figure 6 shows an example of the configuration of an information processing device according to the embodiment. [Figure 7] Figure 7 shows an example of the content configuration according to this embodiment. [Figure 8] Figure 8 shows an example of a section information database according to the embodiment. [Figure 9] Figure 9 shows an example of a dialogue history database according to the embodiment. [Figure 10]Figure 10 shows an example of a performance information database according to the embodiment. [Figure 11] Figure 11 is a flowchart showing the prediction process for predicting the duration of a dialogue. [Figure 12] Figure 12 is a flowchart showing the selection process for selecting the content to be output. [Figure 13] Figure 13 is a hardware configuration diagram showing an example of a computer that implements the functions of an information processing device. [Modes for carrying out the invention]

[0011] Below, an example of an embodiment for implementing an information processing device, an information processing method, and an information processing program (hereinafter referred to as "embodiment") will be described in detail with reference to the drawings. Note that this embodiment does not limit the information processing device, information processing method, and information processing program. Furthermore, the same parts will be denoted by the same reference numerals in the following embodiments, and redundant explanations will be omitted.

[0012] [1. Overview of the Embodiment] For example, some in-vehicle devices, such as car navigation systems, have applications that incorporate an agent function. In such cases, users (for example, the driver or passengers) can receive various services by interacting with the agent function.

[0013] In this situation, it is possible that focusing too much on the conversation could lead to neglecting driving, resulting in dangerous driving that could cause an accident. To prevent this, some in-vehicle devices have a function that outputs content when the driver's driving safety level is judged to be high, and stops outputting content when the driver's safety level is judged to be low, thereby interrupting the conversation.

[0014] The in-vehicle device detects driving behavior (such as the driving situation of the user or the driving situation of the vehicle, etc.) from sensor information detected by sensors of the vehicle, and estimates the driving safety level (which can also be said to be the driving comfort level) based on the detected driving behavior. Then, when it is determined that the safety level is higher than a predetermined value and the driving load of the user is low (there is a margin in driving), the in-vehicle device starts a dialogue based on the content provided by the agent function unit. On the other hand, when it is determined that the safety level is lower than a predetermined value and the driving load of the user is high (there is no margin in driving), the in-vehicle device does not start a dialogue based on the content provided by the agent function unit, or interrupts the ongoing dialogue.

[0015] However, it is required to avoid the situation where the dialogue using the content is interrupted in this way. For example, the agent function unit may conduct a dialogue using content that is considered beneficial to the user, such as content related to spots presumed to be of interest to the user, content related to tourist guides around the driving location, or content related to danger notifications. Therefore, there is a desire to complete the output of the relevant content during the dialogue between the user and the agent function unit.

[0016] The present invention has been made paying attention to the above circumstances. When a dialogue is conducted using the content provided by the agent function unit, considering the time required for the dialogue (how much time is required for the dialogue) and the time during which the user can respond to the dialogue, it is possible to control the progress of the dialogue to prioritize the content whose output can be completed within the time when the dialogue is possible. Therefore, the idea is obtained from the point that the time when the dialogue is possible in the content output can be effectively utilized.

[0017] 〔2. System Configuration〕 First, the configuration of the information processing system according to the embodiment will be described using FIG. 1. FIG. 1 is a diagram showing an example of the information processing system according to the embodiment. In FIG. 1, an information processing system 1 is shown as an example of the information processing system according to the embodiment.

[0018] As shown in Figure 1, the information processing system 1 may include an information processing device 100 and an agent device 300. The information processing device 100 and the agent device 300 are connected to each other via a network N, either by wire or wireless communication. Furthermore, the information processing system 1 shown in Figure 1 may include any number of information processing devices 100 and any number of agent devices 300.

[0019] The information processing device 100 stops outputting content if the level of driving safety deteriorates. For example, if the level of driving safety deteriorates, the information processing device 100 interrupts a conversation between the user and the agent function unit that discusses content provided by the agent function unit. Thus, the information processing device 100 is an example of an information processing device according to the embodiment.

[0020] Furthermore, as shown in Figure 1, the information processing device 100 may be mounted on the vehicle VEx. In other words, the information processing device 100 may be an in-vehicle device. Specifically, the information processing device 100 may be a dedicated navigation device built into or mounted on the vehicle VEx.

[0021] The information processing device 100 may consist of a navigation device and a recording device. As an example, the information processing device 100 may be a composite device in which an independent navigation device and a recording device are communicated with each other. As another example, the information processing device 100 may be a single device having a navigation function and a recording function.

[0022] Furthermore, the information processing device 100 may be equipped with various sensors. For example, the information processing device 100 may be equipped with various sensors such as a camera, an accelerometer, a gyroscope, a GPS sensor, and a barometric pressure sensor. As a result, the information processing device 100 can provide dialogue and information to support the user's driving based on sensor information acquired by the various sensors.

[0023] Furthermore, users can operate their everyday portable terminal devices (e.g., smartphones, tablet devices, notebook PCs, desktop PCs, PDAs, etc.) in the same way as the information processing device 100 by installing a predetermined application on these devices. In other words, users can substitute their own portable terminal devices for in-vehicle devices. For these reasons, a portable terminal device owned by a user can also be understood as an information processing device according to this embodiment.

[0024] The agent device 300 is a server device that actually performs response processing to respond to user utterances. For example, the agent device 300 executes response processing in accordance with the control of the agent function unit described later. Furthermore, the agent device 300 and the agent function unit may be paired for each service, and there may be as many agent devices as there are agent function units.

[0025] For example, if the agent device 300 is unable to obtain an appropriate response to the utterance as a result of response processing, it may return response information to the agent function unit indicating that it is unable to respond. On the other hand, if the agent device 300 is able to obtain an appropriate response to the utterance as a result of response processing, it may return response information to the agent function unit including the answer to the request. The agent function unit then engages in dialogue with the user based on the response information.

[0026] [3. Overview of Information Processing] From here, we will explain, using Figures 2 and 3, a control process that, as an information processing embodiment, selects content that is estimated to be able to conclude the dialogue appropriately within the dialogueable time in which the user can respond to the dialogue, and controls the system to initiate a dialogue corresponding to this content.

[0027] For example, Figures 2 and 3 show a scene in which user U1 is driving vehicle VE1 (an example of vehicle VEx) on road RD1. Also, as shown in Figures 2 and 3, road RD1 includes a driving load reduction section SE11, which is estimated to be a section where the driver's driving load tends to be low based on safety. In this case, the information processing device 100 can perform control processing in real time at a predetermined point just before vehicle VE1 reaches the driving load reduction section SE11. The control processing performed by the information processing device 100 will be explained below using this scene as an example. In the following embodiment, the information processing device 100 is assumed to be an in-vehicle device.

[0028] First, using Figure 2, we will explain the processes of steps S1 and S2 of the control processing performed by the information processing device 100. Figure 2 is an explanatory diagram (1) illustrating the overall structure of the information processing according to this embodiment.

[0029] In the example shown in Figure 2, user U1's vehicle VE1 is approaching the driving load reduction section SE11 at time TM1 while traveling at a speed of Q (m / h). The distance of the driving load reduction section SE11 is P (m). In this example, the information processing device 100 calculates the conversational time XX during which user U1 can respond to a conversation when the agent function unit provides the content to be discussed, based on the speed of vehicle VE1, which is Q (m / h), and the distance of the driving load reduction section SE11, which is P (m) (step S1). Specifically, the information processing device 100 may determine the conversational time XX by calculating the time required P / Q for vehicle VE1 to pass through the driving load reduction section SE11, based on the speed Q (m / h) and the distance P (m).

[0030] The agent function unit referred to here may be one that is part of the information processing device 100, as shown in Figure 6. In the following description, the information processing device 100 will be assumed to have agent function units 31 and 32, as shown in the example in Figure 6.

[0031] For example, the agent function unit 31 may be a function of application AP31 (app AP31), and provides content C1 corresponding to the service of app AP31. More specifically, the agent function unit 31 speaks based on content C1, which consists of message information (a group of texts) about topic NA1. In this way, the agent function unit 31 engages in a dialogue with user U1 about topic NA1.

[0032] Furthermore, the agent function unit 32 may be a function of the application AP32 and provides content C2 corresponding to the service of the application AP32. More specifically, the agent function unit 32 speaks based on content C2, which consists of message information (a group of texts) related to topic NA2. In this way, the agent function unit 32 engages in a dialogue with the user U1 regarding topic NA2.

[0033] Next, the information processing device 100 predicts the time required for the conversation between the corresponding agent function unit and user U1 to be completed, based on the history information of past conversations using the output candidate contents C1 and C2 (step S2). Specifically, the information processing device 100 assumes that an utterance by the agent function unit 31 based on content C1 was output when the vehicle VE1 entered the driving load reduction section SE11, and predicts the time required for the conversation with user U1 regarding topic NA1 to be completed. Also, the information processing device 100 assumes that an utterance by the agent function unit 32 based on content C2 was output when the vehicle VE1 entered the driving load reduction section SE11, and predicts the time required for the conversation with user U1 regarding topic NA2 to be completed.

[0034] According to the example in Figure 2, the information processing device 100 detects a dialogue pattern based on the combination of the output message from the agent function unit 31 corresponding to content C1 and the response message from any user Ux included in the dialogue history LG1. Then, for each detected pattern, the information processing device 100 calculates the actual time required (spent) for the dialogue of that pattern.

[0035] For example, the information processing device 100 may detect the pattern of how the conversation progressed when user Ux responded to an output message with a "positive" message. In the example in Figure 2, the information processing device 100 detects the conversation pattern PT11 as such a pattern. In such a case, the information processing device 100 may also calculate the actual time required for the conversation of conversation pattern PT11 based on the actual conversation time when the exchange progressed according to conversation pattern PT11 in the conversation history LG1. The example in Figure 2 shows an example in which the information processing device 100 calculated the actual time 1y.

[0036] On the other hand, the information processing device 100 also detects the pattern of how the conversation progressed when the user Ux responded to the output message with a "negative" message. In the example in Figure 2, the information processing device 100 detects the conversation pattern PT12 as such a pattern. In this example, the information processing device 100 may also calculate the actual time required for the conversation of conversation pattern PT12 based on the actual conversation time when the exchange progressed in the conversation history LG1 of conversation pattern PT12. In the example in Figure 2, an example is shown in which the information processing device 100 calculated the actual time 1n.

[0037] In this case, the information processing device 100 predicts the longer of the actual time 1y and the actual time 1n as the dialogue duration corresponding to content C1. As shown in Figure 2, let's assume the length relationship is actual time 1y < actual time 1n. In this example, the information processing device 100 predicts the longer actual time 1n as the dialogue duration 1n.

[0038] Regarding step S2, we have shown an example of how the required dialogue time 1n between the agent function unit 31 and the user Ux is calculated, using the output candidate content C1 as an example. Therefore, the information processing device 100 performs the same processing for the output candidate content C2, as shown in Figure 2.

[0039] Specifically, the information processing device 100 detects a dialogue pattern based on a combination of the output message from the agent function unit 32 corresponding to content C2 and the response message from any user Ux included in the dialogue history LG2. Then, for each detected pattern, the information processing device 100 calculates the actual time required for the dialogue of that pattern.

[0040] Similar to the example of content C1, the information processing device 100 detects the pattern of the interaction when user Ux responds to an output message with a "positive" message. In the example in Figure 2, the information processing device 100 detects the interaction pattern PT21. The information processing device 100 also calculates the actual time required for the interaction of interaction pattern PT21 based on the actual interaction time when the interaction progressed in the interaction pattern PT21 from the interaction history LG2. The example in Figure 2 shows an example in which the information processing device 100 calculates the actual time 2y.

[0041] On the other hand, the information processing device 100 also detects the pattern of the interaction when the user Ux responds to an output message with a "negative" message. In the example in Figure 2, the information processing device 100 detects the interaction pattern PT22. In this example, the information processing device 100 also calculates the actual time required for the interaction of interaction pattern PT22 based on the actual interaction time when the interaction progressed in the interaction pattern PT22 from the interaction history LG2. The example in Figure 2 shows an example in which the information processing device 100 calculates the actual time 2n.

[0042] In this case, the information processing device 100 predicts the longer of the actual time 2y and the actual time 2n as the dialogue duration corresponding to content C2. As shown in Figure 2, let's assume the length relationship is actual time 2y > actual time 2n. In this example, the information processing device 100 predicts the longer actual time 2y as the dialogue duration 2y.

[0043] As explained in step S2, the information processing device 100 predicts the longer of the actual times calculated for each content, for each combination of messages (for example, the combination of messages when a "positive" response is given, and the combination of messages when a "negative" response is given), as the dialogue duration. For example, it can be difficult to reliably predict the user's response in a dialogue. Therefore, this processing is performed so that, even in the case of the longest dialogue time, the message exchange will be contained within the operating load reduction section SE11.

[0044] Next, using Figure 3, we will explain the processes of steps S3 and S4 of the control processing performed by the information processing device 100. Figure 3 is an explanatory diagram (2) illustrating the overall structure of the information processing according to this embodiment.

[0045] Up to this point, the information processing device 100 has predicted the dialogue duration for each of the candidate output contents, content C1 and C2. Specifically, the information processing device 100 predicts the dialogue duration 1n corresponding to content C1 and the dialogue duration 2y corresponding to content C2. Based on the relationship between these dialogue durations predicted in step S2 and the dialogue-possible time XX calculated in step S1, the information processing device 100 selects either content C1 or content C2 as the content to be output (step S3). For example, the information processing device 100 selects the content C1 or content C2 that has a predicted dialogue duration shorter than the dialogue-possible time XX as the content to be output.

[0046] Figure 3(a) shows an example in which the predicted dialogue duration 1n is longer than the dialogue-available time XX, while the predicted dialogue duration 2y is shorter than the dialogue-available time XX. In such a relationship, the information processing device 100 may select content C2, for which a predicted dialogue duration 2y is shorter than the dialogue-available time XX, as the content to be output.

[0047] Furthermore, Figure 3(b) shows an example in which the predicted dialogue duration 2y is longer than the dialogue-available time XX, while the predicted dialogue duration 1n is shorter than the dialogue-available time XX. In such a relationship, the information processing device 100 may select content C1, for which a predicted dialogue duration 1n is shorter than the dialogue-available time XX, as the content to be output.

[0048] Furthermore, the process in step S3 can be described as a selection process that takes safety measures into account to ensure that the dialogue is completed within the dialogue time XX, regardless of whether the dialogue between user U1 and content C1 or content C2 is performed, and regardless of whether user U1 gives a positive or negative response during the dialogue.

[0049] Now, let's assume that after some time has passed since step S1 was executed at time TM1, and it is now time TM2, vehicle VE1 has entered the driving load reduction section SE11. In this case, the information processing device 100 controls the output so that a dialogue corresponding to the content to be output selected in step S3 is initiated (step S4). This point will be explained using an example where content C2 is selected as the content to be output in step S3.

[0050] For example, the information processing device 100 may perform the following output control in accordance with the operation of the agent function unit 32. Specifically, the information processing device 100 controls the output to be an utterance made by the agent function unit 32 that corresponds to the text TX21 used to start the dialogue among the texts that make up the content C2. For example, if the agent function unit 32 is a voice agent with a female voice V1, the information processing device 100 may synthesize a voice in which the text TX21 is read aloud in the female voice V1, and control the output to be made from the speaker of the device. As a result, Figure 3 shows an example in which the speaker of the information processing device 100 outputs a synthesized voice saying, "Today is Yakiniku Day. How about yakiniku for lunch today?", in which the text TX21 is read aloud.

[0051] So far, we have used Figures 2 and 3 to explain the overall picture of the processing performed by the information processing device 100. With such an information processing device 100, it is possible to control the conversation to proceed in a way that prioritizes content whose output can be completed within the time available for conversation, thus making effective use of the time available for conversation in content output.

[0052] In the example shown in Figure 2, the information processing device 100 performs the processing of step S2 in real time. Specifically, the information processing device 100 performs in real time the process of predicting the duration of the dialogue for each candidate content based on the history information of the dialogue that took place for that content. However, the information processing device 100 does not necessarily have to perform this process in real time. For example, the information processing device 100 could predict the duration of the dialogue in advance, and then, when performing the process of selecting the content to be output, it could obtain the previously predicted duration of the dialogue and use it in the selection process.

[0053] [4. An example of predictive processing] Next, we will explain in more detail an example of a prediction process for predicting the duration of a dialogue, using Figures 4 and 5.

[0054] First, using Figure 4, we will explain an example of the prediction process until the dialogue duration 1n shown in step S2 of Figure 2 is predicted. Figure 4 is Figure (1) which shows a specific example of the prediction process according to the embodiment.

[0055] Figure 4(a) shows an example of a conversation between user U3 (an example of any user Ux) of vehicle VE3 and the agent function unit 31 that provides content C1. The example in Figure 4(a) also shows an example of a conversation between user U5 (an example of any user Ux) of vehicle VE5 and the agent function unit 31 that provides content C1. In the example in Figure 4(a), both users U3 and U5 respond with "positive" messages to the output messages from the agent function unit 31.

[0056] For example, Figure 4(a) shows an example where the agent function unit 31 initiates a dialogue when the information processing device 100 outputs message TX11, which reads aloud, "There is a delicious ramen restaurant nearby. Do you like ramen?". Also in Figure 4(a), an example is shown where both users U3 and U5 respond with "affirmative" messages in response to the output message TX11. Specifically, both users U3 and U5 utter message AN11, which means, "Yes, I like ramen." Furthermore, according to the example in Figure 4(a), in response to message AN12, the information processing device 100 outputs message TX12, which reads aloud, "I will continue to let you know about other highly-rated ramen restaurants nearby," as an utterance by the agent function unit 31.

[0057] If the information processing device 100 obtains an analysis result based on the dialogue history LG1 that dialogues with "positive" content, as shown in Figure 4(a), tend to include a combination of messages TX11, AN11, TX12, etc., then it can detect this tendency as dialogue pattern PT11. In this case, the information processing device 100 calculates the actual time required for the dialogue of dialogue pattern PT11.

[0058] For example, as shown in Figure 4(a), suppose the actual dialogue time when the dialogue pattern PT11 progressed with user U3 was "DTM13". Also, suppose the actual dialogue time when the dialogue pattern PT11 progressed with user U5 was "DTM15". In such cases, the information processing device 100 may statistically calculate the actual time using the dialogue time "DTM13" and the dialogue time "DTM15". As a result, Figure 4(a) shows an example in which the information processing device 100 calculated the actual time 1y.

[0059] Next, Figure 4(b) shows an example of a conversation between user U4 of vehicle VE4 (an example of any user Ux) and the agent function unit 31 that provides content C1. The example in Figure 4(b) also shows an example of a conversation between user U6 of vehicle VE6 (an example of any user Ux) and the agent function unit 31 that provides content C1. Furthermore, in the example in Figure 4(b), both users U4 and U6 respond with "negative" messages to the output messages from the agent function unit 31.

[0060] Figure 4(b) also shows an example where the agent function unit 31 initiated a dialogue when the information processing device 100 output message TX11, which reads out, "There's a delicious ramen shop nearby. Do you like ramen?" In response to this, Figure 4(b) shows an example where both users U4 and U6 responded with "negative" messages in response to the output message TX11. Specifically, both users U4 and U6 uttered message AN12, which means "No." Furthermore, according to the example in Figure 4(b), in response to message AN12, the information processing device 100 output message TX13, which reads out, "I will refrain from sending you information about ramen shops in the future," as an utterance by the agent function unit 31.

[0061] If the information processing device 100 obtains an analysis result based on the dialogue history LG1 indicating that dialogues with "negative" content, as shown in Figure 4(b), tend to include a combination of messages TX11, AN12, TX13, etc., then it can detect this tendency as dialogue pattern PT12. In this case, the information processing device 100 also calculates the actual time required for the dialogue of dialogue pattern PT12.

[0062] For example, as shown in Figure 4(b), suppose the actual dialogue time when the dialogue pattern PT12 proceeded with user U4 was "DTM14". Also, suppose the actual dialogue time when the dialogue pattern PT12 proceeded with user U6 was "DTM16". In such cases, the information processing device 100 may statistically calculate the actual time using the dialogue time "DTM14" and the dialogue time "DTM16". As a result, Figure 4(b) shows an example in which the information processing device 100 calculated the actual time 1n.

[0063] Here, as explained in step S2 of Figure 2, the information processing device 100 predicts the longer actual time 1n as the dialogue duration 1n, since the length relationship was actual time 1y < actual time 1n.

[0064] Next, using Figure 5, we will explain an example of the prediction process until the dialogue duration 2y shown in step S2 of Figure 2 is predicted. Figure 5 is Figure (2) showing a specific example of the prediction process according to the embodiment. Note that Figure 5 can be explained by following the example in Figure 4, so the details will be omitted.

[0065] Figure 5(a) shows an example of a conversation between user U3 of vehicle VE3 and agent function unit 32 that provides content C2. The example in Figure 5(a) also shows an example of a conversation between user U5 of vehicle VE5 and agent function unit 32 that provides content C2. In the example in Figure 5(a), both users U3 and U5 respond with "positive" messages to the output messages from agent function unit 32.

[0066] If the information processing device 100 obtains an analysis result based on the dialogue history LG2 that dialogues with "positive" content, as shown in Figure 5(a), tend to include a combination of messages TX21, AN21, TX22, etc., then it can detect this tendency as dialogue pattern PT21. In this case, the information processing device 100 calculates the actual time required for the dialogue of dialogue pattern PT21.

[0067] For example, as shown in Figure 5(a), suppose the actual dialogue time when the dialogue pattern PT21 progressed with user U3 was "DTM23". Also, suppose the actual dialogue time when the dialogue pattern PT21 progressed with user U5 was "DTM25". In such cases, the information processing device 100 may statistically calculate the actual time using the dialogue time "DTM23" and the dialogue time "DTM25". As a result, Figure 5(a) shows an example in which the information processing device 100 calculated the actual time 2y.

[0068] Furthermore, Figure 5(b) shows an example of a conversation between user U4 of vehicle VE4 and agent function unit 31 that provides content C2. The example in Figure 4(b) also shows an example of a conversation between user U6 of vehicle VE6 and agent function unit 32 that provides content C2. In the example in Figure 4(b), both users U4 and U6 respond with "negative" messages to the output message from agent function unit 32.

[0069] If the information processing device 100 obtains an analysis result based on the dialogue history LG2 that dialogues with "negative" content, as shown in Figure 5(b), tend to include a combination of messages TX21, AN22, TX23, etc., then it can detect this tendency as dialogue pattern PT22. In this case, the information processing device 100 calculates the actual time required for the dialogue of dialogue pattern PT22.

[0070] For example, as shown in Figure 5(b), suppose the actual dialogue time when the dialogue pattern PT22 proceeded with user U4 was "DTM24". Also, suppose the actual dialogue time when the dialogue pattern PT22 proceeded with user U6 was "DTM26". In such cases, the information processing device 100 may statistically calculate the actual time using the dialogue time "DTM24" and the dialogue time "DTM26". As a result, Figure 5(b) shows an example in which the information processing device 100 calculated the actual time 2n.

[0071] Here, as explained in step S2 of Figure 2, the information processing device 100 predicts the longer actual time 2y as the dialogue duration 2y, since the length relationship was actual time 2y > actual time 2n.

[0072] [5. Variations of prediction processing] Based on the explanation so far, the information processing device 100 has shown an example in which it calculates the longer of the actual time periods, 1y and 1n, as the dialogue duration 1n, given the relationship 1y < 1n. However, the information processing device 100 may also correct the dialogue duration 1n based on the situation of user U1, who is the target user for whom the content is currently being output. For example, if the information processing device 100 can identify user U1 as someone interested in topic NA1, which is the subject of content C1, it may estimate that user U1 wants to enjoy the conversation about topic NA1 with the agent function unit 31 more, and correct the dialogue duration 1n to extend it. As an example, the information processing device 100 may correct the dialogue duration 1n using a weight value corresponding to the degree of user U1's interest in topic NA1. In this case, the information processing device 100 may use a weight value that extends the dialogue duration 1n.

[0073] Furthermore, the information processing device 100 may perform similar correction processing for the dialogue duration 2y. For example, if the information processing device 100 determines that the user U1 is not interested in the topic NA2 targeted by content C2, it may estimate that the user U1 wants to end the dialogue with the agent function unit 32 quickly and correct the dialogue duration 2y to shorten it. As an example, the information processing device 100 may correct the dialogue duration 2y using a weight value corresponding to the degree of interest of the user U1 in topic NA2. In this case, the information processing device 100 may use a weight value that shortens the dialogue duration 2y.

[0074] [6. Configuration of the Information Processing Device] Next, an information processing device 100 according to the embodiment will be described with reference to Figure 6. Figure 6 is a diagram showing an example configuration of the information processing device 100 according to the embodiment. As shown in Figure 6, the information processing device 100 includes an application group 30, a communication unit 110, a storage unit 120, and a control unit 130.

[0075] (Regarding app group 30) The application group 30 is a group of applications that provide services related to driving assistance and useful information. The applications included in the application group 30 may each provide different types of services, or they may provide the same type of service but with different performance characteristics. In addition, each application included in the application group 30 has an agent function unit that acts as an intermediary for providing the service. Furthermore, each application (agent function unit) is associated with an agent device 300 that performs processing to realize the service corresponding to that application.

[0076] As shown in the example in Figure 6, the application group 30 includes an example where multiple applications AP30 are included, such as application AP31 and application AP32. For the sake of explanation, only applications AP31 and AP32 are shown in Figure 6.

[0077] According to the example in Figure 6, application AP31 has an agent function unit 31 as a voice agent function, and agent device 300-1 (an example of agent device 300) is associated with the agent function unit 31. Also according to the example in Figure 6, application AP32 has an agent function unit 32 as a voice agent function, and agent device 300-2 (an example of agent device 300) is associated with the agent function unit 32.

[0078] Furthermore, each application included in the application group 30 (app AP31, app AP32, etc.) passes the content it wants to provide to the user to the situation awareness engine 131, which then plays and outputs the content. At this time, the application may add range information indicating the range to be played and output, category information indicating the category of the content, the length of the content (playback time), etc., to the content.

[0079] Range information, which indicates the range within which content should be played back, is equivalent to conditional information that specifies the geographical range, time range, vehicle VEx mileage range, vehicle VEx passing area range, vehicle speed range, etc., within which content playback should be permitted. For this reason, the app makes a reservation with the situation awareness engine 131 regarding content playback.

[0080] Here, an example of content passed by the agent function unit 31 will be explained using Figure 7. For example, the agent device 300-1 corresponding to the agent function unit 31 may manage content in a predetermined database DA in the configuration shown in Figure 7. Figure 7 is a diagram showing an example of content configuration according to the embodiment.

[0081] In the example shown in Figure 7, the database DA of agent device 300-1 contains items such as "Content ID," "Topic," "Detection Text," and "Action Text."

[0082] The "Content ID" is an identifier that identifies the content of the corresponding "Topic." The "Topic" is information that indicates the topic covered by the content indicated by the "Content ID."

[0083] "Detected text" is text information that indicates candidate messages that the user of the conversation partner may utter in response to a message output as an utterance by the agent function unit 31.

[0084] The "action text" corresponds to conditional information that determines what text the agent function unit 31 should read aloud as an utterance in response to an utterance by the user with whom it is interacting. The "action text" is the text that constitutes the content indicated by the "content ID" (i.e., content C1), and may be text related to topic NA1. Thus, the "action text" can be said to be, in essence, the data of content C1 itself.

[0085] Furthermore, Figure 7 shows that in the first record, the action text "TX11" is registered. This example illustrates a condition in which the agent function unit 31 initiates a dialogue about topic NA1 by outputting message TX11 as an utterance from the agent function unit 31.

[0086] Furthermore, Figure 7 shows an example in which the detection text "AN11" and the action text "TX12" are associated in the second record. This example shows a condition in which, in response to message TX11 output as an utterance by the agent function unit 31, if the user utters message AN11, message TX12 is read aloud as an utterance by the agent function unit 31.

[0087] Figure 7 shows an example of the database DA of agent device 300-1, but agent device 300-2 may have a database DA with a similar configuration. For example, the database DA of agent device 300-2 may have "AN21" and "AN22" registered as detection texts and "TX21", "TX22", and "TX23" registered as operation texts, as shown in Figure 5.

[0088] (Regarding Communications Unit 110) The communication unit 110 is implemented, for example, by a NIC (Network Interface Card). The communication unit 110 is connected to the network N by wire or wireless connection and transmits and receives information with, for example, the agent device 300. The communication unit 110 can also transmit and receive information with a predetermined external device (not shown).

[0089] (Regarding memory unit 120) The storage unit 120 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or storage devices such as hard disks and optical discs. The storage unit 120 includes a section information database 121, an interaction history database 122, and a performance information database 123.

[0090] (Regarding the section information database 121) The section information database 121 stores section information indicating sections where the user's driving burden is estimated to be low based on the degree of driving safety. Here, Figure 8 shows an example of the section information database 121 according to the embodiment. In the example in Figure 8, the section information database 121 has items such as "road information," "section ID," and "section location."

[0091] "Road information" refers to information indicating roads that include sections where the driving load is estimated to be low (driving load reduction sections). "Road information" may be, for example, the name of the road that includes the driving load reduction section, or a link ID that identifies the road that includes the driving load reduction section.

[0092] "Section ID" is identification information that identifies a section of reduced operating load. "Section Location" is location information of the section of reduced operating load identified by "Section ID". "Section Location" may include, for example, location information of one end and location information of the other end that constitute the section of reduced operating load identified by "Section ID".

[0093] (Regarding dialogue history database 122) The dialogue history database 122 stores history information of conversations that take place between the user of the information processing device 100 and the agent function unit installed in the information processing device 100. Here, Figure 9 shows an example of the dialogue history database 122 according to the embodiment. In the example in Figure 9, the dialogue history database 122 has items such as "Content ID", "User ID", "Dialogue ID", "Date and Time", "Agent Message", "User Message", and "Dialogue Time".

[0094] The "Content ID" indicates identification information that identifies the content used in the interaction with the user identified by the "User ID". The "User ID" indicates identification information that identifies the user who interacted with the agent function unit that provided the content identified by the "Content ID". The "Dialogue ID" indicates identification information that identifies the dialogue in question. The dialogue identified by the "Dialogue ID" may be a series of conversations that began with an utterance by the agent function unit, followed by an exchange of messages between the agent function unit and the user, and finally ended with an utterance by the agent function unit.

[0095] The "Date and Time" indicates the date and time when the message was spoken by the agent function unit or the user. In other words, the "Date and Time" indicates the date and time when the information processing device 100 detected that the message was spoken by the agent function unit or the user.

[0096] An "agent message" is information that indicates a message (text) output as an utterance by the agent function unit that provided the content identified by the "content ID". For example, the agent message "AMG311" or "AMG313" associated with the content identified by content ID "C1" (content C1) may be one of the "action texts" explained in Figure 7.

[0097] "User message" is information that indicates the message (text) spoken by the user identified by "User ID" during a conversation (the conversation identified by "Conversation ID") with the agent function unit that provided the content identified by "Content ID". For example, the user message "UMG312" associated with the content identified by Content ID "C1" (Content C1) may be one of the "detection texts" explained in Figure 7.

[0098] "Dialogue time" is information that indicates the actual time spent from the start to the end of a dialogue, identified by the "Dialogue ID," based on the exchange between "Agent messages" and "User messages."

[0099] (Regarding the performance information database 123) The performance information database 123 stores information about actual time. Figure 10 shows an example of the performance information database 123 according to this embodiment. In the example in Figure 10, the performance information database 123 has items such as "Content ID", "Pattern ID", "Message Information", and "Actual Time".

[0100] "Content ID" is identification information that identifies the content used in the interaction with the user. "Pattern ID" is identification information that identifies the pattern of the interaction. For example, "Pattern ID" is identification information that identifies a combination of messages statistically detected based on the messages contained in the interaction conducted using the content identified by "Content ID" as a single pattern.

[0101] "Message information" is information that indicates the combination of messages that constitute the dialogue pattern identified by the "Pattern ID".

[0102] "Actual time" indicates the actual time spent when message exchanges progressed according to the dialogue pattern identified by "Pattern ID". "Actual time" may be statistically calculated based on, for example, the actual dialogue time ("Dialogue time" in the dialogue history database 122) when message exchanges progressed according to the dialogue pattern identified by "Pattern ID".

[0103] (Regarding the control unit 130) Returning to Figure 6, the control unit 130 is realized by a CPU (Central Processing Unit) or MPU (Micro Processing Unit), etc., executing various programs (for example, the information processing program according to the embodiment) stored in the memory device inside the information processing device 100 using RAM as the working area. Alternatively, the control unit 130 can be realized by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array).

[0104] As shown in Figure 6, the control unit 130 includes a situation awareness engine 131, a prediction unit 132, an acquisition unit 133, a calculation unit 134, a selection unit 135, and an output control unit 136, and realizes or executes the information processing functions and operations described below. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in Figure 3, and other configurations are also possible as long as they perform the information processing described later. Also, the connection relationships of the various processing units in the control unit 130 are not limited to the connection relationships shown in Figure 3, and other connection relationships are also possible.

[0105] (Regarding the situation awareness engine 131) The situation awareness engine 131 grasps various situations related to the vehicle VEx and controls the output of content provided by the agent function unit (for example, agent function unit 31 or agent function unit 32, etc.) according to the grasped situation.

[0106] For example, the situation awareness engine 131 detects driving behavior (such as the user's driving status or the driving status of the vehicle VEx) from sensor information detected by sensors on the vehicle VEx, and estimates the degree of driving safety (degree of driving comfort) based on the detected driving behavior.

[0107] Furthermore, if the situation awareness engine 131 determines, for example, that the user's driving load is low (there is leeway in driving) because the safety level is higher than a predetermined value, it will initiate a dialogue based on the content provided by the agent function unit. On the other hand, if the situation awareness engine 131 determines that the user's driving load is high (there is no leeway in driving) because the safety level is lower than a predetermined value, it may not initiate a dialogue based on the content provided by the agent function unit, or it may interrupt the dialogue that is currently in progress.

[0108] (Regarding prediction unit 132) The prediction unit 132 predicts the output time required to complete the output of each candidate content based on predetermined information about that content. For example, for each candidate content, the prediction unit 132 predicts the dialogue time required for a dialogue between the agent function unit providing the content and the user, where the dialogue about the content is discussed, as the output time.

[0109] For example, the prediction unit 132 calculates the actual time required for past conversations based on the history information of past conversations conducted in accordance with the output candidate content (or content related to the output candidate content), and uses the calculated actual time to predict the conversation duration. For example, the prediction unit 132 calculates the actual time required for each combination of messages exchanged between the agent function unit and the user in past conversations, and uses the calculated actual time to predict the conversation duration.

[0110] For example, the prediction unit 132 may predict the dialogue duration as the actual time that satisfies predetermined conditions from the actual time calculated for each message combination. For example, the prediction unit 132 may predict the dialogue duration as the longer actual time from the actual time calculated for each message combination.

[0111] In the example shown in Figure 2, the prediction unit 132 may perform the prediction process described in step S2.

[0112] (Regarding acquisition section 133) The acquisition unit 133 acquires section information indicating sections where the user's driving burden is estimated to be low. For example, the acquisition unit 133 may acquire section information indicating sections where the driving burden is statistically estimated to be low based on the degree of driving safety. Alternatively, the acquisition unit 133 may acquire section information indicating sections where the driving burden is estimated to be low among roads shown in predetermined map information.

[0113] (Regarding calculation unit 134) The calculation unit 134 calculates the time required for the vehicle VEx, which is being driven by the user to whom the content is to be output, to pass through a section where the driving load is estimated to be low, as the output time during which the content can be output. For example, the calculation unit 134 calculates the dialogue time during which the user can respond to a conversation when the agent function unit provides the content that is the topic of discussion, as the output time.

[0114] In the example shown in Figure 2, the calculation unit 134 may perform the calculation process described in step S1.

[0115] (Regarding selection section 135) The selection unit 135 selects content from the output candidate content that is estimated to be output within the output time, based on the output time available and the output time required, as content to be output. For example, the selection unit 135 selects content from the output candidate content that is estimated to be output that will be appropriately completed within the interaction time, based on the interaction time available and the interaction time required, as content to be output.

[0116] For example, the selection unit 135 selects content to be output from among the output candidate content based on the relationship between the predicted dialogue duration for each output candidate content and the available dialogue time. For example, the selection unit 135 may select content from among the output candidate content whose dialogue duration is shorter than the available dialogue time as content to be output.

[0117] In the example shown in Figure 3, the calculation unit 134 may perform the calculation process described in step S3.

[0118] (Regarding the output control unit 136) The output control unit 136 controls the output to produce utterances corresponding to the content being output. For example, if the user's vehicle VEx, which is the target of the content output, enters a section where the driving load is estimated to be low, the output control unit 136 may control the output to produce utterances corresponding to the content being output, which are utterances produced by the agent function unit that provided the content being output.

[0119] In the example shown in Figure 3, the calculation unit 134 may perform the calculation process described in step S4.

[0120] [7. Processing Procedure] Next, the information processing procedure implemented by the information processing method according to the embodiment will be described using Figures 11 and 12. The information processing procedure according to the embodiment can be broadly divided into a preliminary step of predicting the dialogue time in advance and a step of selecting the content to be output in real time using this dialogue time. Therefore, Figures 11 and 12 will explain the processing procedure in each step.

[0121] [7-1. Processing Procedure (1)] First, Figure 11 will be used to explain the procedure for predicting the duration of a conversation. Figure 11 is a flowchart showing the prediction procedure for predicting the duration of a conversation.

[0122] First, the prediction unit 132 determines whether it is time to predict the duration of the conversation (step S1101). For example, the prediction unit 132 may determine that it is time to predict the duration of the conversation when a specific time period arrives each day. Alternatively, the prediction unit 132 may determine that it is time to predict the duration of the conversation when, for example, the agent function unit passes candidate content to the situation awareness engine 131.

[0123] If the prediction unit 132 determines that it is not yet time to predict the dialogue duration (step S1101; No), it waits until it determines that it is now time to predict the dialogue duration.

[0124] On the other hand, if the prediction unit 132 determines that it is time to predict the time required for the dialogue (step S1101; Yes), it determines whether or not there is currently any content that can be output (step S1102).

[0125] For example, if the prediction unit 132 determines that the agent function unit has not provided any candidate content for output and that no candidate content exists (step S1102; No), it may restart the process from step S1101.

[0126] On the other hand, if the prediction unit 132 determines that candidate content exists because the agent function unit has provided candidate content (step S1102; Yes), it extracts the dialogue history of the candidate content from the dialogue history database 122 (step S1103).

[0127] Next, the prediction unit 132 detects multiple dialogue patterns based on the dialogue history (step S1104). For example, the prediction unit 132 may detect dialogue patterns based on the combination of messages exchanged between the agent function unit that provided the output candidate content and the user, which constitute the dialogue history. As an example, the prediction unit 132 may detect the tendency of what kind of message combination will occur when the user responds with a message of "positive" content to a message output as an utterance by the agent function unit, as one dialogue pattern. The prediction unit 132 may also detect the tendency of what kind of message combination will occur when the user responds with a message of "negative" content to a message output as an utterance by the agent function unit, as another dialogue pattern.

[0128] Next, the prediction unit 132 calculates the actual time required for the dialogue corresponding to each pattern detected in step S1104 (step S1105). The prediction unit 132 may statistically calculate the actual time using the dialogue time spent in each dialogue corresponding to the detected pattern.

[0129] Next, the prediction unit 132 predicts the longer of the actual times calculated for each dialogue pattern as the dialogue duration required for a dialogue concerning the topic targeted by the output candidate content, from the time the dialogue between the agent function unit providing the output candidate content and the target user is completed (step S1106).

[0130] Next, the prediction unit 132 determines whether or not it has predicted the dialogue duration for all output candidate content (step S1107). If the prediction unit 132 finds that there is any output candidate content for which it has not predicted the dialogue duration (step S1107; No), it repeats the process from step S1103.

[0131] On the other hand, if there are no output candidate contents for which the dialogue duration cannot be predicted (step S1107; Yes), the prediction unit 132 may, for example, register the predicted dialogue duration in the actual information database 123 and terminate the process.

[0132] [7-2. Processing Procedure (2)] Next, we will explain the procedure for selecting the content to be output using Figure 12. Figure 12 is a flowchart showing the selection procedure for selecting the content to be output.

[0133] First, the acquisition unit 133 acquires section information indicating a section where the operating load is estimated to be low (step S1201).

[0134] Next, the calculation unit 134 determines whether the vehicle on which the device (information processing device 100) is installed is scheduled to pass through the driving load reduction section (step S1202). For example, if the calculation unit 134 determines that the vehicle is not scheduled to pass through the driving load reduction section (step S1202; No), it may wait until it can determine that the vehicle is scheduled to pass through the driving load reduction section.

[0135] On the other hand, if the calculation unit 134 determines that the vehicle is scheduled to pass through the section where the driving load is reduced because the vehicle has reached a predetermined point before the section where the driving load is reduced (step S1202; Yes), it recognizes the driver of the vehicle as a target user for which the content will be output (step S1203).

[0136] Then, the calculation unit 134 calculates the amount of time available for the target user to respond to a conversation initiated by the agent function unit using the topic of conversation, based on the speed of the vehicle and the distance of the section where the driving load is reduced (step S1204). For example, the calculation unit 134 calculates the time required for the vehicle to pass through the section where the driving load is reduced, based on the speed of the vehicle and the distance of the section where the driving load is reduced, and sets this as the available time for conversation.

[0137] Next, the selection unit 135 obtains the predicted dialogue duration for each output candidate content (step S1205). As explained in Figure 11, the dialogue duration is predicted in advance by the prediction unit 132 in a separate phase. Therefore, for example, the selection unit 135 may obtain the dialogue duration from the prediction unit 132.

[0138] The selection unit 135 then determines whether or not there is any content among the output candidate content whose dialogue duration is shorter than the dialogue time (step S1206). If there is no content among the output candidate content whose dialogue duration is shorter than the dialogue time (step S1206; No), the selection unit 135 may determine that there is no content to be output for dialogue between the agent function unit and the target user in the reduced operating load section and terminate the process.

[0139] On the other hand, if the selection unit 135 determines that there is content among the output candidate content whose dialogue duration is shorter than the dialogue time (step S1206; Yes), it determines whether there are multiple pieces of content whose dialogue duration is shorter than the dialogue time (step S1207).

[0140] If the selection unit 135 determines that there are not multiple contents whose dialogue time is shorter than the dialogue time, that is, that there is only one content whose dialogue time is shorter than the dialogue time (step S1207; No), it selects this one output candidate content as the content to be output (step S1208a).

[0141] On the other hand, if the selection unit 135 determines that there are multiple contents whose dialogue duration is shorter than the dialogue time (step S1207; Yes), it may select the content with the shortest dialogue duration from among the multiple contents as the content to be output (step S1208b). However, in step S1208, the selection unit 135 may, instead of selecting the content with the shortest dialogue duration as the content to be output, select the content with the longest dialogue duration as the content to be output.

[0142] Finally, the output control unit 136 controls the output so that a dialogue about the content to be output is initiated by the agent function unit that provided the content (step S1209). For example, if content C2 is selected as the content to be output, the output control unit 136 controls the output so that an utterance made by the agent function unit 32 is output, which corresponds to the text TX21 used to start the dialogue among the texts that make up content C2. For example, if the agent function unit 32 is a voice agent with a female voice V1, the output control unit 136 may synthesize a voice in which the text TX21 is read aloud in the female voice V1, and then control the output so that the synthesized voice is output from the speaker of the device.

[0143] [8. Restriction lifted] In the above embodiment, an example was shown in which the information processing device 100 selects content to be used for interaction with the user. Specifically, the example was shown in which the information processing device 100 calculates the time available for interaction, predicts the time required for interaction for each candidate content to be output, and selects as the content to be output content that is estimated to allow the interaction to be completed appropriately within the time available for interaction, based on the time available for interaction and the time required for interaction.

[0144] However, the content selected by the information processing device 100 does not necessarily have to be content used for interaction with the user; for example, it may be content that is unilaterally output to the user. Examples of such content include content that serves as an advertisement, such as introducing nearby shops or recommended spots through spoken voice from the agent function unit.

[0145] When providing such advertising content to a user, the information processing device 100 calculates the time required for the vehicle VEx to pass through the driving load reduction section, based on the speed of the vehicle VEx driven by the target user to whom the advertising content is to be output, and the distance of the driving load reduction section, as the output time during which the advertising content can be played back. The information processing device 100 also predicts the output time required to complete the playback of each candidate advertising content based on predetermined information related to that advertising content. Then, based on the output time and the output time, the information processing device 100 selects from the candidate advertising content that is estimated to be playable within the output time as the advertising content to be output.

[0146] The specified information regarding advertising content referred to herein may be historical information of advertising content that has been played back in the past. Furthermore, such historical information may include, for example, the actual time required for playback for each piece of advertising content.

[0147] Furthermore, the information processing device 100 may be a server device on the cloud side, rather than an in-vehicle device on the edge side. In other words, the information processing according to the above embodiment may be performed by the information processing device 100 as a server device. In this case, the information processing device 100 transmits the content selected by the information processing according to the embodiment to an in-vehicle device installed in the vehicle VEx of the target user, and the in-vehicle device may, for example, control the output so that the content acquired from the information processing device 100 is output as speech by the agent function unit that provided the content.

[0148] Furthermore, the information processing device 100 may select the content to be output based on the attribute information of the candidate content.

[0149] For example, suppose that conditional information specifying the location where the output should be placed is associated with the output candidate content. In such a case, the information processing device 100 may preferentially select the content among the output candidate content that is conditioned to be closer to the current driving position of the vehicle VEx.

[0150] As another example, suppose that conditional information specifying the time at which output should be generated is associated with the output candidate content. In such a case, the information processing device 100 may prioritize selecting the content from the output candidate content that is conditioned to have a time closer to the current time when the vehicle VEx is running.

[0151] As another example, suppose that categories are associated with the output candidate content. In such a case, the information processing device 100 may preferentially select content that has been assigned a category that matches (or is similar to) matters that the user is presumed to be interested in.

[0152] [9. Hardware Configuration] Furthermore, the information processing device 100 according to the above-described embodiment is realized by a computer 1000 having the configuration shown in Figure 13. Figure 13 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing device 100. The computer 1000 has a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

[0153] The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400, controlling various components. The ROM 1300 stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0154] The HDD1400 stores programs executed by the CPU1100, as well as data used by such programs. The communication interface1500 receives data from other devices via a predetermined communication network and sends it to the CPU1100, and transmits data generated by the CPU1100 to other devices via the predetermined communication network.

[0155] The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice, via the input / output interface 1600. The CPU 1100 acquires data from input devices via the input / output interface 1600. The CPU 1100 also outputs the generated data to output devices via the input / output interface 1600.

[0156] The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1200 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0157] For example, when the computer 1000 functions as an information processing device 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes the functions of the control unit 130 by executing a program loaded on the RAM 1200. The CPU 1100 of the computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be obtained from other devices via a predetermined communication network.

[0158] [10. Other] Furthermore, among the processes described in each of the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above document and drawings can be changed at will unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0159] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0160] Furthermore, the above embodiments can be combined as appropriate, provided that the processing content is not contradictory.

[0161] Although some embodiments of the present invention have been described in detail above with reference to the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention. [Explanation of symbols]

[0162] 1. Information Processing System 100 Information Processing Devices 120 Storage section 121 Section Information Database 122 Dialogue History Database 123 Performance Information Database 130 Control Unit 131 Situation Awareness Engine 132 Prediction Section 133 Acquisition Department 134 Calculation Section 135 Selection Section 136 Output Control Unit

Claims

1. An acquisition unit that acquires section information indicating a section in which the user's driving load is estimated to be low, A calculation unit calculates the time required for the vehicle to pass through the section, based on the speed of the vehicle being driven by the target user for whom the content is to be output, as the output time during which the content can be output. For each candidate content to be output, a prediction unit predicts the time required to complete the output of that content based on predetermined information about that content. A selection unit selects content from the output candidate content that is estimated to be output within the output time based on the output availability time and the output required time, as content to be output. It has, The calculation unit calculates the available time for dialogue, which is the time during which the target user can respond to dialogue regarding the content provided by the agent function unit, as the available output time. The prediction unit predicts, for each of the output candidate contents, the time required for the dialogue related to the content, from the time the dialogue between the agent function unit providing the content and the target user ends, as the output time. Based on the available dialogue time and the required dialogue time, the selection unit selects from the output candidate content that is estimated to allow the dialogue to be completed appropriately within the available dialogue time as the content to be output. An information processing device characterized by the following:

2. The prediction unit calculates the actual time required for past dialogues based on the history information of past dialogues conducted according to the content of the output candidate, and uses the calculated actual time to predict the dialogue duration. The information processing apparatus according to feature 1.

3. The prediction unit calculates the actual time required for each combination of messages exchanged with the user in the past conversations, as the actual time. The information processing apparatus according to feature 2.

4. The prediction unit predicts the longer actual time among the actual times calculated for each combination as the dialogue duration. The information processing apparatus according to claim 3.

5. The selection unit selects the content to be output from among the output candidate content based on the relationship between the predicted dialogue duration for each output candidate content and the available dialogue time. The information processing apparatus according to any one of claims 1 to 4.

6. The selection unit selects from the output candidate content that has a dialogue duration shorter than the dialogue duration as the content to be output. The information processing apparatus according to feature 5.

7. The system further includes an output control unit that controls the output of speech corresponding to the content to be output. The information processing apparatus according to any one of claims 1 to 6.

8. The output control unit controls the output to output speech corresponding to the content to be output, which is speech made by the agent function unit that provided the content to be output, when the vehicle enters the section. The information processing apparatus according to feature 7.

9. The acquisition unit acquires section information indicating sections of roads shown in predetermined map information that are estimated to have low driving load. The information processing apparatus according to feature 1.

10. The selection unit selects the content to be output in real time at a predetermined timing before the vehicle enters the section. The information processing apparatus according to feature 1.

11. An information processing method performed by an information processing device, A process to acquire section information indicating sections where the user's driving load is estimated to be low, A calculation step of calculating the time required for the vehicle to pass through the section, based on the speed of the vehicle being driven by the target user for whom the content is to be output, as the output time during which the content can be output; For each candidate content for output, a prediction step is made to predict the time required to complete the output of that content based on predetermined information about that content. A selection step in which, based on the output availability time and the output time required, content from among the output candidate content that is estimated to be output within the output availability time is selected as content to be output. Includes, The calculation process involves calculating the time during which the target user can respond to dialogue regarding the content provided by the agent function unit, as the outputable time. The prediction step predicts, for each of the candidate output contents, the time required for the dialogue between the agent function unit providing the content and the target user until the dialogue regarding that content is completed, as the output time. The selection step selects, based on the available dialogue time and the required dialogue time, the content from the output candidate content that is estimated to be able to be appropriately completed within the available dialogue time as the content to be output. An information processing method characterized by the following:

12. An information processing program executed by an information processing device, Procedure for obtaining section information indicating sections where the user's driving load is estimated to be low, A calculation procedure for calculating the time required for a vehicle to pass through a given section, based on the speed of the vehicle being driven by the target user to whom the content is to be output, as the available output time for content output; A prediction procedure for each candidate content item, which predicts the time required to complete the output of that content based on predetermined information about that content, A selection procedure for selecting content from the output candidate content that is estimated to be output within the output time based on the output availability time and the output time required, to be output as the content to be output. The information processing device is made to execute this, The calculation procedure involves calculating the time during which the target user can respond to dialogue regarding the content provided by the agent function unit, as the outputable time. The prediction procedure predicts, for each of the candidate output contents, the time required for the dialogue between the agent function unit providing the content and the target user until the dialogue regarding that content is completed, as the output time. The selection procedure, based on the available dialogue time and the required dialogue time, selects from the candidate content for output that is estimated to allow the dialogue to be completed appropriately within the available dialogue time as the content to be output. An information processing program for that purpose.

Images