Vehicle attention reminder device, vehicle attention reminder method and recording medium

Abstract

The present invention provides a vehicle attention reminder device, a vehicle attention reminder method and a recording medium. The vehicle attention reminder device (2000) provides attention reminders when an occupant is having a conversation containing attention information related to driving. The vehicle attention reminder device (2000) comprises: a voice input unit (300) which inputs the voice of an occupant (10, 12, 14, 16, 18, 20) of a vehicle (1000); a speech inference unit (400) which infers speech based on the voice; a attention information determination unit (500) which determines whether the speech contains attention information related to driving of the vehicle; and an output unit (800) which outputs notification information when the speech contains attention information.

Description

Technical Field

[0001] The invention relates to a vehicle attention reminding device, a vehicle attention reminding method and a recording medium. Background Art

[0002] In the past, for example, the following patent document 1 describes an in-vehicle information device that is installed in a vehicle and provides information to assist passengers. The in-vehicle information device includes: a sound acquisition unit that detects and acquires sounds made by passengers; a voice recognition unit that recognizes the speech content of the above sounds acquired by the sound acquisition unit; an output control unit that generates and controls the output of the above speech content recognized by the voice recognition unit as display data or sound data; and an output unit that outputs the display data or sound data generated by the output control unit.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: International Publication No. 2014 / 002128 Summary of the invention

[0006] Technical issues

[0007] When driving a vehicle, passengers sometimes have conversations that include attention information related to driving. In this case, if the attention information is transmitted to the driver of the vehicle in advance, the driver can focus on driving, which is very convenient. However, in the technology described in the above-mentioned patent document 1, although it is assumed that the speech content uttered by the passenger is recognized and output as display data or sound data, there is no consideration of the processing method for the attention information related to driving included in the speech content.

[0008] Therefore, the present invention is completed in view of the above-mentioned problems, and an object of the present invention is to provide a new and improved vehicle attention reminder device, vehicle attention reminder method and recording medium that can provide attention reminders when the occupants are having a conversation containing attention information related to driving.

[0009] Technical Solution

[0010] In order to solve the above-mentioned problems, according to one aspect of the present invention, there is provided a vehicle attention reminder device, which comprises: a voice input unit, which inputs the voice of the vehicle occupant; a speech inference unit, which infers the speech based on the above-mentioned voice; a attention information determination unit, which determines whether the above-mentioned speech contains attention information related to the driving of the above-mentioned vehicle; and a notification unit, which notifies when the above-mentioned speech contains the above-mentioned attention information.

[0011] When the utterance includes the attention information, the output unit may output the utterance.

[0012] When the utterance includes the attention information, the output unit may emphasize and output the utterance. In addition, the output unit may output or send information not only inside the vehicle but also outside the vehicle.

[0013] In addition, the vehicle warning device may include a text conversion unit that converts the inferred speech into text information, and the notification unit includes a voice conversion unit that converts the text information into voice, and outputs the notification to a speaker. In addition, the conversion unit may perform translation, summary, etc.

[0014] Furthermore, the notification unit may include a voice emphasis unit for emphasizing a volume of the converted voice when the attention information is included in the utterance.

[0015] Furthermore, the vehicle attention alerting device may include a text conversion unit that converts the estimated utterance into text information, and the notification unit may output the notification to a display device.

[0016] Furthermore, the notification unit may include a text emphasis unit for emphasizing the text information when the warning information is included in the utterance.

[0017] In addition, the vehicle attention reminding device may include a recipient of information determining unit for determining a recipient of information as an output target of the attention information, and the recipient of information determining unit determines the driver of the vehicle as the recipient of information when the attention information is included in the speech.

[0018] Furthermore, when the caution information is included in the utterance, the information presentation recipient identification unit may identify all occupants of the vehicle as the information presentation recipients.

[0019] In addition, in order to solve the above-mentioned problem, according to another aspect of the present invention, a vehicle attention reminder method is provided, which includes: a step of inputting the voice of a vehicle occupant; a step of inferring speech based on the above-mentioned voice; a step of determining whether the above-mentioned speech contains attention information related to the driving of the above-mentioned vehicle; and a step of notifying when the above-mentioned speech contains the above-mentioned attention information.

[0020] In addition, in order to solve the above-mentioned problems, according to another aspect of the present invention, a recording medium is provided for enabling a computer to function as the following devices: a device for inputting the voice of a vehicle occupant; a device for inferring speech based on the above-mentioned voice; a device for determining whether the above-mentioned speech contains attention information related to the driving of the above-mentioned vehicle; and a device for notifying when the above-mentioned speech contains the above-mentioned attention information.

[0021] Effects of the Invention

[0022] According to the present invention, when the occupant is having a conversation including caution information related to driving, it is possible to call for caution. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a schematic diagram showing a planar structure of a vehicle according to one embodiment of the present invention.

[0024] Figure 2 Schematic diagram showing the configuration of the vehicle attention alerting device according to the present embodiment.

[0025] Figure 3 1 is a flowchart showing the processing performed by the vehicle attention calling device according to the present embodiment.

[0026] Explanation of symbols

[0027] 10, 12, 14, 16, 18, 20: Crew

[0028] 200, 202, 204, 206, 208, 210: Speakers

[0029] 250, 252, 254, 256, 258, 260: Display (display device)

[0030] 300: Voice input unit

[0031] 400: Discourse Inference Department

[0032] 450: Text Conversion Department

[0033] 500: Attention information judgment unit

[0034] 600: Information notification recipient confirmation unit

[0035] 700: Speech Synthesis Department

[0036] 710: Text / speech conversion unit

[0037] 720: Speech Emphasis

[0038] 800: Output

[0039] 900: Text synthesis department

[0040] 910: Text emphasis

[0041] 1000: Vehicle

[0042] 2000: Vehicle warning device DETAILED DESCRIPTION

[0043] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. It should be noted that in this specification and the accompanying drawings, components having substantially the same functional configuration are denoted by the same reference numerals to omit repeated descriptions.

[0044] Figure 1 1 is a schematic diagram showing a planar structure of a vehicle 1000 according to an embodiment of the present invention. Figure 1 As shown, there are six passengers 10, 12, 14, 16, 18, 20 in the vehicle. Among the six passengers, passenger 10 is the driver of the vehicle.

[0045] A plurality of speakers 200, 202, 204, 206, 208, and 210 are provided in the vehicle 1000. The speaker 200 emits sound toward the driver 10. Similarly, the speakers 202, 204, 206, 208, and 210 are provided corresponding to the adjacent passengers 12, 14, 16, 18, and 20, respectively, and emit sound toward the passengers 12, 14, 16, 18, and 20.

[0046] In addition, a plurality of displays (display devices) 250, 252, 254, 256, 258, and 260 are provided in the vehicle 1000. The display 250 displays information to the driver 10. Similarly, the displays 252, 254, 256, 258, and 260 are provided corresponding to the adjacent passengers 12, 14, 16, 18, and 20, and display information to the passengers 12, 14, 16, 18, and 20, respectively.

[0047] In addition, three microphones 100, 102, and 104 are provided in the vehicle 1000. Microphone 100 collects the voices of front row passengers 10 and 12. Microphone 102 collects the voices of middle row passengers 14 and 16. Microphone 104 collects the voices of rear row passengers 18 and 20.

[0048] However, there are cases where sound interferes with conversation in vehicle 1000. For example, when listening to music in the vehicle or when the windows are open, there are cases where sound other than conversation interferes with conversation.

[0049] In the present embodiment, the voices of the passengers 10, 12, 14, 16, 18, 20 are collected by the microphones 100, 102, 104 and outputted from the speakers 200, 202, 204, 206, 208, 210. In particular, in the present embodiment, the voices collected by the microphones 100, 102, 104 are temporarily converted into text information (text information), and then converted again from the text information into voices, and outputted from the speakers 200, 202, 204, 206, 208, 210. In addition, after the voices collected by the microphones 100, 102, 104 are converted into text information, the text information is displayed on the displays 250, 252, 254, 256, 258, 260. Thus, it is possible to suppress the hindrance to the conversation. It should be noted that when the voices are outputted from the speakers 200, 202, 204, 206, 208, 210, the text conversion may not be performed.

[0050] In addition, in the conversation of the passengers, attention information related to the driving of the vehicle is included. For example, when the passengers have conversations with each other such as "A vehicle is approaching from the right!", "The end of the traffic jam is approaching ahead", "The distance between the vehicle in front is shortening", "The road ahead is frozen", etc., these conversations contain attention information. Therefore, in the present embodiment, it is determined whether the conversation contains attention information, and if attention information is included, the speech is emphasized and output from the speakers 200, 202, 204, 206, 208, and 210. In particular, in the present embodiment, if attention information is included in the conversation, the speech is emphasized and output from the speaker 200 to the passenger 10 who is the driver. As a result, the driver can notice the phenomenon that is the root cause of the attention information and can take appropriate measures.

[0051] For example, when the conversation includes the warning information "A vehicle is approaching from the right!", the driver can avoid contact with the approaching vehicle by outputting the words from the speaker 200 of the passenger 10 who is the driver, so that the driver can take appropriate actions such as turning the steering wheel to the left. In addition, when the conversation includes the warning information "The road ahead is frozen", the driver can avoid skidding by taking appropriate actions such as depressing the brakes in advance to reduce the speed.

[0052] Figure 2 2 is a schematic diagram showing the structure of the vehicle warning device 2000 according to the present embodiment. Figure 2As shown, the vehicle attention reminding device 2000 is configured to have microphones 100, 102, 104, speakers 200, 202, 204, 206, 208, 210, displays 250, 252, 254, 256, 258, 260, and also has a voice input unit 300, a speech inference unit 400, a attention information determination unit 500, a person to be reminded of the information determination unit 600, and an output unit 800. The output unit 800 has a voice synthesis unit 700, and the voice synthesis unit 700 includes a text / voice conversion unit 710 and a voice emphasis unit 720. In addition, the output unit 800 has a text synthesis unit 900, and the text synthesis unit 900 includes a text emphasis unit 910. It should be noted that Figure 2 Each component of the illustrated vehicle caution alerting device 2000 may be constituted by a circuit (hardware) or a central processing unit such as a CPU and a program (software) for making it function.

[0053] The voices of the passengers 10, 12, 14, 16, 18, 20 collected by the microphones 100, 102, 104 are input to the voice input unit 300. The input voices are sent to the speech inference unit 400. In the speech inference unit 400, voice recognition is performed to infer the speech contained in the conversation between the passengers. It should be noted that the speech inference unit 400 can infer not only the speech but also the speech (the content of the speech). Examples of speech include "man!", "stone!", etc. uttered by the passengers.

[0054] Specifically, in the speech inference unit 400, the processing is roughly divided into two functions, namely, sound analysis and speech recognition. In the sound analysis, the sound is symbolized, noise is processed, and corrected. In speech recognition, the symbolized speech data obtained by the sound analysis is subjected to speech recognition through sound processing and language processing, and the words or character strings with the highest probability are extracted. Therefore, the speech inference unit 400 includes a text conversion unit 450 that converts the speech data into text information. The sound analysis and speech recognition use a sound model and a language model. In addition, a dictionary is registered with the rigid patterns, words or character strings used by each model, and the recognition rate can be improved by enriching the dictionary. It should be noted that the speech inference performed by the speech inference unit 400 can appropriately use, for example, the method described in the above-mentioned patent document 1, the method described in Japanese Patent Gazette No. 2003-323191, etc.

[0055] If the speech data is recognized by the speech inference unit 400 and converted into text information, the text information is sent to the speech synthesis unit 700 and the text synthesis unit 900 of the output unit (notification unit) 800. In addition, the text information is also sent to the attention information determination unit 500. In the attention information determination unit 500, it is determined whether the text information contains attention information. For example, the attention information determination unit 500 determines whether the text information contains attention information by performing keyword detection. As keywords, for example, "traffic jam", "rain", "snow", "rear vehicle", "adjacent lane", "road surface", "gasoline", etc. can be cited. The attention information determination unit 500 includes a database storing these keywords. When these keywords are included in the text information, the attention information determination unit 500 determines that the text information contains attention information. When the text information contains attention information, the situation is sent to the speech emphasis unit 720 of the speech synthesis unit 700 and the text emphasis unit 810 of the text synthesis unit 800.

[0056] When the text information includes attention information, the output unit 800 outputs notification information for notifying the occupant of the attention information. Therefore, the speech synthesis unit 700 of the output unit 800 includes a speech emphasis unit 720, and the text synthesis unit 900 includes a text emphasis unit 910. The output unit 800 may emit sound or light for calling attention.

[0057] If text information obtained by the inference of the speech inference unit 400 is input to the speech synthesis unit 700, the text / speech conversion unit 710 converts the text information into speech. Specifically, the text / speech conversion unit 710 performs a process of synthesizing speech from a character string. As a method of synthesizing speech from a character string, for example, the method described in Japanese Patent Gazette No. 5512597 can be appropriately used. Speech synthesis is achieved by maintaining a database (DB) that associates a set of character strings, words, and phoneme symbols, and a database (DB) that associates each phoneme symbol with a waveform, converting text information into a sequence of phoneme symbols, and converting it into a waveform.

[0058] When the text information includes attention information, the speech emphasizing unit 720 changes the volume of the speech obtained by the conversion performed by the text / speech conversion unit 710. More specifically, when the text information includes attention information, the speech emphasizing unit 720 increases the volume of the speech obtained by the conversion performed by the text / speech conversion unit 710. In addition, the speech emphasizing unit 720 may emphasize the speech by a method other than the method of increasing the volume. The speech emphasizing unit 720 may emphasize the speech by a method such as changing the sound quality, repeating the attention information multiple times, and changing the speech speed. For example, the speech speed can be changed by changing the frequency of the voltage that changes corresponding to one sound signal. When increasing the volume, the speech emphasizing unit 720 changes the volume by changing a uniform coefficient given to the voltage of the speakers 200, 202, 204, 206, 208, and 210. After the volume is changed by the speech emphasizing unit 720, the speech is output from the speakers 200, 202, 204, 206, 208, and 210. It should be noted that, in the speech synthesis unit 700, the speech of the conversation is basically outputted with emphasis when the conversation of the occupant is outputted from the speaker, but an alarm sound may be outputted for the purpose of calling attention, or speech in a fixed format may be outputted.

[0059] When the text information obtained by the inference of the utterance inference unit 400 is input to the text synthesis unit 900, the text emphasis unit 910 performs a process of emphasizing the attention information when the text information includes attention information. More specifically, the text emphasis unit 910 emphasizes the character string corresponding to the attention information by making it bold, changing the color, increasing the font size, blinking, etc. After the attention information is emphasized by the text emphasis unit 910, the utterance content is displayed on the display 250, 252, 254, 256, 258, 260.

[0060] The information-provided person determination unit 600 determines the occupant to whom the information is to be provided. When the text information includes the caution information, the information-provided person determination unit 600 determines the occupant 10 who is the driver as the information-provided person. On the other hand, when the text information does not include the caution information, the information-provided person determination unit 600 determines all the occupants 10, 12, 14, 16, 18, and 20 as the information-provided persons. In this way, the speaker of the output target is determined so that the information related to the voice is provided only to the information-provided person determined by the information-provided person determination unit 600.

[0061] Note that even when the text information includes caution information, the driver 10 and the passenger 12 in the front passenger seat may be identified as persons to be presented with the information, and the information may be presented to a plurality of passengers.

[0062] It should be noted that whether attention information is included may be determined based on the voice data before text conversion, and the above-mentioned voice emphasis processing and text emphasis processing may be performed based on the determination result.

[0063] The determination of whether the attention information 9 is included in the speech data can be performed by comparing the frequency analysis result of the time series of the speech information with the frequency characteristics of the keywords of the attention information.

[0064] Next, based on Figure 3 The process performed by the vehicle attention reminding device 2000 of this embodiment is described in the flowchart. Figure 3 In the description, the processing of the output unit 800 is described by taking the processing performed by the speech synthesis unit 700 as an example, but the processing performed by the text synthesis unit 900 is also performed in the same manner. First, in step S10, it is determined whether speech is input to the speech input unit 300, and when speech is input to the speech input unit 300, the process proceeds to step S12. In step S12, the speech inference unit 400 infers the speech and textualizes the speech. In the subsequent step S14, the attention information determination unit 500 determines whether attention information is included in the textualized speech, and when attention information is included in the speech, the process proceeds to step S16.

[0065] In step S16, the information presentation recipient identification unit 600 identifies only the driver as the information presentation recipient (passenger 10). In subsequent step S18, the text / speech conversion unit 710 of the speech synthesis unit 700 converts the texted speech into speech.

[0066] In the subsequent step S20, the speech emphasizing unit 720 increases the volume of the speech obtained by the conversion performed by the text / speech converting unit 10. In the subsequent step S22, the speech is output from the speaker 200 near the passenger 10 who is the driver. The process ends after step S22.

[0067] On the other hand, in step S14, if the texted speech does not include the caution information, the process proceeds to step S24. In step S24, the information presented person determination unit 600 determines all the passengers 10, 12, 14, 16, 18, 20 as the information presented persons. In the subsequent step S26, the text / speech conversion unit 710 of the speech synthesis unit 700 converts the texted speech into speech.

[0068] After step S26, the process proceeds to step S22. At this time, since the speech does not include attention information, the volume is not changed by the speech emphasizing unit 720, and the speech is output from all the speakers 200, 202, 204, 206, 208, and 210. After step S22, the process ends.

[0069] It should be noted that Figure 3 In the process of step S14, if the textual speech does not include the attention information, the voice is output from the speaker to all passengers, but if the textual speech does not include the attention information, the sound may not be output from the speaker. In this case, the speaker is not amplified, but the normal conversation is carried out.

[0070] In this embodiment, when the conversation of the occupant contains attention information related to driving, the case of reminding the driver to pay attention is described, but the attention information can also be used for vehicle control. For example, when the conversation contains the attention information "A vehicle is approaching from the right!", the vehicle can be controlled to turn left.

[0071] As described above, according to the present embodiment, when the conversation content of the occupant includes attention information related to driving, the driver can be reminded to pay attention. Therefore, the vehicle can be driven optimally based on the attention information.

[0072] The preferred embodiments of the present invention are described in detail with reference to the accompanying drawings, but the present invention is not limited to the above examples. Anyone with ordinary knowledge in the technical field to which the present invention belongs will clearly understand that various variations or modifications can be thought of within the scope of the technical concept recorded in the claims, and these naturally also belong to the technical scope of the present invention.

Claims

1. A vehicle attention reminder device, characterized in that: have: A voice input unit that inputs the conversation contents of all the occupants of the vehicle; a speech inference unit configured to infer speech based on the conversation content; a caution information determination unit that determines whether the utterance includes caution information related to driving of the vehicle; a notification unit configured to notify the driver only when the utterance includes the caution information; as well as a person to be notified of information specifying unit for specifying a person to be notified of information as a target to which the caution information is to be output, When the utterance includes the caution information, the information-presented-person identification unit identifies only the driver of the vehicle as the information-presented-person among all the occupants of the vehicle, When the caution information is not included in the utterance, the information presentation recipient identification unit identifies all occupants of the vehicle as the information presentation recipients.

2. The vehicle warning device according to claim 1, characterized in that: The notification unit outputs the utterance when the attention information is included in the utterance.

3. The vehicle warning device according to claim 1, characterized in that: When the attention information is included in the utterance, the notification unit emphasizes and outputs the utterance.

4. The vehicle warning device according to any one of claims 1 to 3, characterized in that: The utterance inference unit includes a text conversion unit that converts the utterance inferred by the utterance inference unit into text information. The notification unit includes a voice conversion unit that converts the text information into voice, and outputs the voice to a speaker.

5. The vehicle warning device according to claim 4, characterized in that: The notification unit includes a voice emphasizing unit configured to emphasize the volume of the converted voice when the attention information is included in the utterance.

6. The vehicle warning device according to any one of claims 1 to 3, characterized in that: The utterance inference unit includes a text conversion unit that converts the utterance inferred by the utterance inference unit into text information. The notification unit outputs the text information to a display device.

7. The vehicle warning device according to claim 4, characterized in that: The notification unit includes a text emphasis unit configured to emphasize the text information when the attention information is included in the utterance.

8. A vehicle attention reminder method, characterized in that: include: The step of inputting the conversation contents of all the occupants of the vehicle; The step of inferring an utterance based on the conversation content; a step of determining whether the utterance contains attention information related to driving of the vehicle; a step of notifying the driver only when the utterance includes the caution information; as well as a step of determining a recipient of the information as a target of outputting the attention information, When the attention information is included in the speech, the information prompter determination unit determines only the driver of the vehicle as the information prompter among all the occupants of the vehicle; and when the attention information is not included in the speech, the information prompter determination unit determines all the occupants of the vehicle as the information prompter.

9. A recording medium, characterized in that Used to enable the computer to function as: A device for inputting the conversation contents of all occupants of the vehicle; means for inferring speech based on the content of the conversation; means for determining whether the utterance contains attention information related to driving of the vehicle; means for notifying the driver only when the utterance contains the caution information; as well as A device for determining a person to be prompted as an output target of the attention information, and when the attention information is included in the speech, only the driver of the vehicle is determined as the person to be prompted among all occupants of the vehicle, and when the attention information is not included in the speech, all occupants of the vehicle are determined as the person to be prompted.

Images