JP2019105756A - On-vehicle voice output device - Google Patents

Abstract

To provide an on-vehicle voice output device capable of improving convenience for a user.SOLUTION: A voice output device 12 is an on-vehicle voice output device for outputting a response to utterance by a user in a vehicle with voice from a voice output unit on the basis of information input to the voice output unit, and comprises an acceptance unit 40 which accepts an execution instruction of first processing uttered by the user, a control unit 42 which controls execution of the first processing, a derivation unit 44 which derives second processing executed in association with the first processing on the basis of the user's situation and other user's action tendency and an output unit 46 which outputs information indicating the second processing to the voice output unit.SELECTED DRAWING: Figure 5

Description

  The present invention relates to an on-vehicle audio output device.

  Patent Document 1 discloses a voice control switch for controlling a home appliance according to a voice command uttered by a user.

US Patent Application Publication No. 2017/188437

  By the way, for example, when a user makes an utterance for causing an audio output device such as an AI (Artificial Intelligence) speaker to execute processing such as control of an electric device, the audio output device is predetermined corresponding to the utterance. Execute the process.

  Therefore, in order for the user to execute a plurality of processes continuously on the voice output device, it is necessary for the user to individually perform an utterance for executing each process, which is convenient for the user. There is room for improvement in terms of improvement.

  The present invention has been made in consideration of the above facts, and aims to improve the convenience for the user.

  In order to achieve the above object, the invention according to claim 1 outputs a response to an utterance by a user who is in a vehicle by voice from the voice output unit based on information input to the voice output unit. A vehicle-mounted audio output device, a reception unit receiving an execution instruction of the first process uttered by the user, a control unit performing control to execute the first process, a situation of the user, and other users The output unit includes a deriving unit that derives a second process performed in association with the first process based on the behavior tendency, and an output unit that outputs information indicating the second process to the voice output unit.

  According to the first aspect of the present invention, the second process performed in relation to the first process is derived based on the user's situation and the action tendency of the other user, and the second process derived is shown. Information is output by voice from the voice output unit. Therefore, the convenience for the user can be improved.

  According to the present invention, it is possible to improve the convenience for the user.

It is a figure showing an example of the composition of the speech output system concerning an embodiment. It is a schematic diagram which shows an example of the action tendency model which concerns on embodiment. It is a block diagram showing an example of the hardware constitutions of the voice output device concerning an embodiment. It is a figure showing an example of personal feedback data concerning an embodiment. It is a block diagram showing an example of functional composition of an audio output device concerning an embodiment. It is a flowchart which shows an example of the audio | voice output process which concerns on embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  First, the configuration of an audio output system 10 according to the present embodiment will be described with reference to FIG. As shown in FIG. 1, the voice output system 10 according to the present embodiment receives an utterance by a user 11 who is in a vehicle, executes an application corresponding to the accepted utterance, and responds an execution result by voice. An output device 12 is included. In the present embodiment, the voice output device 12 is brought into the vehicle by the user 11. In addition, the voice output system 10 includes a control device 14 installed in a house which is a home of the user 11 and controlling an electric device to be controlled in the installed house. In addition, as an example of the electric equipment of the control object of the control apparatus 14, the apparatus of the control object in what is called a smart home, such as an air conditioner (henceforth "air-conditioner") and lighting equipment, is mentioned.

  In addition, in the following, when meaning the user who uses the audio | voice output device 12 shown in FIG. 1, a code | symbol is attached with "user 11", and when meaning several users of the audio | voice output system 10 containing the user 11, It is called a "user" without a code.

  The voice output device 12 and the control device 14 are each connected to the network N by wireless communication. Also, the cloud server 16 is connected to the network N. Therefore, the voice output device 12, the control device 14, and the cloud server 16 can communicate with each other via the network N. Further, the behavior tendency model 17 is stored in a predetermined storage area of the cloud server 16.

  The behavior tendency model 17 according to the present embodiment is a model that represents behavior tendencies of a large number of users. Specifically, the behavior tendency model 17 includes the user's situation (for example, "return from work" and "return from exercise", etc.) and an instruction to execute the first process from the user (for example, "turn on the air conditioner"). And a speech such as "bath water" and the like, and is a learned model whose output is a second process performed in relation to the first process. The behavior tendency model 17 performs machine learning using, for example, a plurality of combinations of the user's situation among a large number of users, the first process, and the second process performed in relation to the first process as teacher data. It is a learned model obtained by In addition, as a method of machine learning, a neural network is mentioned, for example.

  An example of the schematic diagram of the action tendency model 17 is shown in FIG. As shown in FIG. 2, in the behavior tendency model 17 according to the present embodiment, the degree of association between the first process and the second process is derived for each situation of the user. The behavior tendency model 17 is a second process corresponding to the input user's situation and the first process, and outputs a second process having a degree of association equal to or higher than a predetermined threshold. In addition, as the degree of association, for example, the probability that the second process is continuously performed to the first process under the situation of the corresponding user can be mentioned.

  Next, the hardware configuration of the audio output device 12 will be described with reference to FIG. As shown in FIG. 3, the audio output device 12 includes a central processing unit (CPU) 21, a memory 22 as a temporary storage area, and a non-volatile storage unit 23. The audio output device 12 also includes information such as a communication I / F (InterFace) 24 used when communicating with an external device, a display unit 25 such as a liquid crystal display, an audio input unit 26 such as a microphone, and a speaker. It includes an audio output unit 27 that outputs the sound into the vehicle compartment. The CPU 21, the memory 22, the storage unit 23, the communication I / F 24, the display unit 25, the voice input unit 26, and the voice output unit 27 are connected to the bus 28. An example of the audio output device 12 is an AI speaker.

  The storage unit 23 stores personal feedback (FB) data 30 and an audio output program 32. An example of the personal FB data 30 is shown in FIG. As shown in FIG. 4, in the personal FB data 30 according to the present embodiment, the voice output device 12 performs the second process on the user 11 corresponding to the combination of the situation of the user 11 and the execution instruction of the first process. The probability (in FIG. 4, expressed as “affirmation rate”) is stored that the response from the user 11 in the case of presenting by voice is a positive response (that is, a response indicating that the second process is to be performed).

  Next, the functional configuration of the audio output device 12 according to the present embodiment will be described with reference to FIG. As shown in FIG. 5, the audio output device 12 includes a reception unit 40, a control unit 42, a derivation unit 44, and an output unit 46. The CPU 21 of the voice output device 12 executes the voice output program 32 stored in the storage unit 23 to function as the reception unit 40, the control unit 42, the derivation unit 44, and the output unit 46 shown in FIG.

  The receiving unit 40 receives the execution instruction of the first process uttered by the user 11 via the voice input unit 26. Also, the receiving unit 40 receives, via the voice input unit 26, a response uttered by the user 11 with respect to the information indicating the second process output by voice from the voice output unit 27. Further, the receiving unit 40 updates the corresponding affirmation rate of the personal FB data 30 according to the received response.

  The control unit 42 performs control to execute the first process received by the receiving unit 40. In the present embodiment, the control unit 42 performs control to execute the first process by transmitting instruction information indicating an instruction to execute the first process to the control device 14 via the communication I / F 24. In addition, when the response uttered by the user 11 with respect to the information indicating the second process accepted by the accepting unit 40 is a positive response, the control unit 42 performs control to execute the second process. In the present embodiment, the control unit 42 performs control to execute the second process by transmitting instruction information indicating an instruction to execute the second process to the control device 14 via the communication I / F 24.

  Further, the control unit 42 transmits (feedbacks) information indicating whether or not the second process has been performed in response to the first process to the cloud server 16 via the communication I / F 24. In the cloud server 16, if the second process is not executed corresponding to the first process, the corresponding degree of association of the behavioral tendency model 17 is lowered, and if it is executed, the corresponding degree of association of the behavioral tendency model 17 Raise

  The deriving unit 44 derives a second process performed in association with the first process received by the receiving unit 40 based on the behavior tendency of another user. In the present embodiment, the derivation unit 44 first identifies the situation of the user 11. For example, the derivation unit 44 specifies the current situation of the user 11 (for example, “work return” or “return from exercise”) by acquiring the schedule of the user 11 from the scheduler software used by the user 11 Do. Note that the derivation unit 44 causes the voice output unit 27 to output a voice for asking the current situation of the user 11, and acquires a response by the user 11 to the outputted voice via the voice input unit 26. The situation of the user 11 may be specified.

  Next, the derivation unit 44 transmits the information indicating the identified status of the user 11 and the first process received by the reception unit 40 to the cloud server 16 via the communication I / F 24. The cloud server 16 receives the information indicating the state of the user 11 and the first process transmitted from the voice output device 12, and inputs the received information indicating the state of the user 11 and the first process to the behavior tendency model 17. In addition, the cloud server 16 transmits, to the voice output device 12, information indicating the second process output from the behavior tendency model 17 in response to the information indicating the input user's 11 situation and the first process. The derivation unit 44 receives the information indicating the second process transmitted from the cloud server 16 through the communication I / F 24.

  Next, when the number of second processes indicated by the received information is one, the derivation unit 44 outputs information indicating the received second processes to the output unit 46. Further, when the number of second processes indicated by the received information is more than one, the derivation unit 44 refers to the personal FB data 30, and identifies the status of the identified user 11, the first process accepted by the accepting unit 40, and Information indicating the second process having the highest positive rate corresponding to the received combination of the second processes is output to the output unit 46.

  Note that, for example, when the affirmation rate is less than the predetermined value, the derivation unit 44 may output, to the output unit 46, information indicating that the second process related to the first process does not exist. Further, for example, when a negative response continues by the user 11 a predetermined number of times for the information indicating the same second process, the derivation unit 44 indicates that the second process related to the first process does not exist. The information may be output to the output unit 46.

  The output unit 46 outputs the information indicating the second process input from the derivation unit 44 to the audio output unit 27. Thereby, the audio output unit 27 outputs the information indicating the second process input from the output unit 46 by audio.

  Next, the operation of the audio output device 12 according to the present embodiment will be described with reference to FIG. FIG. 6 is a flowchart showing an example of the flow of audio output processing executed by the CPU 21 of the audio output device 12 according to the present embodiment. Further, in the present embodiment, the voice output process shown in FIG. 6 is executed by the CPU 21 executing the voice output program 32 stored in advance in the storage unit 23. The voice output process shown in FIG. 6 is executed, for example, when an instruction to execute the first process uttered by the user 11 is input to the voice output device 12.

  In step S10 of FIG. 6, the receiving unit 40 receives, via the voice input unit 26, an instruction to execute the first process uttered by the user 11. In step S12, as described above, the control unit 42 performs control to execute the first process received in the process of step S10.

  In step S14, the derivation unit 44 specifies the current situation of the user 11 as described above. Further, the derivation unit 44 transmits, to the cloud server 16 via the communication I / F 24, the information indicating the identified status of the user 11 and the first process accepted by the process of step S10. Further, the derivation unit 44 receives, via the communication I / F 24, the information indicating the second process returned from the cloud server 16 in response to the transmitted information, so that the second process is accepted in the process of step S10. 1. Deriving a second process performed in connection with the process. Then, as described above, the derivation unit 44 outputs the information indicating the derived second process to the output unit 46 when the number of derived second processes is one, and in the case of multiple, the personal FB data Referring to 30, the information indicating the second process with the highest positive rate is output to the output unit 46.

  In step S16, the output unit 46 outputs, to the voice output unit 27, information indicating the second process input from the derivation unit 44 in the process of step S14. By the process of step S16, the audio output unit 27 outputs the information indicating the second process input from the output unit 46 by voice. For example, in the example shown in FIG. 2 and FIG. 4, when the situation of the user 11 is “the return from exercise” and the utterance “put hot water into the bath” is input from the user 11 to the voice output device 12 The voice output device 12 outputs a voice "Do you also want to turn on the air conditioner?"

  The user 11 utters a positive response (for example, “YES”) or a negative response (for example, “NO”) to the information indicating the second process output by voice from the voice output device 12 . In step S18, the receiving unit 40 receives a response uttered by the user 11 with respect to the information indicating the second process output by voice from the voice output unit 27 in the process of step S16. In step S20, the corresponding affirmation rate of the personal FB data 30 is updated according to the response accepted by the process of step S18.

  In step S22, the control unit 42 determines whether the response accepted in the process of step S18 is a positive response. If the determination is affirmative, the process proceeds to step S24. In step S24, as described above, the control unit 42 performs control to execute the second process derived in the process of step S14.

  On the other hand, if the determination in step S22 is a negative determination, the process proceeds to step S26 without the process of step S24 being performed. In step S26, as described above, the control unit 42 feeds back, to the cloud server 16 via the communication I / F 24, information indicating whether or not the second process has been executed corresponding to the first process. When the process of step S26 ends, the voice output process ends.

  As described above, according to the present embodiment, the second process performed in association with the first process is presented to the user 11 by voice based on the action tendency of another user. Therefore, the user 11 can grasp the second process related to the first process by uttering the execution instruction of the first process to the voice output device 12, and responds to the presented second process. Thus, the execution instruction of the second process can be performed. As a result, the convenience for the user 11 can be improved.

  In addition, although the said embodiment demonstrated the case where the audio | voice output device 12 existed in the vehicle, it is not limited to this. The voice output device 12 may be in a form other than in the vehicle.

  Further, in the above embodiment, the second process related to the first process may be derived using environmental conditions such as the weather at the position where the user 11 is located and environmental temperature in addition to the situation of the user 11 . In this case, the behavior tendency model 17 exemplifies a form in which the degree of association between the first process and the second process is derived for each combination of the situation and the environmental condition.

  Moreover, although the said embodiment demonstrated the case where the action tendency of the user 11 was included in the action tendency model 17, it is not limited to this. The second process may be derived based on the behavior tendency of another user without including the behavior tendency of the user 11 in the behavior tendency model 17. In the above embodiment, the personal FB data 30 may not be used.

  Further, in the above embodiment, the control device 14 may control the execution timing of at least one of the first process and the second process. In this case, for example, the voice output device 12 acquires, from an ECU (Electronic Control Unit) that controls the car navigation system, the time when the vehicle is scheduled to arrive at the home of the user 11 (that is, the scheduled home time of the user 11). The form which transmits the acquired going-home scheduled time to the control apparatus 14 is illustrated. Then, in this case, the control device 14 receives the scheduled return time sent from the voice output device 12, and controls the execution timing of at least one of the first processing and the second processing according to the received scheduled return time. Thereby, the use efficiency of the energy used for the operation of various electric appliances can be improved.

  Further, although the processing performed by the CPU 21 in the above embodiment has been described as software processing performed by executing a program, it may be processing performed by hardware. Further, the process performed by the CPU 21 may be a process performed by combining both software and hardware. Further, the voice output program 32 stored in the storage unit 23 may be stored in various storage media and distributed.

  Further, the present invention is not limited to the above-described embodiment, and it goes without saying that various modifications can be made without departing from the scope of the invention other than the above-described embodiment.

10 voice output system 12 voice output device 21 CPU (reception unit, control unit, derivation unit, output unit)
22 Memory 23 Storage Unit 27 Voice Output Unit 30 Personalized FB Data 32 Voice Output Program 40 Reception Unit 42 Control Unit 44 Derivation Unit 46 Output Unit

Claims (1)

An on-vehicle voice output device for outputting a response to an utterance by a user who is in a vehicle by voice from the voice output unit based on information input to the voice output unit,
A receiving unit that receives an instruction to execute the first process uttered by the user;
A control unit that performs control to execute the first process;
A deriving unit that derives a second process to be performed in relation to the first process based on the user's situation and an action tendency of another user;
An output unit that outputs information indicating the second process to the audio output unit;
Car audio output device equipped with