Speech control device and speech system

Abstract

To provide an utterance control device and an utterance system that allow varied communication between electrical devices and users.SOLUTION: In an air purifier, an utterance control device includes an execution unit that causes an electric device to execute a first utterance when a condition is satisfied, and causes the electric device to execute a second utterance that is different from the first utterance based on the presence of a first reply input to the electric device in response to the first utterance or the characteristics of the first reply, when the next time the condition is satisfied.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present disclosure relates to a speech control device and a speech system.

Background Art

[0002] Patent Document 1 discloses a reproduction device for electronic books. In this reproduction device for electronic books, by selecting a speech synthesis database to be used according to specified information input from a user, it is possible to reproduce with voices of different qualities for each character appearing in the book, or to reproduce with a voice of a quality preferred by the user (paragraphs 0020 and 0027).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the reproduction device for electronic books disclosed in Patent Document 1, the user's response to the reproduced voice is not considered. For this reason, it is impossible to perform rich communication between the reproduction device for electronic books and the user that changes over time.

[0005] The present disclosure has been made in view of this problem. An aspect of the present disclosure aims to provide, for example, a speech control device and a speech system that can perform rich communication between an electric device and a user that changes over time.

Means for Solving the Problems

[0006] One embodiment of the speech control device of the present disclosure includes an execution unit that, when a condition is met, causes an electrical device to execute a first utterance, and the next time the condition is met, causes the electrical device to execute a second utterance that is different from the first utterance based on whether or not a first response is input to the electrical device in response to the first utterance, or the characteristics of the first response.

[0007] Another form of the speech system of the present disclosure comprises a speech control device of the present disclosure and an electrical device. [Brief explanation of the drawing]

[0008] [Figure 1] This is a block diagram of the air purifier according to the first embodiment. [Figure 2] This figure illustrates the speech modification performed by the air purifier of the first embodiment. [Figure 3] This flowchart shows the processing flow performed by the air purifier of the first embodiment. [Figure 4] This flowchart shows the processing flow performed by the air purifier of the first embodiment. [Modes for carrying out the invention]

[0009] The embodiments of this disclosure will be described below with reference to the drawings. In the drawings, the same or equivalent elements are denoted by the same reference numerals, and redundant descriptions are omitted.

[0010] 1. First Embodiment 1.1 Air purifier Figure 1 is a block diagram of an air purifier according to the first embodiment.

[0011] The air purifier 1 of the first embodiment shown in Figure 1 cleans the surrounding air. The air purifier 1 may perform operations other than cleaning the surrounding air. For example, the air purifier 1 may perform operations such as cooling, heating, dehumidifying, humidifying, and ion supply.

[0012] Furthermore, the air purifier 1 executes a speech 11 in conjunction with the occurrence of a speech trigger, and controls the speech 11 to be executed the next time a speech trigger occurs, based on the user's response 12 input to the air purifier 1 in response to the executed speech 11.

[0013] As shown in Figure 1, the air purifier 1 includes an air purification unit 31, a human presence sensor 32, an illuminance sensor 33, an operation unit 34, a speaker 35, a microphone 36, and a speech control device 37.

[0014] The air purification unit 31 draws in air from outside the air purifier 1, purifies the drawn-in air, and blows the purified air out to the outside of the air purifier 1. The air purification unit 31 is equipped with a fan, a filter, etc. The fan generates an airflow that goes from outside the air purifier 1, through the filter, to the outside of the air purifier 1. The filter allows the air flowed by the generated airflow to pass through, and removes substances contained in the air that passes through. The substances to be removed include odor components, dust, pollen, PM2.5, etc.

[0015] The motion sensor 32 detects whether or not there is a person around the air purifier 1 and outputs a detection result 41 indicating whether or not there is a person around the air purifier 1.

[0016] The illuminance sensor 33 detects the illuminance around the air purifier 1 and outputs a detection result 42 indicating the illuminance around the air purifier 1.

[0017] The operation unit 34 detects operations performed on the operation unit 34 and outputs a detection result 43 indicating the operations performed on the operation unit 34. The operation unit 34 can be a button, dial, slider, touch panel, touchpad, etc.

[0018] Speaker 35 emits sound corresponding to the input audio signal 44.

[0019] The microphone 36 outputs an audio signal 45 corresponding to the input audio.

[0020] The speech control device 37 inputs the voice signal 44 to the speaker 35 to make the speaker 35 emit voice. Thereby, the speech control device 37 causes the air cleaner 1 to execute the speech 11.

[0021] The speech control device 37 identifies the voice input to the air cleaner 1 based on the voice signal 45 output by the microphone 36. After the speech control device 37 causes the air cleaner 1 to execute the speech 11, the speech control device 37 identifies the reply 12 of the user 21, which is the voice input to the air cleaner 1.

[0022] In addition, the speech control device 37 controls the timing for causing the air cleaner 1 to execute the speech 11 based on the detection result 41 output by the human presence sensor 32, the detection result 42 output by the illuminance sensor 33, and the detection result 43 output by the operation unit 34.

[0023] In addition, the speech control device 37 controls the speech 11 to be executed by the air cleaner 1 next based on the reply 12 of the user 21 input to the air cleaner 1 after the speech control device 37 causes the air cleaner 1 to execute the speech 11.

[0024] The speech control device 37 and the air cleaner 1 constitute a speech system for controlling the speech 11. The speech control device 37 is built into the air cleaner 1. The speech control device 37 may be built into a device that communicates with the air cleaner 1. For example, the speech control device 37 may be built into a cloud server that communicates with the air cleaner 1. The speech control device 37 may control the speech 11 to be executed by electrical equipment other than the air cleaner 1. For example, the speech control device 37 may control the speech 11 to be executed by an air conditioner, a fan, a humidifier, a dehumidifier, an ion generator, a refrigerator, a vacuum cleaner, a microwave oven, etc. <00001​​​​​1.2 Speech control device As shown in Figure 1, the speech control device 37 includes a voice analysis unit 51 and an execution unit 52.

[0027] The voice analysis unit 51 analyzes the voice signal 45 output by the microphone 36 to identify the voice quality 61 of the voice input to the air purifier 1 and the dialect 62 used in that voice.

[0028] The execution unit 52 causes the air purifier 1 to execute the utterance 11 when the condition is met. In this way, the execution unit 52 causes the air purifier 1 to execute the utterance 11 in conjunction with the occurrence of an utterance trigger, which is the condition being met. The utterance 11 to be executed by the air purifier 1 may be changed depending on the time period to which the time the condition is met belongs.

[0029] The utterance 11 that the air purifier 1 is instructed to perform is an utterance for the air purifier 1 to speak to the user 21, and is not an utterance for the user 21 to respond to the air purifier 1.

[0030] The conditions that must be met for the air purifier 1 to execute the speech 11 include the condition that the detection result 41 output by the human presence sensor 32 indicates that there is a person around the air purifier 1.

[0031] The conditions to be satisfied may include conditions other than the condition that the detection result 41 output by the motion sensor 32 indicates that there is a person around the air purifier 1. For example, the conditions to be satisfied may include the condition that the detection result 42 output by the illuminance sensor 33 indicates that the illuminance around the air purifier 1 is equal to or greater than the standard illuminance, the condition that the current time belongs to a specific time period, the condition that the detection result 43 output by the operation unit 34 indicates that an operation has been performed on the operation unit 34, and the condition that the voice quality 61 identified by the voice analysis unit 51 is a specific voice quality.

[0032] The execution unit 52 may change the voice quality of the utterance 11 according to the voice quality 61 identified by the voice analysis unit 51. For example, if the voice quality 61 identified by the voice analysis unit 51 is male, the execution unit 52 may change the voice quality of the utterance 11 to a male voice quality, and if the voice quality 61 identified by the voice analysis unit 51 is female, the execution unit 52 may change the voice quality of the utterance 11 to a female voice quality.

[0033] When the speech control device 37 controls speech 11 performed by an electrical device equipped with a door, such as a refrigerator, the conditions to be satisfied may include the condition that it has been detected that the door has been opened.

[0034] The conditions that must be met may be a combination of multiple conditions. However, the condition that a voice prompt from user 21 has been detected is not included in the combination of multiple conditions.

[0035] For example, the conditions to be satisfied may be a first condition in which the detection result 41 output by the motion sensor 32 indicates that there is a person around the air purifier 1, and / or a second condition in which it is detected that an operation has been performed on the air purifier 1.

[0036] Furthermore, the conditions to be satisfied may also be that the condition that the current time belongs to a specific time zone is met, and that the first condition and / or the second condition are satisfied.

[0037] Furthermore, the condition to be satisfied may be that the detection result 42 output by the illuminance sensor 33 indicates that the illuminance around the air purifier 1 is equal to or greater than the standard illuminance, and that the first condition and / or the second condition are satisfied.

[0038] Furthermore, the conditions that must be met may also include the condition that the current time belongs to a specific time period and the condition that the detection result 42 output by the illuminance sensor 33 indicates that the illuminance around the air purifier 1 is equal to or greater than the standard illuminance.

[0039] The second condition may be that the detection result 43 output by the operation unit 34 indicates that an operation has been performed on the operation unit 34, or it may be that it has been detected that an operation for normal use has been performed on the air purifier 1. When the speech control device 37 controls speech 11 performed by an electrical device equipped with a door, such as a refrigerator, a normal operation is, for example, opening the door.

[0040] 1.3 Changes to Utterances Figure 2 illustrates the speech modification performed by the air purifier of the first embodiment.

[0041] As shown in Figure 2, the execution unit 52 causes the air purifier 1 to execute the first utterance 71 when the condition is met, and the next time the condition is met, it causes the air purifier 1 to execute the second utterance 72.

[0042] Furthermore, the execution unit 52 makes the second utterance 72 different from the first utterance 71 based on the presence or absence of a first response 73 input to the air purifier 1 in response to the first utterance 71, or the characteristics 101 of the first response 73. The characteristics 101 of the first response 73 include the content 131 and manner of speaking 132 of the first response 73. The manner of speaking 132 of the first response 73 includes the dialect 171 used in the first response 73, the frankness 172 of the first response 73, and the volume 173 of the first response 73. These can be identified by the voice analysis unit 51 using its built-in database to analyze the waveform of the voice signal 45 output by the microphone 36.

[0043] Differentiating the second utterance 72 from the first utterance 71 includes differentiating the content 121 of the second utterance 72 from the content 111 of the first utterance 71, differentiating the manner in which the second utterance 72 is spoken from the manner in which the first utterance 71 is spoken from the manner in which the first utterance 71 is spoken from the manner in which the second utterance 72 first utterance 71.

[0044] The content 111 of the first utterance 71 and the content 121 of the second utterance 72 include matters related to the weather, current events, events such as birthdays, jokes, and introductions to unused modes of the air purifier 1.

[0045] Making the manner of speaking 122 of the second utterance 72 different from the manner of speaking 112 of the first utterance 71 includes making the dialect 161 used in the second utterance 72 different from the dialect 151 used in the first utterance 71, making the frankness 162 of the second utterance 72 different from the frankness 152 of the first utterance 71, making the volume 163 of the second utterance 72 different from the volume 153 of the first utterance 71, and so on.

[0046] For example, if there is a first response 73, the execution unit 52 makes the content 121 of the second utterance 72 different from the content 111 of the first utterance 71.

[0047] Furthermore, the execution unit 52, for example, if a first response 73 is received, makes the manner of speaking 122 of the second utterance 72 different from the manner of speaking 112 of the first utterance 71. For example, if a first response 73 is received, the execution unit 52 makes the volume 163 of the second utterance 72 louder than the volume 153 of the first utterance 71.

[0048] The execution unit 52 may adjust the volume 163 of the second utterance 72 to differ from the volume 153 of the first utterance 71 depending on the quantity of the first response 73. For example, if the first response 73 is infrequent, the execution unit 52 may lower the volume 163 of the second utterance 72 compared to the volume 153 of the first utterance 71. This allows the first utterance 71 to be kept to a minimum when it is highly likely that the user 21 does not want the first utterance 71 to be executed.

[0049] Furthermore, the execution unit 52, for example, makes the content 121 of the second utterance 72 different from the content 111 of the first utterance 71 based on the content 131 of the first response 73. For example, if the air purifier 1 is given a positive first response 73 after the air purifier 1 has delivered the first utterance 71, "Good morning," the execution unit 52 will deliver a second utterance 72 such as "Good morning. It's a nice day today," or "Good morning! Did you sleep well last night?" Preferably, the execution unit 52 includes a question in the content 111 of the first utterance 71 to find out what the user 21 wants to hear.

[0050] Furthermore, the execution unit 52, for example, makes the way of speaking 122 of the second utterance 72 different from the way of speaking 112 of the first utterance 71, based on the way of speaking 132 of the first response 73.

[0051] For example, the execution unit 52 makes the dialect 161 used in the second utterance 72 different from the dialect 151 used in the first utterance 71, based on the dialect 171 used in the first reply 73. For example, if the air purifier 1 has executed the first utterance 71 using standard Japanese, and then the first reply 73 using Kansai dialect is input to the air purifier 1, the execution unit 52 will make the air purifier 1 execute the second utterance 72 using Kansai dialect.

[0052] Furthermore, the execution unit 52, for example, makes the frankness 162 of the second utterance 72 different from the frankness 152 of the first utterance 71 based on the frankness 172 of the first response 73. For example, if the execution unit 52 has the air purifier 1 execute the first utterance 71 using polite language, and then the air purifier 1 receives the first response 73 which does not use polite language, the execution unit 52 will have the air purifier 1 execute the second utterance 72 which does not use polite language.

[0053] Furthermore, the execution unit 52, for example, makes the volume 163 of the second utterance 72 different from the volume 153 of the first utterance 71 based on the volume 173 of the first response 73.

[0054] By making the second utterance 72 different from the first utterance 71, the speech control device 37 can enable varied communication between the air purifier 1 and the user 21. Furthermore, by making the second utterance 72 different from the first utterance 71 based on the presence or absence of the first response 73 or the characteristics 101 of the first response 73, the air purifier 1 can enable communication between the user 21 that is suited to the user 21's personality, language use, preferences, etc.

[0055] If the content 131 of the first response 73 is a request not to perform the first utterance 71, the execution unit 52 may stop the air purifier 1 from performing the first utterance 71 until a voice requesting the air purifier 1 to perform the first utterance 71 is input to the air purifier 1. The first response 73 having content requesting not to perform the first utterance 71 is a response such as "Don't talk to me" or "Be quiet".

[0056] If the conditions to be met are that the current time falls between 6:00 AM and 8:00 AM and that the operation of opening the refrigerator door was detected for the first time that day, then the first utterance 71 is, for example, "Let's do our best today, what shall we make for breakfast?" In this case, the second utterance 72 changes depending on whether or not there is a first response 73 to the first utterance 71, or the characteristics 101 of the first response 73.

[0057] If the conditions to be met are that the current time is between 11 p.m. and 12 p.m., and the detection result output by the illuminance sensor is that the illuminance around the refrigerator is equal to or greater than the standard illuminance, then the first utterance 71, for example, utterance 11, is "You've worked hard today. Get some good rest, good night." In this case, the second utterance 72 changes depending on whether or not there is a first response 73 to the first utterance 71, or the characteristics 101 of the first response 73.

[0058] The presence or absence of a first response 73 to a first utterance 71, or the characteristics 101 of the first response 73, which are executed when the first condition is met, does not affect the utterance 11 executed when the second condition is met. Conversely, the presence or absence of a first response 73 to a first utterance 71, or the characteristics 101 of the first response 73, which are executed when the second condition is met, does not affect the utterance 11 executed when the first condition is met.

[0059] 1.4 Processing performed by an air purifier Figures 3 and 4 are flowcharts showing the processing flow performed by the air purifier of the first embodiment.

[0060] Air purifier 1 performs steps S101 to S114 shown in Figures 3 and 4.

[0061] In step S101, the execution unit 52 determines whether the conditions that must be met for the air purifier 1 to execute the utterance 11 have been met. If it is determined that the conditions have been met, step S102 is executed. If it is determined that the conditions have not been met, step S101 is executed.

[0062] In step S102, the execution unit 52 causes the air purifier 1 to execute the first utterance 71.

[0063] In steps S101 and S102, the execution unit 52 waits without having the air purifier 1 execute the first utterance 71 until the conditions are met, and in conjunction with the conditions being met, it has the air purifier 1 execute the first utterance 71. The execution unit 52 may also have the air purifier 1 execute the first utterance 71 only if the conditions are met during a predetermined time period, for example, first thing in the morning (from 6 a.m. to 7 a.m.), in conjunction with the conditions being met.

[0064] In step S103, following step S102, the execution unit 52 determines whether or not a first response 73 was received. If it is determined that a first response 73 was received, step S104 is executed. If it is determined that there was no first response 73, step S108 is executed.

[0065] In step S104, the execution unit 52 determines whether the content 131 of the first response 73 requests that the utterance 11 not be performed. A first response 73 that requests that the utterance 11 not be performed is a response such as "Don't talk to me" or "Be quiet." If it is determined that the content 131 of the first response 73 requests that the utterance 11 not be performed, step S108 is executed. If it is determined that the content 131 of the first response 73 does not request that the utterance 11 not be performed, step S105 is executed.

[0066] In step S105, the execution unit 52 performs a process to make the second utterance 72 different from the first utterance 71.

[0067] In step S106, following step S105, it is determined whether the conditions that must be met for the air purifier 1 to execute the utterance 11 have been met. If it is determined that the conditions have been met, step S107 is executed. If it is determined that the conditions have not been met, step S106 is executed.

[0068] In step S107, the execution unit 52 causes the air purifier 1 to execute the second utterance 72.

[0069] In steps S106 and S107, the execution unit 52 waits without causing the air purifier 1 to execute the second utterance 72 until the conditions are met, and in conjunction with the conditions being met, causes the air purifier 1 to execute the second utterance 72.

[0070] Furthermore, in steps S103 to S107, the execution unit 52 causes the air purifier 1 to execute a second utterance 72 that is different from the first utterance 71 if there is a first response 73 and the content of the first response 73 does not request that the utterance 11 not be executed. On the other hand, the execution unit 52 does not cause the air purifier 1 to execute the second utterance 72 if there is no first response 73, or if there is a first response 73 but the content 131 of the first response 73 requests that the utterance 11 not be executed.

[0071] In step S108, the execution unit 52 counts the number of consecutive times when there is no first response 73, or when the content of the first response 73 requests not to perform the utterance 11, and determines whether the counted number is equal to or greater than the first number. If it is determined that the counted number is equal to or greater than the first number, step S109 is executed. If it is determined that the counted number is not equal to or greater than the first number, step S101 is executed.

[0072] In step S109, the execution unit 52 stops the air purifier 1 from executing the first utterance 71 and terminates the process.

[0073] In steps S108 and S109, the execution unit 52 stops having the air purifier 1 execute the first utterance 71 if the absence of the first response 73 or the content of the first response 73 requesting that the utterance 11 not be executed is repeated a first number of times or more. This prevents the continuous execution of the first utterance 71 when there is a high probability that the user 21 does not want the first utterance 71 to be executed. On the other hand, the execution unit 52 has the air purifier 1 execute the first utterance 71 in conjunction with the fulfillment of the conditions until the absence of the first response 73 or the content of the first response 73 requesting that the utterance 11 not be executed is repeated a first number of times or more.

[0074] In step S110, following step S107, the execution unit 52 determines whether or not a second response was input to the air purifier 1 in response to the second utterance 72. If it is determined that a second response was received, step S111 is executed. If it is determined that no second response was received, step S113 is executed.

[0075] In step S111, the execution unit 52 determines whether the content of the second response requests that the utterance 11 not be performed. A second response that requests that the utterance 11 not be performed is a response such as "Don't talk to me" or "Be quiet." If it is determined that the content of the second response requests that the utterance 11 not be performed, step S113 is executed. If it is determined that the content of the second response does not request that the utterance 11 not be performed, step S112 is executed.

[0076] In step S112, the execution unit 52 performs a process to make the third utterance that the air purifier 1 will execute when the condition is met next time different from the second utterance 72.

[0077] In step S113, the execution unit 52 counts the number of consecutive times when there is no second response, or when the content of the second response requests not to perform the utterance 11, and determines whether the counted number is equal to or greater than the second number. If it is determined that the counted number is equal to or greater than the second number, step S114 is executed. If it is determined that the counted number is not equal to or greater than the second number, step S106 is executed.

[0078] In step S114, the execution unit 52 stops the air purifier 1 from executing the second utterance 72. Following step S114, step S101 is executed again.

[0079] In steps S113 and S114, if the absence of a second response or the content of the second response requesting that utterance 11 not be executed is repeated two or more times in a row, the execution unit 52 stops having the air purifier 1 execute the second utterance 72 and causes the air purifier 1 to execute the first utterance 71. This makes it possible to suppress the continuous execution of the second utterance 72 when there is a high probability that the user 21 does not want the second utterance 72 to be executed. On the other hand, until the absence of a second response or the content of the second response requesting that utterance 11 not be executed is repeated two or more times in a row, the execution unit 52 causes the air purifier 1 to execute the second utterance 72 in conjunction with the fulfillment of the conditions.

[0080] The second number of times is greater than the first number of times. The reason the second number of times is greater than the first is that the second utterance 72 is executed when it is unlikely that user 21 does not want the second utterance 72 to be executed, and therefore, even if it is executed continuously, it is unlikely to result in a state that does not meet user 21's wishes.

[0081] 1.5 Processing for each voice type The voice analysis unit 51 may record the identified voice quality 61, and the execution unit 52 may perform the processing shown in Figures 3 and 4 for each voice quality 61. This allows the execution unit 52 to control the second utterance 72 to differ from the first utterance 71 for each voice quality 61. This makes it possible to differentiate the change from the first utterance 71 to the second utterance 72 for the first user from the change from the first utterance 71 to the second utterance 72 for the second user. This enables communication between the air purifier 1 and the user 21 that is tailored to the user 21's personality.

[0082] 1.6 Variations Two or more electrical devices may perform the processes shown in Figures 3 and 4. In this case, the two or more electrical devices share the conditions that must be met and the number of times the utterance 11 is counted.

[0083] If the execution unit 52 receives an input to the air purifier 1 of a voice 61 that does not match the recorded voice 61, it may cause the air purifier 1 to execute an utterance that asks for the name of user 21. The utterance that asks for the name of user 21 is such as, "Nice to meet you! What's your name?" Furthermore, if user 21 gives their name in response to the utterance that asks for the name of user 21, the execution unit 52 may cause the air purifier 1 to execute an utterance that includes the given name and provides a greeting. The utterance that includes the given name and provides a greeting is such as, "Hello, Mr. / Ms. XX."

[0084] Furthermore, if the execution unit 52 receives an input to the air purifier 1 of a voice with a voice quality 61 that matches the recorded voice quality 61, but more than the set time has passed since the last time the voice was input to the air purifier 1, the execution unit 52 may cause the air purifier 1 to execute an utterance to reconfirm the name of user 21. The utterance to reconfirm the name of user 21 is an utterance such as, "Are you perhaps Mr. / Ms. XX?". In addition, if user 21 responds to the utterance to reconfirm the name of user 21, the execution unit 52 may cause the air purifier 1 to execute an utterance to welcome the reunion. The utterance to welcome the reunion is an utterance such as, "It's been a long time!". On the other hand, if user 21 does not respond to the utterance to reconfirm the name of user 21, the execution unit 52 may cause the air purifier 1 to execute an utterance to confirm whether user 21's name is correct. The utterance to confirm whether user 21's name is correct is an utterance such as, "Oh, did you mistake me for someone else?".

[0085] The speech control device 37 may have a function for registering a user dictionary. If the speech control device 37 has this function, the execution unit 52 causes the air purifier 1 to execute an utterance that asks about the meaning of a word spoken by the user 21, recognizes the meaning of the word from the answer given by the user 21 in response to the utterance, and associates the recognized meaning with the word. For example, if the user 21 utters the word "ryo", the execution unit 52 causes the air purifier 1 to execute the utterance "What does it mean?" which asks about the meaning of the word "ryo". The execution unit 52 also recognizes the meaning of the word "ryo" as "understood" from the answer given by the user 21 in response to the utterance, "It means 'understood'", associates the meaning of "understood" with the word "ryo", and causes the air purifier 1 to execute the utterance "I see, I'll remember that" to indicate this.

[0086] The execution unit 52 may cause the air purifier 1 to perform a speech in response to a speech prompt input by the user 21, and may also cause the air purifier 1 to perform a speech prompting the user 21 to respond to the performed speech. For example, the execution unit 52 may cause the air purifier 1 to perform a speech in response to the voice "Say something" input by the user 21, and then cause the air purifier 1 to perform the speech "How was it?". Alternatively, the execution unit 52 may change the topic of the speech to be performed by the air purifier 1 next time in response to the speech prompting the user 21 to respond, based on the response input by the user 21 to the air purifier 1. For example, if the execution unit 52 receives the response "It was interesting," it sets the genre of the topic to be spoken by the air purifier 1 next time to the same genre as the topic of the utterance spoken by the air purifier 1 this time. If the response "It wasn't fun" is received, it sets the genre of the topic to be spoken by the air purifier 1 next time to a different genre than the topic of the utterance spoken by the air purifier 1 this time.

[0087] This disclosure is not limited to the embodiments described above, and may be replaced with configurations that are substantially the same as those shown in the embodiments, configurations that produce the same effects, or configurations that can achieve the same purpose. [Explanation of Symbols]

[0088] 1 Air purifier, 11 Utterance, 12 Reply, 21 User, 31 Air purification unit, 32 Human presence sensor, 33 Illuminance sensor, 34 Operation unit, 35 Speaker, 36 Microphone, 37 Speech control device, 41, 42, 43 Detection result, 44, 45 Voice signal, 51 Voice analysis unit, 52 Execution unit, 61 Voice quality, 62 Dialect, 71 First utterance, 72 Second utterance, 73 First reply, 101 Characteristics, 111 Content, 112 Speaking style, 113, 123 Presence or absence of BGM, 121, 131 Content, 122, 132 Speaking style, 151, 161, 171 Dialect, 152, 162, 172 Frankness, 153, 163, 173 Volume

Claims

1. An execution unit that, when a condition is met, causes an electrical device to execute a first utterance, and the next time the condition is met, causes the electrical device to execute a second utterance that is different from the first utterance based on whether or not a first response was input to the electrical device in response to the first utterance, or the characteristics of the first response. Equipped with, The aforementioned conditions do not include the condition that a user's voice prompt has been detected. Speech control device.

2. The execution unit stops causing the electrical device to execute the first utterance if there is no first response or if the content of the first response is a request not to execute the utterance, for a first consecutive number of times. The speech control device according to claim 1.

3. The execution unit, upon receiving the first response, causes the content of the second utterance to differ from the content of the first utterance. The speech control device according to claim 1 or 2.

4. The execution unit, upon receiving the first response, causes the manner in which the second utterance is spoken to differ from the manner in which the first utterance is spoken. A speech control device according to any one of claims 1 to 3.

5. The execution unit stops the electrical device from executing the second utterance if, for a second consecutive number of times, there is no second response input to the electrical device in response to the second utterance, or the content of the second response is such that it requests that the utterance not be executed. A speech control device according to any one of claims 1 to 4.

6. The execution unit, if the absence of the first response or the content of the first response requesting that the utterance not be executed is repeated a first number of times in a row, stops causing the electrical device to execute the first utterance. The second number is greater than the first number. The speech control device according to claim 5.

7. The electrical device includes a voice analysis unit that identifies the voice quality of the voice input to the device. The execution unit performs control to make the second utterance different from the first utterance for each voice quality. A speech control device according to any one of claims 1 to 6.

8. The execution unit causes the way the second utterance is spoken to differ from the way the first utterance is spoken based on the way the first response is spoken. A speech control device according to any one of claims 1 to 7.

9. The execution unit makes the dialect used in the second utterance different from the dialect used in the first utterance, based on the dialect used in the first response. A speech control device according to any one of claims 1 to 8.

10. An execution unit that, when the conditions are met, causes the electrical device to execute a first utterance, and the next time the conditions are met, causes the electrical device to execute a second utterance that is different from the first utterance, based on whether or not a first response was input to the electrical device in response to the first utterance, or the characteristics of the first response. Equipped with, The execution unit, if it is confirmed for a second number of consecutive times that there is no second response input to the electrical device in response to the second utterance, or that the content of the second response requests that the utterance not be executed, then stops the electrical device from executing the second utterance. The execution unit, if the absence of the first response or the content of the first response requesting that the utterance not be executed is repeated a first number of times in a row, stops causing the electrical device to execute the first utterance. The second number is greater than the first number. Speech control device.

11. A speech control device according to any one of claims 1 to 10, The aforementioned electrical equipment, A speech system equipped with the following features.

Images