JP2015064450A - Information processing device, server, and control program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To surely present a previous phrase to a user without leaving it unpresented.SOLUTION: An interactive robot (100) includes: a reply acquisition unit (14) that obtains a phrase (6a) corresponding to a recognition result (4a) of a voice (1a); and a voice output unit (16) that, if a phrase (6b) is newly obtained before the phrase (6a) is presented and the phrase (6a) needs to be presented to a user, presents the phrase (6b) and then presents the phrase (6a).

Description

  The present invention relates to an information processing apparatus or the like that presents a predetermined phrase to a user according to a voice uttered by the user.

  2. Description of the Related Art Dialog systems that allow humans and robots to interact have been extensively studied. For example, Patent Document 1 below discloses an interactive information system that can continue and develop a dialog with a user more naturally. Patent Document 2 below discloses a dialogue method and a dialogue apparatus that maintain continuity of response style when a scenario transition occurs from a focal dialogue scenario to an auxiliary dialogue scenario.

  Here, in the prior art including the techniques disclosed in Patent Documents 1 and 2 above, the “question / answer service” (the user will wait until the answer from the robot to the question is completed). It is premised on one-by-one communication.

JP 2006-171719 A (released June 29, 2006) JP 2007-079397 A (published March 29, 2007)

  In the above interactive system, a delay in the previous answer to the previous call (question) from the user to the robot may cause a phenomenon in which the previous answer and the later answer to the subsequent call are mixed. In the prior art, this phenomenon can be ignored by the above assumption (the user does not make a later call while the previous answer is not presented).

  On the other hand, in the case of “normal communication” that assumes human-like communication (that is, it is assumed that the user will request the next answer even if the answer from the robot to the question is not presented), the above phenomenon Cannot be ignored. When this phenomenon occurs, there may arise a problem that the above-mentioned answer remains unpresented to the user.

  The present invention has been made in view of the above problems, and its purpose is that even when the previous phrase (answer) for the previous call and the subsequent phrase for the subsequent call are interlaced, An object of the present invention is to provide an information processing apparatus and the like that can be reliably presented to the user without leaving the previous phrase unpresented.

  In order to solve the above-described problem, an information processing apparatus according to an aspect of the present invention is an information processing apparatus that presents a predetermined phrase to a user according to a voice uttered by the user, and the voice is recognized. Acquisition means for acquiring the first phrase associated with the obtained result, and before presenting the first phrase, a second phrase different from the first phrase is newly acquired by the acquisition means If it is necessary, when it is necessary to present the first phrase to the user, the information processing apparatus includes a presentation unit that presents the first phrase after presenting the second phrase.

  In order to solve the above-described problem, a server according to one embodiment of the present invention causes the information processing apparatus to present a predetermined phrase to the user in accordance with a voice uttered by the user to the information processing apparatus. A server for controlling, which specifies a phrase associated with a result of recognizing the voice in a predetermined phrase set, and whether or not it is necessary to present the phrase to the user And generating means for generating necessity information and transmitting means for transmitting the phrase and the necessity information to the information processing apparatus.

  According to one aspect of the present invention, the information processing apparatus is a case where a previous phrase (first phrase) for a previous call and a later phrase (second phrase) for a subsequent call cross each other. Also, there is an effect that the above phrase can be surely presented to the user without leaving it unpresented.

  Further, according to one aspect of the present invention, the server may leave the previous phrase unpresented even when the previous phrase for the previous call and the subsequent phrase for the subsequent call cross. Thus, there is an effect that the information processing apparatus can be controlled so as to be surely presented to the user.

It is a block diagram which shows the principal part structure of the dialogue robot which concerns on the 1st Embodiment of this invention. It is a block diagram which shows the principal part structure of the server which concerns on the 1st Embodiment of this invention. It is a mimetic diagram showing roughly the dialog system concerning a 1st embodiment of the present invention. It is a table | surface which shows an example of a phrase set, (a) shows an example of the phrase set which the said interactive robot hold | maintains at a memory | storage part, (b) shows an example of the phrase set which the said server hold | maintains at a memory | storage part . It is a flowchart which shows an example of the process performed in the said interactive system. It is a block diagram which shows the principal part structure of the dialogue robot which concerns on the 2nd Embodiment of this invention. It is a flowchart which shows an example of the process which the said interactive robot performs.

Embodiment 1
The first embodiment (Embodiment 1) of the present invention will be described with reference to FIGS.

(Outline of Dialog System 300)
FIG. 3 is a schematic diagram schematically showing the dialogue system 300. As shown in FIG. 3, the dialogue system 300 includes a dialogue robot 100 and a server 200. According to the dialogue system 300, the user can obtain various information by interacting with the dialogue robot 100 by using a voice using a natural language.

  The interactive robot (information processing device) 100 is a device that presents a predetermined phrase (response text) to the user according to the voice uttered by the user. It is only necessary to be a device that can input voice and can present the predetermined phrase based on the input voice, and is not limited to a dialogue robot (for example, the dialogue robot 100 is a tablet terminal, a smartphone, a personal computer, or the like). It can also be realized by a computer).

  The server 200 is a device that controls the interactive robot 100 so as to present a predetermined phrase to the user in accordance with a voice uttered by the user to the interactive robot 100. As shown in FIG. 3, the interactive robot 100 and the server 200 are communicably connected via a communication network according to a predetermined communication method.

  In the interactive system 300, when the interactive robot 100 acquires a phrase as an answer to the voice uttered by the user, the following two are conceivable.

(1) Acquire a phrase from the local When the result of recognition of the voice by the interactive robot 100 is included in a phrase set (local dictionary) stored in a storage unit included in the interactive robot 100, the interactive robot 100 The predetermined phrase corresponding to the recognized result in the phrase set is acquired from the storage unit.

  For example, when the user calls the interactive robot 100 with “good morning” by voice, the result of recognizing the voice is included in the phrase set, so the interactive robot 100 recognizes the result of the recognition in the phrase set. The phrase “Good morning, how are you today?” (See the first row of the table shown in FIG. 4A) is acquired from the storage unit and the phrase is presented to the user by voice. To do. As in the case of (1) above, the interactive robot 100 can return a phrase for a simple call to the user.

(2) Acquiring a phrase from the cloud When the result of recognition of the voice by the interactive robot 100 is not included in the phrase set, the interactive robot 100 transmits (uploads) the voice to the server 200. The server 200 specifies a predetermined phrase corresponding to the recognized result in the phrase set (cloud dictionary) stored in the storage unit included in the server 200. The interactive robot 100 acquires (downloads) the phrase specified by the server 200.

  For example, when the user calls the interactive robot 100 with a voice saying “What is the weather today?”, The result of recognizing the voice is not included in the phrase set (local dictionary). 100 acquires the phrase “It's raining, bring an umbrella” from the server 200 and presents the phrase to the user by voice. Since there are usually more patterns (detected words) that match the result of speech recognition in the cloud dictionary than in the local dictionary, the interactive robot 100 has the same meaning as in (2) above. It is possible to return an appropriate phrase to the user even for a complicated call.

  In both cases (1) and (2), it may be required to present another phrase before the interactive robot 100 presents a phrase. For example, when the user calls the interactive robot 100 by voice saying “What is the weather today?” (Previous call), the interactive robot 100 “rains, please bring an umbrella” (first Before presenting the phrase "Phrase of" to the user, the user asks "By the way, what is the sports news?" (Later appeal), and "Yesterday, Team A was a big win" (later phrase) May be required to be presented.

  Thus, the reason why the previous phrase for the previous call and the subsequent phrase for the subsequent call cross each other is because the timing of presenting the previous phrase to the user may be delayed. That is, in the case of (1) above, the process of extracting (searching) an appropriate phrase from the local dictionary is heavy, and in the case of (2) above, between the interactive robot 100 and the server 200, The above timing may be delayed due to communication stagnation. In particular, as described in the present embodiment, when the main phrase set is on the server 200 side (takes a cloud configuration), the timing is likely to be delayed due to deterioration of the communication environment. In a dialogue that simulates a human conversation, an unnatural delay (an odd gap) is not allowed, so even if the previous phrase for the previous call is not presented, the user will Is assumed to be required. At this time, the previous phrase may remain unpresented.

  Therefore, when the subsequent phrase is newly acquired before the previous phrase is presented, the interactive robot 100 presents the subsequent phrase when it is necessary to present the previous phrase to the user. Later, the previous phrase is presented. As a result, when it is determined that it is necessary to present the previous phrase to the user, the interactive robot 100 can always present the previous phrase to the user without leaving the previous phrase unpresented. .

  In the following description, as an example, the previous call “What's the weather today?” Is made by the user with the sound 1a, and then the later call “What is sports news?” Is made with the sound 1b. Suppose. Then, as an answer to the above call, the previous phrase (phrase 6a) “It's raining, please bring an umbrella” with the voice 1c, and as an answer to the above call, “Yesterday, Team A was a big win. It is assumed that the dialogue robot 100 presents the subsequent phrase “tayo” (phrase 6b) to the user by the voice 1d.

(Configuration of Dialogue Robot 100)
FIG. 1 is a block diagram showing a main configuration of the interactive robot 100. As shown in FIG. 1, the dialogue robot 100 includes a communication unit 50a (reception unit 51a, transmission unit 52a), a control unit 10a (voice detection unit 11, voice recognition unit 12, answer confirmation unit 13, answer acquisition unit 14, A voice sending unit 15, a voice output unit 16, a flag determination unit 17, an answer storage unit 18), a voice input / output unit 40 (a microphone 41, a speaker 42), and a storage unit 30a are provided.

  The voice input / output unit 40 controls voice input / output with respect to the interactive robot 100. The voice input / output unit 40 includes a microphone 41 and a speaker 42.

  The microphone 41 collects sounds from the surroundings of the interactive robot 100 and outputs the sound signals 2 a and 2 b representing the sounds 1 a and 1 b to the sound detection unit 11.

  The speaker 42 converts the audio signal 2c and the audio signal 2d input from the audio output unit 16 into the audio 1c and the audio 1d, respectively, and outputs the audio 1c and the audio 1d to the outside. The speaker 42 may be built in the interactive robot 100, may be externally connected via an external connection terminal, or may be connected so as to be communicable. Good.

  The control unit 10a comprehensively controls various functions of the interactive robot 100. The control unit 10 a includes a voice detection unit 11, a voice recognition unit 12, a response confirmation unit 13, a response acquisition unit 14, a voice transmission unit 15, a voice output unit 16, a flag determination unit 17, and a response storage unit 18.

  The voice detection unit 11 detects voice uttered by the user. Specifically, when the audio signal 2 a or the audio signal 2 b is input from the microphone 41, the audio signal 2 a and the audio signal 2 b are respectively converted into audio information 3 a and audio information 3 b that can be digitally processed by the interactive robot 100. The voice information 3 a and the voice information 3 b are output to the answer confirmation unit 13 and the voice recognition unit 12.

  The voice recognition unit 12 recognizes a voice uttered by the user with respect to the interactive robot 100. Specifically, when the voice information 3a or the voice information 3b is input from the voice detection unit 11, the voice recognition unit 12 recognizes the voice information 3a or the voice information 3b according to a predetermined voice recognition algorithm. A result (recognition result 4a or recognition result 4b) is obtained. Here, the recognition result 4a or the recognition result 4b includes at least the voice information 3a or text converted from the voice information 3b (representing the contents spoken by the user by characters). As the speech recognition algorithm, known algorithms may be adopted as appropriate. The voice recognition unit 12 outputs the recognition result 4 a and the recognition result 4 b to the answer confirmation unit 13.

  The answer determination unit 13 determines the answer to be returned to the user based on the result of recognizing the voice. Specifically, when the recognition result 4a or the recognition result 4b is input from the speech recognition unit 12, the answer determination unit 13 refers to the phrase set 5a stored in the storage unit 30a, and the recognition result 4a or It is determined whether or not a pattern (detection word) including the text included in the recognition result 4b is included in the phrase set 5a. If it is determined that the phrase is included, the answer determination unit 13 determines the phrase associated with the pattern as an answer to be returned to the user, and outputs the answer to the answer acquisition unit 14. On the other hand, if it is determined not to be included, the answer confirmation unit 13 outputs the voice information 3 a input from the voice detection unit 11 to the voice transmission unit 15. At this time, the answer determination unit 13 may present the phrase to the user by outputting a phrase for suspending the answer to the answer acquisition unit 14.

  FIG. 4 is a table showing an example of a phrase set. (A) shows the phrase set 5a (local dictionary) held in the storage unit 30a by the interactive robot 100, and (b) shows the server 200 by the storage unit 30b. The phrase set 5b (cloud dictionary) held in FIG. Here, the “phrase set” (dictionary) is a data set in which a predetermined phrase (and an important flag 7) is associated with a predetermined pattern (detection word). The “phrase” represents a preferable answer to the predetermined pattern in a predetermined data format (for example, a text format). Further, the “important flag” (necessity information) is information indicating whether or not it is necessary to present the phrase 6a to the user. For example, a binary value having a value of “1” or “0” is used. It may be a flag. At this time, when the important flag 7 is “1”, it indicates “important” (it is necessary to present the phrase 6a to the user), and when it is “0”, “normal” (not necessary) is indicated. May show.

  For example, the text included in the recognition result 4a (the result of recognizing the voice 1a) input from the voice recognition unit 12 is “Today's weather”, but as shown in FIG. A pattern including text does not exist in the phrase set 5a. At this time, the answer confirming unit 13 outputs the voice information 3a to the voice sending unit 15 and also puts the answer on hold phrase (in FIG. 4 (a), “Please wait a bit”, “That's right”, “Hmm” And the like) are output to the answer acquisition unit 14.

  The answer acquisition unit (acquisition means) 14 acquires the phrase (first phrase) 6a associated with the result of recognition of the voice 1a. Specifically, when the phrase 6 a is input from the answer confirmation unit 13 as an answer to be returned to the user, the answer acquisition unit 14 outputs the phrase 6 a to the voice output unit 16. Similarly, when the phrase (second phrase) 6b and the important flag 7 are input from the reception unit 51a, the answer acquisition unit 14 outputs the phrase 6b to the voice output unit 16.

  On the other hand, when the phrase 6a and the important flag 7 are input from the receiving unit 51a, (1) when the phrase 6b is acquired before the phrase 6a is presented (the phrase 6a is output as the audio 1c by the audio output unit 16) Before the answer acquisition unit 14 acquires the phrase 6b), the answer acquisition unit 14 outputs the phrase 6a and the important flag 7 to the flag determination unit 17. (2) In other cases, the answer acquisition unit 14 outputs the phrase 6 a to the voice output unit 16.

  The voice sending unit 15 sends the voice information 3a and the voice information 3b to the server 200 via a communication network according to a predetermined communication method. Specifically, when the voice information 3a or the voice information 3b is input from the answer confirmation unit 13, the voice sending unit 15 outputs the voice information 3a or the voice information 3b to the transmission unit 52a.

  The voice output unit 16 presents the phrase 6a and the phrase 6b to the user by outputting the phrase 6a and the phrase 6b by voice through the speaker 42. Specifically, when the phrase 6 a or the phrase 6 b is input from the answer acquisition unit 14, the phrase 6 a or the phrase 6 b is output to the speaker 42.

  When the phrase 6b is newly acquired before the phrase 6a is presented, the flag determination unit 17 determines whether or not it is necessary to present the phrase 6a to the user based on the important flag 7. . Specifically, when the phrase 6a and the important flag 7 are input from the answer acquiring unit 14, the flag determining unit 17 determines whether the important flag 7 indicates “important” or “normal”. When it is determined that “important” is indicated, the flag determination unit 17 outputs the phrase 6 a to the answer storage unit 18.

  When it is determined that “normal” is indicated, the flag determination unit 17 may discard the phrase 6 a without outputting the phrase 6 a to the answer storage unit 18. Alternatively, the phrase 6a may be stored in a predetermined storage area of the storage unit 30a in preparation for a case where it is requested again to be presented to the user later. For example, it is assumed that a previous call “Let me hear an interesting story” is made from the user by the voice 1a, and a later call “What is sports news?” Is made by the voice 1b. Then, when the previous phrase (phrase 6a) “Old grandfather and grandmother ...” (see the sixth line of the table shown in FIG. 4B) is acquired as an answer to the above call. Since the important flag 7 of the phrase 6a indicates “normal”, the dialogue robot 100 gives the user the following phrase (phrase 6b) “Yesterday, Team A was a big win” as an answer to the later call. Even after the presentation, the phrase 6a may not be presented.

  When the phrase 6b is newly acquired before the phrase 6a is presented, the answer storage unit (storage means) 18 is required to present the phrase 6a to the user when the phrase 6a needs to be presented to the user. Store in the storage unit 30a. Specifically, when the phrase 6a is input from the flag determination unit 17, the answer storage unit 18 stores the phrase 6a in the storage unit 30a.

  Here, when the phrase 6b is newly acquired before the phrase 6a is presented, the voice output unit (presentation means) 16 is required to present the phrase 6a to the user. Then, the phrase 6a is presented. That is, when the flag determination unit 17 determines that the important flag 7 indicates “important”, and the answer storage unit 18 stores the phrase 6a in the storage unit 30a, the voice output unit 16 determines that the phrase 6b Is output to the speaker 42, the phrase 6a is read from the storage unit 30a, and the phrase 6a is output to the speaker 42.

  The communication unit 50a communicates with the outside via a communication network according to a predetermined communication method. It is only necessary to have an essential function for realizing communication with an external device, and the communication line, the communication method, the communication medium, and the like are not limited. The communication unit 50a can be configured by a device such as an Ethernet (registered trademark) adapter, for example. The communication unit 50a can use a communication method or a communication medium such as IEEE802.11 wireless communication or Bluetooth (registered trademark). The communication unit 50a includes a reception unit 51a and a transmission unit 52a.

  The receiving unit 51a receives the phrase 6a and the phrase 6b from the server 200 by communicating with the outside via the communication network according to the predetermined communication method. The receiving unit 51a outputs the received phrase 6a and phrase 6b to the answer acquiring unit 14.

  When the voice information 3a or the voice information 3b is input from the voice sending unit 15, the transmission unit 52a communicates with the outside via the communication network in accordance with the predetermined communication method, so that the server 200 transmits the voice information 3a or the voice information 3a. The voice information 3b is transmitted.

  The storage unit 30a is a storage device that can store the phrase set 5a and the phrase 6a. The storage unit 30a can be configured by, for example, a hard disk, an SSD (silicon state drive), a semiconductor memory, a DVD, or the like.

(Configuration of server 200)
FIG. 2 is a block diagram illustrating a main configuration of the server 200. As shown in FIG. 2, the server 200 includes a communication unit 50b (reception unit 51b, transmission unit 52b), a control unit 10b (voice acquisition unit 21, voice recognition unit 22, answer identification unit 23, flag generation unit 24, information A sending unit 25) and a storage unit 30b are provided.

  Since the communication unit 50b is the same as the communication unit 50a, detailed description thereof is omitted. The communication unit 50b includes a reception unit 51b and a transmission unit 52b.

  The receiving unit 51b receives the voice information 3a and the voice information 3b from the interactive robot 100 by communicating with the outside via a communication network according to a predetermined communication method. The reception unit 51b outputs the received audio information 3a and audio information 3b to the audio acquisition unit 21.

  When the phrase 6a or the phrase 6b and the important flag 7 are input from the information transmission unit 25, the transmission unit 52b communicates with the outside through the communication network according to the predetermined communication method, thereby causing the interactive robot 100 to The phrase 6a or the phrase 6b and the important flag 7 are transmitted.

  The voice acquisition unit 21 acquires the voice information 3a and the voice information 3b from the interactive robot 100 via a communication network according to a predetermined communication method. Specifically, when the voice information 3a or the voice information 3b is input from the reception unit 51b, the voice acquisition unit 21 outputs the voice information 3a or the voice information 3b to the voice recognition unit 22.

  The voice recognition unit 22 recognizes a voice uttered by the user with respect to the interactive robot 100. Specifically, when the voice information 3a or the voice information 3b is input from the voice acquisition unit 21, the voice recognition unit 22 recognizes the voice information 3a or the voice information 3b according to a predetermined voice recognition algorithm. A result (recognition result 4a or recognition result 4b) is obtained. The voice recognition unit 22 outputs the recognition result 4a or the recognition result 4b to the answer specifying unit 23.

  The answer specifying unit (specifying means) 23 specifies the phrase associated with the result of recognition of speech (recognition result 4a or recognition result 4b) in the phrase set 5b. Specifically, when the recognition result 4a or the recognition result 4b is input from the voice recognition unit 22, the phrase set 5b stored in the storage unit 30b is referred to, and the above-described recognition result 4a or the above-mentioned recognition result 4b included in the recognition result 4b. Extract patterns that contain text. Then, the answer specifying unit 23 specifies the phrase (phrase 6a or phrase 6b) associated with the pattern as an answer to be returned to the user, and outputs the answer to the flag generating unit 24.

  For example, the text included in the recognition result 4a (the result of recognizing the voice 1a) input from the voice recognition unit 22 is “Today's weather”, but the text includes the pattern “weather”. According to the first row of the table shown in FIG. 4B, the answer specifying unit 23 returns the phrase 6a to the user, “Today's weather is rain, please bring an umbrella”. And output to the flag generation unit 24. Note that when the pattern including the text is not included in the phrase set 5b (see “-no matching-” in the ninth row of the table shown in FIG. 4B), the speech recognition unit 22 performs speech recognition or A phrase that gives feedback to the user that the phrase has failed to be identified (for example, “I don't know at all”) is identified as an answer to be returned to the user, and is output to the flag generation unit 24 (that is, a pattern that matches the text) Even if it is not included in the phrase set 5b, the phrase is always specified).

  The flag generation unit (generation means) 24 generates an important flag depending on whether or not it is necessary to present the phrase to the user. Specifically, when the phrase 6a or the phrase 6b is input from the answer specifying unit 23, the flag generating unit 24 sets the “importance” (the table shown in (b) of FIG. 4) to the phrase 6a or the phrase 6b. The second column is referred to, and it is determined whether or not the phrase in the row marked with a circle in the second column is “high”. When it is determined that the flag is set, the flag generation unit 24 generates an important flag 7 indicating “important” (representing that it is necessary to present the phrase 6a to the user). The phrase (phrase 6 a or phrase 6 b) corresponding to the important flag 7 is output to the information transmission unit 25. When it is determined that it is not set, the flag generation unit 24 generates an important flag 7 indicating “normal” (representing that it is not necessary to present the phrase 6a to the user), and the important flag 7 is stored as information. The data is output to the sending unit 25.

  The information transmission unit (transmission means) 25 transmits the phrase 6a and the phrase 6b to the interactive robot 100 via a communication network according to a predetermined communication method. Specifically, when the phrase 6a or phrase 6b and the important flag 7 are input from the flag generation unit 24, the information sending unit 25 sends the phrase 6a or the phrase 6b and the important flag 7 to the transmission unit 52b. Output.

  The storage unit 30b is a storage device that can store the phrase set 5b. The storage unit 30b can be configured by, for example, a hard disk, an SSD, a semiconductor memory, a DVD, or the like, similarly to the storage unit 30a.

(Processes executed in the dialogue system 300)
FIG. 5 is a flowchart illustrating an example of processing executed in the interactive system 300. In the following description, parenthesized “˜step” represents each step included in the control method of the interactive robot 100.

  When the voice detection unit 11 detects the voice 1a “What's the weather today?” Issued by the user (YES in step 1, “step” is abbreviated as “S” hereinafter), the voice recognition unit 12 reads the voice 1a. Is recognized (S2). The answer determination unit 13 refers to the phrase set 5a (local dictionary) stored in the storage unit 30a, and whether or not the pattern including the text included in the recognition result 4a is included in the phrase set 5a (matches) (S3). When it is determined that the phrase is included (YES in S3), the answer confirmation unit 13 confirms the phrase associated with the pattern as an answer to be returned to the user (S4). When the answer acquisition unit 14 acquires the phrase (S16), the voice output unit 16 presents the phrase as a voice to the user (reproduces the voice, S17), and the process ends. On the other hand, when it is determined that the answer is not included (NO in S3), the answer determination unit 13 determines a phrase for holding the answer “slightly wait” (S5), and the voice output unit 16 uses the phrase as a voice. To the user (S6). Further, the voice sending unit 15 sends the voice information 3a to the server 200 (S7).

  When the reception unit 51b of the server 200 receives the audio information 3a and the audio acquisition unit 21 acquires the audio information 3a (S8), the audio recognition unit 22 recognizes the audio 1a (S9). The answer specifying unit 23 specifies the phrase 6a associated with the result of recognizing the voice 1a (recognition result 4a) in the phrase set 5b (cloud dictionary) (S10). The flag generation unit 24 determines whether or not it is necessary to present the phrase 6a to the user (whether the importance of the phrase 6a is “high”) (S11), and is determined to be “high”. If it is determined (YES in S11), the important flag 7 indicating "high" is generated (S12). If not determined as "high" (NO in S11), the important flag 7 indicating "normal" is generated (S13). . The information sending unit 25 transmits the phrase 6a and the important flag 7 to the interactive robot 100 (S14), and the receiving unit 51a of the interactive robot 100 receives the phrase 6a and the important flag 7 (S15). The answer acquisition unit 14 acquires the phrase 6a and the important flag 7 (S16, acquisition step).

  When the voice detection unit 11 detects the voice 1b “Where is sports news?” Further issued by the user during the above-described S7 to S15 (when interrupt processing occurs, YES in S18), the flag determination unit 17 determines whether it is necessary to present the phrase 6a to the user based on the important flag 7 received in S15 (S19). When it is determined that it is necessary (when the important flag 7 indicates “important”, YES in S19), the answer storage unit 18 stores the phrase 6a in the storage unit 30a (S20).

  The interactive robot 100 and the server 200 execute the processes of S2 to S17 on the voice 1b (represented as “A” in the flowchart shown in FIG. 5). If no further voice is detected from the user during S7 to S15 (NO in S18), the answer acquisition unit 14 acquires the phrase 6b “Team A was a big win yesterday” as an answer to the voice 1b. (S16, acquisition step), the phrase 6b is presented to the user by voice 1d (S17, presentation step). The series of processes S2 to S17 called from “A” is completed, and the flow of the process returns immediately before S21.

  When the voice output unit 16 needs to present the phrase 6a acquired in S16 to the user (that is, the flag determination unit 17 determines that the important flag 7 indicates “important”, When the phrase 6a is stored in the storage unit 30a by the answer storage unit 18, the phrase 6a is presented to the user by the voice 1c (S17, presentation step). Note that after the phrase 6a is presented to the user, the voice output unit 16 may delete the phrase 6a stored in the storage unit 30a from the storage unit 30a.

  As described above, for example, when the communication is stagnant, the timing at which the interactive robot 100 presents the answer (phrase 6a) to the user is delayed, and the user gives a further voice 1b during S7 to S15. When the phrase 6b is newly acquired before the phrase 6a is presented, when it is necessary to present the phrase 6a to the user, the interactive robot 100 presents the phrase 6b. The phrase 6a is presented.

  In addition, when a further voice is detected from the user while performing the processes of S2 to S16 for the voice 1b (YES in S18 for the second time), the interactive robot 100 performs the processes of S19 and S20. After executing the above, the processes of S2 to S17 can be further executed. Furthermore, the number of times of execution is arbitrary, and the number of times may be set in advance.

[Embodiment 2]
A second embodiment (Embodiment 2) of the present invention will be described based on FIGS. In the present embodiment, only the configuration added to the first embodiment and the configuration different from the configuration of the first embodiment will be described. That is, all the configurations described in the first embodiment can be included in the second embodiment. Moreover, the definition of the term described in Embodiment 1 is the same also in Embodiment 2.

(Differences from Embodiment 1)
FIG. 6 is a block diagram showing a main configuration of the dialogue robot 101. The interactive robot 100 according to the first embodiment is communicably connected to the server 20 to form the interactive system 300. When the phrase 6a or the phrase 6b does not exist in the phrase set 5a, the interactive robot 100 The phrase 6b was obtained from the server 20 (cloud configuration).

  On the other hand, in the present embodiment, the phrase set 5a and the phrase set 5b are stored in the storage unit 30a included in the dialog robot 101, and the dialog robot 101 acquires the phrase 6a or the phrase 6b from the storage unit 30a. Yes (stand-alone configuration). Therefore, as shown in FIG. 6, the dialogue robot 101 does not include the communication unit 50a and the voice sending unit 15 included in the dialogue robot 100 (it is only unnecessary to communicate with the server 20 in order to acquire a phrase). Therefore, the dialogue robot 101 may include the communication unit 50a or the voice sending unit 15).

  Since the interactive robot 101 does not communicate with the server 20 in order to acquire a phrase, the problem that the timing of presenting a phrase to the user is delayed due to “communication between the interactive robot and the server”. It cannot happen. However, the phrase set 5b (cloud dictionary including more patterns than the phrase set 5a) that is centrally managed by the server 200 in the first embodiment is held as a local dictionary in the second embodiment by the interactive robot 101 (storage unit 30a). Therefore, the process of extracting an appropriate phrase from the local dictionary becomes heavier and the timing can be delayed. Therefore, as described above, there is a possibility that the previous phrase for the previous call and the subsequent phrase for the subsequent call are mixed to leave the previous phrase unpresented.

  Therefore, when the later phrase is newly acquired before the previous phrase is presented, the dialogue robot 101 presents the later phrase when it is necessary to present the previous phrase to the user. Later, the previous phrase is presented. As a result, when it is determined that it is necessary to present the previous phrase to the user, the interactive robot 101 can always present the previous phrase to the user without leaving the previous phrase unpresented. .

(Configuration of Dialogue Robot 101)
The answer determination unit 13 determines the answer to be returned to the user based on the result of recognizing the voice. Specifically, when the recognition result 4a or the recognition result 4b is input from the speech recognition unit 12, the answer determination unit 13 refers to the phrase set 5a stored in the storage unit 30a, and the recognition result 4a or It is determined whether or not a pattern including the text included in the recognition result 4b is included in the phrase set 5a.

  If it is determined that the phrase is included, the answer determination unit 13 determines the phrase associated with the pattern as an answer to be returned to the user, and outputs the answer to the answer acquisition unit 14. On the other hand, when it is determined not to be included, the answer confirmation unit 13 refers to the phrase set 5b stored in the storage unit 30a and extracts a pattern including the text. Then, the answer determination unit 13 determines the phrase (phrase 6a or phrase 6b) associated with the pattern as an answer to be returned to the user. At this time, the answer determination unit 13 may present the phrase to the user by outputting a phrase for suspending the answer to the answer acquisition unit 14.

  Next, the answer determination unit 13 generates an important flag according to whether or not it is necessary to present the phrase to the user. Specifically, the answer determination unit 13 determines whether or not “importance” is set for the phrase 6a or the phrase 6b. If it is determined that it is set, the answer determination unit 13 generates an important flag 7 indicating “important”, and the important flag 7 and a phrase (phrase 6a or phrase 6b) corresponding to the important flag 7 Is output to the answer acquisition unit 14. If it is determined that it is not set, the answer determination unit 13 generates an important flag 7 indicating “normal” and outputs the important flag 7 to the answer acquiring unit 14.

  The answer acquisition unit (acquisition means) 14 acquires the phrase 6 a, the phrase 6 b, and the important flag 7 from the answer determination unit 13. The processing after obtaining these is the same as that described in the first embodiment.

(Processing executed by the interactive robot 101)
FIG. 7 is a flowchart illustrating an example of processing executed by the interactive robot 101. The flowchart is a process related to communication (S7, S8, S14, S15) from the flowchart illustrated in FIG. 5 (process executed in the dialogue system 300), and a process in which the server 20 performs voice recognition again (S9). ), And the processing (S10 to S13) executed in the server 200 is executed (S22 to S25) in the interactive robot 101 (answer confirmation unit 13).

  That is, the answer determination unit 13 specifies the phrase 6a associated with the result of recognizing the voice 1a (recognition result 4a) in the phrase set 5b (cloud dictionary) (S22). Then, the answer confirmation unit 13 determines whether or not it is necessary to present the phrase 6a to the user (S23). When it is determined to be “high” (YES in S23), “high” is indicated. The important flag 7 is generated (S24), and if it is not determined to be “high” (NO in S23), the important flag 7 indicating “normal” is generated (S25).

[Embodiment 3]
A third embodiment (Embodiment 3) of the present invention will be described. In the present embodiment, only a configuration added to the first embodiment or the second embodiment or a configuration different from the configuration of the first embodiment or the second embodiment will be described. That is, all the configurations described in the first embodiment or the second embodiment can be included in the third embodiment. Moreover, the definition of the term described in Embodiment 1 or Embodiment 2 is the same also in Embodiment 3.

(Phrase dynamic change)
The interactive robot (including the control unit 10a) may present the dynamically changed phrase to the user so as to reflect the situation at the time when the phrase is acquired. That is, the user calls the dialogue robot “What's the weather today?” With the voice 1a, and the dialogue robot presents the phrase 6a, “It's raining, bring an umbrella” to the user with the voice 1c. As described above, when the “today's weather” is “sunny”, the dialogue robot can present the phrase 6a “sunny, don't need an umbrella”.

  Specifically, the answer determination unit 13 (in the case of a stand-alone configuration) or the answer specifying unit 23 (in the case of a cloud configuration) receives information that dynamically changes from a predetermined web service (for example, text indicating a weather forecast). Acquire and generate a complemented phrase based on the information (for example, insert the above text such as “Sunny”, “Rain” in the updatable part “...” of “... Dayo”) The phrase set 5b is updated with the complemented phrase. And the reply acquisition part 14 acquires the said phrase complemented by the predetermined information which changes dynamically which can be acquired from external services (For example, the said predetermined web service etc. which provide a weather forecast). Thereby, the dialog robot can present the dynamically changing phrase to the user.

(Dynamic setting of important flag)
The important flag 7 included in the phrase set 5b is set in advance by the user in accordance with the pattern or the phrase. That is, the user can register a pattern (detection word) in a predetermined record of the phrase set 5a or the phrase set 5b and set the important flag 7 for the pattern. Further, the user can newly set or change the important flag 7 even after registering the pattern.

  On the other hand, the important flag 7 may be dynamically changed. For example, when the volume of the voice 1a or the voice 1b is higher than normal (when the gain of the voice signal 2a or the voice signal 2b exceeds a predetermined threshold (or an average value of gains obtained so far)) The determination unit 13 (in the case of a stand-alone configuration) or the flag generation unit 24 (in the case of a cloud configuration) may generate the important flag 7 indicating “important”. Conversely, when the volume is lower than normal, the important flag 7 indicating “normal” may be generated.

  Alternatively, when it is determined that the voice 1a or the voice 1b is questionable (when the frequency of the voice 1a or the voice 1b increases, the recognition result 4a or the recognition result 4b ends in a predetermined pattern, etc.) The answer determination unit 13 or the flag generation unit 24 may generate the important flag 7 indicating “important”. On the contrary, when it is determined that it is not questionable, the important flag 7 indicating “normal” may be generated.

[Example of software implementation]
The control blocks (particularly, the control unit 10a and the control unit 10b) of the interactive robot 100 and the server 200 may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or a CPU (Central It may be realized by software using a Processing Unit. In the latter case, the interactive robot 100 and the server 200 have a CPU that executes instructions of a program (control program) that is software for realizing each function, and the program and various data are recorded so as to be readable by the computer (or CPU). A ROM (Read Only Memory) or a storage device (these are referred to as “recording media”), a RAM (Random Access Memory) for expanding the program, and the like are provided. And the objective of this invention is achieved when a computer (or CPU) reads the said program from the said recording medium and runs it. As the recording medium, a “non-temporary tangible medium” such as a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like can be used. The program may be supplied to the computer via an arbitrary transmission medium (such as a communication network or a broadcast wave) that can transmit the program. The present invention can also be realized in the form of a data signal embedded in a carrier wave in which the program is embodied by electronic transmission.

[Summary]
The information processing apparatus according to the first aspect of the present invention is an information processing apparatus (the interactive robot 100 or the interactive robot 101) that presents a predetermined phrase to the user according to the voice (1a, 1b) uttered by the user. , The acquisition means (answer acquisition unit 14) for acquiring the first phrase (6a) associated with the result (4a, 4b) recognized by the voice, and before presenting the first phrase, When it is necessary to present the first phrase to the user when the second phrase (6b) different from the first phrase is newly acquired by the acquisition means, the second phrase And presenting means (speech output unit 16) for presenting the first phrase.

  When it is assumed that a human and a machine communicate naturally, there is a case in which a further call is made from the user to the machine before the presentation due to a delay in the presentation of the phrase for the call from the user to the machine. At this time, the phrase is newly presented only for the further call, and the previous phrase may remain unpresented.

  According to the above configuration, the information processing apparatus needs to present the first phrase to the user when the second phrase is newly acquired before presenting the first phrase. In some cases, after presenting the second phrase, the first phrase is presented. Therefore, the information processing apparatus can reliably present to the user without leaving the previous phrase (first phrase) unpresented.

  In the information processing device according to aspect 2 of the present invention, in the above aspect 1, the acquisition unit needs information (important flag 7) indicating whether or not it is necessary to present the first phrase to the user. ) And the presenting means may present the first phrase after presenting the second phrase when indicating that the necessity information is necessary.

  According to the configuration, the information processing apparatus can know whether or not it is necessary to present the first phrase to the user based on the necessity information. And when it determines with it being necessary to show to a user by the said necessity information, the said information processing apparatus presents the said 1st phrase after presenting the said 2nd phrase. Therefore, the information processing apparatus can reliably present the first phrase to the user without leaving the first phrase unpresented.

  The information processing device according to aspect 3 of the present invention, in the above-described aspect 1 or 2, when the second phrase is newly acquired before presenting the first phrase, When it is necessary to present it to the user, it further comprises storage means (answer storage section 18) for storing the first phrase in a predetermined storage section (30a), wherein the presentation means is configured to store the predetermined storage section. When the first phrase is stored in the section, the first phrase may be read from the predetermined storage section and presented after the second phrase is presented.

  According to the above configuration, when the information processing apparatus needs to present the first phrase to the user, the information processing apparatus causes the storage unit to save the first phrase. Then, when the first phrase is present in the storage unit, the information processing apparatus presents the first phrase to the user after presenting the second phrase. Therefore, the information processing apparatus can reliably present the first phrase to the user without leaving the first phrase unpresented.

  The server according to aspect 4 of the present invention is a server (200) that controls the information processing apparatus so as to present a predetermined phrase to the user in accordance with a voice uttered by the user to the information processing apparatus. , Specifying means (answer specifying unit 23) for specifying a phrase associated with the result of recognizing the voice in a predetermined phrase set (5b), and whether or not it is necessary to present the phrase to the user Accordingly, a generation unit (flag generation unit 24) for generating necessity information and a transmission unit (information transmission unit 25) for transmitting the phrase and the necessity information to the information processing apparatus are provided.

  According to the said structure, the said server transmits the said phrase and the said necessity information to the said information processing apparatus according to the result of having recognized the said audio | voice. Here, since the pattern that matches the result of recognizing the voice is usually included more in the phrase set held by the server than the phrase set held by the information processing apparatus, Even when a complicated call is made to the information processing apparatus, the server can control the information processing apparatus to return an appropriate phrase.

  Further, whether or not it is necessary to present the phrase to the user can be notified to the information processing apparatus by the necessity information. Therefore, the server can control the information processing apparatus so as to reliably present the phrase to the user without leaving the phrase unpresented.

  An interactive system (300) according to another aspect of the present invention includes the information processing apparatus according to any one of the above aspects and the server according to the above aspect.

  Therefore, the dialog system has the same effect as the information processing apparatus according to any one of the aspects or the server according to the aspect.

  An information processing apparatus control method according to another aspect of the present invention is an information processing apparatus control method for presenting a predetermined phrase to a user according to a voice uttered by the user, and the result of the recognition of the voice An acquisition step (S16) for acquiring a first phrase associated with the first phrase, and before presenting the first phrase, a second phrase different from the first phrase is newly acquired in the acquisition step In this case, when it is necessary to present the first phrase to the user, it includes a presenting step (S17) of presenting the first phrase after presenting the second phrase. .

  Therefore, the control method of the information processing apparatus has the same effect as the information processing apparatus according to the aspect.

  The information processing device and the server according to each aspect of the present invention may be realized by a computer. In this case, the information processing device and the server are operated by causing the computer to operate as each unit included in the information processing device and the server. A control program for an information processing apparatus that implements the server in a computer, a control program for the server, and a computer-readable recording medium that records them are also within the scope of the present invention.

  The present invention is not limited to the above-described embodiments, and various modifications can be made within the scope of the claims, and the technical means disclosed in different embodiments can be appropriately combined. Embodiments to be made are also included in the technical scope of the present invention. Furthermore, a new technical feature can be formed by combining the technical means disclosed in each embodiment.

  The present invention can be widely applied to an apparatus that presents a predetermined phrase to the user according to the voice uttered by the user.

1a voice 1b voice 4a recognition result (recognition result)
4b Recognition result (recognized result)
5b Phrase set (predetermined phrase set)
6a Phrase (first phrase)
6b Phrase (second phrase)
7 Important flag (necessity information)
14 Response acquisition unit (acquisition means)
16 Audio output unit (presentation means)
18 Answer storage (storage means)
23 answer identification part (identification means)
24 Flag generator (generator)
25 Information sending part (transmission means)
30a storage unit (predetermined storage unit)
100 Dialogue robot (information processing device)
101 Dialogue robot (information processing device)
200 server 300 interactive system

Claims (5)

An information processing apparatus that presents a predetermined phrase to the user according to the voice uttered by the user,
Obtaining means for obtaining a first phrase associated with the result of recognition of the voice;
Before presenting the first phrase, when a second phrase different from the first phrase is newly acquired by the acquisition unit, it is necessary to present the first phrase to the user. In some cases, an information processing apparatus comprising: a presentation unit that presents the first phrase after presenting the second phrase. The acquisition means further acquires necessity information indicating whether or not it is necessary to present the first phrase to the user;
2. The information processing according to claim 1, wherein the presenting means presents the first phrase after presenting the second phrase when indicating that the necessity information is necessary. apparatus. When the second phrase is newly acquired before presenting the first phrase, when it is necessary to present the first phrase to the user, the first phrase is predetermined. Storage means for storing in the storage unit of
When the first phrase is stored in the predetermined storage unit, the presenting means reads and presents the first phrase from the predetermined storage unit after presenting the second phrase. The information processing apparatus according to claim 1, wherein: A server that controls the information processing apparatus so as to present a predetermined phrase to the user in response to a voice uttered by the user to the information processing apparatus
A specifying means for specifying a phrase associated with the result of recognizing the voice in a predetermined phrase set;
Depending on whether or not it is necessary to present the phrase to the user, generating means for generating necessity information;
A server comprising: transmission means for transmitting the phrase and the necessity information to the information processing apparatus.   A control program for causing a computer to function as the information processing apparatus according to claim 1, wherein the control program causes the computer to function as each of the means.

Images