JP2018159776A - Voice reproduction controller, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make it possible to suppress sound quality deterioration and to reproduce an answer which does not give a user a sense of discomfort, without requiring a large amount of resources.SOLUTION: The provided voice reproduction controller includes a pitch acquiring unit which acquires a pitch based on voice data of an answer corresponding to the voice signal of the inputted question, a target pitch determining unit for determining a first target pitch which maintains a first relation with respect to a pitch of a specific section of the voice signal of the inputted question and falls within a first pitch range determined according to the pitch of the answer, a target pitch changing unit for changing the first target pitch to a second target pitch when the first pitch does not fall within a second pitch range which is narrower than the first pitch range, and a reproduction instructing unit which instructs the voice regeneration unit to reproduce the answer shifting the pitch based on the voice data of the answer to the pitch determined in the target pitch determining unit or the pitch changed in the target pitch changing unit.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an audio reproduction technique.

In recent years, the following has been proposed as an audio reproduction technique. That is, by reproducing a synthesized voice corresponding to the tone and voice quality of a user, a technique for reproducing more human-like voice (see, for example, Patent Document 1), and analyzing a user's voice, the user A technique (for example, refer to Patent Document 2) for diagnosing the psychological state or health state of a child has been proposed.
There has also been proposed a voice dialogue system that recognizes a voice input by a user while synthesizing and playing back a voice of a content specified in a scenario to realize a voice dialogue with the user (for example, Patent Documents). 3).

JP 2003-271194 A Japanese Patent No. 4495907 Japanese Patent No. 4832097

  By the way, a dialogue system is assumed that combines the above-described speech synthesis technology and a speech dialogue system, retrieves data for a question by a user's voice, and synthesizes and reproduces a reply voice by speech synthesis. In this case, if the pitch of the answer voice is far from the pitch of the question voice, there is a problem that the affinity for the question is lacking and an unnatural impression is given to the user. To avoid the occurrence of such problems, prepare multiple types of answer voices with different pitches for the contents of the question, and select one of the answer voices according to the pitch of the question voice. Can be played. However, if there are limited resources, such as the storage capacity of the storage device that stores the answer voice data is not sufficient, multiple answer voices with different pitches cannot be prepared for the contents of the question. . When resource constraints are severe, prepare only one answer voice for each question and synthesize the answer voice with the pitch corresponding to the pitch of the question voice by pitch shift. It is possible to play. However, since the pitch shift is accompanied by sound quality deterioration, there is a problem that the sound quality deterioration of the answer voice becomes more significant as the pitch shift amount increases.

  The present invention has been made in view of such circumstances, and one of the purposes thereof is an answer voice that suppresses deterioration of sound quality and does not give the user an unnatural feeling without requiring a large amount of resources. Is to provide a sound reproduction control device and a sound reproduction control program.

  When considering a man-machine system that plays back voices of answers to questions by users, first of all, what kind of dialogue is made between people, information other than linguistic information, especially the pitch that characterizes the dialogue ( (Frequency) is considered.

  As a dialogue between people, the case where the other person (referred to as b) responds to a question from one person (referred to as a) will be considered. In this case, when a asks a question, not only a but also b to be answered to the question is a pitch (for example, the lowest pitch) in a specific section such as a ending section of the question. Is often left with a strong impression. b, when responding with consent, approval, affirmation, etc., the pitch of the part that characterizes the answer has a specific relationship, for example, the same pitch relationship, with the pitch of the question that remains in the impression Say as follows. A who has heard the answer has a relationship between the pitch that remains in the impression about his question and the pitch of the part that characterizes the answer to the question. It is thought that you will have a good impression.

  In this way, in the dialogue between people, it can be considered that the pitch of the question and the pitch of the answer are not irrelevant and have the above relationship. Based on such considerations, when a dialogue system for reproducing (replying) an answer voice to a user's question was examined, the following configuration was adopted in order to achieve the above-mentioned purpose for the voice reproduction.

  That is, in order to achieve the above object, a sound reproduction control device according to an aspect of the present invention includes a pitch acquisition unit that acquires a pitch based on voice data of an answer corresponding to an input voice signal of a question, A target for determining a first target pitch within a first pitch range that maintains the first relationship with the pitch of a specific section of the input voice signal of the question and is determined according to the pitch of the answer. When the pitch determination unit and the first target pitch do not fall within the second pitch range narrower than the first pitch range, the first target pitch is changed to the second target pitch. The pitch based on the target pitch change section and the answer voice data to be shifted to the pitch determined by the target pitch determination section or the pitch changed by the target pitch change section, and the answer is played back And a playback instruction unit for instructing the audio playback unit to do this.

  Here, as an example of the first relationship, there is a relationship between the same pitch as the pitch of a specific section or a pitch with a different octave. The pitch of octave difference means that the pitch difference between two sounds is an integral multiple of one octave. This is because even if the pitches of two sounds generated at the same time are different, harmony is maintained if the pitch difference is an integral multiple of one octave, and the listener feels less discomfort.

  According to this aspect, it is possible to reproduce the answer to the question issued by the user without being unnatural and preventing deterioration in the quality of hearing. Note that the answer is not limited to a specific answer to the question, but includes an answer (interjection). In addition to human voices, answers include animal calls such as “bow” and “meow”. That is, the answer and the voice here are concepts including not only a voice uttered by a person but also an animal cry.

  The specific section refers to a portion that leaves a strong impression, for example, a section at the end of the question (end section). As a specific example of the pitch of the specific section, there is the lowest pitch of the ending section whose volume is equal to or higher than a predetermined value (that is, a voiced section). The pitch based on the voice data is, for example, the pitch at the characteristic part when the voice data is played back as a standard, and the characteristic part is the part with the highest pitch and volume at the beginning of the word. As well as the average pitch.

  In the above aspect, the target pitch determination unit determines the first target pitch by changing the target pitch in units of the first shift amount until the target pitch falls within the first pitch range, The pitch changing unit is a second shift amount unit that is smaller than the first shift amount unit and is determined according to the pitch of the specific section until it falls within the second pitch range. The second target pitch is determined by changing in units of the second shift amount. The pitch shift amount to the pitch changed by the target pitch changing unit is smaller than the pitch shift amount before the change, and the second target pitch can be set finely. . Note that the first pitch range may be determined in octave units, and the second pitch range may be determined in half octave units.

  Further, as a more preferable configuration, a configuration is conceivable in which the lowest pitch value in the last section when the volume of the input audio signal is equal to or higher than a predetermined value is acquired as the pitch in the specific section. In determining whether the sound volume at this time is equal to or higher than a predetermined value, it may be determined with a hysteresis characteristic.

  The aspect of the present invention can be conceptualized as a program that causes a computer to function as the audio reproduction control device as well as the audio reproduction control device.

It is a block diagram which shows the structure of the audio | voice reproduction | regeneration control apparatus which concerns on embodiment. It is a flowchart which shows operation | movement of an audio | voice reproduction | regeneration control apparatus. It is a flowchart which shows operation | movement of an audio | voice reproduction | regeneration control apparatus. It is a figure which shows the example of a pitch of the question by a user, and the reply by an audio | voice reproduction control apparatus.

Embodiments of the present invention will be described below with reference to the drawings.
(A: Configuration)
FIG. 1 is a diagram showing a configuration of an audio reproduction control apparatus 10 according to the embodiment of the present invention.
The voice reproduction control device 10 is a device that is incorporated in a stuffed animal, for example, and synthesizes and outputs an answer such as a question when a user asks the stuffed animal. The audio reproduction control device 10 includes a CPU (Central Processing Unit), an audio input unit 102, and a speaker 142. When the CPU executes an application program installed in advance, a plurality of functional blocks are as follows. Built in. Specifically, in the audio reproduction control device 10, an audio feature amount acquisition unit 106, an answer selection unit 110, a pitch acquisition unit 112, a reproduction instruction unit 114, and an answer reproduction unit 116 are constructed.

  Although not particularly illustrated, the sound reproduction control device 10 also has a display unit, an operation input unit, and the like, so that the user can check the status of the device or input various operations to the device. Or various settings. The audio reproduction control device 10 is not limited to a toy such as a stuffed toy, and may be a so-called pet robot, a terminal device such as a mobile phone, a tablet personal computer, or the like.

  Although not described in detail, the sound input unit 102 includes a microphone that converts sound into an electrical signal and an A / D converter that converts the converted sound signal into a digital signal.

  The voice feature quantity acquisition unit 106 analyzes the voice signal converted into the digital signal, classifies the voice signal into the utterance period and the non-utterance period, and among the utterance period, the lowest sound of the specific period in the voiced period High is detected, and data indicating the pitch is supplied to the answer selection unit 110 and the reproduction instruction unit 114. Here, the utterance section refers to, for example, a section in which the volume of the audio signal is equal to or higher than a threshold, and conversely, the non-utterance section refers to a section in which the volume of the audio signal is less than the threshold. The voiced section is a section in which the pitch (pitch) of the voice signal can be detected in the utterance section. The section in which the pitch can be detected means that there is a periodic part in the audio signal and that part can be detected. Here, the specific section is the last section of the voiced section. Further, the end section is a section of a predetermined time (for example, 180 msec) that is temporally forward from the end of the voiced section.

  The answer library 124 stores a plurality of voice data of answers to questions by the user in advance. This voice data is a recording of the voice of a model person. For example, “Yes”, “No”, “Yes”, “Yes”, “Fun”, “I see” For example, it is a companion. The voice data of the answer is in a format such as wav or mp3, for example, and a pitch for each waveform sample (or every waveform cycle) when reproduced as a standard and an average pitch obtained by averaging them are obtained in advance. Data indicating the average pitch (pitch based on the answer) is stored in the answer library 124 in association with the voice data. The standard reproduction here means that the audio data is reproduced under the same conditions as the recording conditions (sampling frequency).

  When the voice feature quantity acquisition unit 106 outputs data indicating the minimum pitch of a specific section, the answer selection unit 110 (answer acquisition unit) selects one of the answer data for the voice from the answer library 124. Then, the voice data of the selected answer is read and output together with the data indicating the associated average pitch. For example, the answer selection unit 110 may select a single piece of voice data from among a plurality of pieces of voice data. For example, the answer selection unit 110 may be random, and the average pitch may be lower than the lowest pitch of the specific section in question. The closest audio data may be selected.

  In the present embodiment, the meaning content of the user's question is not taken into account for the selected answer. However, the voice reproduction control device 10 returns a reconciliation as an answer to the question issued by the user. This is sufficient for a device. On the other hand, a language analysis unit 108 is provided as shown by a broken line in the figure, the language analysis unit 108 analyzes the meaning content of the question specified by the audio signal, and the answer selection unit 110 transmits the database via a database or the like. It is good also as a structure which produces the answer with respect to the said question.

  The pitch acquisition unit 112 extracts data indicating the average pitch of answers from the data read by the answer selection unit 110 and supplies the extracted data to the reproduction instruction unit 114.

  The reproduction instructing unit 114 determines the answer voice data based on the difference between the minimum pitch of the specific section in the voice signal output from the voice feature value acquisition unit 106 and the average pitch of the answer output from the pitch acquisition unit 112. Is determined, and a pitch shift amount for shifting the average pitch of the answer to the target pitch is determined. As shown in FIG. 1, the reproduction instruction unit 114 includes a target pitch determination unit 114a and a target pitch change unit 114b.

  The target pitch determination unit 114a is a target pitch that maintains a predetermined first relationship with respect to the minimum pitch of a specific section of the questioned voice, and is determined according to the average pitch of the answers. A first target pitch that falls within the pitch range of 1 is determined. Further, the target pitch determination unit 114a calculates a pitch shift amount for shifting the average pitch of the answer to the first target pitch. In the present embodiment, the first relationship is a relationship of pitches equal to the minimum pitch of a specific section of the questioned voice or a relationship of pitches having a difference in octave units. Of course, if the pitches of two sounds generated at the same time are the same, if the pitch difference is an integer multiple of one octave, harmony is maintained and the listener feels less uncomfortable. is there. The first pitch range is a range of one octave centered on the average pitch of answers, that is, a range of average pitch -600 cents to average pitch +600 cents.

  The target pitch changing unit 114b determines that the first target pitch determined by the target pitch determining unit 114a is not within the second pitch range narrower than the first pitch range. The target pitch of 1 is changed to the second target pitch. More specifically, the target pitch changing unit 114b sets the first target pitch to the lowest pitch of the specific section when the first target pitch is not within the second pitch range. On the other hand, the target pitch that maintains the second relationship different from the first relationship is changed to a second target pitch that falls within the second pitch range, and is calculated by the target pitch determination unit 114a. The pitch shift amount thus corrected is corrected to a pitch shift amount for shifting to the second target pitch. Here, the second relationship refers to a relationship between pitches having a high affinity for the minimum pitch in the specific section, such as a relationship of a consonant such as “So” to “Do”. This is because if the second relationship is maintained, the discomfort given to the listener is small, although not as much as when the first relationship is maintained. The second pitch range is a range of a half octave centered on the average pitch of answers, that is, a range of average pitch -300 cents to average pitch +300 cents.

  The reproduction instruction unit 114 reproduces the answer by shifting the pitch determined by the target pitch determination unit 114a or the pitch shift amount for changing to the pitch changed by the target pitch change unit 114b. To the answer reproducing unit 116. The answer reproducing unit 116 reproduces (synthesizes) the answer voice data read from the answer library 124 by shifting the voice data by the pitch shift amount instructed by the reproduction instruction unit 114. Note that the audio signal whose pitch is shifted is converted into an analog signal by a D / A converter (not shown), and then acoustically converted by the speaker 142 and output. The data associated with the pitch of the answer, that is, the data stored in the answer library 124 and used for determining the pitch shift amount by the playback instruction unit 114 is data other than the data indicating the average pitch. There may be. For example, an intermediate value of the pitch may be used, or an average pitch of a predetermined section of the audio data may be used.

(B: Operation)
Next, the operation of the audio reproduction control device 10 will be described.
FIG. 2 is a flowchart showing the processing operation in the audio reproduction control device 10.
First, when the user asks a question about the stuffed animal to which the voice reproduction control device 10 is applied, the processing shown in this flowchart is started. Here, for convenience, the case where the pitch of the answer voice data is lower than the pitch of the voice (question) of the user will be described as an example.

  First, in step Sa <b> 11, the audio signal converted by the audio input unit 102 is supplied to the audio feature amount acquisition unit 106. Next, in step Sa12, the voice feature quantity acquisition unit 106 performs an analysis process on the voice signal from the voice input unit 102, that is, a process of detecting the pitch of a question issued by the user. In step Sa13, the answer reproducing unit 116 determines whether or not the answer is being reproduced.

  If the answer is not being played back (if the determination result in step Sa13 is “No”), the voice feature acquisition unit 106 determines whether or not the question (speech) of the voice signal from the voice input unit 102 has ended. A determination is made (step Sa14). Note that whether or not the inquiry has ended is specifically determined based on, for example, whether or not a state in which the volume of the audio signal has become less than a predetermined threshold has continued for a predetermined time. If the inquiry has not ended (if the determination result in step Sa14 is “No”), the processing procedure returns to step Sa11, whereby the audio feature quantity acquisition unit 106 receives the audio signal from the audio input unit 102. Continue the analysis process.

  If the inquiry has been completed (if the determination result in step Sa14 is “Yes”), the playback instruction unit 114 sets the pitch shift amount when the voice data of the answer selected by the answer selection unit 110 is played back. The determination is made as described later (step Sa15). Then, the reproduction instruction unit 114 notifies the answer reproduction unit 116 of the determined pitch shift amount, and instructs the reproduction of the answer voice data selected by the answer selection unit 110 (step Sa16). In accordance with this instruction, the answer reproducing unit 116 reproduces the sound data by shifting the audio data by the pitch shift amount notified from the reproduction instruction unit 114 (step Sa17).

  In step Sa13, when the answer reproducing unit 116 determines that the answer is being reproduced (when the determination result in step Sa13 is “Yes”), the answer is being reproduced according to a certain question. For example, when the following question is asked by the user. In this case, the processing procedure does not return the path of steps Sa14 and Sa11, and proceeds to step Sa17, so that the reproduction of the answer has priority.

  FIG. 3 is a flowchart showing details of the process of step Sa15 in FIG. 2, that is, the process of determining the pitch shift amount of the answer voice data. The premise for executing this process is that the answer reproducing unit 116 is not reproducing the answer (the determination result of step Sa13 is “No”), and the user has finished inputting the question ( The determination result of step Sa14 is “Yes”).

  First, in step Sb11, the reproduction instruction unit 114 acquires data indicating the minimum pitch of the specific section in question from the audio feature amount acquisition unit 106.

  On the other hand, the answer selection unit 110 selects voice data of the answer to the question by the user from the answer library 124, the voice data of the selected answer, and data indicating the average pitch associated with the voice data Is read. Among these, the pitch acquisition unit 112 supplies data indicating the average pitch among the read data to the reproduction instruction unit 114. Thereby, the reproduction instruction unit 114 acquires data indicating the average pitch of the answer selected by the answer selection unit 110 (step Sb12).

  Next, the target pitch determination unit 114a provisionally determines the lowest pitch of the specific section in question as the first target pitch described above (step Sb13).

  Subsequently, the target pitch determining unit 114a determines the first target pitch temporarily determined from the average pitch of the answers selected by the answer selecting unit 110 (after step Sb16 and after change by steps Sb16 and Sb18 described later). The pitch shift amount up to (including the first target pitch) is calculated (step Sb14). The target pitch determining unit 114a determines whether or not the temporarily determined first target pitch is lower than the lower limit threshold L1 of the first pitch range described above (step Sb15). The lower limit threshold L1 is a threshold for how much pitch shift is allowed with respect to the average pitch of the answer, and in this embodiment, the average pitch of the answer is −600 cents.

  If the tentatively determined first target pitch is lower than the lower limit threshold L1 (if the determination result in step Sb15 is “Yes”), the target pitch determination unit 114a sets the first target pitch to one octave. (1200 cents) is raised and provisionally determined again (step Sb16). After this, the processing procedure returns to step Sb14, the pitch shift amount is calculated again, and the determination in step Sb15 is executed.

  On the other hand, if the tentatively determined first target pitch is not lower than the lower limit threshold L1 (if the determination result in step Sb15 is “No”), the target pitch determining unit 114a performs the tentatively determined first target pitch. It is determined whether or not the target pitch is higher than the upper limit threshold H1 of the first pitch range (step Sb17). The upper limit threshold H1 is a threshold for how much pitch shift is allowed with respect to the average pitch of the answer, and in this embodiment, the average pitch of the answer is +600 cents. If the tentatively determined first target pitch is higher than the upper limit threshold H1 (if the determination result in step Sb17 is “Yes”), the target pitch determination unit 114a sets the first target pitch to one octave. It is lowered and provisionally determined again (step Sb18). After this, the processing procedure returns to step Sb14, the pitch shift amount is calculated again, and the determinations of steps Sb15 and Sb17 are executed.

  If the tentatively determined first target pitch is not higher than the upper limit threshold value H1 (if the determination result in step Sb17 is “No”), the tentatively determined first target pitch is the first pitch range. Means that The target pitch determination unit 114a sets the first target pitch at the time when the determination result of step Sb17 is “No” as the main determination, and shifts the processing procedure to step Sb19.

  In step Sb19, the target pitch changing unit 114b temporarily determines the pitch after the shift corresponding to the pitch shift amount as the second target pitch, and the temporarily determined second target pitch is the second target pitch described above. It is determined whether the pitch range is lower than the lower limit threshold L2. In step Sb19 executed immediately after the determination result in step Sb17 becomes “No”, the first target pitch determined by the target pitch determination unit 114a is provisionally determined as the second target pitch. . The lower limit threshold L2 is a threshold that more strictly indicates how much pitch shift is allowed to a lower pitch than the average pitch of answers than the lower limit threshold L1 described above, and in this embodiment, the average pitch of answers High-300 cents.

  If the determination result in step Sb19 is “Yes”, the target pitch changing unit 114a temporarily raises the second target pitch by a predetermined amount and recalculates the pitch shift amount (step Sb20). The pitch raising amount in step Sb20 is set smaller than the raising amount (1 octave) in step Sb16. This is because the second pitch range is narrower than the first pitch range as described above. The pitch increase amount in step Sb20 may be set according to the width of the second pitch range and the second relationship described above. Specifically, when the pitch of the question is “do”, the pitch of the answer is “seo” (not the relationship of the same pitch in different octaves, but the affinity within one octave It is conceivable that the pitch increase amount in step Sb20 is set to 700 cents so that the pitch is in a high relationship (for example, a consonant relationship), and this aspect is adopted in this embodiment. Has been. After execution of the process of step Sb20, the process procedure returns to step Sb19, and the determination of step Sb19 is executed again.

  On the other hand, if the tentatively determined second target pitch is not lower than the lower limit threshold L2 (if the determination result in step Sb19 is “No”), the target pitch changing unit 114b performs the tentatively determined second target pitch. It is determined whether or not the target pitch is higher than an upper limit threshold H2 of the second pitch range (step Sb21). This upper limit threshold value H2 is a threshold value that more strictly indicates the upper limit threshold value H1 that indicates how much pitch shift is allowed with respect to the average pitch of the answer. In the present embodiment, Average pitch + 300 cents. If the pitch after the shift is higher than the upper limit threshold value H2 (if the determination result in step Sb21 is “Yes”), the target pitch changing unit 114b lowers the second target pitch by a predetermined amount and again temporarily Then, the pitch shift amount is recalculated (step Sb22). The pitch reduction amount in step Sb22 is also set smaller than the reduction amount (1 octave) in step Sb18. Similar to the pitch increase amount in step Sb20, the pitch decrease amount in step Sb22 may be set according to the width of the second pitch range and the second relationship described above. In this embodiment, when the pitch of the question is “do”, the pitch of the answer is “seo” (the relationship within the octave is not the relationship of the same pitches with different octaves). The pitch reduction amount in step Sb22 is set to 500 cents so that the pitch is high (for example, a pitch that is in the relationship of consonance). After execution of the process of step Sb22, the process procedure returns to step Sb19, and the determinations of steps Sb19 and Sb21 are executed again.

If the tentatively determined second target pitch is not higher than the upper limit threshold H2 (if the determination result in step Sb21 is “No”), the tentatively determined second target pitch is in the second pitch range. Means it is within. The target pitch changing unit 114b finally determines the second target pitch at the time when the determination result in step Sb21 is “No” as the pitch of the answer to be output, and determines the average pitch of the answer. The answer reproduction unit 116 is notified of the pitch shift amount at that time as the pitch shift amount for shifting to the pitch (step Sb23).
The above is the operation of the pitch determination unit 114.

  FIG. 4 exemplifies the relationship between the voice of the question input by the user and the voice of the answer played (synthesized) by the voice playback control device 10 with the pitch on the vertical axis and the time on the horizontal axis. FIG. In this figure, the solid line indicated by the reference symbol T1 simply indicates a change in the pitch of the voice “Ane” by the user as a straight line. The symbol P1 indicates the lowest pitch of the specific section in the question T1, specifically, the pitch of “do”. Further, in the figure, a solid line indicated by reference symbol A1 is a diagram simply showing a change in pitch when the voice data of the answer “Yes” selected for the question T1 is reproduced as a standard. Indicates the average pitch.

  When the question A1 is reproduced without shifting the pitch of the answer A1, an unnatural feeling is easily received. For this reason, in this embodiment, first, the lowest pitch P1 of the specific section (end of word), which is a characteristic and impressive part of the question T1, is provisionally determined as the pitch when the answer A1 is reproduced (FIG. 3). : Step Sb13), a pitch shift amount for shifting the pitch P0 of the answer A1 to the temporarily determined pitch P1 is calculated (step Sb14).

  As shown in FIG. 4, since the temporarily determined pitch P1 exceeds the upper limit threshold value H1 of the first pitch range, the determination result in step Sb15 is “No”, and the determination result in step Sb17 is “ Yes ". As a result, the process of step Sb18 is executed, and the pitch P2 that is one octave lower than the pitch P1 is provisionally determined as the pitch when the answer A1 is reproduced, and the pitch shift amount is recalculated (FIG. 4). : Step Sb14).

  As shown in FIG. 4, the pitch P2 is higher than the lower limit threshold L1 of the first pitch range and lower than the upper limit threshold H1. As a result, the determination result of step Sb15 is “No”, the determination result of step Sb17 is also “No”, and the process of step Sb19 is executed. As shown in FIG. 4, the pitch P2 exceeds the lower limit threshold L2 of the second pitch range, and also exceeds the upper limit threshold H2. Therefore, the determination result of step Sb19 is “No”, the determination process of step Sb21 is executed, and the determination result of step Sb21 is “Yes”. As a result, the process of step Sb22 is executed, and the pitch P3 that is lowered by a predetermined amount (500 cents in this embodiment) from the pitch P2 is provisionally determined as the pitch when the answer A1 is reproduced. The pitch shift amount for shifting P0 to the pitch P3 is recalculated.

  As shown in FIG. 4, the pitch P3 is higher than the lower limit threshold L2 of the second pitch range and lower than the upper limit threshold H2. As a result, the determination result in step Sb19 is “No”, the determination result in step Sb21 is also “No”, and the pitch P3 is determined as the pitch when the answer A1 is reproduced (step Sb23). .

  As a result of the above operation, the answer A1 in which the pitch is shifted to the pitch P3 is reproduced as a response to the question T1. Here, it should be noted that the pitch shift amount D1 when shifting the pitch of the answer A1 to the pitch P3 is greater than the pitch shift amount D2 when shifting the pitch of the answer A1 to the pitch P2. Is also small. The pitch P3 of the answer A1 reproduced according to the present embodiment is not in an octave difference from the lowest pitch P1 (“do”) in the specific section of the question T1, but has a high affinity with this ( "So"). In addition, the pitch shift amount D1 when shifting the pitch of the answer A1 to the pitch P3 is smaller than the pitch shift amount D2, and deterioration in sound quality due to the pitch shift can be reduced.

  As described above, according to the present embodiment, the answer to the question issued by the user can be synthesized (reproduced) without being unnatural and preventing deterioration of the audible quality. Further, in the present embodiment, it is not necessary to prepare voice data of a plurality of answers with different pitches for one question, and it can be realized with few resources.

(C: Deformation and application examples)
The present invention is not limited to the above-described embodiments, and various applications and modifications as described below are possible, for example. In addition, one or more arbitrarily selected aspects of application / deformation described below can be appropriately combined.

  As for the voice data of answers stored in the answer library 124, answers of the same content may be stored across a plurality of people such as persons A, B, C,. For the persons A, B, C,..., For example, celebrities, talents, singers, etc., voice data is stored in a library for each person. In the case of creating a library in this way, answer voice data may be stored in the answer library 124 via a medium such as a memory card, or the voice playback control device 10 may be provided with a network connection function so as to be sent from a specific server. The voice data of the answer may be downloaded and stored in the answer library 124. When the voice data of the answer is obtained from the memory card or the server, it may be free or paid.

  On the other hand, it can be configured so that the user can select the person who wants to answer the model as a model by the operation input unit, etc., or for each condition (day, week, month, etc.) It is good also as a structure determined at random.

  As for the voice data of answers, a library of voices recorded by the user himself / herself, the user's family and acquaintances (or converted into data by a separate device) via the microphone of the voice input unit 102 is a library. May be used. When the answer is made with the voice of a person close to the person like this, it is possible to obtain a feeling as if the person is interacting with the person when the question is made.

  In addition, the answer may be a call from an animal (dog, cat, etc.), or may be configured such that a dog breed or the like can be selected as appropriate. In this way, by using an answer as an animal call, it is possible to obtain a kind of healing effect as if it were interacting with the animal.

  The pitch acquisition unit 112 analyzes the voice data of the answer determined by the answer selection unit 110, acquires an average pitch when the voice data is played back as a standard, and instructs to play back data indicating the pitch It is good also as a structure supplied to the part 114. FIG. According to this configuration, it is not necessary to associate the data indicating the pitch with the voice data of the answer and store it in the answer library 124 in advance. In the above embodiment, the target pitch determination unit 114a and the target pitch change unit 114b are included in the playback instruction unit 114, but the target pitch determination unit 114a and the target pitch change unit 114b are used as the playback instruction unit 114. You may provide separately. In addition, the voice playback control device 10 of the above embodiment includes the answer playback unit 116, but the answer playback unit 116 may be a voice synthesizer separate from the voice playback control device 10.

  In the embodiment, the case where the pitch of the voice data of the answer is lower than the pitch of the question asked by the user has been described as an example. It is also applicable when the pitch of is high.

DESCRIPTION OF SYMBOLS 102 ... Voice input part, 106 ... Voice feature-value acquisition part, 110 ... Reply selection part, 112 ... Pitch acquisition part, 114 ... Reproduction | regeneration instruction | indication part, 114a ... Target pitch determination part, 114b ... Target pitch change part, 116 ... answer playback unit, 124 ... answer library.

Claims (5)

A pitch acquisition unit that acquires a pitch based on voice data of an answer corresponding to the voice signal of the input question;
The target pitch that maintains a predetermined first relationship with the pitch of a specific section of the input question voice signal and that is determined according to the pitch based on the answer voice data. A target pitch determining unit that determines a first target pitch that falls within a pitch range of one;
A target sound for changing the first target pitch to a second target pitch when the first target pitch does not fall within a second pitch range narrower than the first pitch range. High change part,
The pitch based on the voice data of the answer is shifted by the pitch shift amount for changing to the pitch determined by the target pitch determining unit or the pitch changed by the target pitch changing unit. A playback instruction unit for instructing the answer playback unit to play back the answer;
An audio reproduction control device comprising: The target pitch determining unit determines the first target pitch by changing the target pitch in units of a first shift amount until it falls within the first pitch range.
The target pitch changing unit sets the target pitch to a second shift amount unit smaller than the first shift amount unit until the pitch falls within the second pitch range. The audio reproduction control device according to claim 1, wherein the second target pitch is determined by changing in units of second shift amounts determined accordingly.   The sound reproduction control device according to claim 1 or 2, wherein the first pitch range is defined in octave units, and the second pitch range is defined in half octave units. The voice reproduction control according to any one of claims 1 to 3, wherein a minimum pitch value in a section in which a volume of the input voice signal is equal to or higher than a predetermined value is set as a pitch in the specific section. apparatus. Computer
A pitch acquisition unit that acquires a pitch based on voice data of an answer corresponding to the voice signal of the input question;
The target pitch that maintains a predetermined first relationship with the pitch of a specific section of the input question voice signal and that is determined according to the pitch based on the answer voice data. A target pitch determining unit that determines a first target pitch that falls within a pitch range of one;
A target sound for changing the first target pitch to a second target pitch when the first target pitch does not fall within a second pitch range narrower than the first pitch range. High change part,
The pitch based on the voice data of the answer is shifted by the pitch shift amount for changing to the pitch determined by the target pitch determining unit or the pitch changed by the target pitch changing unit. A playback instruction unit for instructing the answer playback unit to play back the answer;
A program characterized by functioning as

Images