JP2018146933A - Evaluation device, evaluation method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform accurate evaluation on a plurality of continuous determined pitches.SOLUTION: A program according to one embodiment of the present invention causes a computer to acquire an input sound, acquire pitch data in time series on the basis of the input sound, accumulate pitch data in a predetermined section from among the acquired pitch data for every predetermined time and sequentially update the pitch data, compare the pitch data in the predetermined section with predetermined reference sound information, calculate an evaluation value of the pitch data in the predetermined section on the basis of a result of comparison, determine whether to switch the reference sound information on the basis of the input sound in a period corresponding to the pitch data in the predetermined section, when switching the reference sound information, compare the pitch data in the predetermined section with predetermined reference sound information different from the predetermined reference sound information, and calculate the evaluation value of the pitch data in the predetermined section on the basis of a result of comparison.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a technique for evaluating sound.

  An evaluation device that evaluates performance sounds and singing voices has a function of analyzing and evaluating input sounds. Various methods are used for sound evaluation. For example, Patent Literature 1 discloses a karaoke device that processes reference data so that it is close to an actual song, compares the singing voice with the processed reference data, and scores the singing voice. Patent Document 2 discloses an evaluation device that compares the pitch of an input sound with a reference pitch in a predetermined evaluation section and evaluates the input sound based on the comparison result.

JP-A-2005-107328 Japanese Patent Laid-Open No. 2006-173562

  However, with the techniques of Patent Document 1 and Patent Document 2, although it is possible to collectively evaluate the input sound input during a predetermined period, it is not possible to evaluate a predetermined predetermined pitch.

  One of the objects of the present invention is to perform highly accurate evaluation on a plurality of predetermined continuous pitches in various input sounds such as performance sounds and singing voices.

  According to an embodiment of the present invention, a computer acquires an input sound, acquires pitch data in time series based on the input sound, and among the acquired pitch data, pitch data of a predetermined section is determined for a predetermined time. Each time it is accumulated and updated sequentially, the pitch data of the predetermined section is compared with predetermined reference sound information, an evaluation value of the pitch data of the predetermined section is calculated based on the comparison result, and the pitch data of the predetermined section is calculated. When determining whether to switch the reference sound information based on an input sound in a corresponding period and switching the reference sound information, a predetermined reference sound different from the pitch data of the predetermined section and the predetermined reference sound information Comparing the information and calculating the evaluation value of the pitch data of the predetermined section based on the comparison result Program is provided.

  Determining whether to switch the reference sound information may be determining whether to switch the reference sound information based on pitch data of the predetermined section.

  Determining whether to switch the reference sound information may be determining whether to switch the reference sound information based on volume data of the predetermined section.

  The computer may be configured to identify a section to be compared with the predetermined reference sound information in the predetermined section based on the input sound of the predetermined section.

  The calculation of the evaluation value of the pitch data of the predetermined section specifies the representative pitch in the predetermined section, and compares the representative pitch with the pitch of the predetermined reference sound information to evaluate the pitch data of the predetermined section. It may be to calculate a value.

  The evaluation value of the pitch data in the predetermined section is calculated based on whether the pitch data in the predetermined section is within a predetermined pitch range based on a pitch corresponding to the predetermined reference sound information. The evaluation value of the pitch data of the predetermined section may be calculated.

  The evaluation value of the pitch data in the predetermined section may be calculated based on the frequency distribution of the pitch data in the predetermined section.

  According to an embodiment of the present invention, an input sound acquisition unit that acquires an input sound, and pitch data is acquired in time series based on the input sound, and pitch data in a predetermined section is acquired from the acquired pitch data. A feature amount acquisition unit that accumulates and sequentially updates each time, and an evaluation that compares the pitch data of the predetermined section with predetermined reference sound information and calculates an evaluation value of the pitch data of the predetermined section based on the comparison result And a switching determination unit that determines whether to switch the reference sound information based on an input sound in a period corresponding to the pitch data of the predetermined section, and the switching determination unit switches the reference sound information When the determination unit determines that the predetermined reference sound information different from the pitch data of the predetermined section and the predetermined reference sound information, Comparison based on the comparison result calculates the evaluation value of the pitch data of the predetermined section, the evaluation device is provided.

  According to an embodiment of the present invention, an input sound is acquired, pitch data is acquired in time series based on the input sound, and pitch data of a predetermined section is accumulated at predetermined intervals among the acquired pitch data. And sequentially updating, comparing the pitch data of the predetermined section with predetermined reference sound information, calculating an evaluation value of the pitch data of the predetermined section based on the comparison result, and a period corresponding to the pitch data of the predetermined section It is determined whether to switch the reference sound information based on the input sound, and when the reference sound information is switched, the pitch data of the predetermined section and predetermined reference sound information different from the predetermined reference sound information are obtained. Comparing and calculating an evaluation value of the pitch data of the predetermined section based on the comparison result is provided.

  According to an embodiment of the present invention, it is possible to perform a highly accurate evaluation on a plurality of continuous pitches determined for various input sounds such as performance sounds and singing sounds.

It is a block diagram which shows the structure of the evaluation apparatus in one Embodiment of this invention. It is a block diagram which shows the structure of the evaluation function in one Embodiment of this invention. It is the flowchart which showed the flow of the evaluation method of the input sound performed by the structure of the evaluation function in one Embodiment of this invention. It is a figure for demonstrating an example of the concept of the effective area identification in a predetermined area. (A) It is a figure which shows the frequency of the pitch in a predetermined area. (B) It is a figure which shows the fluctuation | variation of the time series of the pitch data in a predetermined area. It is a figure for demonstrating an example of the concept of the determination of the presence or absence of switching of the input sound in a predetermined area. It is a figure for demonstrating another example of the concept of the determination of the presence or absence of switching of the input sound in a predetermined area. It is a figure for demonstrating an example of the concept which calculates the evaluation value of the pitch data in this predetermined area. It is a figure for demonstrating another example of the concept which calculates the evaluation value of the pitch data in this predetermined area. It is a figure for demonstrating an example of the concept which calculates the evaluation value of the pitch data in this predetermined area. It is a figure for demonstrating another example of the concept which calculates the evaluation value of the pitch data in this predetermined area. It is a figure for demonstrating an example of the concept of the effective area identification in a predetermined area.

  Hereinafter, an evaluation apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings. The following embodiments are examples of the embodiments of the present invention, and the present invention is not limited to these embodiments.

  An evaluation apparatus according to an embodiment of the present invention will be described in detail with reference to the drawings. An evaluation apparatus according to an embodiment of the present invention is an apparatus having a function of evaluating various input sounds such as an input sound of a musical instrument played by a performer and a singing voice of a singer. This evaluation device acquires pitch data of input sound in time series, accumulates pitch data of a predetermined section in the acquired pitch data, and sequentially updates the pitch data of the predetermined section every predetermined time. Then, the evaluation device compares the pitch data with predetermined reference sound information, and calculates an evaluation value of the pitch data of the predetermined section based on the comparison result. At this time, it corresponds to the pitch data of the predetermined section. When determining whether to switch the reference sound information to be compared based on the input sound of the period to be compared, and switching the reference sound information, the predetermined reference sound information different from the predetermined reference sound information and the The pitch data of a predetermined section is compared, and the pitch data of the predetermined section is evaluated based on the comparison result.

[Hardware]
FIG. 1 is a block diagram showing a configuration of an evaluation apparatus 10 according to an embodiment of the present invention. The evaluation device 10 includes a control unit 11, a storage unit 13, a display unit 15, and a signal processing unit 17. A sound input unit (for example, a microphone) 19 is connected to the signal processing unit 17. Further, the evaluation device 10 may optionally include an operation unit 21 and a communication unit 23. These components are connected to each other via a bus 25. In the following, an example in which the evaluation device 10 according to an embodiment of the present invention evaluates an input sound of a musical instrument played by a performer will be described.

  The control unit 11 includes an arithmetic processing circuit such as a CPU. The control unit 11 causes the CPU to execute the control program 13a stored in the storage unit 13 and implements various functions in the evaluation apparatus 10. The realized functions include a function for evaluating the input sound of a musical instrument performed by the performer. The control unit 11 also includes an input sound storage unit 11a that stores input sound in a FIFO (First In, First Out). The input sound storage unit 11a is a FIFO memory.

  The storage unit 13 is a storage device such as a nonvolatile memory or a hard disk. The storage unit 13 stores a control program 13a for realizing the evaluation function. The control program 13a may be provided in a state stored in a computer-readable recording medium such as a magnetic recording medium, an optical recording medium, a magneto-optical recording medium, or a semiconductor memory. In this case, the evaluation device 10 only needs to include a device that reads the recording medium. The control program 13a may be downloaded via a network such as the Internet. The storage unit 13 stores reference sound information 13b. The reference sound information 13b is reference sound information used as a reference for evaluation of the input sound, and includes pitch information of the reference sound.

  The display unit 15 is a display device such as a liquid crystal display or an organic EL display, and displays a screen representing an evaluation of the input sound by the performer based on control by the control unit 11.

  The signal processing unit 17 includes an A / D converter and the like. The input sound of the musical instrument played by the performer is converted into an electrical signal by the sound input unit 19 and input to the signal processing unit 17, and A / D converted by the signal processing unit 21 and output to the control unit 11.

  The operation unit 21 is a device such as operation buttons, a keyboard, and a mouse provided on the operation panel and the remote controller, and outputs a signal corresponding to the input operation to the control unit 11. The user of the evaluation device 10 may specify a reference sound corresponding to the input sound to be evaluated by the operation unit 21. Based on the control of the control unit 11, the communication unit 23 is connected to a communication line such as the Internet or a LAN, and transmits / receives information to / from an external device such as a server. Note that the function of the storage unit 13 may be realized by an external device capable of communicating in the communication unit 23.

[Evaluation function]
An evaluation function realized when the control unit 11 of the evaluation apparatus 10 executes the control program 13a stored in the storage unit 13 will be described. A part or all of the configuration for realizing the evaluation function described below may be realized by hardware.

  FIG. 2 is a block diagram showing the configuration of the evaluation function 100 in one embodiment of the present invention. Referring to FIG. 2, the evaluation function 100 includes an input sound acquisition unit 103, a feature amount acquisition unit 105, an effective section identification unit 107, a switching determination unit 109, and an evaluation unit 111. In the evaluation function 100, the valid section specifying unit 107 may be omitted.

  FIG. 3 is a flowchart showing the flow of the input sound evaluation method executed by the configuration of the evaluation function 100 according to the embodiment of the present invention shown in FIG. With reference to FIG. 2 and FIG. 3, an example of the input sound evaluation method executed by the configuration of the evaluation function 100 in one embodiment of the present invention will be briefly described.

  First, an input sound is acquired by the input sound acquisition unit 103 (S1). Next, the feature amount acquisition unit 105 acquires a feature amount based on the acquired input sound (S2) and stores it in the FIFO. Based on the feature value of the predetermined section stored in the FIFO, the effective section in the predetermined section is specified by the effective section specifying unit 107 (S3). That is, it is determined whether or not an effective section exists in the predetermined section. When the valid section is not specified, the process proceeds to S8 described later. However, when the valid section specifying unit 107 is omitted in the evaluation function 100, S3 is omitted, and after the feature amount is acquired in S2, the process proceeds to S4 described later. When the valid section is specified (S3) or after the feature value is acquired (S2), the switching determination unit 109 determines whether to switch the reference sound information (S4). When switching the reference sound information, the evaluation unit 111 switches the reference sound information to another reference sound information (S5), compares the pitch data of the predetermined section with the switched reference sound information, and determines the pitch of the predetermined section. An evaluation value for the data is calculated (S6). When the reference sound information is not switched, the evaluation unit 111 calculates the evaluation value for the pitch data of the predetermined section by comparing the pitch data of the predetermined section with the current reference sound information without switching the reference sound information ( S6). When the evaluation values are calculated, it is determined whether or not the evaluation for all input sounds has been completed (S7). If evaluation for all input sounds has not been completed, the process proceeds to identification of an effective section in the next predetermined section (S3). When the valid section specifying unit 107 is omitted, the process proceeds to determination (S4) as to whether or not to switch reference sound information to be compared for evaluation of pitch data in the next predetermined section. After a certain period of time has elapsed since the acquisition of the first input sound, the evaluation for all input sounds may be completed, or may be ended by an operation by the user of the evaluation device 10.

  Hereinafter, the input sound acquisition unit 103, the feature amount acquisition unit 105, the valid section identification unit 107, the switching determination unit 109, and the evaluation unit 111 in the evaluation function 100 according to an embodiment of the present invention will be described.

  The input sound acquisition unit 103 acquires the input sound input from the sound input unit 19. Here, the input sound acquired by the input sound acquisition unit 103 is digitized by the signal processing unit 17. The input sound acquisition unit 103 acquires the input sound directly from the signal processing unit 17, but may acquire the input sound once stored in the storage unit 13. The input sound acquisition unit 103 is not limited to acquiring the input sound to the sound input unit 19, and may acquire the digitized input sound to the external device via the network. In the present embodiment, the input sound acquisition unit 103 sequentially outputs input sounds that are sequentially input to the feature amount acquisition unit 105.

  The feature amount acquisition unit 105 detects the feature amount of the input sound from the input sound acquired by the input sound acquisition unit 103 in time series. The feature amount of the input sound includes the pitch of the input sound. The feature amount acquisition unit 105 detects a zero cross when the waveform of the audio signal indicated by the input sound changes from negative to positive for each frame (data sample divided by a predetermined period), and measures the time interval of the zero cross. Thus, the pitch (frequency) of the input sound is specified. At this time, a high-frequency component that becomes a noise component may be cut from the audio signal by a low-pass filter, or a DC component may be cut by a high-pass filter. The feature quantity acquisition unit 105 may specify the pitch of the input sound from a spectrum obtained by performing FFT (Fast Fourier Transform) on the input sound. Data of a predetermined section of the pitch data indicating the pitch detected in this way is sequentially stored in the FIFO. The buffer size in the FIFO, that is, the length of the predetermined section is not particularly limited and is arbitrary, but may be, for example, the latest to several hundred msec.

  The feature amount acquisition unit 105 may detect the volume of the input sound as the feature amount of the input sound. The feature amount acquisition unit 105 detects the volume of the input sound in time series from the input sound acquired by the input sound acquisition unit 103. The feature amount acquisition unit 105 detects a temporal change (volume waveform) of the volume of the input sound based on the input sound. The feature amount acquisition unit 105 detects the volume based on the amplitude of the audio signal indicated by the input sound. The feature amount acquisition unit 105 associates the detected volume data indicating the volume of the input sound with the pitch data and stores the volume data in the FIFO for each predetermined section.

  The effective section specifying unit 107 specifies the effective section in the predetermined section based on the feature amount of the predetermined section of the input sound acquired by the feature amount acquisition unit 105 and stored in the FIFO. The effective section specifying unit 107 specifies the effective section in the predetermined section based on the pitch data among the feature amounts acquired by the feature amount acquiring unit 105. However, the effective section specifying unit 107 may specify the effective section in the predetermined section based on the volume data among the feature amounts acquired by the feature amount acquiring unit 105. Further, the valid section specifying unit 107 may specify a valid section in the predetermined section based on the pitch data and the volume data. Here, the effective section means a section corresponding to pitch data to be compared with predetermined reference sound information in a predetermined section. In the present embodiment, an example in which the perforated section specifying unit 107 specifies an effective section in the predetermined section based on the pitch data of the predetermined section acquired by the feature amount acquiring unit 105 and stored in the FIFO will be described. .

  The effective section specifying unit 107 sequentially acquires the pitch data of the predetermined section stored in the FIFO from the feature amount acquisition unit 105, and specifies the effective section in the predetermined section based on the acquired predetermined section based on the pitch data. For example, the effective section specifying unit 107 can specify the effective section in the predetermined section based on the absolute value of the slope of the linear approximation curve of the pitch of the predetermined section.

  FIG. 4A is a diagram for explaining an example of a concept of specifying an effective section in a predetermined section by the effective section specifying unit 107. FIG. 4A shows a time-series variation of pitch data in a predetermined section. In FIG. 4A, the predetermined section is t0 to ti. When acquiring the pitch data of the predetermined sections t0 to ti, the valid section specifying unit 107 performs linear approximation of each pitch in the acquired pitch data and obtains the slope of the linear approximation curve. The effective section specifying unit 107 specifies a section in which the absolute value of the slope of the linear approximation curve of the pitch is less than a predetermined threshold in the predetermined sections t0 to ti as the effective section. The predetermined threshold value is arbitrary and may be set by the user of the evaluation device 10. In FIG. 4A, tb to td are shown as effective intervals, and the absolute value of the slope of the linear approximate curve of the pitch in the effective intervals tb to td is less than a predetermined threshold. For example, in the effective sections tb to td in the predetermined section t0 to ti, the absolute value of the slope of the linear approximate curve B of the pitch at tb, the absolute value of the slope of the linear approximate curve C of the pitch at tc, and the linear approximation of the pitch at td The absolute value of the slope of the curve D is less than a predetermined threshold value. On the other hand, the absolute value of the slope of the linear approximate curve of the pitch in the interval t0 to tb and the interval td to ti excluding the valid intervals tb to td in the predetermined interval t0 to ti, for example, the slope of the linear approximate curve A in ta And the absolute value of the slope of the linear approximation curve at te are equal to or greater than the predetermined threshold.

  In the above, the method in which the effective section specifying unit 107 specifies the effective section in the predetermined section based on the absolute value of the slope of the linear approximation curve of the pitch in the predetermined section has been described. The method is not limited to the identification method based on the absolute value of the slope of the linear approximate curve of the pitch in the predetermined section. For example, the effective section specifying unit 107 can specify the effective section in the predetermined section based on the pitch dispersion in the predetermined section.

  When obtaining the pitch data of the predetermined sections t0 to ti, the valid section specifying unit 107 calculates a variance value of the acquired pitch data. When the calculated variance value is equal to or greater than the predetermined value, the effective interval specifying unit 107 determines that there is no effective interval in the predetermined intervals t0 to ti. On the other hand, when the calculated variance value is less than the predetermined value, the effective interval specifying unit 107 determines that an effective interval exists in the predetermined intervals t0 to ti. Here, the predetermined value is arbitrary and can be set by the user of the evaluation apparatus 10.

  When there is an effective section in the predetermined section t0 to ti, the effective section specifying unit 107 specifies the effective section in the predetermined section t0 to ti based on the pitch data of the predetermined section t0 to ti. In a predetermined section t0 to ti, the valid section specifying unit 107 is a section corresponding to a pitch existing in a pitch range having a frequency equal to or higher than a predetermined threshold when a pitch range having a frequency equal to or higher than a predetermined threshold exists. A section in which the pitch existing in the pitch range is continuous for a predetermined period is identified as an effective section.

  FIG. 4B is a diagram for explaining another example of the concept of specifying the effective section in the predetermined section by the effective section specifying unit 107. FIG. 4B (a) shows the frequency of the pitch in the predetermined section, and FIG. 4B (b) shows the time-series fluctuation of the pitch data in the predetermined section. As shown in FIG. 4B (a), the pitches in the pitch data of the predetermined sections t0 to ti are: pitch Pa to pitch Pb, pitch Pb to pitch Pc, pitch Pc to pitch Pd, pitch Pd to pitch Pe, pitch Pe to The pitch Pf, the pitch Pf to the pitch Pg, the pitch Pf to the pitch Ph, and the pitch Ph to the pitch Pi exist. In addition, each pitch range can be set arbitrarily and the pitch in each pitch range may overlap and does not need to overlap. The valid section specifying unit 107 specifies a pitch range having a frequency equal to or higher than a predetermined threshold Va in the predetermined sections t0 to ti. As shown in FIG. 4B (b), the frequency in the pitch range of the pitch Pf to the pitch Pg is equal to or greater than a predetermined threshold value Va. The valid section specifying unit 107 identifies a section corresponding to a pitch existing in the pitch range of the pitch Pf to the pitch Pg and having a pitch corresponding to the pitch range of the pitch Pf to the pitch Pg as a valid section. To do. For example, as shown in FIG. 4B (b), when the pitch corresponding to the pitch range of the pitch Pf to the pitch Pg is continuous in the section of tx1 to tx2, the valid section specifying unit 107 sets the predetermined section t0 to ti. In t, the interval from tx1 to tx2 is specified as the effective interval.

  As described above, the valid section specifying unit 107 sequentially acquires the pitch data of the predetermined section stored in the FIFO from the feature amount acquisition unit 105, and described above based on the acquired pitch data of the predetermined section. As described above, the effective sections in the predetermined section are sequentially specified. When the valid section specifying unit 107 identifies a valid section in the predetermined section, the valid section specifying unit 107 transmits information indicating the valid section in the predetermined section to the evaluation unit 111. If there is no valid section in the predetermined section, information indicating that the predetermined section does not include the valid section is transmitted to the evaluation unit 111. Note that the method for specifying the effective section in the predetermined section is not limited to the method described above. Further, as described above, in the evaluation function 100 according to the embodiment of the present invention, the valid section determination unit 107 may be omitted.

  When the evaluation unit 111 described later calculates an evaluation value of pitch data in a predetermined section, the switching determination unit 109 determines whether to switch reference sound information to be compared with pitch data in the predetermined section. When evaluating a phrase in which a plurality of input sounds are continued, when the input sound is switched, the evaluation unit 111 switches reference sound information to be compared with the switched input sound. The switching determination unit 109 determines whether or not input sound is switched in a predetermined section.

  The switching determination unit 109 can determine whether or not the input sound is switched in the predetermined section based on the input sound in the predetermined section. For example, the switching determination unit 109 may determine whether or not the input sound is switched in the predetermined section based on the pitch data in the predetermined section.

  FIG. 5 is a diagram for explaining an example of the concept of determining whether or not input sound is switched in a predetermined section by the switching determination unit 109. The switching determination unit 109 sequentially acquires pitch data of a predetermined section from the feature amount acquisition unit 105. FIG. 5 shows time-series fluctuations in pitch data in a predetermined section. In FIG. 5, the predetermined section is t0 to ti. When the switching determination unit 109 acquires the pitch data of the predetermined sections t0 to ti, the switching determination unit 109 performs linear approximation of each pitch in the acquired pitch data and obtains the slope of the linear approximation curve. The switching determination unit 109 determines that the input sound is switched in the predetermined section t0 to ti when there is a section in which the absolute value of the slope of the linear approximation curve of the pitch is equal to or greater than a predetermined threshold in the predetermined section t0 to ti. To do. Here, the predetermined threshold value is arbitrary and may be set by the user of the evaluation device 10. In FIG. 5, when the absolute value of the slope of the linear approximate curve of the pitch at tb and tc is equal to or greater than the threshold, the switching determination unit 109 determines that the input sound is switched in the section including tb and tc. When it is determined that the input sound is switched in the predetermined section, the switching determination unit 109 transmits information indicating that the input sound is switched in the predetermined section to the evaluation unit 111. If it is determined that there is no input sound switching in the predetermined section, information indicating that there is no input sound switching in the predetermined section is transmitted to the evaluation unit 111.

  The method for determining the input sound in the predetermined section based on the absolute value of the slope of the linear approximation curve of the pitch in the predetermined section has been described above. The presence / absence is not limited to the determination method based on the absolute value of the slope of the linear approximate curve of the pitch in the predetermined section. For example, when there is a section in which the pitch is not detected in the predetermined section, the switching determination unit 109 may determine that the input sound is switched in the predetermined section.

  Further, the switching determination unit 109 may determine whether or not the input sound is switched in the predetermined section based on the volume data in the predetermined section. FIG. 6 is a diagram for explaining another example of the concept of determining whether or not input sound is switched in a predetermined section by the switching determination unit 109. FIG. 6 shows time-series fluctuations in volume data in a predetermined section. The switching determination unit 109 sequentially acquires volume data of a predetermined section from the feature amount acquisition unit 105. In FIG. 6, the predetermined section is t0 to ti. When the sound volume data of the predetermined sections t0 to ti is acquired, the switching determination unit 109 determines that the input sound is switched in the predetermined sections t0 to ti when detecting the sound volume that is less than the predetermined threshold Vth. FIG. 6 shows that the volume is less than the predetermined threshold Vth in the section between ta and tb among the predetermined sections t0 to ti. The switching determination unit 109 determines that the input sound is switched in the section between ta and tb in the predetermined sections t0 to ti.

  Although not shown, the switching determination unit 109 may perform linear approximation of the volume data of a predetermined section acquired from the feature amount acquisition unit 105 and determine whether or not the input sound is switched in the predetermined section. In this case, the switching determination unit 109 obtains the slope of the linear approximation curve of the volume data in a predetermined section, and there is a section in the predetermined section where the absolute value of the slope of the linear approximation curve of the volume data is greater than or equal to a predetermined threshold. It is determined that the input sound is switched in the predetermined section.

  In the above, the method has been described in which the switching determination unit 109 determines whether or not the input sound is switched in the predetermined section based on the pitch data or the volume data in the predetermined section. The method for determining the presence or absence of switching is not limited to these.

  The evaluation unit 111 sequentially acquires pitch data of a predetermined section stored in the FIFO from the feature amount acquisition unit 105. In addition, when there is no effective section in the acquired predetermined section, the evaluation unit 111 receives information indicating that the predetermined section does not include the effective section from the effective section specifying unit 107, and there is an effective section in the acquired predetermined section. If it exists, information indicating the valid section in the predetermined section is received from the valid section identifying unit 107. When there is no valid section in the acquired predetermined section, the evaluation unit 111 outputs information indicating that evaluation is not possible. On the other hand, when the effective section in the predetermined section is specified by the effective section specifying unit 107, the evaluation unit 111 compares the pitch data in the effective section in the predetermined section with predetermined reference sound information, and the comparison result is obtained. Based on this, an evaluation value of pitch data in the predetermined section is calculated. At this time, when information indicating that the input sound is switched in the predetermined section is transmitted from the switching determination unit 109, the evaluation unit 111 stores another reference sound information corresponding to the switched input sound. 13 and the reference sound information to be compared with the pitch data in the valid section within the predetermined section is switched. Hereinafter, a method for evaluating pitch data in a predetermined section in the evaluation unit 111 will be described.

  The evaluation unit 111 specifies the representative pitch in the pitch data of the specified effective section, compares the specified representative pitch with the pitch of the predetermined reference sound information, and calculates the evaluation value of the pitch data of the predetermined section can do. Here, the representative pitch may be a predetermined pitch in the effective section, or may be an average pitch of pitch data in the effective section. The representative pitch may be a median pitch in the effective section.

  The evaluation unit 111 calculates an evaluation value of pitch data in a predetermined section based on whether or not the identified representative pitch is in a predetermined pitch range based on the pitch of predetermined reference sound information. FIG. 7A is a diagram for explaining an example of a concept by which the evaluation unit 111 calculates an evaluation value of pitch data in a predetermined section based on a representative pitch of an effective section in the predetermined section. As shown in FIG. 7A, when the representative pitch of the valid section exists in the pitch range of pitch Pref-a cent to pitch Pref + a cent with reference to the pitch Pref of the predetermined reference sound information, the evaluation unit 111 Assume that the evaluation value of pitch data in a predetermined section including the section is 100 points (passed). In FIG. 7A, the pitch range of 100 points (accepted) is indicated by hatching. When the representative pitch of the valid section is outside the pitch range of pitch Pref-a cent to pitch Pref + a cent, the evaluation unit 111 sets the evaluation value of the pitch data of the predetermined section including the valid section as 0 (fail). To do.

  The predetermined pitch range based on the pitch Pref of the predetermined reference sound information may be divided into a plurality of sections. The evaluation unit 111 can calculate the evaluation value of the pitch data in the predetermined section based on the section corresponding to the identified representative pitch. FIG. 7B is a diagram for explaining another example of the concept of calculating the evaluation value of the pitch data in the predetermined section based on the representative pitch of the effective section in the predetermined section by the evaluation unit 111. In FIG. 7B, as an example, the predetermined pitch range based on the pitch Pref of the predetermined reference sound information is divided into three sections (section 401, section 403, and section 405).

  In FIG. 7B, a section 401 indicates a pitch range of pitch Pref-b cent to pitch Pref + b cent with reference to a pitch Pref of predetermined reference sound information, and a section 403 indicates a pitch range of pitch Pref + b cent to pitch Pref + a cent. A section 405 indicates a pitch range from pitch Pref-b cent to pitch Pref-a cent. Here, a> b. As shown in FIG. 7B, when the representative pitch of the valid section exists in the pitch range from pitch Pref-b cent to pitch Pref + b cent with reference to the pitch Pref of the predetermined reference sound information, that is, it corresponds to the representative pitch. When the section exists in the section 401, the evaluation unit 111 sets the evaluation value of the pitch data of the predetermined section including the valid section as 100 points (passed). When the section corresponding to the representative pitch exists in the section 403 or the section 405, the evaluation unit 111 sets the evaluation value of the pitch data of the predetermined section in which the valid section exists as 100> x> 0 points (pass). Here, x is an arbitrary number given to the pitch data of the predetermined section by the evaluation unit 111, and as shown in FIG. 7B, the representative pitch approaches the pitch range of pitch Pref-b cent to pitch Pref + b cent. The closer it is, the higher the score may be given, and the lower the score may be, the farther the representative pitch is from the pitch range of pitch Pref-b cent to pitch Pref + b cent. When the representative pitch of the valid section is outside the pitch range of pitch Pref-a cent to pitch Pref + a cent, that is, outside the section 401, section 403, and section 405, the evaluation unit 111 includes the valid section. Assume that the evaluation value of the pitch data in the predetermined section is 0 (failed).

  As shown in FIG. 7B, when dividing a predetermined pitch range based on the pitch Pref of the reference sound information into a plurality of sections and calculating an evaluation value for pitch data of the predetermined section, evaluation values corresponding to the sections May be changeable depending on the instrument played. When the predetermined reference sound information is sharp (#) or flat (♭), the evaluation value corresponding to each section may be set to be unbalanced. Furthermore, the determination criterion may be changed based on the musical instrument played and the pitch of the reference sound information. For example, when a high tone is evaluated using an instrument that does not easily produce a high pitch range, a predetermined pitch range based on the pitch Pref of the reference sound information may be increased, or the pitch range may be finely divided. Further, when the evaluation unit 111 acquires the pitch data of the predetermined section stored in the FIFO from the feature amount acquisition unit 105, the evaluation unit 111 acquires volume data associated with the pitch data of the predetermined section, and the effective section Or when there is little fluctuation | variation of the sound volume in a predetermined area, you may give a weight value to the evaluation value with respect to the pitch data of this predetermined area.

  As described above, the evaluation unit 111 specifies the representative pitch in the pitch data of the specified effective section, and compares the specified representative pitch with the pitch of the predetermined reference sound information, thereby determining the pitch data of the predetermined section. Although the example which calculates an evaluation value was demonstrated, calculation of the evaluation value of the pitch data of the predetermined area by the evaluation part 111 is not necessarily limited to this. For example, the evaluation unit 111 may calculate the evaluation value of the pitch data of the predetermined section by comparing the pitch data of the specified effective section and the pitch of the predetermined reference sound information. That is, the evaluation on the pitch data of the predetermined section by the evaluation unit 111 described with reference to FIGS. 7A and 7B is performed by specifying the representative pitch of the effective section in the predetermined section, and the specified representative pitch and reference sound information. Although the pitch is compared, it is not always necessary to specify the representative pitch. The evaluation unit 111 may calculate the evaluation value of the pitch data of the predetermined section by comparing the pitch data of the whole or a part of the effective section with the pitch of the predetermined reference sound information.

  In addition, when comparing the pitch data of the valid section and the pitch of the predetermined reference sound information, the evaluation unit 111 has all or part of the pitch data of the valid section in the pitch range including the pitch Pref of the predetermined reference sound information. By determining whether or not to perform, an evaluation value for pitch data of a predetermined section including the effective section may be calculated. For example, as shown in FIG. 7A, when all or part of the pitch data of the valid section exists in the pitch range of pitch Pref-a cent to pitch Pref + a cent with reference to the pitch Pref of the predetermined reference sound information. The evaluation unit 111 may set the evaluation value of pitch data in a predetermined section including the effective section as 100 points (pass). Further, as shown in FIG. 7B, a predetermined pitch range based on the pitch Pref of the reference sound information may be divided into a plurality of sections. In this case, the evaluation unit 111 corresponds to the pitch data of the effective section. The evaluation value of the pitch data in the predetermined section may be calculated based on the section to be performed.

  When a predetermined pitch range based on the pitch Pref of the reference sound information is divided into a plurality of sections and an evaluation value for the pitch data of the predetermined section is calculated, the evaluation value corresponding to each section depends on the musical instrument played. It may be changeable. When the predetermined reference sound information is sharp (#) or flat (♭), the evaluation value corresponding to each section may be set to be unbalanced. Furthermore, the determination criterion may be changed based on the musical instrument played and the pitch of the reference sound information. For example, when a high tone is evaluated using an instrument that does not easily produce a high pitch range, a predetermined pitch range based on the pitch Pref of the reference sound information may be increased, or the pitch range may be finely divided. Further, when the evaluation unit 111 acquires the pitch data of the predetermined section stored in the FIFO from the feature amount acquisition unit 105, the evaluation unit 111 acquires volume data associated with the pitch data of the predetermined section, and the volume in the effective section or the predetermined section When there is little fluctuation in the weight, a weight value may be given to the evaluation value for the pitch data in the predetermined section.

  Further, the evaluation unit 111 may take statistics of the pitch data of the effective section and calculate an evaluation value for the pitch data of the predetermined section including the effective section based on the frequency distribution of the pitch of the effective section. FIG. 8A is a diagram for explaining an example of a concept of calculating an evaluation value of pitch data in the predetermined section based on the frequency distribution of the pitch in the effective section by the evaluation unit 111. As shown in FIG. 8A, the evaluation unit 111, based on the pitch Pref of the predetermined reference sound information, uses the pitch Pref of the predetermined reference sound information as a reference, and a pitch range of pitch Pref-a cent to pitch Pref + a cent. 801 is set. Further, the evaluation unit 111 sets a weight value A in the pitch range 801. For example, the weight value A may be 100 points. The evaluation unit 111 multiplies the ratio of the frequency of pitches in the pitch range 801 with respect to all pitches in the effective section by a weight value A (100 points), and uses the obtained score as an evaluation value of pitch data in a predetermined section including the effective section And The higher the frequency of the pitch in the pitch range 801, the higher the evaluation value of the pitch data in the predetermined section including the effective section. When the evaluation value of the pitch data in the predetermined section is equal to or greater than a predetermined score, the evaluation unit 111 may uniformly set the evaluation value of the pitch data in the predetermined section as 100 points (pass). As the frequency of the pitch of the effective section in the pitch range 801 is lower, the evaluation value of the pitch data of the predetermined section including the effective section is lower. For example, when the pitch of the effective section does not exist in the pitch range 801, the evaluation unit 111 sets the evaluation value of the pitch data of the predetermined section including the effective section as 0 points (failed). Further, when the evaluation value of the pitch data in the predetermined section is less than the predetermined score, the evaluation unit 111 may uniformly set the evaluation value of the pitch data in the predetermined section to 0 point (fail).

  The predetermined pitch range based on the pitch Pref of the predetermined reference sound information may be divided into a plurality of sections. The evaluation unit 111 can calculate the evaluation value of the pitch data in the predetermined section based on the frequency distribution of the pitch of the effective section in each section. FIG. 8B is a diagram for explaining another example of the concept of calculating the evaluation value of the pitch data in the predetermined section based on the frequency distribution of the pitch in the effective section by the evaluation unit 111. As shown in FIG. 8B, the evaluation unit 111 determines a pitch range from pitch Pref-a cent to pitch Pref + a cent based on the pitch Pref of the predetermined reference sound information based on the pitch Pref of the predetermined reference sound information. 801 is set. Further, the evaluation unit 111 divides the pitch range 801 into three sections (section 803, section 805, and section 807). Note that the number of divided sections is not limited to three.

  In FIG. 8B, a section 803 indicates a pitch range of pitch Pref-b cent to pitch Pref + b cent based on a pitch Pref of predetermined reference sound information, and a section 805 indicates a pitch range of pitch Pref + b cent to pitch Pref + a cent. A section 807 indicates a pitch range from pitch Pref-b cent to pitch Pref-a cent. Here, a> b. The evaluation unit 111 sets a weight value A, a weight value B, and a weight value C for the sections 803, 805, and 807 in the pitch range 801, respectively. For example, the weight value A for the section 803 may be 100 points. The weight values B and C for the sections 805 and 807 are not particularly limited as long as they are smaller than the weight value A, and the weight values B and C may be the same or different. For example, the weight value B and the weight value C may be 60 points. The weight value corresponding to the divided section is set to be relatively higher as it is closer to the pitch Pref of the predetermined reference sound information, and is set to be relatively lower as it is farther from the pitch Pref of the predetermined reference sound information.

  The evaluation unit 111 calculates the evaluation value of the section 803 by multiplying the ratio of the frequency of the pitch of the section 803 by the weight value A (100 points). Similarly, the evaluation unit 111 calculates the evaluation value of the section 805 by multiplying the ratio of the frequency of the pitch of the section 805 by the weight value B (60 points). In addition, the evaluation unit 111 calculates the evaluation value of the section 807 by multiplying the ratio of the frequency of the pitch of the section 807 by the weight value C (60 points). The evaluation unit 111 sets the score obtained by adding the evaluation value of the section 803, the evaluation value of the section 805, and the evaluation value of the section 807 as the evaluation value of the pitch data of the predetermined section including the effective section. The higher the frequency of the pitch in the pitch range 801, the higher the evaluation value of the pitch data in the predetermined section including the effective section. In particular, the higher the pitch frequency of the section 803 in the pitch range 801, that is, the closer the pitch of the effective section is to the pitch Pref of the reference sound information, the higher the evaluation value of the pitch data of the predetermined section including the effective section is. Become. When the evaluation value of the pitch data in the predetermined section is equal to or greater than a predetermined score, the evaluation unit 111 may uniformly set the evaluation value of the pitch data in the predetermined section as 100 points (pass). On the other hand, as the frequency of the effective section in the pitch range 801 is lower, that is, as the pitch of the effective section is farther from the pitch Pref of the reference sound information, the evaluation value of the pitch data of the predetermined section including the effective section becomes lower. . For example, when there is no pitch data in the valid section in the pitch range 801, the evaluation unit 111 sets the evaluation value of the pitch data in the predetermined section including the valid section as 0 (failed). Further, when the evaluation value of the pitch data in the predetermined section is less than the predetermined score, the evaluation unit 111 may uniformly set the evaluation value of the pitch data in the predetermined section to 0 point (fail).

  As shown in FIG. 8B, when the predetermined pitch range based on the pitch Pref of the reference sound information is divided into a plurality of sections and the evaluation value for the pitch data of the predetermined section is calculated, the weight value corresponding to each section May be changeable depending on the instrument played. Further, when the predetermined reference sound information is sharp (#) or flat (♭), the weight value corresponding to each section may be set to be unbalanced. Furthermore, the determination criterion may be changed based on the musical instrument played and the pitch of the reference sound information. For example, when a high tone is evaluated using an instrument that does not easily produce a high pitch range, a predetermined pitch range based on the pitch Pref of the reference sound information may be increased, or the pitch range may be finely divided. Further, when the evaluation unit 111 acquires the pitch data of the predetermined section stored in the FIFO from the feature amount acquisition unit 105, the evaluation unit 111 acquires volume data associated with the pitch data of the predetermined section, and the volume in the effective section or the predetermined section When there is little fluctuation in the weight, a weight value may be given to the evaluation value for the pitch data in the predetermined section.

  The evaluation unit 111 determines the final value for the input sound when the evaluation value for the pitch data of the predetermined section is equal to or greater than the threshold value for a predetermined number of times or when the evaluation value is continuously for the predetermined number of times. To make a pass. In this case, the evaluation unit 111 may use the highest score among the evaluation values for the pitch data in the predetermined section as the final evaluation value for the input sound, and determine the average score of the evaluation values for the pitch data in the predetermined section. The final evaluation value for the input sound may be used. In addition, the evaluation unit 111 assigns a weight value to the final evaluation value based on the length of the period in which the evaluation value for the pitch data of the predetermined section is equal to or greater than the threshold, that is, the length of the period determined to be acceptable. Alternatively, a weight may be further added as the period during which the pitch of the input sound is within a predetermined pitch range based on the pitch Pref of the reference sound information is longer. Further, the evaluation unit 111 applies a low pass filter to the evaluation value for the pitch data of a predetermined predetermined number or more of continuous intervals, and when the evaluation value from which the noise component is removed continues for a certain level or more, And final evaluation may be performed. In addition, the evaluation unit 111 determines that the final time when the input sound pitch falls within a predetermined pitch range with reference to the pitch Pref of the reference sound information is shorter than the predetermined time from the start point of the input sound pitch acquisition. A weight value may be given to a typical evaluation value. In this case, the predetermined time can be arbitrarily set.

  The evaluation unit 111 calculates an evaluation value for the pitch data of a predetermined section by the method described above, and transmits the calculated evaluation value to the display unit 15. The display unit 15 displays the result based on the transmitted evaluation value so that the musical instrument player can recognize the evaluation of the pitch data in the predetermined section. Although the evaluation value calculation method for the pitch data of the predetermined section by the evaluation unit 111 has been described above, the evaluation value calculation method is not limited to the method described above.

  In the above description, the evaluation unit 111 compares the pitch data of the valid section of the predetermined section with the corresponding reference sound information, and calculates the evaluation value for the pitch data of the predetermined section based on the comparison result. . However, when the valid section specifying unit 107 is omitted in the evaluation function 100, the evaluation unit 111 compares the entire pitch data of the predetermined section or any part of the pitch data with the corresponding reference sound information, and compares them. An evaluation value for the pitch data of the predetermined section may be calculated based on the result. At this time, the evaluation unit 111 specifies the representative pitch of the entire pitch data of the predetermined section or an arbitrary part of the pitch data, and as described with reference to FIGS. 7A and 7B, An evaluation value may be calculated. In addition, the evaluation unit 111 calculates the evaluation value for the pitch data of the predetermined section, as described with reference to FIGS. 8A and 8B, based on the entire pitch data of the predetermined section or an arbitrary part of the frequency distribution of the pitch. It may be calculated.

  The evaluation unit 111 sequentially calculates evaluation values for pitch data in a predetermined section, and evaluates a phrase in which a plurality of input sounds are continuous. The evaluation unit 111 may output the average evaluation value for each sound in the phrase as the final evaluation value for the phrase. Further, the evaluation unit 111 may give a weight value to the final evaluation value when the input sound switching in the phrase is smooth, and the final evaluation value when the input sound switching in the phrase is not smooth. May be deducted. For example, the evaluation unit 111 may determine that the input sound switching is smooth if the input sound switching interval is less than a predetermined period, and if the input sound switching interval is equal to or longer than the predetermined period, the input sound It may be determined that the switching is not smooth. Further, the evaluation unit 111 may determine that the switching of the input sound is smooth if the absolute value of the slope of the linear approximation curve in the input sound switching section is equal to or greater than a predetermined value. Further, the evaluation unit 111 may determine a weighted value or a deduction value for the final evaluation value based on the absolute value of the slope of the linear approximation curve in the input sound switching section.

  Further, the evaluation unit 111 may specify a representative volume of each sound in the phrase, and may assign a weight value to the final evaluation value if the representative volume of each sound is uniform.

  Further, the evaluation unit 111 obtains an average pitch of each sound in the phrase, compares the average pitch of each sound with the corresponding reference sound information, and based on the comparison result, a weighted value or a deduction point for the final evaluation value The value may be determined. For example, if the difference between the average pitch of each sound and the corresponding reference sound information is less than a predetermined value, a weight value may be given to the final evaluation value, and the reference sound information corresponding to the average pitch of each sound If the difference between and is equal to or greater than a predetermined value, the final evaluation value may be deducted.

  According to the evaluation function 100 described above, an effective section is specified from the pitch data of a predetermined section, and the pitch data of the predetermined section is evaluated based on the pitch data of the effective section. The input sound can be evaluated with high accuracy with respect to a plurality of predetermined continuous pitches.

(Modification)
In the evaluation function 100 described above, the effective section specifying unit 107 specifies the effective section in the predetermined section based on the pitch data of the predetermined section. However, the effective section specifying unit 107 can also specify the effective section in the predetermined section based on the volume data among the feature amounts acquired by the feature amount acquiring unit 105.

  The effective section specifying unit 107 sequentially acquires volume data of a predetermined section stored in the FIFO from the feature amount acquisition unit 105, and specifies an effective section in the predetermined section based on the obtained change amount of the volume data of the predetermined section. To do. For example, the effective interval specifying unit 107 can specify an effective interval in the predetermined interval for the interval in which the volume fluctuation in the predetermined interval is not more than a predetermined threshold.

  FIG. 9 is a diagram for explaining an example of a concept of specifying an effective section in a predetermined section by the effective section specifying unit 107. FIG. 9 shows time-series fluctuations in volume data in a predetermined section. In FIG. 9, the predetermined section is t0 to ti. When acquiring the volume data of the predetermined sections t0 to ti, the valid section specifying unit 107 performs linear approximation of the volume in the acquired volume data and obtains the slope of the linear approximation curve. The effective section specifying unit 107 specifies a section where the absolute value of the slope of the linear approximation curve of the volume is less than a predetermined threshold in the predetermined sections t0 to ti as the effective section. The predetermined threshold value is arbitrary and may be set by the user of the evaluation device 10. In FIG. 9, tb to td are shown as effective intervals, and the absolute value of the slope of the linear approximate curve of the sound volume in the effective intervals tb to td is less than a predetermined threshold. For example, in the effective sections tb to td in the predetermined section t0 to ti, the absolute value of the slope of the linear approximate curve B of the volume at tb, the absolute value of the slope of the linear approximate curve C of the volume at tc, and the linear approximation of the volume at td The absolute value of the slope of the curve D is less than a predetermined threshold value. On the other hand, the absolute value of the slope of the linear approximation curve of the volume in the t0 to tb section and the td to ti section, excluding the effective sections tb to td in the predetermined section t0 to ti, for example, the slope of the linear approximation curve A at ta , The absolute value of the slope of the linear approximation curve at te, and the absolute value of the slope of the linear approximation curve at tf are equal to or greater than the predetermined threshold.

  As described above, the valid section specifying unit 107 can specify the valid section in the predetermined section based on the volume data of the predetermined section. Note that the method of specifying the effective section using the volume data of the predetermined section in the effective section specifying unit 107 is not limited to the method described with reference to FIG.

  After the effective section specifying unit 107 specifies the effective section in the predetermined section based on the volume data, the pitch of the predetermined section including the effective section based on the pitch data of the effective section is evaluated by the evaluation unit 111 as described above. An evaluation value is calculated for the data.

  Based on the configuration described as the embodiment of the present invention, those in which a person skilled in the art appropriately added, deleted, or changed the design of the component, or added, omitted, or changed conditions of the process are also included in the present invention. As long as the gist of the present invention is provided, the scope of the present invention is included.

  Of course, other operational effects that are different from the operational effects brought about by the above-described embodiment are obvious from the description of the present specification or can be easily predicted by those skilled in the art. It is understood that this is brought about by the present invention.

10: evaluation device, 11: control unit, 13: storage unit, 15: display unit, 17: signal processing unit, 19: sound input unit, 21: operation unit, 23: communication unit, 100: evaluation function, 103: input Sound acquisition unit, 105: feature amount acquisition unit, 107: effective section identification unit, 109: switching determination unit, 111: evaluation unit

Claims (9)

On the computer,
Get the input sound,
Pitch data is acquired in time series based on the input sound, among the acquired pitch data, the pitch data of a predetermined section is accumulated and updated every predetermined time,
The pitch data of the predetermined section is compared with predetermined reference sound information, and an evaluation value of the pitch data of the predetermined section is calculated based on the comparison result,
Determine whether to switch the reference sound information based on the input sound of the period corresponding to the pitch data of the predetermined section,
When switching the reference sound information, the pitch data of the predetermined section is compared with predetermined reference sound information different from the predetermined reference sound information, and the evaluation value of the pitch data of the predetermined section is calculated based on the comparison result To do,
A program for running   The program according to claim 1, wherein the computer is configured to determine whether to switch the reference sound information based on pitch data of the predetermined section as the determination of whether to switch the reference sound information.   2. The program according to claim 1, wherein the computer is configured to determine whether to switch the reference sound information based on volume data of the predetermined section as the determination of whether to switch the reference sound information.   4. The computer according to claim 1, further comprising: identifying, based on the input sound of the predetermined section, a section to be compared with the predetermined reference sound information in the predetermined section. 5. A program according to any one of the above.   By calculating the evaluation value of the pitch data in the predetermined section, the representative pitch in the predetermined section is specified, and the pitch data in the predetermined section is evaluated by comparing the representative pitch with the pitch of the predetermined reference sound information. The program according to any one of claims 1 to 4, which causes the computer to calculate a value.   Based on whether or not the pitch data in the predetermined section is within a predetermined pitch range based on the pitch corresponding to the predetermined reference sound information, as the evaluation value of the pitch data in the predetermined section is calculated. The program according to any one of claims 1 to 4, which causes the computer to execute an evaluation value of pitch data of the predetermined section.   Calculating the evaluation value of the pitch data in the predetermined section based on the frequency distribution of the pitch data in the predetermined section, and calculating the evaluation value of the pitch data in the predetermined section. The program as described in any one of Claim 1 thru | or 4. An input sound acquisition unit for acquiring the input sound;
A feature amount acquisition unit that acquires pitch data in a time series based on the input sound, and accumulates and sequentially updates pitch data of a predetermined section of the acquired pitch data for each predetermined time;
An evaluation unit that compares the pitch data of the predetermined section with predetermined reference sound information, and calculates an evaluation value of the pitch data of the predetermined section based on a comparison result;
A switching determination unit that determines whether to switch the reference sound information based on an input sound in a period corresponding to the pitch data of the predetermined section;
With
When the switching determination unit determines to switch the reference sound information, the evaluation unit compares the pitch data of the predetermined section with predetermined reference sound information different from the predetermined reference sound information, and determines the comparison result. An evaluation device that calculates an evaluation value of pitch data in the predetermined section based on the evaluation value. Get the input sound,
Pitch data is acquired in time series based on the input sound, among the acquired pitch data, the pitch data of a predetermined section is accumulated and updated every predetermined time,
The pitch data of the predetermined section is compared with predetermined reference sound information, and an evaluation value of the pitch data of the predetermined section is calculated based on the comparison result,
Determine whether to switch the reference sound information based on the input sound of the period corresponding to the pitch data of the predetermined section,
When switching the reference sound information, the pitch data of the predetermined section is compared with predetermined reference sound information different from the predetermined reference sound information, and the evaluation value of the pitch data of the predetermined section is calculated based on the comparison result To do,
Including evaluation method.

Images