JP2010156777A - Singing grading system and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To perform grading in consideration of a user's singing ability and to allow the extent of grading to be known on the basis of the user's singing ability. <P>SOLUTION: Evaluation values are computed on voice related to the user's singing of a musical piece for every section, and after the evaluation values are corrected based on the parameter accuracy of the singing user, the statistics of the respective evaluation values is taken to compute a final grading result. The parameter accuracy shows whether the voice of the ideal parameter can be accurately vocalized for every kind of ideal parameters in each section of the musical piece, i.e., shows the singing ability for the ideal parameter of the singing user. The evaluation value in the section with low parameter accuracy relatively tends to be low, but since the user with low singing ability is considered to have sung well for his or her own singing ability, the grading result is computed after correcting to make high the evaluation of such a section. The singing user's singing ability is thereby reflected in the evaluation value of the section. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a singing scoring system for scoring a song sung by a user.

In the past, scoring of music sung by a user has been performed, and in recent years, it has been proposed to perform scoring with high accuracy according to the characteristics of music (for example, Patent Documents 1 and 2). reference).
JP 2005-107333 A JP 2007-233307 A

  In the conventional technology, the grading is performed based on the same standard regardless of the singing user, and it is useful in comparing the grading result with the grading result of other singing users. Singing power is not taken into consideration, so it is used to know how much the result is based on the singing user's singing power (specifically, the singing power is good enough) It is not possible.

  The present invention has been made to solve such a problem, and the purpose of the present invention is to allow singing user's singing ability to be scored and to what extent the singing user's singing ability is used as a reference. It is to provide a technique for making it possible to know whether or not it is a scoring.

  In order to solve the above-mentioned problem, it is conceivable that the singing scoring system has a first configuration (claim 1) shown below.

  In this configuration, for each of the sections obtained by dividing the voice of the song sung by the user for each predetermined section in time series, the singing parameters relating to the voice of the section are ideal based on the correct voice to be uttered in the section. Section accuracy specifying means for specifying the accuracy (section accuracy) of the singing parameter with respect to the ideal parameter for each section by comparing with the parameters, and each section accuracy specified by the section accuracy specifying means By taking statistics for each type of the ideal parameter, the accuracy (parameter accuracy) when the corresponding user utters the voice defined by the ideal parameter is determined accurately for each type of the ideal parameter. Sex specifying means, and the type and type of the ideal parameter specified by the parameter accuracy specifying means. User storage means for storing the parameter accuracy of each in the storage unit in association with the user who sang the music that triggered the parameter accuracy, and the music to be scored by the user (scoring music) For each section in the case where the voice formed by singing is divided into predetermined sections along the time series, the evaluation value for each section is obtained by comparing the singing parameters relating to the sound of the section with the ideal parameters in the section. The section evaluation means to be calculated and the evaluation value for each section calculated by the section evaluation means, among the parameter accuracy stored in the storage unit by the user storage means, the scoring target music that triggered the evaluation Evaluation correction means for correcting based on the parameter accuracy corresponding to the singing user who sang the song, and the evaluation value after correction by the evaluation correction means. For, respectively, by taking statistics on all sections of the scoring object music pieces, the singing user is provided with a singing scoring means for calculating the scoring results when singing the scoring object music pieces.

  And the said evaluation correction means WHEREIN: Of the parameter accuracy of the said singing user memorize | stored in the memory | storage part by the said user memory means, each evaluation value for every area calculated by the said area evaluation means WHEREIN The configuration is such that the evaluation is improved in accordance with the low parameter accuracy of the ideal parameter.

  According to the singing scoring system configured as described above, after calculating the evaluation value for each section for the voice related to the song singing by the user, and correcting this based on the parameter accuracy of the singing user, By taking statistics, the scoring results for the final song can be calculated.

  This parameter accuracy is specified based on the music sung by the singing user in the past, and the voice of the ideal parameter can be accurately uttered for each type of ideal parameter defined in each section of the music. That is, it shows the singing ability with respect to the ideal parameter of the singing user.

  Then, since the singing ability is lower as the parameter accuracy is lower, the evaluation value for the section corresponding to the parameter accuracy having such a low singing ability tends to be relatively low, but the user having low singing ability For him, he can sing well as his singing ability. In this case, if the singing user's singing ability is used as a reference, it is not preferable that the scoring result is calculated while the evaluation value for the section remains low.

  Therefore, in the above configuration, the scoring result is calculated while the evaluation value for the section remains low by calculating the scoring result after correcting the evaluation value in such a section to be high. Can be prevented. As a result, a scoring result obtained by taking statistics on these evaluation values can be used as a scoring result in consideration of the singing ability of the singing user.

  Since it can be said that the singer's singing ability is reflected in the scoring result calculated in this way, this indicates how much the grading is based on the singing user's singing ability. It is suitable for knowing whether you were singing.

  In this configuration, the “section” in the singing parameters and ideal parameters is obtained by dividing the sound along the time series, for example, when the section is composed of one or more bars in the music or when the sound is represented by the transition of the sound. It is a section of each sound in.

  As in the latter case, in order to specify the section accuracy for each sound section in the case where the sound is represented by sound transitions, the above-described structure is configured as described below (Claim 2) and It is possible to do.

  In this configuration, the section accuracy specifying means is defined by the pitch and length of the sound in each section of each sound when the sound of the user singing the music is represented by the transition of the sound. The section accuracy is specified for each section by comparing the singing parameter with the ideal parameter in the same section.

  With this configuration, it is possible to specify the section accuracy for each section by comparing the singing parameters defined by the pitch and length of each section of the sound with the ideal parameters in the same section. .

  Moreover, in each said structure, "the audio | voice which sings a music" used when specifying area accuracy is memorize | stored in the memory | storage part with which a singing scoring system is equipped beforehand, or is singing data acquired from the outside. It should be shown.

  In this case, if the singing data can identify the user who sang the music, the sung music, and the voice singing the music, the third configuration shown below (claims) Item 3) should be used.

  In this configuration, the section accuracy specifying means is specified by the singing data based on singing data that can specify the user who sang the music, the sung music, and the voice of singing the music. For each section obtained by dividing the voice into predetermined sections along the time series, the singing parameters relating to the voice of the section are ideal parameters based on the correct voice to be uttered in the same section in the music specified by the acquired singing data; By comparing, the interval accuracy with respect to the ideal parameter for each interval is specified. Then, the user storage means provides the parameter accuracy for each type of the ideal parameter specified by the parameter accuracy specifying means to the user specified by the singing data that triggered the specification of the parameter accuracy. The data is stored in the storage unit in association with each other.

  With this configuration, the voice indicated by the singing data can be specified as “voice singing a song” and the section accuracy can be specified, and for each type of ideal parameter specified based on this section accuracy. Can be stored in the storage unit in association with the user specified by the song data.

  In addition, in this configuration, when singing a song, it is possible that the pitch (so-called key) or length (tempo) set for the song is arbitrarily changed. In addition, it is desirable to be able to specify the interval accuracy.

  As a configuration for that purpose, for example, the singing data includes not only the user who sang the song, the sung song, and the voice of singing the song, but also the pitch or length of sound predetermined for the song. In the case of data that can specify the pitch or length of the sound after the change when the song is sung, the third configuration of the fourth configuration (Claim 4) shown below is shown. It is good to do so.

  In this configuration, when the song data acquired from outside the song scoring system is song data that can specify the pitch or length of the sound after the change, Concerning each section obtained by dividing the specified sound into predetermined sections along the time series, the singing parameter relating to the sound of the section is compared with the ideal parameter when the pitch or length of the sound is changed. Thus, the section accuracy for each section is specified.

  With this configuration, even if the user has changed the pitch or length of the sound set for the song at the time of singing the song, the singing parameter of each section is changed to the pitch or length of the sound after the change. By comparing with the ideal parameter corresponding to the length, it is possible to specify an appropriate interval accuracy.

  In addition, this configuration further includes, based on the singing data, a change detection means for detecting that the user has changed the pitch or length of the music when singing the song and the amount of the change, and is provided with section accuracy identification The means compares the singing parameters relating to the sound of each section with the ideal parameter corresponding to the amount of change detected by the change detecting means (the ideal parameter in that section is changed by the amount of change) It is sufficient to specify the accuracy of the section.

  In addition, the “speech made by singing the music for scoring” used when calculating the evaluation value for each section in each of the above configurations is stored in advance in the storage unit included in the singing scoring system or acquired from the outside What is necessary is just to show with song data.

  In this case, if the singing data can identify the user who sang the scoring target music, the scoring target music that has been sung, and the voice formed by singing the scoring target music, each of the above configurations will be described below. It is preferable to use the configuration of No. 5 (Claim 5).

  In this configuration, the section evaluation means is configured to output audio indicated by singing data that can specify a user who sang the music to be scored, a music to be scored that has been sung, and a voice that sings the music to be scored. For each section when divided into predetermined sections along the series, the singing parameters relating to the sound of the section are compared with ideal parameters based on the correct sound to be uttered in the same section in the music specified by the acquired singing data As a result, the evaluation value for each section is calculated. And the said evaluation correction means is the said song data used for the evaluation among the parameter accuracy which the said user memory means memorize | stored each evaluation value for every area calculated by the said area evaluation means. The correction is made based on the parameter accuracy corresponding to the user specified in (1).

  If it is this structure, the evaluation value for every area can be calculated by making the audio | voice shown by singing data into "the audio | voice which sings a music for scoring", and this evaluation value is specified by the singing data. Correction based on the parameter accuracy corresponding to the user.

  Further, in this configuration, all the singing data acquired from the outside may be used for calculation of the evaluation value for each section, but only when the music specified by the singing data is the music to be scored. It may be used to calculate the evaluation value.

  As a configuration for this, for example, the singing data is not only the user who sang the grading target song, the singing target singing song, and the voice singing the grading target song, but also the user singing as the singing target song In the case of data that can specify whether or not it is a music piece, the fifth configuration may be the same as the sixth configuration (claim 6) shown below.

  In this configuration, the section evaluation means indicates the singing data when the singing data acquired from outside the singing scoring system is singing data that can be specified as the singing target music. An evaluation value for each section for the music is calculated using the music to be scored as the scoring music.

  With this configuration, the evaluation value for each section can be calculated only when the music specified by the singing data is the music to be scored, and the final scoring result can be calculated based on this evaluation value.

  In the fifth and sixth configurations, when a song is sung, the pitch (so-called key) or length (tempo) set for the song may be arbitrarily changed. Therefore, it is desirable to be able to specify the interval accuracy in consideration of the change.

  As a configuration for that purpose, for example, the singing data includes not only the user who sang the song, the sung song, and the voice of singing the song, but also the pitch or length of sound predetermined for the song. In the case where the singing is performed by changing the pitch, the seventh and sixth configurations shown below are used as the fifth and sixth configurations when the data can specify the pitch or length of the changed sound. ).

  In this configuration, the section evaluation means is indicated by the singing data when the singing data acquired from outside the singing scoring system is singing data that can specify the pitch or length of the sound after the change. By comparing the singing parameters relating to the sound of the section for each section when the sound is divided into predetermined sections along the time series with the ideal parameters when the sound pitch or length after the change is used The evaluation value for each section is calculated.

  With this configuration, even if the user has changed the pitch or length of the sound set for the song at the time of singing the song, the singing parameter of each section is changed to the pitch or length of the sound after the change. By comparing with the ideal parameter corresponding to the length, an evaluation value for each appropriate section can be calculated.

  In addition, since the “section accuracy” in each of the above configurations is the accuracy of the singing parameter with respect to the ideal parameter for each section, in other words, the degree of similarity of the singing parameter with respect to the ideal parameter, Can also be identified by taking statistics of similarity defined by the interval accuracy.

  As a specific statistical method, for example, it is conceivable to average each section accuracy for each type of ideal parameter. As a configuration for this purpose, for example, the above-described configuration is an eighth configuration shown below. (Claim 8) is preferable.

  In this configuration, the parameter accuracy specifying means averages each section accuracy specified by the section accuracy specifying means for each type of the ideal parameter, thereby obtaining the parameter accuracy for the corresponding user, It is specified for each type of ideal parameter.

  With this configuration, it is possible to specify the parameter accuracy for each type of ideal parameter by averaging each section accuracy for each type of ideal parameter.

  Moreover, the evaluation value for each section in each of the above-described configurations is not particularly limited as long as it is calculated by comparing the singing parameter in each section with the ideal parameter in the same section.

  For example, it is conceivable to calculate an evaluation value as a value indicating the degree of similarity or deviation of the singing parameter with respect to the ideal parameter. In these cases, the evaluation value calculated in this way is the evaluation that occupies the scoring result as the parameter accuracy is lower. After correcting so that the weighting as a value becomes light, a scoring result based on these statistics is calculated.

  As a method of taking statistics in these cases, if similarity is calculated as an evaluation value, for example, averaging the similarities may be considered, and the averaged similarity may be used as a scoring result. . Further, in the case of calculating a deviation amount as an evaluation value, for example, a value left by subtracting all evaluation values from a reference score (full score) may be used as a scoring result.

  As a specific configuration when the amount of deviation is calculated as an evaluation value as in the latter case, it is conceivable to make each of the above configurations as a ninth configuration (claim 9).

  In this configuration, the section evaluation means compares the singing parameters relating to the voices of each section with the ideal parameters in the section, and sets a value indicating the amount of deviation of the singing parameter with respect to the ideal parameter as an evaluation value for each section. The evaluation correction means calculates and corrects each evaluation value for each section calculated by the section evaluation means so as to become lower according to the parameter accuracy for the corresponding ideal parameter, and the singing scoring The means subtracts all of the evaluation values corrected by the evaluation correction means from the reference score, thereby calculating the score remaining after the subtraction as a scoring result when the singing user sings the scoring target music.

  If it is this structure, the sum total of the evaluation value which is the deviation | shift amount of the song parameter with respect to an ideal parameter can be subtracted from a reference | standard score, and the value left in that way can be calculated as a scoring result.

  In addition, the singing scoring system in each said structure may be comprised as one apparatus, and may be comprised so that the several apparatus each connected so that communication may operate | move cooperatively.

  Moreover, in order to solve the said subject, it is good also as a program (Claim 10) for making a computer system perform the various process procedures for functioning as all the means in said 1st-9th.

  If it is a computer system that executes this program, it is possible to configure a part of the singing scoring system according to any of the above first to ninth.

  The above-described program is composed of an ordered sequence of instructions suitable for processing by a computer system, and includes a singing scoring system, each server, and a user who uses this through various recording media and communication lines. Etc. are provided.

Embodiments of the present invention will be described below with reference to the drawings.
(1) Overall Configuration The karaoke system 1 is configured such that a server 2 composed of a well-known computer system and each of one or more karaoke apparatuses 3 are communicably connected via a network 100.

  The server 2 includes a control unit 21 that controls the entire server, a storage unit 23 that stores various information, a communication unit 25 that controls communication via the network 100, and a user interface (U / I) unit 27 including a keyboard and a display. , A media drive 29 for inputting / outputting information via a recording medium and the like, and a function as a difficulty level estimation system of the present invention is implemented.

The karaoke device 3 includes a control unit 31 that controls the entire device, a storage unit 33 that stores music data and video data indicating the accompaniment content and lyrics of the performance music, a communication unit 35 that controls communication via the network 100, and the like. A display unit 41 for displaying video, an operation unit 43 including a plurality of keys and switches, a voice input / output unit 49 for controlling voice input from the microphone 45 and voice output from the speaker 47, and the like are provided. .
(2) Processing by Server 2 Hereinafter, various processing executed by the control unit 21 of the server 2 according to a program stored in the built-in memory or the storage unit 23 will be described.
(2-1) History Accumulating Process First, the processing procedure of the history accumulating process started by receiving singing data from any karaoke apparatus 3 (s110) will be described with reference to FIG.

  This singing data is data that is transmitted after the user of the karaoke device 3 performs a login operation to the karaoke device 3 and then sings the music, and the singing data is in time series of voices related to the singing. It shows the transition pattern of the pitch.

  Moreover, this song data relates to the song related to the song and identification information (music number, user ID) of the user, key information indicating the key set on the karaoke device 3 side at the time of the song, and the song The score information indicating whether or not to score the music is added and transmitted.

  When this history accumulation process is activated, first, ideal data indicating the correct sound to be uttered in the song (hereinafter referred to as “song song”) based on the song number added to the song data received at the time of activation, The data is read from a plurality of types of ideal data stored in advance in the storage area for ideal data in the storage unit 23 (s120).

  Next, the ideal data read in s120 is changed based on the key information added to the singing data received when starting the history accumulation process (s130).

  Here, only when the key information added to the singing data indicates that the setting is different from a predetermined key (original music key) in the singing song, the reading is performed in the above s120. The voice indicated by the ideal data is changed.

  In this change, the ideal parameter relating to the voice to be generated in each unit section obtained by dividing the voice represented by the ideal data for each section of the sound (note in this embodiment) is a parameter corresponding to the key represented by the key information. It is realized by correcting so that Specifically, the ideal parameter in each unit section is changed to the ideal parameter when a song is played with the key setting.

  Next, by comparing the voice indicated by the singing data received at the start of the history accumulation process with the voice indicated by the ideal data read in s120 and changed in s130, the voice is Section accuracy for each unit section divided for each predetermined section along the series is specified (s140).

  Here, for each unit section of the voice indicated by the singing data, the singing parameter regarding the voice of the unit section is compared with the ideal parameter regarding the voice to be generated in the same unit section of the voice indicated by the ideal data. The accuracy of the singing parameter with respect to the ideal parameter for each unit section (hereinafter referred to as “section accuracy”) is specified.

  Specifically, the approximation degree as the pitch of the singing parameter with respect to the ideal parameter and the temporal approximation degree of the timing when the pitch changes are specified as the section accuracy in the unit section. A case where the degree of approximation is equal to or greater than a predetermined threshold is specified as a section accuracy “Yes”, and a case where the degree of approximation is less than the threshold is specified as a section accuracy “No” (see FIG. 3). Note that a value between “0” and “1” may be specified as the interval accuracy according to the degree of approximation.

  Next, the singing history data of the user corresponding to the user ID added to the singing data received when starting the history accumulating process is read from the singing history data storage area in the storage unit 23 (s150).

  This singing history data is data indicating the result of totaling the accuracy (hereinafter referred to as “parameter accuracy”) when the user registered in advance utters the voice defined by the ideal parameter for each type of ideal parameter. And stored in the storage unit 23 in a state associated with each user ID of the corresponding user (see FIG. 4). This singing history data specifies how accurately the user was able to sing a song based on the parameter accuracy of each ideal parameter. The greater the value as each parameter accuracy, the more applicable This indicates that the user can accurately sing according to the ideal parameters.

  In addition, when the user's singing history data corresponding to the corresponding user ID is not stored in the storage unit 23 at the time of s150, in s150, the singing history data having no data substance (that is, before counting) After being generated as corresponding to the corresponding user ID, this is read out.

  Then, the singing history data read in s150 is updated based on the section accuracy for each unit section specified in s140 (s160).

  Here, the section accuracy for each unit section indicated by the singing history data read in s140 is as follows. As shown in FIG. 4, the pitch (sound pitch) and tone value (sound length) are shown. For each of the ideal parameters specified in the above, the ratio between the specific number of times that the unit accuracy corresponding to the ideal parameter is specified as “Yes” and the total number of times that the accuracy of the interval is specified is calculated. It is meant to be shown.

  Therefore, in this s160, if the section accuracy for each of the unit sections specified in s140 is “Yes”, the specified number and the total number of ideal parameters corresponding to the unit section “1” is added to both (numerator and denominator), and if the interval accuracy is “No”, “1” is added only to the total number of times of the ideal parameter (denominator). Thus, each section accuracy of the corresponding user is averaged for each type of ideal parameter, and the parameter accuracy is tabulated for each type of ideal parameter for that user.

When each section accuracy is specified as a value between “0” and “1”, in this s160, the section accuracy is averaged for each type of ideal parameter, thereby obtaining the parameter accuracy. Should be tabulated.
(2-2) Singing scoring process Subsequently, the history data is received from one of the karaoke apparatuses 3 as singing data to which scoring information indicating that the music related to the singing is to be scored (s210). The procedure of the singing scoring process executed in parallel with the accumulation process will be described with reference to FIG.

  When this singing scoring process is activated, first, it is stored in the storage unit 23 in association with the user of the user ID (hereinafter referred to as “singing user”) added to the singing data received prior to the activation of this singing scoring process. The singing history data being read is read (s210).

  Next, based on the song number of the song added to the song data received when starting the singing scoring process, ideal data indicating the correct voice to be uttered in the song (hereinafter referred to as “scoring song”) is stored in the storage unit. 23 is read out from a plurality of types of ideal data stored in advance in the ideal data storage area 23 (s220).

  Next, the ideal data read in s220 is changed based on the key information added to the singing data received when the singing grading process is started (s230).

  Here, only when the key information added to the song data indicates a setting different from a predetermined key (original song key) in the song song, it is read out in s220. The voice indicated by the ideal data is changed.

  In this change, the ideal parameter relating to the voice to be generated in each unit section obtained by dividing the voice represented by the ideal data for each section of the sound (note in this embodiment) is a parameter corresponding to the key represented by the key information. It is realized by correcting so that Specifically, the ideal parameter in each unit section is changed to the ideal parameter when a song is played with the key setting.

  Next, by comparing the voice indicated by the singing data received when starting the singing scoring process with the voice indicated by the ideal data read in s220 and changed in s230, the voice is An evaluation value for each unit section divided for each predetermined section along the series is specified (s240).

  Here, for each unit section of the voice indicated by the singing data, the singing parameter regarding the voice of the unit section is compared with the ideal parameter regarding the voice to be generated in the same unit section of the voice indicated by the ideal data. An evaluation value for each unit section is calculated. In the present embodiment, in each unit section i (i = 1 to I), a value indicating the deviation amount of the singing parameter with respect to the ideal parameter is specified as the evaluation value di.

  In s240, a value indicating the similarity of the singing parameter to the ideal parameter in each unit section i may be specified as an evaluation value di indicating that the higher the similarity is, the higher the evaluation is.

  Next, each evaluation value calculated in s240 is corrected based on the parameter accuracy indicated by the singing history data read in s212 (s250).

  Here, each evaluation value di for each section calculated in s240 is weighted according to the parameter accuracy P (i) for the ideal parameter in the section among the parameter accuracy of the singing user. Specifically, the evaluation value di is corrected so that the weight becomes smaller as the parameter accuracy P (i) corresponding to the evaluation value di is smaller, that is, the singing according to the ideal parameter cannot be performed accurately. . If the deviation amount of the singing parameter with respect to the ideal parameter is calculated as the evaluation value di as in the present embodiment, the value indicating the product of the evaluation value di and the corresponding parameter accuracy P (i) (= P (i) · di).

  Next, for each evaluation value di corrected in the above s250, a value obtained by statistic over the entire section of the scoring target music is calculated as a scoring result when the singing user sings the scoring target music (s260). .

  Here, by subtracting all the evaluation values corrected by the above 250 from the reference score 100, the score remaining after the subtraction is calculated as a scoring result when the singing user sings the scoring target music. .

  As a specific calculation method, when each evaluation value is di, the parameter accuracy for the ideal parameter in the section corresponding to the evaluation value is P (i), and the reference score is 100, the following equation is used: It is conceivable to calculate by 1.

なお、評価値ｄｉとして類似度を特定するように構成した場合には、全ての評価値ｄｉを平均化した値を採点結果として算出することとすればよい。

When the similarity is specified as the evaluation value di, a value obtained by averaging all the evaluation values di may be calculated as a scoring result.

And after the scoring result calculated in said s260 is returned to the karaoke apparatus 3 which has transmitted song data (s270), this singing scoring process is complete | finished. In the karaoke apparatus 3 that has received the scoring result, the scoring result is displayed on the display unit 41.
(3) Action and Effect According to the server 2 of the karaoke system 1 configured as described above, an evaluation value is calculated for each section of the voice related to the song sung by the user (s240 in FIG. 5), and this is the singing user. After correcting based on the parameter accuracy of (No. s250 in the figure), by taking statistics about each evaluation value, it is possible to calculate the scoring result for the final singing (s260 in the figure).

  This parameter accuracy is specified based on the music sung by the singing user in the past (s110 to s160 in FIG. 2), and is ideal for each type of ideal parameter defined in each section of the music. It shows whether the voice of the parameter can be uttered correctly, that is, the singing ability for the ideal parameter of the singing user (see FIG. 3).

  Then, since the singing ability is lower as the parameter accuracy is lower, the evaluation value for the section corresponding to the parameter accuracy having such a low singing ability tends to be relatively low, but the user having low singing ability For him, he can sing well as his singing ability. In this case, if the singing user's singing ability is used as a reference, it is not preferable that the scoring result is calculated while the evaluation value for the section remains low.

  Therefore, in the above configuration, by correcting the evaluation value in such a section so that the evaluation value becomes high and calculating the scoring result (s250 and s260 in FIG. 5), the evaluation value for that section remains low. It is possible to prevent the scoring result from being calculated. As a result, a scoring result obtained by taking statistics on these evaluation values can be used as a scoring result in consideration of the singing ability of the singing user.

  Since it can be said that the singer's singing ability is reflected in the scoring result calculated in this way, this indicates how much the grading is based on the singing user's singing ability. It is suitable for knowing whether you were singing.

  Further, in the above embodiment, the section accuracy for each section is specified by comparing the singing parameters defined by the pitch and length of each section of the sound with the ideal parameters in the same section. (S140 in FIG. 2).

  Moreover, in the said embodiment, the area | region accuracy can be specified as the audio | voice shown by song data as "the audio | voice which sings a music" (s140 of FIG. 2), and based on this area accuracy. The parameter accuracy for each of the identified ideal parameter types can be stored in the storage unit 23 (update the singing history data) in association with the user specified by the singing data (s160 in FIG. 2).

  Moreover, in the said embodiment, even if the user has changed the pitch or length of the sound set for the song at the time of singing the song, the singing parameter of each section is changed to the pitch of the sound after the change. Alternatively, since it can be changed to an ideal parameter corresponding to the length (s130 in FIG. 2), appropriate section accuracy can be specified by comparing with the ideal parameter thus changed (s140 in FIG. 2).

  Moreover, in the said embodiment, the audio | voice shown by song data can specify the evaluation value for every area as "the audio | voice formed by singing a music for scoring" (s240 of FIG. 5), and this evaluation value Can be corrected based on the parameter accuracy corresponding to the user specified by the song data (s250 in the figure).

  Moreover, in the said embodiment, a singing scoring process is started only when the music specified by song data is a music for scoring (refer to Drawing 5), and final scoring is performed based on the evaluation value computed for every section. The result can be calculated.

  Moreover, in the said embodiment, even if the user has changed the pitch or length of the sound set for the music at the time of singing the scoring target music, the singing parameters of each section are changed to the sound parameters after the change. Since it can be changed to an ideal parameter corresponding to the height or length (s230 in FIG. 5), an evaluation value for each appropriate section can be calculated by comparing with the ideal parameter thus changed (same as above). Figure s240).

  Further, in the above embodiment, the parameter accuracy for each type of ideal parameter can be specified by averaging each section accuracy for each type of ideal parameter (s160 in FIG. 2).

Moreover, in the said embodiment, the sum total of the evaluation value which is the deviation | shift amount of the singing parameter with respect to an ideal parameter is subtracted from a reference | standard score, and the remaining value can be calculated as a scoring result (s260 of FIG. 5). ).
(4) Modifications The embodiment of the present invention has been described above. However, the present invention is not limited to the above embodiment, and can take various forms as long as they belong to the technical scope of the present invention. Needless to say.

  For example, in the said embodiment, although the case where the server 2 of the karaoke system 1 was comprised as a song scoring system of this invention was illustrated, a part of process by this server 2, for example, a part or all of a history storage process It goes without saying that a part or all of the singing grading process can be performed in cooperation with other devices such as the karaoke device 3 so as to function as a singing grading system as a whole.

  Moreover, in the said embodiment, the case where the "unit area" in a song parameter and an ideal parameter divided | segmented the sound prescribed | regulated by a musical note along the time series was illustrated. However, the “unit section” may be, for example, a section composed of one or more bars in the music.

Moreover, in the said embodiment, what was comprised so that a log | history storage process (FIG. 2) and a singing scoring process (FIG. 5) might be performed when the singing data from the karaoke apparatus 3 was received was illustrated. However, these processes may be configured to be started at the timing of receiving an external command caused by a user operation or the like based on song data stored in advance in the storage unit 23 or the like.
(5) Correspondence with the present invention In the embodiment described above, the server 2 corresponds to the singing scoring system in the present invention, and s140 in FIG. 2 corresponds to the section accuracy specifying means in the present invention. s160 corresponds to parameter accuracy specifying means user storage means in the present invention, s240 in FIG. 5 corresponds to section evaluation means in the present invention, and s250 in FIG. 5 corresponds to evaluation correction means in the present invention. FIG. S260 corresponds to the singing scoring means in the present invention.

Block diagram showing the overall configuration of the karaoke system Flow chart showing history accumulation processing The figure which showed notionally how to specify the section accuracy for each unit section The figure which shows the data structure of singing history data Flow chart showing singing scoring process

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Karaoke system, 2 ... Server, 3 ... Karaoke apparatus, 21 ... Control part, 23 ... Memory | storage part, 25 ... Communication part, 27 ... part, 29 ... Media drive, 31 ... Control part, 33 ... Memory | storage part, 35 ... Communication unit 41... Display unit 43. Operation unit 45. Microphone 47 47 Speaker 49 49 Voice input / output unit 100 Network

Claims (10)

Contrast the singing parameters related to the voice of the section with the ideal parameters based on the correct voice to be uttered in the section for each section obtained by dividing the voice of the song sung by the user for each predetermined section along the time series. Section accuracy specifying means for specifying the accuracy (section accuracy) of the singing parameter with respect to the ideal parameter for each section;
For each section accuracy specified by the section accuracy specifying means, by taking statistics for each type of the ideal parameter, the accuracy (parameter accuracy) when the corresponding user utters the voice defined by the ideal parameter. Parameter accuracy specifying means for specifying for each type of the ideal parameter,
User memory for storing the parameter accuracy for each of the types of ideal parameters specified by the parameter accuracy specifying means in the storage unit in association with the user who sang the music that triggered the specification of the parameter accuracy Means,
For each of the sections when the user sings the singing target music (scoring target music) for each predetermined section along the time series, the singing parameters relating to the sound of the section are the ideal parameters in the section. By comparing, an interval evaluation means for calculating an evaluation value for each interval,
Each of the evaluation values for each section calculated by the section evaluation means corresponds to the singing user who sang the music to be scored, which is the trigger for the evaluation, out of the parameter accuracy stored in the storage unit by the user storage means Evaluation correction means for correcting based on the parameter accuracy to be performed,
Singing scoring means for calculating a scoring result when the singing user has sung the scoring target music by taking statistics for all sections of the scoring target music for each of the evaluation values after correction by the evaluation correcting means, With
The evaluation correction means includes the ideal parameter in the section among the parameter accuracy of the singing user stored in the storage unit by the user storage means for each evaluation value calculated by the section evaluation means. The singing scoring system is configured to be modified so that the evaluation becomes higher according to the low accuracy of the parameter. The section accuracy specifying means, for each sound section in the case where the sound of the user singing the music is represented by sound transition, singing parameters defined by the pitch and length of the sound in the section The singing scoring system according to claim 1, wherein the section accuracy for each section is specified by comparing with ideal parameters in the same section. The section accuracy specifying means is based on song data that can specify a user who sang a song, a song sung, and a voice formed by singing the song. For each of the sections divided into predetermined sections along the line, the singing parameters relating to the sound of the section are compared with ideal parameters based on the correct sound to be uttered in the same section in the music specified by the acquired singing data. Identifying the interval accuracy for the ideal parameter for each interval;
The user storage means associates the parameter accuracy for each type of the ideal parameter specified by the parameter accuracy specifying means with the user specified by the singing data that triggered the parameter accuracy. The singing scoring system according to claim 1, wherein the singing scoring system is stored in a storage unit. The singing data was sung by changing not only the user who sang the song, the sung song, and the voice of singing the song, but also the predetermined pitch or length of the song. In the case of data that can specify the pitch or length of the sound after the change in the case,
When the song data acquired from the outside of the song scoring system is song data that can specify the pitch or length of the sound after the change, the section accuracy specifying means is a voice specified by the song data. For each section divided into predetermined sections along the time series, the singing parameters relating to the sound of the section are compared with the ideal parameters in the case where the pitch or length of the sound after the change is made. The singing scoring system according to claim 3, wherein the section accuracy is specified. The section evaluation means includes, in chronological order, voices indicated by singing data that can specify a user who sang the music to be scored, a music to be scored that has been sung, and a voice that sings the music to be scored For each section in the case of dividing every predetermined section, by comparing the singing parameters relating to the sound of the section with ideal parameters based on the correct sound to be uttered in the same section in the music specified by the acquired singing data, The evaluation value for each section is calculated,
The evaluation correction means specifies each evaluation value for each section calculated by the section evaluation means with the singing data used for the evaluation out of the parameter accuracy stored in the storage unit by the user storage means. It correct | amends based on the parameter accuracy corresponding to the user to be performed. The singing scoring system in any one of Claim 1 to 4 characterized by the above-mentioned. Whether the singing data is not only the user who sang the grading music, the singing music sung, and the voice singing the singing music, but also the music sung by the user as the grading music Is the data that can be specified,
When the singing data acquired from outside the singing scoring system is singing data that can specify that the singing music is sung as the scoring target music, the section evaluation means scores the music indicated by the singing data. The singing scoring system according to claim 5, wherein an evaluation value for each section of the music is calculated as the target music. The singing data is not only the user who sang the music to be scored, the music to be sung, and the voice singing the music to be scored, but also the pitch or length of the sound determined in advance for the music If it is data that can specify the pitch or length of the sound after the change when singing
The section evaluation means, when singing data acquired from outside the singing scoring system is singing data that can specify the pitch or length of the changed sound, the voice indicated by the singing data is time-series For each section when divided into predetermined sections along the line, the singing parameters relating to the sound of the section are compared with the ideal parameters when the pitch or the length of the sound after the change is made. An evaluation value is calculated. The singing scoring system according to claim 5 or 6 characterized by things. The parameter accuracy specifying means averages each section accuracy specified by the section accuracy specifying means for each type of the ideal parameter, thereby obtaining the parameter accuracy of the user in question for the ideal parameter. It specifies for every kind. The singing scoring system in any one of Claim 1 to 7 characterized by the above-mentioned. The section evaluation means calculates, for each section, a value indicating a deviation amount of the singing parameter with respect to the ideal parameter as an evaluation value by comparing the singing parameter related to the sound of each section with the ideal parameter in the section.
The evaluation correction means corrects each evaluation value for each section calculated by the section evaluation means so as to be low according to the parameter accuracy for the corresponding ideal parameter,
The singing scoring means subtracts all the evaluation values after the correction by the evaluation correcting means from the reference score, thereby obtaining the score remaining after the subtraction as a scoring result when the singing user sings the scoring target music The singing scoring system according to claim 1, wherein the singing scoring system is calculated.   A program for causing a computer system to execute various processing procedures for causing all of the means according to any one of claims 1 to 9 to function.

Images