JP5428458B2 - Evaluation device - Google Patents

Description

  The present invention relates to a technique for evaluating a song.

  A technique for scoring the skill of a user's sing performed using a karaoke device is known. For example, Patent Documents 1 to 3 disclose a technique for scoring a user's singing in a karaoke apparatus that reproduces a karaoke performance. Patent Documents 4 and 5 disclose a technique in which a teacher reproduces a student's singing voice using a computer device, corrects the student's singing voice, and transmits the correction result to the student. Moreover, in patent document 6, when the server of a communication karaoke system distributes karaoke data to the communication terminal which a singer uses, a singer's singing voice signal is received from a communication terminal, and distribution of karaoke data is complete | finished. A technique for determining a singer's singing ability based on a received singing voice signal and transmitting the determination result to a communication terminal is disclosed.

JP 2003-216168 A JP 2007-121550 A Japanese Patent Laid-Open No. 10-222182 Japanese Patent No. 4087087 JP 2003-15673 A JP 2007-17849 A

By the way, if the scoring function which scores a user's song is provided in each karaoke apparatus like patent documents 1-3 mentioned above, the cost of a karaoke apparatus will become high. In addition, since the karaoke apparatus must perform processing such as reproduction of karaoke performance, it cannot perform precise scoring processing with a high processing load. Furthermore, when adding a new scoring function, it is necessary to add a new scoring function to each karaoke device, which takes time and effort.
An object of the present invention is to evaluate a user's singing voice for each evaluation section in an apparatus other than a communication apparatus that performs processing such as playback of a karaoke performance, and sequentially output the evaluation result.

  The present invention includes a first storage unit that stores exemplary voice data representing an exemplary singing voice, an acquisition unit that acquires voice data representing a user's singing voice from a communication device in a streaming manner, and the acquisition unit. Division which divides | segments the time axis | shaft of the said audio | voice data memorize | stored in the 2nd storage means which memorize | stores the acquired audio | voice data, and the said 2nd memory | storage means into the several evaluation area used as the object of evaluation of a song voice And each evaluation section divided by the dividing means in the order corresponding to the position on the time axis, and voice data included in the selected evaluation section is stored in the first storage means. The user's singing voice in the selected evaluation section is evaluated by comparing with the model voice data included in the section corresponding to the evaluation section, and the evaluation result is generated. And means, every time the evaluation results of each evaluation interval is generated by the evaluation unit, generated the evaluation results provide an evaluation device, characterized in that it comprises a transmitting means for transmitting to the communication device.

  In a preferred aspect of the present invention, delimiter data indicating delimiter positions on the time axis of each phrase included in the exemplary singing voice is added to the exemplary voice data, and the dividing means includes the second voice data. The time axis of the voice data stored in the storage means is divided into a plurality of evaluation sections corresponding to each phrase section delimited by the delimiter data in the model voice data stored in the first storage means. May be.

  In a preferred aspect of the present invention, there are a plurality of communication devices of different types, and the evaluation unit determines the predetermined data when the acquisition unit acquires the audio data from a predetermined type of communication device. The evaluation may be performed with simpler content than when the audio data is acquired from a communication device of a type different from the specified type.

  ADVANTAGE OF THE INVENTION According to this invention, in apparatuses other than a communication apparatus which performs processes, such as reproduction | regeneration of a karaoke performance, a user's song voice can be evaluated for every evaluation area and the evaluation result can be output sequentially.

It is a block diagram which shows the structure of the scoring system which concerns on embodiment. It is a block diagram which shows the structure of the karaoke apparatus of the said scoring system. It is a block diagram which shows the structure of the scoring server apparatus of the said scoring system. It is a sequence diagram which shows operation | movement of the said scoring system. It is a figure which shows an example of the song data and model song data which the said scoring system uses.

[Constitution]
FIG. 1 is a block diagram showing a configuration of a scoring system 1 according to this embodiment. As shown in the figure, the scoring system 1 includes a plurality of karaoke apparatuses 10 and a scoring server apparatus 20. The scoring server device 20 and the karaoke device 10 are connected via a network N such as the Internet. This karaoke apparatus 10 functions as a communication apparatus according to the present invention. The scoring server device 20 functions as an evaluation device according to the present invention.

(Karaoke equipment)
Next, the configuration of the karaoke apparatus 10 will be described. This karaoke apparatus 10 is installed in a karaoke shop, for example. FIG. 2 is a block diagram showing the configuration of the karaoke apparatus 10. As shown in the figure, the karaoke apparatus 10 includes a CPU (Central Processing Unit) 11, a memory 12, a communication unit 13, a storage unit 14, an operation unit 15, a display unit 16, and a sound collection unit 17. The sound source unit 18 and the sound emitting unit 19 are provided. The CPU 11 controls each unit of the karaoke apparatus 10 by executing a program stored in the memory 12. The memory 12 includes, for example, a ROM (Read Only Memory) and a RAM (Random Access Memory), and stores programs and data used by the CPU 11. The communication unit 13 communicates with the scoring server device 20 connected via the network N. The memory | storage part 14 is provided with the hard disk, for example, and has memorize | stored apparatus ID used in order to identify each karaoke apparatus 10, and several karaoke data Dk used when a user sings. The karaoke data Dk includes performance data and lyrics data. The performance data is data representing the performance sound of the music. This performance sound does not include singing voice. The lyric data is data indicating the lyrics of the music. The operation unit 15 includes, for example, a plurality of operation buttons, and inputs an operation signal according to a user operation to the CPU 11. The display unit 16 includes a liquid crystal display, for example, and displays an image according to the control of the CPU 11. The sound collection unit 17 includes, for example, a microphone and an A / D conversion unit, generates an analog signal corresponding to the collected voice, converts the generated analog signal into digital data by A / D conversion, and outputs the digital data. . The sound source unit 18 generates a sound signal corresponding to the performance data stored in the storage unit 14 and supplies the sound signal to the sound emitting unit 19. The sound emitting unit 19 includes, for example, a speaker, a D / A conversion unit, and an amplifier. The sound signal supplied from the sound source unit 18 is converted into an analog signal by D / A conversion, and the converted analog signal is amplified. Release as audio.

(Scoring server device)
Next, the configuration of the scoring server device 20 will be described. The scoring server device 20 may be installed in the same karaoke store as the karaoke device 10 or may be installed in a place different from the karaoke store. FIG. 3 is a block diagram illustrating a configuration of the scoring server device 20. As shown in the figure, the scoring server device 20 includes a CPU 21, a memory 22, a communication unit 23, a storage unit 24, and a processing unit 25. The CPU 21 controls each unit of the scoring server device 20 by executing a program stored in the memory 22. The memory 22 includes, for example, a ROM and a RAM, and stores programs and data used by the CPU 21. The communication unit 23 communicates with each karaoke apparatus 10 connected via the network N. The storage unit 24 includes, for example, a hard disk, and stores a plurality of model song data Dr representing model song voices. That is, this memory | storage part 24 functions as a 1st memory | storage means which memorize | stores model audio | voice data (model singing data) showing the singing voice used as model. Separation data k indicating the separation position on the time axis of each phrase included in the exemplary singing voice is added to the exemplary singing data Dr. The processing unit 25 is a DSP (Digital Signal Processor), for example, and compares the singing data Dx transmitted from the karaoke apparatus 10 with the exemplary singing data Dr stored in the storage unit 24, so that the singing of the user is performed. Perform scoring to evaluate the skill of the

[Operation]
Next, the operation of the scoring system 1 according to this embodiment will be described. FIG. 4 is a sequence diagram showing the operation of the scoring system 1. First, the user operates the operation unit 15 of the karaoke apparatus 10 to input a song ID of a song to be sung and instruct to start performance. As the music ID, for example, the name of the music or the number assigned to the music is used. In response to this operation, the CPU 11 establishes communication with the scoring server device 20 by the communication unit 13 (step S11), and then the music ID input by the user and the device ID stored in the storage unit 14 Is transmitted to the scoring server device 20 (step S12). At this time, the communication unit 13 communicates with the scoring server device 20 using a public API (Application Programming Interface). If music ID and apparatus ID are transmitted from the karaoke apparatus 10, CPU21 of the scoring server apparatus 20 will receive this music ID and apparatus ID by the communication part 23 (step S13).

  Further, the CPU 11 of the karaoke apparatus 10 sequentially reads out the karaoke data Dk of the music having the inputted music ID from the storage unit 14 and performs karaoke reproduction processing (step S14). Specifically, the CPU 11 supplies performance data included in the karaoke data Dk read from the storage unit 14 to the sound source unit 18. The sound source unit 18 generates an audio signal corresponding to the performance data supplied by the CPU 11, supplies the generated audio signal to the sound emitting unit 19, and emits sound corresponding to the audio signal. Thereby, the performance of the music designated by the user is started. Further, the CPU 11 supplies the lyric data included in the karaoke data Dk read from the storage unit 14 to the display unit 16 so as to be synchronized with the performance of the music. Thereby, the lyrics of the music designated by the user are displayed.

  The user sings toward the sound collection unit 17 in accordance with the performance of the music. At this time, the sound collection unit 17 collects the user's singing voice and generates singing data Dx representing the collected singing voice. CPU11 supplies the song data Dx produced | generated by the sound collection part 17 to the sound emission part 19 one by one, and releases the song sound according to the song data Dx with the performance of a music. Further, the CPU 11 sequentially supplies the singing data Dx generated by the sound collection unit 17 to the communication unit 13. The communication unit 13 transmits the song data Dx supplied by the CPU 11 to the scoring server device 20 in a streaming manner (step S15). In streaming data transmission, data is divided and transmitted in a certain PDU (Protocol Data Unit) unit (for example, packet unit). Therefore, the data receiving side can receive and process data in units of PDUs without waiting for the completion of reception of the entire data.

  When the song data Dx is transmitted from the karaoke device 10, the CPU 21 of the scoring server device 20 receives the song data Dx by the communication unit 23 (step S <b> 16), and stores the received song data in the memory 22. That is, CPU21 functions as an acquisition means which acquires audio | speech data (singing data) showing a user's song voice from the karaoke apparatus 10 by a streaming system. Further, the memory 22 functions as a second storage unit that stores audio data acquired by the CPU 21. As described above, this song data Dx is transmitted in units of PDUs. Therefore, the singing data transmitted in units of PDUs is sequentially stored in the memory 22.

  Subsequently, the processing unit 25 divides the time axis of the song data Dx stored in the memory 22 into a plurality of scoring sections to be subjected to scoring processing (step S17). This scoring section is used as an evaluation section that is a target of evaluation of the user's singing voice. That is, the processing unit 25 functions as a dividing unit that divides the time axis of the audio data stored in the memory 22 into a plurality of evaluation sections to be evaluated for singing voice. Specifically, with reference to FIG. 5, the processing unit 25 firstly includes phrase sections f1, f2, f3, f4,... Delimited by the delimiter data k in the model song data Dr stored in the storage unit 24.・ Identify Subsequently, the processing unit 25 divides the time axis of the song data Dx stored in the memory 22 into scoring sections corresponding to the specified phrase sections. In this example, the time axis of the singing data Dx is the scoring section h1 corresponding to the phrase section f1, the scoring section h2 corresponding to the phrase section f2, the scoring section h3 corresponding to the phrase section f3, and the scoring section corresponding to the phrase section f4. It is divided into h4. That is, the processing unit 25 sets the time axis of the audio data stored in the memory 22 to a plurality of evaluation intervals corresponding to each phrase interval divided by the delimiter data in the model audio data stored in the storage unit 24. To divide.

  Subsequently, the processing unit 25 selects a scoring section that is a target of the current scoring process from the divided scoring sections (step S18). In this example, since no scoring process is performed for any scoring section, the first scoring section h1 on the time axis is selected. Subsequently, the processing unit 25 performs a scoring process based on the song data Dx included in the selected scoring section and the model song data Dr stored in the storage unit 24 (step S19). Specifically, the processing unit 25 compares the singing data Dx included in the selected scoring section with the model singing data Dr included in the phrase section corresponding to the scoring section, and responds to their similarity. Calculate the score. Subsequently, the processing unit 25 uses the calculated score as a scoring result, and generates scoring result data representing the scoring result. That is, the processing unit 25 converts the voice data included in the evaluation section selected from the plurality of divided evaluation sections into the model voice data included in the section corresponding to the evaluation section stored in the storage unit 24. By comparing with, the user's singing voice in the selected evaluation section is evaluated and functions as an evaluation means for generating an evaluation result. In this example, the score corresponding to the similarity between the singing data Dx included in the scoring section h1 in FIG. 5 and the exemplary singing data Dr included in the phrase section f1 is calculated, and the calculated score is the scoring of the scoring section h1. As a result, scoring result data representing the scoring result of scoring section h1 is generated.

  When the scoring result data is generated, the CPU 21 transmits the scoring result data to the karaoke apparatus 10 having the apparatus ID received in step S16 described above by the communication unit 23 (step S20). In this example, scoring result data representing the scoring result of the scoring section h1 in FIG. Subsequently, the processing unit 25 determines whether or not the scoring section selected in Step S18 described above is the last scoring section on the time axis (Step S21). In this example, as shown in FIG. 5, since the scoring section h1 is not the last scoring section on the time axis, the processing unit 25 determines that the selected scoring section is not the last scoring section (step S21: NO). ). In this case, the processing unit 25 returns to step S18 described above, and selects a scoring section to be subjected to the next scoring process. Specifically, since the scoring process of the scoring section h1 is performed as described above, the processing unit 25 selects the scoring section h2 next to the scoring section h1 on the time axis. That is, the processing unit 25 selects each divided evaluation section in the order corresponding to the position on the time axis. Subsequently, the processing unit 25 performs the scoring process for the scoring section h2 in step S19 as described above. And CPU21 transmits the grading result data showing the scoring result of scoring area h2 to the karaoke apparatus 10 in step S20. That is, the CPU 21 functions as a transmission unit that transmits the generated evaluation result to the karaoke apparatus 10 every time the processing unit 25 generates an evaluation result in each evaluation section. In this way, the processing unit 25 and the CPU 21 repeat the above-described processing of step S18 to step S21 until it is determined in step S21 that the selected scoring section is the last scoring section on the time axis.

  On the other hand, the CPU 11 of the karaoke apparatus 10 stands by until the scoring result data is transmitted from the scoring server apparatus 20 while performing the processes of steps S14 and S15 described above. When the scoring result data is transmitted from the scoring server device 10, the CPU 11 receives the scoring result data by the communication unit 13 (step S22). Subsequently, the CPU 11 displays the scoring result represented by the received scoring result data on the display unit 16 (step S23). For example, as described above, when scoring result data representing the scoring result of the scoring section h1 in FIG. 5 is transmitted from the scoring server device 20, the scoring result of the scoring section h1 is displayed. In addition, since the karaoke reproduction | regeneration processing mentioned above continues while these steps S22-S23 are performed, the lyrics of a music are displayed on the display part 16. FIG. Therefore, when displaying the scoring result in the above-described step S23, the CPU 11 interrupts the display of the lyrics of the music and displays the scoring result or displays the composite of the lyrics of the music and the scoring result. .

  Subsequently, the CPU 11 determines whether or not the user's singing has been completed based on the situation of the karaoke reproduction process of step S14 described above (step S24). For example, when the karaoke playback process has not ended, the CPU 11 determines that the user's singing has not ended (step S24: NO). In this case, the CPU 11 returns to step S22 described above and waits until new scoring result data is transmitted. Then, when new scoring result data is transmitted from the scoring server device 20, the CPU 11 receives the scoring result data by the communication unit 13 in step S21 as described above, and the scoring result in step S23. The scoring result represented by the data is displayed on the display unit 16. Accordingly, the scoring results of the scoring sections in FIG. 5 are sequentially displayed on the display unit 16.

  On the other hand, when the karaoke data Dk of the music designated by the user is read and processed up to the last position on the time axis in the above-described step S14, the karaoke playback process ends. In this case, CPU11 determines with a user's song having been complete | finished (step S24: YES), and after cut | disconnecting communication with the scoring server apparatus 20 by the communication part 13 (step S25), this process is complete | finished. .

  According to the embodiment described above, in the scoring server device 20 that is different from the karaoke device 10 that performs the reproduction processing of the karaoke performance, the user's singing voice is evaluated for each evaluation section, and the evaluation results are sequentially output. it can. Moreover, since it is not necessary to provide each karaoke apparatus 10 with the scoring function which marks a user's song, the cost of the karaoke apparatus 10 can be reduced. Furthermore, since the karaoke performance playback process is not performed on the scoring server device 20 side, a precise scoring process with a high processing load can be performed. In addition, when a new scoring function is added, it is only necessary to add a new scoring function only to the scoring server device 20, and therefore, an operation of adding a new scoring function can be easily performed.

[Modification]
The above is the description of the embodiment, but the contents of this embodiment can be modified as follows. Further, the following modifications may be combined as appropriate.
(Modification 1)
In the embodiment described above, the time axis of the song data is divided into a plurality of scoring sections corresponding to each phrase section of the model song data, but the section used for dividing the scoring section is not limited to the phrase section. For example, the time axis of the song data may be divided into a plurality of scoring sections corresponding to the sections on the time axis of each melody included in the song data. Alternatively, the time axis of the singing data may be divided into a plurality of scoring sections corresponding to sections in which no breathing is performed in the singing data. The section where the breathing is not performed may be specified based on, for example, analyzing the song data, detecting the timing when the breathing is performed, and detecting the timing of the breathing. Alternatively, when the position on the time axis where breathing is performed in the song data is determined in advance, the position may be specified based on the timing of breathing.
Moreover, the karaoke apparatus 10 may designate a scoring section. In this case, the CPU 11 of the karaoke apparatus 10 transmits scoring section information representing the scoring section to the scoring server apparatus 20 together with the music ID and the apparatus ID described above. And the process part 25 of the scoring server apparatus 20 divides | segments the time axis | shaft of song data based on the scoring area which the scoring area information transmitted from the karaoke apparatus 10 represents.

(Modification 2)
In the above-described embodiment, for example, when the song to be sung is composed of No. 1 and No. 2, the first singing is scored as a whole when the first singing is completed. It may be. In this case, delimiter data representing delimiter positions on the time axis between the first and second songs are added to the song data. When the processing unit 25 of the scoring server device 20 finishes the scoring process of the scoring section including the delimiter data, the section from the position where the singing is started to the position where the first singing is finished is performed on the time axis of the singing data. The scoring section is set, and the scoring process similar to that described above is performed for the set scoring section.
Moreover, you may make it perform the scoring process of the whole song when a user's song is complete | finished. In this case, when the processing unit 25 ends the scoring process of the scoring section including the end of the singing data, the section from the position where the singing starts to the position where the singing starts is set as the scoring section on the time axis of the singing data, The same scoring process as described above is performed for the set scoring section.

  Furthermore, when scoring the entire song, the content of the scoring process may be changed. For example, in the scoring process of each evaluation section, only the score representing the similarity between the song data and the model song data is calculated, and in the scoring process for the entire song, the similarity between the song data and the model song data is calculated. In addition to the number of points to be represented, the number of times that a singing technique such as “fist” or “shakuri” is used may be calculated. In addition to scoring according to the degree of similarity, "Vibrato", "Running", "Tame", "Intonation", "Performance time", "Clearness of bass and treble", "Duet match", etc. However, points may be added. In the scoring process for the entire song, the above-described score may be calculated based on whether or not an appropriate singing technique is used at an appropriate timing.

(Modification 3)
In the above-described embodiment, the singing data is transmitted to the scoring server device 20 as it is, but only the feature amount of the singing data may be transmitted to the scoring server device 20 instead of the singing data. As this feature amount, for example, frequency characteristics, pitch or rhythm of song data is used. In this case, the CPU 11 of the karaoke apparatus 10 extracts feature amounts from the song data generated by the sound collection unit 17, and transmits the extracted feature amounts to the scoring server device 20 by the communication unit 13. Then, the CPU 21 and the processing unit 25 of the scoring server device 20 perform the same processing as described above by using the feature amount transmitted from the karaoke device 10.

  Moreover, in the middle of a user's song, the scoring process may be performed based on the feature amount of the song data, and when the user's song is finished, the scoring process for the entire song may be performed based on the song data. In this case, the CPU 11 of the karaoke device 10 extracts the feature amount from the song data and transmits the extracted feature amount to the scoring server device 20 until the user's singing is completed. And the process part 25 of the scoring server apparatus 20 performs the process similar to the above using the feature-value of the song data transmitted from the karaoke apparatus 10. FIG. As described above, when the scoring process is performed using only the feature amount of the singing data, since the load of the scoring process is reduced, the time required for the scoring process is shortened. Therefore, the scoring result data can be quickly transmitted to the karaoke apparatus 10. On the other hand, when the user's singing is finished, the CPU 11 of the karaoke device 10 transmits the singing data itself to the scoring server device 20. When the singing data is transmitted from the karaoke device 10, the processing unit 25 of the scoring server device 20 sets all the sections on the time axis of the singing data as scoring sections, and the set scoring sections are the same as described above. Perform scoring. Thus, when scoring processing is performed using the song data itself, detailed evaluation can be performed in scoring processing. Therefore, the user can know the detailed scoring results.

Moreover, CPU11 of the karaoke apparatus 10 may detect the state of the network N, and may transmit the feature-value of song data instead of song data, when the usable band of the network N is a narrow state. In this case, the CPU 21 and the processing unit 25 of the scoring server device 20 perform the same processing as described above using the feature amount of the song data transmitted from the karaoke device 10.
Moreover, the processing part 25 of the scoring server apparatus 20 extracts the feature-value from the song data memorize | stored in the memory 22, when communication is established between several karaoke apparatuses 10, The extracted feature-value The same processing as described above may be performed using. This is because, when communication is established with a plurality of karaoke apparatuses 10, the singing data transmitted from each karaoke apparatus 10 is scored, so the processing load of the processing unit 25 is reduced. This is because it needs to be kept low.

(Modification 4)
In embodiment mentioned above, CPU21 of the scoring server apparatus 20 may charge with respect to the user who performed the song. In this case, the user operates the operation unit 15 and inputs his / her user ID. In response to this operation, the CPU 11 transmits the input user ID to the scoring server device 20 through the communication unit 13. Then, the CPU 21 of the scoring server device 20 calculates an amount corresponding to the above-described calculation amount of the scoring process, and the calculated amount is charged to the user represented by the user ID transmitted from the karaoke device 10. The billing process is performed as follows. Alternatively, the CPU 21 may perform a charging process so that a predetermined amount is charged to the user.

(Modification 5)
In the embodiment described above, the karaoke device 10 is the only device connected to the scoring server device 20, but the device connected to the scoring server device 20 is not limited to the karaoke device 10. For example, a mobile phone or a mobile game machine may be connected to the scoring server device 20. In this case, similar to the karaoke device 10 described above, these devices transmit singing data representing the user's singing to the scoring server device 20 while performing the karaoke playback process, and have been transmitted from the scoring server device 20. The scoring result represented by the scoring result data is displayed. That is, both the cellular phone and the portable game machine function as the communication device according to the present invention.

  Moreover, the scoring server device 20 may change the content of the scoring process according to the type of the song data transmission source. This type may be a predetermined device classification, such as a karaoke device, a game machine, a mobile phone, etc., or a device used when it is desired to obtain a full-fledged singing voice evaluation. It may be of a type according to the purpose of use, such as a device used when it is desired to obtain a simple singing voice evaluation. Or the classification according to the processing capability of an apparatus may be sufficient. In this case, the storage unit 24 of the scoring server device 20 stores the device type and the scoring algorithm in association with each other. And CPU21 of scoring server apparatus 20 acquires the classification from the transmission source of song data, and specifies the algorithm of the scoring process memorize | stored in association with the acquired classification. And the process part 25 performs the scoring process mentioned above using the specified algorithm.

  Here, in the storage unit 24, the type “karaoke apparatus” and the “scoring algorithm for performing detailed evaluation” are stored in association with each other, and the type “mobile phone” and “simple evaluation are performed. Assume that the “scoring algorithm to be performed” is stored in association with each other. In this case, for example, when the singing data and the type “karaoke device” are transmitted from the karaoke device 10, the processing unit 25 of the scoring server device 20 is associated with the type “karaoke device” in the storage unit 24. The scoring process is performed using the “scoring algorithm for performing detailed evaluation”. On the other hand, when the singing data and the type “mobile phone” are transmitted from the mobile phone, the processing unit 25 performs “scoring process for performing simple evaluation” associated with the type “mobile phone” in the storage unit 24. The scoring process is performed using the “No algorithm”. In other words, the processing unit 25, when acquiring voice data from a predetermined type of communication device, performs a simpler evaluation than when acquiring voice data from a communication device of a type different from the predetermined type. Do. This simple evaluation means that the number of processing steps in the scoring process is small or the processing time when evaluating the same singing voice is short. Specifically, it is realized by reducing evaluation items or omitting processing with a large processing load. Here, the case where two scoring algorithms are provided is described as an example, but three or more scoring algorithms according to the degree of simplicity of evaluation may be provided.

(Modification 6)
In the embodiment described above, the score corresponding to the degree of similarity between the singing data and the model singing data is used as the scoring result, but the information used as the scoring result is not limited to this. For example, after calculating the score according to the degree of similarity between the singing data and the model singing data, the factor that decreases the score in the user's singing voice is specified, and in addition to the calculated score, advice for increasing the score is given. It may be used as a scoring result.

(Modification 7)
In the embodiment described above, the configuration in which the karaoke device 10 and the scoring server device 20 are separate devices has been described as an example, but the karaoke device 10 may have the function of the scoring server device 20. In this case, for example, while exchanging information between the karaoke apparatuses 10, the karaoke apparatus 10 that is not performing the karaoke reproduction process or the karaoke apparatus 10 that is performing the karaoke reproduction process but has a low processing load is determined. The karaoke device 10 functions as the scoring server device 20. Alternatively, when a management device is connected to the karaoke device 10, the management device detects the karaoke device 10 that is not performing karaoke playback processing or the karaoke device 10 that is performing karaoke playback processing but has a low processing load. The detected karaoke device 10 may be operated as the scoring server device 20.

(Modification 8)
In embodiment mentioned above, although the example which the scoring server apparatus 20 scored the skill of a song was given and demonstrated, the object which the scoring server apparatus 20 scores is not restricted to a song. For example, the scoring server device 20 may score the skill of the user's English conversation. Here, it is assumed that the user conducts an English conversation toward a terminal device such as a mobile phone or a computer device, and the terminal device transmits voice data representing the user's English conversation to the scoring server device 20. Similar to the karaoke device 10 described above, this terminal device includes a sound collection unit 17 and a sound emission unit 19, and is connected to the scoring server device 20 via the network N. In this case, exemplary voice data representing an exemplary English conversation is stored in the storage unit 24 of the scoring server device 20 in advance. Then, the processing unit 25 performs a scoring process representing the skill of the user's English conversation based on the voice data transmitted from the terminal device and the model voice data stored in the storage unit 24, as described above. .

(Modification 9)
In the above-described embodiment, the model voice data Dr stored in the storage unit 24 of the scoring server device 20 may represent the model singing voice itself, or may be a model voice note, frequency, It may be a parameter representing a rhythm or the like. In short, the storage unit 24 only needs to store exemplary voice data representing an exemplary singing voice. That is, the “voice data” in the present invention is not limited to data representing the voice itself, but may be a parameter representing the characteristics of the voice.

(Modification 10)
In the embodiment described above, the processing performed by the CPU 21 may be performed by the processing unit 25. Further, the processing performed by the processing unit 25 may be performed by the CPU 21. Furthermore, the processing performed by the CPU 21 or the processing unit 25 may be realized by a single or a plurality of hardware resources, or may be realized by the CPU 21 executing one or a plurality of programs. The program is provided in a state stored in a computer-readable recording medium such as a magnetic recording medium such as a magnetic tape or a magnetic disk, an optical recording medium such as an optical disk, a magneto-optical recording medium, or a semiconductor memory. obtain. It is also possible to download the program via a network such as the Internet.

DESCRIPTION OF SYMBOLS 1 ... Scoring system, 10 ... Karaoke apparatus, 11 ... CPU, 12 ... Memory, 13 ... Communication part, 14 ... Memory | storage part, 15 ... Operation part, 16 ... Display part, 17 ... Sound collection part, 18 ... Sound source part, 19 ... Sound emitting part, 20 ... Scoring server device, 21 ... CPU, 22 ... Memory, 23 ... Communication part, 24 ... Storage part, 25 ... Processing part.

Claims (6)

First storage means for storing exemplary voice data representing an exemplary singing voice;
Acquisition means for acquiring audio data representing a user's singing voice from a communication device in a streaming manner;
Second storage means for storing voice data acquired by the acquisition means;
A dividing unit that divides the time axis of the voice data stored in the second storage unit into a plurality of evaluation sections to be evaluated for singing voice;
Each evaluation section divided by the dividing means is selected in the order according to the position on the time axis, and the audio data included in the selected evaluation section is stored in the first storage means Evaluation means for evaluating the user's singing voice in the selected evaluation section by comparing with the model voice data included in the section corresponding to the evaluation section, and generating an evaluation result;
A transmission unit that transmits the generated evaluation result to the communication device each time an evaluation result in each evaluation section is generated by the evaluation unit ;
The communication device includes a karaoke device and a communication device of a type other than the karaoke device,
The evaluation means changes the processing content of the evaluation between when the acquisition means acquires the voice data from the karaoke device and when the voice data is acquired from a communication device of a type other than the karaoke device. Characteristic evaluation device. Separation data indicating the separation position on the time axis of each phrase included in the exemplary singing voice is added to the exemplary voice data,
The dividing unit corresponds to each phrase section delimited by the delimiter data in the model audio data stored in the first storage unit with respect to the time axis of the audio data stored in the second storage unit. The evaluation apparatus according to claim 1, wherein the evaluation device is divided into a plurality of evaluation sections. Before Symbol evaluation unit, when the acquisition unit has acquired the voice data from the communication device of the type other than the karaoke apparatus, the evaluation with a simple content than from the karaoke apparatus has acquired the voice data The evaluation device according to claim 1, wherein the evaluation device is performed. Until the user finishes singing, the communication device extracts a feature amount from voice data representing the user's singing voice, and transmits the extracted feature amount to the evaluation device.
The evaluation unit evaluates the singing voice based on the feature amount when the feature amount is transmitted from the communication device.
The evaluation apparatus according to any one of claims 1 to 3, wherein The evaluation device and the communication device are connected via a network,
The communication device detects a state of the network, and extracts a feature amount from voice data representing the user's singing voice when an available bandwidth of the network is narrower than a threshold, and the extracted feature Sending the quantity to the evaluation device,
The evaluation unit evaluates the singing voice based on the feature amount when the feature amount is transmitted from the communication device.
The evaluation apparatus according to any one of claims 1 to 4, wherein A communication unit that communicates with a plurality of communication devices,
When the communication unit establishes communication with a plurality of communication devices, the evaluation unit extracts a feature amount from the audio data stored in the second storage unit, and based on the extracted feature amount To evaluate the singing voice
The evaluation apparatus according to any one of claims 1 to 5, wherein

Images