JP7688554B2 - Karaoke System - Google Patents

Description

The present invention relates to a karaoke system.

Karaoke systems can provide a variety of content that utilizes karaoke devices. For example, when singing karaoke, users may imitate the singing of artists (so-called singing imitations). Karaoke systems have been proposed that can provide content that evaluates such singing imitations.

For example, Patent Document 1 discloses a technique for performing speaker-specific word speech recognition processing on feature points of voice extracted from the voice signals of a model singing and the singer's singing of the model singing, comparing the sound quality and length of the generated words obtained by this processing, and outputting and displaying a scoring signal indicating the similarity between the model singing and the singing based on a distance value indicating the degree of separation in this comparison. Patent Document 2 also discloses a technique in which a singing standard reference data storage means stores singing standard reference data that is the singing standard for each karaoke song, a singing imitation reference data storage means stores singing imitation reference data including the singing features, expression positions, and expression sections for each original artist, and a singing scoring means scores the skill of a singing imitation based on singing evaluation data extracted from the user's voice signal input from a microphone, the singing standard reference data for the karaoke song, and the singing imitation reference data corresponding to the original artist of the karaoke song when the user sings any karaoke song.

Japanese Patent Application Publication No. 11-259081 JP 2013-231881 A

The object of the present invention is to provide a karaoke system capable of implementing new content that allows users to guess the singing section in which a specific singer's singing voice is emitted.

One invention for achieving the above object is a karaoke system in which a server device and a karaoke device are communicatively connected, the server device comprising an information storage unit for storing, in association with each other, song identification information of a song, singer identification information of a singer who sang the song, singing voice data corresponding to the singing voice of the singer who sang the song, and feature information extracted from the singing voice data, and when a certain song and a specific singer are selected by a user of the karaoke device, the server device identifies at least one singing voice data corresponding to the singing voice of the other singer based on a similarity obtained by comparing the feature information linked to the song identification information of the certain song and the singer identification information of the specific singer with the song identification information of the certain song and the singer identification information of a singer other than the specific singer. and a transmission processing unit which transmits singing voice data corresponding to the singing voice of the specific singer and singing voice data corresponding to the singing voice of the identified other singer to the karaoke device, wherein the karaoke device has a content generation unit which generates content data by assigning singing voice data of the specific singer corresponding to at least one of the singing sections of the certain song, and assigning singing voice data of the identified other singer corresponding to the other singing sections to other singing sections, and a content execution unit which plays back the content data in conjunction with a karaoke performance of the certain song, thereby emitting a singing voice, and after playback is completed, displays an answer screen for selecting the singing section in which the singing voice of the specific singer is emitted.
Other features of the present invention will become apparent from the following specification and drawings.

The present invention makes it possible to implement content that allows users to guess the singing section in which a specific singer's singing voice is emitted.

FIG. 1 is a diagram showing a karaoke system according to an embodiment. FIG. 2 is a diagram illustrating a server device according to an embodiment. 1 is a diagram showing a karaoke device according to an embodiment; FIG. 2 is a diagram showing a karaoke main unit according to the embodiment. 4 is a flowchart showing a process of the karaoke system according to the embodiment. FIG. 4 is a diagram showing a display screen of the remote control device according to the embodiment. FIG. 11 is a diagram showing similarities in each singing section of a piece of music X in an embodiment. 11 is a diagram showing singing voice data assigned to each singing section of a piece of music X in an embodiment. FIG. FIG. 13 is a diagram showing an answer screen displayed on the display device according to the embodiment. FIG. 13 is a diagram showing an answer screen displayed on the display device according to the embodiment.

<Embodiment>
A karaoke system according to an embodiment will be described with reference to FIGS. 1 to 9B.

==Karaoke System==
1, the karaoke system 1 includes a karaoke device K and a server device S. The karaoke device K is communicably connected to the server device S via a network N. The network N is a transmission path such as a public telephone line network or the Internet.

The server device S is a computer that manages various information. The server device S can also communicate with other karaoke devices (not shown) other than the karaoke device K via the network N.

==Server Device==
2, the server device S includes a storage unit 10, a communication unit 20, and a control unit 30. Each component is connected to a bus B via an interface (not shown).

[Storage means]
The storage unit 10 is a large-capacity storage device that stores various types of data. In this embodiment, a part of the storage area of the storage unit 10 functions as an information storage unit 100.

(Information storage unit)
The information storage unit 100 stores, in association with each other, song identification information of a song, singer identification information of a singer who sang the song, singing voice data corresponding to the singing voice of the singer, and feature information extracted from the singing voice data.

Song identification information is information unique to each song, such as a song ID for identifying each song. Singer identification information is information unique to each singer, such as a singer ID for identifying each singer. A singer is a user who performs karaoke singing on a karaoke device that is communicatively connected to the server device S, or an artist who sings a song that can be karaoke. Singing voice data is data corresponding to the singing voice that sang the song.

Feature information is information extracted from singing voice data. In other words, there is a one-to-one correspondence between singing voice data corresponding to the singing voice of a certain song and the feature information extracted from that singing voice data.

Specifically, the feature information is the transition of pitch and/or volume in the time series of singing voice data. The feature information may also include information on special singing methods, etc. Types of special singing methods include vibrato type (ascending, descending, pennant type), sharp tendency (weak or strong tendency to sharpen relative to the pitch of the reference data), flat tendency (weak or strong tendency to flatten relative to the pitch of the reference data), deviation of the pronunciation position relative to the accompaniment (a method of delaying the timing of karaoke singing relative to the accompaniment (so-called pause), a method of speeding up the timing of karaoke singing relative to the accompaniment (so-called run)), fall, high cup, shake, falsetto, fist, voice quality (breathy, husky, clear, etc.), etc.

For example, suppose that a user inputs his/her singer ID via karaoke device K and logs into karaoke system 1. Then, suppose that the user sings a karaoke version of a song selected on karaoke device K. In this case, karaoke device K analyzes singing voice data corresponding to the singing voice of the karaoke performance using a known method. Specifically, karaoke device K extracts pitch and/or volume sample by sample from the singing voice data in a time series for each frame of a predetermined time length (e.g., 10 to 20 msec), and obtains feature information by associating the extracted pitch and/or volume with the elapsed time in frame units based on the start of the performance of the song. Alternatively, karaoke device K may detect a special singing style during karaoke singing by using a known singing scoring function, and include in the feature information the type of singing style, the elapsed time at the time of detection (the elapsed time in frame units based on the start of the performance of the song), the duration for which the singing style is detected, etc.

The karaoke device K links the song ID of a song, the singer ID of the user, singing voice data corresponding to the singing voice of the user, and feature information extracted from the singing voice data, and transmits them to the server device S. The server device S stores the received various information in the information storage unit 100.

On the other hand, the singing voice data of the artist himself can be, for example, singing voice data acquired when recording the guide vocals. The feature information can be extracted using the same method as when extracting it from singing voice data of karaoke singing.

[Communication means]
The communication means 20 provides an interface for communicating with the karaoke device K.

[Control Means]
The control means 30 performs various controls in the server device S. The control means 30 includes a CPU and a memory (neither of which are shown in the figure). The CPU realizes various functions by executing programs stored in the memory.

In this embodiment, the CPU executes a program of the singing section guessing content stored in the memory, and the control means 30 functions as the identification unit 200 and the transmission processing unit 300. The singing section guessing content is content for guessing the singing section in which the singing voice of a specific singer is emitted.

(Specific section)
When a certain song and a specific singer are selected by a user of the karaoke device, the identification unit 200 identifies at least one singing voice data corresponding to the singing voice of the other singer based on the degree of similarity obtained by comparing the song identification information of the certain song and the feature information linked to the singer identification information of the specific singer with the song identification information of the certain song and the singer identification information of the other singer.

Typically, singing section guessing content is performed by multiple users. Hereinafter, the user who selects the song and the specific singer may be referred to as the "question setter," and the user who plays the game to guess the singing section may be referred to as the "answerer." The question setter operates the remote control device 50 of the karaoke device K to select the song and the specific singer to be used in the singing section guessing content. The karaoke device K transmits the song identification information of the selected song and the singer identification information of the specific singer to the server device S.

The specific singer is one of the singers who have sung the selected song in the past (i.e., one of the singers for whom singing voice data corresponding to the singing voice of the selected song is stored in the information storage unit 100). The other singers are singers different from the specific singer and who have sung the selected song in the past.

Similarity indicates the degree of agreement between feature information extracted from singing voice data of a particular singer and feature information extracted from singing voice data of other singers. Similarity can be determined by comparing the changes in pitch and/or volume that constitute the feature information.

For example, suppose that the average pitch value in frames F1 to F20 of the feature information extracted from the singing voice data of a particular singer and the average pitch value in frames F1 to F20 of the feature information extracted from the singing voice data of another singer are within a predetermined pitch range (for example, within ±30 cents). In this case, the identification unit 200 determines that the feature information in frames F1 to F20 match (high similarity). On the other hand, suppose that the average pitch value in frames F1 to F20 of the feature information extracted from the singing voice data of a particular singer and the average pitch value in frames F1 to F20 of the feature information extracted from the singing voice data of another singer are outside the predetermined pitch range. In this case, the identification unit 200 determines that the feature information in frames F1 to F20 do not match (low similarity).

Or, suppose that ascending vibrato is detected as a type of singing style in multiple frames F100 to F150 of the feature information extracted from the singing voice data of a specific singer, and ascending vibrato is detected as a type of singing style in multiple frames F90 to F140 of the feature information extracted from the singing voice data of another singer. In this case, the identification unit 200 determines that the feature information in the multiple frames F100 to F140 matches. On the other hand, suppose that ascending vibrato is detected as a type of singing style in multiple frames F100 to F150 of the feature information extracted from the singing voice data of a specific singer, and descending vibrato is detected as a type of singing style in multiple frames F90 to F140 of the feature information extracted from the singing voice data of another singer. In this case, the identification unit 200 determines that the feature information in the multiple frames F100 to F140 does not match.

The identification unit 200 determines whether the feature information matches in all singing sections (all frames) that make up a song, thereby determining the similarity between the feature information extracted from the singing voice data of a particular singer and the feature information extracted from the singing voice data of other singers. The greater the number of frames in which the feature information matches, the higher the similarity of the overall feature information. The similarity can be expressed, for example, as a percentage from "0% (feature information does not match in all frames) to 100% (feature information matches in all frames)."

The identification unit 200 identifies singing voice data corresponding to the singing voice of another singer based on the similarity.

The number of singing voice data to be identified needs to be at least one. For example, the identification unit 200 can identify singing voice data whose similarity satisfies a preset reference value (e.g., 80% or more).

Alternatively, the identification unit 200 can identify singing voice data according to the type of performance section that constitutes a song, more specifically, the singing section. A performance section is a section in which karaoke is performed. A performance section includes a singing section and a non-singing section. A singing section is a section in which lyrics to be sung are set in a song, such as an A verse, a B verse, a C verse, a chorus, etc. A non-singing section is a section in which lyrics to be sung are not set in a song, such as a measure that constitutes an introduction, interlude, or postlude.

Let us now assume that a certain song is composed of the following performance sections: introduction, first verse A melody, first verse B melody, first chorus, interlude, second verse A melody, second verse B melody, second chorus, interlude, chorus, and postlude. In this case, there are three types of singing sections (A melody, B melody, chorus). Of these, singing voice data corresponding to the singing voice of a specific singer needs to be assigned to at least one singing section (details will be described later). The identification unit 200 can then identify two singing voice data in descending order of similarity as singing voice data corresponding to the singing voice of another singer.

(Transmission Processing Section)
The transmission processing unit 300 transmits singing voice data corresponding to the singing voice of the specific singer and singing voice data corresponding to the singing voices of the other specified singers to the karaoke device.

In the above example, the transmission processing unit 300 reads out, from the information storage unit 100, singing voice data linked to the singer identification information of a specific singer received from the karaoke device K. The transmission processing unit 300 also reads out, from the information storage unit 100, singing voice data corresponding to the singing voice of another singer identified by the identification unit 200. The transmission processing unit 300 transmits the read singing voice data to the karaoke device K.

==Karaoke Equipment==
The karaoke device K is a device for playing karaoke songs and for users to sing karaoke. As shown in FIG. 3, the karaoke device K includes a karaoke main unit 40, a speaker 50, a display device 60, a microphone 70, and a remote control device 80.

The karaoke main unit 40 performs various controls related to karaoke performance and karaoke singing, such as controlling the performance of the selected song, controlling the display of lyrics and background images, and processing audio signals input through the microphone 70. The speaker 50 is configured to emit sound based on a signal from the karaoke main unit 40. The display device 60 is configured to display videos and images on a screen based on a signal from the karaoke main unit 40. The microphone 70 is configured to convert the singing voice based on the singer's karaoke singing into an analog audio signal and input it to the karaoke main unit 40. The remote control device 80 is a device for performing various operations on the karaoke main unit 40.

As shown in FIG. 4, the karaoke main unit 40 includes a storage unit 40a, a communication unit 40b, an input unit 40c, a performance unit 40d, and a control unit 40e. Each component is connected to a bus B via an interface (not shown).

[Storage means]
The storage unit 40a is a large-capacity storage device that stores various types of data, including music data.

The song data is provided with song identification information. The song data includes accompaniment data, reference data, and performance section information. The accompaniment data is data that is the source of the karaoke performance sound. The reference data is data that indicates the main melody of the song. The performance section information is information that indicates the order and type of the performance sections that make up the song. In addition, the storage means 40a stores, for each song, lyric subtitle data for displaying the lyrics of the song on the display device 60 or the like in sync with the karaoke performance, and background video data such as background video that is displayed on the display device 60 or the like during the karaoke performance.

[Communication means/input means]
The communication means 40b provides an interface for communicating with the remote control device 80 and the server device S. The input means 40c is configured to allow the user to input various instructions. The input means 40c is a button or the like provided on the karaoke main unit 40. Alternatively, the remote control device 80 may function as the input means 40c.

[Means of performance]
The performance means 40d, under the control of the control means 40e, performs karaoke performance of a song and processes a signal based on a singing voice input through the microphone 70. The performance means 40d includes a sound source, a mixer, an amplifier, etc. (none of which are shown).

[Control Means]
The control means 40e performs various controls in the karaoke device K. The control means 40e includes a CPU and a memory (neither of which are shown in the figures). The CPU realizes various functions by executing programs stored in the memory.

In this embodiment, the CPU executes the singing section matching content program stored in memory, and the control means 40e functions as a content generation unit 400 and a content execution unit 500.

(Content Generation Unit)
A content generating section 400 generates content data by allocating singing voice data of a specific singer corresponding to at least one of the singing sections of a certain song, and by allocating singing voice data of other specified singers corresponding to the other singing sections to the other singing sections.

The content data is used when implementing the singing section guessing content.

The content generating unit 400 assigns singing voice data of a specific singer corresponding to the singing section to at least one of the singing sections of the selected song. On the other hand, the content generating unit 400 assigns singing voice data of other identified singers corresponding to the other singing sections to the other singing sections. The content generating unit 400 completes the content data by assigning one singing voice data to each of all singing sections of the selected song.

Here, the content generating unit 400 can randomly select the singing voice data to be assigned to a certain singing section. For example, suppose that a certain song has three singing sections, an A melody, a B melody, and a chorus, and two singing voice data (singing voice data SD1 and SD2) are identified as singing voice data corresponding to the singing voice of another singer.

In this case, the content generating unit 400 assigns, to the singing section of the A melody of a certain song, singing voice data corresponding to the singing section of the A melody from among the singing voice data of a specific singer. On the other hand, the content generating unit 400 assigns, to the singing section of the B melody of a certain song, singing voice data corresponding to the singing section of the B melody from among the singing voice data SD1, and assigns, to the singing section of the chorus of a certain song, singing voice data corresponding to the singing section of the chorus from among the singing voice data SD2.

In addition, when the content generating unit 400 assigns singing voice data of another singer to another singing section, it can assign the singing voice data that has the highest similarity in that other singing section.

For example, in the above example, the content generation unit 400 refers to the similarity of the characteristic information determined by the identification unit 200 for the singing voice data SD1 and SD2 of the other singer in the singing section of the B melody. The content generation unit 400 assigns the singing voice data of the other singer corresponding to the characteristic information with the higher similarity to the singing section of the B melody. Similarly, the content generation unit 400 assigns the singing voice data of the other singer corresponding to the characteristic information with the higher similarity to the singing section of the chorus.

In the above example, singing voice data of the same other singer may be assigned to different singing sections. For example, singing voice data SD1 of the same other singer may be assigned to the singing section of the bridge and the singing section of the chorus.

(Content Execution Department)
A content execution section 500 reproduces the content data in accordance with the karaoke performance of a certain piece of music to emit a singing voice, and after the reproduction is completed, displays an answer screen for selecting a singing section in which the singing voice of a specific singer is emitted.

After the content data is completed, the content execution unit 500 reads out the accompaniment data of a certain song from the storage means 40a and controls the performance means 40d to start a karaoke performance of the certain song. The content execution unit 500 also controls the performance means 40d to play the content data in sync with the karaoke performance. A singing voice based on the singing voice data is emitted from the speaker 50 in sync with the karaoke performance. The contestant listens to the singing voice that is being emitted and guesses who is singing.

After playback ends, the content execution unit 500 displays an answer screen on the display screen of the display device 60 for selecting the singing section in which the singing voice of a specific singer is emitted.

The contestant operates the remote control device 80 to select the singing section that he or she considers to be the correct answer. The content execution unit 500 announces "correct answer" if the selected singing section is a singing section in which the singing voice of a specific singer is emitted, and announces "incorrect answer" if the selected singing section is not a singing section in which the singing voice of a specific singer is emitted.

For example, the content execution unit 500 notifies the user of the correct answer by displaying the words "Correct" or "Incorrect" on the display screen of the display device 60. Alternatively, the content execution unit 500 notifies the user of the correct answer by emitting the sound "Correct" or "Incorrect" from the speaker 50.

If application software for the singing section guessing content is pre-installed on a mobile device such as a smartphone owned by the contestant, the contestant can operate the mobile device to select the singing section that he or she thinks is correct. In this case, the content execution unit 500 transmits a judgment result of "correct" or "incorrect" to the mobile device according to the selection result transmitted from the mobile device. The mobile device displays the received judgment result on the display screen.

==About processing in the karaoke system==
Next, the process in the karaoke system 1 according to this embodiment will be described with reference to Fig. 5 to Fig. 9B. Fig. 5 is a flow chart showing the process in the karaoke system 1. Fig. 6 shows a display screen of the remote control device 80. Fig. 7 shows the similarity in each singing section of a song X (described later). Fig. 8 shows singing voice data assigned to each singing section of a song X (described later). Figs. 9A and 9B show answer screens displayed on the display device 60. In this example, it is assumed that two users U1 and U2 use the karaoke device K. It is also assumed that the information storage unit 100 stores, in association with each other, song identification information of a song, singer identification information of a singer who sang the song, singing voice data corresponding to the singing voice of the singer who sang the song, and feature information extracted from the singing voice data.

When a user operates the remote control device 80 of the karaoke device K and selects the singing section guessing content, the karaoke device K starts the singing section guessing content (Start singing section guessing content; step 10).

A user operates the remote control device 80 to select a song to be used in the singing section guessing content. The karaoke device K transmits the song ID of the selected song to the server device S (transmitting the song ID of the selected song; step 11).

The server device S creates a list of singers based on the singer IDs linked to the song IDs of the received songs. The server device S transmits the created list of singers to the karaoke device K (transmitting list of singers. Step 12).

A user operates the remote control device 80 to select a specific singer. The karaoke device K transmits the singer ID of the selected specific singer to the server device S (transmitting the singer ID of the selected specific singer. Step 13).

Specifically, the remote control device 80 displays a search list of songs. User U1 uses the search list to select song X. Song X is composed of the following performance sections: introduction, A melody of the first verse, B melody of the first verse, chorus of the first verse, interlude, A melody of the second verse, B melody of the second verse, chorus of the second verse, interlude, C melody, chorus, and epilogue.

The karaoke device K transmits the song ID of the song X selected by the user U1 to the server device S via the remote control device 80. The server device S refers to the information storage unit 100 and creates a list of singers based on the singer IDs linked to the song ID of the song X. The server device S transmits the created list of singers to the karaoke device K. The remote control device 80 of the karaoke device K displays the list of singers (see FIG. 6). The user U1 selects himself/herself from the list of singers and selects the execution icon for the singing section guessing content. The karaoke device K transmits the singer ID of the selected user U1 to the server device S. In this example, the user U1 corresponds to the "question setter" and the "specific singer".

Next, the identification unit 200 of the server device S identifies at least one piece of singing voice data corresponding to the singing voice of the other singer based on the similarity obtained by comparing the feature information linked to the song ID of the song transmitted in step 11 and the singer ID of the singer transmitted in step 13 with the feature information linked to the song ID of the song and the singer ID of another singer different from the singer (identifying singing voice data corresponding to the singing voice of the other singer; step 14).

The transmission processing unit 300 of the server device S transmits singing voice data corresponding to the singing voice of the specific singer and singing voice data corresponding to the singing voice of the other singers identified in step 14 to the karaoke device K (transmitting singing voice data; step 15).

Specifically, the identification unit 200 refers to the information storage unit 100 and reads out characteristic information CU1 linked to the song ID of song X and the singer ID of user U1. Similarly, the identification unit 200 reads out characteristic information CP, CU3 to CUn linked to the song ID of song X and the singer IDs of other singers (artist P, and users U3 to Un) different from user U1 to whom the song ID of song X is linked.

The identification unit 200 compares the read characteristic information CU1 with each of the characteristic information CP and CU3 to CUn to determine the degree of similarity. In this example, the identification unit 200 identifies singing voice data SD7 to SD9 that correspond to the singing voices of other singers, users U7 to U9, based on the degree of similarity.

The transmission processing unit 300 reads out the singing voice data linked to the singer ID of user U1 received from the karaoke device K from the information storage unit 100. The transmission processing unit 300 also reads out the singing voice data SD7 to SD9 corresponding to the singing voices of users U7 to U9 identified by the identification unit 200 from the information storage unit 100. The transmission processing unit 300 transmits the read singing voice data to the karaoke device K.

Next, the content generation unit 400 generates content data by assigning singing voice data of the singer corresponding to at least one of the singing sections of the song, and by assigning singing voice data of the other singers identified in step 14 corresponding to the other singing sections to the other singing sections (generating content data; step 16).

Specifically, when the content generating unit 400 assigns the singing voice data SD7 to SD9 of the users U7 to U9 to another singing section, it assigns the singing voice data with the highest similarity in the other singing section.

Figure 7 shows the similarity in each singing section of song X, determined by the determination unit 200. In this example, the similarity is shown in five levels (the higher the number, the higher the similarity).

For example, in the singing section of the A melody, the similarity corresponding to the singing voice data SD7 is "2", the similarity corresponding to the singing voice data SD8 is "3", and the similarity corresponding to the singing voice data SD9 is "5". In this case, the content generation unit 400 assigns the singing voice data from the singing voice data SD9 corresponding to the singing section of the A melody to the singing section of the A melody. Note that there may be multiple singing voice data showing the same similarity, such as the singing section of the chorus shown in Figure 7. In this case, the content generation unit 400 selects and assigns one of the singing voice data.

The content generating unit 400 also needs to assign the singing voice data of user U1 to one of the singing sections. Here, as shown in FIG. 7, the similarity between the singing voice data SD7 to SD9 is low in the singing section of the B melody. Therefore, the content generating unit 400 assigns, to the singing section of the B melody, the singing voice data of user U1 that corresponds to the singing section of the B melody.

Figure 8 shows the singing voice data assigned to each singing section of song X. In this example, singing voice data SD9 is assigned to the singing section of the A melody of song X, singing voice data of user U1 is assigned to the singing section of the B melody of song X, singing voice data SD7 is assigned to the singing section of the C melody of song X, and singing voice data SD8 is assigned to the singing section of the chorus of song X.

Next, the content execution unit 500 plays the content data in time with the karaoke performance of the song to emit a singing voice (emit singing voice based on the content data; step 17). After playback is completed (Y in step 18), the content execution unit 500 displays an answer screen for selecting a singing section in which the singing voice of a specific singer is emitted (display answer screen; step 19). Based on the answerer's selection, the content execution unit 500 displays whether the answer is correct or incorrect (display whether the answer is correct or incorrect; step 20).

Specifically, the content execution unit 500 controls the performance means 40d to play the content data generated in step 16 in accordance with the karaoke performance of the song X. The speaker 50 emits a singing voice based on the singing voice data in accordance with the karaoke performance.

After playback is completed, the content execution unit 500 displays an answer screen on the display screen of the display device 60 for selecting the singing section in which the singing voice of user U1 is emitted (see FIG. 9A). User U2 operates the remote control device 80 to select the singing section that he or she considers to be the correct answer. In this example, user U2 corresponds to the "answerer."

If the selected singing section is a singing section in which the singing voice of a specific singer is emitted, the content execution unit 500 displays the word "Correct" on the answer screen (see Figure 9B).

As is clear from the above, the karaoke system 1 according to this embodiment has a server device S and a karaoke device K connected to each other so that they can communicate with each other. The server device S has an information storage unit 100 that stores, in association with each other, song identification information of a song, singer identification information of a singer who sang the song, singing voice data corresponding to the singing voice of the singer who sang the song, and characteristic information extracted from the singing voice data, and an identification unit 200 that, when a song and a specific singer are selected by a user of the karaoke device K, identifies at least one singing voice data corresponding to the singing voice of the other singer based on the similarity obtained by comparing the song identification information of the song and the characteristic information linked to the singer identification information of the specific singer with the song identification information of the song and the characteristic information linked to the singer identification information of another singer different from the specific singer, and a transmission processing unit 300 that transmits the singing voice data corresponding to the singing voice of the specific singer and the singing voice data corresponding to the identified singing voice of the other singer to the karaoke device K. The karaoke device K has a content generating unit 400 that generates content data by assigning singing voice data corresponding to a specific singer to at least one singing section of a certain song, and by assigning singing voice data corresponding to the other singing sections of the specified other singer to the other singing sections, and a content executing unit 500 that reproduces the content data in accordance with the karaoke performance of a certain song to emit a singing voice, and after the reproduction is completed, displays an answer screen for selecting the singing section in which the singing voice of the specific singer is emitted.

With this karaoke device K, singing section guessing content can be enjoyed using content data generated by assigning singing voice data of a specific singer to at least one singing section of a song, and assigning singing voice data of other singers identified based on similarity to the remaining singing sections. In other words, with the karaoke device K of this embodiment, new content can be implemented in which the singing section in which the singing voice of a specific singer is emitted can be guessed.

In addition, when allocating singing voice data of another singer to another singing section, the content generation unit 400 in the karaoke device K according to this embodiment can allocate singing voice data with the highest similarity in the other singing section. With such a karaoke device K, content data can be generated using singing voice data corresponding to the singing voice of another singer that is similar to the singing voice of a specific singer.

<Modification>
In the above embodiment, there is a possibility that the similarity of the singing voice data corresponding to the singing voice of the specified other singers is low. In this case, even if the content data is created using the singing voice data, the difference between the singing voice of the specified singer and the singing voice of the specified singer is clear, so that the content may not be fully enjoyed.

When the content generating unit 400 assigns singing voice data of another singer to another singing section, the content generating unit 400 can assign singing voice data that has been corrected according to the characteristic information of a specific singer in the other singing section. The correction can be performed using known techniques (for example, JP-A-10-187187).

For example, in an embodiment, when the content generating unit 400 assigns singing voice data SD1 to the singing section of the B melody of a certain song, the content generating unit 400 uses feature information extracted from the singing voice data of a specific singer and performs correction to approximate the pitch and/or volume of the singing voice data SD1 to the pitch and/or volume of the singing voice data of the specific singer. The content generating unit 400 assigns the corrected singing voice data SD1 to the singing section of the B melody.

With this karaoke device K, even if the similarity of the singing voice data corresponding to the singing voices of the other identified singers is low, it is possible to generate content data consisting of singing voices that resemble the singing voice of a specific singer. This allows contestants to enjoy the singing section guessing content even more.

In addition, the content generating unit 400 may determine whether the similarity satisfies a preset reference value, and perform correction only if it is determined that the similarity does not satisfy the reference value.

In addition, the karaoke device K may be configured to allow the difficulty level to be set when selecting content. When a high difficulty level is set, the content generation unit 400 performs the above-mentioned correction on singing voice data with low similarity and generates content data using the corrected singing voice data. On the other hand, when a low difficulty level is set, the content generation unit 400 does not correct the singing voice data even if the similarity is low, and generates content data using the identified singing voice data as is. With such a karaoke device K, content data can be generated according to the difficulty level. Therefore, contestants can enjoy singing section guessing content with various levels of difficulty.

＜Other＞
The above embodiment is presented as an example, and does not limit the scope of the invention. The above configurations can be implemented in appropriate combinations, and various omissions, substitutions, and modifications can be made without departing from the gist of the invention. The above embodiment and its modifications are included in the scope of the invention and its equivalents described in the claims, as well as the scope and gist of the invention.

REFERENCE SIGNS LIST 1 karaoke system 100 information storage unit 200 identification unit 300 transmission processing unit 400 content generation unit 500 content execution unit K karaoke device S server device

Claims (3)

A karaoke system in which a server device and a karaoke device are communicatively connected,
The server device includes:
an information storage unit that stores, in association with each other, song identification information of a song, singer identification information of a singer who sang the song, singing voice data corresponding to the singing voice of the singer who sang the song, and feature information extracted from the singing voice data;
an identification unit which, when a certain song and a certain singer are selected by a user of the karaoke device, identifies at least one singing voice data corresponding to the singing voice of the other singer based on a degree of similarity obtained by comparing feature information linked to song identification information of the certain song and singer identification information of the specific singer with feature information linked to song identification information of the certain song and singer identification information of another singer different from the specific singer;
a transmission processing unit for transmitting singing voice data corresponding to the specific singer's singing voice and singing voice data corresponding to the specified other singer's singing voice to the karaoke device;
having
The karaoke device includes:
a content generating unit that generates content data by allocating singing voice data of the specific singer corresponding to at least one of the singing sections of the certain song, and allocating singing voice data of the other specific singer corresponding to the other singing sections to the other singing sections;
a content execution unit that reproduces the content data in accordance with the karaoke performance of the certain song to emit a singing voice, and after the reproduction ends, displays an answer screen for selecting a singing section in which the singing voice of the specific singer is emitted;
A karaoke system having: The karaoke system according to claim 1, characterized in that, when the content generation unit assigns the singing voice data of the other singer to the other singing section, it assigns the singing voice data with the highest similarity in the other singing section. The karaoke system according to claim 1 or 2, characterized in that when the content generating unit assigns the singing voice data of the other singer to the other singing section, the content generating unit assigns singing voice data that has been corrected according to characteristic information of a specific singer in the other singing section.

Images