Karaoke device

Abstract

To provide a karaoke device which can determine whether note end pulling-off singing is included in karaoke singing or not.SOLUTION: A karaoke device includes a singing technique determination part which determines whether note end pulling-off singing is performed in one note when a horizontal section and a descending section exist in a singing pitch detected in a period from sounding start timing to sounding end timing of one note included in reference data of karaoke music, the singing pitch when the descending section ends is included between a reference pitch of one note and a singing pitch which is higher than the reference pitch by a first prescribed value and the singing pitch when the descending section ends is less than the singing pitch of the horizontal section by a second prescribed value.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a karaoke device. [Background technology]

[0002] The karaoke device is equipped with a scoring function for scoring the skill of karaoke singing by comparing singing voice data extracted from singing voice input by a microphone with reference data showing the main melody of the song performed as karaoke.

[0003] Some karaoke singers imitate the singing of professional singers and use special singing techniques such as shakuri, falls, fists, shouts, etc. Here, Patent Document 1 discloses a technology that can identify so-called "pulling-off singing." [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7158282 Summary of the Invention [Problem to be solved by the invention]

[0005] An object of the present invention is to provide a karaoke device capable of determining whether or not a karaoke performance includes a pull-off performance at the end of a note (described later). [Means for solving the problem]

[0006] The inventor analyzed the singing pitch included in the singing voice signal judged to be a distinctive singing style by the auditory sense, and found that, in one note, there exists a horizontal section in which the pitch difference between two consecutive singing pitches is within a predetermined range and continues for more than a first predetermined number of times, and a descending section after the horizontal section in which the pitch difference between the two consecutive singing pitches becomes negative, the singing pitch at the end of the descending section is included between the reference pitch of the one note and a singing pitch that is higher than the reference pitch by a first predetermined value, and the singing pitch at the end of the descending section is equal to or lower than a second predetermined value with respect to the singing pitch in the horizontal section. The present invention was completed based on this discovery, and is a technology that can judge this singing style by detecting the pitch change unique to this singing style.

[0007] This type of singing shows a similar singing pitch transition (horizontal section, descending section) to the pull-off singing described in Patent Document 1, but because the singing pitch near the end of a note shows a unique transition, this type of singing is hereinafter referred to as "note-end pull-off singing."

[0008] Specifically, the invention for achieving the above object is a karaoke device having a singing pitch detection unit which detects a singing pitch for each predetermined section from a singing voice signal obtained by singing a karaoke song, and a singing technique determination unit which determines that a note-end pulling-off is being performed for a note when, among the singing pitch detected during a period from the start timing to the end timing of the pronunciation of a note included in reference data of the karaoke song, there is a horizontal section in which the pitch difference between two consecutive singing pitches is within a predetermined range and continues for more than a first predetermined number of times, and a descending section after the horizontal section in which the pitch difference between two consecutive singing pitches is negative, the singing pitch at the end of the descending section is between a reference pitch for the note and a singing pitch that is higher than the reference pitch by a first predetermined value, and the singing pitch at the end of the descending section is equal to or lower than a second predetermined value relative to the singing pitch of the horizontal section. Other features of the present invention will become apparent from the following specification and drawings. Effect of the Invention

[0009] According to the present invention, it is possible to determine whether karaoke singing includes a pull-off at the end of a note. [Brief description of the drawings]

[0010] [Figure 1] 1 is a diagram illustrating an example of a hardware configuration of a karaoke device according to an embodiment. [Diagram 2] FIG. 2 is a diagram showing an example of a software configuration of a karaoke main unit according to an embodiment. [Diagram 3] 11 is a flowchart showing a process for determining whether singing is performed with a pull-off at the end of a note according to the embodiment. [Figure 4] FIG. 4 is a diagram showing singing pitch and pitch difference according to the embodiment. [Diagram 5] FIG. 2 is a diagram showing a relationship between one note and a singing pitch according to an embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] <Embodiment> A karaoke device 1 according to an embodiment will be described with reference to FIGS. 1 to 5.

[0012] ==Karaoke Equipment== The karaoke device 1 is a device for performing karaoke and for singers to sing karaoke. The karaoke device 1 registers karaoke songs selected by singers in a reservation queue and performs karaoke in order. As shown in Fig. 1, the karaoke device 1 includes a karaoke main unit 10, a speaker 20, a display device 30, a microphone 40, and a remote control device 50.

[0013] The speaker 20 is configured to emit sound based on a sound emission signal from the karaoke main unit 10. The display device 30 is configured to display videos and images on a screen based on a signal from the karaoke main unit 10. The microphone 40 is configured to convert the singer's singing voice (voice input to the microphone 40) into an analog singing voice signal and input it to the karaoke main unit 10. The remote control device 50 is a device for performing various operations on the karaoke main unit 10. The singer can use the remote control device 50 to select a karaoke song that he or she wishes to sing. The display screen of the remote control device 50 displays icons and the like for inputting instructions for various operations.

[0014] The karaoke main unit 10 performs various controls related to karaoke singing, such as controlling the performance of the selected karaoke piece, controlling the display of lyrics and background images, and processing singing voice signals input through a microphone 40. As shown in Fig. 1, the karaoke main unit 10 comprises a control unit 11, a communication unit 12, a storage unit 13, an audio processing unit 14, a display processing unit 15, and an operation unit 16. Each component is connected to a bus B via an interface (not shown).

[0015] The control unit 11 includes a CPU 11a and a memory 11b. The CPU 11a executes operation programs stored in the memory 11b to realize various control functions. The memory 11b is a storage device that stores the programs executed by the CPU 11a and temporarily stores various information when the programs are executed.

[0016] The communication section 12 provides an interface for connecting the karaoke machine 10 to a communication line via a router (not shown).

[0017] The storage unit 13 is a large-capacity storage device that stores various data, such as a hard disk drive, etc. The storage unit 13 stores data of a plurality of songs to be used by the karaoke device 1 to perform karaoke.

[0018] The song data is assigned a song ID for identifying each karaoke song. The song data includes accompaniment data, reference data, etc. The accompaniment data is the data that is the source of the karaoke performance sound. The accompaniment data includes information indicating the tempo at which the karaoke is performed. A predetermined tempo value is set for each song. The reference data is data used as a standard for scoring a singer's karaoke singing. The reference data is made up of a number of notes, and a predetermined pitch (i.e., a reference pitch) is set for each note.

[0019] The memory unit 13 also stores lyric subtitle data for displaying lyrics corresponding to each karaoke song on the display device 30, etc., background image data such as background images to be displayed on the display device 30, etc. during karaoke performance, performance time data indicating the karaoke performance time for each song, and song attribute information (information related to the song such as the singer's name, lyricist / composer's name, genre, etc.).

[0020] The sound processor 14 controls the performance of the karaoke music piece and processes the singing voice signal input through the microphone 40 under the control of the controller 11. The display processor 15 performs various processes related to the display on the display device 30 and the remote controller 50 under the control of the controller 11. For example, the display processor 15 controls the display device 30 to display an image in which lyrics subtitles and various icons are superimposed on a background image during the performance of the karaoke music piece. Alternatively, the display processor 15 displays various icons for operation input on the display screen of the remote controller 50. The operation unit 16 is composed of a panel switch and a remote controller receiving circuit, and outputs operation signals such as a song selection signal and a performance stop signal to the controller 11 in response to the operation of the panel switch of the karaoke device 1 or the remote controller 50 by the singer. The controller 11 detects the operation signal from the operation unit 16 and executes the corresponding process.

[0021] (Software configuration) 2 is a diagram showing an example of the software configuration of the karaoke main unit 10. The karaoke main unit 10 includes a singing pitch detection unit 100, a singing technique judgment unit 200, a presentation unit 300, and a scoring processing unit 400. The singing pitch detection unit 100, the singing technique judgment unit 200, the presentation unit 300, and the scoring processing unit 400 are realized by the CPU 11a executing programs stored in the memory 11b.

[0022] [Singing pitch detection section] The singing pitch detection section 100 detects the singing pitch for each predetermined section from a singing voice signal obtained by singing a karaoke piece.

[0023] Specifically, the singing pitch detection unit 100 analyzes the singing voice signal to detect the singing pitch. The singing pitch is detected sample by sample in time series in frame units of a predetermined time length (for example, 20 msec). The singing pitch detection unit 100 sequentially outputs the detected singing pitch to the singing technique determination unit 200. Note that the singing pitch may be detected sequentially as the karaoke piece is sung, or may be detected all at once after the singing of one karaoke piece has been completed. That is, in this embodiment, a frame of 20 msec length corresponds to the predetermined section in the singing pitch detection.

[0024] [Singing technique judgment department] The singing technique determination unit 200 determines that a note-end pulling-off is being performed for a note when a horizontal section and a descending section are present in the singing pitch detected during the period from the pronunciation start timing to the pronunciation end timing of a note included in the reference data of a karaoke song, the singing pitch at the end of the descending section is included between the reference pitch of the note and a singing pitch that is higher than the reference pitch by a first predetermined value, and the singing pitch at the end of the descending section is equal to or lower than a second predetermined value relative to the singing pitch of the horizontal section.

[0025] A horizontal section is a section in which the pitch difference between two consecutive singing pitches is within a predetermined range and continues for a first predetermined number or more. The predetermined range is the allowable deviation between two consecutive singing pitches when specifying a horizontal section. The first predetermined number is the minimum number of consecutive pitch differences that are included within the predetermined range when specifying a horizontal section. The predetermined range and the first predetermined number are preset to arbitrary values ​​according to the degree of note-end pulling-off singing to be determined. For example, the predetermined range is ±1 cent to ±10 cents, and the first predetermined number is 3 to 8.

[0026] The descending section is a section after the horizontal section where the pitch difference between two consecutive singing pitches becomes negative. In the descending section, the singing pitch decreases (i.e., the singing pitch value decreases) from the starting singing pitch to the ending singing pitch. Note that the singing pitch at the end of the descending section is higher than the reference pitch of one note.

[0027] The first predetermined value is a value for determining whether or not the singing pitch at the end of the descending section is a singing pitch of the note-end pulling-off singing with respect to the reference pitch of one note. The second predetermined value is a value for determining whether or not the singing pitch at the end of the descending section is a singing pitch of the note-end pulling-off singing with respect to the singing pitch of the horizontal section (for example, a statistical quantity such as the average singing pitch of the horizontal section, the singing pitch at the end of the horizontal section, the highest or lowest singing pitch in the horizontal section). The first predetermined value and the second predetermined value are preset to any value according to the degree of the note-end pulling-off singing to be determined. For example, the first predetermined value is +100 cents to +150 cents with respect to the reference pitch of one note. Also, for example, the second predetermined value is -75 cents to -100 cents with respect to the average singing pitch of the horizontal section.

[0028] Here, the process of determining whether or not a singing is pulled-off at the end of a note will be described in detail with reference to Fig. 3 and Fig. 4. Fig. 3 is a flow chart showing the process of determining whether or not a singing is pulled-off at the end of a note. Fig. 4 is a diagram showing the singing pitch for each predetermined section detected by analyzing the singing voice signal obtained by singing karaoke piece X, and the pitch difference between two consecutive singing pitches. In the example of Fig. 4, the pitch difference is shown in cents.

[0029] ===Note Specific= First, the singing technique determination unit 200 identifies one note for determining whether or not the singing technique is a pull-off at the end of a note (identifying one note; step 10). The determination of whether or not the singing technique is a pull-off at the end of a note is performed for each individual note. The note can be identified by referring to the reference data for karaoke X. In this example, note N is identified as the one note.

[0030] ==Singing pitch extraction== Next, the singing technique determination unit 200 extracts the singing pitch detected in the period from the pronunciation start timing to the pronunciation end timing of the identified one note in time series (extraction of singing pitch for one note; step 11). The detection of the singing pitch is performed by the singing pitch detection unit 100.

[0031] The pronunciation start timing indicates the time when pronunciation should start when singing a note. The pronunciation start timing of a note corresponds to the elapsed time until that note, assuming that the time when the karaoke song starts to be played is 0. The pronunciation end timing indicates the time when pronunciation should end when singing a note. In this example, it is assumed that singing pitches P(1) to P(n) detected in the period from the pronunciation start timing to the pronunciation end timing of note N are extracted. Also, in this example, it is assumed that there were three predetermined sections (i.e., sections of 0 cent) in which singing pitches P(n+1) to P(n+3) could not be detected before the pronunciation end timing of note N.

[0032] ===Calculating the pitch difference= Next, the singing style determining section 200 calculates the pitch difference between two consecutive singing pitches, starting from the first pitch extracted in step 11 (calculating the pitch difference; step 12).

[0033] Specifically, the singing technique judgment unit 200 calculates the pitch difference between two consecutive singing pitches starting from pitch P(1). For example, the singing technique judgment unit 200 obtains the pitch difference Pd(1) between singing pitch P(1) and singing pitch P(2). Here, the pitch Pd(x) is calculated as P(x+1)-P(x), and so on.

[0034] ==Identifying the descending section== Next, the singing style determining section 200 identifies a descending section.

[0035] First, if there are more than a second predetermined number of consecutive pitch differences in which the singing pitch furthest from the beginning of the two singing pitches used to calculate the pitch difference is not 0 but is negative, among the pitch differences included in the period from near the end of the pronunciation period of one note to the pronunciation end timing, the singing technique judgment unit 200 determines the singing pitch closest to the beginning used to calculate the multiple negative pitch differences as the singing pitch at the start of the descent section (determines the singing pitch at the start of the descent section; step 13).

[0036] Then, when there is a negative pitch difference in the period after the singing pitch at the start of the descending section, and the singing pitch furthest from the beginning of the two singing pitches used to calculate the pitch difference is 0, and there are a third predetermined number or more consecutive singing pitches that are 0 after the singing pitch that is 0, and the singing pitch closer to the beginning of the two singing pitches used to calculate the pitch difference is included between the reference pitch of a note and a singing pitch that is higher than the reference pitch by a first predetermined value, the singing technique judgment unit 200 identifies the descending section by determining the singing pitch at the end of the descending section as the singing pitch at the end of the descending section (the descending section is identified by determining the singing pitch at the end of the descending section; step 14).

[0037] The end of the sounding period of one note is preset as a range based on the sounding end timing. The range can be set based on the length of the note, such as the last note n10 when note N is divided into 10 parts (note n1 to note n10) in time series. A plurality of singing pitches are included in the range.

[0038] The second predetermined number is the minimum number of consecutive negative pitch differences in which the singing pitch farthest from the beginning of the two singing pitches used in calculating the pitch difference to identify the descending section is not 0. The third predetermined number is the minimum number of consecutive zero singing pitches to identify the descending section. The second and third predetermined numbers are preset to arbitrary values ​​according to the degree of note-end pulling-off singing to be determined. For example, the second predetermined number is 3 to 5, and the third predetermined number is 2 to 4.

[0039] Specifically, the singing technique determination unit 200 checks whether the singing pitch of the two singing pitches used to calculate the pitch difference, which is the farthest from the beginning, is not 0 and whether there are a predetermined number or more consecutive negative pitch differences, starting from pitch difference Pd(n-19) included in the period from near the end of the sounding period of note N to the sounding end timing. In this example, it is assumed that the period from near the end of the sounding period of note N to the sounding end timing includes singing pitches P(n-20) to P(n). The second predetermined number is "3".

[0040] 4, the pitch differences Pd(n-17) and Pd(n-16) are negative, and the singing pitches P(n-17) and P(n-16) used to calculate the pitch differences, which are the furthest from the beginning, are not zero, but the pitch difference Pd(n-15) is positive. In this case, the singing technique determination unit 200 resets the count and checks whether the singing pitch used to calculate the pitch difference from the pitch difference Pd(n-14) is not zero, and whether the negative pitch difference occurs a predetermined number of times in a row.

[0041] On the other hand, as shown in Fig. 4, from the pitch difference Pd(n-6) to the pitch difference Pd(n), the singing pitch farthest from the beginning of the two singing pitches used to calculate the pitch difference is not 0 and there are seven consecutive negative values. In this case, the singing technique judgment unit 200 judges that the singing pitch farthest from the beginning of the two singing pitches used to calculate the pitch difference is not 0 and there are consecutive negative pitch differences of a second predetermined number or more. Note that, as in this example, the singing technique judgment unit 200 counts regardless of the second predetermined number when the singing pitch farthest from the beginning of the two singing pitches used to calculate the pitch difference is not 0 and there are consecutive negative values.

[0042] The singing technique determination unit 200 determines that the singing pitch furthest from the beginning of the two singing pitches used to calculate the pitch difference is not 0, and that the singing pitch closest to the beginning used to calculate the multiple pitch differences that are negative (i.e., singing pitch P(n-7)) is the singing pitch at the start of the descent section DZ.

[0043] Next, the singing style judgment unit 200 checks whether there is a negative pitch difference in the period after the singing pitch P(n-7) at the start of the descending section DZ, and whether the singing pitch farthest from the beginning of the two singing pitches used in calculating the pitch difference is 0. In this example, the value of the pitch difference Pd(n+1) is negative, as shown in Figure 4.

[0044] In this case, the singing style judgment unit 200 checks whether the singing pitch P(n+1), which is the farthest from the beginning of note N, of the two singing pitches (P(n) and P(n+1)) used in calculating the pitch difference Pd(n+1), is 0. As shown in FIG. 4, P(n+1) is 0.

[0045] Next, the singing style judgment unit 200 checks whether or not there are a third predetermined number of consecutive singing pitches that are 0 after the singing pitch P(n+1) that is 0. In this example, the third predetermined number is set to "2".

[0046] As shown in FIG. 4, singing pitch P(n+1) is followed in succession by singing pitches P(n+2) and P(n+3) which are both 0.

[0047] Furthermore, the singing technique determination unit 200 checks whether the singing pitch P(n) closer to the beginning of note N of the two singing pitches (P(n) and P(n+1)) used in calculating the pitch difference Pd(n+1) is included between the reference pitch Pn of note N and a singing pitch that is higher than the reference pitch Pn by a first predetermined value. In this example, the reference pitch Pn of note N is "6000 cents" and the first predetermined value is "+150 cents".

[0048] 4, the value of the singing pitch P(n) is 6010 cents. Therefore, the singing technique determination unit 200 determines that the singing pitch P(n) is included between the reference pitch Pn of note N, "6000 cents," and the singing pitch "6150 cents," which is "+150 cents" higher than the reference pitch Pn.

[0049] In this case, the singing technique determination unit 200 identifies the descent section DZ by using the singing pitch P(n) closer to the beginning of the two singing pitches used when calculating the pitch difference Pd(n+1) as the singing pitch at the end of the descent section DZ.

[0050] ==Identifying horizontal intervals== Next, the singing style determining section 200 identifies horizontal sections.

[0051] First, the singing technique judgment unit 200 checks the calculated pitch differences toward the beginning of a note in the period before the identified descending section, and if there are a first predetermined number or more consecutive pitch differences within a predetermined range, it determines the singing pitch just before the singing pitch farthest from the beginning used when calculating the multiple pitch differences as the singing pitch at the end of the horizontal section (determines the singing pitch at the end of the horizontal section; step 15).

[0052] Then, if there is a pitch difference outside a specified range in the period before the singing pitch at the end of the horizontal section, the singing technique judgment unit 200 identifies the horizontal section by taking the singing pitch closer to the beginning of the two singing pitches used when calculating the pitch difference outside the specified range as the singing pitch at the start of the horizontal section (the horizontal section is identified by determining the singing pitch at the start of the horizontal section; step 16).

[0053] Specifically, the singing technique determination unit 200 checks the calculated pitch difference in the period before the specified descending section DZ toward the beginning of the note N, and determines whether or not the pitch difference within a predetermined range continues for a first predetermined number or more. In this example, the predetermined range is "±5 cents" and the first predetermined number is "6".

[0054] As shown in Fig. 4, there are 11 consecutive pitch differences within the predetermined range (±5 cents) from the pitch difference Pd(n-7) that was first determined to be within the predetermined range in the period before the specified descending section DZ to the pitch difference Pd(n-17). In this example, the descending section DZ and the horizontal section HZ are consecutive. In this case, the singing technique determination unit 200 can determine the singing pitch P(n-8) at the end of the horizontal section HZ to be the singing pitch just before the singing pitch P(n-7) that is the furthest from the beginning of note N and was used when calculating the pitch difference Pd(n-17) from the pitch difference Pd(n-7).

[0055] Next, the singing style determining unit 200 checks whether or not there is a pitch difference that is outside a predetermined range in the period before the singing pitch P(n-8) at the end of the horizontal section HZ.

[0056] In this example, as shown in Fig. 4, the pitch difference Pd(n-18) is outside the predetermined range in the period before the singing pitch P(n-8). In this case, the singing technique determination unit 200 specifies the horizontal section HZ by taking the singing pitch P(n-18) farthest from the beginning of the note N out of the two singing pitches P(n-18) and P(n-19) used when calculating the pitch difference Pd(n-18) as the singing pitch at the start of the horizontal section HZ. If the maximum and minimum values ​​of the pitch difference included in the horizontal section are specified as horizontal sections only when they are within a predetermined range (e.g. ±10 cents), sections in which the pitch difference gradually rises or falls, for example, even if they are within the predetermined range, are not specified as horizontal sections, thereby preventing erroneous detection of pulling-off at the end of a note.

[0057] == Judgment of pulling-off at the end of a note == If the singing pitch at the end of the identified descending section is equal to or lower than a second predetermined value relative to the singing pitch of the identified horizontal section, the singing technique determination unit 200 determines that the identified note is being sung with a pull-off at the end of the note (determining whether a pull-off at the end of the note is being sung with a pull-off at the end of the note; step 17).

[0058] Specifically, the singing technique determination unit 200 checks whether the singing pitch P(n) at the end of the identified descending section DZ is equal to or less than a second predetermined value with respect to the average singing pitch of the identified horizontal section HZ. In this example, the second predetermined value is assumed to be "-75 cents."

[0059] 4, the average singing pitch of the identified horizontal section HZ is approximately "6092 cents," and the value of the singing pitch P(n) at the end of the identified descending section DZ is "6010 cents." In other words, the singing pitch at the end of the identified descending section is equal to or smaller than the second predetermined value (-75 cents) relative to the average singing pitch of the identified horizontal section.

[0060] Therefore, the singing technique determining unit 200 determines that the specified note N is sung with a pull-off at the end of the note.

[0061] Fig. 5 is a diagram showing the relationship between note N and singing pitch. In Fig. 5, the pitch of note N (reference pitch) is indicated by pitch Pn. Also, start time Ts is the timing when note N starts to be sounded, and end time Te is the timing when note N ends to be sounded. The vertical axis in Fig. 5 indicates singing pitch, and the horizontal axis indicates time.

[0062] As shown in Fig. 5, the entire horizontal section HZ and the entire descending section DZ are included between the start time Ts and the end time Te of note N, the singing pitch at the end of the descending section DZ is included between the reference pitch Pn of note N and a singing pitch that is +150 cents higher than the reference pitch Pn, and the singing pitch at the end of the descending section DZ is -75 cents or lower compared to the singing pitch of the horizontal section HZ. In other words, it can be said that note-end pulling-off is being performed for note N. Note that in order to determine that note-end pulling-off has been performed for note N, it is sufficient that part of a horizontal section (a section in which a pitch difference within a specified range continues for a specified number of times or more) is included.

[0063] [Presentation part] The presenting unit 300 presents the result of the judgment made by the singing technique judging unit 200 to the singer. For example, when the judgment result indicates that the note N has been sung with a pull-off at the end of the note, the presenting unit 300 can display an icon indicating that the note N has been sung with a pull-off at the end of the note near the guide melody image corresponding to the note N displayed on the display device 30. The guide melody image is an image of the melody of a karaoke piece displayed to support the singer in singing karaoke. A known method can be used to display the guide melody image (see, for example, Japanese Patent Application Laid-Open No. 2004-205817).

[0064] The presentation unit 300 may display only an icon indicating that the note-end pulling-off has been performed, regardless of the guide melody image. Alternatively, the presentation unit 300 may notify the user of the note-end pulling-off by voice (applause, cheers, etc.) via the speaker 20, instead of displaying an icon indicating that the note-end pulling-off has been performed.

[0065] [Scoring Processing Section] The scoring processor 400 compares the singing voice signal obtained by singing a karaoke piece with the reference data of the karaoke piece to calculate a score. The scoring processor 400 may use a known method for the process of calculating the score. For example, the scoring processor 400 performs a scoring process on the accuracy of the singing pitch based on the singing voice signal input from the microphone 40 and the standard pitch based on the reference data.

[0066] Here, in the scoring processing unit 400 according to this embodiment, when the scoring result of a note is judged to be a fail and when a pull-off at the end of the note is performed for that note, the scoring result of that note is judged to be a pass.

[0067] 5, singing with a pulled-off at the end of a note is likely to fail a typical scoring process because the difference between the reference pitch of the note and the singing pitch is large. Therefore, even if the scoring result of a note is failed, if the singing technique judgment unit 200 judges that the pulled-off at the end of the note was performed for that note, the scoring processing unit 400 judges that the note as passing.

[0068] As is clear from the above, the karaoke device 1 according to the present embodiment includes a singing pitch detection unit 100 for detecting a singing pitch for each predetermined section from a singing voice signal obtained by singing a karaoke piece, and a singing technique determination unit 200 for determining that a note is being pulled-off at the end of a note when the singing pitch detected during the period from the start timing to the end timing of a note included in the reference data of the karaoke piece includes a horizontal section in which the pitch difference between two consecutive singing pitches is within a predetermined range and continues for more than a first predetermined number of times, and a descending section after the horizontal section in which the pitch difference between the two consecutive singing pitches is negative, the singing pitch at the end of the descending section is between the reference pitch of the note and a singing pitch that is higher than the reference pitch by a first predetermined value, and the singing pitch at the end of the descending section is equal to or lower than a second predetermined value with respect to the singing pitch of the horizontal section. With such a karaoke device 1, it is possible to determine whether or not a note is being pulled-off at the end of a note.

[0069] More specifically, the singing technique judgment unit 200 identifies a note, extracts singing pitches detected in a period from the pronunciation start timing to the pronunciation end timing of the identified note in chronological order, calculates a pitch difference between two consecutive singing pitches starting from the extracted first pitch, and if, among the pitch differences included in the period from near the end of the pronunciation period of the note to the pronunciation end timing, the singing pitch furthest from the beginning of the two singing pitches used in calculating the pitch difference is not 0 and there are consecutive negative pitch differences for a second predetermined number or more, the singing pitch closest to the beginning used in calculating the multiple negative pitch differences is set to the singing pitch at the start of the descending section, and if, in the period after the singing pitch at the start of the descending section, there is a negative pitch difference, the singing pitch furthest from the beginning of the two singing pitches used in calculating the pitch difference is 0, there are consecutive singing pitches that are 0 after the singing pitch that is 0 for a third predetermined number or more, and the singing pitch closest to the beginning of the two singing pitches used in calculating the pitch difference is set to If the pitch difference is included between a reference pitch of one note and a singing pitch that is higher than the reference pitch by a first predetermined value, the descending section is specified by setting the singing pitch at the end of the descending section to the singing pitch closest to the beginning of the two singing pitches used in calculating the pitch difference, and if there are a first predetermined number or more consecutive pitch differences within the predetermined range, the singing pitch just before the singing pitch farthest from the beginning used in calculating the multiple pitch differences is specified as the singing pitch at the end of the horizontal section. If there is a pitch difference outside the predetermined range in the period before the singing pitch at the end of the horizontal section, the horizontal section is specified by setting the singing pitch at the start of the horizontal section to the singing pitch farthest from the beginning of the two singing pitches used in calculating the pitch difference outside the predetermined range, and if the singing pitch at the end of the specified descending section is equal to or lower than a second predetermined value relative to the singing pitch of the specified horizontal section, it is determined that a note-end pulling-off is being performed for the specified note.By performing such a process, the singing style determining unit 200 can determine whether the karaoke singing includes a pulling-off technique at the end of a note.

[0070] Furthermore, the karaoke device 1 according to the present embodiment has a scoring processor 400 that, when the scoring result of a note is determined to be "fail" and when the singing technique of "pulling off at the end of the note" is performed for the note, determines the scoring result of the note as "pass." With this karaoke device, it is possible to avoid a situation where the scoring result does not reflect the singing technique of "pulling off at the end of the note" even though the singing technique has been used.

[0071] <Modification> The singing technique determination unit 200 described in the above embodiment can determine that a pull-off at the end of a note is being performed if the voice uttered in the identified horizontal section and the identified descending section is the same vowel.

[0072] Whether or not a voice contains a vowel can be determined using a known voice recognition technique.

[0073] For example, the singing technique determination unit 200 acquires, via the microphone 40, a singing voice signal corresponding to a singing voice obtained when a singer sings karaoke of note N. The singing technique determination unit 200 analyzes the singing voice signal corresponding to the identified horizontal section and identifies a vowel contained in the singing voice. Similarly, the singing technique determination unit 200 analyzes the singing voice signal corresponding to the identified descending section and identifies a vowel contained in the singing voice.

[0074] In addition to the conditions described in the embodiment, the singing technique determination unit 200 determines that a pulling-off at the end of a note is being performed on note N if the voices uttered in the specified horizontal section and the specified descending section are the same vowel. Since a pulling-off at the end of a note is performed by lowering the pitch at the end of the note while pronouncing a vowel, by performing such a determination process, it is possible to distinguish this from normal singing in which different vowels are used in the horizontal section and the descending section, or a consonant is included in the descending section. In other words, this karaoke device can prevent erroneous detection of a pulling-off at the end of a note.

[0075] <Other> In the above embodiment, an example has been described in which the horizontal section and the descending section are continuous. However, in note-end pull-off singing, the horizontal section and the descending section do not necessarily have to be continuous.

[0076] In the above embodiment, when identifying a horizontal section, the calculated pitch difference is checked toward the beginning of a note in the period before the identified descending section, but this is not limited to the above.

[0077] For example, when a descending section is specified, the singing technique judging unit 200 checks the calculated pitch differences in sequence from the beginning of a note in the period before the specified descending section, and if there are a first predetermined number or more consecutive pitch differences within the predetermined range, it sets the singing pitch closest to the beginning used in calculating the multiple pitch differences as the singing pitch at the start of the horizontal section. Also, if there is a pitch difference outside the predetermined range in the period after the singing pitch at the start of the horizontal section, the singing technique judging unit 200 can specify the horizontal section by setting the singing pitch just before the singing pitch closest to the beginning of the two singing pitches used in calculating the pitch difference outside the predetermined range as the singing pitch at the end of the horizontal section.

[0078] Alternatively, the singing style determining section 200 may first identify the horizontal section and then identify the descending section.

[0079] Specifically, the singing technique judging unit 200 checks the calculated pitch differences from the beginning of a note, and if there are a first predetermined number or more consecutive pitch differences within the predetermined range, it sets the singing pitch closest to the beginning used in calculating the multiple pitch differences as the singing pitch at the start of the horizontal section. Also, if there is a pitch difference outside the predetermined range in the period after the singing pitch at the start of the horizontal section, the singing technique judging unit 200 specifies the horizontal section by setting the singing pitch just before the singing pitch closest to the beginning of the two singing pitches used in calculating the pitch difference outside the predetermined range as the singing pitch at the end of the horizontal section.

[0080] In the period after the specified horizontal section, if the singing pitch farthest from the beginning of the two singing pitches used in calculating the pitch difference among the pitch differences included in the period from near the end of the sounding period of one note to the sounding end timing is not 0 and there are a second predetermined number or more consecutive negative pitch differences, the singing technique judgment unit 200 determines the singing pitch closest to the beginning of the negative pitch difference used in calculating the multiple negative pitch differences to be the singing pitch at the start of the descending section, and determines whether or not there is a negative pitch difference in the period after the singing pitch at the start of the descending section. When the singing pitch furthest from the beginning of the two singing pitches used to calculate the pitch difference is 0, and there are a third predetermined number or more consecutive singing pitches that are 0 after the singing pitch that is 0, and the singing pitch closer to the beginning of the two singing pitches used to calculate the pitch difference is included between a reference pitch of a note and a singing pitch that is higher than the reference pitch by a first predetermined value, the descending section can be identified by setting the singing pitch closer to the beginning of the two singing pitches used to calculate the pitch difference as the singing pitch at the end of the descending section.

[0081] It is also possible to provide a program for the process of identifying and determining the pulling-off at the end of a note, as described in the above embodiment. For example, the program causes a computer (e.g., a karaoke device) to detect the singing pitch for each predetermined section from a singing voice signal obtained by singing a karaoke song, and to determine that the pulling-off at the end of a note is being performed when the singing pitch detected during the period from the start timing to the end timing of the pronunciation of a note included in the reference data of the karaoke song includes a horizontal section in which the pitch difference between two consecutive singing pitches is within a predetermined range and continues for more than a first predetermined number of times, and a descending section after the horizontal section in which the pitch difference between two consecutive singing pitches is negative, the singing pitch at the end of the descending section is between the reference pitch of the note and a singing pitch that is higher than the reference pitch by a first predetermined value, and the singing pitch at the end of the descending section is equal to or lower than a second predetermined value with respect to the singing pitch of the horizontal section.

[0082] The program can also be supplied to a computer using a non-transitory computer readable medium with an executable program thereon. Examples of the non-transitory computer readable medium include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), CD-ROMs (Read Only Memory), etc.

[0083] 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. [Explanation of symbols]

[0084] 1 Karaoke equipment 10 Karaoke unit 11 Control section 100 Singing pitch detection section 200 Singing Technique Judgment Department 300 Presentation section 400 Scoring Processing Unit

Claims

1. a singing pitch detection unit that detects a singing pitch for each predetermined section from a singing voice signal obtained by singing a karaoke piece; a singing technique determination unit which determines that a note-end pulling-off technique is being performed for a note when the singing pitch detected during the period from the sounding start timing to the sounding end timing of a note included in the reference data of the karaoke piece includes a horizontal section in which the pitch difference between two consecutive singing pitches is within a predetermined range and continues for more than a first predetermined number of times, and a descending section after the horizontal section in which the pitch difference between the two consecutive singing pitches is negative, the singing pitch at the end of the descending section is included between the reference pitch of the note and a singing pitch that is higher than the reference pitch by a first predetermined value, and the singing pitch at the end of the descending section is equal to or lower than a second predetermined value relative to the singing pitch of the horizontal section; A karaoke device having the following features:

2. The singing technique determination unit is Identifying said one notebook; Extracting, in a time series, the singing pitches detected during the period from the sounding start timing to the sounding end timing of the identified one note; Calculate the pitch difference between two consecutive singing pitches starting from the extracted first pitch, among the pitch differences included in the period from near the end of the sounding period of the one note to the sounding end timing, if the singing pitch farthest from the beginning of the two singing pitches used in calculating the pitch difference is not 0 and there are more than the second predetermined number of consecutive negative pitch differences, the singing pitch closest to the beginning used in calculating the multiple negative pitch differences is set as the singing pitch at the start of the descending section, if there is a negative pitch difference in a period following the singing pitch at the start of the descending section, and the singing pitch furthest from the beginning of the two singing pitches used in calculating the pitch difference is 0, and there are a third predetermined number or more consecutive singing pitches that are 0 after the singing pitch that is 0, and the singing pitch closer to the beginning of the two singing pitches used in calculating the pitch difference is included between a reference pitch of the one note and a singing pitch that is higher than the reference pitch by the first predetermined value, then the descending section is identified by setting the singing pitch closer to the beginning of the two singing pitches used in calculating the pitch difference as the singing pitch at the end of the descending section, during the period before the identified descending section, checking the calculated pitch difference toward the beginning of the one note, and if there are at least the first predetermined number of consecutive pitch differences within the predetermined range, determining the singing pitch immediately preceding the singing pitch farthest from the beginning used in calculating the plurality of pitch differences as the singing pitch at the end of the horizontal section; if there is a pitch difference outside a predetermined range in a period before the singing pitch at the end of the horizontal section, the singing pitch furthest from the beginning of the two singing pitches used in calculating the pitch difference outside the predetermined range is set as the singing pitch at the start of the horizontal section, thereby identifying the horizontal section; 2. The karaoke apparatus according to claim 1, wherein if the singing pitch at the end of the identified descending section is equal to or lower than the second predetermined value relative to the singing pitch of the identified horizontal section, it is determined that the identified note is being pulled off with its end pulled off.

3. The singing technique determination unit is 3. The karaoke apparatus according to claim 2, wherein if the voices uttered in the identified horizontal section and the identified descending section are the same vowel, it is determined that the note-end pulling-off is being performed.

4. 4. The karaoke apparatus according to claim 1, further comprising a scoring processing unit that judges the scoring result of the one note as passing if the scoring result of the one note is judged to be failing and if the note end pulling-off has been performed for the one note.

Images