JP6340303B2 - Karaoke scoring system for scoring humming songs - Google Patents

Description

  The present invention relates to a karaoke scoring device for scoring the skill of karaoke singing by a singer.

  In a commercial communication karaoke system installed in a karaoke room or the like, a singer can enjoy not only karaoke singing but also various services called contents. Singing is one of them, and various services are offered from simple scoring to competing national rankings. On the other hand, services for enjoying singing by a plurality of singers such as duets and choruses (hamori) are also widespread, and techniques for scoring such songs are known.

  For example, in Patent Document 1, each pitch and volume are independently extracted from audio signals input from two microphones, and compared with the pitch and volume of reference data of different singing parts, A technique for scoring each singer or comprehensively is disclosed.

  Further, Patent Document 2 compares a voice signal input from two microphones with reference data of a melody (main song melody), and uses a more similar microphone input as a melody part singing voice, A technique for discriminating others as chorus part singing sounds is disclosed. If this technique is used, it is said that the scoring process can be performed appropriately regardless of which microphone is sung by two singers.

  Further, Patent Document 3 compares the volume transitions and pitch transitions of audio signals input from two microphones, and extracts similarities, whereby two singers in unison singing and cross singing singing. Techniques for determining compatibility are disclosed. If this technique is used, even if the scoring result based on the reference data is not good, that is, it is out of the exemplary singing of the music, if the two singers are out of the same way, the compatibility is good. It can be evaluated.

Japanese Patent Laid-Open No. 10-74093 Japanese Patent Laid-Open No. 10-161672 JP 2006-259237 A

  When melody singing and chorus singing by a plurality of singers are to be scored independently, for example, the technique described in Patent Document 2 identifies the singing voice signal of the melody part and the singing voice signal of the chorus part, Using the prepared melody part reference data and chorus part reference data, each singing may be scored by the technique described in Patent Document 1.

  However, since the chorus part singer tries to perform the chorus singing from the pitch actually sung by the melody part singer, that is, the melody part singing pitch is well, the melody part singing pitch is the reference for the melody part. If it deviates from the pitch of the data, the singing pitch of the chorus part also deviates from the pitch of the chorus part reference data. In other words, the chorus part singer sings at a predetermined pitch from the melody part's singing pitch (successfully sung), but the problem is that the conventional technique is unfairly deducted. was there.

  Here, the above-mentioned Patent Document 3 discloses a technique for evaluating singing by a plurality of singers regardless of reference data, but this is only to evaluate compatibility by similarity, and therefore, the correct pitch. There is no disclosure or suggestion about evaluating the skill of chorus singing whether or not you are addicted.

  The present invention has been made in view of such circumstances, and its purpose is to appropriately evaluate the singing ability of the chorus part singer, that is, the skill of the hammer, using resources and reference data equivalent to those of the prior art. There is to do.

  In order to achieve the above-mentioned object, the present invention is a karaoke scoring device for scoring karaoke songs sung by one melody part singer and one or more chorus part singers in accordance with the karaoke performance of the music. A melody part reference data for evaluating a singing of the melody part and data for evaluating a singing of the chorus part, including at least pitch information, and a chorus part reference corresponding to the chorus part Storage means for storing data, extraction means for independently extracting the singing pitch from the singing voice signal of the melody part and the singing voice signal of the chorus part, and the singing pitch of the melody part and the melody The melody part is sung according to the degree of coincidence with the pitch of the part reference data. A melody part scoring means that calculates a singing tone pitch that is a pitch difference between a singing pitch of the chorus part and a singing pitch of the melody part for the chorus part, and the pitch of the chorus part reference data A chorus part scoring means for calculating a reference pitch which is a pitch difference between the pitch of the melody part reference data and the melody part reference data, and scoring the chorus part singing based on a degree of coincidence between the singing pitch and the reference pitch And.

  According to the karaoke scoring device of the present invention, the chorus part scoring means calculates the pitch difference between the extracted melody part singing pitch and the chorus part singing pitch as the singing humming pitch, and reads it from the storage means. The pitch difference between the pitch of the melody part reference data and the pitch of the chorus part reference data is calculated as a reference pitch. The chorus part scoring means evaluates whether or not the chorus part singer is in the correct pitch with respect to the actual melody part singing pitch, based on the degree of coincidence between the singing hamori pitch and the reference hamori pitch. Scoring. As a result, the singing ability of the chorus part singer, that is, the skill of the hammer can be appropriately evaluated.

  In the above karaoke scoring device, the extracting means extracts the singing pitch of the melody part and the singing pitch of the chorus part divided into minute sections, and the chorus part scoring means extracts the singing tone pitch from the minute pitch. It is preferable that the degree of coincidence be a ratio of the minute section calculated for each section and having a difference from the reference hammer pitch within the acceptable range to the entire minute section in the scoring section. In this karaoke scoring device, the singing hamori pitch is calculated for each minute section, and scoring is made based on the ratio of the minute section determined to pass to all the minute sections, so even if the singing pitch trembles, it is appropriate Can be obtained.

  In the above karaoke scoring device, the melody part reference data and the chorus part reference data include information indicating a break of each note in a singing melody, and the chorus part scoring means includes the melody part reference data and the chorus part reference. It is preferable that the chorus part singing is scored for each musical note break in the data. In this karaoke scoring device, the singing of the chorus part can be performed for each note in the melody of the melody part and the chorus part. As a result, not only the melody part singing pitch and the chorus part singing pitch change at the same timing while taking a predetermined pitch, but the chorus part singing pitch is constant and only the melody part singing pitch is In the case of changing or vice versa, scoring is performed every time the pitch of both changes, so the skill of the hammer can be evaluated more precisely.

  In the karaoke scoring device described above, the karaoke scoring device further includes setting means for setting information indicating whether or not the song is a hamori song section in accordance with the karaoke performance of the music, and the chorus part scoring means includes information set by the setting means. It is preferable that the chorus part singing is scored when indicates the humming singing section. In this karaoke scoring device, since the information indicating whether or not it is a humor song section is used, the recognition process of the harm song section can be simplified.

  According to the present invention, it is possible to appropriately evaluate the singing power of a chorus part singer, that is, the skill of a hammer, using resources and reference data equivalent to those of the prior art.

It is a block diagram explaining the structure of a karaoke apparatus. (A) is a block diagram explaining the structure of a karaoke main body. (B) is a figure explaining the partial area | region of the memory which a control part has. It is a figure explaining a melody part pitch group, a chorus part 1 pitch group, a chorus part 2 pitch group, a melody part reference pitch, a chorus part 1 reference pitch, a chorus part 2 reference pitch, and the like. It is a flowchart (the 1) explaining pass / fail judgment and scoring processing. It is a flowchart (the 2) explaining pass / fail judgment and scoring processing. It is a flowchart (the 3) explaining a pass / fail determination and scoring process. It is a figure explaining the pass / fail determination of a melody part. It is a figure explaining the pass / fail judgment of chorus part 1. It is a figure explaining the pass / fail judgment of chorus part 2.

  Hereinafter, embodiments of the present invention will be described. First, the configuration of the karaoke apparatus 1 will be described with reference to FIG. The karaoke apparatus 1 has a karaoke scoring function for scoring the skill of karaoke singing by a singer. For this reason, the karaoke apparatus 1 of this embodiment is corresponded to the karaoke scoring apparatus which concerns on this invention.

  As shown in FIG. 1, the karaoke apparatus 1 includes a karaoke main body 11, a speaker 12, a monitor 13, a singing microphone 14 (14 </ b> A to 14 </ b> C), and a remote control device 15.

  The karaoke main body 11 performs various controls relating to karaoke singing, such as performance control of karaoke music designated by the singer, display control of lyrics and background video, processing of a microphone signal generated by the singing microphone 14, and scoring of karaoke singing. It is a part to do. The speaker 12, the monitor 13, the singing microphone 14, and the remote control device 15 are connected to the karaoke main body 11 by wire or wirelessly. The karaoke main body 11 is connected to the transmission path 2 via the router device RT. The karaoke main body 11 can communicate with a host device (not shown) through the transmission path 2. The karaoke main body 11 can download music data, for example, through communication with the host device. The karaoke main body 11 will be described in detail later.

  The speaker 12 outputs a sound based on the sound emission signal output from the karaoke main body 11. For example, a mixed sound in which singing voice and karaoke performance sound are mixed is output. The monitor 13 displays a video on the screen based on the video signal from the karaoke main body 11. For example, the lyrics telop and background video of the singing karaoke music are displayed.

  The singing microphone 14 collects the singing voice of the singer, converts it into an analog microphone signal, and inputs it to the karaoke main body 11. And the group of the singing microphone 14 and the karaoke main body 11 is equivalent to the singing voice input means for inputting the singing voice of the singer. In the present embodiment, three singing microphones 14A to 14C are used. Although these singing microphones 14A-14C are equivalent, for convenience of explanation, the singing microphone 14A is used by a singer who sings the melody part of the music, and the singing microphone 14B is a singer who sings the chorus part 1 of the music. It is assumed that the singing microphone 14C is used by a singer who sings the chorus part 2 of the music. Thereby, the microphone signal of the melody part is output from the singing microphone 14A, the microphone signal of the chorus part 1 is output from the singing microphone 14B, and the microphone signal of the chorus part 2 is output from the singing microphone 14C. The number of singing microphones 14 connected to the karaoke main body 11 is not limited to three, and may be two or more.

  The remote control device 15 includes a short-range wireless communication unit capable of bidirectional communication for transmitting and receiving information to and from the karaoke main body 11, and is operated, for example, when a karaoke song is reserved. When reserving karaoke music, the remote controller 15 transmits an operation signal including a music ID for identifying the music to be played. And the karaoke main body 11 manages the selected karaoke music in a queue. Further, the remote control device 15 is configured to be communicable with the router device RT, and is also operated when the singer logs in.

  Next, the karaoke main body 11 will be described in detail. As shown in FIG. 2A, the karaoke main body 11 includes a control unit 21, a communication unit 22, a storage unit 23, a sound source unit 24, an acoustic processing unit 25, a display processing unit 26, and an operation unit 27. And have.

  The control unit 21 is a central part of control in the karaoke main body 11. The communication unit 22 is a part that provides an interface for connecting the karaoke main body 11 to the transmission path 2 via the router device RT. The storage unit 23 is a large-capacity storage device that stores various types of data, and includes a hard disk drive or the like. The sound source unit 24 is a part that generates a musical sound signal that is the basis of the karaoke performance sound signal. The acoustic processing unit 25 is a part that performs processing related to karaoke performance sound and singing voice. Singing microphones 14 (14A to 14C) are connected to the acoustic processing unit 25, and analog microphone signals output from the singing microphones 14A to 14C are individually input and digitally converted. As a result, as will be described later, the acoustic processing unit 25 generates three types of singing voice signals for each minute section. The display processing unit 26 is a part that performs display control of a background video or the like during karaoke performance. The operation unit 27 is a part that outputs an operation signal corresponding to the operation of the panel switch or the remote control device 15. These units are connected to a communicable state via the bus BS.

  Next, the control unit 21 will be described in detail. The control unit 21 includes a CPU 21a and a memory 21b. The CPU 21a performs various controls according to the operation program stored in the memory 21b. For example, an operation input process for accepting an operation from the operation unit 27 and a sequencer process that operates as a sequencer are performed. The control unit 21 performs a scoring process by reading and executing a scoring program stored in the storage unit 23. In the present embodiment, scoring processing for the melody part and scoring processing for the chorus part are performed. Therefore, the control unit 21 functions as a melody part scoring means for scoring a melody part song, and also functions as a chorus part scoring means for scoring a chorus part song. The scoring process will be described in detail later.

  The memory 21b included in the control unit 21 is a storage element that stores a program to be executed by the CPU 21a or temporarily stores various types of information when the program is executed. The memory 21b also stores the music reserved for performance by the remote controller 15 in a queue. A part of the memory 21b is used as a work area, and as shown in FIG. 2B, a memory flag storage area 21c, a reference pitch storage area 21d, a singing pitch storage area 21f, and a melody part determination result storage. It is used as an area 21h, a melody part scoring result storage area 21i, a chorus part determination result storage area 21j, and a chorus part scoring result storage area 21k. For convenience, information stored in these areas will be described later. As can be seen from FIG. 2B, the “part” portion in the melody part or chorus part may be abbreviated as “P” in the drawing for the convenience of illustration.

  Next, the storage unit 23 will be described in detail. The storage unit 23 is provided with a music data storage area 23a in addition to the above-described scoring program storage area 23b in which the scoring program is stored. The music data storage area 23a stores music data for playing karaoke music. The music data includes MIDI data, reference data, and lyrics data.

  The MIDI data is data for automatically playing a sound source of an electronic musical instrument, that is, performance data for karaoke, and is composed of time-series MIDI events defined by delta time. MIDI events include, for example, note-on indicating the start of sound generation and note-off indicating the end of sound generation. These note-on and note-off include a pitch number representing the pitch of the sound and a velocity representing the strength of the sound. The control unit 21 reads out and accumulates the MIDI data from the music data storage area 23a of the storage unit 23, and transmits a MIDI event to the sound source unit 24 at a timing defined by the delta time.

  The sound source unit 24 generates a musical sound signal based on the received MIDI event. The tone generator 24 stores waveform data of various sounds played by the target musical instrument so as to be readable. As the sequencer process, the control unit 21 acquires MIDI data from the music data storage area 23a and transmits each MIDI event included in the MIDI data to the sound source unit 24 at a predetermined timing. Then, the sound source unit 24 processes the waveform data according to the received MIDI event, and generates a digital musical tone signal. The generated musical sound signal is output to the acoustic processing unit 25.

  The reference data is used when scoring a karaoke song by a singer, and is information representing the pitch of each note in the singing melody in time series. This reference data is also in MIDI format. That is, it is composed of time-series MIDI events defined by delta time. Therefore, note-on and note-off constituting the reference data correspond to information indicating the separation of each note in the singing melody. The pitch number corresponds to the pitch information of each note in the singing melody. The control unit 21 reads this reference data from the music data storage area 23a of the storage unit 23 at a timing specified by the delta time, and stores the reference pitch based on the pitch number in the reference pitch storage area 21d of the memory 21b. .

  As described above, in the karaoke apparatus 1 of the present embodiment, each of the melody part, the chorus part 1 and the chorus part 2 is targeted. Correspondingly, melody part reference data for scoring a melody part, chorus part 1 reference data for scoring chorus part 1, and chorus part 2 are scored as reference data in the music data storage area 23a. The chorus part 2 reference data for storing is stored. Therefore, the storage unit 23 (music data storage area 23a) corresponds to storage means for storing melody part reference data and chorus part reference data.

  As shown in FIG. 3, the melody part reference data includes a reference pitch Pmr. Similarly, the chorus part 1 reference data includes the reference pitch Pc1r, and the chorus part 2 reference data includes the reference pitch Pc2r. At the time of scoring the chorus part, the control unit 21 calculates the difference between the reference pitch Pc1r and the reference pitch Pmr as the reference tone pitch Dc1r and compares it with the tone pitch of the melody part and the chorus part 1 singing voice. Similarly, the difference between the reference pitch Pc2r and the reference pitch Pmr is also calculated by the control unit 21 as the reference tone pitch Dc2r and compared with the tone pitch of the singing voice of the melody part and the chorus part 2. That is, the reference tone pitch Dc1r is used for scoring the chorus part 1, and the reference tone pitch Dc2r is used for scoring the chorus part 2.

  Returning to FIG. 2A, the lyric data is character string data for displaying on the monitor 13 the lyric telop of the music being played. This lyric data is transmitted from the music data storage area 23 a to the display processing unit 26 by the sequencer processing of the control unit 21. The display processing unit 26 displays the lyrics telop on the monitor 13 based on the received lyrics data. Further, the display processing unit 26 changes the display color of the lyrics telop according to the color change command output by the sequencer process of the control unit 21. As a result, the display color of the lyrics telop is changed on the monitor 13 in synchronization with the performance of the music.

  Next, the acoustic processing unit 25 will be described in detail. In the acoustic processing unit 25, a musical tone signal process and a microphone signal process are performed. For example, in the performance control of music, the sound processing unit 25 performs analog conversion on the musical sound signal output from the sound source unit 24 to generate a karaoke performance sound signal. When the microphone signals from the singing microphones 14A to 14C are input, the acoustic processing unit 25 branches each microphone signal and mixes each branched microphone signal with the karaoke performance sound signal. The signal after mixing is output to the speaker 12 as a sound emission signal. Thereby, from the speaker 12, the mixed sound in which the singing voice by each singer and the karaoke performance sound are mixed is output. And the group of the control part 21, the memory | storage part 23, the sound source part 24, the sound process part 25, and the speaker 12 is equivalent to the performance means which performs the karaoke performance of a music.

  Each of the other branched microphone signals is independently digitally converted inside the acoustic processing unit 25, and a singing voice signal is individually generated. Each singing voice signal is generated by the acoustic processing unit 25 repeatedly sampling each microphone signal, for example, in units of 10 to 30 ms. Each generated singing voice signal includes a volume for each minute section and a pitch for each minute section. The acoustic processing unit 25 outputs each generated singing voice signal to the control unit 21. And the control part 21 functions as an extraction means, takes in each song audio | voice signal, and extracts a pitch. The extracted pitch is stored in the singing pitch storage area 21f of the memory 21b.

  A specific example shown in FIG. In this specific example, the melody part pitch group SP0 is extracted from the singing voice signal of the melody part. The melody part pitch group SP0 includes a time-series sample pitch sp0 (t). These sample pitches sp0 (t) are data in which the singing pitch and the sounding timing are paired. That is, these sample pitches sp0 (t) include the singing pitch Pmk (t). Further, the chorus part 1 pitch group SP1 is extracted from the singing voice signal of the chorus part 1. The chorus part 1 pitch group SP1 includes a time-series sample pitch sp1 (t). These sample pitches sp1 (t) are data in which the singing pitch Pc1k (t) and the sound generation timing are paired. Similarly, the chorus part 2 pitch group SP2 is extracted from the singing voice signal of the chorus part 2. The chorus part 2 pitch group SP2 includes a time-series sample pitch sp2 (t). These sample pitches sp2 (t) are data in which the singing pitch Pc2k (t) and the sound generation timing are paired.

  When scoring the chorus part, the control unit 21 uses the singing tone pitch Dc1k (t), which is the difference between the singing pitch Pc1k (t) and the singing pitch Pmk (t), for scoring the chorus part 1. Similarly, the singing tone pitch Dc2k (t), which is the difference between the singing pitch Pc2k (t) and the singing pitch Pmk (t), is used for scoring the chorus part 2.

  Next, the scoring process performed by the karaoke apparatus 1 will be described. This scoring process is performed by the control unit 21 executing a scoring program.

  As shown in FIG. 4A, the control unit 21 first initializes the hammer flag 1 and the hammer flag 2 and sets both to “0”. When the hamori flag 1 is set to “1”, the section is a hamori song section by a melody part song and a chorus part 1 song. Similarly, when the hamori flag 2 is set to “1”, the section is a hamori song section by a melody part song and a chorus part 2 song. If it is not a humming song section, the humming flag 1 and the humming flag 2 are each set to “0”. t is a variable indicating the number of a minute section (for example, 20 ms in length) in the music from the first minute section, and is initialized at t = 0 first and then incremented sequentially until the music ends. Is done. The reference data note-on or note-off timing is detected as 0 or 1 within each minute interval.

  Next, each reference data is acquired by the sequencer process of the control unit 21, and the hammer flag 1 and the hammer flag 2 are set (S11). That is, if each reference data is a note-on timing, the control unit 21 acquires melody part reference data, chorus part 1 reference data, and chorus part 2 reference data from the music data storage area 23a of the storage unit 23. Then, the control unit 21 acquires reference pitches Pmr, Pc1r, and Pc2r from each of the acquired melody part reference data, chorus part 1 reference data, and chorus part 2 reference data, and a reference pitch storage area in the memory 21b. Store in 21d. That is, Pmr, Pc1r, and Pc2r stored in the reference pitch storage area 21d are always updated to the latest values.

  Further, if the melody part reference data is in the note-on state and the chorus part 1 reference data is in the note-on state, the control unit 21 sets the hammer flag 1 to “1”. That is, since the section is a section that a melody part singer should sing and a chorus part 1 singer should sing, the section is regarded as a chorus part 1 humming song section. Similarly, if the melody part reference data is in the note-on state and the chorus part 2 reference data is in the note-on state, the hammer flag 2 is set to “1”. That is, since the section is a section that a melody part singer should sing and a chorus part 2 singer should sing, the section is regarded as a chorus part 2 humming section. The settings of the hammer flag 1 and the hammer flag 2 are stored in the hammer flag storage area 21c. It can be determined by flag management whether or not each reference data is in note-on in each section (details are omitted).

  The reference pitches Pmr, Pc1r, and Pc2r may be in any format as long as the pitch can be recognized. In this embodiment, the reference pitches Pmr, Pc1r, Pc2r, and other types of pitches are acquired as cent values.

  As shown in FIG. 4B, the control unit 21 monitors the output from the sound processing unit 25. When a singing voice signal is input, the control unit 21 takes in the input singing voice signal and the melody part in the section. The pass / fail judgment of each chorus part 1 and chorus part 2 is performed (S12). For convenience, the singing voice signal of the melody part based on the microphone signal from the singing microphone 14A is referred to as a singing voice signal 0. Similarly, the singing voice signal of the chorus part 1 based on the microphone signal from the singing microphone 14B is referred to as a singing voice signal 1, and the singing voice signal of the chorus part 2 based on the microphone signal from the singing microphone 14C is referred to as the singing voice signal 2. First, the control unit 21 extracts the melody part pitch Pmk (t) included in the singing voice signal 0, and stores it as Pmk in the singing pitch storage area 21f of the memory 21b. Similarly, the control unit 21 extracts the chorus part 1 pitch Pc1k (t) included in the singing voice signal 1 and the chorus part 2 pitch Pc2k (t) included in the singing voice signal 2, and each of the songs in the memory 21b. The pitch is stored as Pc1k and Pc2k in the pitch storage area 21f. That is, Pmk, Pc1k, and Pc2k stored in the singing pitch storage area 21f are always updated to the latest values.

  Next, the melody part pass / fail determination process will be described. In this pass / fail determination process, the control unit 21 determines whether or not the melody part reference data is being turned on after updating Pmk in the singing pitch storage area 21f. Here, when it is determined that the note-on is being performed, since the melody part is a section to be sung, the control unit 21 performs pass / fail determination of the melody part singing in the section.

  In this pass / fail determination process, the control unit 21 reads the reference pitch Pmr from the reference pitch storage area 21d, reads the singing pitch Pmk from the singing pitch storage area 21f, and subtracts the reference pitch Pmr from the singing pitch Pmk. To obtain the deviation value. Then, the control unit 21 determines pass / fail of the minute section from the magnitude of the divergence value, and associates the pass / fail determination result with the value of t indicating the minute section and stores it in the melody part determination result storage area 21h of the memory 21b. To do.

  In the example of FIG. 5, at time ta, the control unit 21 calculates the singing pitch Pmk (ta) −reference pitch Pmr, and whether or not the obtained divergence value is in the range of −α1 to β1. Determine. In the present embodiment, the threshold values −α1 and β1 are cent values. In this case, the divergence value is smaller than 0 but larger than -α1, so that the minute section is determined to be acceptable. Similarly, at time tb, the control unit 21 calculates the singing pitch Pmk (tb) −reference pitch Pmr. Also in this case, the divergence value is smaller than 0 but larger than -α1, so that the minute section is determined to be acceptable.

  In this embodiment, the singing voice signal is not output at a volume higher than a specified value while the reference data is note-on, and Pmk (t) cannot be obtained, or the volume of the reference data is higher than the specified value during note-off. When Pmk (t) is acquired by outputting a singing voice signal, the problem is a rhythm (accumulation, previous twist, etc.), and it is excluded from the determination target. This also applies to the scoring process for each chorus part described later.

  Returning to FIG. 4B, the pass / fail determination process for chorus part 1 will be described. In this acceptance / rejection determination process, the control unit 21 determines whether or not it is a hamori song section after updating Pc1k of the song pitch storage area 21f. In this embodiment, this determination process is performed based on the hame flag 1 set in the hame flag storage area 21c of the memory 21b by the control unit 21 as setting means. As described above, the hamori flag 1 is set to “1” when it is a hamori song section, and is set to “0” when it is not a hamori song section.

  When it is determined that the hamori flag 1 = 1 (the section is the chorus part 1 hamori song section), the control unit 21 determines the reference pitch Pc1r of the chorus part 1 reference data and the reference pitch Pmr of the melody part reference data. Is read from the reference pitch storage area 21d, and Pmr is subtracted from Pc1r to obtain the reference tone pitch Dc1r.

  The control unit 21 reads the singing pitch Pc1k of the chorus part 1 and the singing pitch Pmk of the melody part from the singing pitch storage area 21f, and subtracts Pmk from Pc1k to obtain the singing tone pitch Dc1k.

  Next, the control unit 21 performs a pass / fail determination for the chorus part 1, that is, a pass / fail determination for the minute section. In this acceptance / rejection determination process, the control unit 21 obtains a divergence value by subtracting the reference hammer tone pitch Dc1r from the singing hammer tone pitch Dc1k. Then, the control unit 21 determines pass / fail of the minute section from the magnitude of the divergence value, and associates the pass / fail determination result with the value of t indicating the minute section and the identifier of the singing part, and the chorus part determination result of the memory 21b. The data is stored in the storage area 21j.

  In the example of FIG. 6, at the time tc, the control unit 21 calculates the singing hammer tone pitch Dc1k (tc) −reference hammer tone pitch Dc1r, and whether or not the obtained divergence value is in the range of −α2 or more and β2 or less. Determine. In the present embodiment, the threshold values −α2 and β2 are cent values. In this case, since the deviation value is larger than 0 but smaller than β2, the minute section is determined to be acceptable. Similarly, at the time td, the control unit 21 calculates the singing hammer tone pitch Dc1k (td) -reference hammer tone pitch Dc1r. In this case, the divergence value is smaller than 0 but larger than -α2, so that the minute section is also determined to be acceptable.

  Returning to FIG. 4B, the pass / fail determination process for chorus part 2 will be described. Even in this acceptance / rejection determination process, the control unit 21 determines whether or not it is a hamori song section after updating Pc2k in the song pitch storage area 21f. This determination is the same as the pass / fail determination process of the chorus part 1, but is performed based on the hammer flag 2.

  When it is determined that the hamori flag 2 = 1 (the section is the chorus part 2 hamori singing section), the control unit 21 sets the reference pitch Pc2r of the chorus part 2 reference data and the reference pitch Pmr of the melody part reference data. Is read from the reference pitch storage area 21d, and Pmr is subtracted from Pc2r to obtain the reference tone pitch Dc2r.

  The control unit 21 reads the singing pitch Pc2k of the chorus part 2 and the singing pitch Pmk of the melody part from the singing pitch storage area 21f, and subtracts Pmk from Pc2k to obtain the singing tone pitch Dc2k.

  Next, the control unit 21 performs a pass / fail determination for the chorus part 2, that is, a pass / fail determination for the minute section. In this acceptance / rejection determination process, the control unit 21 obtains a divergence value by subtracting the reference hammer tone pitch Dc2r from the singing hammer tone pitch Dc2k. Then, the control unit 21 determines pass / fail of the minute section from the magnitude of the divergence value, and associates the pass / fail determination result with the value of t indicating the minute section and the identifier of the singing part, and the chorus part determination result of the memory 21b. The data is stored in the storage area 21j.

  In the example of FIG. 7, at the time te, the control unit 21 calculates the singing hamori pitch Dc2k (te) −reference hamori pitch Dc2r, and whether or not the obtained divergence value is in the range of −α3 to β3. Determine. In the present embodiment, the threshold values −α3 and β3 are cent values. In this case, the divergence value is smaller than 0 but sufficiently larger than -α3, so that the minute section is determined to be acceptable. Similarly, at the time tf, the control unit 21 calculates the singing morimori pitch Dc2k (tf) −the reference morimori pitch Dc2r. In this case, since the deviation value is larger than 0 but smaller than β3, the minute section is also determined to be acceptable.

  Next, as shown to FIG. 4C, the control part 21 performs the scoring process of a melody part and each chorus part (S13). Then, the processes of S11 to S13 are repeated while incrementing t until the music performance is completed (N in S14).

  The scoring process of S13 includes a melody part scoring process, a chorus part 1 scoring process, and a chorus part 2 scoring process. Hereinafter, each scoring process will be described.

  The control unit 21 determines whether or not the melody part reference data is note-off timing. Here, when it is determined that it is the note-off timing, since it is the timing at which the pronunciation of a certain note constituting the melody part singing melody is to be terminated, the control unit 21 determines the corresponding note in the melody part. The scoring process is performed. For example, the control unit 21 reads out the pass / fail judgment results of all the minute sections included in the period of the notes to be scored from the melody part judgment result storage area 21h of the memory 21b, and calculates the ratio between the number of passes and the number of failures. Let the ratio of the number of passes be the score. Specifically, if the ratio of the minute section determined to be acceptable is 90% and the ratio of the minute section determined to be unacceptable is 10%, the scoring result for the note is 90 points. Then, the control unit 21 stores the scoring result for the note in the melody part scoring result storage area 21i of the memory 21b in association with the note identifier.

  At this time, if the hamori flag 1 = 1, that is, if the chorus part 1 is a hamori song section, the end of pronunciation of the note of the melody part means the end of the hamori song section of the chorus part 1, so the control unit 21 The scoring process for the corresponding note in Part 1 is performed, and the hammer flag 1 is set to “0”. Similarly, if the hamori flag 2 = 1, that is, if the chorus part 2 is a hamori singing section, the end of pronunciation of the note of the melody part means the end of the humor singing section of the chorus part 2, so that the control unit 21 performs the chorus part. The scoring process for the corresponding note in 2 is performed, and the hammer flag 2 is set to “0”. The scoring process for chorus part 1 and chorus part 2 will be described later.

  Next, the control unit 21 determines whether or not the chorus part 1 reference data is the note-off timing and the hammer flag 1 = 1. Here, when it is determined that it is the note-off timing and the hammer flag 1 = 1, it is the timing at which the pronunciation of a certain note constituting the singing melody of the chorus part 1 should be terminated, so that the control unit 21 Performs the scoring process for the corresponding note in chorus part 1 and sets the hammer flag 1 to “0”. For example, the control unit 21 reads out the pass / fail judgment results of all the minute sections included in the period of the note to be scored from the chorus part judgment result storage area, calculates the ratio of the number of passes and the number of passes, The ratio is a score. Specifically, if the ratio of the minute section determined to be acceptable is 80% and the ratio of the minute section determined to be unacceptable is 20%, the scoring result for the note is 80 points. Then, the control unit 21 stores the scoring result for the note in the chorus part scoring result storage area 21k of the memory 21b in association with the note identifier and the chorus part 1 identifier.

  Further, the control unit 21 determines whether or not the chorus part 2 reference data is the note-off timing and the hammer flag 2 = 1. Here, when it is determined that the note-off timing and the hammer flag 2 = 1, it is the timing at which the sound generation of a certain note constituting the singing melody of the chorus part 2 should be terminated. Performs the scoring process for the corresponding note in the chorus part 2 and sets the hammer flag 2 to “0”. For example, the control unit 21 reads out the pass / fail judgment results of all the minute sections included in the period of the note to be scored from the chorus part judgment result storage area, calculates the ratio of the number of passes and the number of passes, The ratio is a score. Specifically, if the ratio of the minute section determined to be acceptable is 75% and the ratio of the minute section determined to be unacceptable is 25%, the scoring result for the note is 75 points. Then, the control unit 21 stores the scoring result for the note in the chorus part scoring result storage area 21k of the memory 21b in association with the note identifier and the chorus part 2 identifier.

  If the processes of S11 to S13 described above are performed until the performance of the music is completed (Y in S14), the control unit 21 notifies the scoring result (S15). For example, the control unit 21 calculates the score for each of the melody part, the chorus part 1 and the chorus part 2 by summing up the scoring results of all the notes in the singing melody of the karaoke song, and displays the calculated score on the monitor 13. Let In addition, the control unit 21 controls the communication unit 22 and the router device RT, and transmits a scoring result to the host device (not shown) through the transmission path 2. In the host device, the received scoring result is added to the singing history of the singer.

  Thus, according to the karaoke apparatus 1 of the present embodiment, the control unit 21 (chorus part scoring means) is configured to extract the singing pitch Pmk (t) of the extracted melody part and the singing pitch Pc1k (t) of each chorus part. ), Pc2k (t), the pitch difference between the pitches Dc1k (t) and Dc2k (t) is calculated as a singing hammer tone pitch, and the reference pitch Pmr and each chorus of the melody part reference data read from the storage unit 23 (storage means). Differences in pitch between the reference pitches Pc1r and Pc2r of the part reference data are calculated as reference pitches Dc1r and Dc2r. Then, the control unit 21 allows the chorus part singer to perform the melody part singing pitch with respect to the actual melody part singing pitch, based on the degree of coincidence between the singing humor pitches Dc1k (t) and Dc2k (t) and the reference humor pitches Dc1r and Dc2r. Evaluate and score whether the pitch is correct. As a result, the singing ability of the chorus part singer, that is, the skill of the hammer can be appropriately evaluated.

  In this karaoke apparatus 1, the control unit 21 (extracting means) extracts the singing pitch Pmk (t) of the melody part and the singing pitches Pc1k (t) and Pc2k (t) of each chorus part into small sections. ing. Then, the control unit 21 (chorus part scoring means) calculates the singing hamori pitches Dc1k (t) and Dc2k (t) for each minute interval, and the minute interval in which the difference from the reference hammeriness intervals Dc1r and Dc2r is within the acceptable range. The degree of coincidence is determined based on the ratio of the whole note interval (scoring interval) to the entire minute interval. In this way, the singing pitches Dc1k (t) and Dc2k (t) are calculated for each minute section, and the scoring is performed based on the ratio of the minute sections determined to be acceptable to all the minute sections. Appropriate scoring results can be obtained even if.

  In addition, the melody part reference data and each chorus part reference data include note-off indicating the separation of each note in the singing melody, and the control unit 21 (chorus part scoring means) includes the melody part reference data and the chorus part reference data. Since the chorus part singing is scored for each one of the note breaks, the singing of each chorus part can be performed in units of notes in the melody part and chorus part singing melody. As a result, not only the melody part singing pitch and the chorus part singing pitch change at the same timing while taking a predetermined pitch, but the chorus part singing pitch is constant and only the melody part singing pitch is In the case of changing or vice versa, scoring is performed each time the pitch of the two changes, so the skill of the hammer can be evaluated more precisely.

  Moreover, the control part 21 (setting means) has set the hamori flag 1 and the hamori flag 2 which show whether it is a hamori song section according to the karaoke performance of a music. And the control part 21 (chorus part scoring means) scores the song of each chorus part, when the set hamori flag 1 or the hamori flag 2 shows a hamori song section. In this way, the use of the hamori flag 1 and the hamori flag 2 indicating whether or not it is a hamori singing section makes it possible to simplify the recognition processing of the hamori singing section.

  The above description of the embodiment is for facilitating the understanding of the present invention, and does not limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and the present invention includes equivalents thereof. For example, you may comprise as follows.

  First, in the above-described embodiment, the singing part of each singer and the singing microphones 14A to 14C are associated with each other in advance, but the singing voice of each singer is compared with the reference data of the melody. The singing voice of the singer may be identified as the singing voice of the melody part, and the other as the singing voice of the chorus part.

  For each chorus part, evaluation and scoring may be performed based on the degree of coincidence between the pitch of the melody part and the reference pitch of the reference pitch and the chorus part. Alternatively, the reference tone pitch may be calculated in advance, and only the melody part reference data and the reference tone pitch may be stored in the storage unit 23.

  Also, the scoring results based on the degree of coincidence between the chorus part singing pitch and the reference pitch of the chorus part reference data may be used in combination for evaluation. Or you may employ | adopt as a scoring result whichever is higher or lower rather than using together.

  In the above-described embodiment, the singing section without the chorus part is not particularly mentioned. For example, the following processing can be performed. First, the chorus part singer can sing at the same pitch as the melody part. In this case, since the reference pitch is zero, the evaluation is based on the degree of coincidence between the singing pitch of the melody part and the singing pitch of the chorus part, and the scoring is performed. Similarly, even if the chorus part singer sings at the same pitch as the melody part singing pitch, the evaluation based on the degree of coincidence between the pitch of the reference data for the melody part and the singing pitch of the chorus part, It is good also as a scoring. In other words, it is the same idea as the prior art. Also, chorus part singers may not sing

  In the above-described embodiment, the chorus parts 1 and 2 are used, but there may be three or more chorus parts. In this case, a singing hamori pitch and a reference hamori pitch may be calculated between each chorus part and the melody part.

  In the above-described embodiment, when there are a plurality of chorus parts, the singing humor pitch and the reference humor pitch are calculated between each chorus part and the melody part. It may be calculated. For example, in the case of three chorus parts, a singing hamori pitch and a reference hamori pitch are calculated between chorus part 1 and the melody part. Alternatively, the singing humor pitch and the reference humor pitch may be calculated between the chorus part 2 and the chorus part 1, and the singing humor pitch and the reference humor pitch may be calculated between the chorus part 3 and the chorus part 2.

  Moreover, even when singing one chorus part with two or more singers, the singing voices of each singer can be input by the singing microphones 14 for the number of persons, and can be independently evaluated and scored.

  With respect to the threshold values -α (-α1 to -α3) and β (-β1 to -β3), they may be determined according to the difficulty level (singing range, note density, tempo, rhythm, etc.) of each singing part. Moreover, you may correct | amend according to singer's attribute information (age, sex, singing history, etc.) and the attribute information (music genre, original music singer, etc.) of music.

  Record the new singing video while downloading and playing the uploaded and released singing video, synthesize each video and singing audio, upload again, and perform the scoring process of this embodiment in the so-called collaboration video, You may provide the content which competes for the scoring result of the chorus part singer who collaborates with the melody part which the predetermined singer sang.

DESCRIPTION OF SYMBOLS 1 ... Karaoke apparatus, 2 ... Transmission path, 11 ... Karaoke main body, 12 ... Speaker, 13 ... Monitor, 14 (14A-14C) ... Singing microphone, 15 ... Remote control device, 21 ... Control part, 21a ... CPU, 21b ... Memory , 21c... Hamori flag storage area, 21d. Reference pitch storage area, 21f. Singing pitch storage area, 21h ... Melody part determination result storage area, 21i ... Melody part scoring result storage area, 21j ... Chorus part determination result storage area, 21k ... Chorus part scoring result storage area, 22 ... Communication section, 23 ... Storage section, 23a ... Music data storage area, 23b ... Scoring program storage area, 24 ... Sound source section, 25 ... Sound processing section, 26 ... Display processing section 27 ... Operating unit, RT ... Router device, BS ... Bus, Pmr ... Reference pitch, Pc1r ... Reference High, Pc2r ... reference pitch, SP0 ... melody part pitch group, sp0 (t) ... sample pitch, Pmk (t) ... singing pitch, SP1 ... chorus part 1 pitch group, sp1 (t) ... sample sound High, Pc1k (t) ... singing pitch, SP2 ... chorus part 2 pitch, sp2 (t) ... sample pitch, Pc2k (t) ... singing pitch, Dc1r ... reference pitch, Dc2r ... reference pitch, Dc1k (t) ... Singing harpoon pitch, Dc2k (t) ... Singing harpoon pitch

Claims (4)

A karaoke scoring device for scoring karaoke songs sung by one melody part singer and one or more chorus part singers according to the karaoke performance of the music,
Melody part reference data for evaluating a song of the melody part, including data of at least pitch, and data for evaluating a song of the chorus part, the chorus part reference data corresponding to the chorus part; Storage means for storing
Extracting means for independently extracting the singing pitch from the singing voice signal of the melody part and the singing voice signal of the chorus part;
Melody part scoring means for scoring the melody part singing according to the degree of coincidence between the melody part singing pitch and the pitch of the melody part reference data;
For the chorus part, a singing tone pitch that is a pitch difference between the singing pitch of the chorus part and the singing pitch of the melody part is calculated, and the pitch of the chorus part reference data and the sound of the melody part reference data are calculated. A chorus part scoring means for scoring the chorus part singing according to the degree of coincidence between the singing hammer sound pitch and the reference hammer sound pitch;
A karaoke scoring device comprising: The extraction means extracts the singing pitch of the melody part and the singing pitch of the chorus part divided into minute sections,
The chorus part scoring means calculates the singing humor pitch for each minute section, and the ratio of the minute section whose difference from the reference humor pitch is within the acceptable range to the entire minute section in the scoring section is the same The karaoke scoring device according to claim 1, wherein the karaoke scoring device is a degree. The melody part reference data and the chorus part reference data include information indicating a break of each note in a singing melody,
The chorus part scoring means scores the chorus part singing for each break of one of the notes in the melody part reference data and the chorus part reference data. Karaoke scoring device. According to the karaoke performance of the music, further comprising a setting means for setting information indicating whether or not it is a hamori song section,
4. The chorus part scoring means scores the chorus part singing when the information set by the setting means indicates the hamori singing section. 5. Karaoke scoring device.

Images