JPH09127968A - Automatic scoring device for karaoke device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic scoring device for a karaoke device capable of clearly and accurately evaluating the singing skill of a song by awarding marks according to the note number difference corresponding to the drift degree between an accompaniment and the musical step and the musical pitch based on the note number. SOLUTION: A pitch calculation section CPU1 continuously stores eight musical note numbers S10 and eight accompaniment note numbers S14 respectively and extracts and stores those having the largest number of appearance times. Eight note numbers S10 are stored as one partition, the musical note number S10 and the accompaniment note number S14 having the largest number of appearance times for each partition interval are stored respectively, and the difference between the musical note number S10 and the accompaniment note number S14 extracted and stored for each partition interval is obtained. The key adjustment signal S5 is generated as required after one chorus is finished. This key adjustment quantity is automatically reflected on the scoring about the musical pitch.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention automatically scores a song that is sung in accordance with the performance of a karaoke device, and therefore can be used for evaluation of good or bad singing of a song and singing lessons. The present invention relates to an automatic scoring device for a device (hereinafter, referred to as an “automatic scoring device” as appropriate), and particularly to a technique for accurately scoring a sung song.

[0002]

2. Description of the Related Art In recent years, it has become very popular to sing along with the accompaniment of a karaoke apparatus, as shown in the popularization of karaoke houses. However, it is not enough to just sing along with the accompaniment. Therefore, an automatic scoring device for a karaoke device is sometimes placed together with the karaoke device in order to increase the satisfaction of the singer (customer). The automatic scoring device for a karaoke device is a device for automatically scoring a sung song as soon as the customer finishes singing a song and evaluating the customer's song.

[0003]

However, since the conventional automatic scoring device is an evaluation system that scores by observing the tempo of the song and the voice volume of the singer, it is said that it accurately indicates the skill of the song. The fact is that the singer has not reached the point where he is completely satisfied.

In view of the above circumstances, it is a first object of the present invention to provide an automatic scoring device for a karaoke device, which can make an accurate evaluation so that the good or bad of the sung song can be clearly understood. The second problem is to provide an automatic scoring device for a karaoke device in which the evaluation of the karaoke device is performed quickly, and a third problem is to provide an automatic scoring device for the karaoke device in which a karaoke accompaniment that is easy for the singer to sing is played. And

[0005]

[Means for Solving the Problems] In order to solve the above-mentioned problems, the inventors of the present invention have conducted extensive studies and as a result, have determined MIDI (Musical Instrument) that is specified for electronic musical instruments such as synthesizers.
We paid attention to the existence of karaoke software that applied the Digital Interface standard. In the case of the karaoke software to which the MIDI standard is applied, which is considered to be already used in about half of the karaoke devices at present, performance information is constituted by a digital signal which can be transmitted through a telephone line, which is typical. As an example of a digital signal, as shown in FIG. 13, it is composed of 3 bytes, and the first 1 byte is the status byte, the middle 1 byte is the first data byte for note number (pitch), and the end. 1 byte is the second data byte for velocity (volume).

As shown in the upper part of FIG. 14, in the case of serial transmission, the status byte is followed by the first and second data bytes and is transmitted with the most significant bit (MSB) side as the head. In the present invention, serial / parallel conversion is performed and the lower signal in FIG. 14 is used. In the status byte, the upper 4 bits (MSB to 4SB) specify a message, and the lower 4 bits (5SB to LSB) specify a channel (instrument specification) so that a performance system configuration with a plurality of electronic musical instruments can be configured. . The MSB (most significant bit) of the first and second data bytes is always 0, and the remaining 7 bits specify the scale (pitch) or volume.

In order to play one note, a sound generation start (note on) and sound generation stop (note off) signal is required. In the note-on message, as shown in FIG. 13A, the upper 4 bits of the status byte are “1001”.
The message for note-on is shown in Fig. 13
As shown in (b), the upper 4 bits of the status byte are set to "1000". Each MSB "1" will catch the arrival of a signal. In the case of a note-on signal, the pitch and volume are designated by the first and second data bytes. On the other hand, in the case of the note-off signal, the scale designation of the first data byte is left as it is and the volume designation of the second data byte is set to "00000000" and the volume is zero.

In the case of serial transmission, as shown in FIG. 14, one note-on signal of about 1 millisecond (mSEC) and about 1
When one note-off signal of millisecond is combined with the note-on signal, the sound starts to sound, and when the note-off signal arrives, the sound stops and the single note is played. There is usually about 300 milliseconds or more between the note-on signal and the note-off signal. When a set of note-on and note-off signals ends, the note-on and note-off signals for playing the next note arrive repeatedly in the same manner. In this way, the karaoke performance is performed.

Then, the inventors of the present invention have sung a melody (main melody) signal in a number of channels designated by the lower 4 bits of the status byte, that is, a signal for playing the same melody as the singer's song (in the present invention, Noted that there is a channel for accompaniment signals). Usually, the attention is paid to the first channel designated by the message "0000" of the lower 4 bits of the status byte. The note-on signal and note-off signal shown in FIGS. 13 and 14 are accompaniment signals from the first channel. Among the signals of other channels for playing the non-main melody, there is a signal composed of 4 bytes or more. As a result of further diligent study on the accompaniment signal, the inventors have found that the accompaniment signal is a song melody. The inventors found that the accompaniment signal is a good pitch standard for measuring the degree of pitch deviation of an actual song.
The present invention has been completed.

From the above, the present invention for solving the first problem has the following configuration. That is, the automatic scoring device for a karaoke device according to claim 1 is an accompaniment note number creating means for creating an accompaniment note number indicating a note number in an accompaniment signal according to a standard set for an electronic musical instrument, and a song according to the accompaniment. A song note number creating means for creating a song note number indicating the pitch of a song signal obtained from a song being played in the form of a note number conforming to the standard, and a note number difference calculation for obtaining a note number difference between the accompaniment note number and the song note number. Means and pitch scoring means for scoring the pitch of the song based on the note number difference.

According to a second aspect of the present invention, in the automatic scoring device for a karaoke device according to the first aspect, tempo difference detecting means for detecting a tempo difference between the tempo in the accompaniment signal and the tempo in the song signal, and the tempo difference. And a tempo scoring means for scoring the tempo of the song based on.

According to a third aspect of the present invention, in the automatic scoring device for a karaoke device according to the first or second aspect, a frequency detecting means for detecting the frequency of appearance of a harmonic in a song signal per unit time, and the harmonic detecting unit And a tone color scoring means for scoring the tone color of a song based on the frequency of appearance.

The invention as defined in claim 4 is based on claims 1 to 3.
In the automatic scoring device for karaoke device according to any one of the above, intensity measuring means for measuring the intensity of the song signal, and a voice volume scoring means for scoring the voice volume of the singer based on the intensity measurement result of the song signal. It is equipped with.

The invention described in claim 5 is from claim 1 to claim 4.
In the automatic scoring device for a karaoke device described in any of the above, a fermata time difference detecting means for detecting a time difference between the fermata (reverberation) time in the accompaniment signal and the fermata time in the song signal, and a song based on the time difference between both fermata times. And a fermata scoring means for scoring the fermata.

[0015] The invention according to claim 6 is the invention according to claims 1 to 5.
In the automatic scoring device for karaoke device according to any one of the above, a rest time difference detection means for detecting a time difference between the rest time in the accompaniment signal and the rest time in the song signal, and the time difference between the rests. Rest scoring means for scoring the rest of the song based on the above.

The invention described in claim 7 is from claim 1 to claim 6.
In the automatic scoring device for the karaoke device according to any one of the above, the pitch of the song, the tempo of the song, the tone of the song, the voice of the singer, the fermata of the song, and, based on each scoring result about the rest of the song, It is equipped with a comprehensive scoring means for comprehensively scoring songs.

[0018] The invention described in claim 8 is the invention according to claims 1 to 7.
In the automatic scoring device for a karaoke device described in any one of the above, the song note number creating means for the electronic musical instrument according to the frequency discriminating means for discriminating the frequency of the song signal and the discriminating frequency by the frequency discriminating means. It is provided with a pitch code assigning means for assigning a digital pitch code signal corresponding to the digital pitch code of the standard defined in 1.

According to the ninth aspect of the present invention, there is provided an image processing apparatus comprising:
In the automatic scoring device for a karaoke device described in any of the above, based on the note number difference, a key that creates a key adjustment signal for driving the key controller of the karaoke device so as to eliminate the difference between the accompaniment and the singer's key The adjustment signal creating means is provided.

[0019]

The operation of the automatic scoring device for a karaoke machine of the present invention is as follows. In the case of the invention according to claim 1, the accompaniment note number creating means creates an accompaniment note number indicating a note number (digital pitch code) in an accompaniment signal conforming to the electronic musical instrument standard at the same time when the karaoke performance is started, and at the same time, sings the song. The note number creating means creates a song note number indicating the pitch of the song signal obtained from the song actually sung in accordance with the accompaniment in the form of a note number conforming to the electronic musical instrument standard. On the other hand, the note number difference calculation means obtains the note number difference between the accompaniment note number and the song note number obtained one after another. Furthermore, the pitch of the song is scored by the pitch scoring means from the obtained note number difference.

According to the second aspect of the present invention, the tempo difference detecting means detects the tempo difference between the tempo in the accompaniment signal and the tempo in the song signal, and the tempo scoring means from the determined tempo difference. The song tempo is scored.

According to the third aspect of the present invention, while the frequency detection means detects the frequency of appearance of the harmonic in the song signal per unit time, the timbre scoring means calculates the song from the obtained frequency of appearance of the harmonic. The timbre of the sound is scored.

According to the invention described in claim 4, the intensity of the singing signal is measured by the intensity measuring means, while the voice volume scoring means scores the voice volume of the singer based on the obtained intensity measurement result of the song signal. It

According to the invention of claim 5, the difference between the fermata time in the accompaniment signal and the fermata time in the singing signal is detected by the fermata time difference detecting means, and the fermata score is obtained from the obtained difference between both fermata times. By means of the song, the scoring of the song Fermata is done.

According to the sixth aspect of the present invention, the rest time difference detecting means detects the difference between the rest time in the accompaniment signal and the rest time in the song signal, and at the same time, finds the difference in rest time obtained. From the rest rest scoring means, the rest of the song is scored.

According to the invention described in claim 7, by the total scoring means, all scoring results for the song pitch, the song tempo, the timbre of the song, the voice volume of the singer, the fermata of the song, and the rest of the song. Scores are made from a comprehensive perspective such as the average value of.

According to the eighth aspect of the invention, after the frequency discrimination means discriminates the frequency of the song signal obtained with the singer's song, the pitch code imparting means responds to the discrimination frequency in accordance with the discrimination frequency. A digital pitch code signal corresponding to the digital pitch code of the electronic musical instrument standard is added to create a song note number for the song signal.
The note number difference necessary for pitch scoring can be obtained by subtraction calculation of both digital pitch codes.

According to the ninth aspect of the present invention, when there is a difference between the accompaniment and the singer's keys, the key adjustment signal creating means transmits a key adjustment signal to the key controller of the karaoke apparatus to accompaniment and singer's keys. Eliminate the difference between.

[0028]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of an automatic scoring device for a karaoke device of the present invention (hereinafter appropriately referred to as "automatic scoring device") will now be described in detail with reference to the drawings. FIG.
FIG. 1 is a block diagram showing a schematic configuration of an entire karaoke system in which an automatic scoring device according to an embodiment is incorporated. FIG.
Is a block diagram showing a configuration of a signal source device, FIG. 3 is a block diagram showing a configuration of a mixing amplifier device, and FIG. 4 is a block diagram showing a configuration of an automatic scoring device.

In the karaoke system shown in FIG. 1, the performance signal S1 is sent from the signal source device 1 to the speaker 3 through the mixing amplifier device 2 so that the karaoke performance is played from the speaker 3 while the video signal is sent from the signal source device 1. The signal S2 is sent to the display device 4 via the mixing amplifier device 2, and an image is displayed on the display device 4 at the same time as the karaoke performance. And the singer's song is the song signal S3
Is sent from the microphone 5 to the speaker 3 via the mixing amplifier device 2. From the speaker 3, the singer's song is played on the karaoke performance.

On the other hand, a performance signal S1, an accompaniment signal S4 and a song signal S3 are sent to the scoring device 6, and based on these signals, the scoring device 6 makes various scoring items including the pitch of the singer's song. Is performed, and the result is sent to the display device 4 for display, and the key adjustment signal S5 is sent from the scoring device 6 to the mixing amplifier device 2. Hereinafter, the configuration of the main part will be described in more detail.

In the signal source device 1 shown in FIG. 2, each signal of channels 1 to 16 is sent from a MIDI sound source (a disk storing karaoke software, etc.) 7 to a karaoke interface 8 for signal processing, and then D / The analog signal is converted by the A converter 9 and then sent to the low-pass filter 10. On the other hand, the signal of channel 1 which is an accompaniment signal is a bandpass filter 1 having a pass band of 22 kHz to 42 kHz.
An interface 12 is used to prepare a serial signal as shown in the upper side of FIG. The analog signal from the low pass filter 10 is the performance signal S1, and the digital signal from the interface 12 is the accompaniment signal S4. Further, the video signal S2 is also output from the video signal source 13.

In the mixing amplifier device 2 shown in FIG.
After the performance signal S1 is amplified by the preamplifier 15, the key controller 16 passes the mixing amplifier 17, and the power amplifier 19 sends it to the speaker 3. On the other hand, the video signal S2 is sent through the video distributor 14 provided as necessary to the display device. Send to 4. The singing signal S3 is amplified in advance by the microphone amplifier 18, and then the mixing amplifier 17 adds the singing signal S3 to the performance signal S1 and sends it from the power amplifier 19 to the speaker 3.

Next, the automatic scoring device 6, which is a feature of the present invention, will be specifically described with reference to FIG. (1) First, a configuration for creating a song note number indicating the pitch in the song signal S3 in the form of a note number according to the MIDI standard will be described. The analog singing signal S3 obtained through the microphone 5 is adjusted to an appropriate level by the amplifier 21, and then converted into a parallel digital signal S6 of about 8 bits by the A / D converter 22, and the singing pitch signal extracting section 23. The frequency of the song signal S3 is discriminated. With respect to the discrimination frequency of the song signal S3, the song MIDI signal generating section 24 gives a digital pitch code, and finally: the song note number / flag creating section 25 gives a song note number in the form of a note number conforming to the MIDI standard. S8 will be output.

As shown in FIG. 5, in the singing pitch signal extracting section 23, 37 frequency discriminating sections BF1 to BF37 are provided in parallel after the low pass filter LF1. Each frequency discriminating unit is a series connection circuit of a bandpass filter, an absolute value converter, and an OR element, and each bandpass filter has a different pass band. 110H
The bandpass filters each have a pass band corresponding to one of 37 pitches in the range of z to 880 Hz and about 3 octaves, and the pitch of the parallel digital signal S6 is 1: 1 with the pitch of the parallel digital signal S6. The frequency discrimination corresponding to is performed. Each frequency discriminator BF1 to BF37
16 shows the correspondence between the numbers (No) and frequencies. If the parallel digital signal S6 from which harmonics that exceed the object to be discriminated by the low-pass filter LF1 are removed in advance is, for example, 220 Hz, the frequency discriminator BF corresponding thereto
The output signal SE13 of 13 becomes "H".

A digital code is given by the song MIDI signal generating section 24 shown in FIG. 6 to each frequency from which "H" discriminated by each frequency discriminating section BF1 to BF37 is output. The important point here is that the song MIDI signal generator 24
The correspondence relationship (algorithm) between the digital code and the frequency (pitch) given in step 3 is that it matches the correspondence relationship between the digital code of the note number in the accompaniment signal S4 and the pitch.

In the song MIDI signal generating section 24, the same number of coding sections CD1 to CD1 as the frequency discriminating sections BF1 to BF37 are provided.
CD37 is provided, and each coding unit CD1 to CD
At 37, the bit connected to "L" is always a code value of "0" and the bit connected to the output of the AND element is given a code value by the AND element. The output signal SE of the frequency discriminator corresponding to one input of each AND element
1 to SE37 are given respectively, and the control signals S7-1 to S7-37 are given respectively to the other inputs. When both signals input to the AND element become "H", the output becomes "H" and the code value becomes "1". If both signals input to the AND element are "L" even if one is output, the output becomes "L" and the code value becomes "0".

For example, the output signal S of the frequency discriminator BF1
If E1 and the control signal S7-1 of the tone calculation unit CPU3 are both "H", coding section CD1 is, as shown in the upper part of FIG. 6, (corresponding to A ₂ in FIG. 16) a digital code SF1 Will be output. Upper control signal S7
Is a signal for avoiding confusion caused by a plurality of coding sections simultaneously outputting digital pitch code signals for one song signal. The output signals of the plurality of frequency discriminators BF1 to BF37 may be "H" at the same time because of the double frequency relationship. In this case, it is confusing that different digital code signals are simultaneously output from the plurality of encoders. Cause The output signals SE1 to SE37 of the frequency discriminating units BF1 to BF37 are also sent to the tone color computing unit CPU3, and only the digital code corresponding to the frequency discriminating unit from which the tone color computing unit CPU3 first outputs "H" is output. As described above, only one line of the control signals S7-1 to S7-37 is set to "H". The tone color calculation unit CP
U3 outputs one of the control signals S7-1 to S7-37 so that only the digital code corresponding to the lowest frequency discriminating unit among the frequency discriminating units that output "H" is output. It may be that only the line is “H”.

In response to the digital code SF from the song MIDI signal generating section 24, the song note number / flag creating section 25 shown in FIG. 7 converts the song note number S8 of the note number conforming to the MIDI standard into an 8-bit parallel signal. To output. The song note number / flag creation unit 25 includes eight OR elements having 37 input terminals, and as shown in FIG. 7, the OR elements corresponding to the MSBs of the parallel signals have the output signals SE1 to SE37 of the frequency discriminator. The OR element corresponding to 2SB has a total of 3 coding units.
Seven 2SBs, OR elements corresponding to 3SBs receive a total of 37 3SBs of coding units, and OR elements corresponding to LSBs receive a total of 37 LSBs of coding units. Has been done. The song note number S8 is a signal having the same code as each digital code SF except that the MSB of each digital code SF is "1". Therefore, the song note number S8 is a signal indicating a pitch according to the MIDI standard. While the song note number S8 is output, M
Since SB is always "1", M of the song note number S8
SB also becomes a song note-on signal (flag) S9 indicating the occurrence of the song note number. Therefore, by counting the number of song note-on signals S9, the number of song note numbers S8 can be known.

The song chorus signal creating section 26 inputs the song note-on signal S9 and outputs a song chorus signal S10 which maintains "H" for one chorus. In the present invention, the chorus refers to one song melody (main melody), and in the case of a song having lyrics up to the third, the main melody is played three times, resulting in a three-chorus structure. As shown in FIG. 7, in the song chorus signal creation unit 26, while the song note-on signal S9 is "H", the integration circuit stores electricity,
If the output of the integrating circuit is equal to or more than a certain value, the output of the voltage comparing circuit becomes "H". During one chorus, the song note-on signals S9 are input to the integrator circuit one after another and the integrator circuit continues to maintain an output above a certain level, so that the output of the voltage comparison circuit maintains "H". There are occasional rest times of about 3 seconds during one chorus, but by adjusting the time constant of the integrating circuit, the voltage comparison circuit maintains "H" for rest times of about 3 seconds, and the song chorus signal S10 So that there is no interruption.

(2) Next, a configuration related to the creation of an accompaniment note number indicating the note number in the accompaniment signal will be described. The accompaniment signal S1 for the song melody is the accompaniment M of FIG.
The IDI signal conversion unit 27 causes the parallel digital signal S
After being converted into 12, it is sent to the accompaniment note number / flag creation unit 28. Accompaniment note number / flag creation unit 28
Then, as shown in FIG. 8, the accompaniment note-on signal (flag) S13 and accompaniment note number S14 which are “H” during the occurrence of the accompaniment note number are utilized by using the three signals of status, note number and velocity. create. The accompaniment note-on signal S13 is a signal that becomes “H” when the note-on signal arrives and becomes “L” when the note-off signal arrives. Status MS when a note-on signal arrives
B and 4SB are "H", and at least one bit of both the note number and velocity is "H".
Therefore, three AND elements 1 to 3 and three ORs
The outputs of the elements 1 to 3 are all "H", and the output of the D flip-flop 1 is "H". When the note-off signal arrives, even if the status 4SB is "L", the MSB is "H", all the velocity bits are "L", but at least one bit of the note number is "H". Therefore, although the outputs of the AND elements 1 and 3 and the OR element 2 are all "L", the outputs of the AND element 2 and the OR elements 1 and 3 are "H", and the output of the D flip-flop 1 is "L". Invert. When the output from the D flip-flop 1 becomes "H", it means that the accompaniment note-on signal S13 is generated. Therefore, only one "H" pulse of the accompaniment note-on signal S13 causes the accompaniment note number S1.
This means that one 4 is out.

In the accompaniment note number S14, as shown in FIG. 8, the note number 2S of the incoming note-off signal is input.
B to LSB form 2SB to LSB as they are, accompaniment note-on signal S13 forms MSB, and accompaniment note number S14 is an 8-bit parallel digital signal. The MSB of the status of the incoming note-on signal is latched by the D flip-flop 1 and output as "1", while "1" of 2SB to LSB is latched by the D flip-flops 2-8 and is "1". Is output as "0" of 2SB to LSB.
Is output as "0" as it is, and the accompaniment note number S14 is output. The correspondence between the pitch and the digital chord in the accompaniment note number S14 is as shown in FIG. 16, and matches the song note number S8.

As shown in FIG. 8, the accompaniment chorus signal forming section 29 inputs the accompaniment note-on signal S13 and outputs an accompaniment chorus signal S15 which maintains "H" while the accompaniment continues for one chorus. In the accompaniment chorus signal creation unit 29,
While the accompaniment note-on signal S13 is "H", the integration circuit stores electricity, while if the output of the integration circuit is equal to or more than a certain value, the output of the voltage comparison circuit becomes "H". For one chorus,
Since the accompaniment note-on signal S13 is input to the integrator circuit one after another and the integrator circuit can keep the output above a certain level,
The output of the voltage comparison circuit maintains "H". There are occasional rest times of about 3 seconds during one chorus, but by adjusting the time constant of the integrating circuit, the voltage comparison circuit maintains "H" for rest times of about 3 seconds, and the accompaniment chorus signal S15 So that there is no interruption.

(3) Next, a description will be given of a configuration related to the creation of a song tempo signal and an accompaniment tempo signal necessary for scoring the song tempo. As shown in FIG. 9 (a), the song tempo creating unit 31 that creates the song tempo signal has three toggle flip-flops connected in series and receives the song note-on signal S 9 as an input, while the song chorus is input. In this configuration, the signal S10 is used as a reset input. The output of the toggle flip-flop in the third stage is the song tempo signal S17. This song tempo signal S17 is "H" every time four song note-on signals S9 are input, as shown in FIG. 9 (b).
And "L" are repeated digital signals.

As shown in FIG. 10A, the accompaniment tempo creating section 32 for creating the accompaniment tempo signal has three toggle flip-flops connected in series, and an accompaniment note-on signal created based on the accompaniment signal. While S13 is input, the accompaniment chorus signal S15 is reset input. The output of the toggle flip-flop in the third stage is the accompaniment tempo signal S18. As shown in FIG. 10B, it is a digital signal in which "H" and "L" are repeated every time four accompaniment note-on signals S13 are input.

(4) The configuration related to the extraction of harmonics by song pitch is as follows. As shown in FIG. 5, the high-pass filter HF1 is used in the harmonic signal extracting section 33 for each song pitch.
34 frequency discriminating units BF38 to BF71 are provided in parallel at the subsequent stage. Then, the frequency discriminator BF38-
The output signals SE38 to SE71 of the BF71 are output together with the output signals SE1 to SE37 of the frequency discriminating units BF1 to BF37 as a harmonic signal S19 for each pitch. Each frequency discriminating unit is a series connection circuit of a bandpass filter, an absolute value converter, and an OR element, and each bandpass filter has a different pass band. 932Hz
Frequency discriminator BF1 in the range of about 6160 Hz
~ Corresponding to the pitch of the parallel digital signal S6 in a one-to-one correspondence with the frequency of the parallel digital signal S6 by the action of the bandpass filters each having a pass band corresponding to one of the 34 pitches corresponding to the higher order frequencies (pitch) of the BF37 Frequency discrimination is performed. FIG. 17 shows the correspondence between the frequency discriminators BF38 to BF71 and the pitch. When the frequency of the parallel digital signal S6, in which the low frequency components exceeding the object to be discriminated by the high-pass filter HF1 are removed in advance, is, for example, 988 Hz, the output signal SE3 of the corresponding frequency discriminating unit BF39.
9 becomes “H”.

(5) The configurations related to the extraction of the song rest signal and the accompaniment rest signal will be described. In the case of the song rest signal extraction unit 35, as shown in FIG. 11A, the song note-on signal S9 and the song chorus signal S10 are input signals,
The song note-on signal S9 is integrated by the integrator circuit, while the output of the integrator circuit goes through a voltage comparator that inverts when the output exceeds a certain level.
Input to one input of D element. A song chorus signal S10 is input to the other input of the AND element, and a song rest signal S21 which is "H" only during rests is obtained at the output of the AND element, as shown in FIG. 11 (b). . During the rests in which the song is cut off for a long time, the input of the song note-on signal S9 is cut off, so that the output of the integrating circuit drops below a certain level, and as a result, it becomes "H" only during the rests. The reason why the song chorus signal S10 is ANDed is to avoid the inconvenience that a song rest signal is erroneously output during an interlude period when there is no song signal.

In the case of the accompaniment rest signal extracting section 36, FIG.
As shown in (a), the accompaniment note-on signal S13 and the accompaniment chorus signal S15 are used as input signals, and the accompaniment note-on signal S13 is integrated by the integrator circuit, while the output of the integrator circuit is inverted when it exceeds a certain level. After that, the signal is input from the signal inversion circuit to one input of the AND element as the output signal SM2. The accompaniment chorus signal S15 is input to the other input of the AND element, and the output of the AND element is
As shown in (b), an accompaniment rest signal S22 which is "H" only during the rest is obtained. During the rest where the note is cut off for a long time, the input of the accompaniment note-on signal S13 is cut off, and the output of the integration circuit drops below a certain level.
As a result, it becomes "H" only during the rest. The AND operation is performed with the accompaniment chorus signal S15 in order to avoid inconvenience such as erroneously outputting the accompaniment rest signal during an interlude period in which there is no accompaniment signal.

(6) The configuration related to the generation of the performance end signal is as follows. The accompaniment signal is practically only during the singing of the singer, but the performance signal S1 is necessary not only during the singing of the singer but also for the prelude and the interlude to the afterplay.
It is virtually uninterrupted. As shown in FIG. 8, the performance end signal creating section 37 accumulates the performance signal S1 which is an analog signal in a rectification / integration circuit, and if the output of the circuit is above a certain level, a performance end signal composed of a comparator or the like. A performance end signal S23 is output from the take-out section. Even if the signal amount of the performance signal S1 is minute or there is a moment when the signal is interrupted, the integration circuit keeps the output above a certain level for a while, so that the performance end signal creating section 37 keeps "H" during the performance.
And the performance end signal S23 is sent out. The end of the performance is indicated when shifting from "H" to "L". A song note-on signal S9 and a song chorus signal S10 that appear during the performance of one song.
The accompaniment note-on signal S14, accompaniment chorus signal S15, and performance end signal S23 are shown together in FIG.

(7) Next, the scoring means for each scoring item will be described. Each scoring means is mainly CPU1 to CPU
7 and its control program. First, the configuration around the pitch calculation unit CPU1 for pitch scoring will be described. In the pitch calculator CPU1, the song note number S10 and the accompaniment note number S1 sent one after another.
Eight consecutive 4's are stored, and the one with the highest number of appearances is extracted and stored repeatedly. If there are the same number of appearances, the one with the highest pitch is stored. In order to check the number of appearances of a song note number of a specific pitch, subtraction processing with other song note numbers may be performed to find the number of times that the difference becomes zero. The time for each note number varies depending on the song, but is usually 0.3-
Since it is about 0.5, if eight note numbers are divided into sections, the interval of one division is about 2.4 to 4 seconds. However, the accompaniment note number S14 is sent according to the musical score, but the song note number S10 is not always sent according to the musical score. Therefore, accompaniment note number S
A method in which the division is performed when the number of 14 becomes 8 is appropriate. In this case, the number of song note numbers S10 may not be eight.

The song note number and the accompaniment note number with the highest number of appearances are stored for each division interval, and the note number between the song note number and the accompaniment note number extracted and stored for each division interval is stored. Subtract and subtract the result (♭ 5 to # 5: -1/3 below)
Data corresponding to octave to +1/3 octave) is stored at each delimiter interval. Then, based on the result of the subtraction, after one chorus is completed, a key adjustment signal (pitch correction control signal) S5 is created as necessary and sent to the key controller 16 of the mixing amplifier device 2. ♭ 5
If there are 20 pieces of data corresponding to # 5, only the data having the largest number of appearances is calculated and used as the key adjustment signal S5. In the calculation of the key adjustment amount, if there is the same number of appearances, the data on the ♭ side close to = (natural) is preferentially used. When the key adjustment is executed, the key adjustment amount is reflected in the pitch adjustment by the key adjustment amount reflecting means performing a correction corresponding to the key adjustment amount by the key controller on the note number difference. It is preferable to reflect it automatically. The fact that the key adjustment has been performed is one of the pitch scoring information.
It is transmitted to the scoring calculation unit CPU 7.

♭ 5000000101 = 0000000000 # 1 11111111 ♭ 4 00000100 # 2 11111110 ♭ 3 00000011 # 3 11111101 ♭ 2 00000010 # 4 11111100 ♭ 1 00000001 # 5 11111011

Between a male voice and a female voice, one octave (1
There is a difference of two chromatic scales. Since the accompaniment signal S4 is matched with the female voice, the singing note number is one octave lower than the accompaniment note number even if the male voice is sung correctly. Therefore, if the difference between the song note number and the accompaniment note number is ♭ 6 (6 semitones) or more, it is judged as a male voice, and as a result of carrying out the carry process of one octave to the calculation process of both note numbers, The subtraction result of both note numbers is ♭ 5
The range is # 5. In the case of the undertone, the singing note number is one octave higher than the accompaniment note number, which is a phenomenon that is rarely seen. Therefore, if the difference between the song note number and the accompaniment note number is # 6 (six chromatic scales) or more, it is determined to be a back voice, and as a result of carrying out a one-octave carry-down process to the arithmetic processing of both note numbers, The subtraction result of both note numbers is in the range of ♭ 5 to # 5. The pitch calculator CPU1 is also responsible for such octave correction processing.

On the other hand, the following points are given to each of the stored 20 data items 5 to 5 (scoring process). However, since the pitch of singing tends to be incorrect, the following points are doubled only for the first 2 seconds (scoring the weight twice). ♭ 5 / # 5 -5 points for each time: ♭ 4 / # 4 -4 points for each time ♭ 3 / # 3 -3 points for each time: ♭ 2 / # 2-2 points for each time ♭ 1 / # 1 -1 point for each time In addition, when the key adjustment is performed, it may be possible to perform -10 point marking (point deduction processing) from the second chorus because the keys do not match.

Furthermore, it is possible to add a scoring item that the singer does not sing at all or interrupts the song. For example, it is conceivable that the pitch calculation unit CPU1 measures the "H" time difference between the song chorus signal and the accompaniment chorus signal, and gives a score according to the degree of the time difference as described below. 30 seconds or more −70 points: 20 to 29 seconds −50 points 10 to 19 seconds −30 points: 5 to 9 seconds −10 points 0 to 4 seconds 0 points However, if the total point reaches −100 points, it will be capped at −100 points. And The pitch scoring information signal S31 obtained by the pitch computing unit CPU1 is sent to the scoring computing unit CPU7 in a timely manner.

(8) Tempo calculator CPU for tempo scoring
The structure around 2 is as follows. Tempo calculator C
In PU2, the rising of the song tempo signal S17 and the accompaniment tempo signal S18 that are inverted between "H" and "L" for each input of four song note numbers and accompaniment note numbers [S17 of FIG. 9 (b), And S18 in FIG.
A calculation for obtaining the time difference is performed. This time difference represents the degree of tempo deviation between the accompaniment and the song. The larger the tempo deviation, the larger the time difference between the rising of the song tempo signal S17 and the accompaniment tempo signal S18. Data of about 40 time differences can be obtained with one chorus. As shown below, each data is scored according to the time difference. However, since the tempo tends to shift when singing, the score may be doubled (double weighting) only for the singing data depending on the time difference. The cases (1) to (3) below are virtually equivalent to a state in which they are not sung or are sung intermittently.

1000 milliseconds or more -40 points 999 to 600 milliseconds -20 points 599 to 400 milliseconds -10 points 399 to 200 milliseconds -5 points 199 to 150 milliseconds -4 points 149 to 100 milliseconds -3 points 99 to 60 milliseconds -2 points 59 to 20 milliseconds -1 point 19 to 0 milliseconds -0 point However, if the total point reaches -100 points, it will be capped at -100 points. The tempo scoring information signal S32 obtained by the tempo computing unit CPU2 is sent to the scoring computing unit CPU7 in a timely manner.

(9) Tone color calculation unit The configuration around the CPU 3 is as follows. The timbre calculation unit CPU3 counts the number of simultaneous occurrences (becomes "H" at the same time) of the output signals SE38 to SE71 of the frequency discriminating units BF38 to BF71 for detecting harmonics at regular time intervals, and at the same time at regular time intervals. The maximum counts are detected and the following points are assigned to each maximum count. Count number 5 or more -2 points: Count number 3 to 4 -1 point Count number 1 to 2 0 point: Count number 0 +2 points Tone color scoring information signal S3 obtained by tone color calculation unit CPU3
3 is sent to the scoring calculation unit CPU 7 in a timely manner.

(10) The structure around the voice volume calculator CPU4 will be described. The voice volume calculator CPU4 inputs the parallel digital signal S6 of the song signal output from the A / D converter 22, extracts only the positive polarity signal, and detects the highest bit MB every 2 seconds. The following scoring is performed on the detected highest bit MB. Since the voice tends to be small when singing, only the data at the singing is given a double weight. Highest bit MB: 2SB + 2 points, highest bit MB: 3SB + 1 point Highest bit MB: 4SB 0 point, highest bit MB: 5SB -1 point Highest bit MB: 6SB -2 points Highest bit MB: 7SB -3 points No bits are set except for LSB -50 points However, if the total points exceed -100 points, it will be -100 points. The voice volume scoring information signal S34 obtained by the voice volume computing unit CPU4 is sent to the scoring computation unit CPU7 in a timely manner.

(11) The configuration around the fermata arithmetic unit CPU 5 will be described. In the fermata calculation unit CPU5, the song note-on signal S9 and the accompaniment note-on signal S13
And the accompaniment chorus signal S15. Normally, the fermata (reverberation) is intended for a note length of a whole note or longer, and usually takes about 1 to 4 seconds. However, in a high-tempo song, the fermata often has 2-3 whole notes, so here , See Fermata for 3 seconds,
When the accompaniment note-on signal S13 of 3 seconds or more is input,
The time difference between the rising edge of the singing note-on signal S9 and the rising edge of the accompaniment note-on signal S13 is calculated, and the following scores are given at each chorus end. It should be noted that the fermata usually appears once at the end of each chorus. 0 to 0.19 seconds 0 0.2 to 0.39 seconds -5 points 0.4 to 0.59 seconds -10 points 0.6 to 0.99 seconds -15 points 1.0 to 1.49 seconds -20 Points 1.5 to 1.99 seconds -25 points 2.0 to 2.99 seconds -30 points 3 seconds or more -50 points The fermata scoring information signal S35 obtained by the fermata arithmetic unit CPU5 is timely calculated by the scoring arithmetic unit CPU7. Sent to.

(12) The configuration around the rest calculation unit CPU 6 will be described. In the rest calculator CPU6, the song rest signal S
21 and the accompaniment rest signal S22, the time difference (pulse width difference) is obtained, and the following scores are given. 1 second or more −5 points: 0.5 to less than 1 second −3 points 0.3 to less than 0.5 seconds −2 points: 0.1 to less than 0.3 seconds −1 point less than 0.1 seconds 0 points

The rest calculator CPU 6 causes a bar graph on the screen of the display device 4, for example, the following bar graph of two-color display of red and green, that is, □□□□□□□□
(“1”, “2”, ..., “8” in order from the left end) may be displayed so that the arrival of a rest can be recognized.
By inputting the accompaniment note-on signal S13 and the accompaniment rest signal S22, for example, red is lit when the accompaniment note-on signal S13 comes, and green is lit when the accompaniment rest signal S22 comes. If the accompaniment note-on signal S13 continues, "1"
Light up red in order from 1 to "8". If you proceed to "8", all lights up. If it lights up to 8, it will be turned off and the process will be repeated. Meanwhile, if there is a rest, it lights up green. Rest calculation information signal S36 obtained by rest calculator CPU6
Are sent to the scoring calculation unit CPU 7 in a timely manner.

(13) The configuration around the scoring calculation unit CPU 7 will be described. In the scoring calculation unit CPU7, each calculation unit CP
Based on the scoring information signals S31 to S36 from the U1 to CPU6, after completion of a poor performance during the performance, the scores of the respective scoring items are totaled to score each chorus, or the average value of all scoring items is calculated. Then, the scoring result data S37 obtained by performing comprehensive scoring is sent to the display device 4 and displayed.

Next, the concrete scoring of a song by the embodiment apparatus having the above-described structure will be described with reference to the drawings. Initially, the process of obtaining the pitch scoring information by the pitch calculator CPU1 will be described with reference to the flowchart shown in FIG.

[Step A1] Whether or not accompaniment is being performed is checked based on the accompaniment chorus signal S15.
If "H", it is accompaniment, and if "L", it is not accompaniment.

[Step A2] Whether the accompaniment note number S14 is input during accompaniment is checked based on the accompaniment note-on signal S13. If "H", the accompaniment note number S14 is input, and if "L", no input.

[Step A3] Accompaniment Note Number S1
If 4 is input, the number of the count of the accompaniment note number S14 is incremented by 1.

[Step A4] Accompaniment Note Number S1
It is checked whether or not the number count number of 4 has reached 8.

[Step A5] If the accompaniment note number S14 has not been input in step A2 or the number of pieces has not reached 8 in step A4, it is determined whether or not the song note number S8 has been input. Check based on the ON signal S9. If it is "H", the song note number S8 is input, if it is "L", there is no input. If the song note number S8 is not input, the process returns to step A1.

[Step A6] If the song note number S8 is input in step 5, the song note number S
Increase the number of 8 by 1 and then step A
Return to 1.

[Step A7] When the number of counts reaches 8 in step A4, the accompaniment note number S14 and the song note number S8 having the maximum number of appearances in the section are obtained.

[Step A8] The note number difference between the accompaniment note number S14 and the song note number S8 having the maximum number of appearances is obtained.

[Step A9] After scoring the difference between the accompaniment note number and the song note number of the maximum number of appearances, the pitch scoring information is transmitted to the CPU 7 for scoring, and the process returns to step A1.

[Step A10] 1 if not in accompaniment
Whether or not the chorus has ended is checked by the fall (“H” to “L”) of the accompaniment chorus signal S15. If one chorus has not ended, the process returns to step A1.

[Step A11] If the end of one chorus in step 10, the key adjustment signal is sent to the key controller 16 according to the value of the maximum number of appearances of the note number difference obtained in step A8, and the key adjustment information is sent. Is transmitted to the scoring calculation unit CPU7 and then the process returns to step A1.

The process of obtaining the tempo scoring information by the tempo calculator CPU2 will be described with reference to the flowchart shown in FIG. [Step B1] Whether or not accompaniment is being performed is checked based on the accompaniment chorus signal S15.

[Step B2] If accompaniment tempo signal S18 is input during accompaniment, from the rising edge thereof,
A timer TA1 measures the time TBA until the rising of the song tempo signal S17. Specifically, the accompaniment tempo signal S1
After the timer TA1 is reset and started at the rising edge of 8, the timer is stopped at the rising edge of the song tempo signal S17 to obtain the time TBA.

[Step B3] A score is assigned to the measured time TBA.

[Step B4] Measured time TBA
However, if it is for singing and is in the range of 20 milliseconds to 399 milliseconds, weighting for doubling the score is performed.

[Step B5] The result of the scoring process is transmitted to the scoring calculator CPU7 as tempo scoring information, and the process returns to step B1.

The process until the tone color scoring information is obtained by the tone color calculator CPU3 will be described with reference to the flowchart shown in FIG. [Step C1] Based on the song chorus signal S10, it is checked whether or not the user is singing. If it is "L", the check is repeated because it is not singing.

[Step C2] If singing is in progress (if “H”), it is checked in order whether or not there is “H” in the output signals SE1 to 37.

[Step C3] The control signals S7-1 to S7-37 corresponding to the output signal in which "H" is first detected.
Only the line of "H" is set.

[Step C4] The number HC of “H” of the output signals SE38 to SE71 for 100 milliseconds is counted every 100 milliseconds.

[Step C5] The maximum value of the number HC of "H" in step C4 is detected and stored.

[Step C6] It is checked whether or not 2 seconds have elapsed, and if not, the process returns to step C1.

[Step C7] If it is confirmed that two seconds have passed in step C6, a score is assigned to the maximum value of the count number of "H".

[Step C8] The result of the scoring process is transmitted to the scoring calculator CPU7 as tone color scoring information, and the process returns to step C1.

The process until the voice volume scoring unit CPU4 obtains the voice volume scoring information will be described with reference to the flowchart shown in FIG. [Step D1] Whether or not singing is checked based on the song chorus signal S10, and if not, the check is repeated.

[Step D2] The parallel digital signal S6 of the song signal output from the A / D converter 22 is input, only the positive polarity signal is extracted, and the highest bit MB is detected and stored.

[Step D3] It is checked whether or not 2 seconds have elapsed, and if not, the process returns to step D1.

[Step D4] If it is confirmed that 2 seconds have passed in step D3, the detected highest bit MB
The scoring process for is performed.

[Step D5] If the detected highest-order bit MB is a singing start, weighting for doubling the score is performed.

[Step D6] The result of the scoring process is transmitted to the scoring calculator CPU7 as voice volume scoring information, and then the process returns to step D1.

The process until the fermata scoring information is obtained by the fermata arithmetic unit CPU 5 will be described with reference to the flowchart shown in FIG. [Step E1] Whether or not accompaniment is being performed is checked based on the accompaniment chorus signal S15.

[Step E2] The timer TE1 measures the “H” time TEA of each accompaniment note-on signal S13, while the timer TE2 measures the “H” time TEB of each song note-on signal S9. go. At the rising edge of the accompaniment note-on signal S13 or the song note-on signal S9, the timer TE1 or the timer TE2 is reset / started and stopped at the falling edge to stop the time TEA, TE.
Each B can be measured.

[Step E3] It is checked whether the measured time TEA of the timer TE1 is 3 seconds or more. If it is less than 3 seconds, the process returns to step E1.

[Step E4] In step E3, if the measurement time TEA of the timer TE1 is 3 seconds or more, the time difference ΔTE between the measurement time TEA of the timer TE1 and the measurement time TEB of the timer TE2 is calculated, and the scoring process for the time difference ΔTE is performed. I do.

[Step E5] The result of the scoring process is transmitted to the scoring calculator CPU7 as fermata scoring information, and then the process returns to step E1.

The process until the rest calculation CPU 6 obtains rest scoring information will be described with reference to the flowchart shown in FIG. [Step F1] Whether or not accompaniment is being performed is checked based on the accompaniment chorus signal S15.

[Step F2] If the accompaniment note number S14 is input during accompaniment, the N-th of the eight display portions is lit in red. If eight LEDs are already lit, only the first display should be lit in red and then the next step F
Proceed to 3. Accompaniment note number S1 in step F2
When there is no input of 4, the operation of lighting the display unit is not performed and the process proceeds to the next step F3.

[Step F3] It is checked whether or not the accompaniment rest signal S22 is input. Accompaniment rest signal S2
If there is no input of 2, jump to step F6.

[Step F4] If the accompaniment rest signal S22 is input, the Nth of the eight display portions is lit in green. If eight of them are already lit, only the first display is lit in green and the process proceeds to step F5.

[Step F5] The length of the accompaniment rest signal S22 is measured by the timer TF1. The timer TF1 measures the time TFA from the rise to the fall of the accompaniment rest signal S22.

[Step F6] The length of the song rest signal S21 is measured by the timer TF2. The time T from the rise to the fall of the song rest signal S21 by the timer TF2
Measure FB.

[Step F7] The difference ΔTF between the time TFA and the time TFB obtained in steps F5 and F6 is measured.

[Step F8] A score is assigned to the difference ΔTF.

[Step F9] The result of the scoring process is transmitted to the scoring calculator CPU7, and then step F9.
Return to 1.

Finally, the process of obtaining the scoring result by the scoring calculation unit CPU 7 which has input the above scoring information,
Description will be given according to the flowchart shown in FIG. [Step G1] A performance end signal S23 indicating whether or not performance is in progress
Check based on. If it is "H", it is playing,
If it is "L", it is not playing. If a performance is in progress, an intermediate scoring is performed for each scoring item in the following steps on the basis of each scoring information sent one by one.

[Step G2] Regarding pitch scoring, the accompaniment chorus signal S15 is “L” while processing is performed as necessary during accompaniment based on the received pitch scoring information.
When the first chorus is completed, the scoring for the first chorus song will be performed. If the following pitch scoring information (20 pieces) is obtained, a total of 12 points will be deducted. 12 points are deducted from 100 points and 88 points are deducted from the first chorus. When the key adjustment is performed in the first chorus, the key adjustment is performed 10 after the second chorus.
Points will be deducted (the maximum score will be reduced from 100 to 90).

1 ... ♭ 1 −1 point, 2 ... = 0 point, 3 ... = 0 point, 4 ... = 0 point 5 ... # 1 −1 point, 6 ... = 0 point, 7 ... = 0 point, 8 ... ♭ 2 -2 points 9 ... ♭ 2 -2 points, 10 ... = 0 points, 11 ... = 0 points, 12 ... = 0 points 13 ... # 1 -1 points, 14 ... # 1 -1 points, 15 ... = 0 Point, 16 ... = 0 point 17 ... = 0 point, 18 ... ♭ 1 -1 point, 19 ... ♭ 2-2 point, 20 ... = 0 point Data unit: millisecond Singing weight added ... -1 point Total- 12 points

[Step G3] Regarding the tempo, one chorus is performed at the time when the first chorus in which the accompaniment chorus signal S15 becomes “L” is finished, while processing is performed as necessary during the accompaniment based on the received tempo scoring information. You will be scoring the eyes song. If the following tempo scoring information (24 pieces) is obtained, a total of 26 points will be deducted. 26 points are deducted from 100 points and 74 points are the first chorus evaluation. The second and subsequent choruses will be scored similarly.

1 ... 130 -3 points, 2 ... 180 -4 points, 3 ... 80-2 points, 4 ... 40 -1 points 5 ... 50 -1 points, 6 ... 100 points, 7 ... 100 points, 8 ... 30 -1 point 9 ... 25 -1 point, 10 ... 15 0 point, 11 ... 25 -1 point, 12 ... 40 -1 point 13 ... 35 -1 point, 14 ... 10 0 point, 15 ... 30 -1 point , 16 ... 70 -2 points 17 ... 25 -1 point, 18 ... 25 -1 point, 19 ... 30 pieces -1 point, 20 ... 10 0 points 21 ... 5 0 points, 22 ... 45 -1 point, 23 ... 30 Pieces-1 point, 24 ... 150 points Data unit: millisecond Singing weights added ...- 3 points Total -26 points

[Step G4] With regard to the timbre, one chorus is performed at the time when the first chorus at which the accompaniment chorus signal S15 becomes “L” is finished while being processed as necessary during the accompaniment based on the received timbre scoring information. You will be scoring the eyes song. If the following tone color scoring information is obtained, a total of 7 points will be deducted. 7 points are deducted from 100 points and 93 points are the first chorus evaluation. The second and subsequent choruses will be scored similarly.

1 ... 2 0 points, 2 ... 3 points -1, 1 point, 3 ... 3 points -1, 4 ... 4 points -1 point 5 ... 5 points-2 points, 6 ... 1 point 0 points, 7 ... 5 pieces -2 points, 8 ... 2 pieces 0 points 9 ... 0 pieces +2 points, 10 ... 3 pieces -1 point, 11 ... 0 pieces +2 points, 12 ... 3 pieces -1 point 13 ... 3 pieces -1, point, 14 … 2 -1 points, 15… 1 piece 0 points, 16… 0 pieces + 2 points 17… 2 pieces 0 points, 18… 3 pieces −1 point, 19… 3 pieces −1 point, 20… 1 piece 0 points Total -7 points

[Step G5] Regarding the voice volume, the voice volume is processed as needed during the accompaniment based on the received voice volume scoring information, and is also set to 1 when the first chorus at which the accompaniment chorus signal S15 becomes “L” is finished. The chorus song will be scored. If the following voice volume scoring information is obtained, a total of 12 points will be deducted. 12 points for 1
88 points are deducted from 00 points to evaluate the first chorus.
The second and subsequent choruses will be scored similarly.

Maximum bit MB is 2SB +2 points: 3 times +6 points Maximum bit MB is 3SB + 1 point: 2 times +2 points Maximum bit MB is 4SB 0 points: 5 times 0 points Maximum bit MB is 5SB -1 point: 5 times -5 points Maximum bit MB is 6SB -2 points: 2 times (singing out) -4 points Maximum bit MB is 7SB -3 points: 3 times -9 points There is no bit other than LSB -50 points: 0 times 0 points Sing out Weighting (bits in 6SB) -2 points Total -12 points

[Step G6] With regard to the fermata, based on the fermata scoring information, the singing of the first chorus is scored when the first chorus at which the accompaniment chorus signal S15 becomes “L” is finished. . A time difference of 1.5 to 1.
If the score information is 1.7 seconds in the 99-second interval, 25
The points will be deducted. 25 points are deducted from 100 points and 75 points are the first chorus evaluation. The second and subsequent choruses will be scored similarly.

[Step G7] With respect to rests, the rest chorus signal S15 becomes “L” while the first chorus ends while being processed as necessary during the accompaniment based on the received rest scoring information. The first chorus song will be scored. If the following rest scoring information is obtained, a total of 8 points will be deducted. Eight points are deducted from 100 points and 92 points are evaluations of the first chorus. The second and subsequent choruses will be scored similarly.

Rest 1 ... Time difference ΔTF 0.4 seconds Scoring -2 points Rest 2 ... Time difference ΔTF 0.1 seconds Scoring -1 point Rest 3 ... Time difference ΔTF 0.02 seconds Score Mark 0 points Rest 4… Time difference ΔTF 0.6 seconds Score -3 points Rest 5… Time difference ΔTF 0.35 seconds Score −2 points Rest 6… Time difference ΔTF 0.05 seconds Score Addition-0 points Total-8 points

[Step G8] 1 after step G7
Whether or not the chorus has ended is checked based on the accompaniment chorus signal S15. If not completed, the process returns to step G1.

[Step G9] In step G8, if one chorus has ended, the above-mentioned steps G2 to
The intermediate scoring results obtained in G7 are displayed on the display device 4, and then the process returns to step G1.

[Step G10] When it is confirmed in step G1 that the performance is not being performed, the trailing edge of the performance end signal S23 (transition from "H" to "L") is checked. If not, step G1 Return to.

[Step G11] When the trailing edge of the signal S23 is confirmed in step G10, the scores of all scoring items for all choruses are summed up for each scoring item and stored separately in the number of choruses (three chorus configurations). In this case, divide by 3) to give the average score of 3 chorus for each scoring item, and divide the total of the average score for each scoring item by the number of scoring items (6 items at the moment) to obtain the total average score. Obtain as a grade.

[Step G12] Both the average score and the total score for each scoring item obtained in step G11 are displayed on the display device 4, and then the process returns to step G1. With the above, the evaluation of one song is completed.

The present invention is not limited to the above embodiment, but can be modified as follows, for example. (1) In the embodiment, the automatic scoring device has a separate structure from other devices such as the signal source device, however, the entire automatic scoring device is integrated with the karaoke device. Good.

(2) In the embodiment, the scoring result is shown by a score, but it is excellent / good / good / not good or "the pitch is greatly deviated upward" or "the pitch is slightly deviated upward". It is also possible to use a method in which the scoring results such as "the singing starts are greatly deviated" and "the singing starts are right" are indicated by words.

(3) In pitch scoring in the embodiment,
Of the eight note numbers, the note number with the maximum number of appearances was calculated, and the note number difference was calculated by subtracting both note numbers. For example, the average value of the eight note numbers was calculated, and The average note number may be subtracted to calculate the note number difference.

(4) Although there were 6 scoring items in the embodiment, the automatic scoring device for a karaoke device of the present invention does not have to be a device capable of scoring all 6 scoring items.
It is sufficient if at least the pitch can be scored.

[0129]

According to the automatic scoring device for a karaoke device according to the invention described in claim 1, as a result of scoring with the note number difference corresponding to the degree of deviation between the pitch of the accompaniment and the song, most of the singing is good or bad. In addition to the evaluation of the essential pitch, the main melody (song melody) called the accompaniment signal is used as a standard for measuring the deviation of the pitch of the song, and the degree of deviation of the pitch of the singer's song with respect to the accompaniment is accurately grasped and scored. As a result, since the scoring accuracy is high, the song to be sung is accurately evaluated.

According to the automatic scoring device of the second aspect of the present invention, since the quality of the song tempo is also scored, the song can be evaluated more accurately.

According to the automatic scoring device of the third aspect of the present invention, since the timbre of the song is also scored,
The song is evaluated more accurately.

According to the automatic scoring device of the fourth aspect of the present invention, since the presence or absence of the voice volume of the singer is also scored, the song can be evaluated more accurately.

According to the automatic scoring device of the fifth aspect of the present invention, the singing person's fermata (reverberation) singing style is also graded, so that the song is evaluated more accurately.

According to the automatic scoring device of the sixth aspect of the present invention, the singer's resting method is also graded, so that the song can be evaluated more accurately.

According to the automatic scoring device of the seventh aspect of the invention, since the goodness and badness of the song is scored from a comprehensive viewpoint, the song can be evaluated more accurately.

According to the automatic scoring device of the eighth aspect of the invention, since the digital pitch code signal corresponding to the digital pitch code of the electronic musical instrument standard is added to the pitch indicated by the song signal, the pitch As a result of rapid progress of the signal processing required to obtain the note number difference as the scoring information, the song can be evaluated quickly.

According to the automatic scoring device of the ninth aspect of the invention, since the difference between the accompaniment and the singer's key is eliminated with the operation of the key controller by the key adjustment signal, the karaoke accompaniment is easy for the singer to sing. Will be things.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a schematic configuration of a karaoke system of an embodiment.

FIG. 2 is a block diagram showing a configuration of a signal source device according to an embodiment.

FIG. 3 is a block diagram showing a configuration of a mixing amplifier device according to an embodiment.

FIG. 4 is a block diagram showing a configuration of an automatic scoring device according to an embodiment.

FIG. 5 is a circuit diagram showing a configuration around the extraction of a song pitch in a song signal.

FIG. 6 is a circuit diagram showing a configuration relating to adding a pitch code to a song signal.

FIG. 7 is a circuit diagram showing a configuration of an output unit for a song note number and a song chorus signal.

FIG. 8 is a circuit diagram showing a configuration related to accompaniment note number creation.

FIG. 9 is an explanatory diagram showing a circuit configuration and a signal waveform related to creation of a song tempo signal.

FIG. 10 is an explanatory diagram showing a circuit configuration and a signal waveform related to creation of an accompaniment tempo signal.

FIG. 11 is an explanatory diagram showing a circuit configuration and a signal waveform related to creation of a song rest signal.

FIG. 12 is an explanatory diagram showing a circuit configuration and a signal waveform related to creation of an accompaniment rest signal.

FIG. 13 is an explanatory diagram showing a format example of an accompaniment signal according to the MIDI standard.

FIG. 14 is a graph showing an example of an accompaniment signal according to the MIDI standard.

FIG. 15 is a graph showing a main signal waveform in the example.

FIG. 16 is an explanatory diagram showing a digital pitch code (fundamental wave side).

FIG. 17 is an explanatory diagram showing a digital pitch code (on the harmonic side).

FIG. 18 is a flowchart showing the flow of processing in the pitch calculation section.

FIG. 19 is a flowchart showing a processing flow in a tempo calculation unit.

FIG. 20 is a flowchart showing the flow of processing in a tone color calculation section.

FIG. 21 is a flowchart showing a flow of processing in a voice volume calculator.

FIG. 22 is a flowchart showing the flow of processing in the fermata arithmetic unit.

FIG. 23 is a flowchart showing a processing flow in a rest calculation unit.

FIG. 24 is a flowchart showing the flow of processing in a scoring calculation unit.

[Explanation of symbols]

 6 ... Automatic scoring device 25 ... Song note number / flag creating section 28 ... Accompaniment note number / flag creating section 31 ... Song tempo signal creating section 32 ... Accompaniment tempo signal creating section 35 ... Song rest signal creating section 36 ... Accompaniment rest Signal creation section S3 ... Song signal S4 ... Accompaniment signal S8 ... Song note number S14 ... Accompaniment note number CPU1 ... Pitch calculation section CPU2 ... Tempo calculation section CPU3 ... Tone color calculation section CPU4 ... Voice volume calculation section CPU5 ... Fermata calculation section CPU6 ... Rest Calculation unit CPU7 ... Scoring calculation unit

Claims (9)

[Claims] 1. An accompaniment note number creating means for creating an accompaniment note number indicating a note number in an accompaniment signal according to a standard defined for an electronic musical instrument, and a pitch in a song signal obtained from a song sung in accordance with the accompaniment. A song note number creating means for creating a song note number in the form of a note number conforming to the standard, a note number difference calculating means for obtaining a note number difference between the accompaniment note number and the song note number, and a song based on the note number difference An automatic scoring device for a karaoke device, comprising: a pitch scoring means for scoring the pitch. 2. The automatic scoring device for a karaoke device according to claim 1, wherein tempo difference detecting means for detecting a tempo difference between the tempo in the accompaniment signal and the tempo in the song signal, and a song based on the tempo difference. An automatic scoring device for a karaoke device, comprising: tempo scoring means for scoring tempo. 3. The automatic scoring device for a karaoke device according to claim 1 or 2, based on frequency detection means for detecting the frequency of appearance of harmonics in a song signal per unit time, and based on the frequency of appearance of the harmonics. An automatic scoring device for a karaoke device, comprising timbre scoring means for scoring the timbre of a song. 4. The automatic scoring device for a karaoke device according to claim 1, wherein intensity measuring means for measuring the intensity of the singing signal, and the singing person's intensity based on the intensity measurement result of the singing signal An automatic scoring device for a karaoke device, comprising: a voice volume scoring means for scoring a voice volume. 5. The automatic scoring device for a karaoke device according to claim 1, further comprising a fermata time difference detecting means for detecting a time difference between a fermata time in the accompaniment signal and a fermata time in the song signal. , An automatic scoring device for a karaoke device, comprising: a fermata scoring means for scoring the fermata of a song based on the time difference between both fermata times. 6. The automatic scoring device for a karaoke device according to claim 1, wherein a rest time difference detection is performed to detect a time difference between a rest time in an accompaniment signal and a rest time in a song signal. An automatic scoring device for a karaoke device, comprising: means and a rest scoring means for scoring a rest of a song based on a time difference between the rests. 7. The automatic scoring device for a karaoke device according to claim 1, wherein the pitch of the song, the tempo of the song, the tone of the song, the voice of the singer, the fermata of the song, and the rest of the song. An automatic scoring device for a karaoke device, which comprises a comprehensive scoring means for scoring a song comprehensively based on each scoring result of a mark. 8. The automatic scoring device for a karaoke device according to claim 1, wherein the song note number creating means discriminates the frequency of the song signal and a frequency discriminating means. An automatic scoring device for a karaoke device, comprising a pitch code assigning means for assigning a digital pitch code signal corresponding to the standard digital pitch code defined for the electronic musical instrument in accordance with the discrimination frequency according to. 9. The automatic scoring device for a karaoke device according to claim 1, wherein a key controller of the karaoke device is driven so as to eliminate the difference between the accompaniment and singer keys based on the note number difference. An automatic scoring device for a karaoke device, which comprises a key adjustment signal creating means for creating a key adjustment signal for performing.