JPH05323978A - Recording and reproducing method for accompaniment signal and automatic key controller for orchestral accompaniment device - Google Patents

Abstract

PURPOSE:To automatically perform key control by adding an accompaniment signal having the same frequency with notes of an accompaniment as a modulated wave in a specific band in a musical sound signal band at the time of recording and demodulating the modulated wave at the time of reproduction. CONSTITUTION:The musical sound signal is supplied to a BPF 202 with a DC-16 KHz to remove a speech signal except in this band. The accompaniment signal is supplied to a BPF 204 with a 100-600Hz to remove an accompaniment signal except in the band, and an FM modulator 206 imposes FM modulation on the accompaniment signal of 100-500Hz by using a carrier of 18KHz in center frequency f0. The musical sound signal outputted from the BPF 202 and the modulated accompaniment signal outputted from the BPF 204 are put together and converted into a digital signal. Thus, the FM modulated accompaniment signal is recorded. This accompaniment signal is a sine wave signal, so accurate key control can be performed by the key controller device at the time of reproduction.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a key controller of a karaoke apparatus, and more particularly to automation of the key controller.

[0002]

2. Description of the Related Art A karaoke device is a device in which a singer selects a song that he wants to sing and sings it along with a karaoke accompaniment. Recently, a karaoke device may be equipped with a key controller capable of setting an accompaniment key according to the pitch (key) of the singer. For example, if one's own key is a semitone higher than the natural key, by setting the key of the key controller to # 1 in advance, the accompaniment key can be made a semitone higher than the natural key to sing.

Conventionally, the key controller of the karaoke apparatus is set in this way.

[0004]

However, the above-described key controller of the conventional karaoke apparatus has the following problems.

Cassette tapes used in karaoke machines, compact disc graphics (hereinafter referred to as CD-
The key of the accompaniment recorded on the laser disk (hereinafter referred to as “G”) or the laser disk (hereinafter referred to as “LD”) may be different for each manufacturer. For this reason, even if the singer selects his or her own key, since the reference key is different, it does not match the accompaniment key, and there is a problem that the key needs to be changed again during the song.

When a female singer sings with a male accompaniment or a male singer sings with a female accompaniment, the former key is too high and the latter key is too low, making it difficult to match the keys. There was also the problem of being.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the above problems and provide an automatic key controller device for a karaoke device which can automatically perform key control.

[0008]

According to the accompaniment signal recording / reproducing method of claim 1, at the time of recording, an accompaniment signal having the same frequency as the frequency of the accompaniment note is added as a modulation wave to a predetermined band of the music signal band. And record it on a recording medium,
At the time of reproduction, the modulated wave is demodulated to obtain an accompaniment signal.

The accompaniment signal recording / reproducing method of claim 2 is characterized in that, in the accompaniment signal recording / reproducing method of claim 1, the modulated wave is an FM wave.

The accompaniment signal recording / reproducing method of claim 3 is characterized in that, in the accompaniment signal recording / reproducing method of claim 1, the modulating wave is an AM wave.

According to a fourth aspect of the present invention, there is provided an automatic key controller device for a karaoke apparatus, wherein the accompaniment frequency extracting means for obtaining a digital frequency signal representing the frequency of each note of the accompaniment signal obtained by the accompaniment signal recording / reproducing method according to claim 1, and a song signal. Of the accompaniment signal from the accompaniment signal obtained from the accompaniment frequency extraction means and the accompaniment frequency extraction means for obtaining a digital frequency signal representing the frequency of each note Among the digital frequency signals of the song signal obtained from the extracting means, the highest frequency of the song obtained from the highest frequency detecting means and the highest frequency of the accompaniment signal obtained from the highest accompaniment frequency detecting means The calculation means for calculating the frequency difference from the highest frequency of the Selecting means for selecting the most frequent difference frequency within a predetermined time among the difference frequencies, and key correcting means for correcting the accompaniment key based on the shift of the key corresponding to the difference frequency selected by the selecting means. It is characterized by having.

According to a fifth aspect of the present invention, there is provided an automatic key controller device for a karaoke device according to the fourth aspect, wherein the calculating means divides the highest frequency of the accompaniment signal by the highest frequency of the song signal. Quotient is 2
If it is equal to or greater than or equal to or less than 1/2, the difference between the highest frequency of the accompaniment signal multiplied by 1/2 or 2 and the highest frequency of the song signal is calculated.

[0013]

According to the accompaniment signal recording / reproducing method of claim 1, at the time of recording, an accompaniment signal having the same frequency as the frequency of the accompaniment note is added as a modulated wave to a predetermined band of the music signal band and recorded on the recording medium. At the same time, during reproduction, the modulated wave is demodulated to obtain an accompaniment signal. Therefore, the sinusoidal accompaniment signal can be recorded as a modulation wave in a band where the music signal of the recording medium is not used, and the sinusoidal accompaniment signal can be obtained from the recording medium.

The accompaniment signal recording / reproducing method according to a second aspect is the accompaniment signal recording / reproducing method according to the first aspect, wherein the modulated wave is an FM wave. Therefore, the sine wave accompaniment signal can be FM-modulated and recorded in a band where the music signal of the recording medium is not used.

According to a third aspect of the accompaniment signal recording / reproducing method, in the accompaniment signal recording / reproducing method of the first aspect, the modulated wave is an AM wave. Therefore, the sine wave accompaniment signal can be AM-modulated and recorded in a band where the music signal of the recording medium is not used.

In the automatic key controller device for a karaoke apparatus according to a fourth aspect, the accompaniment frequency extracting means obtains a digital frequency signal representing the frequency of each note of the accompaniment signal obtained by the accompaniment signal recording / reproducing method according to the first aspect. The song frequency extracting means obtains a digital frequency signal representing the frequency of each note of the song signal. The accompaniment highest frequency detecting means obtains the highest frequency for every predetermined unit time among the digital frequency signals of the accompaniment signal obtained from the accompaniment frequency extracting means. The highest song frequency detecting means obtains the highest frequency every predetermined unit time from the digital frequency signals of the song signal obtained from the song frequency extracting means. The calculating means calculates a difference frequency between the highest frequency of the accompaniment signal obtained from the highest accompaniment frequency detecting means and the highest frequency of the singing signal obtained from the highest singing frequency detecting means, every predetermined unit time. The selecting means selects, from the difference frequencies obtained by the calculating means, the most frequent difference frequency within a predetermined time. The key correction means corrects the accompaniment key based on the shift of the key corresponding to the difference frequency selected by the selection means.

Therefore, the difference frequency between the highest frequency of the song signal and the highest frequency of the accompaniment signal is calculated every predetermined unit time,
By selecting the difference frequency having the highest frequency in a predetermined time, a shift between the song key and the accompaniment key is detected, and the accompaniment key is corrected based on the shift between the keys.

In the automatic key controller device for a karaoke apparatus according to claim 5, the arithmetic means further divides the highest frequency of the accompaniment signal by the highest frequency of the song signal, and the quotient is 2
If it is more than or equal to or less than 1/2, the difference between the highest frequency of the accompaniment signal multiplied by 1/2 or 2 and the highest frequency of the song signal is calculated. Therefore, even if there is a difference in accompaniment or voice octave between a female and a male, it is possible to detect the shift between the song key and the accompaniment key.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention is for automatically performing key control in a karaoke apparatus. In the following embodiments, an accompaniment signal used for key control is added as a modulated wave and recorded. After the recording and extraction of the accompaniment signal is explained, the explanation of the key control will be made.

On a laser disk (LD) used in a karaoke apparatus, an accompaniment of an orchestra and a song of a professional singer are recorded. In this embodiment, the accompaniment of the orchestra and the song of a professional singer are called music signals. This accompaniment signal recording / reproducing method uses an L recording medium.
In D, an accompaniment signal used for key control is to be recorded together with the music signal. However, even if the accompaniment signal is recorded as it is, it is not possible to extract only the accompaniment signal because the music signal and the accompaniment signal are not distinguished. Therefore, it is necessary to record the accompaniment signal in a frequency band different from the frequency band in which the music signal is recorded.

The music signal used in the ordinary audio equipment is a signal in the audible frequency band of DC to 20 kHz, as shown in FIG. 6a. However, in the music signal recorded on the karaoke machine LD, the frequency band of DC to 16 kHz is sufficient for use. That is, the frequency band of 16 kHz to 20 kHz is not normally used. Therefore, as shown in FIG. 6b, the accompaniment signal can be recorded in this frequency band.

Here, the accompaniment signal used for this key control will be described. FIG. 7 shows the score of one song. Each note in this song can be represented as an array of frequencies. For example, the frequency of the first "do" note appearing on the musical score is 262 Hz, and the frequency of the next "re" note is 294 Hz. The state of this frequency array is shown in FIG.

The accompaniment signal recorded in the present invention is a sinusoidal recording signal created so as to have the frequency of the note of such an accompaniment. In this example, the accompaniment signal is 100H.
It is assumed that the frequency is z to 500 Hz.

As described above, when attempting to record an accompaniment signal having a frequency of 100 Hz to 500 Hz in the frequency band of 16 kHz to 20 kHz of a music signal, it is necessary to modulate the frequency of the accompaniment signal to a frequency of 16 kHz to 20 kHz and record it. There is. According to the present invention, the accompaniment signal is FM-modulated and recorded in this band.

FIG. 1 shows an accompaniment signal recording circuit according to an embodiment of the present invention. This circuit consists of a band pass filter (BPF) 202, BPF 204, FM modulator 206, mixer 20.
8 and AD converter 210. The operation of recording the accompaniment signal as an FM modulated wave in the LD by this circuit will be described below.

First, the music signal is converted into a BP in the band of DC to 16 kHz.
It is given to F202 to remove the voice signal other than this band.
Further, the accompaniment signal is given to the BPF 204 in the band of 100 Hz to 500 Hz to remove the accompaniment signal other than this band.

Next, the accompaniment signal of 100 Hz to 500 Hz is FM-modulated by the FM modulator 206 using a carrier wave (FIG. 6B) having a center frequency f ₀ of 18 kHz. Then, the music signal output from the BPF 202 and the modulated accompaniment signal output from the BPF 204 are input to the mixer 208. Next, the combined analog signal is digitally converted by the AD converter 210 to obtain a total digital signal. As described above,
The FM-accompanied accompaniment signal is recorded in a band where the LD music signal is not used.

FIG. 2 shows an accompaniment signal reproducing circuit for taking out the recorded accompaniment signal. This circuit is a DA converter 21
2, BPF 214, BPF 216, and FM detector 218. The operation of extracting the accompaniment signal recorded as an FM modulated wave in the LD by this circuit will be described below.

First, the recorded total digital signal is D
An analog signal is converted by the A converter 212. This analog signal is then provided to BPFs 214,216. Therefore,
The accompaniment signal is output and taken out from the BPF 216 in the band of 16 kHz to 20 kHz. Next, the signal is demodulated by the FM detector 218 and extracted as an accompaniment signal of 100 Hz to 500 Hz. Further, the music signal is output and taken out from the BPF 214 in the band of DC to 16 kHz. As described above, by using the FM-modulated wave as the accompaniment signal, the accompaniment signal is recorded / reproduced in / from the LD.

Next, as another embodiment, the accompaniment signal is set to AM.
A method of recording and reproducing the accompaniment signal on the LD by using the modulated wave will be described. As described above, the frequency band of the music signal used in the karaoke device is DC to 16 kHz. In this frequency band, as shown in FIG. 3a, it is difficult for the human ear to hear frequencies around 3 kHz, and it is considered that it is audibly preferable to remove frequencies in this band. This invention, as shown in FIG.
The accompaniment signal is AM-modulated and recorded in this band.

FIG. 4 shows an accompaniment signal recording circuit according to this embodiment. This circuit has a narrow band rejection filter (BRF) 22
0, AM modulator 222, mixer 224, AD converter 226. Hereinafter, the operation of recording an accompaniment signal as an AM modulated wave on the LD by this circuit will be described.

First, the frequency of the accompaniment signal is set to, for example, 1/5.
Count down to 20-100 Hz. This is because the accompaniment signal to be recorded has a frequency in a narrow band as much as possible to narrow the removal frequency band of the music signal and reduce the influence on the music signal. This signal is converted to an AM modulator 2 using a carrier (FIG. 3b) having a center frequency f ₀ of 3 kHz.
AM modulation by 22. On the other hand, music signals from 2.9kHz to 3.1k
It is applied to the BRF 220 in the Hz band to remove the music signal in this band. Then, the music signal output from the BRF 220 and the modulated accompaniment signal output from the AM modulator 222 are input to the mixer 224. Next, the combined analog signal is digitally converted by the AD converter 226 to form a total digital signal. As described above, the AM-accompanied accompaniment signal is recorded in the band where the LD music signal is not used.

FIG. 5 shows an accompaniment signal reproducing circuit for taking out the recorded accompaniment signal. This circuit is a DA converter 22
8, BRF230, BPF232, AM detector 234. The operation of extracting the accompaniment signal recorded as an AM modulated wave in the LD by this circuit will be described below.

First, the recorded total digital signal is D
An analog signal is converted by the A converter 228. Next, this analog signal is given to BRF230 and BPF232. As for music signals, accompaniment signals are recorded by BRF230.
It is taken out after removing the band from 2.9kHz to 3.1kHz. The accompaniment signal is output from the BPF 232 in the band of 2.9 kHz to 3.1 kHz and taken out. Next, demodulate by AM detector 234,
Extracted as an accompaniment signal of 0Hz to 100Hz. This signal is multiplied to 100 Hz to 500 Hz. In this way, the accompaniment signal is recorded and reproduced on the LD by using the AM-modulated wave as the accompaniment signal.

In this way, the accompaniment signal is recorded as a modulated wave in the LD, and is reproduced and taken out from the LD. Since this accompaniment signal is a sine wave signal, in the following key controller device, when the digital frequency signal of the accompaniment signal is taken out, it is possible to perform accurate key control without erroneously recognizing the accompaniment signal.

Next, FIG. 9 shows the configuration of an automatic key controller device for a karaoke machine according to an embodiment of the present invention. The accompaniment frequency extracting means 10 obtains a digital frequency signal representing the frequency of each note of the accompaniment signal reproduced by the accompaniment signal recording / reproducing method described above. Song frequency extraction means 1
2 obtains a digital frequency signal representing the frequency of each note of the song signal. The accompaniment highest frequency detecting means 14 obtains the highest frequency for every predetermined unit time among the digital frequency signals of the accompaniment signal obtained from the accompaniment frequency extracting means 10. The song highest frequency detecting means 16 obtains the highest frequency of the digital frequency signals of the song signal obtained from the song frequency extracting means 12 every predetermined unit time.

The calculating means 18 calculates the difference frequency between the highest frequency of the accompaniment signal obtained from the highest accompaniment frequency detecting means 14 and the highest frequency of the singing signal obtained from the highest singing frequency detecting means 16 every predetermined unit time. The selecting means 20 selects the most frequent difference frequency within a predetermined time period from the difference frequencies obtained by the calculating means 18. The key correction means 22 is
The accompaniment key is corrected based on the shift of the key corresponding to the difference frequency selected by the selection means 20, and the key controller is operated.

Generally, as shown in FIG. 10, the pitch (key) of the female accompaniment signal in a karaoke piece is G ₅
~ C ₄ is almost the same. In contrast, considering that singer singing keys varies to # 4～B4, the frequency band of the female singer singing signal, 988Hz (B ₅₎ ~208Hz
Can be regarded as

On the other hand, the keys of the accompaniment signal for men are G ₄ to
Almost aggregated to C ₃ . Similarly, considering that the keys of the singer change, the frequency band of the song signal of the male singer can be regarded as 494 Hz to 104 Hz (G ₂ ). Thus, the key accompaniment signals substantially are aggregated into G ₅ -C _3, key songs signals are substantially summarized in B ₅ ~G _2.

Further, regarding the keys of the accompaniment signal and the keys # 4 to b4 of the song signal, when the frequency of the key of the accompaniment signal is 1,
The frequency of the key of the song signal varies from 1.26 to 0.79. For example, 988 Hz of # 4 is 1.26 for 784 Hz (natural) of the accompaniment signal G ₅ . This device is premised on the above.

11 and 12 are detailed views of FIG.
FIG. 11 shows an accompaniment signal digital conversion circuit 10a of an accompaniment frequency extraction circuit 10 which is an accompaniment frequency extraction means, a song signal digital conversion circuit 12a of a song frequency extraction circuit 12 which is a song frequency extraction means, and a control signal generation circuit 10c. ..
Further, FIG. 12 shows an accompaniment signal digital frequency generation circuit 10b and a song signal digital frequency generation circuit 12b. The operation of this automatic key controller device for a karaoke machine will be described below with reference to FIGS. 9 and 11 and 12.

FIG. 11 shows an accompaniment signal digital conversion circuit 1
0a is shown. As described above, the accompaniment signal used here is a signal recorded and reproduced in the LD. Therefore,
Since this signal is a sine wave representing each note in the accompaniment,
It can be used as it is as an accompaniment signal. Less than,
The operation of extracting the digital frequency of the accompaniment signal will be described with reference to FIG.

First, the accompaniment signal is, for example, a 16-bit A
It is input to the D converter 40. The AD converter 40 converts the input analog signal into a sampling frequency fs (for example, fs = 4).
It is converted to a digital frequency signal at 4.1 kHz), but in this embodiment, a parallel system is used to detect a zero cross which will be described later.

FIG. 13 shows the level of this digital signal.
An example is shown using a 16-bit parallel bipolar code.
When the input analog signal is positive, the MSB is "0", and when it is negative, the MSB is "1". This indicates whether the digital signal has a positive polarity or a negative polarity. In the case of positive polarity, the level is expressed by sequentially setting “1” in 2SB to LSB. When the polarity is negative, the level is expressed by sequentially setting “0” in 2SB to LSB.

Next, the accompaniment signal digital conversion circuit 10a.
The accompaniment signal digitally converted by is sent to the accompaniment signal digital frequency generation circuit 10b as shown in FIG. The accompaniment signal digital frequency generation circuit 10b is
Digital BPF 52 _{1 to} 52 ₃₂ , SW2, SW3, logic circuit 5
6, 62, 68, 70, 72, 74, a buffer 58, a flip-flop 64, and a delay circuit 66.

FIG. 14 shows the state of the frequency sequence of the digital BPFs 52 _{1 to} 52 ₃₂ . The digitally converted accompaniment signals are 32 digital BPs in the frequency band of G _{5 to} C _3.
F521～52 ₃₂ will be parallel input to one of the.

Next, the data signals taken into the digital BPFs 52 _{1 to} 52 ₃₂ are detected. FIG. 15 shows a zero cross state of the data signal. To detect the data signal, so that the data signal shown in FIG. 15A, when the time of change from the positive polarity signal to a negative polarity signal (P1), and a negative polarity signal changes to the positive polarity signal (P ₂₎ i.e. the zero-crossing Should be taken out as a signal. That is, as shown in FIG. 15B, the zero cross is output as a pulse signal,
If the number of pulse signals is counted within the detection time T _K ,
It can be extracted as a digital frequency signal representing the frequency of each note of the accompaniment signal. For example, if the digital signal has a positive polarity, the zero-cross is the MSB in the case of FIG.
~ 15 SB is represented by "0000000000000001" where 0 is 0 and LSB is 1.

Returning to FIG. 12, in this way, the BPF is
The signals taken out from 52 _{1 to} 52 ₃₂ are converted into negative polarity signals by SW2. As a result, all zero crosses are displayed as "0000000000000001", which makes it easy to detect zero crosses. After that, 2S of data signal
Input B to 15SB into OR56. When there is a zero cross, there is no bit in 2SB to 15SB of the logic circuit 54, and the output of the OR 56 is "0".

If the output of OR56 is "0", buffer 5
2SB to 15SB of the data signal passed through 8 are forcibly set to "1" by SW2 to be a full-bit signal. When these signals are input to AND62, the output of AND62 becomes "1". Therefore, a data signal is generated. As a result, as shown in FIG.
Pulse signal is generated. The number of occurrences of this pulse is counted within the detection time T _k (for example, 100 msec) and extracted as a digital frequency signal of the accompaniment signal.

On the other hand, other than zero crossing, that is,
If bits exist in 2SB to 15SB, the output of OR56 is
It is "1". If the output of OR56 is "1", the buffer
2SB to 15SB of the data signal through 58 are all SW3,
Forced to "0". Therefore, the output of AND62 is "0".
Therefore, no accompaniment signal is generated.

As shown in FIG. 16, the detection time T _k is 100 ms.
When ec is set, there are pulse numbers corresponding to the respective frequencies. For example, with a frequency of 988 Hz, there are 197 pulses. Thus, the digital frequency of the accompaniment signal can be specified by the number of pulses extracted by the accompaniment frequency extraction circuit 10. Here, the detection time T _{k is set} to 100 msec for the following reason.

In a normal song used for karaoke, the length of each note of the song is about 1 sec for a half note, about 500 msec for a quarter note, about 250 msec for an eighth note, and about 125 msec for a 16th note. is there. Therefore, if 16th notes continue, 125 msec
Each time the frequency of the accompaniment signal changes. Furthermore, a pause time of 20 msec for recognizing the pulse signal is also required. Therefore, in order to recognize the digital frequency signal, it must be 100 msec or less. Therefore, in this embodiment, the detection time T _k for detecting the accompaniment signal is 100 ms.
ec

The detection time T _k of 100 msec is generated by the accompaniment signal digital frequency generation circuit 10b as shown in FIG. First, AND to the flip-flop 64
From the OR 56 to which the data signal is input from each BPF 52 by the start of accompaniment, using the output of 62 as the clock signal,
A signal that becomes "1" is output only for the duration of the data signal, and this "1" is input as a clear signal.

Then, the output of the flip-flop 64 and the output delayed by 100 msec by the delay circuit 66 are AND 70
To enter. As a result, if there is an output from the AND62, the signal that becomes "1" is always output from the AND70 for 100 msec.

Next, the output of the AND70 and the AND62
The output of and is input to the AND68, a serial signal with a duration of 100 msec is obtained, and this signal and a performance signal described later are input to the AND72, and the performance signal is maintained for 100 msec only when the performance signal is "1". Create a serial signal. The detection time T _k is created as described above.

On the other hand, with respect to the song signal, the song frequency extracting circuit 12 extracts the digital frequency signal of the song signal. This state is shown in FIGS. The song signal is passed through the microphone 44 and the song signal digital conversion circuit 12a.
Entered in.

Next, the song signal has a bandwidth of 100 Hz to 1000 Hz.
The high frequency component of the accompaniment signal that is input to the analog pass band filter (BPF) 46 and is taken out is removed as much as possible. After that, it is input to the 16-bit AD converter 48,
Like the accompaniment signal, it is converted into a digital signal.

Incidentally, the song signal digital frequency generation circuit 1
The 2b, considering that the singer singing keys varies to # 4~b4, as shown in FIG. 17, B ₅ ~ # with total of 56 installed BPF in the frequency permutation to G ₂ There is.

With such a circuit, the digital frequency of the song signal is similarly extracted as the number of pulses.

FIG. 18 shows an example of concrete hardware when the accompaniment highest frequency detecting means 14, the song highest frequency detecting means 16, the calculating means 18, and the selecting means 20 of FIG. 9 are constituted by a CPU. The CPU 60 has a ROM 112, R
AM114, CRT116, keyboard 118, accompaniment signal input 12
0, song signal input 122, measurement signal input 124, performance signal input 12
6, key correction output 128 is connected. CPU60 is R
Each unit is controlled according to the program stored in the OM 112. The RAM 114 stores the difference frequency between the maximum digital frequency MM1 of the accompaniment signal and the maximum digital frequency MM2 of the song signal.

FIG. 19 is a basic flowchart of the digital signal processing of the accompaniment signal and the song signal.
22, 23 and 24 show detailed flow charts of this. The operation of signal processing the digital frequencies of the accompaniment signal and the song signal will be described below with reference to FIGS.

First, as shown in the flowchart of FIG. 19, in step S1, frequency measurement A of the accompaniment signal and frequency measurement B of the song signal are alternately performed at a detection time of 100 msec to detect M1 and M2, and a predetermined unit time is detected. Measurement time T which is
The maximum digital frequency MM1 of the accompaniment signal and the maximum digital frequency MM2 of the song signal in _m (for example, for 2 seconds) are detected. This state is shown in the flowcharts of FIGS. Hereinafter, the frequency measurement A of the accompaniment signal shown in the flowchart of FIG. 20 will be mainly described.

[0063] First, CPU 60, all flags, the memory is initialized (step S1 _0). Next, the pulse of the performance signal input 126 determines whether "1" or "0" (Step S1 _1). If it is "1", the process proceeds to step S1 _3. "0"
If, end flag is determined whether "1" or "0" (Step S1 _1). If it is "0", the flow returns to step S1 _1. If it is "1", the process proceeds to step S3 (step S
1 ₂ ).

Here, the performance signal input 126 is a signal in which "1" or "0" lasts for one chorus (for example, about 60 seconds) which is a predetermined time, and the end flag means that one chorus ends. It means a flag indicating that it has been done.

This performance signal is rectified and integrated by the rectifier 76 and the integrator 78 from the start to the end of the accompaniment in the control signal generation circuit 10c shown in FIG.
By shaping the waveform with the voltage comparator 80, it is created as a signal having one chorus duration. Also, the measurement time T _m is 2
sec is created by inputting the output of the 2 sec pulse generator 104 and the performance signal created as described above to the AND 106 and outputting the result.

[0066] Returning to the flowchart of FIG. 20, then, in step S1 _3, CPU 60 is a pulse accompaniment signal input 120 to determine whether "1" or "0" (Step S1 _3a). If the pulse of the accompaniment signal input 120 is "1", step S1
_Continue to _3b . In step S1 _3b , it is confirmed whether the timer flag 1 is set. The timer flag 1 is a flag for recognizing that the pulse "0" of the accompaniment signal input 120 continues for 20 msec. The timer flag 1 is set to "1" when 20 msec has elapsed from the start of the timer of the 20 msec timer 1, so that the timer flag 1 being "1" means that the accompaniment signal pulse has not been generated for 20 msec or more. ..

Next, the measurement flag 1 is set to "1" and the 100 msec timer 1 is started (step S
1 _3c ). The measurement flag 1 is a flag that permits counting of the number of pulses of the accompaniment signal input 120. After that, the timer flag 1 is set to "0" to start the 20 msec timer (step S1 _3d ). Note that when the timer flag 1 in step S1 _3b is "0", it is the pulse of the accompaniment signal input 120 after the second time after the rising edge, so the step S1 is directly performed.
Go to _3d .

Next, the CPU 60 confirms whether or not the input flag 1 is set (step S1 _3e ). The input flag 1 is a flag for recognizing the rising edge of the accompaniment signal input 120. When the input flag 1 is "0", the input flag
1 is set to "1" (step S1 _3f ). Note that step S
When the input flag 1 is 1 in 1 _3e , the accompaniment signal input 120 has not risen, so steps S1 _3f to S1 _3h.
The frequency measurement B of the song signal is performed without performing.

Next, after setting the input flag 1 to "1", it is confirmed whether or not the measurement flag 1 is set (step S1 _3g ). When the measurement flag 1 is "1", a 100 msec timer is started to count the number of pulses of the accompaniment signal input 120 (step S1 _3h ). Then, the frequency measurement B of the song signal is performed in the same manner as when the measurement flag 1 is "0".

As described above, the CPU 60 uses the 100 msec timer 1
When 100 msec has elapsed from the start of the timer, the measurement flag 1 is set to "0", so that the number of pulses of the accompaniment signal is counted only while the measurement flag 1 is "1".

Next, in step _S13a , when the measurement flag 1 is "1", that is, when the counting of the number of pulses of the accompaniment signal input 120 is permitted, frequency measurement B of the song signal is performed. When the accompaniment signal input 120 is "0", step S1 _3i
Go to. In step S1 _3i , the input flag 1 is set to "0".

Next, it is confirmed whether or not the measurement flag 1 is set (step S1 _3j ). Measurement flag 1 is "0"
That is, when the counting of the number of pulses of the accompaniment signal input 120 is not permitted, it is determined whether the measurement signal input 124 is "1" or "0" (step S1 _3k ). Measurement signal input 124 is "1"
, That is, when the measurement signal is the measurement time T _m (2 sec)
When continuing, M1 counted in step S1 _3h and maximum value MM1 already stored in RAM 114
And are compared (step S1 _3l ).

Next, the counted M1 is compared with the already stored MM1. If M1 is large, the maximum value MM
Replace 1 and store in RAM 114 (step S
1 _3m ). As a result, the maximum value MM1 of the pulse count number of the accompaniment signal input 120 is detected.

FIG. 25 shows an example of this state. Figure 25A
The accompaniment signal shown in is detected as the number of pulses for 100 msec by the AND68 shown in FIG.
29 waves, ..., which are serially output.
Therefore, as shown in FIG. 25B, the first measurement time T
_{At m1} (2 sec), the number of pulses is 39 and the maximum value of the accompaniment signal is MM.
1. At the next measurement time T _m2 (2 sec), 29 pulses are detected as the maximum value MM1 of the accompaniment signal.

Next, when the measurement signal input 124 is "0", it is confirmed whether or not the measurement of the song signal input 122 of the measurement flag 2 is completed (step S1 _3n ). If the measurement end is "0", the process proceeds to step S2. If the value is "1" that has not been measured, frequency measurement B of the song signal is performed.

[0076] On the other hand, as shown in the flowchart of FIG. 21, CPU 60 performs alternately also frequency measurement B of the song signal (Step _{S1 3 'a ~1 3' n} ). Then, the maximum value MM2 of the pulse count number of the song signal input 122 is detected, and similarly, MM2 at the measurement times T _m1 , T _{m 2} , ... Is detected and stored in the RAM 114.

In this way, the frequency measurement A of the accompaniment signal and the frequency measurement B of the song signal are alternately performed, and the measurement times T _m1 and T
The maximum value MM1 of the pulse count number of the accompaniment signal input 120 and the maximum value MM2 of the pulse count number of the song signal input 122 are detected and stored in the RAM 114 every 2 seconds, such as _m2 .

Next, as shown in FIG. 19, step S1
To step S2. In step S2, the CPU 60
In step S1, the measurement times T _m1 , T _m2 , ...
MM1 and MM2 stored in the RAM 114 for each measurement time, the difference frequency R = MM2 / M for each measurement time T _m1 , T _m2 , ...
Calculate M1.

Here, the difference frequency R corresponds to any of # 4 to b4. As shown in FIGS. 10 and 16, for example, MM1 is 784 Hz (pulse number 156) and MM2 is 988H.
If z (the number of pulses is 197), R is 1.26 and corresponds to # 4. Therefore, within one chorus, the difference frequency # 4 to b4
Is detected. Then, the calculated appearance frequencies of the difference frequencies # 4 to b4 are added and stored in the RAM 114. The details are shown in the flowchart of FIG. The operation of this flowchart will be described below.

[0080] First of all, all of the output to clear (step S2 _0). Then, CPU 60 calculates the R = MM2 / MM1 (Step S2 _1). Then, the quotient R is large or small determination of whether than 1.5 (step S2 _2). If R ≧ 1.5, R is halved (step S2 ₃ ) and step S2 ₆
Proceed to.

If R <1.5, it is determined whether R is larger than 0.667 (step S24). If R ≦ 0.667, R is doubled (step S2 ₅ ) and step S2
Go to ₆ . In this way, when a female singer sings with a male accompaniment or a male singer sings with a female accompaniment, the accompaniment key is automatically doubled or halved to the song key.

[0082] Next, in step S2 ₆ to 2 _35, the key output (# 4~b4) according to the value of the resulting R, # 4~b4
The number of appearances is added for each. For example, if 1.23 ≦ R <1.30, the process proceeds to step S2 _8, the key output # 4 (Step S2 _9), is added to the number of times that appeared it previously (Step S2 _10). If 1.16 ≦ R <1.23,
The process proceeds to step S2 _11, # 3 of the key output (step S2 _12), is added to the number of times that appeared it previously (Step S2 _13). Hereinafter, similarly, the number of appearances is added for each of # 4 to b4.

However, if R ≧ 1.30 or R <0.77, it is determined that the key cannot be corrected, an error is output, and the process returns to A (steps S2 ₆ and S2 ₃₅ ).

In this way, the number of times each of the difference frequencies # 4 to b4 detected during one chorus is added, and the RAM 11 is added.
Remember to 4.

Next, if one chorus has ended and the end flag is "1", as shown in FIG. 19, step S3
Advances to (Step S1 _2). In step S3, RAM1
The frequency of occurrence of the difference frequencies # 4 to b4 stored in 14 is compared, and the difference frequency # 4 to b4 having a high frequency is selected. The details are shown in the flowchart of FIG. The operation of this flowchart will be described below.

[0086] First of all, all of the output to clear (step S3 _0). Next, MAX1 is set to M # 0 (the number of occurrences of natural) and MAX2 is set to 0 (step S3 ₁ ). M
AX2 corresponds to MAX1, for example, MAX1 is M
If # 0, MAX2 represents 0, if MAX1 is Mb1 (the number of appearances of b1), MAX2 is 1, and if MAX1 is M # 1 (the number of appearances of # 1), MAX2 is 2.

Next, M # 0 (the number of appearances of natural) of MAX1 is compared with Mb1 (the number of appearances of b1) (step S3 ₂ ). If M # 0 of MAX1 is smaller than Mb1, M
It rewrites AX1 to Mb1 (Step S3 _3). MAX1
If the M # 0 is greater than Mb1, the process proceeds to step S3 _4. In step S3 _4, M being written to MAX1
# 0 or Mb1 is compared with M # 1 (the number of appearances of # 1). If MAX1 is less than M # 1, rewrites the MAX1 to M # 1 (Step S3 _5). If MAX1 is greater than M # 1, the process proceeds to step S3 _6. In the same way,
Performed until M # 4 (Step S31 _7), for selecting the difference frequency # 4～B4 most frequent among # 4~b4.

Next, as shown in FIG. 19, step S4
Proceed to. In step S4, the selected difference frequency # 4 ...
The key correction output 128 is performed according to b4. This detail is shown in FIG.
Is shown in the flow chart. The operation of this flowchart will be described below.

[0089] First, comparing the MAX1 and number 3 (Step S4 _0). It is smaller than the number 3 Natural output (Step S4 _1). Here, the reason for setting 3 times is as follows:
This is to ensure accuracy, and if it is twice or less, it is considered that there is no need to correct the key. Proceed larger than the number 3 in step S4 _2, MAX2 is 0, i.e. MAX1 determines whether it is natural. M
If AX2 is 0, and natural outputs proceeds to step S4 _1. If MAX2 is not 0, it is determined whether MAX2 is 1, that is, MAX1 is b1 (step S4).
₃ ). If MAX2 is 1, the routine proceeds to step S4 ₄ b1
Output. If MAX2 is not 1, it is determined whether MAX2 is 2, that is, MAX1 is # 1 (step S4).
₃ ). Thereafter, in the same manner, it is determined whether MAX2 is 7 or not, and # 4 to b4 selected in step S2 are subjected to key correction output 128.

Next, the key controller is operated based on this key correction output 128.

As described above, the most frequent difference frequency is selected three times or more in one chorus, the key is corrected and output, and the key controller is operated. Note that b1, # 1, b2, # 2, ...
・ In order to judge that, if there is a difference frequency with the same frequency three or more times, a key close to natural is output.

As described above, according to this accompaniment signal recording / reproducing method, at the time of recording, the accompaniment signal having the same frequency as the frequency of the accompaniment note is added as a modulation wave to the unused band of the music signal and recorded on the LD. At the same time, during playback,
This modulated wave is demodulated to obtain an accompaniment signal. Therefore, it is possible to obtain a sinusoidal accompaniment signal from the LD by recording the sinusoidal accompaniment signal as a modulation wave in a band in which the LD music signal is not used. As a result, an accurate accompaniment signal can be input to the automatic key controller device for a karaoke device.

This automatic key controller device for karaoke apparatus calculates the difference frequency between the highest frequency of the accompaniment signal and the highest frequency of the singing signal obtained by the accompaniment signal recording / reproducing method for each measurement time, and determines the highest frequency within the performance time. Select the frequency with a high frequency and correct and output it. Therefore, the shift between the singing key and the accompaniment key is detected, and the accompaniment key is corrected based on the shift between the keys. As a result, even if the singing key is deviated from the accompaniment key, the accompaniment key can be automatically matched with the singing key automatically, so that it is possible to sing with the accompaniment that matches the own key.

Further, even if there is a difference in accompaniment or voice octave between female and male, it is possible to detect the shift between the song key and the accompaniment key. As a result, regardless of whether the character is female or male, the accompaniment key can be automatically set to the singing key, so that the accompaniment that matches the own key can be sung as it is.

Since the deviation between the accompaniment key and the key of one's own song can be displayed on the CRT 116 at each measurement time, it is possible to confirm on the screen which part of one's own song is deviated.

Further, in this embodiment, the duration of the performance signal is set to 1 chorus (about 60 seconds), but it may be set to ten and several seconds.

[0097]

According to the accompaniment signal recording / reproducing method of the first aspect, at the time of recording, an accompaniment original signal having the same frequency as the frequency of the accompaniment note is added as a modulated wave to a predetermined band of the music signal band. The data is recorded in the (LD), and at the time of reproduction, the accompaniment original signal is taken out to obtain an accompaniment signal. Therefore, the sinusoidal accompaniment signal can be recorded as a modulation wave in a band where the music signal of the recording medium is not used, and the sinusoidal accompaniment signal can be obtained from the recording medium. As a result, an accurate accompaniment signal can be input to the automatic key controller device for a karaoke device.

According to a second aspect of the accompaniment signal recording / reproducing method, in the accompaniment signal recording / reproducing method of the first aspect, the modulated wave is an FM wave. Therefore, the sine wave accompaniment signal can be FM-modulated and recorded in a band where the music signal of the recording medium is not used.

According to a third aspect of the accompaniment signal recording / reproducing method, in the accompaniment signal recording / reproducing method of the first aspect, the modulated wave is an AM wave. Therefore, the sine wave accompaniment signal can be AM-modulated and recorded in a band where the music signal of the recording medium is not used.

In the automatic key controller device for a karaoke device according to claim 4, the accompaniment frequency extracting means obtains a digital frequency signal representing the frequency of each note of the accompaniment signal obtained by the accompaniment signal recording / reproducing method according to claim 1. The song frequency extracting means obtains a digital frequency signal representing the frequency of each note of the song signal. The accompaniment highest frequency detecting means obtains the highest frequency for every predetermined unit time among the digital frequency signals of the accompaniment signal obtained from the accompaniment frequency extracting means. The highest song frequency detecting means obtains the highest frequency every predetermined unit time from the digital frequency signals of the song signal obtained from the song frequency extracting means. The calculating means calculates a difference frequency between the highest frequency of the accompaniment signal obtained from the highest accompaniment frequency detecting means and the highest frequency of the singing signal obtained from the highest singing frequency detecting means, every predetermined unit time. The selecting means selects, from the difference frequencies obtained by the calculating means, the most frequent difference frequency within a predetermined time. The key correction means corrects the accompaniment key based on the shift of the key corresponding to the difference frequency selected by the selection means.

Therefore, the difference frequency between the highest frequency of the song signal and the highest frequency of the accompaniment signal is calculated every predetermined unit time,
By selecting the difference frequency having the highest frequency in a predetermined time, a shift between the song key and the accompaniment key is detected, and the accompaniment key is corrected based on the shift between the keys. As a result, even if the singing key is deviated from the accompaniment key, the accompaniment key can be automatically matched with the singing key automatically, so that it is possible to sing with the accompaniment that matches the own key.

In the automatic key controller device for a karaoke device according to claim 5, the arithmetic means further divides the highest frequency of the accompaniment signal by the highest frequency of the song signal, and the quotient is 2
If it is more than or equal to or less than 1/2, the difference between the highest frequency of the accompaniment signal multiplied by 1/2 or 2 and the highest frequency of the song signal is calculated. Therefore, even if there is a difference in accompaniment or voice octave between a female and a male, it is possible to detect the shift between the song key and the accompaniment key. As a result, regardless of whether the character is female or male, the accompaniment key can be automatically set to the singing key, so that the accompaniment that matches the own key can be sung as it is.

[Brief description of drawings]

FIG. 1 is a diagram showing an accompaniment signal recording circuit according to an embodiment of the present invention.

FIG. 2 is a diagram showing an accompaniment signal reproducing circuit according to an embodiment of the present invention.

FIG. 3 is a diagram showing a deafness band of a music signal.

FIG. 4 is a diagram showing an accompaniment signal recording circuit according to another embodiment.

FIG. 5 is a diagram showing an accompaniment signal reproducing circuit according to another embodiment.

FIG. 6 is a diagram showing bands of music signals.

FIG. 7 is a diagram showing a musical score of one song.

FIG. 8 is a diagram showing each frequency of musical notes on a musical score.

FIG. 9 is a diagram showing a basic configuration of an automatic key controller device for a karaoke device according to an embodiment of the present invention.

FIG. 10 is a diagram showing a range of pitches in a song.

FIG. 11 is a diagram showing a digital conversion circuit for accompaniment signals and song signals according to an embodiment of the present invention.

FIG. 12 is a diagram showing a digital frequency generation circuit for accompaniment signals and song signals according to an embodiment of the present invention.

FIG. 13 is a diagram showing a 16-bit parallel bipolar code.

FIG. 14 is a diagram showing frequency permutations of accompaniment signals.

FIG. 15 is a diagram showing a zero-cross state of a signal.

FIG. 16 is a diagram showing the number of pulses corresponding to each frequency.

FIG. 17 is a diagram showing frequency permutations of a song signal.

FIG. 18 is a diagram showing a hardware configuration of a CPU 60.

FIG. 19 is a diagram showing a basic flowchart of digital signal processing of an accompaniment signal and a song signal.

FIG. 20 is a diagram showing a flowchart of frequency measurement of an accompaniment signal.

FIG. 21 is a view showing a flowchart of frequency measurement of a song signal.

FIG. 22 is a diagram showing a flowchart for calculating a difference frequency and counting the number of appearances thereof.

FIG. 23 is a diagram showing a flowchart for selecting the most frequent difference frequency.

FIG. 24 is a diagram showing a flowchart for performing key correction output according to a selected difference frequency.

FIG. 25 is a diagram showing a state in which a maximum value of an accompaniment signal is detected for 2 seconds.

[Explanation of symbols]

 202 ・ ・ ・ BPF (band pass filter) 204 ・ ・ ・ BPF 206 ・ ・ ・ FM modulator 208 ・ ・ ・ Mixer 210 ・ ・ ・ AD converter

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yoshiyoshi Ito 2-7-18 Shibata, Kita-ku, Osaka, Toshiba Toshiba Corporation

Claims (5)

[Claims] 1. At the time of recording, an accompaniment signal having the same frequency as the frequency of an accompaniment note is added as a modulation wave to a predetermined band of the music signal band and recorded on a recording medium, and at the time of reproduction, the modulation wave is recorded. Accompaniment signal recording / reproducing method characterized by demodulating to obtain an accompaniment signal. 2. The accompaniment signal recording / reproducing method according to claim 1, wherein the modulating wave is an FM wave. 3. The accompaniment signal recording / reproducing method according to claim 1, wherein the modulation wave is an AM wave. 4. An accompaniment frequency extracting means for obtaining a digital frequency signal representing a frequency of each note of an accompaniment signal reproduced by the accompaniment signal recording / reproducing method according to claim 1, and a digital frequency signal representing a frequency of each note of a song signal. Acquiring song frequency extracting means, Accompaniment frequency digital means of accompaniment signal obtained from accompaniment frequency extracting means, maximum accompaniment frequency detecting means for obtaining the highest frequency every predetermined unit time, song frequency digital signal of song signal obtained from extracting means Of the highest frequency of the song signal obtained from the highest accompaniment signal obtained from the accompaniment highest frequency detection means and the highest accompaniment signal obtained from the highest accompaniment frequency detection means. The calculating means for calculating every predetermined unit time, the difference frequency most frequent within the predetermined time among the difference frequencies obtained by the calculating means An automatic key controller device for a karaoke device, comprising: selecting means for selecting a wave number; and key correcting means for correcting an accompaniment key on the basis of a key shift corresponding to a difference frequency selected by the selecting means. . 5. The automatic key controller device for a karaoke apparatus according to claim 4, wherein the calculating means divides the highest frequency of the accompaniment signal by the highest frequency of the song signal, and the quotient is 2 or more or 1/2 or less. For example, an automatic key controller device for a karaoke device, characterized in that the difference between the highest frequency of the accompaniment signal multiplied by 1/2 or 2 and the highest frequency of the song signal is calculated.