JPH0338600B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a device that stores music information having at least scale data and note length data as bit patterns in a semiconductor memory, etc., and plays music according to reading of this music information, without using information bits for tremolo. The present invention relates to a music playing device that uses note length data read from a music information memory to produce a tremolo effect and exhibit even more acoustic effects. In this type of device, it is desirable to reduce the amount of data in the song information memory as much as possible because of the limited capacity of the song information memory and the amount of control required for various reproductions. The tremolo effect brings an even greater acoustic effect to the played sound, but rather than applying the tremolo effect to each note, applying it only to long notes creates a lingering sound on the long notes, making the sound effect more pronounced. appears. The present invention uses note length data read out from the music information to produce an effective tremolo effect with slight hardware improvements without using dedicated bits. An embodiment of the present invention will be described below with reference to the drawings. <Basic Structure> FIG. 1 shows an overall circuit block diagram. The device can be roughly divided into an oscillation/frequency dividing circuit 1, a tone generator section 2, an envelope generation section 3,
It is composed of an external input circuit 4, a control circuit 5, an address counter 6, a music information memory 7, an amplifier 9, and a speaker 10. The oscillation/frequency divider circuit 1 includes, for example, a crystal oscillation circuit and a frequency divider circuit, and generates a basic clock for the device. The tone generator section 2 is a section that generates a corresponding frequency signal according to data output from the music information memory 7 (for example, composed of a ROM). The envelope generator 3 generates the length and envelope of the sound according to the data output from the music information memory 7. The input circuit 4 is a switch circuit that controls music selection, music start, transposition/modulation, tempo, etc. The control circuit 5 converts the switch input of the input circuit 4 into a signal and transmits it to each block. The address counter 6 specifies the address of the music information memory 7, and sequentially reads the music data stored in the music information memory 7. FIG. 2 shows the bit assignment of data stored in the music information memory 7. That is, 16 of B ₀ to B ₁₅
Bits B〓~ _B3 (4 bits)...Scale bits _B4 ~ _B7 (4 bits)...Note length bits _B8 , _B9 (2 bits)...Octave bits _B10 , _B11 (2 bits)... Instrument recombination bits _B12 , _B13 (2 bits)...Sound strength bit _B14 (1 bit)...Redundancy bit _B15 (1 bit)...Assigned as a stop bit. <Specific configuration and operation> 1. First, a song is selected using the song designation switch of the input circuit 4. By operating this switch, the control circuit 5 decodes 3 into M ₁ to M ₃ .
Outputs a bit pattern and specifies the top address of the music information memory 7 for reading. 2. Next, when the start switch of the input circuit 4 is turned on, the music information bit pattern at the address specified above is read out. Among the bit patterns read out, B〓～ _B3 (scale bits),
The 8 bits from _B8 to _B11 (octave bits, instrument recombination bits) are the tone generator section 2.
Also, B ₄ ~ _{B 7} (note length bit), B ₁₀ ~ _{B 13}
The 8 bits (instrument recombination bit, sound strength bit) are transmitted to the envelope generator 3. B ₁₅ (stop bit) is transmitted to the control circuit 5. When "1" is output to the _B15 bit, the operation is stopped, but at the time of start and when playing music, it is "0" and does not affect the operation in any way. 3 Of the 8 bits transmitted to the tone generator section 2, 6 bits B ₀ to _{B 3} , B ₈ , and B ₉ are input to the adder/subtractor circuit 2-6, and the addresses of the scale division ratio memory 2-5 are input to the adder/subtracter circuit 2-6. At the same time, the octave is selected using the octave selector 2-2. The addition/subtraction circuit 2-6 controls the data of B≦ to B ₃ , B ₈ , and B ₉ by inputting an external switch from the input circuit 4 to enable transposition and modulation. 4. One of the fundamental frequencies ₀₁ to ₀₄ corresponding to each octave is selected by the octave selector 2-2. The frequency divider circuit 2-1 is connected to the oscillation/frequency divider circuit 1, and is multiplied by (1,
2, 4, 8 times) fundamental frequencies ₀₁ to ₀₄ are prepared. The fundamental frequency _0i selected by the octave selector 2-2 is input to the frequency dividing circuit 2-3 and divided. The 9-bit binary frequency division output of the frequency dividing circuit 2-3 is compared with the same 9-bit frequency division ratio output from the frequency division ratio memory 2-5 specified by the above address in the matching circuit 2-4,
When they match, a pulse is output and the frequency divider circuit 2-3 is reset. The frequency of the pulses output at the time of this coincidence corresponds to each scale in each octave. By the way, if the frequency division ratio is 12 values between 478 and 253 (expressed by 9 bits in binary code) with respect to the reference frequency of 500 KHz, each scale frequency in the range of 1046 to 1975 Hz can be obtained. The octave can be changed by doubling the reference frequency. 5. The coincidence detection pulse is further input to the frequency dividing circuit 2-7. The frequency dividing circuit 2-7 uses its 4-bit binary frequency divided output to generate the next stage basic waveform memory 2-8.
Specify the address. That is, the frequency dividing circuit 2-7 operates as a so-called address counter. Basic waveform memory 2-8 consists of 8 bits and has 16
Data is stored so that each step forms a waveform corresponding to one period of the scale frequency. That is, octave selector 2-2, matching circuit 2-
The frequency output from 4 is the actual scale frequency.
This corresponds to 16 times, and one cycle of each scale frequency is divided into 16 by the frequency dividing circuit 2-7, and the addresses of 16 steps in the basic waveform memory 2-8 are sequentially designated. The waveform data read out by this addressing is input to the D/A converter 2-9, where it is D/A converted and
A basic waveform of the sound signal is formed every cycle. Further, the basic waveform memory 2-8 receives 2 bits B ₁₀ and B ₁₁ from the music information memory 7 and the upper 2 bits binary frequency divided output C of the frequency dividing circuit 3-7 of the envelope generating section 3. _B10 ,
_{The 11} bits in B select the waveform memory to be read out in accordance with instrument recombination control, and the 2 bits in C further divide the envelope into predetermined regions in time, and select the basic waveform to which higher-order frequencies are added as appropriate. This is to increase the naturalness and ease of listening to the sound. 6. Of the 8 bits transmitted to the envelope generator 3, 4 bits _B4 to _B7 are input to the note length division ratio memory 3-6 as address designations. The two bits _B10 and _B11 are used as control bits for instrument recombination and are stored in the envelope waveform memory 3-8.
Furthermore, the two bits B ₁₂ and B ₁₃ are input to the arithmetic circuit 3-9 as sound intensity control bits. 7 In the envelope generating section 3, when the start switch is turned on, the frequency dividing circuit 3-1 starts operating first. The 6-bit binary frequency division output of the frequency division circuit 3-1 is input to the matching circuit 3-2, and is compared with the 6-bit frequency division ratio output of the tempo frequency division ratio memory 3-3. If they match, a pulse is generated and input to the subsequent frequency divider circuit 3-4. This pulse determines the time interval of the shortest note length. If necessary, an addition/subtraction circuit (not shown) may be provided.
By operating the tempo control switch of the input circuit 4, the frequency division ratio output of the frequency division ratio memory 3-3 is appropriately added or subtracted, and the frequency division ratio data can be changed to set an arbitrary tempo. 8 Note length division ratio memory 3-4 is 4 of B ₄ to _{B 7}
Using bits as addresses, 8-bit frequency division ratio data corresponding to each note length is selected and output. Correspondingly, the match circuit 3-5 compares it with the 8-bit binary frequency division output of the frequency divider circuit 3-4, and outputs a pulse when they match. The output time interval of this pulse corresponds to each note length specified by 4 bits _B4 to _B7 . However, here again, due to the following reasons, 1
It outputs 32 pulses per note length. That is, the coincidence detection circuits 3-2, 3-5
The frequency of the pulses output from etc. is usually 32 times that of the general case. This pulse is transmitted to the frequency dividing circuit 3-
7 and is frequency-divided. 9. The envelope waveform memory 3-8 has an 8-bit configuration and stores data to form one envelope waveform in 32 steps. Like the basic waveform described above, the envelope waveform is only compressed and expanded in time for each note length, and one envelope waveform must be read out for one note length. The frequency dividing circuit 3-7 serves as a so-called address counter for the envelope waveform memory 3-8, and the envelope waveform memory 3-8 is divided into 5-bit binary frequency divided outputs per note length.
Specify the addresses of 32 steps of 8 in sequence. The B ₁₀ and B ₁₁ bits input to the envelope waveform memory 3-8 select the waveform memory to be read out, and are combined with the basic waveform selected in the basic waveform memory 2-8 to perform musical instrument recombination control. The 2 bits of C are the high-order ones of the 5-bit binary frequency division output of the frequency divider circuit 3-7, and divide the envelope period into four equal parts. 10 The data read from the envelope waveform memory 3-8 is processed in the arithmetic circuit 3-9.
Multiplication based on the strength bit data of B ₁₂ and B ₁₃ and addition/subtraction based on the output of the tremolo control circuit 3-11 are performed to modify the envelope waveform data. The modified data is D/A converted by a D/A converter 3-10 to generate an envelope. 11 The envelope is sent as a level control signal to the D/A converter 2-9 of the tone generator section 2, and the D/A converter 2-9 multiplies it by the basic waveform and outputs an enveloped sound signal.
The sound signal is emitted via an amplifier 9 and a speaker 10. 12 When the frequency divider circuit 3-7 counts 32 steps (one envelope read), the carry pulse is input to the address counter 6 and advances the address by one. As a result, the next music information bit pattern is read out from the music information memory 7, and the operations 1) to 12) above are repeated. 13 In this way, the song information bit patterns are sequentially read out from the song information memory 7, and when "1" is output to the _B15 bit, a stop signal is output from the control circuit 5, and each frequency dividing circuit 2-
1, 2-3, 2-7, 3-1, 3-4, 3-7
At the same time, the internal gate circuit is closed and the series of operations ends. <Data Capacity of Memory> Incidentally, the data capacity of each memory in the above embodiment is as follows. Scale division ratio memory 2-5...9 bits x 12 scale basic waveform memory 2-8...8 bits x 16 steps x 4 basic waveforms (unless controlled by C signal) Note length frequency division ratio memory 3-3... 8 bits x 1 shortest note length note length frequency division ratio memory 3-6...8 bits x 16 notes length envelope waveform memory 3-8...8 bits x 32 steps x 4 envelope waveforms <basic waveform and envelope> As the basic waveform, for example, As shown in the time chart in Figure 3, there are (a) sine waves, (b) sawtooth waves, (c) rectangular waves, and (d) triangular waves. Basic waveform memory 2-
8, one cycle of these waveforms A is divided into 16 parts and stored as 8-bit, 16-step digital data. The time charts in FIGS. 1 to 4 show examples of envelope B waveforms. The envelope is divided into 32 parts and stored in envelope waveform memory 3-
8 is stored as 8-bit, 32-step digital data. The basic waveform and envelope waveform described above are just examples, and there are various other basic waveforms and envelope waveforms, and the present invention is not limited to these. If the sound of a certain instrument is as shown in Figure 5,
The basic waveform A is (a) in Figure 3, and the envelope B is the 4th waveform.
It consists of 2 in the figure. The frequency of the fundamental waveform corresponds to each scale, and the length of the envelope corresponds to each note length. Data specifying instrument sounds is stored in song information memory 7.
These are the instrument recombination bits B ₁₀ and B ₁₁ outputted from the instrument, and here the instrument sound can be switched for each note at most. Instrument recombination bits
The data of B ₁₀ and B ₁₁ are stored in basic waveform memory 2-8,
The signals are input to the envelope waveform memory 3-8, and the corresponding fundamental waveform and envelope waveform are selected. In this way, in this embodiment, the envelope waveform and the basic waveform are combined to reproduce a song with a desired musical instrument sound (or timbre). <Scale data, note length data> A chord diagram of the scale data (B≓ to B ₃ bits) is shown in FIG. 6, and a chord diagram of the note length data (B ₄ to B ₇ bits) is shown in FIG. Here, as shown in the figure, scale data and note length data are encoded in sequential correspondence with 4-bit binary codes. The scale data is “0000” (code OH)
indicates a rest, and the rest length is set using note length data. When we consider the above-mentioned scale transposition and transposition using the above code, it becomes addition and subtraction of a certain number of binary codes. For example, moving up a semitone from C major changes it to D flat major. The first octave of C major is “0100”
(Code 4H) to “1111” (Code FH), and the first octave of D-flat major is “0101” (Code FH).
5H) ~ “0100” one octave higher (code 4H)
So, it becomes the C major chord plus "0001". Also, if you move down a semitone from C major, you get B major. The first octave of B major at this time is "1111" (code FH) one octave below, "0101" (code 4H) to "1110" (code FH) in the same octave.
EH), which is C major minus "0001". The upper and lower octaves are represented by carries and borrows from "0100" (code 4H) to "1111" (code FH) using the above addition and subtraction. Addition/subtraction circuit 2-6 of tone generator section 2
executes this, and in this example, the scale from Do to Shi, including the chromatic part, is set to “0100” (code 4H) to
Since "1111" (code FH) is made to correspond to the binary code in sequence, transposition and transposition can be performed using a simple addition/subtraction circuit. The tempo can also be changed by adding or subtracting an arbitrary number (binary code) to the output of the 6-bit frequency division ratio memory 3-3 using a similar addition/subtraction circuit (not shown) to change the shortest note length. Also, as will be described in detail later, by sequentially associating note length data with such a binary code, it is possible to detect a predetermined bit, for example, the data of the upper two bits of B ₆ and B ₇ (which must be "1"). You can easily create whole notes (〓), dotted half notes (〓.),
Only long notes such as half notes (〓) can be identified. <Naturalness and audibility of the sound> The 2-bit C signal input to the basic waveform memory 2-8 increases the naturalness and audibility of the sound. In general, as shown in the typical envelope waveform in Figure 4-1, the time from the rise of the sound to the peak: attack time A, the time from the peak to the hold level: decay time D, hold level. There is a sustain time S, a falling time, and a release time R, and the basic waveform can change in line density during these periods. In this embodiment, the envelope period is equally divided into four by a 2-bit C signal and approximated to this. The basic waveform can be changed in various ways, such as slight changes by adding higher-order frequencies, or a certain period of a certain musical instrument sound having a completely different basic waveform. In this case, since it is specified by the instrument recombination bits B ₁₀ and B ₁₁ , and it is necessary to select the basic waveform in each of the four periods of the envelope, a 4 x 4 basic waveform is stored in memory, and the data in basic waveform memory 2-8 is The capacity is 8 bits x 16 steps x (4 x 4) basic waveform. <Sound strength and tremolo effect> The strength of the sound is determined by the magnitude of the amplitude, so here we will adjust the level of the entire envelope, and since the tremolo effect is so-called low frequency amplitude modulation, we will adjust the envelope waveform. This can be achieved by amplitude modulation at a low frequency. FIG. 8 shows the calculation circuit 3-9 of the envelope section 3,
Details of the tremolo control circuit 3-11 and its peripheral circuits are shown. The tremolo effect is applied by detecting the B ₆ and B ₇ bits of the note length data. That is, the B ₆ and B ₇ bits of the note length data are input to the tremolo control circuit 3-11, and the AND gate A detects only long notes such as whole notes, dotted half notes, and half notes. Although it is possible to apply the tremolo anywhere or to any part of the envelope, it is more appropriate to apply the tremolo effect to the latter half of the envelope in order to increase the effect even further. For this purpose, in this example, the envelope waveform memory 3
A signal _t1 is taken out from the most significant bit output _O5 of the frequency divider circuit 3-7, which specifies the address of -8, and is applied to the AND gate A. The low frequency (Hz) signal _T determines the period of amplitude modulation. The song information is read from the song information memory 7, and by detecting the B ₆ and B ₇ bits of the note length data, it is determined that the note is a long note (B ₆ and B ₇ bits "1"), and then a signal is sent when the note is in the latter half of the envelope. _t1 also becomes "1", and the low frequency signal _T is output from the AND gate A. This low frequency signal _T is input to the addition/subtraction circuit 3-9-1 of the arithmetic circuit 3-9, and the low frequency signal T
Based on _{the T} "1" and "0" signals, a predetermined number is added or subtracted from the 8-bit binary code representing the envelope waveform. That is, this causes the digital quantity of the envelope waveform to change at a low frequency of Hz. Assuming that the next stage shift circuit 3-9-2 is not affected, the further downstream D/A converter 3-10
is input to the lower half of the envelope as a low frequency
It will output an analog signal amplitude modulated at Hz. The strength of the sound is determined by the bits specified by the strength control bits _B11 and _B12 in the shift circuit 3-9-2.
This is done by appropriately shifting the output of the addition/subtraction circuit 3-9-1. for example,

[Table] According to the B ₁₂ and B ₁₃ bit data, each of the 32-step binary codes supplied from is sequentially shifted by a specified number of bits, and the digital amount of the entire envelope is changed by a predetermined amount. The shifted binary output is converted into an analog signal by the D/A converter 3-10, but it is doubled for a 1-bit shift, quadrupled for a 2-bit shift, and
With a 3-bit shift, an analog signal with eight times the amplitude is output. That is, the intensity of the sound is controlled by changing the amplitude by a factor of 1, 2, 4, or 8 using B ₁₂ and B ₁₃ bit data. As described above, the present invention effectively produces a tremolo effect without using dedicated information bits for tremolo, reduces the amount of data in the song information memory, and can be achieved with slight hardware improvements. A useful music playing device with excellent acoustic effects can be provided.

[Brief explanation of the drawing]

FIG. 1 is an overall circuit block diagram showing one embodiment of the present invention, FIG. 2 is a diagram showing the allocation of song information bits, FIGS. 3 a to d are time charts showing examples of basic waveforms, and FIG. 4 ~4 is a time chart showing an example of an envelope waveform, Figure 5 is a time chart showing an example of mixing the basic waveform and envelope waveform, Figure 6 is a diagram showing an example of chords corresponding to each scale, and Figure 7 is a time chart showing an example of mixing the basic waveform and envelope waveform.
The figure shows chord examples corresponding to each note length.
This figure is a circuit block diagram showing the main part of FIG. 1 in more detail. DESCRIPTION OF SYMBOLS 1...Oscillation/frequency division circuit, 2...Tone/generator section, 3...Envelope generation section, 6...Address counter, 7...Song information memory, 3-9...Arithmetic circuit, 3-10...D/A converter, 3- 11...Tremolo control circuit, 3-9-1...Addition/subtraction circuit, 3-
9-2...Shift circuit, B〓~ _B3 ...Scale bit, _B4
~B ₇ …Note length bit.

Claims (1)

[Scope of Claims] 1. In an apparatus that stores music information in which at least scale data and note length data are used as bit patterns, and plays a music according to reading of the music information, long notes are extracted from the bit patterns of the note length data. A music playing device comprising: a detecting means; and a tremolo effect generating means for generating a tremolo effect in reproduced sound using the detection output.