JPH0310050B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a musical note display device, and an object of the present invention is to provide a musical note display device that can accurately display the pitch and length of an input note on a musical staff on a display surface.

The applicant previously filed Japanese Patent Application No. 57-232457 (Japanese Unexamined Patent Publication No.
59-124384) and Japanese Patent Application No. 58-3670 (Japanese Unexamined Patent Publication No. 59-128576) "Musical note display device" converts the input audio signal into a digital signal, analyzes the frequency of this, and determines the pitch. Determine the pitch of the sound and display it on the CRT
proposed a device that displays music notation in real time.
This method performs an FFT operation on audio signal data converted to a digital signal, then calculates the result into a power spectrum to obtain a spectral value, then calculates the maximum value among these spectral values, and calculates the maximum value. Find the frequency value of the spectrum that has the lowest frequency among the spectra that have a frequency lower than the spectral frequency that indicates the frequency and that has a smaller spectrum within a predetermined range than the maximum spectrum, and use it as the pitch of the input audio signal. judge,
On the other hand, when there is no spectrum that satisfies the above conditions, the frequency value indicating the maximum spectrum is determined to be the pitch of the input audio signal, and this determination result is processed by a video display processor (hereinafter referred to as
Video RAM (hereinafter referred to as VDP)
(VRAM) and display it on a CRT.

However, this method only displays the pitch of the input audio signal, but cannot display the length of the sound. Therefore, it would be easier to see and more convenient if both the pitch and length of the input audio signal were analyzed and displayed as actual notes on the musical staff in real time.

The present invention satisfies the above requirements, and one embodiment thereof will be described below with reference to the drawings.

FIG. 1 shows a block system diagram of an embodiment of a musical note display device according to the present invention. In the same figure, terminal 1
The incoming input audio signal is supplied to a graphic equalizer 2, where it is varied to have a frequency characteristic that is easy to analyze, and then passed through an anti-aliasing filter 3.
The signals are supplied together to the AD converter 4 after removing unnecessary high frequency components so as not to cause aliasing noise. Initial settings (system initialization) are performed in the control device 5 (step a in FIG. 2A), and each operation of the steps shown in FIG. 2A is executed as long as no interrupt occurs.

When an interrupt occurs every sampling period of the AD converter 4 (step S in FIG. 2B), the operation shown in FIG.
AD conversion is performed in AD converter 4 according to the command from (step t), and the digital signal is
Stored in RAM7. At step u, it is determined whether or not the AD conversion operation has been performed a predetermined number of times. If NO, the process returns. On the other hand, if YES, the process presets the AD and returns (step v).

In this way, a predetermined number of digital data is
Once stored in RAM 7, calculation and pitch analysis section 8
FFT calculation is performed at (step b)
The calculation result is stored in the RAM 7, and then a power spectrum calculation is performed (step c) and the calculation result is stored in the RAM 7. The maximum spectral value is then determined, and then a spectral value that exists in a frequency region lower than the frequency of the spectrum exhibiting the maximum spectral value and is smaller within a predetermined range than the spectral size of the maximum spectral value is determined. The frequency value of the spectrum that shows the lowest frequency value among the spectra that it has is determined to be the pitch (fundamental tone) of the input audio signal, and if there is no spectrum that meets the above conditions, The frequency value of the spectrum indicating the maximum spectrum value is determined to be the pitch (fundamental tone) of the input audio signal and is stored in the RAM 7.

The time required for the operations of steps b to d above is the third
1/2 of the time corresponding to the eighth note display shown in Figure A
The fourth step is set in advance in steps b to d.
This means that the pitch of the input audio signal at time _t1 shown in the figure has been analyzed.

For example, if we assume that the sound of a quarter note C shown in pattern 15 in Figure 5 is being input by an instrument such as a piano whose sound decays over time, the pitch will be changed according to steps b to d. Following the analysis, the same operations as in steps b-d are performed in steps e-g (time t ₂ shown in FIG. 4).
Next, the number of times the pitch analysis was performed in step g is counted (step h), and then the continuous sound detecting section 9 determines whether or not the sound is continuous (step i). Here, the results of the previous pitch analysis and the results of the current pitch analysis are compared to determine whether the pitches of the two are the same or not, and the level difference between the two is determined within a preset level difference range. At the same time, the results of the current pitch analysis are determined at the preset threshold level L (Figure 4).
It is determined whether or not the above is true. Here, the length of the shortest note to be displayed is divided into two and the frequency is analyzed in each section, so even if the input sound is a continuous sound that changes slightly over time, it is possible to check whether the input sound is within a predetermined frequency difference range. The continuity and pitch can be detected with high accuracy.

In the case of a quarter note, the continuous note detection unit 9 detects it as a continuous note, and the analysis number detection and pattern data command unit 10 detects whether the count number at step h has reached 16 (step j ). Since the count number is 1 at the stage up to time t ₂ ,
FFT calculation is performed again at step e, pitch analysis is performed at step f, and count-up is performed at step h.

Thereafter, the same operation as above is carried out until time t ₄ (the count number is 3), and then at time t ₅ , the E sound of the eighth note shown in Figure 5 is input. Suppose that At time _t5 , pitch analysis is performed in steps e to g, and the count number is set to 4 in step h.

Next, in step i, since the difference in pitch and level is different from the previous analysis (quarter note C note), it is determined that the sound is not a continuous sound, and the pattern data command signal is extracted from the detection and command section 10 and the pattern is The data is supplied to the data determining section 11. In the pattern data determination unit 11, the analysis unit 8 determines the time.
Pattern data is created from the pitches analyzed up to _t4 and the number of analyzes detected by the detection unit 10 (count number 4) (step K). Here, the relationship between the count number and the note length is as shown in Figure 3 A to P, where the count number ranges from 1 to 16 from an eighth note to a whole note.
are associated with each. In this case, since the count number is 4 and the pitch is C, pattern data corresponding to the C note of a quarter note is extracted,
It is supplied to the VRAM 13 via the VDP 12 and written into a predetermined table.

As shown in FIG. 6, the memory map of the V-RAM 13 includes a sprite generator table SGT and a pattern generator table PGT that store a plurality of types of patterns, and a display screen (32 columns x 24 rows) on the CRT 14 shown in FIG. ) pattern name table PNT, which has a capacity corresponding to each section of
It is composed of a sprite attribute table SAT that stores information on the vertical and horizontal positions of the display surface, a color table CT that stores color information, and unused tables. VDP1
In 2, the pattern name table of VRAM13
In accordance with the data written in the PNT and the sprite attribute table SAT, a pattern to be displayed on the display screen of the CRT 14 is selected and the movement mode of the pattern is specified, and a composite video signal is created.
The signal is supplied to the CRT 14, and pattern 15 (quarter note C note) is displayed on the display screen here (step 1). Note that the staff score, clef, etc. are displayed in advance based on data stored in the RAM 7 in advance.

When pattern 15 is displayed, it is determined in step m whether the count number is an odd number or not. In this case, since the count number is 4, the count is cleared in step n, and the count number at time _t5 is cleared to zero. be done. At the next time _t6 , as in the case above, steps e to g perform pitch analysis, step h counts up and sets the count number to 1, then step i determines that it is a continuous tone, and step i determines the count number. It is determined that the number has not reached 16, and the process returns to step e.

When an eighth rest occurs at times _t7 and _t8 , it is determined at step i that the pitch analysis result is below the threshold level L, and the count number 1 at time _t6 and time _t6
Based on the pitch analysis results, the above pattern 15
Pattern data is created in the same way as CRT14
A pattern 16 is displayed at a predetermined position on the display screen. Similarly, when the next eighth note treble in FIG. 5 is input, eighth rest pattern 17 is displayed.

Here, in FIG. 5, when the A note of an eighth note is inputted several times independently from time _t1 , at step i, the level difference between time _t2 and time _t3 , and the level difference between time t4 and time _t4 . Since the level difference between the sound and the time _t5 is outside the preset level difference range, it is determined each time that the sound is not a continuous sound. Patterns 18, 19, . . . are displayed based on the count number 2 detected each time and the pitch A (FIG. 3B). next,
For example, when using an instrument that can sustain the sound produced, such as a bowed string instrument, the mouth sounds of whole notes in Figure 5 are
When input from _t1 , the count number is set to 15 at step h from time _t1 to time _t16 , continuous sound is determined at step i, and discontinuous sound is determined at the next time _t17 . Pattern 20 is displayed with the count number 15 and the high-pitched mouth sound (FIG. 3 O).

Next, in Figure 5, when a mouth sound consisting of a whole note and a dotted quarter note is input from time _t1 , the count number is 16 at time _t17 , and a continuous sound is input at step i. be judged. It is determined in step j that the count number has reached 16, and in step O, pattern data is created from the count number 16 and the high-pitched mouth sound (Fig. 3 P), and in step p, it is converted into a composite video signal and sent to the CRT 14. Pattern 21 is displayed. When this display ends, the count is cleared in step q, and the FFT calculation is performed again in step e.
is displayed.

On the other hand, time t ₁ to 4 as in Statzcato performance.
If a diacritic note is input but interrupted at time _t4 , the count at this point is 3, and a discontinuous note is determined at step i. At time _t4 , a pattern of quarter notes is displayed, at step m it is determined that the count is an odd number, and at step r the count is cleared in preparation for analysis of the incoming note at time _t5 .

As described above, the musical note display device according to the present invention detects the difference in pitch and level between two tones analyzed before and after the time of the shortest note of the notes to be displayed in the frequency analysis means. It is determined whether the sound is a continuous sound or not, and if it is determined that the sound is a continuous sound, the frequency of the input signal is analyzed again in the latter half of the frequency analysis means to determine the pitch,
If a discontinuous sound is determined, data corresponding to the pitch and length of the sound before the discontinuous sound determination is obtained from the pitch analysis results and the number of frequency analyzes of the sound before the discontinuous sound determination, and this data is used. Based on this, notes based on the pitch and length of the notes are displayed at the corresponding positions on the staff on the display diagram, so that the pitch and length of continuous sounds that vary slightly over time can be displayed on the display screen. In addition, unlike devices that print the notes of the input sound on paper, etc., the musical score of the music being played can be displayed immediately, so it is possible to display the musical score of the music being played with high accuracy. It has features such as being able to clearly define the relationship between music and music, and being convenient for music education.

[Brief explanation of the drawing]

FIG. 1 is a block system diagram of an embodiment of the device of the present invention, FIGS. 2A and B are flowcharts for explaining the operation of the device of the present invention, and FIGS. 3A to 3P show the relationship between note length and count number. Figure 4 is a diagram showing the relationship between sound level and musical notes, Figure 5 is a diagram of the CRT display screen,
Figure 6 is the memory map of VRAM, Figure 7 is
FIG. 3 is an explanatory diagram of sections on a CRT display screen. 1...Audio signal input terminal, 4...AD converter,
5...Control device, 6...Conversion command and digital signal output section, 7...RAM, 8...Calculation and pitch analysis section, 9...Continuous sound detection section, 10...Analysis circuit detection and pattern data detection Section, 11...Pattern data determination section, 12...Video display
Processor, 13... V RAM, 14...
CRT.

Claims (1)

[Claims] 1. AD that converts an input audio signal into a digital signal
a conversion means; a frequency analysis means for frequency-analyzing the signal from the AD conversion means twice before and after each other within a time corresponding to the length of the shortest note among the notes to be displayed to determine the pitch; and the frequency analysis frequency analysis frequency detection means for detecting the number of times the frequency has been analyzed by the means;
continuous sound determination means for detecting a difference in pitch and level between two sounds analyzed before and after by the frequency analysis means to determine whether or not the sounds before and after are continuous sounds; and the continuous sound. If the determining means determines that the sound is continuous, the frequency analysis means analyzes the frequency of the input signal again in the second half of the frequency analyzing means to determine the pitch, and the continuous sound determining means determines the pitch of the continuous sound. If a sound is determined, a data output means for obtaining data corresponding to the pitch and length of the sound before the discontinuous sound determination from the pitch analysis result of the sound before the discontinuous sound determination and the frequency analysis number; A musical note display device comprising a display means for displaying a musical note according to pitch and length on a display surface at a corresponding position on a musical staff based on data from the data output means. 2. The display means converts the data from the data output means into predetermined pattern information, converts the pattern information into a composite video signal using a video display processor, and displays the signal on a CRT. Claim 1 characterized in that
Musical note display device as described in section.