KR950012086B1 - Midi data compressing and resting method - Google Patents

Abstract

The method comprises the steps of detecting tone and rhythm data from MIDI data and setting a maximum record limit value to compress the tone and rhythm data; forming data of a compressure tone and rhythm up to the maximum record limit value; coding the tone data as a value of the compressure tone data table, coding the rhythm as a value of a second data table, and converting the coded rhythm data into data of 1 byte, if the tone data is positioned in the compressure tone data table; and coding the tone data as a value except the maximum record limit value, coding the rhythm as a value of the second data table, converting the coded rhythm data into data of 1 byte, and storing the original tone data in a following data of 1 byte, if the tone data is not positioned in the compressure tone data table;

Description

MIDI data compression and playback

1 is an exemplary embodiment showing a configuration of a flexible accompaniment system to which the compression and reproducing method of the present invention is applied.

2A to 2C of FIG. 2 show a structure of MIDI data and a structure of compressed MIDI data for explaining the compression and reproduction method of the present invention.

3 is a signal flow diagram showing a compression method of the present invention.

4 is a signal flow diagram showing a reproduction method of the present invention.

* Explanation of symbols for main parts of the drawings

100: microprocessor 102: system RAM

104: system ROM 106: key / remote control interface

108: Arrangement data ROM 109: FM or PCM type sound source

110: lyrics super impose circuit 114: audio mixer

117: audio output terminal 119: video output terminal

301: Arrangement data memory 302: Modem

303: public telecommunications network

The present invention relates to a method for processing MIDI data in an image flexible system. In particular, a compressed MIDI data is compressed, stored, and converted into original data to store a large amount of MIDI data in a small amount of arranged data memory. The present invention relates to a method of compressing and playing back MIDI data.

The video accompaniment system is a system that is widely used in a place recently called "karaoke room".

This video accompaniment system stores video and song data in memory, and reads the corresponding video and song data from memory according to the user's choice to sing a song. Will output

Therefore, the users of the video accompaniment system can enjoy the song along with the accompaniment using the microphone while watching the video and lyrics displayed on the monitor.

Karaoke is a place where everyone sings and sings without any distinction. In order to satisfy all the needs of users, the karaoke should store accompaniment songs of all the songs that users select to sing.

In order to meet the needs of users, the storage device must store a variety of songs that can be enjoyed and enjoyed by everyone from children to adults as well as from old songs to recent songs. The songs sung in) should also be stored in order to achieve the ideal visual accompaniment system.

On the other hand, a flexible accompaniment system is a system that achieves the same effect as an image flexible accompaniment system without performing image processing.

This flexible accompaniment system has the same configuration as that in FIG.

When power is applied to the system, the microprocessor 100 executes an initialization operation and returns to an initial state.

In the initialization operation, the microprocessor 100 reads the entire system control program data from the system ROM 104 and writes it to the system RAM 102.

In the method in which the microprocessor 100 reads system control program data from the system ROM 104, the microprocessor 100 generates a control signal through the control bus CB to read the system ROM 104. Enable mode, generate an address signal via the address bus AB, specify the read range of the data, read the data through the data bus DB, and move to the system RAM 102 for recording.

The method of recording the system control program data read by the microprocessor 100 into the system ROM 104 to the system RAM 102 may generate a control signal through the control bus CB to generate the system RAM 102. Enable, specify the write mode, designate the area to be written through the address bus AB, and then read the system control program data for system control read from the system ROM 104 through the data bus DB. In the area of).

The key / remote control unit 106, the arrangement data ROM 108, and the FM or PCM system sound source unit 109 are initialized.

In this state, when the user inputs a song selection number for designating a song to be sung by operating the key input unit (not shown in the drawing) or the remote controller, the song selection number is inputted to the key / remote control interface unit 106. The microprocessor 100 reads the data, writes it to the system RAM 102, and processes the program according to the program.

The microprocessor 100 generates a control signal through the control bus CB according to the song selection number to enable the arrangement data ROM 108 in the read mode, and stores the corresponding song data through the address bus AB. An address signal for designating a designated area is applied to the generation and arrangement data ROM 108 to read song data, and is then input to the FM or PCM type sound source unit 109 in the same manner as described above.

At this time, the microprocessor 100 inputs control data to the FM or PCM type sound source unit 109 via a data bus DB to generate an accompaniment signal according to the input data.

Therefore, the generated accompaniment signal is mixed with the user's voice signal input through the microphone stage in the audio mixer 114 and output to the speaker through the audio output terminal 117.

In addition, since the audio signal is output through the audio mixer 114 and the video signal is output to the video output terminal 118 through the lyrics superimpose circuit 110, the user views the output image and sings according to the accompaniment. It is supposed to be enjoyed by calling.

The lyrics superimpose circuit 110 also processes the input video signal of the video input terminal 118 under the control of the microprocessor 100, outputs the video signal to the video output terminal 118, and displays the same on the monitor.

In order to add new song data to the arrangement data ROM 108 that stores the arrangement data, such a flexible accompaniment system needs to use a large-capacity memory chip that can store all the new song data of previously stored song data. Additional work is required from time to time to save the data.

The additional work described above cannot be done in the light of being able to store tens or hundreds of songs in one arrangement data ROM 108.

This is because, when new songs are added to the arrangement data ROM 108, already arranged songs are out of fashion.

In order to solve this problem, the arrangement data ROM 108 uses a memory device such as EPROM, which can be freely written and erased, and the MIDI of a new song transmitted through a modem 302 for computer music according to a predetermined serial communication protocol. There has been a way to easily add new songs that are popular with the data.

However, since new songs are released a lot and the amount is very large, even if a memory with a large capacity is used, there is no fundamental solution such as the need to add new memory to store new song data.

Accordingly, an object of the present invention is to provide a method of compressing and reproducing a MIDI data that compresses flexible accompaniment data, i.e., MIDI data, to increase storage capacity within a limited memory capacity, and does not affect existing MIDI data even when the data is reproduced. There is.

To achieve this object, the MIDI data compression method of the present invention detects both pitch and negative intensity data in the MIDI data, sets the maximum recording boundary to compress the pitch and negative intensity data, and detects the detected pitch and tone. Compressed pitch data is stored in the compressed pitch data table up to the set maximum record threshold. If the read pitch data is in the compressed pitch data table, the pitch data is encoded as the table value of the compressed pitch data. The negative strength is encoded into the table value of the second data and converted into one byte, and if the pitch data is not in the compressed first data table, the first data is encoded into a value other than the maximum recording boundary value and the second data is stored. Encode into table value of 1 to convert into 1 byte and store in 1 byte It is characterized by storing the original pitch data.

The MIDI data reproducing method of the present invention reads the table values of the pitch data and the negative intensity data, reads the stored status byte, the code value of the pitch data, and the code value of the negative intensity data, and records the maximum code value of the pitch data. If it is within the boundary value, the data corresponding to the code value in the table of the pitch data is the original pitch data, and if it is above the maximum recording boundary value, the following 1 byte value is the pitch data, and the negative intensity is the code value. The corresponding data is read from a negative strength and weakness data table and output.

Hereinafter, the compression and reproducing method of the present invention will be described in detail with reference to the accompanying drawings of FIGS. 2 to 4.

2 shows the form of MIDI data compressed in accordance with the present invention.

Here, Figure 2a is a diagram showing the configuration of the MIDI data that is generally input or stored via the modem (302).

The MIDI data consists of a status byte 201, first data 202 indicating pitch, and second data 203 indicating negative intensity.

The status byte 201 is generated only when there is an input change, for example, when the accompaniment song output is turned on and off, and when there is a change in the output musical instrument. In this case, the first data 202 and the second data 203 are generated continuously.

Each of these data 201, 202, and 203 consists of 8 bits each, and the status byte 201 always has the most significant bit (MSB (Most Significant Bit)), and the first and second data (202, 203) have the most significant bit. The status byte 201 and the first and second data 202 and 203 are divided into most significant bits MSB with (MSB) always "0".

The first data 202 and the second data 203 represent pitch and negative intensity steps with values from '0' to '127' as sixth to zero bits, and the first and second data for the fidelity of the sound. It is sometimes necessary to have a value of (202, 203) from '0' to 127, but two data bytes can be compressed into one data byte at the expense of some fidelity.

Such MIDI data is compressed in the present invention as shown in FIG. 2B.

That is, the MSB (B7) always indicates that it is a data byte as '0', and the 4-bit code value of the 6th to 3rd bits (B6 to B3) indicates the pitch of 15 steps, which indicates the pitch used in each track. Used as an index into a table that has data.

For example, if the value of the fifth to third bits (B5 to B3) is '0000', it corresponds to the pitch data of the zeroth item in the pitch table, and the value of the sixth to third bits (B6-B3) is '0001'. 'Corresponds to the pitch data of the first item in the pitch table. Thus, the pitch data of the 0th to 14th items of the pitch table is set according to the values' 0000' to '1110' of the 6th to 3rd bits (B6-B3). Yes.

The pitch table is created by classifying the pitches used in the corresponding tracks of the MIDI data in advance, which will be described in detail in the compression method described below.

Three bits of the second to zero bits B2 to B0 represent negative strengths of eight levels.

For example, if the value of the original second data 203 is from 0 _H (where 'H' represents a hexadecimal number) to 19 _H , then 1, 40 _H to 59 _H days from 0.20 _H to 39 _H 2, 60 _H to 79 _H 3, 80 _H to 99H 4, 100 to 119 5, 120 _H to 122 _H 5, 122 _H to 124H 7 To have a value.

Here, the negative intensity is 3 bits, and when output is divided into 8 stages as described above, the negative intensity is slightly lower than the resolution of the negative intensity of the original second data 203, but there is no practical problem. Does not occur.

However, the pitch is 4 bits, and when the output is divided into 15 stages as described above, the original sound cannot be reproduced and the user is offended.

Therefore, in the present invention, the sixth to third bits (B5-B3) represent the pitch in 15 steps from '0000' to '1110', and the pitches of the more than 15 steps are the sixth to third bits (B5-B3) as shown in FIG. The values were all coded as '1111' and the original next data was accepted in the next one byte of data.

Therefore, according to the present invention, the compression of the first data 202 and the first data 203 of the MIDI data is variable depending on the size of the pitch data, and when the value of the pitch data is within the range of 15 steps. If it is compressed to 1 byte and goes out of step 15, it is compressed to 2 bytes so that the amount of MIDI data can be reduced to about 1/3.

3 is a signal flow diagram illustrating a compression method of the present invention. As shown in the drawing, according to the present invention, the microprocessor 100 reads and processes predetermined MIDI data transmitted through the modem 302 and stored in the system RAM 102.

In step 3a, the microprocessor 100 is capable of processing data stored in the system RAM 102 as shown in FIG. 2a, which is the first and second data 202 and 203 in one track. Initialize each flag to store the state of the pitch and negative strength to the maximum, and the table number of the pitch to be stored and the table number of the negative strength and weakness when compressed.

Step (3b), we determine that the value of the status byte 201 in reading the status byte 201, the step (3d) 80 _H or 90 _H.

That is, the status byte 201 in the MIDI data is off the output of the audio signal if the value is 80 _H day, and 90 _H are ordered on the audio signal output, the value if is not the 80 _H or 90 _H As specified in advance as a command for specifying the sound of the musical instrument, it is determined in step 3d whether the value of the status byte 201 is an on or off command of the audio signal output.

If the value of the status byte 201 is not _80H or _{90H and} is not an on or off command of the audio signal output, the next data is not pitch data and negative strength and weakness data, and then in step 3b. Reads the status byte of.

In the step (3d) Step (3e) in the case of turning on or off command of the audio signal output as a 80 _H or 90 _H value of status byte 201. In reading the first data 202 and second data 203, Set the appropriate flag.

That is, the first data 202 and the second data 203 are read, and if there is a value, the corresponding flag of the first data 202 and the second data 203 is set to "1".

Next, in step 3f, it is determined whether the end of the current track is the value of the status byte 201.

In other words, when the end of one track in the MIDI data, the value of the status byte 201 is FF _H, and the value of the consecutive data byte is defined to be set to 2F _{H. In the} step 3e, the status byte 201 is specified. Determines if the value of FF _H is the end of the track, and the value of the next data byte is 2F _H.

If it is not the end of the track in step 3f, the operation of reading the status byte 201 is repeatedly performed in step 3b. If it is the end of the track, the first data 202 and the first data are made in step 3i. An index for initializing a counter for counting two data 203, initializing a maximum recording boundary value of the compressed first data 202, and checking a flag of the first data 202 and the second data 203. (i) is reset.

In this state, in step 3j, the value of the i-th flag of the first data 202 is "1", and the value of the counter of the second data 202 that counts it is smaller than 15, which is the set maximum recording boundary value. In step 3k, it is determined whether the i-th flag of the second data is "1".

If the value of the flag in step 3j is " 1 " and the counter value is less than 15, in step 3g, the number of the index i is stored as the maximum recording boundary value of the compressed first data 202. The data value of the i-th flag is designated as the count value of the first data 202 of the table of the first data 202, and the first data 202 count value is assigned to the i-th flag of the first data 202. After storing, add 1 to the first data count value.

If the i-th flag of the second data 203 is "1" in step 3k, the index i of the second data table is designated in step 3r, and the flag of the second data 203 is removed. The second data count value is increased by one after designating it as the two data count value.

In step 3l, it is determined whether the index i has reached 128. If the index i has not reached 128, in step 3m, 1 is added to the value of the index i and then this value is 128. Repeated from step 3j until it reaches. When the value of index i reaches 128, the value of k is set to 0 in step 3n and the second data count is received in step 30. The value is 8 or more, that is, the present invention encodes the negative strength and weakness in eight steps, and determines whether the negative strength and weaknesses coded so far are eight or more levels.

If the count value of the second data 203 is 8 or more in step 30, in step 3s, the k + 1th value is stored in the k-th of the second data table, and 1 is added to k, and the k-th is added. Subsequently, all the values of the flags of the second data 203 are decreased by one, and the operation of decreasing the number of ones from the second data count value is repeated until the count value of the second data 203 becomes eight.

In step 3p, the table of the first and second data 202 and 203 created as described above is stored, moved to the beginning of the track in step 3t, and again in step 3v, the status byte ( in step (3v) to read 201) discriminates that the value of the status byte (201) 80 _H or 90 _H.

Not the intensity of the pitch and negative as described above, when the value of status byte 201 is not the 80 _H or 90 _H, and stored as the status byte 201 and data as specified in the musical instrument sound or the like, the status byte the flags of the first data 202 and second data, read the second data 203 to 203 in the case 201, one of the values is 80 _H or 90 _H, the step as the pitch and the sound intensity of the (3w) of claim A code of two data 203 is designated, and it is determined in step 3y whether the value of the first data 202 is less than the maximum recording boundary value.

In step 3y, when the value of the first data 202 is less than the maximum recording boundary value, the first data 202 is shifted 3 to the left in step 30b, and the first data (bits B2 to B0) is shifted. If the value of the first data 202 is less than the maximum recording boundary value, the pitch of the first data 202 and the negative strength of the second data 203 are stored as shown in FIG. Compress and save

First data (202) values up to record boundary adding 78 _H to the code of the second data (203) in step a, if not less than the value (3z) by bit as the 2c help (B6 in step (3y) ~ The values of B3) are all " 1 ", and the negative and weak code values are stored in the bits B2 to B0, and the original pitch value is stored in the next byte.

In step 30c, it is determined whether or not the end of the track, and repeatedly performed from step 3v until the end of the track to compress and store the first and second data.

On the other hand, Fig. 4 reads compressed MIDI data compressed and stored as described above from the arrangement data memory 301, converts them into original first and second data, and seals the FM or PCM type sound source unit 109, and The performance is performed by outputting through the mixer 119.

To do this, in step 4a, table values of the first data 202 and the second data 203 are read from the arrangement data memory 301, and in step 4b, the status byte 201 and the first data 202 are read. ) and the value of the second data 203 to read the status byte (201) in step (4C) determines whether 80 _H or 90 _H.

If it is not the value is 80 _H or 90 _H in the status byte (201) in step (4c) has a the step (4j) as not the data was compressed status byte 201, a first data 202 and second data 203 is output as it is and input to the FM or PCM system sound source unit 109.

And if the value of status byte 201. In step (4c) 80 _H or 90 _H, the means that the code of the first data 202 by a 3-bit shift right data in the step (4d) as compressed data, the bit (B6 a ~ B3) to the sensing and the first code data, and detects the code in the 7 _H to the data mask words, for each bit 7 _H and multiplying logic bit (the second data (203) B2 ~ B0).

In step 4e, it is determined whether the code of the first data 202 is OF _H or whether the values of the three bits shifted to the right by bits B5 to B3 are all '1111'. In 4f), the value of the first data 201 is read with a code value obtained by shifting bits B6 to B3 to the right from the table of the first data 202, and if all are '1111', the step 4k. Reads the next 1-byte data into the first data 202.

In step 4g, the value of the second data 203 corresponding to the code of the second data 203 is read from the table of the second data 203, and in step 4h, the status byte 201 and the pitch data are read. The first data 202 and the second data 203, which are negative strength and weakness data, are sequentially output, and in step 4i, it is determined whether the track is the end, and the process is repeated from step 4b until the end of the track. do.

As described above, the present invention compresses and stores the first data, which is the pitch data, and the second data, which is negative strength and weakness data, into one or two bytes of data according to the value of the first data, and outputs the data. The amount can be reduced to about one third, thereby reducing the consumption of the memory area and having the advantage of storing a lot of MIDI data in the limited memory area.

Claims (5)

A first step of detecting both pitch and negative strength and weakness data from the MIDI data and setting a maximum recording boundary value for compressing the pitch and negative strength and weakness data; and setting the pitch and negative intensity data detected in the first process A second step of creating and storing a table of compressed pitch data and a strength and weakness table of the compressed sound up to a maximum recording boundary value; and when the pitch data read in the first process is in the table of compressed pitch data created in the second process Coded data is coded as a table value of compressed pitch data, negative strength is coded as a table value of second data, converted into one byte, and stored if the pitch data is not in the compressed first data table. Coded with a value other than the boundary value, coded with the table value of the second data, converted to 1 byte, and stored A method of compressing MIDI data, comprising storing original pitch data in one byte. The method of claim 1, wherein the first process comprises: a first step of initializing a table number of the pitch and the negative intensity to be stored at the time of compression and each flag to maximum store the state of the pitch and the negative intensity data in one track; A second step of determining whether the next data is pitch or negative strength data by the byte value, and when the next data in the second step is pitch data and negative strength data, Initiate a third step of setting a flag and determining whether the track is at the end of the track, and repeating the second step if the track is at the end of the track, and counting pitch and negative intensity data at the end of the track. And a fourth step of specifying a maximum recording boundary value of the compressed pitch data. The method of claim 1, wherein the second process includes: a fifth step of checking whether the value of the pitch data flag is '1' and the count value is smaller than the maximum recording boundary value and determining whether the flag of the negative intensity data flag is '1'; In the fifth step, when the value of the pitch data flag is '1' and the value of the counter is smaller than the maximum record boundary value, the index number is stored as the maximum record boundary value of the compressed pitch data, and the pitch data is stored in the table of the pitch data. A sixth step of storing the data value of the flag as a count value of and assigning the count value of the pitch data to the flag of the pitch data, and then adding 1 to the count value of the pitch data; and the flag of the negative intensity data in the fifth step. Is '1', the index is assigned to the table of negative strength and weakness data, the flag of negative strength and weakness data is designated as the count value of negative strength and weakness data, and the count of negative strength and weakness data is set. A seventh step of increasing the value, an eighth step of determining whether the index has reached a maximum value in the sixth and seventh steps, and an increase of the index value if the index has not reached the maximum value in the eighth step. A ninth step, which is repeated from the fifth step, a tenth step of determining whether the coded negative intensity is eight or more when the index reaches the maximum value in the eighth step, and the tenth step. In step 11, if the two or more pairs of values are encoded in one step, re-creating the negative strength and weakness data table, and in step 10, if the negative strength and weakness table is step 8, the sixth and eleventh steps 12. A method of compressing MIDI data, comprising a twelfth step of storing a table of pitch data and negative strength and weakness data generated in a process. The method of claim 1, wherein the third process includes: a thirteenth step of reading the status byte to determine whether it is pitch and negative intensity data; and in the case of the thirteenth step, the state byte and the following is not the case. In the fourteenth step of storing the data as it is, and in the thirteenth step, in the case of the pitch and the negative strength data, the flag value of the negative strength data is set as the negative strength code value, and it is determined whether the pitch data is less than the maximum recording boundary value. Step 15 and Step 16, when the pitch data is less than the maximum recording boundary value in step 15, shifts the pitch data 3 to the left, and stores the code values of negative strength and weakness data in bits B2 to B0. When the value of the maximum recording boundary is not less than the maximum recording boundary in step 15, all the values of the bits B6 to B3 are set to '1' in the code of the negative strength and weakness data, and the negative strength and weakness are set to the bits B2 to B0. The code value is After the chapter to be MIDI data compression method of claim claim 17, made of an step of storing the value of the original pitch to the next one byte. The twenty-first step of reading the table values of the pitch data and the negative strength data, reading the stored state byte, the code value of the pitch data, and the code value of the negative strength data, and the code value of the pitch data read in the step 21 A process of determining whether the code value of the pitch data is within the maximum recording boundary value, and reading the original pitch data corresponding to the code value from the table of the pitch data when the code value of the pitch data is within the maximum recording boundary value. If the threshold value is greater than or equal to the recording boundary value, the twenty-third step of using the subsequent 1-byte value as pitch data; and if the pitch data is read from the twenty-third step, data corresponding to the negative strength and weakness code value is read from the negative strength and weakness data table. And a twenty-fourth process of outputting the status byte and the pitch data together.