JP2000221966A - Control data generating device for waveform regenerating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To make easily and automatically preparable control data appropriate for the waveform data of plural channels for stereo sound system by detecting a change point, in which the characteristic of the audio sound expressed with the waveform data is quickly changed on the time base, about one channel, and generating the control data for showing position of the change point on the time base. SOLUTION: Sample value data are read out in order from the head of the waveform data (S1). A rise of the waveform data is detected so as to discriminate whether a rising part of a syllable is detected or not (S2). In the case where a rise of the syllable is not detected, reading address of the waveform data is advanced by one (S3). This reading address is added to a syllable table (S4). When one rise is detected, the data is continuously read for the set time from the detecting position (S5). Judgement on whether the renewed reading address exceeds the address corresponding to the final sample value data among the waveform data or not is performed (S6).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a waveform reproducing apparatus for storing waveform data of a plurality of channels of audio sound forming a three-dimensional sound and reproducing the waveform data of the plurality of channels as a three-dimensional sound. The present invention relates to a control data generation device that generates control data (for example, mark data indicating the beginning of a syllable).

[0002] A series of played musical tones and audio sounds such as human voices are sampled, a time series of sampled data is stored as waveform data, and the waveform data is reproduced to generate musical tones. There is a waveform reproducing device, and electronic musical instruments equipped with such a waveform reproducing device as a sound source are becoming widespread. In such an electronic musical instrument, it is possible to detect the start timing position (head position) of a syllable, a musical tone, or the like in the stored waveform data, and to start sounding from a desired syllable or a musical sound according to a sounding instruction.
It is considered to change the pitch for each syllable or musical tone. As a method of detecting this timing position, a portion where the volume increases rapidly is judged and detected as an attack portion of a syllable rising or a musical tone, and a mark indicating a position on a time axis, which is a detection result, is used as a control data. Created as
It is general to store this mark.

[0003]

The waveform data used in a conventional waveform reproducing apparatus of this type is a monaural signal, and does not correspond to a stereo signal forming stereophonic sound.
On the other hand, considering the case where the above-mentioned mark is generated by the waveform data of the stereo signal, for example, even for the same syllable, there may be a time difference in the rise of the syllable between the right channel and the left channel of the stereo signal depending on the position of the sound source. In this case, if mark information is separately added to each of the right channel and the left channel for the same syllable or the like, the number of marks becomes complicated and it becomes difficult to handle the waveform reproduction control. In particular, there has been a conventional method of independently creating the right channel signal and the left channel signal separately for the same musical tone in a stereo system. However, in this case, the time difference between the left and right channels is often large, which is a problem. Becomes

The present invention has been made in view of such a problem, and has as its object to easily and automatically create control data suitable for waveform data of a plurality of channels for stereophonic sound.

[0005]

According to the present invention, there is provided a waveform reproducing apparatus for storing waveform data of a plurality of channels of audio sound forming stereophonic sound and reproducing the waveform data of the plurality of channels so as to form stereophonic sound. The present invention relates to a control data generation device for generating control data used in the above.

In order to solve the above-mentioned problems, a control data generating apparatus according to the present invention provides, as one mode, audio data represented by waveform data of one channel selected from a plurality of channels. A change point detecting means for detecting a change point at which the characteristic of the sound changes rapidly on the time axis and generating control data indicating a position on the time axis of the change point; and control data generated by the change point detection means And a control table generating means for generating a control table for storing The change point at which the characteristic of the audio sound represented by the waveform data changes abruptly on the time axis includes a point at which the amplitude value sharply increases (that is, a rising portion of a syllable or a musical tone) and a point at which the pitch sharply changes. And so on. In this control data generation device, the change point detecting means detects a change point at which the characteristic of the audio sound represented by the waveform data of one channel suddenly changes on the time axis, and detects the change point on the time axis. Is generated. The control table generating means generates a control table for storing the control data generated by the change point detecting means. Thereby, it is possible to prevent a plurality of control data from being generated based on a change in the characteristic of the same sound and recorded in the control table in duplicate.
When reproducing the waveform, the control operation can be performed with reference to the control data of the control table.

According to another aspect of the present invention, there is provided a control data generating apparatus in which, for each of the waveform data of a plurality of channels, the characteristic of the audio sound represented by the waveform data changes abruptly on the time axis. A change point detecting means for detecting a point and generating control data indicating a position on the time axis of the change point, and a control table for storing control data generated by the change point detecting means, based on the same sound. A control table generating means for integrating and generating control data of a change point generated in each of the plurality of channels into any one of them. The change point at which the characteristic of the audio sound represented by the waveform data changes abruptly on the time axis includes a point at which the amplitude value sharply increases (that is, a rising portion of a syllable or a musical tone) and a point at which the pitch sharply changes. And so on. In this control data generation device, the change point detecting means detects, for each of the waveform data of a plurality of channels, a change point at which the characteristic of the audio sound represented by the waveform data sharply changes on the time axis, and detects the change point. Control data indicating the position of the point on the time axis is generated. In generating the control table storing the control data generated by the change point detecting means, the control table generating means may select one of the control data of the change points generated in each of the plurality of channels due to the same sound. To integrate. Thereby, it is possible to prevent a plurality of control data from being generated based on a change in the characteristic of the same sound and recorded in the control table in duplicate. When reproducing the waveform, the control operation can be performed with reference to the control data of the control table.

The control table generating means in the control data generating apparatus according to the other aspect described above once generates control data for all of the waveform data of the plurality of channels, and generates the control data of the plurality of channels. Of the control data, control data within a predetermined time on the time axis is integrated with control data of any one channel, for example, control data first detected within the predetermined time is integrated as control data of the sound. Can be composed of Alternatively, the change point detecting means in the control data generation device of the other embodiment sequentially generates control data for each of the waveform data of the plurality of channels, and when a change point is detected in one channel, the change point is detected. It can be configured to detect the next change point after jumping a predetermined time from any change point to any channel.

In another aspect, the control data generating apparatus according to the present invention is represented by synthesizing means for synthesizing waveform data of a plurality of channels into one waveform data, and waveform data synthesized by the synthesizing means. Change point detecting means for detecting a change point at which the characteristic of the audio sound changes rapidly on the time axis and generating control data indicating a position on the time axis of the change point; and control generated by the change point detection means Control table generating means for generating a control table for storing data. The change point at which the characteristic of the audio sound represented by the waveform data changes abruptly on the time axis includes a point at which the amplitude value sharply increases (that is, a rising portion of a syllable or a musical tone) and a point at which the pitch sharply changes. And so on. In this control data generation device, the synthesizing unit synthesizes the waveform data of a plurality of channels into one waveform data to obtain a monaural signal. The change point detecting means detects a change point at which the characteristic of the audio sound represented by the monaural waveform data suddenly changes on the time axis, and generates control data indicating a position on the time axis of the change point. Generate. The control table generating means generates a control table for storing the control data generated by the change point detecting means. Thereby, it is possible to prevent a plurality of control data from being generated based on a change in the characteristic of the same sound and recorded in the control table in duplicate. When reproducing the waveform, the control operation can be performed with reference to the control data of the control table.

Further, a recording medium according to the present invention comprises a computer-readable recording medium storing a program for functioning as each means in the control data generating apparatus according to any one of the above embodiments.

[0011]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 2 shows a hardware configuration when the control data generation device of the waveform reproducing device as one embodiment of the present invention is realized by a personal computer. In FIG. 2, a CPU 1 is a central processing unit that controls the entire apparatus. The ROM 2 is a read-only memory that stores a control program (such as an OS) and a table, and the RAM 3 is a random access memory that provides a working memory for the CPU. The operation unit 4 includes a computer keyboard, various switches attached to the computer main body, a keyboard device (keyboard), and the like. As the display unit 5, a display device such as a CRT display device or a liquid crystal display device can be used.

Reference numeral 6 denotes a hard disk device which stores various installed application programs and various audio waveform data. Generally, audio waveform data is a time series of sampled data obtained by sampling one frame of audio sound, such as a musical sound or a human voice, by two left and right channels so as to form a three-dimensional sound. However, in this embodiment, a human voice waveform is used as an example. Note that the waveform data may be a time series of amplitude value data of a musical tone created by a sequencer or the like in addition to sampling recording. The D / A converter 7 converts the audio waveform signal (digital value) reproduced by the present system from digital to analog, and outputs it.
This audio waveform signal (analog value) is input to an audio system (not shown) and emitted.

FIG. 3 is a functional block diagram of the computer system. Roughly, it comprises a marking device 20 and a waveform reproducing device 21. The marking device 20 generates mark data indicating a position corresponding to the beginning of the syllable or the like based on the waveform data, and creates a control table. . Hereinafter, this control table is referred to as a syllable table in this embodiment.

The waveform reproducing device 21 is a device that stores waveform data of left and right two-channel audio sounds forming stereophonic sound and reproduces the audio data by reproducing the waveform data of the plurality of channels. When playing back waveform data, various operations such as playing each syllable at the specified pitch, starting playing from the beginning of the specified syllable, performing time stretching, tone processing, etc. Is possible.

More specifically, the mark providing section 20 stores a waveform memory 22 for storing waveform data of an audio sound, detects the beginning of a syllable or the like from the waveform data in the waveform memory 22, adds a mark, and writes the mark. A mark processing unit 23 for generating an syllable table, a waveform memory 24 for storing the syllable table together with waveform data, and the like are provided. The waveform reproducing device 21 includes a reproducing unit 26 that reproduces waveform data based on the waveform data in the waveform memory 24 and a syllable table, and a pitch instructing unit 25 that instructs the pitch of the reproduced waveform.
Etc.

FIG. 3 shows the structure of the above-mentioned voice table. The time series of the sample value data constituting the waveform data is sequentially inspected, and a portion (change point) where the amplitude value data rapidly rises is determined as the head position of the syllable (syllable), and the position information on the time axis is determined. Marks M are sequentially stored in the syllable table. Waveform data is 1
The data is stored in the sequential address area of the memory at a rate of one sample value data per address. The position information on the time axis is the number of the address from the start address of the waveform data (that is, the number of the address). Sample value data) by relative addressing. This waveform data is composed of a plurality of syllables arranged in a time series, and a mark (address) indicating the head of the nth syllable is represented by M (n), and is applied to the syllable table. Marks of n syllables M (1) to M
(N) are stored in order. Further, in this syllable table, the pitch N at the time of reproducing each syllable is also stored for each syllable, and the pitch of the nth syllable is N (n)
It is represented by

In the process of additionally writing data to the syllable table, basically, as shown in FIG. 5, syllables in the waveform data are sequentially detected from the leading side thereof. When the head of the syllable is detected, a mark M (n) is written in the n-th column of the syllable table.
Next, the process of incrementing the value of n by one in preparation for writing the detected syllable mark is repeated.

The syllable table creation operation of this embodiment will be described below. In this embodiment, the syllable table can be created by a method arbitrarily selected from the following four methods I to VI. First, these methods I ~
The outline of VI is as follows.

[Method I] In method I, the rise (leading) of a syllable is detected only for one channel (one channel) of the left and right (L, R) audio waveforms forming stereophonic sound. A syllable table is created for the mark corresponding to the head position.

[Methods II and III] In methods II and III, as shown in FIG. 1 (1), of the audio waveforms of the left and right channels forming the stereophonic sound, for the channel which rises first for the same syllable, A syllable table is created by creating a mark and ignoring the left and right channels for the rise within a predetermined time β from that position, thereby integrating the marks of the left and right channels for the same syllable into one.

[Method VI] In method VI, as shown in FIG. 1 (2), the absolute values of the audio waveforms of the left and right channels forming the stereophonic sound are obtained and added and synthesized (monaural). A rising edge of a syllable is detected in the monaural audio waveform, a mark is generated, and a voice table is created.

The syllable table creation processing of the above-described method I will be described below. FIG. 7 is a flowchart showing the procedure of the syllable table creation processing of the method I. First, one of the left and right channels is selected (in this example, left (L)
The waveform data of the left channel is copied from the hard disk device 6 to the RAM 3. Then, with respect to the waveform data of the left channel, the sample value data is sequentially read from the leading side (step S1). The read address of the sample value data is referred to as a read reference address
Then, the rising detection described below is performed on the waveform data in the vicinity of the read reference address a (a window to be described later) to determine whether or not the rising portion of the syllable has been detected (step S2).

This rising detection is performed by the method shown in FIG.
FIG. 6A shows an audio waveform based on the waveform data, and shows a state in which two syllables continue. FIG. 6B is an enlarged view of the head of the syllable on the rear side (the vertical amplitude axis is compressed). A window composed of three points (addresses) a, b, and c is applied to this audio waveform to detect the rise of the syllable waveform. This window has a first section a to b on the front side of the read reference address a and a second section a to c on the rear side, and the first section is a second section. Take wider than.
The principle of detection is as follows.

1. The waveform data of the first predetermined section (ab) from the read reference address a to the previous address b is read and its absolute value is obtained, and the first predetermined section (ab) is obtained.
An average value of the data is calculated to be a first average value. 2. The waveform data of the second predetermined section (ac) from the read reference address a to the address c subsequent to the read reference address a is read, its absolute value is obtained, and the average value of the data of the second predetermined section (ac) is calculated. To obtain a second average value. 3. Divide the second average by the first average. 4. If the result of the division is equal to or greater than a predetermined value, it is determined that the position of the read reference address a is a location where the sound volume (that is, the amplitude value) is rapidly increased, and the read reference address a is set to the rising of the syllable. Part (for music, attack part)
Judge.

The length of the first predetermined section (ab), the length of the second predetermined section (ac), the value for judging the rise from the quotient calculated by division, and the like are as follows. Appropriate values are determined in advance by experiments.

If no rising syllable is detected in step S3, the waveform data read address is advanced by one. On the other hand, when a rising syllable is detected,
The read address (read reference address a) at that time is added to the syllable table as mark data M (n). Thereafter, the read address for reading the next sample value data is advanced by an address width corresponding to a preset time α. This is because, near the rise of a syllable, when the rise is detected by the above-described rise detection processing, the rise may be detected at a plurality of locations. Therefore, when one rising edge is detected, the reading is skipped from the position for the set time α to suppress the generation of noise-like mark data.

Next, it is determined whether the read address updated in step S3 or S5 has exceeded the address corresponding to the last sample value data in the waveform data (step S6). If the address has not been exceeded, the processing from step S1 is repeated. If the last address has been exceeded, the syllable table creation processing ends. Through the above processing, a syllable table according to the method I can be created.

Next, the syllable table creation processing of the system II will be described. FIG. 8 is a flowchart showing the procedure of the syllable table creation process of the method II. As described above, schemes II and III detect the rise of syllables for the left and right channels and integrate the marks of the left and right channels for the same syllable into one.
II is all of the left (L) channel waveform data and the right (R)
After performing syllable rise detection separately for all channel waveform data and creating each syllable table once, the syllable table of the left channel and the syllable table of the right channel are combined into one, and the approach of the left and right channels is This is to integrate the marks at the specified positions into one.

In FIG. 8, steps S11 to S16 correspond to the processing of the above-described method I (steps S1 to S1).
This is the same as S6), and the syllable table for the left channel can be created by this processing. The following step S17 is a process of returning the read address to the head address of the waveform data in order to process the waveform data of the right channel. Steps S18 to S23 are the same as the processing of the above-described method I (steps S1 to S6) except that the waveform data of the right channel is to be processed, and the syllable table of the right channel is created by this processing. can do.

The syllable table of the left channel and the syllable table of the right channel created in this way are integrated into one (step S24). In this processing, when there are a plurality of marks within the set time β for the marks of each syllable table, the marks are integrated into one by removing the other marks except for the first one of them. .

FIG. 9 is a flowchart showing a detailed procedure of the mark removal processing. First, the syllable table of the left channel and the marks of the syllable table of the right channel are sorted so as to be arranged in order on the time axis to form one syllable table (step S241).

Thereafter, a variable n for indicating the syllable number is reset to "0". It is determined whether the value (n + 1) indicating the syllable next to the nth syllable has exceeded the number of the last syllable in the waveform data (step S243). If it exceeds, it means that the processing has been completed for the n-th syllable of interest, and the mark removal processing is terminated.
If not, is the time interval between the mark M (n) of the n-th syllable of interest and the mark M (n + 1) of the next (n + 1) -th syllable at least a set time β or more apart? It is determined whether or not it is (step S244).

If this time interval is smaller than the set time β (step S244), the n-th and (n + 1) -th marks M (n) and M (n + 1) are marks generated based on the same syllable. Judge the latter mark M
(N + 1) is erased (step S246), and marks are integrated. In the erasing process, the mark M (n + 2) next to the erased mark M (n + 1) is set as a new mark M (n + 1), and the subsequent marks are sequentially packed.

On the other hand, if the time intervals are separated by the set time β or more (step S244), the n-th and (n +
It is determined that the 1) th marks M (n) and M (n + 1) are mark data generated based on different syllables, and the value of the variable n is incremented by one (step S245). The processing after S243 is repeated.

Next, the syllable table creation processing of the method III will be described. FIG. 10 is a flowchart showing the procedure of the syllable table creation processing of the method III. This method III
Similarly to method II, the rising and falling of syllables are detected for the left and right channels and the marks of the left and right channels for the same syllable are integrated into one.
Is different from That is, in this method III, the sample value data at the same time is simultaneously read from both the left channel and the left channel to detect the rise, and when the rise is detected, the detection result is added to the syllable table as a mark, and Is skipped from the detection position by a predetermined time β.

In FIG. 10, first, sample value data is read from the waveform data of the left channel (step S).
31) The rising of the syllable is detected using the address of the sample value data as the read reference address a (step S3).
2) If no rise is detected, sample value data is read from the waveform data of the right channel at the same read address (step S33), and the rise of a syllable is detected (step S34). When the rise of a syllable is detected from the waveform data of the left channel or the right channel (steps S32 and S34), the detection result is added to the syllable cable (step S3).
6) The read address is advanced by the set time β (step S37), and if no rise is detected, the read address is incremented by one (step S3).
5) It is determined whether the updated read address has exceeded the last address of the waveform data (step S38). If not, the process returns to step S31 and repeats the process. .

Next, the syllable table creation processing of the method VI will be described below. FIG. 11 is a flowchart showing the procedure of the syllable table creation processing of this method VI. As described above, this method VI creates a monaural signal for audio table creation work from the waveform data of the left and right channels, detects a rising portion of a syllable from the waveform data of the monaural signal, and creates a syllable table. .

In FIG. 11, sample values of the waveform data of the left and right channels are read using the same read address (step S41), their absolute values are taken (step S42), and the sum is added to create a monaural signal. (Step S43). This processing is performed over the entire length of the waveform data of the left and right channels (all sample value data) (steps S41 to S43). The reason why the absolute value is taken when the waveform data of the left and right channels is converted to monaural as described above is to prevent the two from canceling out due to the phase relationship between the sample value data.

Thereafter, the read address is returned to the head address (step S44), and a syllable table creation process is performed for the created monaural signal waveform data (steps S45 to S49). The processing in steps S45 to S49 is the syllable table creation processing (S
1 to S6), and a detailed description thereof will be omitted.

Next, a control operation when reproducing waveform data having syllables in the waveform reproducing apparatus using the syllable table will be described. FIG. 12 shows an outline of this control operation. In this waveform reproduction, the phrase itself is reproduced at the original reproduction speed, and the pitch of each syllable is specified in advance before they are reproduced, and when the reproduction time has elapsed, the phrase is reproduced at the indicated pitch. . As shown in FIG. 12, the waveform data has a plurality of syllables of 1, 2, 3, 4,
... And the pitch of the syllable is specified in advance by an operator prior to its reproduction (the vertical axis in the figure indicates the magnitude of the pitch). Shown), and each syllable to be reproduced has the designated pitch.

FIG. 13 is a flow chart showing a process for instructing the pitch of each syllable. Before reproducing the waveform data, the pitch of each syllable in the syllable table is indicated and the syllable table is assigned to each syllable. Record it. First, a variable n indicating a syllable number is reset to “0” (step S51),
It is determined whether or not this syllable number n has reached the value of the last syllable (step S52). If it is not the last syllable, it is determined whether or not the pitch instruction operator has been operated (step S53), and the pitch instruction operation has been performed. If so, the designated pitch N (n) is written in the syllable number n column of the syllable table (step S5).
4), the syllable number n is incremented by one (step S)
55), returning to step S52 to repeat the processing; The same applies to step S when the pitch instruction operation element is not operated.
Returning to step 52, the process is repeated. When this process has been performed up to the last syllable in the syllable table, this flow ends. As a result, the pitch N (n) at the time of reproducing each syllable in the syllable table can be recorded.

Next, the operation of reproducing the waveform while controlling the pitch using this syllable table will be described. FIG. 14 is a flowchart showing the waveform reproduction processing. The variable n indicating the syllable number is reset to "0" (step S560), and for the waveform data of the left and right channels, the waveform data after the syllable indicated by the mark M (n) is written in the syllable table. Sound at high N (n) (step S6)
1), it is determined whether this syllable number is the last syllable in the syllable table (step S62). If not, the syllable number n is incremented by one and the processing from step S61 on is repeated. When the last syllable is reached (step S6
2) End the reproduction process.

In practicing the present invention, various modifications are possible. For example, in the above-described embodiment, a human voice is used as an example of the waveform data, and therefore, the mark is set at each rising of the syllable. However, the present invention is not limited to this, and the audio waveform data of the musical instrument is used as the audio waveform data. A sound waveform may be used. In this case, the rise of each musical tone becomes an attack portion, and there is also a sharp increase in amplitude here, so that detection can be performed by the same principle as described above. Further, the position at which the mark is applied is not necessarily limited to the head position of the syllable or musical sound, and the mark may be applied at a position where the volume of the audio sound suddenly changes.

Further, in the above-described embodiment, the point at which the volume (amplitude) suddenly changes is set as a change point of the waveform, and the change point is marked, but the present invention is not limited to this.
For example, a point at which the pitch suddenly changes (a position on the time axis) may be detected and a mark may be added to the change point. For such detection, for example, the detected pitch may be compared with the previous detected pitch, and a portion having a pitch change corresponding to a semitone or more of the scale may be set as a change point. Since the portion where the pitch changes in this way is often a break in the sound (or a point where the quality of the sound changes) in the waveform data, it is also possible to add the mark to the change point in this manner. The information can be used for controlling waveform reproduction.

In the above embodiment, the case where the present invention is realized by a personal computer has been described. However, the present invention is not limited to this, and the present invention is realized by a dedicated hardware device called an electronic musical instrument. There may be. When implemented by a personal computer, a computer-readable recording medium, such as a compact disk (C
D), stored in a recording medium such as a floppy disk (FD), a magnetic tape, or an optical disk, and installed in a personal computer to realize the apparatus of the present invention.

[0046]

As described above, according to the present invention, control data suitable for waveform data of a plurality of channels for stereophonic sound can be easily and automatically created.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining an outline of a control data creation method according to the present invention.

FIG. 2 is a diagram showing a hardware configuration of a personal computer system for realizing a control data generation device of a waveform reproduction device as one embodiment according to the present invention.

FIG. 3 is a diagram illustrating an example apparatus in a functional block configuration.

FIG. 4 is a diagram showing a data structure of a syllable table used in the embodiment device.

FIG. 5 is a flowchart illustrating a general method for performing a syllable table mark adding process in the embodiment device.

FIG. 6 is a diagram for explaining a syllable or other rising detection calculation in the apparatus of the embodiment.

FIG. 7 is a flowchart illustrating a procedure of a method I syllable table creation process in the apparatus of the embodiment.

FIG. 8 is a flowchart illustrating a procedure of a syllable table creation process of method II in the apparatus of the embodiment.

FIG. 9 is a flowchart illustrating a detailed procedure of a mark removal process in a system II syllable table creation process routine of the embodiment device.

FIG. 10 is a flowchart illustrating a procedure of a method III syllable table creation process in the apparatus of the embodiment.

FIG. 11 is a flowchart illustrating a procedure of a syllable table creation process of the method VI in the apparatus according to the embodiment.

FIG. 12 is a diagram schematically illustrating an example of a control operation at the time of reproduction by the device of the embodiment.

FIG. 13 is a flowchart illustrating an example in which a pitch instruction of a reproduced waveform is instructed in a syllable table in the apparatus of the embodiment.

FIG. 14 is a flowchart illustrating an example in which pitch control is performed based on a syllable table during reproduction of waveform data in the apparatus of the embodiment.

[Description of Signs] 1 CPU (Central Processing Unit) 2 ROM (Read Only Memory) 3 RAM (Random Access Memory) 4 Operator 5 Display Unit 6 Hard Disk Device

Continuation of front page (72) Inventor Hiroharu Taniguchi 1-4-16-1 Dojimahama, Kita-ku, Osaka-shi, Osaka F-term in Roland Corporation (reference) 5D378 AD22 AD67 AD70 JA02 JA03

Claims (6)

[Claims] 1. A control data generating apparatus for a waveform reproducing apparatus for storing waveform data of audio sound of a plurality of channels forming a stereophonic sound and reproducing the waveform data of the plurality of channels so as to form a stereophonic sound. For a waveform data of one channel selected from the plurality of channels, a change point where the characteristic of the audio sound represented by the waveform data rapidly changes on the time axis is detected, and the change point on the time axis is detected. A control data generating apparatus, comprising: a change point detecting means for generating control data indicating the position of the control point; and a control table generating means for generating a control table for storing the control data generated by the change point detecting means. 2. A control data generating apparatus for a waveform reproducing apparatus for storing waveform data of audio sound of a plurality of channels forming stereophonic sound and reproducing the waveform data of the plurality of channels so as to form stereophonic sound. Control data indicating, for each of the waveform data of the plurality of channels, a change point where the characteristic of the audio sound represented by the waveform data rapidly changes on the time axis, and indicating a position on the time axis of the change point And a control table for storing control data generated by the change point detecting means. The control data of a change point generated in each of the plurality of channels due to the same sound is set to one of the following. A control data generation device, comprising: a control table generation unit that integrates and generates a control table. 3. The control table generating means once generates control data for all of the waveform data of the plurality of channels and over a predetermined time period on the time axis among the generated control data of the plurality of channels. 3. The control data generation device according to claim 2, wherein the control data within the range is integrated with control data of any one channel. 4. The change point detecting means sequentially generates control data for each of the waveform data of the plurality of channels, and when a change point is detected in one channel, any one of predetermined time from the change point is determined. 3. The control data generation device according to claim 2, wherein the next change point is detected after jumping to the channel of (1). 5. A control data generating apparatus for a waveform reproducing apparatus for storing waveform data of audio sound of a plurality of channels forming stereophonic sound and reproducing the waveform data of the plurality of channels so as to form stereophonic sound. Synthesizing means for synthesizing the waveform data of the plurality of channels into one waveform data; and detecting a change point at which a characteristic of an audio sound represented by the waveform data synthesized by the synthesizing means rapidly changes on a time axis. Change point detecting means for generating control data indicating the position of the change point on the time axis, and control table generating means for generating a control table for storing the control data generated by the change point detecting means. Control data generator. 6. A computer-readable recording medium storing a program for causing a computer to function as each means in the control data generation device according to claim 1.