JP4213835B2 - Waveform playback device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a waveform reproduction device for reproducing a series of stored waveforms.
[0002]
[Prior art]
2. Description of the Related Art Conventionally, a waveform reproducing apparatus that stores a series of waveforms in the form of a musical sound signal obtained by sampling a series of sounds such as musical sounds in a sampler or the like and reproduces the waveforms is known.
[0003]
In Japanese Patent Laid-Open No. 11-52954, in order to enrich the music expression at the time of reproduction of a sampled musical sound signal, the waveform reproduction speed can be controlled to reproduce the waveform faster or slower, and the reproduction speed is set to “0”. A waveform reproduction apparatus that can also be reproduced by stopping the reproduction position on the waveform by setting to, and further reproducing in the direction of turning the time axis of the waveform by setting the reproduction speed to a negative speed. Proposed.
[0004]
[Problems to be solved by the invention]
However, if the waveform reproduction speed is controlled as in the previous period, it is difficult to reproduce the waveform position desired by the performer, and the performance operation is very difficult.
[0005]
Therefore, there is a demand for a waveform reproducing apparatus that can reproduce sampled musical sound signals with rich musical expressions and easy performance operations.
[0006]
In view of the above circumstances, the present invention allows a user to directly specify a playback position on a waveform, and to reproduce a musical sound with rich musical expression desired by the user with an easy performance operation. An object is to provide an apparatus.
[0007]
[Means for Solving the Problems]
The waveform reproduction apparatus of the present invention that achieves the above object includes waveform storage means for storing a series of waveforms,
A position specification operator that specifies the position on the waveform stored in the waveform storage means and allows the specified position to be changed continuously. The position control is an ON state in which the position specification is valid and an OFF state in which the position specification is invalid. A position specification operator that can be switched between,
When the position specifying operator is in the ON state, the position specified by the position specifying operator on the series of waveforms stored in the waveform storage means is included in the reproduction range determined according to the specified position. When the position is reproduced and the playback range is changed according to the continuous change of the specified position and the position specifying operator is in the OFF state, the series of waveforms stored in the waveform storage means are changed to the series of waveforms. When playback is performed while moving the playback position at a desired speed, and the position specifying operator is switched from the OFF state to the ON state, the progress of the playback position on the waveform is delayed, and the playback position is set. And a waveform reproduction means for shifting to waveform reproduction at a designated position.
[0008]
Here, the position designation operator may be switched between an on state and an off state by an operation of the position designation operator itself, and is an operator (for example, a pedal operator) independent of the position designation operator. It may be switched by an operation or the like.
[0009]
Further, the waveform reproducing means may be one that sequentially reproduces the reproduction range when in the on state, or one that reproduces the reproduction range randomly. Furthermore, the reproduction range may include only the designated position.
[0010]
According to the waveform reproducing apparatus of the present invention, a series of sounds such as musical sounds are sampled and stored in advance in the waveform storage means, and when the user turns off the position specifying operator, the stored series of sounds are reproduced. It is played back at the desired speed. Then, when listening to the reproduced sound, the user switches the state of the position specifying operator to the on state and directly designates the position of the sound having a desired tone color among the series of sounds. Thus, the sound at the designated position is reproduced repeatedly. When the user continuously changes the designated position by the position designation operator, the reproduction position is continuously changed according to the continuous change.
[0011]
As a result, the user can easily find and specify the position of the sound desired to be reproduced on the waveform, and can realize the musical sound reproduction with rich music expression by an easy performance operation.
[0012]
Further, the waveform reproducing device of the present invention is
It is preferable that the waveform reproducing means acquires pitch information indicating the pitch of the divided waveform and reproduces the waveform at the pitch indicated by the pitch information.
[0013]
According to the waveform reproducing apparatus having such a preferable configuration, it is possible to perform performances using, for example, a keyboard or a MIDI signal input device.
[0014]
Furthermore, in the waveform reproduction apparatus of the present invention, the waveform reproduction means may reproduce the waveform at a pitch corresponding to the speed of changing the designated position by the position designation operator.
[0015]
According to the waveform reproducing device having the waveform reproducing means for reproducing the waveform at a pitch corresponding to the speed of changing the designated position, the operation speed can be increased to a so-called scratch that is performed by moving an analog record by hand. Correspondingly, an effect of changing the pitch of the reproduced sound can be obtained.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described.
[0017]
FIG. 1 is a diagram showing an embodiment of a waveform reproducing device of the present invention.
[0018]
The waveform reproducing device 10 is a kind of electronic musical instrument, and is a device that designates a position on a series of stored waveforms and reproduces a sound represented by a waveform around the position as a performance sound. Here, as will be described below, a time-axis compression / expansion technique that compresses and expands the waveform in the time-axis direction is used to reproduce a waveform near a specified position on the waveform with a natural tone color.
[0019]
The waveform reproduction apparatus 10 includes a waveform memory 11, a DSP 12, a CPU 13, a RAM 14, a ROM 15, a ribbon controller 16, a keyboard 17, an operator group 18, and a D / A converter 19. The waveform memory 11 stores a series of waveforms obtained by sampling a series of sounds as PCM (Pulse Code Modulation) waveform data. Therefore, the waveform memory 11 is an example of the waveform storage means referred to in the present invention. Further, the CPU 13 detects the operation state of the ribbon controller 16, the keyboard 17, and the operator group 18, and controls the DSP 12 according to the detection result. As will be described later, the waveform stored in the waveform memory 11 is reproduced by the DSP 12. That is, the DSP 12 and the CPU 13 constitute an example of the waveform reproducing means referred to in the present invention.
[0020]
The ROM 15 stores a program representing the operation of the CPU 13 and the DSP 12, and the program representing the operation of the DSP 12 is transferred to the DSP 12 by the CPU 13. The RAM 14 is used as a working memory for the CPU 13.
[0021]
The ribbon controller 16 is an example of a position designation operator according to the present invention, and designates a position on the waveform stored in the waveform memory 11.
[0022]
The keyboard 17 instructs the CPU 13 to start and end waveform reproduction. In addition, pitch information indicating the pitch associated with the pressed key is input to the CPU 13 to specify the pitch at the time of waveform reproduction. The pitch is specified by the pitch shift amount corresponding to the pitch between the pitch associated with the key pressed by the performance operation and the predetermined reference pitch.
[0023]
The operator group 18 is for instructing the CPU 13 of the waveform compression / decompression amount and the like. The operators included in the operator group 18 will be described later as appropriate.
[0024]
The D / A converter 19 converts a digital signal corresponding to the waveform reproduced by the DSP 12 into an analog signal, and emits a sound corresponding to the analog signal from the speaker 19a.
[0025]
FIG. 2 is an explanatory diagram of a position specifying operation using the ribbon controller 16.
[0026]
The upper part of FIG. 2 shows a waveform 11a represented by the waveform data stored in the waveform memory 11, and the lower part shows a ribbon controller 16.
[0027]
When the user's finger F is away from the ribbon controller 16, the position designation is invalid, and the waveform 11a is reproduced at a predetermined reproduction speed. Then, when the user touches the ribbon controller 16 with the finger F, the position designation by the ribbon controller 16 becomes effective, and any part on the ribbon controller 16 is reproduced when the finger F on the waveform 11a touches. The position P is handled as specified. Further, the ribbon controller 16 can detect the position where the user's finger F is touching, and if the finger F moves in the A1 direction or the B1 direction shown in the figure while touching the ribbon controller 16, it will be on the waveform 11a. The designated position moves continuously in the A2 direction and the B2 direction.
[0028]
FIG. 3 is a flowchart of the main routine of the CPU 13 shown in FIG.
[0029]
When the main routine of the CPU 13 is started, initialization is first performed (step S10).
[0030]
Then, an operator group processing routine for detecting the operation state of the operator group 18 shown in FIG. 1 and performing processing in accordance with the operation state is executed (step S11), and the operation state of the keyboard 17 is detected and changed to the operation state. A keyboard processing routine for performing a corresponding process is executed (step S12), and a ribbon controller processing routine for detecting an operation state of the ribbon controller 16 and performing a process according to the operation state is executed (step S13).
[0031]
Thereafter, steps S11 to S13 are repeatedly executed.
[0032]
FIG. 4 is a flowchart of the operator group processing routine executed in step S11 of the main routine shown in FIG.
[0033]
In this operator group processing routine, the value of the formant volume representing the size of the instrument or person as a resonator, and the waveform reproduction speed are set according to the operation of the operator group 18 shown in FIG. The control group 18 includes control elements other than the control elements for setting the formant volume value, the reproduction speed value, and the like, but the description thereof is omitted here.
[0034]
When this operator group processing routine is started, a series of operators included in the operator group 18 shown in FIG. 1 is scanned (step S11_1), and any one of the series of operators is selected. It is determined whether or not a state change has occurred (step S11_2).
[0035]
If it is determined that no state change has occurred, the operator group processing is terminated as it is, and the process returns to the main routine shown in FIG. On the other hand, if it is determined that a state change has occurred, it is determined whether or not the state change is a state change of an operator that designates a formant volume F_VR representing the size of a musical instrument or a person as a resonator. S11_3). If it is determined that the state of the operator specifying the formant volume F_VR is changed, the value of the register F_VR storing the formant volume F_VR of the CPU 13 shown in FIG. 1 is updated (step S11_4). The updated value is set in the register f_vr for storing the formant volume F_VR of the DSP 12 (step S11_5).
[0036]
In the following description, a register that stores a numerical value and a numerical value that is stored in the register may be used without distinction, and they may be represented by the same symbol. Unless otherwise specified, numerical values and registers for the CPU 13 are represented by capital letters, and numerical values and registers for the DSP 12 are represented by small letters.
[0037]
When the change detection or update of the formant volume F_VR is completed, it is next determined whether or not the state change detected in step S11_2 is a state change of an operator that specifies the compression / decompression amount TCOMP that determines the waveform reproduction speed. (Step S11_6). If it is determined that the state of the operator specifying the compression / decompression amount TCOMP is changed, the compression / decompression amount TCOMP of the CPU 13 shown in FIG. 1 is updated (step S11_7), and the updated value is updated by the DSP 12. The compression / decompression amount tcomp is set (step S11_8).
[0038]
Thereafter, the operator group processing is terminated, and the process returns to the main routine shown in FIG.
[0039]
FIG. 5 is a flowchart of the keyboard processing routine executed in step S12 of the main routine shown in FIG.
[0040]
In this keyboard processing routine, in accordance with the operation of the keyboard 17 shown in FIG. 1, the value of the gate flag gate indicating the generation and disappearance of the reproduced sound and the shift amount p_shift of the reproduced pitch from the reference pitch are set in the DSP 12. .
[0041]
When this keyboard processing routine is started, a series of keys constituting the keyboard 17 are scanned (step S12_1), and it is determined whether or not a change has occurred in the operation state of any one of the series of keys. (Step S12_2).
[0042]
If it is determined that no change has occurred in the operation state, the keyboard processing routine is terminated as it is, and the process returns to the main routine shown in FIG. On the other hand, if it is determined that the operation state has changed, it is determined whether or not all keys on the keyboard 17 have been released (step S12_3). If it is determined that all keys have been released, the value “0” representing the disappearance of the reproduced sound is substituted for the gate flag GATE (step S12_10), and the value of the gate flag GATE of the CPU 13 is the gate flag. gate is set (step S12_9), and the process returns to the main routine shown in FIG.
[0043]
If it is determined in step S12_3 described above that any key is depressed, is the current operation state a state where a new key is depressed from a state where all keys are released? It is determined whether or not (step S12_4), and if it is determined that the key is newly pressed, a value “1” representing the occurrence of the reproduced sound is substituted for the gate flag GATE (step S12_5).
[0044]
Thereafter, pitch data KEYP representing the pitch of the highest pitch among the pitches associated with each of the one or more keys currently pressed is obtained from the keyboard 17 (step S12_6). A ratio KEYP / ORGP between the value of the data KEYP and the value of the original pitch data ORGP representing a predetermined reference pitch is substituted as a playback pitch shift amount P_SHIFT from the reference pitch (step S12_7).
[0045]
Then, the current value of the shift amount P_SHIFT of the CPU 13 is set to the shift amount p_shift of the DSP 12 (step S12_8), the value of the gate flag GATE is set to the gate flag gate (step S12_9), and the process returns to the main routine shown in FIG. .
[0046]
FIG. 6 is a flowchart of the ribbon controller processing routine executed in step S13 of the main routine shown in FIG.
[0047]
In this ribbon controller processing routine, in accordance with the operation of the ribbon controller 16 shown in FIG. 1, the value of the touch flag r_touch indicating whether or not the ribbon controller 16 is touched and the amount of movement from the initially touched position are set. The corresponding offset offset value is set in the DSP 12.
[0048]
When this ribbon controller processing routine is started, it is determined whether or not the ribbon controller 16 is currently touched (step S13_1). If it is determined that the ribbon controller 16 is not touched, it is touched as the value of the touch flag R_TOUCH. A value “0” indicating that there is no value is substituted (step S13_9), the value of the touch flag R_TOUCH is set to the touch flag r_touch of the DSP 12 (step S13_8), and the process returns to the main routine shown in FIG.
[0049]
If it is determined in step S13_1 that the ribbon controller 16 is currently touched, a value “1” indicating that the ribbon controller 16 is touched is substituted as the value of the touch flag R_TOUCH (step S13_2). If the value of the touch flag R_TOUCH has just risen from “0” to “1”, that is, if the ribbon controller has just been touched (step S13_3: Y), the current touch position Is detected, and a value indicating the touch position is set as the reference position BPOINT (step S13_4). Also, regardless of whether or not it is immediately after a new touch, the current touch position is detected, and a value indicating the touch position is set as the current position POINT (step S13_5). Then, the difference POINT−BPOINT between the current position POINT and the reference position BPOINT is substituted for the offset OFFSET of the CPU 13 (step S13_6), and the value of the offset OFFSET is set to the offset offset of the DSP 12 (step S13_7).
[0050]
Thereafter, the value of the current touch flag R_TOUCH is set to the touch flag r_touch of the DSP 12, and the process returns to the main routine shown in FIG.
[0051]
FIG. 7 is a waveform diagram showing an example of a waveform stored in the waveform memory.
[0052]
The horizontal axis of FIG. 7 indicates the address of the waveform memory.
[0053]
The waveform stored in the waveform memory is obtained by analyzing the periodic structure of the waveform in advance and setting a waveform section in which the waveform is divided every one period. The start addresses sadrs0, sadrs1, sadrs2,. And pitches pitch0, pitch1, pitch2,... Corresponding to the length of each waveform section. These waveform sections are examples of the reproduction range referred to in the present invention.
[0054]
FIG. 8 is a data structure diagram of the waveform memory 11 shown in FIG.
[0055]
The waveform memory is composed of a waveform-related data part 11a and a waveform data part 11b. The waveform-related data part 11a stores the start address wavestart and the final address waveend of the waveform data, and the waveform represented by the waveform data. Are alternately stored as start addresses sadrs0, sadrs1, sadrs2,... Representing the start position of each waveform section and pitches pitch0, pitch1, pitch2,. The waveform data portion 11b stores PCM waveform data.
[0056]
FIG. 9 is an equivalent circuit diagram showing the operation of the DSP 12 shown in FIG.
[0057]
FIG. 9 shows an equivalent circuit representing the operation of the DSP 12 and the waveform memory 11 described above.
[0058]
This equivalent circuit comprises position information generating means 20, start address generating means 21, reproduction pitch period generating means 22, window signal generating means 23, address change signal generating means 24, address generating means 25, reading means 26, and window providing means 27. , And envelope applying means 28. Further, the compression / expansion amount tcomp, the touch flag r_touch, the offset offset, the shift amount p_shift, the gate flag gate, and the formant volume f_vr are input to the equivalent circuit from the CPU 13.
[0059]
Hereinafter, the operation of the equivalent circuit will be generally described, and then described in detail with reference to a flowchart.
[0060]
In this equivalent circuit, there are two playback channels, and waveforms are alternately read out of the waveform memory 11 for those playback channels. Thereafter, the read waveform is processed by each of the two playback channels and synthesized with each other to generate a playback waveform.
[0061]
The position information generation means 20 is a means for generating position information phase representing the position on the waveform, and depending on whether the value of the touch flag r_touch input from the CPU 13 is “0” or “1”, The value of the position information phase is obtained using the compression / decompression amount tcomp or offset offset, respectively. In addition, the position information generating means 20 causes the reading means 26 to read out the waveform data start address wavestart and final address waveend from the waveform memory 11 and obtain them.
[0062]
The start address generating means 21 causes the reading means 26 to read out the start addresses sadrs0, sadrs1, sadrs2,... Of each waveform section and the pitches pitch0, pitch1, pitch2,. The start address sadrs corresponding to the position information phase generated by the position information generating means 20 is selected and output in response to the rise of the gate flag gate or the input of the pulse signals sp1 and sp2, and corresponds to the start address sadrs. It is also means for outputting a pitch pitch.
[0063]
Based on the shift amount p_shift input from the CPU 13 and the pitch pitch output from the start address generating unit 21, the playback pitch cycle generation unit 22 generates a playback pitch pitch indicating the cycle of the playback sound, and plays back the playback pitch. This is means for generating pulse signals sp1 and sp2 indicating the read timing at which each of the two reproduction channels described above reads out the waveform piecewise in accordance with the pitch.
[0064]
The window signal generation means 23 receives the pulse signals sp1 and sp2 generated by the reproduction pitch period generation means 22 or the rising edge of the gate flag gate input from the CPU 13, and receives the reproduction generated by the reproduction pitch period generation means 22. Triangular waveform window functions window1 and window2 corresponding to the pitch pitch, the pitch pitch output by the start address generator 21 and the formant volume f_vr input from the CPU 13 are generated, and the rise or fall of the gate flag gate is generated. This is a means for receiving an input and generating an envelope function env that naturally starts and stops the reproduced sound.
[0065]
The address change signal generating means 24 receives the pulse signals sp1 and sp2 generated by the reproduction pitch period generating means 22, and receives sawtooth address change signals pph1 and pph2 corresponding to the formant volume f_vr input from the CPU 13. It is a means to generate.
[0066]
Based on the start address sadrs selected and output by the start address generation means 21 and the address change signals pph1 and pph2 generated by the address change signal generation means 24, the address generation means 25 actually stores each reproduction channel in the waveform memory. 11 is a means for generating read addresses adrs1 and adrs2 for reading waveform data from 11.
[0067]
The reading means 26 is means for reading the waveform data data1 and data2 for each reproduction channel from the waveform memory 11 in accordance with the read addresses adrs1 and adrs2 generated by the address generating means 25. However, since the address generation means 25 also generates decimal values as the read addresses adrs1 and adrs2, the read means 26 generates waveform data data1 and data2 corresponding to the read values adrs1 and adrs2 of decimal values using an interpolation technique. To do.
[0068]
The window providing means 27 causes the window functions window 1 and window 2 generated by the window signal generating means 23 to act on the waveform data data 1 and data 2 for each reproduction channel read from the waveform memory 11 by the reading means 26. A means for synthesizing the waveform data of these reproduction channels with each other.
[0069]
The envelope applying means 28 is a means for generating a reproduced waveform signal by applying the envelope function env generated by the window signal generating means 23 to the waveform data synthesized by the window applying means 27.
[0070]
FIG. 10 is a block diagram of the window signal generating means 23 shown in FIG.
[0071]
The window signal generating means 23 includes a window function control means 231, a first window function generating means 232, a second window function generating means 233, and an envelope generating means 234. The second window function generating means 232 is means for generating the window functions window 1 and window 2 for the two playback channels described above, and the envelope generating means is means for generating the envelope function env.
[0072]
The window function control unit 231 receives the pulse signals sp1 and sp2, the gate flag gate, the reproduction pitch pitch, the formant volume f_vr, and the pitch pitch, and receives the first window function generation unit 232 and the second window function. This is a means for controlling the generating means 233 and the envelope generating means 234.
[0073]
The equivalent circuit shown in FIG. 9 reproduces the waveform according to the value of the position information sphase indicating the position on the waveform. When the user lifts his / her finger from the ribbon controller 16, the position indicated by the position information “sphase” moves on the waveform at a moving speed corresponding to the compression / decompression amount specified by the operator group 18. On the other hand, when the user touches the ribbon controller 16, the position information “phase” indicates a position on the waveform corresponding to the position where the user touches the ribbon controller 16.
[0074]
Hereinafter, changes in variable values and the like in the case where the position indicated by the position information phase moves on the waveform at a moving speed corresponding to the compression / decompression amount will be described with reference to the waveform diagrams.
[0075]
FIG. 11 is a waveform diagram for explaining the operation when the waveform stored in the waveform memory 11 is reproduced without being compressed or expanded in the time axis direction. Note that the playback pitch is processed so as to be slightly lower than the pitch of the stored original waveform. In this case, a reproduced sound having the same length as the length of a series of sampled sounds is emitted.
[0076]
The horizontal axis of each waveform diagram shown in FIG. 11 indicates time.
[0077]
FIG. 11H shows an example of the value of the gate flag. The value “1” is displayed when the key on the keyboard 17 shown in FIG. 1 is pressed, and the value “0” is displayed when all keys are released. "
[0078]
FIG. 11G shows the envelope function env, which rises gently when the gate flag rises and falls gently when the gate flag falls.
[0079]
FIG. 11A shows the time change of the position indicated by the position information sphase, and the position indicated by the position information sphase sequentially passes through the start addresses sadrs0, sadrs1, sadrs2,. .
[0080]
FIG. 11B shows the start address corresponding to the head of the waveform section including the position indicated by the position information sphase among the start addresses sadrs0, sadrs1, sadrs2,. As described above, the start address generating means 21 shown in FIG. 9 receives and inputs the pulse signals sp1, sp2, etc., and outputs the start address. The start address shown in FIG. Of these, the start address sadrs corresponding to the selected time is selected and output.
[0081]
FIG. 11C shows pulse signals sp1 and sp2 that are generated with a period twice as high as the reproduction pitch, and each of the two reproduction channels fragments a waveform based on the timing of each pulse signal. Read out automatically. This fragmentary reading is started from the start address sadrs selected by the start address generating means 21 at the timing of each pulse signal, and the waveform data for about two waveform sections are read.
[0082]
FIG. 11D shows address change signals pph1 and pph2 that represent the read address of the waveform data by the amount of displacement from the start address sadrs. The waveforms of the address change signals pph1 and pph2 are the pulse signals described above. A sawtooth wave is reset to the value “0” at each rise of sp1 and sp2.
[0083]
FIG. 11E shows waveform data data1 and data2 that are read out piecewise for each reproduction channel, and also shows the start address at which the waveform reading is started. Here, since the compression / expansion in the time axis direction is not performed, the start address changes at the same pace as the temporal change of the original waveform data, and changes sequentially. However, since the reproduction pitch is reproduced so as to be slightly lower than the pitch of the original waveform, the waveform data from the start address sadrs8 is skipped.
[0084]
FIG. 11 (F) shows window functions window1 and window2 for each reproduction channel. The triangular wave window functions window 1 and window 2 shown here are multiplied by the waveform data data 1 and data 2 for each reproduction channel, the waveform data after multiplication for each reproduction channel are added together, and the envelope function env is further multiplied. To generate a reproduced waveform.
[0085]
FIG. 12 is a waveform diagram for explaining the operation when the waveform is compressed and reproduced in the time axis direction. However, as in FIG. 11, the reproduction pitch is reproduced so as to be slightly lower than the pitch of the original waveform.
[0086]
FIG. 12A shows a time change of the position indicated by the position information “phase”. Here, a time change faster than the time change shown in FIG. 11A is shown. For this reason, the position indicated by the position information sphase in FIG. 12A passes through the start addresses sadrs0, sadrs1, sadrs2,... At a timing earlier than the timing shown in FIG. As a result, the change timing of the start address sadrs shown in FIG. 12B is earlier than the change timing of the start address sadrs shown in FIG.
[0087]
The pulse signals sp1 and sp2 shown in FIG. 12C and the address change signals pph1 and pph2 shown in FIG. 12D reproduce the same playback pitch as that shown in FIG. 11, and the pulses shown in FIG. These are the same signals as the signals sp1 and sp2 and the address change signals pph1 and pph2 shown in FIG.
[0088]
As described above, since the change timing of the start address sadrs shown in FIG. 12B is earlier than the change timing of the start address sadrs shown in FIG. 11B, the read address is read according to the pulse signals sp1 and sp2. The start address sadrs3, the start address sadrs6, and the start address sadrs10 are skipped at the start address where the reading of the fragmented waveform data data1 and data2 shown in FIG. That is, since the waveform data stored in the waveform memory is read out with skipping of the waveform sections in some places, the reproduction time of the entire waveform is shortened by the skipping amount.
[0089]
FIG. 13 is a waveform diagram for explaining the operation when the waveform is reproduced while being expanded in the time axis direction. Here, as in FIG. 11, the reproduction pitch is reproduced so that it is slightly lower than the pitch of the original waveform.
[0090]
FIG. 13A shows a time change of the position indicated by the position information “phase”. Here, a time change slower than the time change shown in FIG. 11A is shown. For this reason, the position indicated by the position information phase passes through the start addresses sadrs0, sadrs1, sadrs2,... At a timing later than the timing shown in FIG. The change timing of the start address sadrs shown in FIG. 13B is later than the change timing of the start address sadrs shown in FIG.
[0091]
Similarly to the above, the pulse signals sp1 and sp2 shown in FIG. 13C and the address change signals pph1 and pph2 shown in FIG. 13D reproduce the same playback pitch as in FIG. This is the same signal as the pulse signals sp1 and sp2 shown in FIG. 11C and the address change signals pph1 and pph2 shown in FIG.
[0092]
Contrary to the occurrence of skipping at the start address where the reading of the waveform data data1 and data2 shown in FIG. 12 (E) is started, the start address shown in FIG. 13 (E) includes the start address sadrs0 and the start address sadrs6. Overlap occurs, and the reproduction time of the entire waveform becomes longer by the overlap.
[0093]
As described above, when the position indicated by the position information phase moves on the waveform at a moving speed corresponding to the compression / expansion amount, the waveform is compressed and expanded in the time axis direction and reproduced.
[0094]
On the other hand, when the position information “phase” indicates the position on the waveform corresponding to the position where the user touches the ribbon controller 16, the time-axis compression / decompression technique of the waveform is used as it is and includes the designated position. The waveform section is repeatedly read and reproduced with a natural tone. When the position touching the ribbon controller 16 is changed, the position indicated by the position information sphase on the waveform is also moved, and the waveform section to be reproduced is changed to a waveform section including the position indicated by the current position information sphase. To do.
[0095]
FIG. 14 is a flowchart of the main routine of the DSP 12 shown in FIG.
[0096]
This main routine is a routine that is repeatedly executed at a predetermined sampling period.
[0097]
In this main routine, the position information generating means 20, the start address generating means 21, the reproduction pitch period generating means 22, the window signal generating means 23, the address change signal generating means 24, and the address generating means 25 constituting the equivalent circuit shown in FIG. The operations of the reading unit 26, the window applying unit 27, and the envelope applying unit 28 are sequentially executed (step S20 to step S28). Finally, the calculated waveform signal is output from the DSP 12 (step S29).
[0098]
However, step S23 in which the operation of the window signal generation means 23 is executed is the same as step S23_1 in which the operation of the window function control means 231 shown in FIG. 10 is executed, the first window function generation means 232, and the second window function. The generation unit 233 and the operation of the envelope generation unit 234 are configured in step S23_2.
[0099]
Of the steps shown in the flowchart of FIG. 14, the steps marked with the symbol “*” will be described in detail below.
[0100]
FIG. 15 is a flowchart of a routine for executing the operation of the position information generating means 20 shown in FIG.
[0101]
When this routine is started, it is first determined whether or not the value of the gate flag gate has risen (step S20_1). If it is determined that it is at a time other than immediately after the rise, the touch flag r_touch is determined. Is determined to be a value “0” indicating that the user's finger is away from the ribbon controller (step S20_2).
[0102]
If it is determined that the value of the touch flag r_touch is “0”, the user's finger is away from the ribbon controller, and the value of the position information phase is incremented by the compression / decompression amount tcomp (step S20_3).
[0103]
Thereafter, if the value of the position information “phase” is smaller than the start address “wavestart” (step S20_4: Y), the start address “wavestart” is substituted as the value of the position information “phase” (step S20_5), and the value of the position information “phase” is the final address waveend. If greater than (step S20_6: Y), the final address waveend is substituted as the value of the position information sphase (step S20_7). Accordingly, the position indicated by the position information “phase” is limited so that it always exists on the waveform.
[0104]
After that, the position information “phase” is output (step S20_8), and the process returns to the DSP main routine shown in FIG.
[0105]
In step S20_1 described above, when it is determined that the value of the gate flag gate has just risen (step S20_1: Y), the start address wavestart is substituted as the position information phase value (step S20_5). It is determined that the value of “phase” is equal to or smaller than the final address “waveend” (step S20_6: N), the position information “phase” is output (step S20_8), and the process returns to the main routine of the DSP shown in FIG.
[0106]
If it is determined in step S20_2 described above that the value of the touch flag r_touch is a value “1” indicating that the user's finger is touching the ribbon controller (step S20_2: N), the touch is performed. It is determined whether or not the value of the flag r_touch is immediately after rising from the value “0” to the value “1” (step S20_9). If it is determined that it is immediately after the start-up, that is, immediately after the user's finger touches the ribbon controller, the current position as the value of the touch point touchp indicating the position on the waveform reproduced at the time of the touch. The value of information phase is substituted (step S20_10), and the value of the current offset cur_offset indicating the amount of movement from the position indicated by the touch point touchp is reset (step S20_11).
[0107]
Thereafter, the value of the current offset cur_offset is updated so as to gradually approach the value of the offset offset corresponding to the movement amount of the touch position on the ribbon controller (step S20_12). That is, the current offset cur_offset is incremented by a value obtained by multiplying the difference between the current offset cur_offset and the offset offset by the time constant K. The reason why the offset offset value is not directly assigned to the current offset cur_offset is to prevent the playback sound from becoming unnatural due to a large jump in the playback position when the user suddenly moves his / her finger. .
[0108]
When the update of the current offset cur_offset in step S20_12 is finished, the value of the sum of the touch point touchp and the current offset cur_offset is substituted for the position information phase (step S20_13). The position indicated by is always limited to be present on the waveform, the position information “phase” is output (step S20_8), and the process returns to the DSP main routine shown in FIG.
[0109]
FIG. 16 is a flowchart of a routine for executing the operation of the start address generating means 21 shown in FIG.
[0110]
When this routine is started, it is first determined whether or not any one of the pulse signal sp1, the pulse signal sp2 and the rising edge of the gate flag gate has been input (step S21_1), and it is determined that it has been input. In this case, a waveform section including the position indicated by the position information sphase is selected based on a series of start addresses sadrs0, sadrs1, sadrs2,... Read from the waveform memory 11 shown in FIG. The start address and the pitch are set as the selection output start address sadrs and pitch pitch (step S21_2). That is, as long as the position indicated by the position information “phase” does not move beyond the waveform section, one waveform section is repeatedly reproduced.
[0111]
Thereafter, the start address sadrs is output (step S21_3), the pitch pitch is output (step S21_4), and the process returns to the main routine of the DSP shown in FIG.
[0112]
FIG. 17 is a flowchart of a routine for executing the operation of the reproduction pitch period generating means 22 shown in FIG.
[0113]
When this routine is started, it is first determined whether or not the value of the gate flag gate has risen (step S22_1). If it is determined that it is at a time other than immediately after the rise, a sawtooth wave is generated. The value of the control variable “phase” to be represented is incremented by the progressive value “phinc” (step S22_2). Then, it is determined whether or not the value of the control variable phase has reached the maximum value pmax indicating that the value has been continuously increased for a time corresponding to the pitch pitch of the reproduced sound (step S22_3), and it is determined that it has not been reached. Then, the routine of FIG. 17 is finished as it is, and the process returns to the main routine of the DSP shown in FIG.
[0114]
When it is determined that the value of the control variable phase has reached the maximum value pmax (step S22_3: Y), the value of the control variable phase is subtracted by the maximum value pmax (step S22_4), and the two playback channels described above The value of the flag ff_flg indicating the difference between the value “1” and the value “−1” is determined (step S22_5). When the value of the flag ff_flg is a negative value “−1”, the pulse signal sp2 is output and the value “1” is substituted for the value of the flag ff_flg (step S22_6), and the value of the flag ff_flg is positive. When the value is “1”, the pulse signal sp1 is output and the value “−1” is substituted for the value of the flag ff_flg (step S22_7). As a result, the pulse signals sp1 and sp2 of the two reproduction channels are output alternately.
[0115]
Thereafter, the value of the product of pitch pitch selected and output by the start address generating means and shift amount p_shift set by the CPU is substituted as the pitch pitch value of the reproduced sound (step S22_8). The value is converted into a progressive value phinc in a predetermined table (step S22_9), and the process returns to the DSP main routine shown in FIG.
[0116]
If it is determined in step S22_1 described above that the value of the gate flag gate has just risen, the value of the control variable phase is reset to “0”, the pulse signal sp1 is output, and the value of the flag ff_flg is set. The value “−1” is substituted (step S22_7). Then, similarly to the above, the value of the pitch pitch of the reproduced sound is set (step S22_8), the value of the pitch pitch is converted into the progressive value phinc (step S22_9), and the process returns to the DSP main routine shown in FIG.
[0117]
FIG. 18 is a flowchart of a routine for executing the operation of the window function control means 231 shown in FIG.
[0118]
Also when this routine is started, it is first determined whether or not the value of the gate flag gate has risen (step S23_1_1). If it has just been risen, the first and second shown in FIG. The stop instruction signal Wstop for stopping the generation of the window function by the window function generation means 232 and 233 and the envelope start signal Gstart for instructing the envelope generation means 234 to rise of the envelope function env are output (step S23_1_2). It is also determined whether or not the value of the gate flag gate has just fallen (step S23_1_3). If it is just after the fall, the envelope stop signal Gstop for instructing the envelope generating means 234 to fall the envelope function env. Is output (step S23_1_4).
[0119]
Thereafter, it is determined whether or not the pulse signal sp1 is input (step S23_1_5). If it is determined that the pulse signal sp2 is not input, it is determined whether or not the pulse signal sp2 is input (step S23_1_13). When it is determined that the pulse signal sp2 has not been input, the process returns to the DSP main routine shown in FIG.
[0120]
When it is determined that the pulse signal sp1 is input (step S23_1_5: Y), function generation control is performed for the window function window1 for the first reproduction channel. First, the smaller value of the value obtained by multiplying the pitch pitch by the formant volume f_vr and the value of the pitch pitch of the reproduced sound is determined (step S23_1_6), and the reciprocal of the smaller value is the triangular wave of the window function window1 Is substituted as a slope wp1 (step S23_1_7, step S23_1_8). Then, the value of the slope wp1 and the start instruction signal st1 instructing the start of window function generation are output to the first window function generating means 232 (WG1) shown in FIG. 10 (step S23_1_9).
[0121]
Thereafter, it is determined whether or not the slope wp1 obtained here is larger than the current value of the slope wp2 for the window function window2 (step S23_1_10), and if it is determined that the slope wp1 is larger, Since there is a possibility that the pulse signal sp2 is input before the triangular wave of the window function window2 is lowered, the value of the slope wp1 is substituted for the slope wp2 (step S23_1_11), and the value of the slope wp2 is the second value. Is output to the window function generating means 233 (WG2) (step S23_1_12).
[0122]
In exactly the same manner as in steps S23_1_5 to S23_1_12 described above, in step S23_1_13 to step S23_1_20, function generation control is performed for the window function window2 for the second reproduction channel.
[0123]
Thereafter, the process returns to the DSP main routine shown in FIG.
[0124]
FIG. 19 is a flowchart of a routine for executing the operation of the address change signal generator 24 shown in FIG.
[0125]
When this routine is started, first, it is determined whether or not the pulse signal sp1 is input (step S24_1). If it is determined that the address has been input, the address change ph1 with respect to the start address sadrs described above is reset to the value “0” (step S24_2). If it is determined that the address has not been input, the address change ph1 is set. The step is advanced (step S24_3). It is also determined whether or not the pulse signal sp2 has been input (step S24_4). If it is determined that the pulse signal sp2 has been input, the address change ph2 is reset to the value “0” (step S24_5) and input. If it is determined that there is no address, the address change ph2 is incremented (step S24_6).
[0126]
Thereafter, the product of the address change amount ph1, ph2 and the formant volume f_vr is substituted as the value of the address change signal pph1, pph2 (step S24_7), and the address change signal pph1, pph2 is output (step S24_8). Return to the main routine of the DSP shown in FIG.
[0127]
FIG. 20 is a flowchart of a routine for executing the operation of the address generating means 25 shown in FIG.
[0128]
Even when this routine is started, it is first determined whether or not the pulse signal sp1 is input (step S25_1). If it is determined that it has been input, the value of the start address sadrs currently selected and output by the above-described start address generating means is substituted for the start address start1 for the first reproduction channel (step S25_2). ). It is also determined whether or not the pulse signal sp2 has been input (step S25_3). If it is determined that the pulse signal sp2 has been input, the value of the start address sadrs is substituted into the start address start2 for the second playback channel ( Step S25_4).
[0129]
Thereafter, as the read addresses adrs1 and adrs2 for reading the waveform data from the waveform memory 11, the sum of the start addresses start1 and start2 for each reproduction channel and the address change signals pph1 and pph2 is substituted, and the read addresses adrs1 and adrs2 Is output to return to the DSP main routine shown in FIG.
[0130]
Above, description of embodiment shown in FIG. 1 is complete | finished.
[0131]
The position specifying operator referred to in the present invention is not limited to the ribbon controller provided in the above-described embodiment, and may be an operator described below.
[0132]
FIG. 21 is a diagram showing another example of the position specifying operator according to the present invention.
[0133]
The operation element 30 is an operation element that is rotatable in the C direction and the D direction shown in FIG. Further, the operation element 30 is biased in the B direction and can be pushed in the A direction.
[0134]
The operation in which the operating element 30 is pushed in the A direction corresponds to the operation in which the ribbon controller is touched, and the operation in which the operating element 30 returns to the B direction by urging in the B direction is an operation in which the finger is released from the ribbon controller. It corresponds to. Further, the operation of rotating the operation element 30 in the C direction or the D direction while being pushed in the A direction corresponds to the operation of rubbing the ribbon controller.
[0135]
Further, the position specifying operator referred to in the present invention may be a rotary operator shaped like a turntable of a record player. As a means for turning on / off the effect of position designation by such a rotary operation element, for example, a means for performing on / off control by detecting that pressure has been applied to the rotation operation can be considered.
[0136]
Further, in the method of assigning a part of the waveform to the operation unit of the ribbon controller as in the ribbon controller described above, the position of the waveform other than the assigned waveform part cannot be indicated, so the operation amount to the ribbon controller is accumulated. The position of the waveform may be indicated by calculation. That is, the accumulated amount indicates the position of the waveform. In addition, even with a rotary operator shaped like a turntable as described above, using an endlessly rotatable operator such as a rotary encoder, the operability can be improved by indicating the position of the waveform by accumulating the scanning amount. Good.
[0137]
In the above-described embodiment, the pitch of the reproduced sound depends only on the keyboard operation. However, the waveform reproducing device of the present invention has a pitch according to the change speed of the position specified by the position specifying operator. The waveform may be reproduced. The embodiment in which the waveform is reproduced with the pitch corresponding to the change speed of the designated position can be realized by partially changing the flowchart of the above-described embodiment.
[0138]
Hereinafter, the changed part of the flowchart will be described.
[0139]
FIG. 22 is a diagram showing a changed portion of the flowchart of the ribbon controller processing routine.
[0140]
Step S13_5a shown in FIG. 22 is a step executed in place of step S13_5 shown in FIG.
[0141]
In step S13_5 shown in FIG. 6, the touch position is only detected and set to the current position POINT, whereas in step S13_5a shown in FIG. 22, the touch position detected last time and this time are further detected. The operation speed is calculated based on the difference from the touched position, set to the CPU speed V, and set to the DSP speed v.
[0142]
FIG. 23 is a diagram showing a part changed from the flowchart for executing the operation of the reproduction pitch period generating means.
[0143]
Step S22_8a shown in FIG. 23 is a step executed in place of step S22_8 shown in FIG.
[0144]
In step S22_8 shown in FIG. 17, the value of the product of the pitch pitch of the waveform section and the shift amount p_shift is substituted as the pitch pitch value of the reproduced sound. In step S22_8a shown in FIG. As the pitch value, a value obtained by multiplying the product of the pitch pitch of the waveform section and the shift amount p_shift by the ratio between the compression / decompression amount tcomp and the speed v is substituted.
[0145]
As described above, by partially changing the flowcharts shown in FIGS. 6 and 17, waveform reproduction at a pitch corresponding to the moving speed of the designated position is realized. In the waveform reproducing apparatus that performs such a reproducing operation, it is possible to perform a performance similar to a performance called “scratch” in which an analog record is moved by hand to make a sound of “ku”.
[0146]
The ribbon controller and the operation element as described above are integrated with a means for turning on / off the position designation effect. However, the position designation operation element according to the present invention has an ON state and an OFF state, which are It may be switched with another operation element.
[0147]
In the above embodiment, the waveform is reproduced according to the value of the formant volume f_vr specified by the operator for specifying the formant volume f_vr. However, the waveform reproducing apparatus of the present invention uses the value of the formant volume f_vr described above as the designated position. The waveform may be reproduced by changing to a value corresponding to the moving speed.
[0148]
The waveform section described above is one in which the waveform is divided every one period, but is not limited to one period, and may be divided every plural periods. In addition, the reproduction range referred to in the present invention is not limited to the waveform section divided in advance in this way, and may be, for example, a range from a specified position on the waveform to two cycles ahead.
[0149]
Furthermore, the technique used for realizing the waveform reproduction apparatus of the present invention is not limited to the above-described time axis compression / decompression technique.
[0150]
【The invention's effect】
As described above, according to the waveform reproduction device of the present invention, the user can directly designate the reproduction position on the waveform and can continuously change the reproduction position. As a result, rich music expression can be realized.
[Brief description of the drawings]
FIG. 1 is a diagram showing an embodiment of a waveform reproducing device of the present invention.
FIG. 2 is an explanatory diagram of a position designation operation using a ribbon controller.
FIG. 3 is a flowchart of a main routine of a CPU.
FIG. 4 is a flowchart of an operator group processing routine.
FIG. 5 is a flowchart of a keyboard processing routine.
FIG. 6 is a flowchart of a ribbon controller processing routine.
FIG. 7 is a waveform diagram showing an example of a waveform stored in a waveform memory.
FIG. 8 is a data structure diagram of a waveform memory.
FIG. 9 is an equivalent circuit diagram showing the operation of the DSP.
FIG. 10 is a configuration diagram of a window signal generating means.
FIG. 11 is a waveform diagram for explaining the operation when the waveform stored in the waveform memory is reproduced without being compressed or expanded in the time axis direction.
FIG. 12 is a waveform diagram for explaining the operation when the waveform is compressed and reproduced in the time axis direction.
FIG. 13 is a waveform diagram for explaining the operation when the waveform is reproduced by being expanded in the time axis direction.
FIG. 14 is a flowchart of a DSP main routine.
FIG. 15 is a flowchart of a routine for executing the operation of the position information generating means.
FIG. 16 is a flowchart of a routine for executing the operation of the start address generating means.
FIG. 17 is a flowchart of a routine for executing the operation of the reproduction pitch period generation means.
FIG. 18 is a flowchart of a routine for executing the operation of the window function control means.
FIG. 19 is a flowchart of a routine for executing the operation of the address change signal generating means.
FIG. 20 is a flowchart of a routine for executing the operation of the address generating means.
FIG. 21 is a diagram showing another example of the position specifying operator according to the present invention.
FIG. 22 is a diagram showing a changed part of the flowchart of the ribbon controller processing routine.
FIG. 23 is a diagram showing a change to the flowchart for executing the operation of the reproduction pitch period generation means.
[Explanation of symbols]
10 Waveform playback device
11 Waveform memory
11a Waveform related data part
11b Waveform data part
12 DSP
13 CPU
14 RAM
15 ROM
16 Ribbon controller
17 keyboard
18 controls
19 D / A converter
20 Location information generating means
21 Start address generation means
22 Reproduction pitch period generating means
23 Window signal generating means
24 Address change signal generating means
25 Address generation means
26 Reading means
27 Window giving means
28 Envelope giving means
30 operator
231 Window function control means
232 First window function generating means
233 Second window function generating means
234 Envelope generating means

Claims (3)

Waveform storage means for storing a series of waveforms consisting of a plurality of waveform sections;
There is an on state and an off state, and a position specifying operator that outputs the position being operated when in the on state,
Means for outputting position information representing the position on the waveform stored in the waveform storage means, and when the position specifying operator is in an off state, the position specifying operator is moved at a predetermined speed; Position information generation that moves the position information to a position corresponding to the amount of operation from the reference position when the position specifying operator is in the on state, with the position operated when the is switched from the off state to the on state as the reference position Means,
Waveform section selection means for selecting a waveform section including a position on the waveform represented by the position information;
A waveform reproduction apparatus comprising: reproduction means for repeatedly reproducing a waveform of a selected waveform section at a desired speed.   2. The waveform reproducing apparatus according to claim 1, wherein the reproducing means reproduces a waveform at a speed corresponding to a designated pitch.   2. The waveform reproduction apparatus according to claim 1, wherein the reproduction means reproduces a waveform at a speed corresponding to an operation speed for the position specifying operator.

Images