JP5509849B2 - Musical sound generating apparatus and program - Google Patents

Description

  The present invention relates to a tone generation system that generates a tone signal to which a modulation effect such as vibrato is applied in a bowed instrument such as a violin.

  2. Description of the Related Art Conventionally, various musical sound generating apparatuses such as electronic musical instruments have been proposed for generating a musical sound signal (vibrato musical sound signal) to which a modulation effect such as vibrato in a bowed musical instrument such as a violin is given. For example, the electronic musical instrument disclosed in Patent Document 1 generates a plurality (two series) of tone signals having different waveforms and periodically changing the pitch according to the vibrato signal, and each tone signal is modulated by a modulation signal synchronized with the vibrato signal. After weighting, by synthesizing them and generating a tone, the pitch of the tone is changed according to the vibrato signal, and the weighting ratio for each tone signal is changed according to the vibrato signal. A musical tone with changing tone is obtained.

  Further, Patent Document 2 proposes a method of creating two series of waveform data used to generate such two series of vibrato musical sound signals, and frequency analysis is performed on the waveform data played by vibrato performance. Then, upper and lower waveform data are created based on the maximum value (upper point) and the minimum value (upper point) of the pitch change (FIGS. 2 to 10), respectively. In addition, based on the created upper and lower waveform data, a vibrato musical tone is generated by a synthesis method similar to that of Patent Document 1. That is, the upper and lower waveform data are read at a common speed, and the weighting of both waveform data is varied while oscillating the read speed in accordance with the triangular vibrato control signal output from the low frequency oscillator (LFO) (FIG. 11, FIG. 11). 12).

JP 60-90391 A JP 2007-248551 A JP 2000-122665 A

  However, in the musical tone generation method using two vibrato musical tone signal sequences as in the above-described conventional example, when vibrato is not applied, musical tone is synthesized by synthesizing two extreme musical sound waveforms whose tone color and volume change inversely. Therefore, the sound quality is poor compared to the performance sound of a bowed instrument such as a violin, and a realistic bowed instrument performance sound cannot be obtained. In addition, in a stringed instrument such as a violin, the performance sound when vibrato is sufficiently applied has a unique “sound”, but when vibrato is sufficiently applied by the conventional musical sound generation method, such friction is applied. Musical sounds with the “sound” characteristic of vibrato performance of stringed instruments are not generated, and it is impossible to obtain realistic stringed instrument sounds.

  In view of such circumstances, the present invention provides a third musical tone signal generation sequence in addition to two musical tone signal generation sequences for imparting a timbre modulation effect with vibrato or the like. An object of the present invention is to provide a musical sound generation system capable of realizing realistic performance sounds.

  According to the main feature of the present invention, the musical sound for generating the first, second and third musical sound waveform signals (Wp1, Wf1; Wp2, Wf2; Wp3, Wf3) corresponding to one sounding instruction (note-on). Designated by waveform generation means (RP: RP1 to RP3, RF: RF1 to RF3), pitch shift designation means (PS) for designating a pitch shift value [Ps (Lf, Vd)], and pitch shift designation means (PS) Based on the pitch shift value (Ps), the first, second and third musical sound waveform signals (Wp1, Wf1; generated by the musical sound waveform generating means (RP1, RF1; RP2, RF2; RP3, RF3); Pitch changing means (PG) for changing the pitch (Pi) of Wp2, Wf2; Wp3, Wf3) and musical tone waveform generating means (RP1, RF1; RP2, RF2; RP3, RF3) A predetermined weight (Pu, Pd, Vd) is given to the first, second and third musical sound waveform signals (Wp1, Wf1; Wp2, Wf2; Wp3, Wf3) generated by the first, second and third, respectively. A tone signal generating means (EP: EP1 to EP3, EF: EF1 to EF3) for generating a third tone element signal (Sp1, Sf1; Sp2, Sf2; Sp3, Sf3), wherein the first and second tone For the waveform signals (Wp1, Wf1; Wp2, Wf2), when the pitch shift value (Ps) specified by the pitch shift specifying means (PS) increases, the weight for the first musical sound waveform signal (Wp1, Wf1) When (Pu) is increased and the weight (Pd) for the second musical sound waveform signal (Wp2, Wf2) is decreased, and the pitch shift value (Ps) decreases, the first musical sound waveform The weight (Pu) for the signal (Wp1, Wf1) is decreased and the weight (Pd) for the second musical sound waveform signal (Wp2, Wf2) is increased (EP1, EF1; EP2, EF2), and the third musical sound waveform signal As for (Wp3, Wf3), a predetermined weight (Vd) that does not depend on the increase / decrease of the pitch shift value (Ps) is given (EP3, EF3) music signal generation means (EP, EF), and music signal generation means (EP , EF), the first, second and third musical tone element signals (Sp1, Sf1; Sp2, Sf2; Sp3, Sf3) are mixed (mixed) to output the body part musical tone signal (Sb). In response to one tone generation instruction (note-on), the first, second, and third tone waveform signals (Wp1, Wp1, Musical sound waveform generation step (RP: RP1 to RP3, RF: RF1 to RF3) for generating Wf1; Wp2, Wf2; Wp3, Wf3) and a pitch shift specification step for specifying a pitch shift value [Ps (Lf, Vd)] (PS) and the pitch shift value (Ps) specified in the pitch shift specifying step (PS), the first generated in the musical sound waveform generating step (RP1, RF1; RP2, RF2; RP3, RF3), It is generated in a pitch change step (PG) for changing the pitch (Pi) of the second and third musical sound waveform signals (Wp1, Wf1; Wp2, Wf2; Wp3, Wf3) and a musical sound waveform generation step (RP, RF). The first, second and third musical sound waveform signals (Wp1, Wf1; Wp2, Wf2; Wp3, Wf3) are respectively given weights (Pu, Pd, Vd). Is a musical sound signal generation step (EP: EP1 to EP3, EF: EF1 to EF3) for generating first, second and third musical sound element signals (Sp1, Sf1; Sp2, Sf2; Sp3, Sf3). For the first and second musical sound waveform signals (Wp1, Wf1; Wp2, Wf2), when the pitch shift value (Ps) designated in the pitch shift designation step (PS) increases, the first musical tone When the weight for the waveform signal (Wp1, Wf1) is increased and the weight for the second musical sound waveform signal (Wp2, Wf2) is decreased and the pitch shift value (Ps) decreases, the first musical sound waveform signal (Wp1) , Wf1) is decreased and the weight for the second musical sound waveform signal (Wp2, Wf2) is increased (EP1, EF1; EP2, EF2). The musical sound waveform signals (Wp3, Wf3) are given a predetermined weight independent of the increase / decrease of the pitch shift value (Ps) (EP3, EF3), and the musical sound signal generation step (EP, EF) The first, second and third musical tone element signals (Sp1, Sf1; Sp2, Sf2; Sp3, Sf3) generated by EP, EF) are mixed (mixed) to output a body part musical tone signal (Sb). A musical sound generation program (claim 5) for causing a computer to execute a procedure comprising a musical sound signal output step (MX) is provided. Note that the parentheses are reference symbols, terms and the like of the examples added for convenience of understanding, and the same applies to the following.

  The musical sound generating apparatus according to the present invention further comprises modulation means (ML) for designating the pitch modulation depth (Vd) by changing with time, and the pitch shift designating means (PS) is controlled by the modulation means (ML). A pitch shift value (Ps) having a designated pitch modulation depth (Vd) and a pitch modulation amount (Lf) that periodically changes at a predetermined time interval (T) is designated, and a tone signal generating means ( EP: EP1 to EP3, EF: EF1 to EF3) changes the weight for the third musical sound waveform signal (Wp3, Wf3) according to the pitch modulation depth (Vd) specified by the modulation means (ML). (Claim 2).

  In this musical sound generating device, the musical sound signal generating means (EP: EP1 to EP3, EF: EF1 to EF3) is the third when the pitch modulation depth (Vd) designated by the modulating means (ML) is zero. A predetermined weight (va, vb) is given to the musical tone waveform signal (Wp3, Wf3), and the third musical tone waveform signal (Wp3, Wf3) is increased as the pitch modulation depth (Vd) increases. ) Is reduced (claim 3).

  In this musical tone generator, the musical tone waveform generating means (RP: RP1 to RP3, RF: RF1 to RF3) includes first and second musical tone waveform data (Dp1, Df1; Dp2, Df2) having predetermined frequency characteristics. ), The first and second musical sound waveform signals (Wp1, Wf1; Wp2, Wf2) are generated, respectively, and the intermediate frequency characteristics of both musical sound waveform data (Dp1, Df1; Dp2, Df2) are obtained. The third musical sound waveform signal (Wp3, Wf3) is generated based on the third musical sound waveform data (Dp3, Df3), and the musical sound signal generating means (EP: EP1 to EP3, EF: EF1 to EF3) is modulated. As the modulation depth (Vd) specified by the means (ML) increases, the weight (Vd) for the third musical sound waveform signal (Wp3, Wf3) increases. To can be configured to [claim 4] as.

  In the tone generation system according to the main feature of the present invention (Claims 1 and 5), the first, second and third tone waveform signals (Wp1, Wf1; Wp2, Wp2, corresponding to one sounding instruction (note-on)) Wf2; Wp3, Wf3) is generated (RP: RP1-RP3, RF: RF1-RF3). At this time, a pitch shift value [Ps (Lf, Vd)] is designated (PS), and the pitch (Pi) of the first, second and third musical sound waveform signals (Wp1, Wf1; Wp2, Wf2; Wp3, Wf3). ) Is changed based on the designated pitch shift value (Ps) (PG). These first, second and third musical sound waveform signals (Wp1, Wf1; Wp2, Wf2; Wp3, Wf3) are given predetermined weights (Pu, Pd, Vd), respectively. 3 musical tone element signals (Sp1, Sf1; Sp2, Sf2; Sp3, Sf3) are generated (EP: EP1 to EP3, EF: EF1 to EF3), and the generated first, second and third musical tone element signals are generated. (Sp1, Sf1; Sp2, Sf2; Sp3, Sf3) are further mixed (mixed) to output a body part tone signal (Sb) (MX). Here, when the weight is given, when the designated pitch shift value (Ps) increases for the first and second musical sound waveform signals (Wp1, Wf1; Wp2, Wf2), the first musical sound waveform. The weight (Pu) for the signal (Wp1, Wf1) is increased and the weight (Pd) for the second musical sound waveform signal (Wp2, Wf2) is decreased. Conversely, the pitch shift value (Ps) is decreased. Sometimes, the weight (Pu) for the first musical sound waveform signal (Wp1, Wf1) is decreased and the weight (Pd) for the second musical sound waveform signal (Wp2, Wf2) is increased (EP1, EF1; EP2, EF2). ), The third musical sound waveform signal (Wp3, Wf3) is given a predetermined weight (Vd) that does not depend on the increase / decrease of the pitch shift value (Ps) (EP3, EF3). .

  That is, in a bowed instrument such as a violin, the performance sound when the vibrato is weakly applied or when the vibrato is sufficiently applied includes a constant timbre component regardless of the fluctuation of the pitch. In order to generate a timbre modulation effect by vibrato or the like, a first musical tone signal generation sequence (RP1-EP1, RF1-EF1) for generating a first musical tone element signal (Sp1, Sf1) and a second musical tone element signal ( Two musical sound signal generation sequences whose weight changes are contradictory to each other in response to increase / decrease in the pitch shift value (Ps) are provided as second musical sound signal generation sequences (RP2-EP2, RF2-EF2) for generating Sp2, Sf2). In addition to these two musical tone signal generation sequences, in addition to these two musical tone signal generation sequences, the pitch shift value (Ps) Special third tone signal generating sequence given predetermined weights independent (RP3-EP3, RF3-EF3) is provided in the reduction. Therefore, according to the present invention, by providing a third musical tone signal generation sequence in addition to two musical tone signal generation sequences for imparting a timbre modulation effect by vibrato or the like, a realistic performance sound in a bowed instrument can be obtained. Can be realized.

  In the musical sound generating apparatus according to the present invention (Claim 2), modulation means (ML) for changing the pitch modulation depth (Vd) with time is provided, and the pitch modulation specified by the modulation means (ML) is provided. A pitch shift value (Ps) having a component of depth (Vd) and a component of pitch modulation amount (Lf) that periodically changes at a predetermined time interval (T) is designated (PS), and modulation means In accordance with the pitch modulation depth (Vd) specified by (ML), the weight for the third musical sound waveform signal (Wp3, Wf3) is changed (EP, EF).

  That is, the component of the pitch shift value (Ps) includes not only the pitch modulation amount (Lf) that periodically changes at a predetermined time interval (T) but also the depth of pitch modulation specified by the modulation means (ML). The third musical tone element signal generated by the third musical tone signal generation sequence (RP3-EP3, RF3-EF3) for generating a musical tone having a constant timbre component that does not depend on the pitch variation but includes (Vd). The weights (Sp3, Sf3) change according to only the pitch modulation depth (Vd) from the modulation means (ML) (with a monotonically increasing or monotonically decreasing correlation). In this case, the first and second musical tone signal generation sequences (RP1-EP1, RF1-EF1; RP2-EP2, RF2-EF2) for generating a timbre modulation effect by vibrato or the like are generated. The weights of the second musical tone element signals (Sp1, Sf1; Sp2, Sf2) vary periodically according to the pitch shift value (Ps), but the average level is determined by the pitch modulation from the modulation means (ML). Change in response to depth (Vd). Therefore, according to the present invention, the modulation means assigns appropriate weights to the two tone signal generation sequences for applying a modulation effect such as vibrato and the tone signal generation sequence for obtaining a constant tone color tone. More realistic bowed instrument performance sounds.

  In the musical sound generating device according to the present invention (Claim 3), when the pitch modulation depth (Vd) designated by the modulation means (ML) is zero, the third musical sound waveform signal (Wp3, Wf3) is applied. When a predetermined weight (va, vb) is given and the pitch modulation depth (Vd) increases, the weight for the third musical sound waveform signal (Wp3, Wf3) decreases accordingly (EP: EP1-EP3, EF: EF1-EF3).

  In other words, in a bowed instrument such as a violin, the performance sound when vibrato is not yet applied includes a certain tone without periodic fluctuations, and such a constant tone even when the vibrato is weakly applied. And is called "non-vibrato sound". In the present invention, in order to obtain such a “non-vibrato sound”, the third musical sound signal (Wp3, Wf3) generated by the third musical sound signal generation sequence (RP3-EP3, RF3-EF3) is modulated. When the pitch modulation depth (Vd) from the means (ML) is zero, predetermined weights (va, vb) are given, and the pitch modulation depth (Vd) increases (monotonically decreases). Decrease the weight). In this case, the first and second musical tone signal generation sequences (RP1-EP1, RF1-EF1; RP2-EP2, RF2-EF2) for generating a timbre modulation effect by vibrato or the like have a pitch modulation depth. When (Vd) is zero, a zero weight value (Pu = Pd = 0) is assigned and the first and second musical element signals (Sp1, Sf1; Sp2, Sf2) are not generated, and pitch modulation is performed. When the depth (Vd) increases, the weights of both musical tone element signals (Sp1, Sf1; Sp2, Sf2) vary periodically according to the pitch shift value (Ps), but the average level is pitch-modulated. Increase in response to the depth (Vd). Therefore, according to the present invention, it is possible to satisfactorily reproduce a realistic bowed instrument performance sound when vibrato is not applied to a bowed instrument such as a violin and when vibrato is applied.

  In the musical sound generating apparatus according to the present invention (Claim 4), the first and second musical sound waveform signals (Wp1, Wf1; Wp2, Wf2) are first and second musical sound waveform data having predetermined frequency characteristics, respectively. The third musical sound waveform signal (Wp3, Wf3) is generated based on (Dp1, Df1; Dp2, Df2), while the third musical sound waveform signal (Wp3, Wf3) Generated based on the third musical sound waveform data (Dp3, Df3) having an intermediate frequency characteristic of Dp2, Df2) (RP3, RF3), and the weight for this is specified by the modulation means (ML) (EP3, EF3) as the depth (Vd) increases.

  In other words, when a vibrato is sufficiently applied to a bowed instrument such as a violin, it has a frequency characteristic that corresponds to almost the midpoint of a periodically changing pitch, and is called a “sounding” that does not involve periodic fluctuations in timbre components. In the present invention, in order to obtain such “sound”, the first and second musical tones in the third musical tone signal generation sequence (RP3-EP3, RF3-EF3) are included in the present invention. Third musical sound waveform data having a frequency characteristic intermediate between the first and second musical sound waveform data (Dp1, Df1; Dp2, Df2) used to generate the waveform signals (Wp1, Wf1; Wp2, Wf2). A third musical sound waveform signal (Wp3, Wf3) is generated based on (Dp3, Df3), and the weight for this is changed by the modulation means (ML). Depending on the depth (Vd) increases (with a correlation of monotonically increasing) increase (EP3, EF3). In this case, the first and second musical tone signal generation sequences (RP1-EP1, RF1-EF1; RP2-EP2, RF2-EF2) for generating a modulation effect such as vibrato are generated. Although the weights of the musical tone element signals (Sp1, Sf1; Sp2, Sf2) of the sound signal periodically change according to the pitch shift value (Ps), the pitch modulation depth (Vd) increases for the average level. Will be reduced accordingly. Therefore, according to the present invention, it is possible to satisfactorily reproduce a realistic bowed instrument performance sound when vibrato is sufficiently applied to a bowed instrument such as a violin.

[Other features]
In the musical sound generating system according to the present invention, the first, second and third musical sound waveform signals (Wp1, Wf1; Wp2, Wf2; Wp3, Wf3) are used to generate the first, second and third musical sound waveform signals, respectively. For the third musical sound waveform data (element waveform data Dp1, Df1; Dp2, Df2; Dp3, Df3), two sets each for strong sound (Df1; Df2; Df3) and weak sound (Dp1; Dp2; Dp3) Prepared (element storage unit ES), and according to this, the first, second and third musical sound signal generation sequences (RP1-EP1, RF1-EF1; RP2-EP2, RF2-EF2; RP3-EP3, RF3) -EF3) also for strong sound (RF1-EF1; RF2-EF2; RF3-EF3) and weak sound (RP1-EP1; RP2-EP, respectively) RP3-EP3) are provided, and the body part tone signal (Sb) is generated by mixing the tone element signals (element tone signals Sp1, Sf1; Sp2, Sf2; Sp3, Sf3) generated in a total of six series. It can be constituted as follows. In this case, in the first, second and third musical sound signal generation sequences, the distribution of weights (Df, Dp) is controlled (DF, DP) between strong sounds and weak sounds.

  In the musical sound generation system according to the present invention, in order to generate an attack part musical sound signal, attack part musical sound waveform data (attack element waveform data Da) is prepared (ES), and an attack part musical sound signal generation function ( RA) and an attack part musical sound signal generation sequence (RA-EA) including an attack part musical element signal generation function (EA) can be provided. Then, by mixing (mixing) the attack part tone signal (Sa) generated in this series with the body part tone signal (Sb), the entire tone signal (So) corresponding to one sounding instruction (note-on) is obtained. Can be generated.

1 shows a hardware configuration of a musical sound generating apparatus according to an embodiment of the present invention. 1 is a functional block diagram of a tone generation system according to one embodiment of the present invention. The figure for demonstrating the musical sound production | generation operation | movement in 1st Embodiment of this invention is shown. The figure for demonstrating the musical sound production | generation operation | movement in 2nd Embodiment of this invention is shown.

[Hardware configuration example]
FIG. 1 is an example of a hardware configuration of a musical sound generating apparatus according to an embodiment of the present invention. The musical sound generating apparatus according to one embodiment of the present invention uses an electronic music apparatus such as an electronic musical instrument, and includes a central processing unit (CPU) 1, a random access memory (RAM) 2, a read only memory (ROM) 3, and an external storage. Device 4, performance operation detection unit 5, modulation operation detection unit 6, panel operation detection unit 7, display unit 8, sound source and effect applying unit 9, MIDI interface (MIDII / F) 10, communication interface (communication I / F) 11 These elements 1 to 11 are connected to each other via a bus 12.

  The CPU 1 that controls the entire apparatus, together with the RAM 2 and the ROM 3, constitutes a data processing unit that executes various processes according to various control programs, and executes various processes using a clock by the timer 13 according to the various control programs. The RAM 2 is used as a work area for temporarily storing various data necessary for these processes, and the ROM 3 stores various control programs and various necessary data.

  The external storage device 4 includes a compact disk read only memory (CD-ROM), a flexible disk in addition to a built-in storage medium such as a hard disk (HD) and a rewritable nonvolatile semiconductor memory such as a flash memory. Includes various portable external storage media such as (FD), magneto-optical (MO) disk, digital multi-purpose disk (DVD), smart media (registered trademark), etc. Can be stored in the storage medium.

  The performance operation detection unit 5 can detect the operation of a performance operator such as a keyboard, and can introduce performance data corresponding to the detected performance operation into the data processing unit. The modulation operation detection unit 6 can detect an operation of a modulation operator such as a modulation wheel (ML), and can introduce modulation data corresponding to the detected modulation operation into the data processing unit. The panel operation detection unit 7 can detect the operation content of a panel operation element (setting operation element) such as a switch, and can introduce various corresponding setting information into the data processing unit. The display unit 8 controls lighting / display contents of various display elements and a display (display device) such as an LCD according to a command from the CPU 1 and performs display assistance for various operations.

  The sound source and effect applying unit 9 includes a sound source unit and an effect providing unit, and the sound source unit includes a musical sound waveform memory and a sound source processor. The tone generator processor is a musical sound waveform memory based on performance data and modulation data from the performance operation detection unit 5 and modulation operation detection unit 6 or automatic performance data read from the storage means 3 and 4 in accordance with a command from the CPU 1. The musical sound waveform data is read out from, and an audio musical signal representing the musical sound waveform is generated. The effect imparting unit imparts a predetermined effect to the musical sound signal generated by the sound source unit, and outputs it to the sound system 14 connected to the sound source and effect providing unit 9. The sound system 14 includes a D / A conversion unit, an amplifier, and a speaker, and generates a musical sound based on a musical sound signal output from the sound source and the effect applying unit 9.

  An external MIDI device MD having a MIDI music information processing function is connected to the MIDII / F10 in the same manner as this musical sound generating device, and MIDI performance data is exchanged between the musical sound generating device and the external MIDI device MD through the MIDII / F10. can do. A communication network CN such as the Internet or a local area network (LAN) is connected to the communication I / F 11, and a control program and various data are downloaded from an external distribution device SV etc. and stored in the external storage device 4 to generate this musical sound. Can be used in the device.

[Configuration of Music Generation System]
In the musical tone generating apparatus according to the embodiment of the present invention, when the musical tone signal having a bowed instrument tone color such as a violin is generated in the sound source and effect applying unit 9, one sound generation instruction based on performance data is used in order to correspond to the vibrato playing method. Corresponding to the above, a musical tone generation system is constructed that generates musical tone signals based on a plurality of characteristic element waveform data. FIG. 2 shows a functional block diagram of a tone generation system according to one embodiment of the present invention.

  In this musical tone generation system, a musical tone signal generation sequence including element waveform data reading units RA, RP, RF, envelope applying units EA, EP, EF, and a mixing unit MX is provided for the element waveform data storage unit ES. . Further, for control of each part of the musical tone signal generation sequence, a low frequency oscillator (LFO) LF, a modulation means ML, a pitch shift calculation unit PS, a pitch generation unit PG, a strong sound component and a weak sound component weight generation unit DF, DP, Pitch shift up and pitch shift down weight generation units SU, SD, etc. are provided.

  The element waveform data storage unit ES is provided in the musical tone waveform memory of the sound source and effect applying unit 9 to generate a musical tone signal corresponding to one note-on, the attack element waveform data Da, the first element waveform for weak sound Data Dp1, Weak second element waveform data Dp2, Weak sound third element waveform data Dp3, Strong sound first element waveform data Df1, Strong sound second element waveform data Df2, Strong sound third element waveform data Df3 7 types of element waveform data are stored.

  Here, the attack element waveform data Da is waveform data for generating the attack portion Sa in the musical sound output signal So finally output from the mixing unit MX in accordance with one note-on, and the dynamic value (Dy ), Multiple types of data are prepared. The remaining six types of element waveform data Dp1 to Dp3 and Df1 to Df3, excluding the attack element waveform data Da, are body portion waveform data for generating the body portion Sb of the musical tone output signal So, and include three types of waveform data. It consists of musical sound waveform data indicating characteristic timbres, which are called first, second and third elements, respectively. In this example, these three types of musical sound waveform data (first to third elements) are used for strong sounds and weak sounds in order to give a subtle difference in tone when playing strongly and weakly. Are prepared, and “p” or “P” in the reference symbol indicates a strong sound, and “f” or “F” indicates a weak sound.

  Of the element waveform data Dp1 to Dp3 and Df1 to Df3, the first element waveform data Dp1 and Df1 and the second element waveform data Dp2 and Df2 are prepared to generate a modulation effect of the timbre component in vibrato and the like. The instrument is vibrated with a stringed instrument such as a violin, and the performance sound is recorded and time-frequency analyzed, and the components extracted from the upper end (maximum part) and lower end (minimum part) of the pitch change are recorded. is there. For example, the first element waveform data Dp1 and Df1 are frequency components extracted at the upper end of the pitch change, and the waveform data when the vibrato is weakly played is recorded as “first element waveform data Dp1 for weak sound” and strongly The waveform data in the case of playing is recorded as “first element waveform data Df1 for strong sound”. On the other hand, the second element waveform data Dp2 and Df2 are frequency components extracted at the lower end of the pitch change. When the vibrato is weakly played, the second element waveform data Dp2 and Df2 is recorded as “second element waveform data Dp2 for weak sound”. When played, it is recorded as “second element waveform data Df2 for strong sound”.

  The third element waveform data Dp3 and Df3 are used to generate certain timbre components included in the performance sound when vibrato is not applied, when vibrato is applied weakly, or when vibrato is applied sufficiently. Prepared. (A) In the first embodiment, the third element waveform data Dp3 and Df3 are created by recording a performance sound that does not vibrate at all with a bowed instrument such as a violin. It is recorded as “3 element waveform data Dp3”, and when played strongly, it is recorded as “third element waveform data Df3 for strong sound”. On the other hand, (b) in the second embodiment, the third element waveform data Dp3 and Df3 are recorded as a performance sound when the vibrato is performed with a bowed instrument such as a violin and subjected to time-frequency analysis. Components extracted from characteristic portions other than the upper and lower ends of the pitch change are used. For example, the frequency characteristic at the time when the pitch rises or falls near the intermediate position in the pitch change when the pitch changes almost periodically is extracted and created. Also in this case, when the vibrato is played weakly, it is recorded as “third element waveform data Dp3 for weak sound”, and when played strongly, it is recorded as “third element waveform data Df3 for strong sound”.

  The element reading unit includes an attack element selection reading unit RA, a weak sound element reading unit RP, and a strong sound element reading unit RF. The weak sound element reading unit RP includes a weak first sound reading unit RP1 and a weak second sound reading unit. RP2 and a weak third element reading unit RP3, and the strong element reading unit RF includes a strong first element reading unit RF1, a strong second element reading unit RF2, and a strong third element reading unit RF3. Is provided. Each element reading unit RA, RP1 to RP3, RF1 to RF3 generates the corresponding predetermined element waveform data Da, Dp1 to Dp3, Df1 to Df3 from the element waveform data storage unit ES as shown in the figure. The corresponding musical tone waveform signals Wa, Wp1 to Wp3, Wf1 to Wf3 are generated by sequentially reading from the part PG at a speed according to the designated pitch Pi. Here, the pitch generation unit PG generates a pitch Pi from a note number Nt indicated by note-on information in the performance data and a pitch shift value (also referred to as pitch shift amount) Ps described later, and each element reading unit RA, RP1. -RP3, RF1-RF3 are designated.

  The attack element selection reading unit RA starts reading at the note-on timing in the performance data, and selects one from a plurality of prepared attack element waveform data Da according to the dynamics value Dy determined by the note-on velocity information or the like. Select and read. Further, when a predetermined time has elapsed since the reading of the attack element waveform data Da is started, the body part element waveform data Dp1 to Dp3, Df1 to Df3 are read by the body part element reading parts RP1 to RP3 and RF1 to RF3. It is.

  The envelope providing unit is composed of an attack element envelope providing unit EA, a weak element envelope providing unit EP, and a strong element envelope providing unit EF. 1 element envelope provision part EP1, 2nd element envelope provision part EP2 for weak sounds, and 3rd element envelope provision part EP3 for weak sounds, and the strong sound element envelope provision part EF 1 element envelope provision part EF1, 2nd element envelope provision part EF2 for strong sounds, and 3rd element envelope provision part EF3 for strong sounds are provided. As shown in the figure, the element envelope applying units EA, EP1 to EP3, and EF1 to EF3 respectively receive element waveform data Da, input from corresponding corresponding element reading units RA, RP1 to RP3, and RF1 to RF3. A predetermined envelope (weight) is given to Dp1 to Dp3, Df1 to Df3, and corresponding element musical sound signals Sa, Sp1 to Sp3, Sf1 to Sf3 are generated and output to the mixing unit MX.

  Specifically, the attack element envelope assigning unit EA weights the attack element waveform data Da according to the above-described dynamics value Dy, and generates an attack element musical sound signal Sa representing a rising component of the musical sound. On the other hand, the body part element envelope providing parts RP1 to RP3 and RF1 to RF3 generate the body part musical sound signal Sb representing the body part component of the musical sound in order to generate the body part element waveform data Dp1 to Dp3. , Df1 to Df3 are subjected to various weights.

  The first element envelope assigning units EP1 and EF1 and the second element envelope assigning units EP2 and EF2 are weight values Pu, designated by the pitch shift up weight generating unit SU and the pitch shift down weight generating unit SD, respectively. In accordance with Pd, the first element musical sound waveform signals Wp1, Wf1 and the second element musical sound waveform signals Wp1, Wf1; Wp2, Wf2 are weighted in response to the increase / decrease of the pitch shift value Ps that varies mainly periodically, First element musical sound signals Sp1 and Sf1 representing the characteristic component when the pitch shift value Ps increases (pitch shift up component) and the characteristic component when the pitch shift value Ps decreases (pitch shift up component), respectively. Two-element musical sound signals Sp2 and Sf2 are generated.

  Further, the third element envelope assigning sections EP3 and EF3 do not respond to the increase / decrease of the periodically changing pitch shift value Ps, and the vibrato depth amount (also referred to as the vibrato depth value) Vd specified by the modulation means ML. In response to this, the third element musical sound wave signals Wp3, Wf3 are weighted to generate third element musical sound signals Sp3, Sf3 that mainly represent certain timbre components included in the musical sound during vibrato performance. The third element musical sound signals Sp3, Sf3 and the corresponding musical sounds are referred to as (a) “non-vibrato component” or “non-vibrato sound” in the first embodiment, and (b) “sound component” in the second embodiment. Or “sounding sound”.

  When the vibrato depth Vd specified by the modulation means ML is zero, that is, during non-vibrato performance, (a) in the first embodiment, a constant weight is given to the third element musical sound waveform signals Wp3, Wf3, and (b In the second embodiment, the body portion Sb of the musical tone signal can be generated by giving a constant weight to the first and second element musical sound waveform signals Wp1, Wf1; Wp2, Wf2.

  Further, the weak sound element envelope providing section EP, that is, the weak sound first to third element envelope providing sections EP1 to EP3, is based on the weak sound component weight value Dp specified by the weak sound component weight generating section DP. In accordance with the increase or decrease, the weight for the weak element waveform data Sp1 to Sp3 is decreased or increased, respectively. On the other hand, the strong sound element envelope giving section EP, that is, the strong sound first to third element envelope giving sections EF1 to EF3, is the strong sound component weight specified by the strong sound component weight generating section DF. Depending on the value Df, the weights for the strong tone element waveform data Sf1 to Sf3 are increased or decreased as the dynamics value Dy increases or decreases, respectively. Thereby, a subtle change in timbre according to the increase or decrease in the dynamics value Dy can be expressed.

  Element musical sound signals Sa to Sf3 output from each element envelope applying unit EA to EF3 are introduced into a mixing unit MX as shown in the figure, and the mixing unit MX mixes these element musical sound signals Sa to Sf3, Further, the amplitude of the mixed tone signal is controlled in accordance with the dynamics value Dy to adjust the overall volume, and the tone output signal So is introduced into the subsequent stage of the sound source and effect applying unit 9, that is, the effect applying unit.

The control unit in the upper part of FIG. 2 will be described. The pitch shift value Ps is designated by the pitch shift calculation unit PS by a predetermined modulation frequency signal Lf generated by a low-frequency transmitter LF called “LFO”, and a modulation wheel. The pitch shift value Ps is calculated based on the vibrato depth amount Vd specified by the modulation means WL such as the above, and is output to the pitch generation unit PG. For example, if the instantaneous value of the modulation frequency signal Lf is expressed as “Lf” as it is and a predetermined constant for expressing the calculation result in units of pitch is “A”, the pitch shift value Ps is obtained by the following equation (1). .
Ps = A × Vd × Lf (1)

Further, when the pitch conversion value of the note number Nt is expressed by “Pnt”, the pitch generation unit PG generates a pitch Pi obtained by the following equation (2), and sets the reading speed of each element waveform data Da to Df3 to each element. The waveform data reading unit RA to RF3 is instructed.
Pi = Pnt + Ps (2)

  When the modulation wheel is not operated (operation amount = 0), the modulation means WL sets the vibrato depth amount Vd to zero (vibrato is not played), and has a value corresponding to the operation amount when the modulation wheel is operated. The vibrato depth amount Vd is output (vibrato performance starts), and the vibrato depth amount Vd is increased (vibrato performance is deepened) as the operation amount increases, and the vibrato depth amount Vd is equal to or greater than a predetermined operation amount. Then, the vibrato depth Vd is set to a predetermined maximum value (d1).

  The vibrato depth Vd generated by the modulation means WL and the pitch shift value Ps obtained by the pitch shift calculation unit PS are also output to the pitch shift up and pitch shift down weight generation unit SU; SD. The pitch shift up and pitch shift down weight generation unit SU; SD converts the pitch shift up component and pitch shift down component weight values Pu; Pd into first and second element envelope assigning units EP1, EF1, EP2, EF2, respectively. , The ratios of the pitch shift up components (Sp1, Sf1) and the pitch shift down components (Sp2, Sf2) in the modulated musical tone during the vibrato performance are periodically opposite to each other according to the pitch shift value Ps. [The pitch shift up component and the pitch shift down component are reciprocally changed according to the amount (Ps) of pitch shift by vibrato] The average level of both components is changed according to the vibrato depth amount Vd.

For example, for the modulation frequency signal Lf, the instantaneous value “Lf” in the equation (1) is represented by “Lf (−)” and the corresponding pitch shift value is represented by “Ps (−)”. Represent,
Ps (−) = A × Vd × Lf (−) (1 ′)
Then, (a) In the first embodiment, the pitch shift up component weight value Pu is obtained by the following equation (3a), and the pitch shift down component weight value Pd is obtained by the following equation (4a). Note that “B”, “C”, and “M” in the formula are predetermined constants (positive).
Pu = B × Ps + C × Vd + M (3a)
Pd = B × Ps (−) + C × Vd + M (4a)

On the other hand, (b) in the second embodiment, the first and second element envelope providing units EP1, EF1; EP2, EF2 also contribute to the generation of certain musical sounds included in the non-vibrato performance sounds. When playing vibrato, this constant tone gradually decreases. Therefore, in the second embodiment, weights for generating such a constant musical tone are added, so the pitch shift up component weight value Pu is obtained by the following equation (3b), and the pitch shift down component weight value Pd is It is obtained by the equation (4b). “D” and “E” in the formula are predetermined constants.
Pu = B × Ps−D × Vd + E (3b)
Pd = B × Ps (−) − D × Vd + E (4b)

The modulation means ML outputs a weight value corresponding to the vibrato depth amount Vd to the third element envelope assigning sections EP3 and EF3. (A) In the first embodiment, the ratio of the non-vibrato component (non-vibrato sound) is controlled. (B) In the second embodiment, the ratio of the sound component (sound sound) is controlled. For example, (a) the weight value Vda output from the modulation means ML to the third element envelope applying units EP3 and EF3 in the first embodiment is expressed by the following equation (5a), and (b) the second embodiment. In the embodiment, the weight value Vdb output from the modulation means ML to the third element envelope providing units EP3 and EF3 is expressed by the following equation (5b). Note that “F”, “G”, and “H” in the formula are predetermined constants.
Vda = F−G × Vd (5a)
Vdb = H × Vd (5b)

In this example, the weak sound component weight generation unit DP and the strong sound component weight generation unit DF each have a weak sound component weight so that the weak sound component decreases and the strong sound component increases as the dynamics value Dy increases. A value (proportion) Dp and a strong sound component weight value (proportion) Df are generated and output to the weak sound element envelope providing section EP: EP1 to EP3 and the strong sound element envelope providing section EP: EF1 to EF3. . That is, the weight values Dp and Df decrease or increase as the dynamics value Dy increases, for example, as represented by the following equations (6) and (7). In the formula, “J” and “K” are predetermined constants.
Dp = −J × Dy + K (6)
Df = J × Dy (7)

  As described above, in the tone generation system according to the embodiment of the present invention, in addition to the first tone signal generation sequence RP1-EP1, RF1-EF1 and the second tone signal generation sequence RP2-EP2, RF2-EF2, Music signal generation series RP3-EP3, RF3-EF3 are provided. In the first and second musical sound signal generation sequences, the first musical element signal (first musical element signal) in which the changes in the weights Pu and Pd conflict with each other in response to the increase / decrease in the pitch shift value Ps including the LFO cycle variation component Lf and the modulation depth component Vd. 1 element musical tone signal) Sp1, Sf1 and second musical tone element signals (second element musical tone signal) Sp2, Sf2 are generated, and a modulation effect such as vibrato is obtained by the period variation component Lf. On the other hand, in the third tone signal generation series RP3-EP3, RF3-EF3, third tone element signals (third element tone signals) Sp3, Sf3 whose weight changes in response to the magnitude of the modulation depth Vd are generated. Thus, it is possible to obtain a musical tone (“non-vibrato sound” or “sound”) having a certain tone color independent of pitch fluctuation.

[Operation Example in First Embodiment]
FIG. 3 shows an operation example in the first embodiment of the tone generation system according to the present invention. In the first embodiment, as described above, the first musical tone signal generation sequence RP1-EP1, RF1-EF1 and the second musical tone signal generation sequence RP2-EP2, RF2-EF2 for the modulation component are used for “non-vibrato”. The third musical tone signal generation series RP3-EP3, RF3-EF3 is added, and the third element waveform data Dp3, Df3 for “non-vibrato” used in this series is completely vibrato by a bowed instrument such as a violin. Created by recording no performance.

  As shown in FIG. 3A, the low frequency transmitter LF outputs a modulation frequency signal Lf that changes between a maximum value + L and a minimum value −L in a cycle T. Here, if there is a key-on at time to, as shown in FIG. 3 (3), an attack element tone signal Sa based on the attack element waveform data Da is generated by the attack tone signal generation sequence RA-EA, and at time ta. Then, the “non-vibrato” music signal generation sequence (third music signal generation sequence) RP3-EP3, RF3-EF3, and the “non-vibrato” music signal (third element music signal based on the third element waveform data Dp3, Df3) ) Generation of Sp3 and Sf3 starts, and generation of the attack element musical sound signal Sa ends at time tb. Further, as shown in FIG. 3 (2), at time tc, the modulation means ML gives the vibrato depth Vd by the modulation depth, and thereafter the vibrato depth Vd is saturated at the maximum value dm at time td. Gradually increases.

  In this case, the volume levels (“non-vibrato” level) of the “non-vibrato” musical sound signals Sp3 and Sf3 are constant levels va and vb corresponding to the constant “F” in the equation (5a) until the time tc. However, after the time tc, it attenuates as the vibrato depth Vd becomes deeper according to the second term of the equation (5a), and after the time td, it becomes the predetermined values vc and vd. On the other hand, the first element tone signal Sp1, Sf1 and the second element tone signal Sp2, Sf2 output from the first tone signal generation sequence RP1-EP1, RF1-EF1 and the second tone signal generation sequence RP2-EP2, RF2-EF2. Is changed according to the equations (3a) and (4a) and is at the zero level (Ps = 0, Vd = 0 → Pd = Pu = 0) until time tc, but rises at time tc, After tc, the average level increases as the vibrato depth Vd becomes deeper according to the second term of the equations (3a) and (4a) while periodically changing according to the first term of the equations (3a) and (4a). To do.

[Operation Example in Second Embodiment]
FIG. 4 shows an operation example in the second embodiment of the tone generation system according to the present invention. In the first embodiment, as described above, the first sound signal generation sequence RP1-EP1, RF1-EF1 and the second music signal generation sequence RP2-EP2, RF2-EF2 for the modulation component are used for “sound”. Third musical tone signal generation series RP3-EP3, RF3-EF3 is added. In addition, the third element waveform data Dp3, Df3 for “sound” used in this series is to record a performance with sufficient vibrato with a bowed instrument such as a violin, perform time frequency analysis, and perform a pitch change. Components extracted from characteristic parts other than the upper and lower ends are used. For example, frequency characteristics at the time when the pitch rises or falls near the center frequency of the pitch when the pitch changes almost periodically are extracted. Created.

  As shown in FIGS. 4 (1) and (2), the output mode of the low frequency transmitter LF and the indication mode of the vibrato depth amount Vd from the modulation means ML are shown in the operation example of the first embodiment [FIG. 1) and (2)]. As shown in FIG. 4 (3), when there is a key-on at time to, an attack element music signal Sa is generated, and at time ta, the first and second music signal generation sequences RP1-EP1, RF1-EF1; RP2 -EP2, RF2-EF2, generation of the first and second element musical sound signals Sp1, Sf1; Sp2, Sf2 based on the first and second element waveform data Dp1, Df1; Dp2, Df2 is started, and time tb This completes the generation of the attack element musical sound signal Sa. Then, when the vibrato depth Vd is given at time tc, the “sounding” tone signal generation sequence (third tone signal generation sequence) RP3-EP3, RF3-EF3 “sound” based on the third element waveform data Dp3, Df3. “Generation of musical tone signals (third element musical tone signals) Sp3 and Sf3 starts.

  In this case, the volume levels of the first and second element musical sound signals Sp1, Sf1; Sp2, Sf2 are constant levels corresponding to the constant “E” in the expressions (3b) and (4b) until the time tc. “non-vibrato” level is given by ve, vf; vg, vh (Ps = 0, Vd = 0 → Pd = Pu = E), but after time tc, as the vibrato depth Vd becomes deeper, the equation (3b) , (4b), the average level attenuates according to the second term, but the periodic modulation component varies at a desired ratio according to the first term of the equations (3b), (4b). On the other hand, the volume of the “sounding” musical sound signals Sp3 and Sf3 increases as the vibrato depth Vd becomes deeper according to the equation (5b), and reaches the maximum volume vj and vk at time td. A musical sound signal So including a unique “sound” when played over time is obtained.

[Various Embodiments]
The preferred embodiment of the present invention has been described above with reference to the drawings. However, this is merely an example, and the present invention can be variously modified without departing from the spirit of the invention. For example, in the embodiment, a waveform readout type sound source composed of an attack part and a body part has been described, but the present invention is not limited to this. As disclosed in Patent Document 3, “AEM” or “SAEM” is used to create a series of sound waveforms by connecting a plurality of rendition style waveform modules in time series combination rather than a portion composed of an attack portion and a body portion. It may be applied to a technique called “.”

  In the embodiment, the pitch Pi is equally supplied to all the element reading units so that all the pitches of the signals output from the element reading units are exactly the same, but each element reading unit is replaced with this. It is also possible to slightly shift the value of the pitch supplied to the signals so that the pitches of the signals output from the element reading units do not match. In particular, in the second embodiment, a pitch shift calculation unit PS ′ and a pitch generation unit PG ′ dedicated to the third element are further provided, and each third element reading unit has a pitch generation unit PG ′ instead of the pitch Pi. Is preferably supplied with the pitch Pi ′. In this case, the pitch shift calculation unit PS ′ and the pitch generation unit PG ′ generate the pitch Pi ′ in the same manner as the pitch shift calculation unit PS and the pitch generation unit PG, respectively. However, the effectiveness of the vibrato depth Vd from the pitch Pi. May be different, or a predetermined offset value may be added. By doing so, the phase of the third element is slightly shifted from the phase of the other elements, so that a rich sound peculiar to stringed instruments can be obtained.

  Furthermore, in the embodiment, the body part element readout units RP1 to RP3 and RF1 to RF3 read the body part element waveform data Dp1 to Dp3 and Df1 to Df3 after a predetermined time has elapsed from the note-on timing. Instead, the body element reading unit reads the body element waveform data from the note-on timing, and the envelope applying unit reduces the volume to “0” until the predetermined time elapses, and the predetermined time period elapses. Then, the envelope applying unit may apply a predetermined envelope, whereby the connection from the attack part to the body part can be made smooth.

  In the embodiment, a triangular wave is used as the LFO waveform, but another periodic waveform may be used instead. In particular, a more realistic sound may be realized by using a sine wave or the like.

Da, Dp1 to Dp3, Df1 to Df3 element waveform data,
RA, EA Attack element selection readout section and envelope provision section,
RP, EP Weak element reading unit and envelope applying unit,
RP1 to RP3 weak sound first to third element reading unit,
EP1 to EP3 first to third element envelope applying sections for weak sound,
RF, EF strong sound element reading unit and envelope applying unit,
RF1 to RF3 first to third element readout units for strong sound,
EF1 to EF3 first to third element envelope applying portions for strong sound,
Wa, Wp1 to Wp3, Wf1 to Wf3 element musical sound waveform signals,
Sa attack element music signal (attack part music signal),
Sp1-Sp3, Sf1-Sf3 Element musical sound signals (musical element signals) for body parts,
Sb Body music signal,
Sc Music output signal (all music signals).

Claims (5)

Musical tone waveform generating means for generating first, second and third musical tone waveform signals in response to one sounding instruction;
A pitch shift specifying means for specifying a pitch shift value;
Pitch changing means for changing the pitches of the first, second and third musical sound waveform signals generated by the musical sound waveform generating means based on the pitch shift value specified by the pitch shift specifying means;
A tone signal generating means for generating first, second and third tone element signals by giving predetermined weights to the first, second and third tone waveform signals generated by the tone waveform generating means, respectively. As for the first and second musical sound waveform signals, when the pitch shift value designated by the pitch shift designation means increases, the weight for the first musical sound waveform signal is increased and the weight for the second musical sound waveform signal is increased. When the pitch shift value decreases, the weight for the first musical sound waveform signal is decreased and the weight for the second musical sound waveform signal is increased. For the third musical sound waveform signal, the pitch shift value is reduced. A musical sound signal generating means for giving a predetermined weight independent of increase or decrease of
A music sound generating apparatus comprising: a music signal output means for outputting a body part music signal by mixing the first, second and third music element signals generated by the music signal generating means. further,
Comprising modulation means for changing the pitch modulation depth over time and specifying it;
The pitch shift designating means designates a pitch shift value having a pitch modulation depth that periodically changes at a predetermined time interval, and a depth of pitch modulation designated by the modulation means,
2. The musical tone generation apparatus according to claim 1, wherein the musical tone signal generation unit changes a weight for the third musical tone waveform signal in accordance with a pitch modulation depth designated by the modulation unit. The musical sound signal generating means includes
When the depth of the pitch modulation designated by the modulation means is zero, a predetermined weight is given to the third musical sound waveform signal,
3. The musical tone generation apparatus according to claim 2, wherein the weight for the third musical sound waveform signal is decreased as the depth of the pitch modulation increases. The musical sound waveform generating means generates first and second musical sound waveform signals based on the first and second musical sound waveform data having a predetermined frequency characteristic, respectively, and is intermediate between both musical sound waveform data. Generating a third musical sound waveform signal based on the third musical sound waveform data having frequency characteristics;
3. The musical tone generation apparatus according to claim 2, wherein the musical tone signal generating means increases the weight for the third musical sound waveform signal as the modulation depth specified by the modulation means increases. . A musical sound waveform generating step for generating first, second and third musical sound waveform signals in response to one sounding instruction;
A pitch shift specifying step for specifying a pitch shift value;
A pitch changing step for changing the pitch of the first, second, and third musical sound waveform signals generated in the musical sound waveform generating step based on the pitch shift value specified in the pitch shift specifying step;
A musical sound signal generating step for generating first, second and third musical tone element signals by giving predetermined weights to the first, second and third musical sound waveform signals generated in the musical sound waveform generating step, respectively; For the first and second musical sound waveform signals, when the pitch shift value designated in the pitch shift designation step increases, the weight for the first musical sound waveform signal is increased and the weight for the second musical sound waveform signal is increased. When the pitch shift value decreases, the weight for the first musical sound waveform signal is decreased and the weight for the second musical sound waveform signal is increased. For the third musical sound waveform signal, the pitch shift value is reduced. A musical sound signal generating step for giving a predetermined weight independent of increase or decrease of
A musical sound generation program for causing a computer to execute a procedure consisting of:

Images