JP4867940B2 - Musical sound generating apparatus and program - Google Patents

Description

  The present invention generates a plurality of element sounds based on a plurality of element data, synthesizes the generated plurality of element sounds to generate one musical sound, and a computer using such a musical sound generation apparatus. Related to the program to function as.

Conventionally, as voice data (tone data) that defines one timbre, a plurality of element data indicating the contents of waveforms, envelopes, etc. are prepared, and when sounding, element sounds are generated based on each element data. There is known a musical tone generating apparatus that synthesizes each element sound and outputs it as a tone color defined by voice data.
Such an apparatus is described in Patent Document 1, for example.
Japanese Patent No. 2842232

  In addition, in the tone generator capable of using a plurality of element data as described above, in which range of note numbers and velocities a note-on event having a velocity is used for each element (used) for sound generation. In the event that a note-on event occurs, element data used to generate a musical tone is determined according to the note number and velocity.

In Patent Document 2, by utilizing such a function, by appropriately determining the range in which element data is valid and the content of waveform data specified by the element data, waveform data of a plurality of rendition styles can be obtained. In addition, it is described that mapping is performed on an NV plane formed with a note number as a horizontal axis and a velocity as a vertical axis. If such a mapping is performed, the performance method is automatically selected according to the note number and velocity related to the key pressing operation in response to a single key pressing operation, and the performance is performed with the desired performance method. Musical sound can be output.
Japanese Patent Laying-Open No. 2005-107029

In addition, electronic musical instruments such as an electronic organ and a keyboard are provided with a function of changing the contents of musical sounds to be generated even when the same tone color is apparently selected according to the performance contents and settings.
As such a function, for example, when a legato setting is made, and a legato performance that partially overlaps the note-on periods of the preceding and following key presses, A legato function that performs portamento processing that smoothly connects pitches to musical notes corresponding to note-on and generates musical sounds corresponding to subsequent note-on using legato sound data with a low attack feeling is known. ing.

There is also known a note-off sound generation function that starts sound generation of a note-on sound in response to note-on and starts sound generation of another note-off sound in response to note-off. This function is used to reproduce the harpsichord's scratching sound. Such a note-off sound generation function is described in Patent Document 3, for example.
There is also known a replacement (random) sound generation function for generating musical tones having different characteristics such as frequency characteristics and pitch according to random values each time a note is turned on. This function is used to reproduce sounds that cannot be played. Such a replacement sound generation function is described in Patent Document 4, for example.
There is also a rendition style switching function that provides a rendition style switch and generates a musical tone of a rendition style according to the rendition style switch operated at that time in response to note-on. Such a rendition style switch is described in Patent Document 5, for example.
Japanese Examined Patent Publication No. 05-46955 Japanese Patent No. 2722665 Japanese Patent No. 3407610

  By the way, as described above, the function of changing the content of a musical sound to be generated even when the same timbre is seemingly selected according to the performance content and settings is conventionally used when using timbre data created by the manufacturer. It was only available. And when making a user produce timbre data, the function which performs the setting for implement | achieving a legato function etc. using the timbre data was not performed.

  According to the technique described in Patent Document 2, waveforms corresponding to different performance methods can be selectively used for sound generation using one timbre data. However, in this technique, since waveform mapping corresponding to a performance style is performed, there is a problem that only one performance style can be associated with a specific combination of note number and velocity, and the degree of freedom in creating a timbre is low. In addition, the operation to select a waveform according to the relationship between multiple note-ons as in the above-mentioned legato function, the operation to select a waveform according to the type of event, such as the note-off sound generation function, and replacement Even if the note-on note number and the velocity are the same as in the sound generation function and the rendition style switching function, there is a problem that the technique described in Patent Document 2 cannot cope with the operation of switching the waveform depending on the case.

  In addition, in a synthesizer that can edit timbre data, the timbre data is generally composed of a plurality of elements, but each element starts to sound in response to note-on, legato sound generation, random sound generation, Conditional pronunciation control, such as note-off pronunciation, was not possible. That is, a plurality of elements sound independently of each other, and the sound of one element cannot be controlled by the sound state of another element.

The present invention has been made based on such a background, and in a tone generator capable of editing timbre data having a plurality of element data defining a plurality of element sounds , the timbre changes dynamically for each pronunciation. The purpose is to be able to generate a variety of musical sounds .

In order to achieve the above object, the present invention provides a musical sound generating apparatus that generates a plurality of element sounds based on a plurality of element data and generates one musical sound by synthesizing the generated plurality of element sounds. A plurality of element data corresponding to a plurality of element sounds constituting a musical tone of one tone color, each corresponding to a characteristic control data for controlling a musical sound characteristic of the corresponding element sound, and a selection rule for each element data Storage means for storing a plurality of element data including pronunciation type data indicating one pronunciation type among the plurality of pronunciation types, and the plurality of element data stored in the storage means in response to an instruction from the user for each, modify the settings of the sound type data indicating the pronunciation type desired one of said plurality of sound types The sound generation control event generated in response to the generation of a sound generation control event that instructs the sound generation control of one musical sound and includes a parameter that controls the characteristics of the one musical sound; Element selection means for selecting one or more element data used for generating an element sound from the plurality of element data based on the pronunciation type indicated by the pronunciation type data of each of the plurality of element data, and the element Based on one or a plurality of element data selected by the selection means, for each element data, a musical tone characteristic corresponding to the characteristic control data included in the element data and the parameter included in the generated sound generation control event is obtained. Element sound generating means for generating element sounds .
In such a musical sound generating apparatus, the editing means may be means for editing the characteristic control data for each of the plurality of element data in accordance with an instruction from a user.

Further, in the above-described musical sound generation device, the sound generation control event is a note-on event, the non-replacement sound and the replacement sound are included in the plurality of sound generation types, the element selection means includes a first element selection means, Second element selection means is provided, and the first element selection means is element data in which the pronunciation type data indicates the non-replacement sound among the plurality of element data in response to the occurrence of the note-on event. Is selected as element data to be used for generating the element sound, and the second element selecting means determines that the pronunciation type data is the replacement sound among the plurality of element data in response to the occurrence of the note-on event. Any one of the element data indicating the element according to a predetermined rule or at random It may be so selected as element data used to generate the element tone.
Furthermore, the non-replacement sound may be a normal sound, a legato sound, a switch-on sound, or a switch-off sound.

Further, in the above-described musical sound generation device, the sound generation control event is a note-on event, and the element selection means includes a legato sound and a non-legato sound in the plurality of sound generation types, and the element selection means responds to the occurrence of the note-on event Determining whether the non-legato sound or the element sound generated based on the element data of the legato sound is still in the note-on state according to the note-on event that occurred last time. If there is, select the element data in which the pronunciation type data indicates the legato sound among the plurality of element data as the element data used for generating the element sound. Among the plurality of element data, the element in which the pronunciation type data indicates the non-legato sound The chromatography data, may be selected as the element data to be used to generate the above element tones.
Further, the non-legato sound may be a normal sound, a switch-on sound, or a switch-off sound.

In the above musical tone generation apparatus, the sound generation control event is a note-on event and a note-off event, the note-on sound and the note-off sound are included in the plurality of sound generation types, and the element selection means includes a first element. A selection means and a second element selection means; the element sound generation means is provided with a first element sound generation means, an element sound identification means and a second element sound generation means; and the first element selection means is In response to the occurrence of the note-on event, element data in which the pronunciation type data indicates the note-on sound is selected from among the plurality of element data as element data used for generating the element sound, and 1 element sound generating means is the element selected by the first element selecting means. Generating the element sound having a musical tone characteristic according to the characteristic control data included in the data, and the second element selecting means generates the sound of the plurality of element data in response to the occurrence of the note-off event. Element data whose type data indicates the note-off sound is selected as element data to be used for generating the element sound, and the element sound specifying means selects the element sound generated by the first element sound generating means. Among them, one element sound is specified as a characteristic takeover source, the second element sound generating means takes over some of the musical sound characteristics of the element sound specified by the element sound specifying means, and the second element sound generating means The musical tone characteristics according to the characteristic control data included in the element data selected by the element selection means It may generate the element sounds.
Further, the note-on sound may be a normal sound, legato sound, random sound, cycle sound, switch-on sound, or switch-off sound .

Further, in the above-described musical sound generation device, a switch that can be switched between an off state and an on state by a user operation is provided, the sound generation control event is a note-on event, and the switch-off sound, the switch-on sound, The element selection means, including a constant sound, in response to the occurrence of the note-on event, (1) if the switch is off when the note-on event occurs, The element data indicating that the sound generation type data indicates the switch-off sound is selected as element data used to generate the element sound, and (2) the switch is on when the note-on event occurs. Indicates that the pronunciation type data of the plurality of element data indicates a switch-on sound. Element data is selected as element data to be used for generating the element sound, and (3) among the plurality of element data, the pronunciation type data indicates the constant sound regardless of the on / off state of the switch. Element data may also be selected as element data used for generating the element sound.
Further, the continuous sound may be a normal sound, legato sound, random sound or cycle sound.
The program of the present invention is a program for causing a computer to function as any one of the above-described musical tone generation apparatuses.

According to the musical tone generating apparatus of the present invention as described above, in the musical tone generating apparatus capable of editing timbre data having a plurality of element data defining a plurality of element sounds, a change in which the timbre changes dynamically for each pronunciation. It is possible to generate rich musical sounds.
Further, according to the program of the present invention, the computer can function as the above-described musical tone generating apparatus to realize the characteristics thereof, and the same effect can be obtained.

Hereinafter, the best mode for carrying out the present invention will be specifically described with reference to the drawings.
First, FIG. 1 shows a hardware configuration of a synthesizer that is an embodiment of a musical tone generating apparatus according to the present invention.
As shown in FIG. 1, a synthesizer 10 includes a CPU 11, a ROM 12, a RAM 13, a MIDI (Musical Instruments Digital Interface: registered trademark) I / F (interface) 14, a performance operator 15, a timer 16, a display unit 17, and a panel operation. A child 18, a storage device 19, and a musical tone signal generation circuit 20 are provided, and these are connected by a system bus 34. In addition, a waveform memory 31, a DAC (digital / analog converter) 32, and a sound system 33 are also provided.

  The CPU 11 is a control unit that performs overall control of the synthesizer 10, and detects and displays operation contents of the performance operator 15 and the panel operator 18 by executing a required control program stored in the ROM 12 or the storage device 19. Control operations such as display control of the device 17, MIDI data transmission / reception control via the MIDI_I / F 14, and control of musical tone signal generation processing in the musical tone signal generation circuit 20 are performed.

The ROM 12 is a storage unit that stores a control program executed by the CPU 11, data that does not need to be changed, and the like. It is conceivable that the ROM 12 is constituted by rewritable nonvolatile storage means such as a flash memory so that these data can be updated.
The RAM 13 is a storage unit that is used as a work memory for the CPU 11 or that temporarily stores parameter values and the like.

The MIDI_I / F 14 is an interface for transmitting / receiving MIDI data to / from an external device.
The performance operator 15 is an operator for accepting a performance operation by the user, such as a keyboard or a pedal.
The timer 16 is a time measuring means for measuring the time used for interrupt control timing control and the like.
The display unit 17 is configured by a liquid crystal display (LCD), a light emitting diode (LED) lamp, or the like, and a graphical user interface (for receiving an operation state and setting contents of the synthesizer 10, a message to the user, an instruction from the user) (GUI) and the like.

The panel operation element 18 is an operation element for receiving a setting operation related to the operation of the synthesizer 10 such as a key, a knob, a slider, and a pitch bend. Further, the panel operator 18 includes switches A and B (not shown) for element selection which will be described later.
The storage device 19 includes an HDD (hard disk drive), a CD-ROM drive, a magneto-optical (MO) disk drive, and the like. The storage device 19 stores control programs executed by the CPU 11 and data such as voice data and music data changed by editing. Nonvolatile storage means for storing.

The musical sound signal generation circuit 20 is a MIDI event (performance event for controlling the generation of musical sound) generated by the CPU 11 according to the operation of the performance operator 15 or generated by the CPU 11 according to the contents of the music data during automatic performance of music. Waveform generating means for generating a musical tone waveform according to
More specifically, a sound source register 21, a waveform data read processing unit 22, a filter processing unit 23, an envelope processing unit 24, a ch (channel) accumulation unit 25, and an effect processing unit 26 are provided.

  The tone generator register 21 has a plurality of sets of registers corresponding to a plurality of sound generation channels, and each of the registers 22 is a component 22 of the tone signal generation circuit 20 written by the CPU 11 via the system bus 34. Various data for controlling .about.26 are temporarily stored. This data includes data for designating the storage address of the waveform data to be read, parameters for generating filters and envelopes, the contents of effect processing applied to the waveform data, and the like.

The waveform data read processing unit 22 reads waveform data used for sound generation from an address set in the sound source register 21 among a plurality of waveform data stored in the waveform memory 31, and outputs the waveform data to the filter processing unit 23. It has a function. The read address at that time advances at a speed specified according to the pitch of the musical sound to be generated, and interpolation between samples is performed in order to suppress aliasing noise.
The waveform memory 31 stores a plurality of waveform data corresponding to elements to be described later. Different waveform data may be prepared for each element for each predetermined pitch range (may be for each pitch) or for each velocity range. The waveform data can be obtained by sampling a sound of a musical instrument or the like, or can be generated by calculation. Further, the waveform memory 31 may be constituted by a ROM or a RAM. In the latter case, the waveform data set stored in the storage device 19 or the like is read out and stored in the waveform memory 31 when the power is turned on.

The filter processing unit 23 performs filter processing for adjusting the frequency characteristics according to the parameter values set in the sound source register 21 on the waveform data output from the waveform data reading processing unit 22, and the result is used as an envelope processing unit. 24.
The envelope processing unit 24 generates an envelope waveform indicating a temporal change in volume (Amplitude) based on the parameter value set in the sound source register 21, and the waveform data output from the filter processing unit 23 is subjected to the envelope. It has a function of adjusting the volume according to the waveform. The waveform data after this processing is output to the ch accumulation unit 25.

  The waveform data read processing unit 22, the filter processing unit 23, and the envelope processing unit 24 execute each process in a time division manner at a plurality of timings corresponding to a plurality of sound generation channels within one sampling period. In the synthesizer 10, 128 sound generation channels are provided, and this number of processing timings is provided. Further, as will be described later, the synthesizer 10 prepares a plurality of element data for one tone color, synthesizes element sounds to be generated based on each element data, and generates a musical tone of a certain tone color. For example, when one sound generation channel is used for sound generation using one element data and, for example, four element sounds are synthesized to generate a sound of one tone color, four sound generations for generating the sound of that sound color are generated. ch is used.

The channel accumulation unit 25 has a function of accumulating waveform data output from the envelope processing unit 24 at a timing corresponding to a plurality of sound generation channels for one sampling period, and outputting the result to the effect processing unit 26.
The effect processing unit 26 gives a musical effect such as chorus or reverb to the waveform data output from the channel accumulation unit 25 according to the parameter value set in the sound source register 21, and supplies it to the DAC 32 at every sampling period. Has a function to output.

The DAC 32 has a function of converting digital waveform data output from the tone signal generation circuit 20 (effect processing unit 26) into an analog tone signal and supplying the analog tone signal to the sound system 33.
The sound system 33 is constituted by a speaker or the like, and is an audio output unit that outputs sound according to a musical sound signal supplied from the DAC 32.

  The synthesizer 10 as described above can generate and output musical tones specified by the user in accordance with the user's performance operation. Musical sounds according to automatic performance can be output in the same way. In this case, one tone color tone can be generated by synthesizing a plurality of element sounds.

  In this case, in response to the occurrence of a note-on event (a sound generation control event for instructing the start of sound generation), the CPU 11 instructs the sound signal generation circuit 20 to start sounding a required element sound, and Start generation. Then, in response to the occurrence of a note-off event corresponding to the note-on event (a sound generation control event for instructing the start of musical sound attenuation), the CPU 11 causes the element sound that has started sounding to the music signal generation circuit 20 during the note-on event. Is instructed to start attenuation (release start), and the attenuation of the element sound is started.

  For each element sound, a period from when the sound generation starts in response to the note-on event to when the sound begins to decay in response to the note-off event is referred to as the “note-on state” of the element sound. The period from when attenuation starts until the sound becomes smaller and disappears is called the “release period” of the element sound. As will be described later, the sound of the element sound may be started in response to the occurrence of a note-off event.

The synthesizer 10 has a plurality of element data used for generating element sounds, and tone color data (voice data) including specification of rules for automatically determining element data used for tone generation according to various rules. Can be easily edited by the user.
Hereinafter, the configuration of the voice data and the sound generation control process according to the voice data will be described in detail. In the following description, “element data” as a part constituting voice data is simply referred to as “element”.

First, FIG. 2 shows the configuration of voice data for one tone color as a data map.
The voice data shown in this figure is stored in advance in the ROM 12 or the storage device 19 for a plurality of timbres. Then, when any voice data is selected for use in pronunciation by the operation of the user (performer) or the specification of automatic performance data, the CPU 11 reads the voice data and stores each part secured in the RAM 13. It is stored in each voice buffer area and referred to during sound generation processing. This storage also sets element grouping and sound generation type, which will be described later. In this process, the CPU 11 functions as a group setting means and a type setting means.

As shown in FIG. 2, the voice data is composed of a plurality of element data and common data indicating information on the entire voice. Here, the number of element data is eight.
Among these, the common data includes a voice name, a voice select number, a program change number, a temperament designation data, an effect parameter, poly / mono data, and other data.

The voice name is data indicating the name of the voice.
The voice select number and the program change number are data indicating numbers to be specified by an operator or a MIDI program change event when selecting a voice.
The temperament designation data is data that designates a temperament specific to the voice.
The effect parameter is data indicating the content of a common effect to be given to all element sounds.

The poly / mono data is data for designating whether the voice is sounded in a poly mode that allows simultaneous sound generation of a plurality of pitches or in a mono mode that does not allow sound. In poly mode, when multiple keys are pressed on the keyboard at the same time (with overlapping timing), a sound corresponding to all the keys is sounded, but in mono mode, after pressing one key, the next key is pressed. When the key is played, the sound corresponding to the previous key press starts to be attenuated (note-off) even if the previous key is not released.
A voice may be compatible with both poly / mono modes, and the mode may be switched by a user operation during performance.

Each element data includes element on / off, sound generation type, group number, waveform set number, note range, velocity range, velocity curve, filter parameter, EG parameter, and other data.
The element on / off is data for designating valid (on) / invalid (off) of the element. Only elements with this on will be used for pronunciation.
The pronunciation type is data specifying a rule used when determining whether or not an element is used for pronunciation as an element type. The sound generation types that can be specified are shown in Table 1 below. The parenthesis written in addition to the specific pronunciation type name is a classification according to the function of the pronunciation type.

  The group number is data indicating grouping of elements, and the rule indicated by the pronunciation type is applied to each group. Here, eight groups from 1 to 8 can be set, and all elements in one voice can be made into separate groups.

  The waveform set number indicates the number of waveform set data used for generating an element sound using the corresponding element. In the waveform set data, a plurality of sets of data are numbered and stored in the waveform memory 31. When an element sound is generated, a plurality of waveforms included in the waveform set data designated by the waveform set number are stored. Any one of the data is selected according to the note (pitch) of the note on / off event, and is read out by the waveform data reading processing unit 21. The waveform set data may be freely edited by the user. Further, when selecting waveform data, the velocity (intensity) of the note on / off event may be used in addition to or instead of the note.

  The waveform set data may include a plurality of waveform data, and different waveform data may be used according to the range range or velocity range of the note on / off event. Further, in the element data, an address for reading waveform data from the waveform memory 31 may be directly specified instead of the waveform set number.

  Each of the note range and velocity range is limit data, and the element is used for sound generation when the note number and velocity of the generated note-on event (note-off event when the sound type is key-off sound) is in any range. It is data that specifies. If the note number or velocity is out of these ranges, the element for which the range is set is not used for sound generation.

The velocity curve is data defining a velocity curve used for generating the element sound.
The filter parameter is data defining the type of filter processing performed by the filter processing unit 23 and the characteristic value thereof when generating element sounds.
The EG parameter is data defining characteristics of an envelope used for various processes performed by the envelope processing unit 24 when generating an element sound.

Among the above, the waveform set number, velocity curve, filter parameter, and EG parameter are characteristic control data for controlling the musical tone characteristics (amplitude change, pitch change, waveform shape, etc.) of the element sound.
The voice data as described above is edited by the CPU 11 functioning as an editing means, using the panel operator 18 of the synthesizer 10, or by using a required application on an external PC (personal computer) or the like. Can be stored in the storage device 19.

Next, FIG. 3 shows a configuration of sound generation management data used when the CPU 11 instructs the musical tone signal generation circuit 20 to generate sound using the above voice data.
The sound management data shown in this figure is stored in the RAM 13 by securing a storage area in the RAM 13, and is rewritten as needed based on the contents of the voice data used for sound generation and the contents of the generated note on / off event. This element is referred to when an element to be used for pronunciation corresponding to the performance event is selected (selected from a plurality of elements included in the voice data).

  Since the synthesizer 10 is capable of sound generation in a plurality of parts, it is possible to store, for each part, management data related to events that have occurred in that part and voice data used for sound generation in that part. Yes. For example, the part can be a part for each performance operator such as an upper keyboard, a lower keyboard, a foot keyboard, etc., each key is divided into a plurality of key ranges, and each part of the automatic performance. It can be considered to be a part for each track. The synthesizer 10 can set a timbre for each part. Also, when a performance event (MIDI event) of a part occurs, a musical tone to be generated is controlled using voice data (tone color data) indicating a timbre set for that part.

The sound management data for each part includes a voice number, a group sound status flag GS, a valid element flag EE, a legato status flag LS, data regarding each group, and other data.
The voice number is a number indicating voice data used for pronunciation of the corresponding part.

  The group sounding state flag GS describes whether or not there is an element sound that is still in a note-on state among non-legato or legato element sounds that are sounded in response to a past note-on event in each group. Flag data for GS is 8-bit data, and each bit corresponds to the 1st to 8th groups, and indicates information related to the group.

  The valid element flag EE is flag data for temporarily determining and describing which element can be used for sound generation according to the event when a note-on event is detected. This determination refers to the on / off and limit data of each element. If the sound generation type is switch-compatible (all switch off, switch A on or switch B on sound generation type), the sound generation type and the switch Also refer to the state. EE is also 8-bit data, and each bit corresponds to the 1st to 8th elements and indicates information related to the element.

The legato state flag LS is flag data for determining whether or not the event is a note-on event related to a legato performance when a note-on event is detected, and temporarily describing the result. The LS may be 1-bit data.
These values of GS, EE, and LS are updated every time there is a note-on event.

The data relating to each group includes a non-legato element flag NLE, a legato element flag LE, a cycle element flag CE, a random element flag RE, a key-off sounding element flag OSE, and a cycle sounding register CX.
Of these, the five flag data excluding the cycle sound register CX correspond to the five sound types of non-legato, legato, cycle, random, and key-off sound, respectively. Elements that belong to a group and are set to the corresponding pronunciation type. Here, the elements set to the four sound generation types of normal, all switch off, switch A on, and switch B on are collectively indicated by a non-legato element flag NLE as a non-legato element.

  Each flag data is 8-bit data, and each bit corresponds to each element in the voice data. In each flag data of each group, the bits of the elements belonging to the corresponding group and set to the sound generation type corresponding to the flag data are set to “1”, and the remaining bits are set to “0”. . Since each element can belong to only one group and only one sound generation type can be set, each bit can be set to “1” only by one flag data through all flags of all groups. However, a plurality of bits may be “1” with one flag data.

That is, if an e-th element belongs to a certain group and that element is a legato element, the legato element flag LE of that group is set to “1” in the e-th bit. In addition, the e-th bit of all other flag data is “0”. Further, if the first and second elements belong to a certain group and both of them belong to the cycle, the first and second bits of the cycle element flag CE of that group are “1”.
These NLE, LE, CE, RE, and OSE flags are set when voice data is selected in each part, and are not changed unless the voice data is changed thereafter.

The cycle sound generation register CX is a register that records the element of the cycle used for sound generation corresponding to the latest note-on event, and determines the cycle element used for sound generation corresponding to the next note-on event. refer. The value of the cycle sound register CX is updated every time there is a note-on event. Further, CX may be provided for all the group data of the first to eighth groups, or may be provided only for the group data of the group including the cycle element.
The above is the configuration of the pronunciation management data.

Next, FIG. 4 shows a flowchart of the sound generation instruction process corresponding to the event, which is executed when the CPU 11 detects the note-on event. This process proceeds while referring to and rewriting each data shown in FIG.
When the CPU 11 detects that a note-on event has occurred by being generated in response to a user's performance operation or by automatic performance processing, the CPU 11 starts the processing shown in the flowchart of FIG.

First, data of part p, note number n, velocity v possessed by the detected note-on event is acquired (S11). The note number basically corresponds to the pitch of the sound to be generated, and the velocity basically corresponds to the intensity of the performance operation, but is not necessarily limited thereto.
Next, referring to the voice number corresponding to the part p acquired in step S11, the voice data set in the part is selected and determined to be used for the subsequent processing (S12).
Then, the validity determination process (S13) and the mono mode process (S14) are executed to set the valid element flag EE and the legato state flag LS.

  Thereafter, the group number register g is sequentially increased from 1 to 8, and the non-legato & legato sound generation processing (S16), cycle sound generation processing (S17), and random sound generation processing (S18) are executed for each of the first to eighth groups. Then, the element used for sound generation is determined according to the sound generation type of the elements belonging to the group, the sound generation channel of the tone signal generation circuit 20 is assigned, and the sound generation channel register assigned to the sound source register 21 is required for sound generation. Settings are made (S15 to S21).

  Also, the current note number n and the sound channel to which the selected element is assigned are shown as elements for selecting one of the elements for which sound is set and passing the envelope information to the sound at the corresponding note-off time. The takeover information is added to the takeover list HL (g) of the group being processed (S19). In the process of step S19, the CPU 11 functions as element sound specifying means.

The takeover list HL (g) may be provided in the group data shown in FIG. 3, but the number of takeover information to be registered is not limited to one. Since the amount varies, it is preferable to store the amount separately from the group data.
In addition, the takeover information is deleted from the takeover list HL (g) when the described sounding channel is instructed to start attenuation (release) or the sounding volume is lowered and the sounding channel is released. Then, the selection of the element to carry over the envelope information may be automatically performed according to the element number in ascending order, or the user may be able to specify the element selection order. The element selected here is preferably an element sound having the longest sound generation duration among a plurality of element sounds that start sounding simultaneously in response to a note-on event.
The handover of envelope information will be described later.

  When YES is determined in the step S21, the musical sound signal generation circuit 20 is instructed to start sounding of all sounding channels set to sounding in the sounding processes in the steps S16 to S18 (S22), and the process is ended. In response to the sounding start instruction, the musical tone signal generation circuit 20 corresponds to the pitch and velocity v corresponding to the note number n based on the element data of each element set in the register of these sounding channels. By generating element sounds having the same volume and synthesizing them, a musical sound corresponding to the detected note-on event can be generated. In this case, the CPU 11 and the tone signal generation circuit 20 function as element sound generation means.

Next, the contents of each process shown in FIG. 4 will be described in more detail.
First, FIG. 5 shows a flowchart of the effectiveness determination process in step S13.
In this process, the CPU 11 increments the value of the element number register e by 1, and sequentially executes the processes of steps S32 to S37 for each of the first to eighth elements (S31, S38, S39).

In this part of the processing, whether the element on / off of the e-th element is on (S32), and the note number n and velocity v acquired in step S11 of FIG. It is sequentially determined whether it is within the effective range defined by the velocity range (S33). If the e-th element is a switch-compatible element (all switch-off, switch A-on or switch-B-on sound generation type) (S34), the element selection switch is depressed (on-off state) when the e-th element It is also determined whether or not the element requirement is satisfied (S35).
If all these determinations are YES (even if NO in step S34), the e-th bit of the valid element flag EE is set to “1”, and the e-th element is detected as a note-on detected this time. It shows that it can be used for the pronunciation corresponding to an event (S36).

If NO in any of steps S32, S33, and S35, the e-th bit of the valid element flag EE is set to “0”, and the e-th element corresponds to the note-on event detected this time. It indicates that it is not used for the pronunciation to be performed (S37).
Then, the above process is repeated, and when YES is determined in the step S39, the process returns to the original process.
With the above processing, it is possible to record in the valid element flag EE which element can be used for the current pronunciation among the elements included in the voice data used for the pronunciation of the part p.

Next, FIG. 6 shows a flowchart of the mono-mode process in step S14.
In this process, the CPU 11 first refers to the poly / mono data of the voice to be used, and determines whether or not the voice is in the mono mode (S41). If it is the mono mode, the legato function can be used, so it is determined whether or not there is a current note-on event while the previous note-on is being performed in part p (S42).

  In this case, since it is known that the next key is pressed before the previous key is released, a note-off event corresponding to the previous note-on (same note number in the same part) is generated. The musical tone signal generation circuit 20 starts attenuation of the element sound that has been generated by the previous note-on (S43), and sets the legato state flag LS to “1” to confirm that the legato performance has been performed. In step S44, the process returns to the original process.

If NO in step S41, it is a poly mode and the legato function cannot be used. If NO in step S42, the legato performance is not performed, so that each legato state flag LS is set to “0”. Set to indicate that no legato performance has been performed, and return to the original process.
With the above processing, whether or not the legato playing method has been performed can be recorded in the legato state flag LS in the mono mode in which the legato function is valid.

Next, FIG. 7 shows a flowchart of the non-legato & legato sound generation process executed for the g-th group in step S16.
In this process, the CPU 11 first determines whether or not the legato state flag LS set in the process shown in FIG. 6 is “0” (S51). If it is “0” here, it can be seen that the current note-on has not been performed with the legato playing method, and therefore, the process proceeds to the processing related to the pronunciation of the non-legato sound in step S52 and subsequent steps.

First, for each bit, the logical product of the valid element flag EE and the non-legato element flag NLE (g) existing in the g-th group being processed is calculated (S52). The element data indicated by the bit whose result is “1” due to this logical product is one in which the condition used for pronunciation is satisfied among the non-legato elements of the g-th group.
In this flow, the elements of the g-th group are processed, and the bit “1” in the non-legato element flag NLE (g) indicates that the corresponding element is non-legato, and the valid element The bit “1” in the flag EE indicates that the corresponding element satisfies the requirements for the note range and velocity range (and switch correspondence) and is an element that can be used for the current sound generation.

  Therefore, if there is a bit whose result is “1” (S53), the sound generation using the element corresponding to each bit is assigned to the sound generation channel of the musical sound signal generation circuit 20 to generate the sound for generating each element sound. Ch is secured (S54), and at the same time the tone information (sounding parameters) defined by the element data of each element is set for the assigned sounding channel (S55).

  The musical tone information is set by selecting the waveform data corresponding to the note number n from the waveform set data indicated by the waveform set number, and taking note of the read start address, the loop start / end address, and the waveform data. The reading speed for shifting the pitch to the pitch corresponding to the number n is set in the area for the waveform data reading processing unit 22 of the assigned sounding channel register in the tone generator register 21, or the filter parameter or EG parameter is set. The filter of the register of the same sound channel is applied to the note number n and the velocity v in part (including the change of the EG parameter for changing the envelope waveform value to the value indicating the volume corresponding to the velocity v). This is processing for setting the area for the processing unit 23 and the envelope processing unit 24. If there are a plurality of bits with a result of “1”, the sound generation channel is assigned and the register is set for a plurality of elements corresponding to each bit. The start of pronunciation itself is performed later in step S22 of FIG.

Thereafter, GS (g), which is the g-th bit of the group sounding state flag GS, is set to “1” to indicate that there is sounding using a non-legato or legato element in the gth group (S56). Return to the original process.
On the other hand, if there is no bit of “1” in step S53, it is found that there is no non-legato element to be used for sound generation in the g-th group, so that “0” is set to GS (g) and g It is indicated that there is no pronunciation using a non-legato or legato element in the th group (S57), and the processing returns to the original processing.
The group sounding state flag GS is used only in the mono mode. If NO is determined in step S53 in the mono mode, the non-legato sound and the legato sound are not in the note-on state. GS (g) is set to “0”.

If LS = 0 is not satisfied in step S51, the current note-on is performed by the legato playing method, and the process proceeds to the processing relating to the legato sound generation in step S58 and subsequent steps.
First, the logical product of the valid element flag EE and the legato element flag LE (g) of the g-th group being processed is calculated for each bit (S58). The element indicated by the bit whose result is “1” due to the logical product is one that satisfies the requirements of the note range and velocity range of the legato elements in the g-th group.

  Here, in the synthesizer 10, the legato elements are used for sound generation only when the sound corresponding to the previous note-on event in the same group is performed using the non-legato or legato elements, and these elements are used. If the sound was not played due to a note number or velocity limitation, a non-legato element is used for sounding even when a legato performance is performed.

Therefore, only when there is a bit of “1” as the result of the logical product in step S58 and GS (g) = 1 (S59), the process proceeds to step S60 and subsequent steps to set the sound generation using the legato element. It performs (S60, S61). The processing performed here is the same as in steps S54 and S55, but the elements to be used are legato elements.
In general, in a legato element, by setting a waveform set number that collects weakly attacked waveform data, or by setting an EG parameter with a small first decay, the sense of attack of the generated element sound is reduced. Has been. Further, GS (g) = 1 here indicates that in the mono mode, the non-legato sound or the legato sound is continuing the note-on state for the g-th group.

After step S61, the process returns to the original process. In this case, since GS (g) is already “1”, there is no process for setting “1” again, but this process may be added.
If NO in step S59, the process proceeds to step S52 to return to the original process in order to examine the pronunciation using a non-legato element.
By the above processing, if there is a non-legato and / or legato element in the g-th group, search for one that satisfies the conditions used for pronunciation, and make settings necessary for pronunciation using that element. Can do. In this case, whether or not the legato performance is performed is one criterion for whether to use a non-legato element or a legato element for pronunciation.

Next, FIG. 8 shows a flowchart of the cycle sound generation process executed for the g-th group in step S17.
In this process, the CPU 11 first performs a logical product of the valid element flag EE and the cycle element flag CE (g) of the g-th group being processed for each bit (S71). The element data indicated by the bit whose result is “1” due to the logical product satisfies the requirements of the note range and velocity range of the element among the elements of the cycle existing in the g-th group.
Therefore, if there is a bit whose result is “1” (S72), the cycle sound register CX (g) of the g-th group is referred to and the sound corresponding to the previous note-on event recorded therein is recorded. Based on the value indicating what the element of the used cycle is, the element used for the current pronunciation is selected from the elements whose logical product is “1” (S73).

The selection rule may be, for example, to select the candidate with the next smallest number after CX (g) and select the candidate with the smallest number when there is no candidate. In this case, for example, if the first to third elements are selection candidates, each candidate is periodically selected in sequence every time there is a note-on, such as 1 → 2 → 3 → 1. . In addition, when an EE is “0” due to a note number or velocity restriction for an element at a certain note-on, that element is excluded from the selection candidates at the time of the note-on, and therefore You will skip the element and select the next.
Of course, the selection rule is not limited to this, and it is conceivable that the selection is made in the order of decreasing numbers, or a register for counting the number of selections is provided, and the selection is made a plurality of times in order. Alternatively, the user may set the selection order of elements and select according to the order.

Then, after performing the selection as described above, the number of the selected element is recorded in CX (g) for reference at the next note-on event detection (S74), and in the case of steps S54 and S55 in FIG. In the same manner as described above, assignment of sounding channels for sounding using the selected element and setting of musical tone information are performed (S75, S76). Thereafter, the process returns to the original process.
With the above processing, if there is a cycle element in the g-th group, one to be used for sound generation can be selected from the elements and settings necessary for sound generation using that element can be made.

Next, FIG. 9 shows a flowchart of the random sound generation process executed for the g-th group in step S18.
In this process, the CPU 11 first ANDs the valid element flag EE and the random element flag RE (g) of the g-th group being processed for each bit (S81). The element data indicated by the bit whose result is “1” due to the logical product satisfies the requirements of the note range and velocity range of the random element in the g-th group.

Therefore, if there is a bit whose result is “1” (S82), an element used for the current sound generation is randomly selected from these elements (S83). At this time, it may be possible to set a ratio at which each selection candidate is selected instead of being completely random.
After this selection, as in the case of steps S54 and S55 of FIG. 7, assignment of sounding channels and sound information for sounding using the selected element are performed (S85, S86). Return to the process.
As a result of the above processing, if there is a random element in the g-th group, one to be used for sound generation can be selected from the random elements, and settings necessary for sound generation using that element can be made. Note that the random and cycle sound types are the sound types of the selected sound.

  This is the end of the description of the processing executed when a note-on event is detected. By these processing, the element data used to generate the musical sound corresponding to the detected note-on event according to the rules indicated by the pronunciation type of each element. The musical tone signal generation circuit 20 can be instructed to generate an element sound waveform based on the element data. In this case, the rule indicated by the pronunciation type is applied to each group of elements. In these processes, the CPU 11 functions as an element determination unit.

Next, FIG. 10 shows a flowchart of a sound generation instruction process corresponding to the event, which is executed when the CPU 11 detects a note-off event (an event instructing the start of decay of a musical sound). This process also proceeds while referring to and rewriting each data shown in FIG.
When the CPU 11 detects that a note-off event has occurred by being generated according to a user's performance operation or generated by automatic performance processing, the CPU 11 starts the processing shown in the flowchart of FIG.

  First, data of part p and note number n which the detected note-off event has is acquired (S91). Then, information on the sound generation channel that produces the sound corresponding to the part p and the note number n and the element that is used for the sound on that channel are acquired (S92). This pronunciation is a pronunciation corresponding to the note-on event (note-on event immediately before the same note number in the same part as the detected note-off event) corresponding to the detected note-off event. An element may be discovered. In addition, all the pronunciations corresponding to the note-on event have already been stopped, and there is a case where no pronunciation channel is found.

  After step S92, the group number register g is sequentially increased from 1 to 8, and the key-off sound generation process (S94) is executed for each of the first to eighth groups to use the sound corresponding to the note-off event. The tone generation channel of the tone signal generation circuit 20 is assigned, and settings necessary for tone generation are made in the register of the tone generation channel assigned by the tone generator register 21 (S93 to S96).

If YES is determined in the step S96, the musical sound signal generation circuit 20 is instructed to start the sound generation of all sounding channels set to sound in the step S94 (S97). In this case, the CPU 11 and the tone signal generation circuit 20 function as element sound generation means.
If a sound generation channel is found in step S92 (S98), the musical sound signal generation circuit 20 is instructed to start attenuation (release start) of the sound generation channel (S99), and the process is terminated.
Through the above processing, the tone signal generation circuit 20 starts to generate a tone corresponding to the note-off event, and attenuates the tone being generated according to the corresponding note-on event according to the release setting in the EG parameter. Can be started.

Next, FIG. 11 shows a flowchart of the key-off sound generation process in step S94.
In this process, the CPU 11 first determines whether or not there is a bit of “1” in the key-off sounding element flag OSE (g) of the g-th group (S101). That is, it is determined whether or not the g-th group has an element for key-off pronunciation.
If there is, the validity of each element is determined (S102). This criterion is determined to be valid when the element on / off of the target element is “ON” and the note number n is in the note range. Here, the velocity is not considered, but the velocity v of the corresponding note-on event may be stored, and the value may be within the velocity range or may be an effective condition.
In the processing so far, the CPU 11 functions as an element determination unit.

Further, after the above determination, it is determined whether or not there is an effective element (S103). If there is, each element is sequentially processed (S104, S112, S113), and the sound generation setting in steps S105 to S111 is performed. Execute the process.
In this part, first, the sound generation using the element to be processed is assigned to the sound generation channel (S105).

  Thereafter, it is determined whether there is one or more handover information in the handover list HL (g) of the g-th group stored in step S19 of FIG. 4 (S106). If this is not the case, the note information n and the musical tone information based on the element to be processed are set in the sound generation channel assigned in step S105 without taking over the envelope information (S107). In this case, according to the note-off event, sound generation using the key-off sound generation element is performed independently. Even when there is no pronunciation corresponding to the note-on event, the pronunciation according to the note-off event is possible.

On the other hand, if there is takeover information in step S106, it is further determined whether or not there is takeover information indicating the note number n in the takeover information (S108). If not, the element to be processed is used. If the sound generation is stopped (S109), the current envelope waveform value is acquired from the sound generation channel indicated by the inheritance information (S110), and the acquired envelope value, the note number n, and the sound generation channel assigned in step S105 are acquired. The tone information based on the element to be processed is set (S111).
In either case, the process proceeds to step S112. If NO is determined here, the process returns to the original process.

  As a result of the above processing, if there is an element for key-off pronunciation in the g-th group, it is possible to search for an element satisfying the conditions used for pronunciation and to make settings necessary for pronunciation using that element. In this case, the envelope waveform value of a specific element sound among the g-th group element sounds generated in response to the note-on event can be taken over. Furthermore, even when there are a plurality of elements used for key-off sounding, the correspondence between the takeover source and the takeover destination can be easily set by grouping the elements.

Note that the key-off sound that is sounded in response to the note-off event is generally an element sound with a short sounding time, so even if there is no instruction to start attenuation, it automatically disappears after a predetermined time from sounding and the sounding ch Opened.
Key-off sounds like the ones mentioned above may occur when playing a musical instrument, such as the sound of a damper hitting a vibrator when a damper is applied to the vibrator at the time of key-off to attenuate the vibration. This is a function provided to reproduce this. However, the application is not limited to this. Further, the inherited musical sound characteristic is not limited to the value of the envelope waveform.

Here, taking over of envelope information will be described with reference to FIG.
Note that the envelope values in this figure are obtained by converting (scaling down) an envelope waveform of a decibel scale (13 bits to 16 bits: about 10000 steps) into a MIDI volume parameter scale (7 bits: 128 steps). is there. This figure schematically shows the characteristic part of the envelope waveform. Actually, the length of the key-off sound is much shorter than that of the key-on sound, and each envelope waveform is a little more complicated.

First, the handover of envelope information refers to the value of the envelope waveform used for the processing in the envelope processing unit 24 for the sound being sounded, and the envelope waveform used for the processing in the envelope processing unit 24 for the newly sounded sound. Is reflected in the value of.
For this purpose, the value of the envelope waveform (value at the time of occurrence of the note-off event) acquired from the sounding channel that is sounding may be simply used as the volume parameter of the sound that is newly generated as it is. This volume parameter is included in the EG parameter and controls the maximum value of the generated envelope waveform. Alternatively, the obtained envelope waveform value may be multiplied by a predetermined coefficient (0 or more and 1 or less), and the multiplication result may be used as a volume parameter. Of course, other methods may be used.
FIG. 12 shows an example of the former case or an example of the latter case where the predetermined coefficient is “1”.

  (A) shows an example of the case where the envelope value is kept at the maximum value 127 (converted value) until note-off, as indicated by the solid line 41, for the sound generation corresponding to the note-on event. In this case, in step S111 in FIG. 11, an EG parameter for generating an envelope waveform having a maximum value of 127 (converted value) is set. The thick line 42 shows the time change of the envelope value in this case.

On the other hand, (b) shows a case where the envelope value attenuates from the maximum value 127 (converted value) to 64 (converted value) by note-off as shown by the solid line 43 for the sound generation corresponding to the note-on event. It is an example. In this case, in step S111, an EG parameter for generating an envelope waveform having a maximum value of 64 (converted value) is set. The thick line 44 shows the time change of the envelope value in this case. For comparison, the same time change as the thick line 42 in FIG.
As can be seen from these figures, by performing such a takeover, the characteristics of the sound produced in response to the note-on event and the sound produced in response to the corresponding note-off event are continuously changed. It can be made into a natural sound.

  Further, by performing the processing described above, the synthesizer 10 performs sound generation control based on the sound generation type for each of the elements belonging to the same group among a plurality of elements included in the voice data, so that two or more different These elements can be associated with each other to control pronunciation. In particular, select which non-legato element from multiple elements to which legato element to legato, which note-on sound to which note-off sound will carry over the musical sound characteristics, in order or randomly. It can be easily set by group setting whether to use for pronunciation.

In addition, by making it possible to set the note range and velocity range for each element, the number of element sounds that can be pronounced during cycle pronunciation and random pronunciation (replacement pronunciation) is set as the sound generation instruction event (note-on event). It can be different depending on the range of pitch (note number) and performance intensity (velocity). The element sound for non-legato sound generation and the element sound for legato sound generation can be switched independently of each other according to the pitch of the sound generation instruction event and the range of performance intensity, and the number can be freely selected. The same applies to the element sounded when the element selection switch is operated and the element sound sounded when the element selection switch is not operated. Therefore, various timbre changes can be easily obtained.
If a normal sound (non-replacement sound) sound generation type is prepared, a replacement sound or switch-corresponding sound can be generated for only an arbitrary element among a plurality of elements.

  In addition, when a note-on event related to a legato performance occurs, depending on whether the element sound corresponding to the previous note-on event continues in the note-on state, the pronunciation corresponding to the generated note-on event is legato-sounded. And non-legato sound generation, so that it produces a more natural legato sound than the conventional method of determining whether or not a legato sound is generated only by whether or not the note-on event is related to legato performance. Can be made.

Next, a specific example of selecting an element to be used for sound generation according to the process shown in FIG. 4 will be described with reference to FIGS.
In these figures, “eleNO” indicates an element number, “GID” indicates a group ID, and “Type” indicates an element pronunciation type. In addition, the arrangement of the boxes described on the right side of these columns indicates the respective notes from note numbers 60 (C3) to 78 (F # 4), and the filled boxes use elements for the pronunciation of the notes. It shows that. The hatched box indicates that the note range is set so as not to include the range, and the element is invalidated within the range.
Further, these drawings show examples in which performance operations are performed in ascending order of note numbers, but it goes without saying that the order of performance of notes is arbitrary in the actual performance operation.

First, FIG. 13 and FIG. 14 show examples in the case where legato elements exist. In these figures, elements when a note-on event of legato performance occurs in this order on notes (E3, B3, E4) at three positions A (64), B (71), and C (76). An example of selection is shown.
FIG. 13A shows the most typical example, in which one normal (non-legato) element and one legato element are provided in one group. In this case, at the first note-on event, the normal element 1 is used for sound generation, and after that, the legato performance is performed, so that the normal element is used and the legato element 2 is used for sound generation.

(B) shows a case where a normal element is restricted by a note range. Even in this case, the legato element can be used at the time of the note-on events B and C, and the condition that the non-legato or legato element is used at the time of the previous note-on event is satisfied. The sound is produced using similar elements.
(C) shows a case where a legato element is constrained by a note range. In this case, there is no valid legato element in the group at the time of the note-on events B and C, and the determination in step S59 in FIG. 7 is NO, so that sound generation using normal elements is performed.

  (D) shows a case where the group is different between the normal element and the legato element. Since the element selection rule is applied to each group, there is no valid legato element in group 0. Therefore, as in the case of (c), even when the legato performance is performed, all the normal elements are used for sound generation. Group 1 has only legato element 2. Therefore, since the condition that a non-legato or legato element is used at the time of the previous note-on event is not satisfied, the element 2 is not used for sound generation.

FIG. 14A shows an example in which two normal and legato elements are provided in group 0, and FIG. 14B shows an example in which two normal elements and three legato elements are provided. It was. In this way, even if the number of elements in the group changes, the normal element is used for the sound corresponding to the first note-on event A, and the legato element is used for the sound corresponding to the note-on events B and C by the legato playing technique. There is no change in using.
In these cases (a) and (b), it is possible to generate musical tones having different timbres by synthesizing a plurality of element sounds during normal performance and legato performance. Also, by simply changing the group and pronunciation type settings for each element, the number of element sounds to be synthesized can be changed individually for normal performance and legato performance, making editing voice data easy and high It can be said that it can be done with a degree of freedom.

For example,
Element 1 (group 1, normal): 1st flute normal sound Element 2 (group 1, legato): 1st flute If there is flute tone data like legato sound, add an element to this,
Element 1 (Group 1, Normal): First Flute Normal Sound Element 2 (Group 1, Legato): First Flute Legato Sound Element 3 (Group 2, Normal): Second Flute Normal Sound Element 4 (Group 2, Legato) : 2nd flute It is possible to easily make a setting for changing to two flute tone data like legato pronunciation.
Although not shown in this figure, element sounds that are synthesized separately for note-on and note-off pronunciations can also be obtained for key-off pronunciation types by simply changing the group and pronunciation type settings for each element. Similarly, it can be said that voice data can be easily edited with a high degree of freedom.

  (C) shows an example in which a part of the legato element is restricted by the note range. In this case, only legato elements that fall within the note range are used for pronunciation during legato performance. Therefore, it is possible to cope with changing the tone color of the sound to be generated during legato performance depending on the sound range.

  (D) shows an example in which normal and legato elements are provided for group 0 and group 1, respectively, and for group 1, normal elements are restricted by the note range. Since the element selection rule is applied to each group, the pronunciation related to group 0 is performed in the same manner as in (a). However, for group 1, at the first note-on event A, there is no valid normal element, so no sound is generated. Then, at the next note-on event B, even if it is a legato performance, the condition that the non-legato or legato element is used at the time of the previous note-on event is not satisfied, so the normal element is used.

Next, FIG. 15 shows an example in the case where there are cycles and random elements as replacement sounds. In these figures, notes (C # 3, D # 3, F3, G3, A3, B3, C # 4, D # 4, F4) at nine positions A (61) to H (77) An example of element selection when note-on events occur in this order is shown.
(A) shows an example in which the sound generation type of the cycle is set for the four elements of group 0. In this case, each time a note-on event occurs, the elements used for pronunciation are selected in order from the smallest number in the group, and when the element with the largest number is selected, the element with the smallest number is returned. Even if the element numbers are not consecutive, there is no problem in operation.

  (B) shows an example in which two cycle elements are provided in group 0, three cycle elements are provided in group 1, and one normal element that is a non-replacement sound is provided. In this case, every time a note-on event occurs, an element used for sound generation is selected for each group in numerical order. Since the number of elements in the cycle is different for each group, the selection cycle is also different for each group. The normal element is always used for sound generation regardless of the cycle element.

  In this way, the synthesizer 10 can generate a sound with a highly novel timbre in which the timbre changes periodically by simply specifying a cycle as a sound generation characteristic for the element. In addition, the element selection cycle can be changed simply by changing the number of cycle elements in the group. Furthermore, considering the combination of pronunciation by group 0 and pronunciation by group 1, it is possible to generate sounds by switching between six timbres with five elements periodically, allowing for a variety of pronunciations with a small amount of data. It can be said that.

  (C) shows an example in which a part of a cycle element is restricted by a note range. In this case, in the range where the element is invalidated by the setting of the note range, the element is selected on the assumption that the element does not exist. Therefore, the selection candidate changes depending on the note number of the note-on event.

(D) shows an example in which random elements are provided. In this case, the operation is the same as in the case of the cycle except that an element used for sound generation is randomly selected. However, this randomness can produce pronunciations that are more varied and surprising than in the case of cycles.
It is possible to have both a cycle element and a random element in the same group, but in this case, the cycle is a cycle, the random is random, and one element is selected from each of the elements with the same pronunciation type. Select the elements to be used for pronunciation.

Next, FIG. 16 shows an example when there is an element corresponding to the switch. These figures show an example of element selection when note-on events occur in this order in the notes (D # 3, G3, B3, D # 4) at four positions A (61) to D (75). Is shown. Further, it is assumed that the state of the element selection switches A and B when the event occurs is as shown by “SW_A” and “SW_B”.
In the case of the example shown in (a), a sound generation type corresponding to a switch is set for the elements 1 to 6. Whether these elements are used for sound generation is determined according to the state of the element selection switch. For normal elements, the state of the element selection switch is not relevant. Of course, the restrictions by the note range and velocity range are reflected.

  (B) shows an example in which the legato sound type is set in the elements 7 and 8 and the note-on event is based on the legato playing technique. The switch-compatible pronunciation type is also “non-legato”, so if there are legato elements in the same group and the usage conditions are met, the legato elements are used for pronunciation instead of the switch-compatible elements. To do. Whether or not legato is used is irrelevant to the state of the element selection switch, but the restrictions due to the note range and velocity range are reflected.

Using such switch-compatible elements, it is possible to easily reproduce playing styles, turning on / off the chorus effect, adding instruments to the performance section, and the like.
For example, for guitar sounds
Element 1 (Group 1, all switches off): Attack noise Element 2 (Group 1, all switches off): Normal sounding Element 3 (Group 1, switch A on): Harmonics sounding Element 4 (Group 1, switch B on): If a mute tone is provided, the normal selection method, the harmonics performance method, and the mute performance method can be easily switched by the operation of the element selection switches A and B to obtain a tone that can be played.

In the organ tone,
Element 1 (Group 1, Normal): Percussion sound Element 2 (Group 1, all switches off): Chorus off sound Element 3 (Group 1, Switch A on): Chorus on sound is provided, while percussion sound is based The chorus can be switched to ON / OFF by turning ON / OFF the element selection switch A.

In the string instrument section tone,
Element 1 (Group 1, Normal): Violin Tone Element 2 (Group 1, Switch A On): Cello Tone Element 3 (Group 1, Switch B On): If a contrabass tone is provided, the violin performance is the basis. By pressing the element selection switches A and B, it is possible to obtain a tone color to which a cello or a contrabass can be added.

In the orchestra section tone,
Element 1 (Group 1, Normal): String instrument timbre Element 2 (Group 1, switch A on): Wind instrument timbre Element 3 (Group 1, switch B on): Timpani timbre provides a base for the string instrument part. By pressing the selection switches A and B, it is possible to obtain a tone color that can add wind instrument parts and timpani parts.

Also, in any case, the part which is one element in the above example can be constituted by a plurality of elements, and in this way, a plurality of pronunciations corresponding to one rendition, instrument, part, etc. It is possible to synthesize element sounds.
For example, in the above string instrument section timbre, if you want to be able to pronounce legato for each instrument, use element grouping,
Element 1 (Group 1, Normal): Violin tone (Normal)
Element 2 (group 2, switch A on): Cello tone (normal)
Element 3 (group 3, switch B on): Contrabass tone (normal)
Element 4 (Group 1, Legato): Violin tone (legato)
Element 5 (Group 2, Legato): Cello tone (Legato)
Element 6 (Group 3, legato): Contrabass tone (legato)
With this, the legato sound can be generated while maintaining the addition of the musical instrument by pressing the element selection switch.
Such a setting change can also be made by simply adding an element or changing the pronunciation type and group. In this respect, too, it can be said that the synthesizer 10 can edit voice data easily and with a high degree of freedom. .

This is the end of the description of the embodiment, but it goes without saying that the configuration of the apparatus, the configuration of data, the contents of voice data and element data, specific processing contents, etc. are not limited to those described in the above-described embodiment. is there.
For example, in the above-described embodiment, the envelope value is inherited only when the key-off sounding element is used for sounding. However, when other elements are used for sounding (when sounding according to the note-on event is performed). ) May also be performed. In particular, when a legato sound is generated, when a portamento is set, or when a sound is generated using a monomode element, it is preferable that the envelope value can be taken over.

Further, the number of elements included in one voice data and the number of groups that can be set are not limited to eight. Further, the number of elements may be variable for each voice data. The grouping itself is not essential.
Furthermore, the rules used when determining whether or not to use the element data to generate the musical sound corresponding to the performance event, that is, the musical sound corresponding to the performance event, are not limited to those described above. On the contrary, it is not necessary to prepare all the pronunciation types described above. Even in this case, the effect for each pronunciation type can be obtained.

Further, it is not always necessary to use both the note range and the velocity range for the limit data, and only one of them may be used.
Also, note on (off) events include parameters other than note number and velocity, such as attack rate data (data for controlling the rising speed), brilliance data (data for controlling brightness), depth data (fluctuation) Data for controlling the depth of the image may be included. In this case, the range in which the element is used for sound generation may be limited based on parameter values other than the note and velocity.

Or, conversely, a configuration in which no limit data is used, that is, an element used for sound generation is not limited by the limit data is also possible.
Regardless of the case, a sound generation control event is performed according to the rules according to the sound generation type for each sound generation, which element is used for sound generation among the multiple elements included in the voice data for one tone color. By selecting and referring to the contents of the above, it is possible to generate a richly varied musical tone in which the tone changes dynamically for each pronunciation. For this reason, necessary settings and control can be made relatively simple.

In the above-described embodiment, the normal and switch sound generation types are set to non-legato, while the random and cycle sound generation types are neither non-legato nor legato. However, which pronunciation type is non-legato and whether or not the pronunciation is stopped when the legato sound is pronounced, or which pronunciation type is legato and is pronounced when the legato is played, is not limited to that described here. It can be arbitrarily determined.
The user may be able to set whether each sound generation type is non-legato, legato, or irrelevant to the legato playing style. Further, instead of setting for each sound generation type, each element may be set to non-legato, legato, or unrelated.

  In addition, regarding the legato sound generation, in the above-described embodiment, even if the legato sound element exists and GS (g) = 1, the note number related to the note-on event is the range of the note range related to the legato sound element. In the other case (NO in step S59 in FIG. 7), a non-legato sound is generated instead of a legato sound. However, in this case, neither legato sound nor non-legato sound may be generated.

  Further, if the generated note-on event is related to legato performance in the mono mode (YES in step S42 in FIG. 4), the previous sound is started unconditionally. However, instead of starting the release, dumping (rapid decay) may be performed. Even in legato performance, when there is no element to be used for sound generation in response to note-on and no new sound is generated (steps S53 and S59 in FIG. 7, step S72 in FIG. 8, step S82 in FIG. 9). If all of the above are NO), the previous sound may not be released.

  In the above-described embodiment, an example in which two switches A and B are provided as the element selector switch has been described. However, the number of switches may be an arbitrary number of one or more. In addition, when multiple switches are turned on at the same time, all elements whose sound generation type is “switch on” for any one of the switches are used for sound generation, but depending on the combination of switches that are turned on Alternatively, a “switch-on” pronunciation type may be prepared, and only elements for which the combination of the turned on switches completely matches the pronunciation type may be used for pronunciation.

In the above-mentioned embodiment, the non-replacement sound (the sound that is not replaced during sound generation) is a normal sound, legato sound, switch-on sound (switch A) that is not subject to cycle sound generation or random sound generation. ON sound and switch B ON sound) and switch OFF sound correspond to this. However, it is not necessary to mount all of these pronunciation types in the musical tone generation device, and it is sufficient that at least one of them is mounted.
Regarding the non-legato sound, in the above-described embodiment, the normal sound, the switch-on sound, and the switch-off sound are collectively used as the non-legato sound. However, at least one of these may be implemented as the non-legato sound.

The note-on sound is a sound that is a source of the characteristic (amplitude envelope, etc.) to the note-off sound. In the above-described embodiment, the normal sound, the switch-on sound, the switch-off sound, the legato sound, the cycle sound, Random sounds fall under this category. However, it is not necessary to mount all of them, and it is sufficient that at least one of them is mounted.
In addition, the continuous sound is a sound that is generated regardless of the operation state of the element selection switch. In the above-described embodiment, the sound generation type that branches to NO in step S34 in FIG. 5, that is, a normal sound, a legato sound, and a cycle. Sounds and random sounds fall under this category. However, it is not necessary to mount all of them, and it is sufficient that at least one of them is mounted.

  Further, in the above-described embodiment, the selection of the element that inherits the envelope information in the sound generation at the time of note-off is performed in step S19 in FIG. However, no selection is made at that time, and in step S106 of FIG. 11, which is processing in response to the occurrence of the note-off event, for each element group, the time of the element sound that has been started to sound at the corresponding note-on event. It is also possible to select the one having the largest envelope waveform value and take over the value.

  In the above-described embodiment, the synthesizer 10 having a keyboard has been described as an example. However, the present invention is not limited to an electronic musical instrument having a performance operator other than a keyboard, a sound source device having no performance operator, or musical tone generation. Needless to say, the present invention can be applied to other musical tone generation apparatuses such as expansion boards and chips for use. Further, a device that uses data other than the MIDI format as a performance event for driving the musical tone signal generation circuit 20 may be used.

Further, the program of the present invention is a program for causing a computer to function as a musical sound generation device and realizing the above-described functions, and is stored in advance in a ROM, an HDD, etc., a CD-ROM, a flexible disk, etc. This program is recorded on a non-volatile recording medium (memory) and provided, and the program is read from the memory to the RAM and executed by the CPU, or an external device or a program including the recording medium storing the program is stored in an HDD or the like. The same effect can be obtained even when downloaded from an external device stored in the means and executed.
In addition, the configurations and modifications described above can be applied in appropriate combinations within a consistent range.

As is apparent from the above description, according to the musical tone generating apparatus or program of the present invention, in the musical tone generating apparatus capable of editing timbre data having a plurality of element data that defines a plurality of element sounds , a dynamic sound is generated for each pronunciation. It is possible to generate a variety of musical tones that vary in tone.
Therefore, by applying the present invention, it is possible to provide a musical tone generating apparatus that can generate a variety of timbres .

It is a figure which shows the outline of the synthesizer which is embodiment of the musical sound production | generation apparatus of this invention. It is a figure which shows the structure of the voice data for 1 tone color in the synthesizer shown in FIG. It is a figure which similarly shows the structure of the data for pronunciation management. It is a flowchart of the sound generation instruction | indication process according to the note on event which CPU shown in FIG. 1 performs. 5 is a flowchart of validity determination processing shown in FIG. 4.

It is a flowchart of a mono mode process similarly. It is a non-legato & legato pronunciation processing flowchart. It is also a flowchart of cycle sound generation processing. It is also a flowchart of a random sound generation process. It is a flowchart of the sound generation instruction | indication process according to the note-off event which CPU shown in FIG. 1 performs.

It is a flowchart of the key-off sound generation process shown in FIG. It is explanatory drawing about taking over of envelope information. It is a figure which shows the specific example of selection of the element used for pronunciation according to the process shown in FIG. It is a figure which shows the other example. It is a figure which shows another example. It is a figure which shows another example.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Synthesizer, 11 ... CPU, 12 ... ROM, 13 ... RAM, 14 ... MIDI_I / F, 15 ... Performance operator, 16 ... Timer, 17 ... Display, 18 ... Panel operator, 19 ... Storage device, 20 ... Music signal generation circuit, 21 ... sound source register, 22 ... waveform data read processing unit, 23 ... filter processing unit, 24 ... envelope processing unit, 25 ... ch accumulation unit, 26 ... effect processing unit, 31 ... waveform memory, 32 ... DAC, 33 ... Sound system

Claims (11)

A musical sound generating device that generates a plurality of element sounds based on a plurality of element data, and synthesizes the plurality of generated element sounds to generate one musical sound,
A plurality of element data corresponding to a plurality of element sounds constituting a musical tone of one tone color, each corresponding to a characteristic control data for controlling a musical sound characteristic of the corresponding element sound, and a selection rule for each element data Storage means for storing a plurality of element data including pronunciation type data indicating one of the plurality of pronunciation types
Editing means for changing a setting of pronunciation type data indicating a desired one of the plurality of pronunciation types for each of the plurality of element data stored in the storage means in response to an instruction from a user;
In response to the generation of a sound generation control event that instructs the sound generation control of one musical sound and includes a parameter that controls the characteristics of the one musical sound, the generated sound control event and each of the plurality of element data Element selection means for selecting one or more element data to be used for generating an element sound from the plurality of element data based on the pronunciation type indicated by the pronunciation type data of
Based on one or a plurality of element data selected by the element selection means, for each element data, a musical tone corresponding to the characteristic control data included in the element data and the parameter included in the generated sound generation control event A musical sound generating apparatus comprising element sound generating means for generating an element sound having characteristics. The musical sound generating device according to claim 1,
The musical sound generating apparatus according to claim 1, wherein the editing means is means for editing the characteristic control data for each of the plurality of element data in accordance with an instruction from a user. The musical sound generating device according to claim 1 or 2,
The pronunciation control event is a note-on event,
The plurality of pronunciation types include non-replacement sounds and replacement sounds,
The element selection means has a first element selection means and a second element selection means,
The first element selection means uses element data in which the pronunciation type data indicates the non-replacement sound among the plurality of element data for generating the element sound in response to the occurrence of the note-on event. Select as element data,
The second element selection means is configured according to a predetermined rule or at random from element data in which the pronunciation type data indicates the replacement sound among the plurality of element data in response to the occurrence of the note-on event. Any one of the element data is selected as element data to be used for generating the element sound. A musical sound generating device according to claim 3,
The musical sound generating apparatus, wherein the non-replacement sound is a normal sound, a legato sound, a switch-on sound, or a switch-off sound. The musical sound generating device according to claim 1 or 2,
The pronunciation control event is a note-on event,
The plurality of pronunciation types include legato and non-legato sounds,
In response to the occurrence of the note-on event, the element selection means may be configured that the element sound generated based on the element data of the non-legato sound or the legato sound according to the note-on event that occurred last time is still in the note-on state. (1) If the note-on state is still present, element data in which the pronunciation type data indicates the legato sound among the plurality of element data is used as element data used for generating the element sound. (2) If the note-on state is no longer present, the element data in which the pronunciation type data indicates the non-legato sound among the plurality of element data is selected as element data used to generate the element sound. A musical sound generator characterized by. The musical sound generating device according to claim 5,
The musical sound generating apparatus, wherein the non-legato sound is a normal sound, a switch-on sound, or a switch-off sound. The musical sound generating device according to claim 1 or 2,
The pronunciation control events are a note-on event and a note-off event,
The plurality of pronunciation types include a note-on sound and a note-off sound,
The element selection means has a first element selection means and a second element selection means,
The element sound generating means includes first element sound generating means, element sound specifying means, and second element sound generating means,
The first element selection means uses element data in which the pronunciation type data indicates the note-on sound among the plurality of element data for generating the element sound in response to the occurrence of the note-on event. Select as element data,
The first element sound generating means generates the element sound having a musical sound characteristic according to characteristic control data included in the element data selected by the first element selecting means,
The second element selection means uses, in the generation of the element sound, element data in which the pronunciation type data indicates the note-off sound among the plurality of element data in response to the occurrence of the note-off event. Select as element data,
The element sound specifying means specifies one element sound as a characteristic takeover source from among the element sounds generated by the first element sound generating means,
The second element sound generation means takes over the musical sound characteristics of a part of the element sound specified by the element sound specification means, and includes characteristic control data included in the element data selected by the second element selection means. A musical sound generating apparatus characterized by generating the element sound having a corresponding musical sound characteristic. It is a musical sound production | generation apparatus of Claim 7, Comprising:
A musical sound generating apparatus, wherein the note-on sound is a normal sound, legato sound, random sound, cycle sound, switch-on sound, or switch-off sound. The musical sound generating device according to claim 1 or 2,
It has a switch that can be switched between the off state and the on state by user operation,
The pronunciation control event is a note-on event,
The plurality of pronunciation types include a switch-off sound, a switch-on sound and a constant sound,
In response to the occurrence of the note-on event, the element selection means (1) if the switch is in an off state at the time of occurrence of the note-on event, the sound generation type data among the plurality of element data Selecting element data indicating the switch-off sound as element data to be used for generating the element sound, and (2) when the switch is on when the note-on event occurs, The element data in which the pronunciation type data of the element data indicates a switch-on sound is selected as element data used for generating the element sound, and (3) regardless of the on / off state of the switch. Among the element data, the sound type data indicates the element data indicating the continuous sound. The musical tone generating apparatus is also characterized in that data is also selected as element data used for generating the element sound. The musical sound generating device according to claim 9, wherein
A musical sound generating apparatus, wherein the continuous sound is a normal sound, a legato sound, a random sound, or a cycle sound.   The program for functioning a computer as a musical sound production | generation apparatus as described in any one of Claims 1 thru | or 10.

Images