JP2003216151A - Tone generation apparatus and method for the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To easily control a fine waveform control in a partial section. <P>SOLUTION: Each segment template is arranged so that at least parts of characteristic variation curves in the segment templates of different time lengths overlap with each other. A synthesized envelope for a desired section is produced by combining all arranged characteristic variation curves with an envelope based on inputted performance information. A tone generation means generates a tone by using the synthesized envelope thus formed. Thereby, a user can generate the tone only by changing the combination of the segment templates of different time lengths while easily carrying out the fine control in the partial section. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tone generating apparatus and method for generating a tone, a voice, or any other tone. In particular, the present invention relates to a musical tone generating apparatus and method for generating musical tone waveforms with rich expressiveness by controlling various envelopes of waveform data in accordance with a rendition style. INDUSTRIAL APPLICABILITY The present invention is applicable not only to electronic musical instruments, but also to automatic playing devices, computers, electronic game devices, and other multimedia devices in any field or device or method of any field having a function of generating a musical sound or voice or any other sound. It can be applied in a wide range.
In this specification, the musical tone waveform is not limited to a musical sound waveform, but may be a voice or any other arbitrary sound waveform.

[0002]

2. Description of the Related Art PCM (Pulse Code Modulation) or DP
CM (differential PCM) or ADPCM (adaptive differential PCM)
The waveform data of one or a plurality of cycles encoded by an arbitrary encoding method such as is stored in the waveform memory, and an appropriate musical tone waveform is generated by repeatedly reading this waveform data at a desired pitch. The so-called “waveform memory read” technique is already known, and various types of “waveform memory read” techniques are also known. In the sound source using such "waveform memory read" technology,
By not only outputting the waveform data stored in the memory as it is and outputting it as a musical tone, but also controlling the read waveform data for each musical tone element such as pitch, volume, tone color, etc. It has been performed conventionally. For controlling waveforms in such a sound source, for example, a pitch EG built in the sound source
There are various EGs (Envelope Generators) such as an amplifier (amplitude) EG or a filter EG. The pitch EG controls the pitch envelope relating to the pitch (that is, the pitch), and by appropriately changing the reading speed of the waveform data according to an arbitrary pitch envelope,
Generates a musical tone waveform whose pitch changes over time. Amplifier EG
Forms a required envelope waveform for volume (that is, volume amplitude envelope), and adds the volume amplitude envelope to the read waveform data to control the volume from the start to the end of sound generation. Also, the filter EG
Controls the filter characteristics for tone color control. By processing the read waveform data with a filter whose characteristics are controlled by the filter EG, a tone waveform whose tone color changes with time is generated.

In addition to the various EGs described above, waveform control is performed to suppress the performance of a musical piece (for example, expression (expression), Pitch bend, Mo).
dulation depth, Modulati
Each time the user wants to change the on Speed (modulation speed) etc., the user selects a specific control element (eg, expression pedal, expression wheel, bend wheel, modulation whe).
Conventionally, continuous control is known in which the pitch, volume, timbre, etc. are continuously changed by operating el (modulation wheel). The various control values set by the continuous control can be stored in the sequencer, and the stored various control values can be appropriately edited on the sequencer. In addition to this, waveform control by a rendition style template function provided as a unique function by a sequencer has been conventionally known. This is because, when a music symbol is specified, macro pattern data (also called a rendition style template) corresponding to each music symbol prepared in advance is added to the music data, whereby the pitch
It controls the volume and timbre. The musical symbols for which the rendition style template is prepared include, for example, dynamic symbols (for example, crescendo, diminuendo, piano, mezzo, etc.).
Forte, etc.) and end symbol (eg, fermata)
And speed change symbols (such as accelerando, ritardando), or performance and performance symbols (such as glissando, portamento, choking, tremolo, staccato, accent). In order to perform waveform control, a plurality of pitch modulation waveforms of musical tones over the entire sound generation period from the rising edge to the falling edge of one note are stored in advance (for example, pitch modulation waveforms for attack pitch, vibrato, portamento, etc.), and turned on. It has already been filed by the applicant of the present invention to perform control by synthesizing only the pitch modulation waveforms that have been created and changing the pitch (that is, the pitch) of the musical sound based on the synthesized pitch modulation waveforms. (JP-A-60-60693).

[0004]

However, the one that generates a musical tone waveform according to the various waveform controls as described above is
There are problems in the following points. That is, in waveform control by various EGs, it is common to perform control in units of one sound, such as rising and falling of one sound.
There is a problem in that it is not possible to successfully control a gradual change in the waveform in phrase units (for example, in units of one bar). In waveform control with continuous control, continuous control is independent of the notes being played, so control is performed only for a specified partial section such as the attack part of the sound, or one note or There is a problem that it is difficult to control only one phrase.
In the waveform control by the rendition style template function of the sequencer, all the rendition style templates have the same time length (that is, in the case of synthesizing a change curve using a plurality of templates hierarchically divided by the time length). Therefore, there is a problem in that it cannot be used when the expression is desired to be changed for each desired partial section and is not flexible. Further, since the rendition style template generates continuous control, when controlling a voice (tone color waveform) composed of a plurality of elements, it is not possible to give a unique rendition style to each element.
There is also a problem. In the waveform control by the pitch modulation waveform of the musical tone over the entire sound generation period, the pitch modulation waveform that can be turned on and off controls the whole one sound from the rising to the falling of one sound, and it is possible to control only the attack part or phrase. However, there is a problem that the control cannot be performed. As described above, most of the conventionally known sound sources have waveform control for generating musical tones that change a lot at predetermined intervals within one tone, and waveforms for generating musical tones that appropriately change over a plurality of tones. Controlling was very difficult.

The present invention has been made in view of the above points and is not limited to waveform control in units of one note, but waveform control for a partial section within one note, or waveform control over a plurality of notes. It is an object of the present invention to provide a musical tone generating apparatus and method capable of generating a musical tone while easily performing. For example, a template including a characteristic change curve to be given to a part of a phrase or a part of a sound is prepared for different time lengths such as a phrase unit, a note unit, an attack part unit, and a release part unit. , By combining a plurality of templates for different time lengths as appropriate to synthesize an envelope and generating a musical tone waveform in accordance with the synthesized envelope, it is possible to easily perform minute waveform control in a partial section. To a musical tone generating apparatus and method.

[0006]

A musical tone generating apparatus according to claim 1 of the present invention supplies performance information supplying means for supplying performance information and segment templates including characteristic change curves, and has different lengths. Segment template supply means capable of supplying a plurality of segment templates of the above and at least a part of characteristic change curves of a plurality of segment templates of different lengths corresponding to a given performance section based on the performance information. And a plurality of characteristic change curves arranged so as to combine to form a synthetic envelope for the given performance section, and for the given performance section using the formed synthetic envelope. And a musical sound generating means for generating the musical sound.

According to the present invention, the characteristic change curves of a plurality of segment templates having different lengths are arranged so that at least some of them overlap, and a synthetic envelope is formed by the combination thereof. If you want to perform fine envelope control in a part,
A characteristic change curve corresponding to a fine envelope having a required length may be pasted. As described above, by simply combining the segment templates having different time lengths, fine envelope control in a part of the performance section can be easily performed.

A musical sound generating apparatus according to claim 2 of the present invention is
A performance information supply means for supplying performance information and a segment template including a characteristic change curve are supplied, which corresponds to a note segment template corresponding to almost the entire length of one note and a part of one note. Segment template supply means capable of supplying a partial segment template, and one note segment template and one or a plurality of partial segment templates corresponding to a given one note performance section based on the performance information, The selected note segment template is arranged in the performance section of the one note, the selected partial segment template is arranged for a part of the performance section of the one note, and the characteristic change curves of the arranged templates are calculated. Form a composite envelope for the given one note playing interval based on combining Comprises that the envelope synthesis means, and tone generating means for generating musical tones for using the formed synthetic envelope playing interval of said given one tone. In this case as well, similar to the above, it is possible to easily perform fine envelope control in a part of the performance section of one note.

The present invention may be constructed and implemented not only as the apparatus invention as described above but also as a method invention. Further, the present invention can be implemented in the form of a program of a processor such as a computer or a DSP, and can also be implemented in the form of a storage medium storing such a program.

[0010]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a block diagram showing an example of the hardware configuration of a musical tone generating apparatus according to the present invention. The hardware configuration example shown here is configured by using a computer, in which the tone generation processing is performed by the computer executing a predetermined software program for realizing the tone generation processing according to the present invention. Be implemented. Of course, this musical tone generation processing is not limited to the form of computer software, and DSP (Digital Signal Processo)
It is also possible to carry out in the form of a microprogram processed by r), and not limited to the form of this kind of program, a dedicated hardware device configured to include a discrete circuit or an integrated circuit or a large-scale integrated circuit, etc. You may implement by a form. Further, the musical sound generating device may take any product application form such as an electronic musical instrument, a karaoke device, an electronic game device, other multimedia equipment or a personal computer. Note that the above-described musical sound generating device may have hardware other than these, but here, the case where the minimum necessary resources are used will be described.

In the hardware configuration example shown in FIG. 1, a CPU 1 as a main control unit of a computer is connected to a read-only memory (ROM) 2 and a random number via a bus line BL (data or address bus). An access memory (RAM) 3, an input device 4, a display device 5, a drive 6, a waveform capturing section 7, a sound source section 8, a hard disk 9, and an input / output interface 10 are connected to each other. The CPU 1 executes various kinds of processing such as “musical sound generation processing” (see FIG. 2 described later) based on a predetermined program. These programs are external electronic musical instruments connected to the communication network via the input / output interface 10, or CDs or MOs mounted on the drive 6.
It is supplied from the external storage medium 6A or the like and stored in the hard disk 9. Then, it is loaded from the hard disk 9 to the RAM 3 at the time of execution. Alternatively, the program may be recorded in the ROM 2 in advance.

The ROM 2 stores various programs and various data executed or referred to by the CPU 1. The RAM 3 is used as a working memory for temporarily storing various data generated when the CPU 1 executes a program, or as a memory for storing a currently executed program and data related thereto. A predetermined address area of the RAM 3 is assigned to each function and used as a register, a flag, a table, a memory, or the like. The input device 4 has an instruction to sample a musical sound, edits sampled waveform data (that is, waveform control), selects various segment templates used in the waveform control (details will be described later), inputs various information, etc. It is configured to include various operators for performing. For example, it is a switch for selecting various segment templates, a numeric keypad for inputting numerical data, a keyboard for inputting character data, or a pointing device such as a mouse. In addition to this, various operators for selecting, setting, and controlling pitch, tone color, effect, etc. may be included. The display device 5 displays various information input by the input device 4, sampled waveform data, waveform data after waveform control, various segment templates, etc., for example, a display such as a liquid crystal display panel (LCD) or a CRT. Is.

The waveform capturing unit 7 has a built-in A / D converter (not shown), samples an analog tone signal externally input from a microphone or the like, converts it into digital data, and converts the digital data into original waveform data ( That is, it is captured on the hard disk 9 as waveform data) which is a material of a musical tone waveform to be generated. The fetched original waveform data is stored in the “waveform database” and the “rendition style database” as vector data and rendition style module according to a predetermined process.
The vector data is not the original waveform data that has been imported, but for each partial waveform that represents the shape of the imported original waveform (for example, attack waveform, body waveform, release waveform, joint waveform, etc.)
It is stored in the waveform database in a data compressed form using a hierarchical compression method. The rendition style module includes data for designating vector data and data necessary for returning the compressed vector data to the original waveform data. Therefore, the rendition style modules are the Entrance-type module, Finish-type module, and Join-type module.
It is stored in the rendition style database as either a t (joint) module or a Body module. That is, the Entrance-type module indicates vector data representing a rising performance section in one note such as an attack part, and the Finish-type module represents vector data representing a falling performance section in one note like a release part. The Joint system module indicates vector data representing a performance section that connects sounds, and the Body system module
It indicates vector data representing a performance section between the Entrance system module and the Finish system module. Entrance-type, Finish-type, and Joint-type rendition style modules point to vector data generated from high-quality waveforms that have characteristics such as rendition style (or articulation), and Body-type modules are NSB (Normal
Short Body: Normal short body and VB (Vibrato
(Body: Vibrato body), etc., indicating vector data generated from a unit waveform (that is, a loop waveform) of a relatively monotonous sound portion having a waveform for one cycle or an appropriate plurality of cycles. The vector data for forming the waveform includes the waveform (timbre) vector, the amplitude (amplitude) envelope vector, and the pitch (pitch).
It is composed of elements such as an envelope vector and a time vector.

The tone generator 8 can generate a series of musical tone waveforms by connecting a plurality of vector data read from the "waveform database" and synthesizing the waveforms, and sends the generated musical tone waveforms to the sound system 8A. When generating this musical sound, the tone generator 8 reads various segment templates from the "segment database" (described later).
Can be used to control the tone waveform for a part of one sound. Since the details of this will be described later, the description thereof is omitted here. In the sound system 8A, the tone signal output from the sound source section 8 is D / A converted to an analog signal and output to the outside. Of course, it is possible to simultaneously output a plurality of tone signals. The hard disk 9 is a storage device for storing various databases such as a rendition style database accumulating various rendition style modules, a segment database accumulating various segment templates, a waveform database accumulating vector data, various programs executed by the CPU 1, and the like. Is. Needless to say, the sound source unit 8 may be realized by a so-called software sound source that generates a musical sound by software.

The drive 6 includes various data such as various rendition style modules and various segment templates, and the C
It drives a removable external storage medium 6A for storing various programs executed by the PU 1.
External storage medium 6A driven by the drive 6
Is a flexible disk (FD), compact disk (CD-ROM / CD-RW), magneto-optical disk (M
O), DVD (Digital Versatile Disk), or any other type of medium that can be used in various removable external storage media. Or
It may be a semiconductor memory or the like. When the external storage medium 6A storing various programs is set in the drive 6, the stored content (that is, the program) may be directly loaded in the RAM 3 without being dropped in the hard disk 9. In addition, using the external storage medium 6A,
Alternatively, the method of providing the program via the input / output interface 10 is convenient because the program can be easily added or the version can be upgraded.

The input / output interface 10 is, for example, L
It is connected to a communication network (not shown) such as an AN, the Internet, or a telephone line, and is connected to a server computer or the like (not shown) via the communication network,
This is a communication interface for loading a program, various rendition style modules, various segment templates, or performance information from the server computer or the like into the musical tone generating device side. That is, when a program, various rendition style modules, various segment templates, etc. are not stored in the ROM 2 or the hard disk 9, it is used to download the programs, various rendition style modules, various segment templates, etc. from the server computer. The musical sound generating device serving as a client transmits a command requesting a download of a program, various rendition style modules, various segment templates, etc. to the server computer via the input / output interface 10. Upon receiving this command, the server computer stores the requested program, various rendition style modules, various segment templates, etc. in the hard disk 9 via the input / output interface 10, thereby completing the download. The input / output interface 10 may be a MIDI interface. In such a case, MIDI performance information is input / output to / from an external MIDI device such as a sequencer or an electronic musical instrument. When using a MIDI interface,
It is also possible to connect a musical performance keyboard and a musical performance operating device and use these to supply MIDI musical performance information to the musical sound generating apparatus in real time.

In the musical tone generating apparatus shown in FIG. 1, the musical tone is generated by the computer executing a predetermined software program for realizing the musical tone generating process. Alternatively, the musical tone generation processing is not limited to the form of such a program, and may be performed in the form of a dedicated hardware device. Therefore, a musical sound generation process executed by the musical sound generation device according to the present invention will be described with reference to FIG. FIG. 2 is a block diagram showing an embodiment in which the tone generation processing is configured in the form of a dedicated hardware device. An outline of the operation of the entire tone generation processing will be briefly described with reference to FIG.

The music piece data reproducing section 1A performs reproduction processing of music piece data with rendition style symbols. That is, first, the music data reproducing section 1A determines the music data with the rendition style symbol (that is, performance information).
To receive. In ordinary music, MI
Musical symbols such as dynamic symbols (such as crescendo and decrescendo) that cannot be used as DI data, tempo symbols (such as Allegro and ritardando), slur symbols, tenuto symbols, and accent symbols are added. Therefore, these musical symbols are converted into data as “playing style symbols”. The MIDI song data including this "playing style symbol" is "music data with playing style symbol", and the song data reproducing section 1
A receives the music data with such a rendition style symbol. The score interpretation unit (player) 1B performs score interpretation processing.
Specifically, the MIDI data included in the received music data with the rendition style symbol and the above-mentioned "rendition style symbol" are converted into predetermined rendition style designation information (for example, rendition style ID and rendition style parameter), and the rendition style synthesizing section (article) together with the time information. Curator) 1
Output to C. The musical score interpreting unit (player) 1B generates an original envelope based on the performance information (hereinafter referred to as an “input envelope” or a “basic envelope”) when generating the rendition style designation information, and a user selection or The various segment templates read from the segment database according to the automatic selection processing are added to this "input envelope" (that is, "basic envelope") to generate a synthetic envelope, and the generated synthetic envelope is used as a rendition style synthesis unit (article). Output as a rendition style parameter to the curator 1C. In general, even with the same musical symbol, the interpretation of the musical symbol differs depending on the performer or musical instrument, and a performance method (that is, performance method or articulation) may be performed for each performer or musical instrument. . Alternatively, the performance may be performed by different performers or musical instruments depending on the arrangement of notes. Therefore, the musical score interpreting unit 1B is a system in which the knowledge for interpreting the symbols (music symbols, arrangement of notes, etc.) on such a musical score is made into an expert system, and the musical score interpreting unit 1B interprets the musical score according to a predetermined standard. Do against. At this time, the musical score interpretation unit 1B interprets the musical score by changing the musical score interpretation method in accordance with the player's designation (for example, who's playing or musical instrument). As one of the different musical score interpretation methods corresponding to such a plurality of players, a plurality of various segment templates are accumulated in the segment database, and the musical score interpretation unit 1B should use the segment template to be used according to the player designation or musical instrument designation from the user. Determine the template.

Rendition style synthesis section (articulator) 1C
Refers to the rendition style database on the basis of the predetermined rendition style designation information (rendition style ID and rendition style parameters) converted by the musical score interpretation section (player) 1B, and selects the packet stream and the rendition style parameter corresponding to the rendition style specification information. A vector parameter relating to the stream is generated and supplied to the waveform synthesizing unit 1D. The data supplied to the waveform synthesizing unit 1D as a packet stream is a vector ID, time information, and the like. At this time, the rendition style synthesis unit (articulator) 1C reads out the rendition style module from the rendition style database based on predetermined rendition style designation information, arranges the rendition style module on the time axis to generate a packet stream, and on the time axis. The synthesized envelope generated by the musical score interpretation unit (player) 1B is distributed to the arranged rendition style modules. The rendition style module is stored in the “rendition style database” on the hard disk 9. For example, one rendition style module is specified by the “rendition style ID”.
The content of a predetermined rendition style module specified according to the "rendition style ID" is given as a rendition style parameter that characterizes or controls the waveform data corresponding to the rendition style module. The waveform synthesizing unit 1D sequentially extracts vector data from the waveform database according to the generated packet stream, deforms the vector data according to vector parameters, and connects the deformed vector data in waveform.
An arbitrary musical tone waveform is generated based on the connected series of vector data. The sound system 8A outputs a musical tone based on the generated musical tone waveform.

The segment template used when the score interpretation unit (player) 1B described above synthesizes the input envelope will be described with reference to FIG. FIG. 3 is a conceptual diagram showing an example of the data structure of the segment template, and FIG.
(Note) Indicates a segment template. Figure 3
FIG. 3B shows an example of a table that specifies an Entrance segment template, and FIG. 3C shows an example of a table that specifies a Finish segment template.
Each of these segment templates is stored in advance in the hard disk 9 or the like as a tabulated "segment database". Of course, these embodiments are merely examples, and needless to say, the present invention is not limited to these.

The Note segment template (hereinafter, simply referred to as Note segment) is control data used to express the intonation of one note. As shown in FIG. 3A, the Note segment is stored on the hard disk 9. It is organized as a table. Each note that composes the table
A segment is composed of a characteristic change curve to be given to one sound as a whole. That is, each Note segment is data having a characteristic change curve of a predetermined shape for each predetermined table number. Each Note segment is classified into several types according to the shape of its characteristic change curve. For example, in this embodiment, it is roughly classified into six types. In other words, table numbers 0 to 7 have a Note segment having a substantially straight characteristic change curve, and table numbers 8 to 15 have a Note segment having a rising characteristic change curve, table number 1
6 to 23 are Note segments with a downward-sloping characteristic change curve, table numbers 24 to 31 are Note segments with a mountain-shaped characteristic change curve, and table numbers 32 to 39 are valley-shaped Note segment having the characteristic change curve of No., and table numbers 40 to 47 are Note segments having the skip type characteristic change curve.

Each note classified according to these types
Each segment has a characteristic change characteristic curve that belongs to each type. For example, the Note segment of table number 8 classified into the upward rising type is data having a characteristic change curve of rising upward with a constant slope, and the Note segment of table number 9 is compared with the slope of the first half part. This data has a characteristic change curve in which the slope suddenly increases from the second half. In addition, the Note segment of table number 24, which is classified as a mountain type, is data having a characteristic change curve having a mountain peak in the first half, and the Note segment of table number 25 has a mountain peak in the center. It is data having a certain characteristic change curve. Note that the Note segment with table number 0 in this embodiment is set as data used by default, and the Note segment has a completely flat shape. That is, when the Note segment of table number 0 is used when changing the input envelope, the original input envelope waveform is reproduced. Note that the Phrase segment template (not shown) is control data used to express the inflection of one phrase (for example, one measure), and the above Note
Since the time length differs from that of the segment template, and the other data structure and the characteristic change curve shape are similar to those of the Note segment template, the description thereof will be omitted.

On the other hand, the Entrance segment template and the Finish segment template (hereinafter simply referred to as Entr
ance segment and Finish segment)
The time length of the data is shorter than that of the segment template, and the Entrance segment expresses the intonation in the specified section of the beginning of one note, so the Finish segment represents the intonation in the specified section of the end of one note. This is control data used for expression, and the Entrance segment and the Finish segment are tabulated and configured on the hard disk 9 like the Note segment described above. Each Entrance segment constituting the table is data including a characteristic change curve and position data (specifically note-on timing) to be given to a predetermined section at the beginning of the ringing within one note. On the other hand, each Finish segment constituting the table is data including a characteristic change curve and position data (specifically note-off timing) to be given to a predetermined section at the end of the ringing within one sound. That is, each Entrance segment and Finish segment is data having a characteristic change curve (not shown) of a predetermined shape for each predetermined table number. The position data is the Entrance segment and Fini.
This data is set at the position corresponding to the shape of each characteristic change curve in the sh segment. For example, in the Entrance segment, position data for a sound with a slow rise is set to an appropriate position from the latter half of the Entrance segment, and position data for a sound with a sharp rise is compared to the first half of the Entrance segment. Is set to an appropriate position. When the position data is set in this way, the Entrance segment and the Finish segment can be arranged on the time axis near note-on and note-off, respectively (for each segment template shown in FIG. 6 to be described later on the time axis). (See Arrangement), so that the attack sound and the release sound can be generated at an appropriate position where the user feels the attack sound and the release sound.

In the Entrance segment, the characteristic change curve of the Entrance segment is the note value (quarter note or less, half note or less, whole note or less, whole note or more), relationship with the previous note (Finish / Joint) and hooking. Strength (strong / medium /
Weak / slur). Hooking strength is the strength of placing a bow on the string when starting to play a bow in the case of string instruments such as violin, the strength of tongue in the case of wind instruments such as sax, and the key pressing in the case of keyboard instruments such as piano. Represents the strength of. In the embodiment shown in FIG. 3 (b),
For example, when the note value to which the Entrance segment is applied is "quarter note or less", the relationship with the preceding note is "Joint", and the hooking strength is "medium",
Entr applied to the Entrance segment of table number 68
It will be selected as the ance segment. On the other hand, Fi
The characteristic change curve of the nish segment is the note value (quarter note or less, half note or less, whole note or less, whole note or more), relationship with the back note (Entrance / Joint) and type (sticky / medium / light / easy / Slur). In the embodiment shown in FIG. 3C, for example, the note value to which the Finish segment is applied is “less than a whole note”,
When the relation with the back sound is "Entrance" and the type is "light", the finish of table number 48
The segment will be selected as the Finish segment to apply. As with the Note segment template, the Entrance segment template and Finish
Also in the segment template, the Entrance and Finish segments of table number 0 are set as data used by default, and when the Entrance and Finish segments of table number 0 are used when changing the input envelope. Will reproduce the original waveform.

Each of these Phrase / Note / Entrance / Finish segment templates has characteristic change curves of dynamics (sustaining performance intensity) / Pitch (pitch) / vibrato depth (vibrato depth) / vibrato Speed (vibrato speed). Hold as a set. However, in the above-described embodiment, for simplicity of explanation, only the segment template regarding dynamics is shown as a representative. That is, FIG. 3A shows an example of a characteristic change curve for controlling dynamics. Pitch / is omitted for each table number so as to correspond to each segment template relating to this dynamics. vibrato depth /
Each segment template relating to vibrato Speed is tabulated in advance and stored on the hard disk 9. Phrase / Note / Entrance / Finish segment templates are dynamics / Pitch / vibrato d
By increasing the dynamics gradually by having a total of 4 types of each characteristic change curve of epth / vibrato speed
It is possible to simultaneously control the envelopes of different musical sound elements by associating them with each other, such as lowering the pitch. Such dynamics / Pi
Phrase / Note / Entrance / Fini with a set of characteristic change curves for tch / vibrato depth / vibrato Speed
Let each segment template of sh be S in this example.
Such an SAT is called an AT (Segment Articulation Template) and is stored in the segment database for each player or for each musical instrument. In this way, even if the SAT with the same table number is designated, a different SAT can be used for each player or musical instrument, and the Phrase / No
dynamics / Pitch / vibrato depth / vibr by simply specifying each segment template of te / Entrance / Finish
Since it is possible to specify all ato Speed characteristic change curves at the same time, it is convenient to specify each one individually.

As mentioned above, Phrase / Note / Entrance
Each segment template of / Finish is data that is hierarchically configured for each time length, and these segment templates are combined as appropriate and combined with the input envelope that is the original envelope possessed by the performance information. As a result, fine envelope control can be performed for each phrase and each sound. That is, in the musical sound generating apparatus according to the present invention, the characteristic change curves are combined with the input envelope by using the Phrase segment and the Note segment in an overlapping manner, and waveform control is performed for each phrase and for each sound based on the combined envelope. To generate an arbitrary tone waveform. Also, set the Entrance segment to the Fini
The sh segment is used by overlapping it with the specified section at the end of the Note segment, and each characteristic change curve is combined with the input envelope, and waveform control is performed for each specified partial section within one note based on the combined envelope. Generate a musical tone waveform of. That is, it is possible to perform waveform control for each phrase and for each sound or for each predetermined section within one sound by appropriately combining and using the segment templates of the hierarchical structure having different time lengths. Thus, it is possible to generate a musical tone waveform with a slight change. Therefore, the waveform control using such a segment template will be described using a specific example.

First, the process of determining each segment template will be described. Phrase and Note segment templates are determined according to user selection. Phrase
For segment templates, Phrase segment
Since the Phrase segment template is valid only for the predetermined performance section starting from the first (including the same time) note-on event data in each predetermined performance section after the template is designated to be used, the user Appropriately select the Phrase segment template to be used for each performance section. A predetermined Phrase segment template is used as a default for the predetermined performance section in which such selection is not made. On the other hand, in the case of the Note segment template, the Note segment template is valid only for one note starting from the first (including the same time) note-on event data after the Note segment template is specified to be used. Therefore, the user appropriately selects the Note segment template to be used for each sound. For one note that is not selected, a default Note segment template is used. For example, if the Note segment template having the complete flat characteristic change curve of table number 0 as shown in FIG. 3A is set as the default, the Note segment template is used. .

On the other hand, the Entrance and Finish segment templates are determined for each sound, according to the selection if the user selection is made, and automatically according to a predetermined process if the user selection is not made. It is determined.
Therefore, the determination processing of such Entrance and Finish segment templates will be described with reference to FIGS. 4 and 5. The process of determining the Entrance segment template will be described with reference to FIG. Figure 4 shows "Entr
15 is a flowchart showing an example of "ance segment template automatic determination processing".

In step S1, the Entrance segment
It is determined whether the template is selectively designated by the table number or the like. If the Entrance segment template is specified (YE in step S1)
S), and determines the Entrance segment template according to the designated table number and the like (step S2). Entr
If the ance segment template is not specified (NO in step S1), the Entrance in the Note
It is determined whether or not the segment template is required, that is, whether or not it is necessary to control the rising section of the sound (step S3). Entrance segment
No template required, ie Entr in the Note
When it is not necessary to perform the waveform control by the ance segment template (NO in step S3), the Entrance segment template of table number 0 (that is, SA
(No T) is determined (step S4). That is, in this case, the characteristic change curve is "flat" Entrance
Since it is determined to be the segment template, control for the rising section of the sound is not performed. On the other hand, we need the Entrance segment template,
That is, when it is necessary to control the waveform by the Entrance segment template in the Note (YES in step S3), it is determined whether or not the automatic selection of the Entrance segment template is set (step S5). . If the automatic selection is not set to be executed (NO in step S5), the Entrance segment template of table number 0 (that is, no SAT) is determined (step S4). If it is set to execute automatic selection (step S5)
YES), it is determined whether or not the relationship with the previous sound (Note) is a joint (step S6). If the relationship with the previous sound is not a joint (NO in step S6),
The relationship with the preceding sound in the table is set to "Finish" (step S9). If the relationship with the previous sound is a joint (YES in step S6), it is further determined whether the relationship with the previous sound is slur (step S).
7). When the relationship with the previous sound is not slur (NO in step S7), the relationship with the previous sound in the table is “Jo
"int" (step S10). If the relationship with the previous sound is slur (YES in step S7), it is determined to be the Entrance segment template of table number 1 (that is, data for performing slur control) (step S8).

In step S11, it is determined whether the entrance segment template is automatically selected. If the entrance segment template is automatically selected (YES in step S11)
S), the hooking of the Entrance segment template is determined according to the playing strength (particularly the velocity of the sound in the case of MIDI data) (step S13). If the type of the Entrance segment template is not automatic selection (NO in step S11), the Entr selected by the user in advance
It is decided to hook the ance segment template (step S14). In step S12, it is determined whether the length of the Entrance segment template is automatic selection. If the length of the Entrance segment template is automatically selected (YES in step S12)
S), the length of the Entrance segment template is determined according to the length of the note (step S15). If the length of the Entrance segment template is not automatically selected (NO in step S12), the En selected by the user in advance
The length of the trance segment template is determined (step S16). The relationship with the preceding sound in the table thus determined (see steps S9 and S10), Entran
ce segment template type (step S13
And S14), and according to the length of the Entrance segment template (see steps S15 and S16), by referring to the table regarding the Entrance segment template as shown in FIG. Entran to be used for waveform control in the rising section
Determine the ce segment template.

The process of determining the Finish segment template will be described with reference to FIG. Figure 5 shows "Fi
is a flowchart showing an example of "nish segment template automatic determination processing".

At step S21, the user selects Finish.
It is determined whether the segment template is selectively designated by the table number. If the Finish segment template is specified (step S21)
YES), the finish segment template according to the designated table number is determined (step S22).
If the Finish segment template is not specified (NO in step S21), Fini in the Note
It is determined according to a predetermined parameter whether or not the sh segment template is required, that is, whether or not it is necessary to control the falling edge of the sound (step S2).
3). No Finish segment template required,
In other words, if it is not necessary to control the waveform with the Finish segment template in the Note (step S
No. 23), the finish segment template of table number 0 (that is, no SAT) is determined (step S24). In other words, in this case, since the characteristic change curve is determined to be the Finish segment template in which the characteristic curve is “flat”, control is not performed for the falling edge of the sound. On the other hand, if the Finish segment template is required, that is, if the waveform control by the Finish segment template is required in the Note (YES in step S23), the setting is made to execute the automatic selection of the Finish segment template. It is determined whether it has been done (step S25). If it is not set to execute the automatic selection (step S
No of No. 25), the finish segment template of table number 0 (that is, no SAT) is determined (step S24). When it is set to execute the automatic selection (YES in step S25), it is determined whether or not the relationship with the next sound (Note) is a joint (step S26). If the relationship with the next sound is not a joint (NO in step S26), the relationship with the rear sound in the table is set to "Entrance" (step S29). If the relationship with the next sound is a joint (step S
26), it is determined whether or not the relationship with the next sound is slur (step S27). If the relationship with the next sound is not slur (NO in step S27), the relationship with the rear sound in the table is set to "Joint" (step S30). If the relationship with the next sound is slur (YES in step S27), Finish of table number 1
A segment template (that is, data for performing slur control) is determined (step S28).

In step S31, the Finish segment ...
It is determined whether the template type is automatic selection. When the type of the Finish segment template is automatic selection (YES in step S31), the type of the Finish segment template is determined according to the velocity of the sound (step S33). When the type of the Finish segment template is not automatic selection (NO in step S31), the type of the Finish segment template previously selected by the user is determined (step S34). In step S32, it is determined whether the length of the Finish segment template is automatic selection.
When the length of the Finish segment template is automatically selected (YES in step S32), the length of the Finish segment template is determined according to the length of the sound (step S35). If the length of the Finish segment template is not automatically selected (N in step S32)
O), the length of the Finish segment template previously selected by the user is determined (step S36). The relationship with the preceding sound in the table thus determined (step S
29 and S30), Finish segment template type (see steps S33 and S34), Finish
According to the length of the segment template (see steps S35 and S36), as shown in FIG.
By referring to the table regarding the Finish segment template, the Finish segment template to be used for waveform control of the falling edge of the sound in each note is determined.

As described above, if the user does not specify each of the Entrance / Finish segment templates, the connection sequence relationship with the preceding and following adjacent notes (Note) is used as a reference, or the segment template It is automatically selected based on the intensity in the section to which the template is applied, or based on the tone length in the section to which the segment template is applied. Note that the automatic selection criteria for the segment template are not limited to these, and the pitch in the section to which the segment template is applied may be the reference. in this way,
Since the segment template to be used for waveform control is automatically determined when the user does not select the segment template, the user bothers to select the Entrance / Finish for each note over the entire song. You do not have to select the segment template. This allows the user to perform efficient waveform control.

When the segment templates to be used for waveform control are determined by user selection or automatic processing as described above, the determined segment templates are then arranged on the time axis and arranged. The segment template and the input envelope are combined. Then, by applying the envelope of the synthesized result to each module, waveform control is performed and an arbitrary musical tone waveform is generated. Therefore, the arrangement of the determined segment templates on the time axis and the synthesis of the input envelope performed based on the arranged segment templates will be described with reference to FIGS. 6 and 7.
However, in the following explanation, Entr for one note (Note)
Placement of each segment template of ance / Finish / Note and composition with the input envelope will be explained. That is, such processing is performed for each sound.

First, the arrangement of each segment template on the time axis will be described. FIG. 6 is a conceptual diagram for explaining the arrangement of each segment template on the time axis. Figure 6 (a) shows the Entrance segment.
It is a conceptual diagram for explaining the arrangement of the template and Finish segment template on the time axis,
FIG. 6B is a conceptual diagram for explaining the arrangement of the Note segment template on the time axis.

As shown in FIG. 6A, first, the Entrance segment template and the Finish segment template are arranged on the time axis. By matching the note-on timing preset in the Entrance segment template with the note-on event (NoteOn), the Entrance segment template is arranged on the time axis near the note-on. On the other hand, Finish segment
The Finish segment template is arranged on the time axis near the note-off by matching the note-off timing preset in the template with the note-off event (Note Off). At this time, Entrance / Fi
The shape of the characteristic change curve in each segment template can be expanded or contracted in the time axis direction by enlarging or reducing the time length of each segment template of nish using a predetermined time control parameter or the like. The scaling of the time length of each segment template of Entrance / Finish using this time control parameter scales around the note-on timing or note-off timing. For example, the first half of the note-on timing or note-off timing in each of the Entrance / Finish segment templates is pre-timed (Pre Tim
e) If the second half is Post Time, the pre-time and post-time are controlled separately, and the first half or the second half of each Entrance / Finish segment template is expanded separately. Can be reduced. In this way, by allowing the user to edit the time length appropriately, the placed segment templates can be customized in the time axis direction, and each segment template of Entrance / Finish can be applied to the section desired by the user. You will be able to.

When the arrangement of the segment templates of Entrance and Finish on the time axis is completed, then, as shown in FIG.
As shown in (b), the already placed Entrance and Fini
Place Note segment templates based on each segment template in sh. This Note segment
The placement of the template is such that the start position of the Note segment template is the start point of the Entrance segment template and the end position of the Note segment template is Fini.
It is done so that it matches the end point of the sh segment template. That is, for the Note segment template, the time length is scaled according to the time length from the start point of the already placed Entrance segment template to the end point of the Finish segment template. In this way, the segment templates of Entrance and Finish and the Note segment templates are arranged on the time axis so as to overlap each other.

Next, envelope synthesis performed by combining each segment template will be described.
FIG. 7 is a conceptual diagram for explaining envelope synthesis performed by combining segment templates. In this example, dynamics / Pitch / vibrato d
The dynamics (amplitude) of the envelope segment for the four types of musical sound elements of epth / vibrato speed will be described as an example. That is, the respective figures shown in FIG. 7 are, in order from the top, a dynamics value curve of an input envelope (basic envelope) generated based on MIDI input, an input dB value curve calculated from the dynamics value curve based on the MIDI input, Note segment template dynamics dB value curve, Entrance and Finish segment template
It shows the dB value curve of dynamics and the input dB value curve of the synthesized result, respectively.

The dynamics value curve (ie input envelope or basic envelope) based on the input of MIDI information is
According to the key number of the Note, score interpretation section (player)
Using the Dynamics scale data assigned to
Converted to dB value (this is called input dB value curve). Note / Entrance / Fi arranged on the time axis as described above for the input dB value curve calculated by this conversion
Add each segment template of nish. At that time, the shape of the characteristic change curve in each segment template can be changed in the level direction by enlarging or reducing each level of each segment template of Note / Entrance / Finish by a predetermined level control parameter. The level control parameter may be specified by the user directly inputting a numerical value at a predetermined position (for example, a representative point characterizing the shape of a predetermined characteristic change curve), or the display device 5 may be used. The representative point on the characteristic change curve displayed above may be designated by the user moving it up and down using a mouse or the like. By thus editing the level by the user, each segment template can be customized in the level direction. And note / for the input dB value curve
When each segment template of Entrance / Finish is added, each segment template is relatively synthesized to generate a characteristic dB value curve. The generated synthesized dB value curve is input to the rendition style synthesis section (articulator) as, for example, Amplitude Shift or Dynamics.

Of course, similar to the synthesis of the envelope in the above dynamics, other Pitch / vibrato depth /
Also in vibrato Speed, a composite envelope is generated by calculating Note / Entrance / Finish segment templates for a predetermined input envelope. Also, similar to the level control for the envelope in the above dynamics, other Pitch / vibra
It goes without saying that the level control for the envelope in the to depth / vibrato speed can be performed respectively. Like this, Note / Entrance / Fini
Each segment template of sh (or Phrase) is designed to generate one composite envelope by acting on each other relative to the input envelope.

The synthetic envelope generated by the musical score interpretation unit (player) 1B, that is, the synthetic envelope calculated by calculating each segment template such as Note / Entrance / Finish for the input envelope is the rendition style synthesis unit (artist). Curator) 1C is output as a rendition style parameter. Then, the rendition style synthesis unit 1C reads out the rendition style modules from the rendition style database based on predetermined rendition style designation information (rendition style ID and rendition style parameters), and synthesizes the read rendition style modules by the score interpretation unit 1B. A packet stream in which the envelope is distributed is generated (see FIG. 2). Here, the arrangement of each rendition style module on the time axis and the distribution of the synthetic envelope to each rendition style module when the rendition style synthesis unit 1C creates a packet stream will be described with reference to FIG. FIG. 8 is a conceptual diagram for explaining the arrangement of each rendition style module on the time axis and the distribution of the synthetic envelope to each rendition style module when creating a packet stream. Figure 8 (a) shows En
FIG. 9 is a conceptual diagram for explaining the arrangement of trance system modules and Finish system modules on the time axis.
(B) is a conceptual diagram for explaining the arrangement of the Body system module or the Joint system module on the time axis and the distribution of the composite envelope.

As shown in FIG. 8A, first, the rendition style synthesizing unit 1C reads the Entrance-type module and the Finish-type module from the rendition-style database based on predetermined rendition-style designating information (rendition-style ID and rendition-style parameters), and on the time axis. It is placed at a predetermined time position of. The Entrance module is placed at a predetermined time position on the time axis by matching the sound-on timing of the Entrance module to the start point of the Entrance segment template. The sound-on timing here is recorded in the Entrance type module as the time from the beginning of the rendition style module. On the other hand, the Finish module is placed at a predetermined time position on the time axis by matching the sound off timing of the Finish module to the end point of the Finish segment template. The sound-off timing is recorded as the time from the beginning of the rendition style module in the Finish system module. Next, as shown in FIG. 8B, an Entrance system module and a Finish module arranged at a predetermined time position on the time axis.
Body module is placed between the system modules. En
When the time interval between the trance system module and the Finish system module is less than or equal to the specified value, NSB (Normal Short Body) is placed as the Body module, and when the time interval is greater than or equal to the specified value. Places VB (Vibrato Body) as a Body module. However, if the Body module is not inserted between the Entrance module and the Finish module that have already been arranged, the Body module may be omitted without being arranged. Then, when the interval between the next note-off and the note-on is a time interval equal to or less than a predetermined specified value, the Body module and the next Body module are jointed and the Joint module is inserted. In this way, each rendition style module is sequentially arranged on the time axis. Needless to say, if the user has selected a module to be used directly, that module may be used.

As can be understood from FIG. 8B, the Note / Entrance / Finish is applied to the synthetic envelope generated by the musical score interpretation unit 1B, that is, the input envelope, in accordance with the position of each rendition style module arranged. A packet envelope is generated by acquiring the composite envelope calculated by calculating each segment template such as, and distributing the acquired composite envelope to each rendition style module. The packet stream thus generated is supplied to the waveform synthesizing unit 1D, and the waveform synthesizing unit 1D transforms the vector data (that is, the waveform represented by the rendition style module) extracted from the waveform database based on the packet stream according to the synthetic envelope. It is possible to generate arbitrary musical tone waveforms by waveform-connecting the waveforms represented by the individual rendition style modules thus modified. That is, each rendition style module is deformed in accordance with the synthetic envelope distributed to each rendition style module, and the deformed rendition style modules are waveform-connected to form a musical tone waveform having a shape according to each characteristic change curve given as a segment template. It can be generated.

The rendition style synthesis unit 1C may reflect not only the synthesis envelope generated by the musical score interpretation unit 1B as described above in the packet stream but also vector data. That is, the rendition style synthesis unit 1
In C, when generating a vector parameter, a vector parameter may be generated in which the synthetic envelope generated by the score interpretation unit 1B is distributed to the rendition style module arranged at a predetermined time position on the time axis. In such a case, the waveform synthesizing unit 1D sequentially extracts vector data from the waveform database according to the generated packet stream, deforms the vector data according to the vector parameter, and connects the deformed vector data by waveform connection. It is possible to generate sounds that reflect the dynamics envelope. Of course, Pitch / vibrato depth / vi
Even in brato Speed, it is possible to generate sounds in which each envelope is reflected in the same manner as the dynamics envelope.

As described above, in the musical sound generating apparatus shown in this embodiment, the musical sound in the unit of phrase or in the unit of sound is used by using the hierarchical templates having different time lengths such as Phrase / Note / Entrance / Finish segment templates. Since it is possible to partially change the expression, it becomes possible to easily generate a musical tone waveform having a complicated changing shape particularly in a part of an attack portion or a release portion within one sound. . Each of these segments
By synthesizing the envelope for dynamics represented by the template and further synthesizing it with the envelope for each note, the envelope for all dynamics is formed. In addition, the above Phrase / Note / Entrance / Finish
Each segment template of each is dynamics / Pitc
By configuring as a set with characteristic change curves (that is, envelope curves) related to h / vibrato depth / vibrato Speed, the user only needs to specify the rendition style for dynamics.
In addition to this, waveform control for Pitch / vibrato depth / vibrato Speed can be performed at the same time. It is also possible to synthesize the envelopes based only on the combination of the segment templates without generating the original input envelope (basic envelope) based on the performance information.

When the musical tone generating apparatus as described above is used in an electronic musical instrument, the electronic musical instrument is not limited to a keyboard musical instrument, but may be any type of musical instrument such as a string instrument, a wind instrument, or a percussion instrument. Also, in that case, the music data reproducing unit 1A, the musical score interpreting unit 1B, the rendition style synthesis unit 1C, and the waveform synthesis unit 1
Not only D etc. built in one electronic musical instrument body,
It is needless to say that the invention can be similarly applied to a configuration in which each component is separately configured and each component is connected using a communication means such as a MIDI interface or various networks. Further, it may be configured by a personal computer and application software. In this case, the processing program may be supplied from a storage medium such as a magnetic disk, an optical disk or a semiconductor memory, or may be supplied via a network. Further, it may be applied to an automatic performance device such as an automatic performance piano.

[0049]

According to the present invention, characteristic change curves of a plurality of segment templates having different lengths are arranged so as to overlap at least part of each other, and a synthetic envelope is formed by the combination thereof. If you want to perform fine envelope control in a part of the section,
A characteristic change curve corresponding to a fine envelope having a required length may be pasted. As described above, there is an excellent effect that it is possible to easily perform fine envelope control in a part of the performance section simply by appropriately combining segment templates having different time lengths. For example,
Templates composed of characteristic change curves to be given to a part of one phrase or one note within one sound are prepared for different time lengths, and a plurality of templates corresponding to the different time lengths are appropriately combined and used. By synthesizing envelopes and generating musical tone waveforms in accordance with the synthesized envelopes, the user can easily perform minute waveform control that achieves more delicate musical expression, simply by appropriately combining templates for different time lengths. Can be done.

[Brief description of drawings]

FIG. 1 is a block diagram showing a hardware configuration example of a musical sound generating apparatus according to the present invention.

FIG. 2 is a block diagram showing an embodiment in which the musical sound generation processing is configured in the form of a dedicated hardware device.

FIG. 3 is a conceptual diagram showing an example of a data structure of a segment template.

FIG. 4 is a flowchart showing an example of “Entrance segment template automatic determination processing”.

FIG. 5 is a flowchart showing an example of “Finish segment template automatic determination processing”.

FIG. 6 is a conceptual diagram for explaining the arrangement of each segment template on the time axis.

FIG. 7 is a conceptual diagram for explaining envelope synthesis performed by combining each segment template.

FIG. 8 is a conceptual diagram for explaining arrangement of each rendition style module on the time axis and distribution of a synthetic envelope to each rendition style module when creating a packet stream.

[Explanation of symbols]

1 ... CPU, 2 ... Read-only memory (ROM),
3. Random access memory (RAM), 4 ... Input device, 5 ... Display device, 6 ... Drive, 6A ... External storage medium, 7 ... Waveform capture unit, 8. .. Sound source section, 8A ... sound system, 9 ... hard disk, 10 ... communication interface, BL ... bus line, 1A ... song data playback section, 1B ... score interpretation section (player ) 1C ... Rendition style synthesis section (articulator), 1D ... Waveform synthesis section

   ─────────────────────────────────────────────────── ─── Continued front page    F term (reference) 5D378 BB19 FF22 FF27 GG21 HB12                       HB36 KK40 KK44 MM29 MM42                       MM62 XX22 XX25

Claims (9)

[Claims] 1. Performance information supply means for supplying performance information, segment template supply means for supplying segment templates including characteristic change curves, and segment template supply means for supplying a plurality of segment templates of different lengths, Corresponding to a given performance section based on performance information, a plurality of segment templates of different lengths are arranged so that at least part of their characteristic change curves overlap, and based on combining the arranged characteristic change curves. A musical tone generating apparatus comprising: envelope synthesizing means for forming a synthetic envelope for the given performance section; and musical tone generating means for generating a musical tone for the given performance section using the formed synthetic envelope. 2. A performance information supply means for supplying performance information, and a segment template including a characteristic change curve, wherein a note segment template corresponding to the length of almost one note and a part of one note. Segment template supply means capable of supplying a partial segment template corresponding to the length; one note segment template and one or a plurality of partial segment templates corresponding to a given musical performance section based on the performance information Is selected, the selected note segment template is arranged in the performance section of the one note, the selected partial segment template is arranged for a part of the performance section of the one note, and Based on the combination of the characteristic change curves, the synthetic en- Tone generation device comprising an envelope synthesis means for forming a rope, and a musical tone generating means for generating musical tones for using the formed synthetic envelope playing interval of said given one tone. 3. The envelope synthesizing means generates a basic envelope based on the performance information, and forms the synthetic envelope by combining the generated basic envelope with a plurality of arranged characteristic change curves. The musical sound generating device according to claim 1 or 2. 4. Each of the plurality of segment templates having different lengths has a set of characteristic change curves for a plurality of types of musical tone elements, and the envelope synthesizing means sets a plurality of the plurality of types of musical tone elements for each of the plurality of types of musical tone elements. 4. The musical tone generating apparatus according to claim 1, wherein the synthetic envelopes for each of the plurality of types of musical tone elements for the given performance section are formed based on a combination of characteristic change curves. . 5. The envelope synthesizing means arranges each of the characteristic change curves in the given performance section based on a note-on or note-off timing included in the performance information. The musical sound generating apparatus according to any one of 1 to 4. 6. At least one of the plurality of segment templates of different length to be combined includes position information, and the envelope synthesizing means renders the at least one segment template in response to the position information. The musical sound generating apparatus according to claim 1, wherein the musical sound generating apparatus is arranged in a section. 7. A step of supplying performance information, a segment template including a characteristic change curve, wherein a plurality of segment templates having different lengths are supplied, and a given performance section based on the performance information is provided. Correspondingly, a plurality of segment templates of different lengths are arranged such that at least some of their characteristic change curves overlap,
Forming a synthetic envelope for the given performance section based on combining the arranged characteristic change curves; and generating a musical tone for the given performance section using the formed synthetic envelope. A method for generating a musical sound that includes and. 8. A segment template including a step of supplying performance information and a characteristic change curve, the segment segment template corresponding to substantially the entire length of one note, and the segment corresponding to the length of a part of one note. Providing a segment template, and selecting one note segment template and one or a plurality of partial segment templates corresponding to a given performance segment of one note based on the performance information, and selecting the selected note segment The template is arranged in the performance section of the one note, the selected partial segment template is arranged for a part of the performance section of the one note, and the characteristic change curves of the arranged templates are combined to generate the template. Forming a synthetic envelope for a given musical interval of a note, and the formed synthetic Generating a musical tone for the given one note playing section using an envelope. 9. A computer program comprising a group of instructions for causing a computer to execute the musical tone generating method according to claim 7.