JPH08185174A - Voice generating device - Google Patents

Abstract

PURPOSE: To voice an expressive voice according to strength and brightness to be voiced in a voice generating device capable of voicing the voice. CONSTITUTION: The brightness of the voice is decided by a voice brightness switch, and the strength of the voice is decided by the strength of a touch of a key board 1. Then, the consonant waveform data corresponding to the brightness of the voice decided by the voice brightness switch and the vowel waveform data corresponding to the strength of the voice decided by the strength of the touch of the key board 1 are read out respectively, and these consonant and vowel waveform data are converted based on a key code and the velocity data, and voicing is instructed to a sound source 6 based on the result. Thus, the strength and the brightness such as a sound constituting language are expressed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice generating device capable of generating voice as a musical sound.

[0002]

2. Description of the Related Art Conventionally, there is a voice generating device provided in an electronic musical instrument or the like, which can produce a voice (for example, a chorus "Ah") as a musical sound.

[0003]

By the way, in the above-mentioned conventional voice generating device, the voice is only captured as a part of the timbre, and the voice to be pronounced is composed of the voices constituting the words. There was a problem in that it was not possible to give a pronunciation that would make a person hum the song (humming) because the character and the difference in the expression depending on the playing method were not taken into consideration, and thus it was not possible to perform with a higher expressiveness.

Therefore, according to the present invention, it is possible to pronounce a voice having a facial expression according to the strength and the brightness to be pronounced, paying attention to the strength and the brightness of each of the sounds constituting the word. It is intended to provide a sound generation device.

[0005]

In order to achieve the above object, a voice generating apparatus according to the invention of claim 1 has a storage means for storing a plurality of voice waveform data and a strength for designating a sound intensity. A designating means, a brightness designating means for designating the brightness of the sound, and a voice having a facial expression according to the brightness of the sound designated by the brightness designating means and the strength of the sound designated by the strength designating means. And sound generation control means for selecting sound waveform data corresponding to the above from the storage means and designating sound generation based on the selected sound waveform data. Further, as a preferred mode, for example, as in the invention described in claim 2, the sound generation control means may change the waveform data to be selected according to the operation state of the operator.

According to a third aspect of the present invention, there is provided a voice generating device, which has a storage means for storing a plurality of vowel sound waveform data and a consonant sound waveform data, and an operation strength for detecting an operation strength of an operator for instructing sounding. Detecting means, brightness designating means for designating the brightness of the generated sound, and consonant sound waveform data stored in the storage means in correspondence with the brightness of the sound designated by the brightness designating means. Select, select the vowel sound waveform data stored in the storage means corresponding to the strength of the operation on the operator detected by the operation strength detection means, based on the selected consonant and vowel sound waveform data, A sounding control means for instructing sounding in response to an operation of the operation element is provided. In a preferred embodiment, the manipulator may be a key as in the invention according to claim 4.

For example, as in the invention described in claim 5, the sound generation control means selects specific consonant sound waveform data from the storage means if the operation time is a first predetermined time when the operation on the operation element is released. However, if the time from the release of the operation to the start of the next operation is the second predetermined time or more, the sounding based on the selected consonant waveform data is instructed, and the second
If it is less than or equal to the predetermined time, a function of not instructing sound generation based on the selected consonant waveform data may be provided.

[0008]

In the present invention, the sound intensity is designated by the intensity designating means, and the brightness of the sound is designated by the brightness designating means. Then, the voice waveform data corresponding to the voice having the expression corresponding to the designated sound intensity and brightness is selected from the plurality of voice waveform data stored in the storage means, and the selected voice waveform data is selected. The pronunciation is specified based on. Therefore, strength and brightness can be expressed like each voice that constitutes a word, and a voice having an emotion according to a performance style (operation state of an operator) can be pronounced, so that a more expressive performance can be achieved. It will be possible.

[0009]

Embodiments of the present invention will be described below with reference to the drawings. (I) Example 1 A. First Embodiment Structure of Embodiment 1 FIG. 1 is a block diagram showing the structure of an electronic musical instrument equipped with a sound generation device according to a first embodiment of the present invention. In this figure, reference numeral 1 denotes a keyboard, which detects a key release operation and key release speed (strength) for each key to detect a key-on signal, a key number,
A key-off signal or performance information such as velocity indicating a touch and release is output. Reference numeral 2 denotes a console panel of this electronic musical instrument, which is composed of various operation switches and a liquid crystal display section for displaying the setting states of these switches. Among the operation switches arranged on the console panel 2,
There is a voice mode switch 20 and a voice brightness switch 21 shown in FIG. The voice mode switch 20 is for setting the voice mode, and the voice mode and the normal performance mode are repeatedly set each time the switch is pressed. An LED 22 is provided immediately above the voice mode switch 20 and is lit while the voice mode is selected. The voice brightness switch 21 sets the brightness of voice in three steps. When the knob 21a is moved to the right, it becomes bright, and when it is moved to the left, it becomes dark.

Returning to FIG. 1, reference numeral 3 is a CPU having a function of controlling each part of the musical instrument, and its operation will be described later. Reference numeral 4 denotes a ROM that stores various control programs loaded into the CPU 3 and various data used in these programs. The ROM 4 also stores a voice group table described later. 5 is CPU
A RAM used as a work area 3 for temporarily storing various register / flag data. The following flags and registers are set in the RAM 5, for example. VF: Voice mode flag, voice mode switch 2
It is inverted every time 0 is pressed. VF = 1: voice mode VF = 0: normal performance mode KC: key code register used to store key code data. VL: Velocity register, used for storing velocity data.

Reference numeral 6 is a sound source constituted by a well-known waveform memory reading method, and is provided with a normal sound source and an accompaniment sound source for generating a musical sound according to a normal melody performance. Reference numeral 7 is a waveform memory for storing waveform data of various tones by the ADPCM system, for example. The waveform memory 7 also stores consonant sound waveform data corresponding to a consonant sound and vowel sound waveform data corresponding to a vowel sound. Reference numeral 10 denotes a D / A conversion unit for performing D / A conversion and outputting the musical sound signal W output from the sound source 6 after performing filtering for imparting a sound effect. An amplifier 11 amplifies the analog musical tone signal supplied from the D / A conversion unit 10 and causes the speaker SP to generate a musical tone. When the CPU 3 searches the operating states of various switches on the console panel 2 and detects that the voice mode is set, the consonant waveform data is read from the waveform memory 7 based on the set value of the voice brightness switch 21, and further the keyboard 1 The vowel waveform data is read from the waveform memory 7 according to the strength of the touch (velocity) of. Then, the consonant and vowel sound waveform data read from the waveform memory 7 are frequency- and amplitude-converted based on the key code and velocity data, and the sound source 6 is instructed to sound based on the result.

The selection of the consonant sound waveform data and the mother sound waveform data is performed according to the voice group table shown in FIG. This voice group table is "DI, TI,
PI, DA, TA, PA, DU, TU, PU ”, and the sound waveform data corresponding to each of“ D, T, P ”is selected by the setting of the sound brightness switch 21, and the keyboard is selected. The vowel waveform data corresponding to each of "I, A, U" is selected according to the strength of the touch (velocity) when 1 is played. In addition, in this figure,
"Pi, Pa, Pu" is a bright sound, and "Di, Da, Do"
Is a dark sound.

This voice group table is created as follows. That is, the voice has lightness and darkness. For example, if a dakuten is added to "ha", it becomes "ba" and becomes dark, and if a dakuten is added, it becomes "pa" and becomes bright.
On the other hand, in the case of “ta”, adding a dakuten gives a “da”, but since a semi-voiced mark cannot be given, this is “pa”. However, for example, when the sound brightness switch 21 is set to the center and the strength of the touch when the keyboard 1 is played is at a medium level, the consonant sound waveform data corresponding to "T" is selected and "A" is selected. The vowel waveform data corresponding to "" is selected. As a result, the sound of "ta" is pronounced. In addition,
FIG. 3 is a waveform diagram showing a voice signal obtained as described above. The consonant part is “D,
In "T, P", the vowel part corresponds to "I, A, U".
The latter half of the vowel part is attenuated like a normal keyboard instrument.

B. Operation of First Embodiment Next, the operation of the first embodiment having the above configuration will be described with reference to FIGS. Here, the operation of the main routine will be described as a general operation, and then the operations of various processing routines called in the main routine will be sequentially described. (A) Operation of Main Routine First, when the electronic musical instrument according to the present embodiment is powered on,
The CPU 3 loads a predetermined control program from the ROM 4 and executes the main routine shown in FIG.
The process proceeds to 2. In step SA2, various registers and flags set in the RAM 5 are reset and the sound source 6 is instructed to clear the tone generation channel register to zero. The tone generation channel register is a register provided inside the tone generator 6 and temporarily stores the tone parameter assigned to each tone generation channel under the instruction of the CPU 3.

When the initialization for initializing the various registers is completed, the CPU 3 advances the processing to the next step SA4, performs the switch processing according to the various switch operations of the console panel 2, and then advances to step SA6, whereupon the keyboard Perform processing. Next, when proceeding to step SA8, other processing such as updating display of the liquid crystal display unit is performed. After that, the CPU 3 returns the process to step SA4,
Thereafter, steps SA4 to SA8 are repeated to instruct the sound source 6 to generate a musical sound according to the contents of the switch operation and the keyboard operation.

(B) Operation of Switch Processing Routine First, the CPU 3 determines in step SB2 whether or not the voice mode switch 20 has been pressed. In this decision,
If it is determined that the voice mode switch 20 is not pressed, the process proceeds to step SB6. On the other hand, if it is determined that the voice mode switch 20 has been pressed, the process proceeds to step SB4 and the value of the voice mode flag VF is inverted. After performing this process, the process proceeds to step SB6. Step SB6
When the process proceeds to step S1, the set position of the voice brightness switch 21 is detected. Then, in step SB8, the consonant waveform data corresponding to the detection result of the set position of the voice brightness switch 21 is selected. For example, if the sound brightness switch 21 is set to the middle level, the consonant waveform data corresponding to "T" is selected.
After selecting the consonant sound waveform data, other switch processing is performed in step SB10, and then the processing ends.

(C) Keyboard processing routine After the above switch processing is completed, the keyboard processing is started.
First, in step SC2, it is determined whether or not the state of the keyboard 1 has changed. In this determination, if it is determined that there is no change, the process directly ends. On the other hand, if it is determined that the key has been released, the process proceeds to step SC4 to instruct the mute of the sound corresponding to the key code (pitch).
Exit the process. Note that the sound is naturally attenuated when the key is released, so it is not always necessary to forcibly silence the sound. On the other hand, if it is determined that the key is released after the key is released, the process proceeds to step SC6, the key code of the depressed key is stored in the key code register KC, and the velocity data indicating the touch strength at that time is stored in the velocity register VL. To store.

After the key code and the velocity data are stored in the respective registers, the process proceeds to step SC8 and it is determined whether or not the value of the voice mode flag VF is "1", that is, whether or not the voice mode is set. Do. If it is determined in this determination that the mode is not the voice mode, that is, the normal performance mode, the process proceeds to step SC10 to perform a sounding process based on the contents of the key code register KC and the velocity register VL, that is, a sounding process by simply playing a key on the keyboard 1. To do. After performing this tone generation processing, the processing is exited. On the other hand, if it is determined that the voice mode is set in the determination in step SC8, the process proceeds to step SC12,
The vowel waveform data is selected based on the contents of the velocity register VL. For example, if the touch is strong, the vowel waveform data corresponding to "I" is selected. After selecting the vowel sound waveform data, the process proceeds to step SC14, and the already selected child vowel sound wave data and vowel sound wave data are stored in the key code register KC.
And the velocity register VL are converted to give a sounding instruction. After performing this process, the process is exited.

As described above, according to the first embodiment, the consonant sound waveform data is selected by the sound brightness switch 21,
The vowel sound waveform data is selected according to the strength of touch by playing a key on the keyboard 1, and the selected sub-vowel sound waveform data and the vowel sound waveform data are converted in frequency and amplitude based on the key code and velocity data to give a sounding instruction. . Therefore, strength and brightness can be expressed like each voice constituting a word.

(II) Example 2 A. Configuration of Second Embodiment Since the configuration of the second embodiment is similar to the configuration of the first embodiment described above, it will be replaced with FIG. However, the following flags, registers and counters are set in the RAM 5, for example. VF: Voice mode flag, voice mode switch 2
It is inverted every time 0 is pressed. VF = 1: voice mode VF = 0: normal performance mode

KC: A key code register used to store key code data. VL: Velocity register, used for storing velocity data. T: Counter for counting time from last key release to current key release t: Counter for counting time from last key release to current key release In addition, A is data corresponding to a quarter note length. , B are data corresponding to the slur discrimination time.

In the first embodiment, the voice brightness switch 2
The voice to be generated is specified according to the strength of the touch (velocity) of the keys of 1 and the keyboard 1.
Then, the sound to be pronounced is changed according to the playing method. Here, FIG. 8 is a diagram showing a voice group table. As shown in FIG. 8, “DI, TI, PI,
RI, DA, TA, PA, RA, DU, TU, PU, R
It consists of 12 voice information of "U". This voice group table is stored in the ROM 4. Based on this voice group table, the consonant sound waveform data corresponding to each of “D, T, P, R” is selected by the setting of the voice brightness switch 21, and the strength of the touch when the keyboard 1 is played is selected. The vowel waveform data corresponding to each of "I, A, U" is selected. Further, one of the consonant sound waveform data is selected depending on the rendition style. That is, either "D" or "R", "T", "R", or "P" or "R" is selected.

Here, for example, the voice brightness switch 21
When the setting of is medium level and the consonant waveform data of “T and R” is selected, if the touch strength when the key is first pressed is at the medium level, the mother waveform data of “A” is selected. It At this time, it becomes "TA" and "TA" is pronounced. Then, after the key is released, when the key is again pressed for more than the slur discrimination time b, "TA" is generated and "TA" is pronounced.
In this case, if the time from the previous key press to the subsequent key release is a quarter note length or less, for example, if it is an eighth note or a sixteenth note, it will be “ta” as it is, but from the previous key press to the subsequent key release. If the time is longer than a quarter note, the pronunciation of "n" is added after "ta". Further, in the case of continuous keystrokes such as chords without key release, for example, "TA, TA, TA" is sounded, and "tatata" is pronounced substantially at the same time. On the other hand, if the time from one key release to the next key depression is within the slur discrimination time b, the "T" consonant waveform data is not selected, but the "R" consonant waveform data is selected. The result is "RA" and "LA" is pronounced.

As shown in FIG. 9, the time from the previous key press to the key release is an eighth note length, the time to the next key press exceeds the slur discrimination time b, and the time from the next key press to the key release. If the time is 8th note length, it is pronounced as "Tata". Further, the time from the previous key depression to the key release is a quarter note length, the time to the next key depression exceeds the slur discrimination time b, and the time from the next key depression to the key release is a quarter note length. When pronounced as "tantan". Also, the time from the previous key press to the key release is a half note length, the time to the next key press exceeds the slur discrimination time b, and the time from the next key press to the key release is a quarter note length. In case of
Is pronounced.

Further, the time from the previous key depression to the key release is a quarter note length, the time to the next key depression exceeds the slur discrimination time b, and the time from the next key depression to the key release is 2 minutes. In the case of note length, it is pronounced as "tanta-n". Also, the time from key depression to key release is an eighth note length, and in the case of substantially simultaneous continuous keying without key release, "tata" is pronounced at substantially the same time. Also,
If the time from the previous key press to the key release is the eighth note length, the time to the next key press is within the slur determination time b, and the time from the next key release to the key release is the eighth note length. Pronounced "cod". Also, the time from the previous key depression to the key release is a quarter note length, the time to the next key depression is within the slur discrimination time b, and the time from the next key depression to the key release is a quarter note length. The case is pronounced "Taran".

Further, the time from the previous key depression to the key release is a quarter note length, the time to the next key depression is within the slur discrimination time b, and the time from the next key depression to the key release is 8 minutes. In case of note length, it is pronounced "ta-la". Also, the time from the previous key press to the key release is an eighth note length, the time to the next key press is within the slur discrimination time b, and the time from the next key press to the key release is a quarter note length. The case is pronounced "Taran". Also, the time from the previous key depression to the key release is a half note length, the time to the next key depression is within the slur discrimination time b, and the time from the next key depression to the key release is a quarter note length. The case is pronounced "ta-ran".
Also, the time from the previous key depression to the key release is a quarter note length, the time to the next key depression is within the slur determination time b, and the time from the next key depression to the key release is a half note length. In the case, "Tara"
Is pronounced.

B. Operation of Second Embodiment Next, the operation of the second embodiment will be described with reference to FIGS. 11 to 13. Here, since the general operation of the main routine is similar to that of the first embodiment, it will be replaced with the description of the first embodiment, and the operations of the processing routines different from the first embodiment, that is, the switch processing routine, the keyboard processing routine, and the interrupt processing routine. Will be sequentially described.

(A) Operation of Switch Processing Routine First, the CPU 3 determines in step SD2 whether or not the voice mode switch 20 has been pressed. In this decision,
If it is determined that the voice mode switch 20 is not pressed, the process proceeds to step SD6. On the other hand, if it is determined that the voice mode switch 20 is pressed, the process proceeds to step SD4 and the value of the voice mode flag VF is inverted. After performing this processing, the process proceeds to step SD6. Step SD6
When the process proceeds to step S1, the set position of the voice brightness switch 21 is detected. Next, at step SD8, a set of consonant sound waveform data corresponding to the detection result of the set position of the sound brightness switch 21 is selected. For example, when the sound brightness switch 21 is set to the middle level, the consonant sound waveform data corresponding to "T and R" are selected. After selecting the consonant sound waveform data, other switch processing is performed in step SD10, and then the processing ends.

(B) Keyboard processing routine After the above switch processing is completed, the keyboard processing is started.
First, in step SE2, it is determined whether or not the state of the keyboard 1 has changed. In this determination, if it is determined that there is no change, the process directly ends. On the other hand, if it is determined that the key has been released, the process proceeds to step SE4, where it is determined whether the value of the voice mode flag VF is "1".
That is, it is determined whether or not the voice mode is set. If it is determined in this determination that the mode is not the voice mode, that is, the normal performance mode, the process proceeds to step SE6 to instruct to mute the sound corresponding to the key code. That is, the sound that has been produced so far is muted. After doing this process,
Exit the process. Note that the sound is naturally attenuated when the key is released, so it is not always necessary to forcibly silence the sound.

If it is determined in the above-mentioned step SE4 that the voice mode is set, the process proceeds to step SE8 and whether or not the time T from the last key depression to the current key release is longer than the quarter note length. Is determined. In this determination, if the time T from the last key depression to the current key release is not longer than the quarter note length, for example, if it is determined that it is the eighth note or 16th note length, the process proceeds to step SE6. On the other hand, when it is determined that the time T from the last key depression to the current key release is longer than the quarter note length, the flow proceeds to step SE10, and the waveform data of "N (N)" is selected. After performing this process, the process proceeds to step SE6.

When it is determined in step SE2 that a new key has been pressed, the process proceeds to step SE12, the key code of the pressed key is stored in the key code register KC, and the velocity data is stored in the velocity register VL.
To be stored. After performing this processing, step SE14
Then, it is determined whether or not the voice mode is set. In this determination, if it is determined that the mode is not the voice mode, that is, the normal performance mode, the process proceeds to step SE16, and a tone generation process based on the contents of the key code register KC and the velocity register VL, that is, a tone generated by simply playing a key of the keyboard 1. Perform processing. After performing this tone generation processing, the processing is exited.

If it is determined in the step SE14 that the mode is the voice mode, the process proceeds to step SE18 to select the vowel waveform data based on the contents of the velocity register VL. For example, if the strength of key depression is medium level, "A"
The vowel sound waveform data corresponding to is selected. After performing this process, the process proceeds to step SE20, and it is determined whether or not the time t from the last key release to the current key depression is longer than the slur determination time b. In this determination, when it is determined that the time t from key release to key depression is longer than the slur determination time b, the process proceeds to step SE22 and step SE1.
The vowel sound waveform data selected in step 8 and one of the consonant sound waveform data selected in the switch processing routine are converted based on the contents of the key code register KC and the velocity register VL to give a sounding instruction. As a result, for example, "T
A "is pronounced. After performing this process, the process is exited.

If it is determined in step SE20 that the time t from the key release to the key depression is shorter than the slur determination time b, the process proceeds to step SE24 to cancel the selection of the waveform data "n". That is, if it is a slur, do not pronounce "n". After performing this processing, the flow proceeds to step SE26, and the mother sound waveform data selected in step SE18 and the other one of the child sound waveform data selected in the switch processing routine are converted based on the contents of the key code register KC and the velocity register VL. And give pronunciation instructions. As a result, for example, "RA" is pronounced. After performing this process, the process is exited.

(C) Interrupt processing routine This interrupt processing routine is executed a plurality of times while the main routine is being executed once. First, in step SF2, it is determined whether or not the voice mode is set. If it is determined in this determination that the mode is not the voice mode, the process directly ends. On the other hand, if it is determined that the mode is the voice mode, the process proceeds to step SF4 to determine the key state. In this determination, if it is determined that the state of the key is the initial state, that is, no key has been pressed, the process directly ends.
On the other hand, if it is determined that the key is new, the process proceeds to step SF6, and the count value of the time T from key depression to key release is set to "0". After that, the processing is exited.

If it is determined in step SF4 that the key is a new key-off, the process proceeds to step SF8, and the count value of the time t from key release to key depression is set to "0". After that, the processing is exited. If it is determined in step SF4 that the key is on, the process proceeds to step SF10, the count value of the time T from key depression to key release is incremented by "1", and the process exits.

If it is determined in step SF4 that the key is off, the process proceeds to step SF12 and the count value of the time t from key release to key depression is increased by "1". Next, in step SF14, it is determined whether the count value of the time t from key release to key depression is equal to the slur determination time b. In this determination, if it is determined that the count value of the time t from the key release to the key depression is not equal to the value of the slur determination time b, the process directly ends. On the other hand, if it is determined that the count value of the time t from the key release to the key depression is equal to the value of the slur determination time b, the process proceeds to step SF16 to determine whether or not the waveform data of "n" is selected. To do. In this judgment, "n"
If it is determined that the waveform data of is not selected, the process is exited. On the other hand, when it is determined that the waveform data of "n" is selected, the process proceeds to step SF18,
The pronunciation instruction is given based on the waveform data of "n".

The operation of this embodiment will now be described with further reference to FIG. In this case, it is assumed that the voice brightness switch 21 is set to the middle level. (B) When each key on the keyboard 1 is pressed one by one First, when a key is pressed with a weak force, "to" is pronounced. When the key is released within the time of the quarter note length or less, "n" is not selected and the operation waits until the next key is pressed. There is no new key depression from the time the key is released until the slur discrimination time b is reached, and when the key is depressed with a strong force thereafter, a "Tee" is sounded. When this key-depressed state is released for more than the quarter note length, "n" is selected.

If there is no new key depression from the time the key is released until the slur determination time b is reached, "n" is sounded when the slur determination time b is reached. Next, after the slur determination time b has been exceeded, when the key is pressed with a medium force, "ta"
Is pronounced. When this key-depressed state is released for more than the quarter note length, "n" is selected. If there is no new key depression from the time the key is released until the slur determination time b is reached, "n" is sounded when the slur determination time b is reached. Next, after the slur determination time b has been exceeded, when a key is pressed with a medium force, "ta" is sounded. When this key-depressed state is released for more than the quarter note length, "n" is selected. If there is no new key depression from the time the key is released until the slur determination time b is reached, "n" is sounded when the slur determination time b is reached.

(B) When each key is continuously pressed on the keyboard 1. First, when a key is pressed with a medium force, "ta" is pronounced. Then, before the determination of the key release is made, when the key is continuously depressed twice with a medium force, "ta, ta" is sounded. In this case, since the time from the last key depression to key release is approximately eighth note length, "n" is not selected. Then, after the slur determination time b has been exceeded, if the keys are pressed three times at the same time with a medium force, "ta, ta, ta" is sounded at substantially the same time. In this case, since the time from the key depression to the last key release exceeds the quarter note length, "n" is selected. Therefore, when the slur determination time b is reached after the key is released, "n" is pronounced.

After the "n" is pronounced, when the key is pressed with a strong force, the "ti" is pronounced. After the "ti" is pronounced, when the key is subsequently pressed with a medium force, the "ta" is pronounced. Then, after the "ta" is pronounced, if a key is subsequently pressed with a weak touch, "to" is pronounced. After that, when the key that pronounced "Ti" is released, it exceeds the quarter note length since the last key was pressed.
When the slur determination time b has elapsed, "n" is pronounced. Then, when the keys released with "ta" and "to" are released at the same time, the time from the last key release to the key release exceeds the quarter note length. When the slur determination time b is reached, "n" is pronounced.

(C) When each key on the keyboard 1 is pressed smoothly First, when the key is pressed with a medium force, "ta" is pronounced. When the key is released for more than the quarter note length, "n" is selected. Then, when the key is newly pressed with the medium force within the slurring determination time b after the key is released, "LA" is pronounced,
"N" is destroyed. When this key-depressed state is released for more than the quarter note length, "n" is selected. When a key is pressed with a weak force before the key that has pronounced "la" is released, "to" is pronounced. After that, when the key is released for more than the quarter note length and at the same time, the key is newly pressed with a weak force, "le" is pronounced and "n" is discarded. In this case, if the new key depression is a little earlier than the previous key depression, "to" is sounded. If the key is pressed with a weak force and then released in a time shorter than the quarter note length, "n" is not selected. afterwards,
When the key is newly pressed with a strong force, a "ti" is pronounced, and when the key is released for more than a quarter note length, "n" is selected and the slur judgment time b is reached from the key release. "N" is pronounced when you do.

As described above, according to the second embodiment, since the processing is performed as described below, it is possible to change the sounding voice depending on the difference in playing style such as note length, slur, staccato, and chord. Become. (B) If the time from the last key depression to the current key release is a quarter note or more, other than the slur playing style or chord, "n" is sounded when releasing the key. (B) The time from key release to the next key depression is the slur discrimination time b
If it is within the range, then "LA, LI or LE (depending on the strength of key depression)" is pronounced. (C) The time from key release to the next key depression is the slur discrimination time b
If it is longer, the pronunciation of "n" is performed normally.

The contents of the voice group table in each of the above embodiments are merely examples, and the present invention is not limited to this. Also, the number of audio information is not limited to this. Further, instead of one voice group table, a plurality of voice group tables having different contents may be provided so that a desired one can be freely selected.
Further, in the second embodiment, as a condition for sounding "n" at the time of releasing the key, the time from the previous key depression to the current key release is equal to or more than a quarter note, and further as a condition for causing "La, Li or Le" to be pronounced. Although the time from the key release to the next key depression is within the slur determination time b, these conditions are not limited and may be arbitrarily determined. Preferably, these conditions may be set arbitrarily. Further, in each of the above-described embodiments, the example is applied to the electronic musical instrument, but other than this, the invention may be applied to computer music by music software.

[0044]

According to the present invention, it is possible to generate a voice having a facial expression according to the strength and brightness of the sound, and further, it is pronounced according to the difference in playing style such as note length, slur, staccato or chord. Since it is also possible to change the voice, it is possible to express strength and brightness like each voice that composes a word, and a voice with emotion according to the performance style (operation state of the operator) is pronounced. it can. As a result,
A more expressive performance is possible.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of an electronic musical instrument including a voice generation device according to a first embodiment of the present invention.

FIG. 2 is a diagram illustrating a voice information group table according to the first embodiment.

FIG. 3 is a waveform diagram showing an audio signal according to the first embodiment.

FIG. 4 is a diagram showing appearances of a voice mode switch and a voice brightness switch in the first embodiment.

FIG. 5 is a flowchart for explaining the operation of the main routine in the first embodiment.

FIG. 6 is a flowchart for explaining the operation of a switch processing routine in the first embodiment.

FIG. 7 is a flowchart illustrating an operation of a keyboard processing routine according to the first embodiment.

FIG. 8 is a diagram illustrating a voice information group table in an electronic musical instrument including a voice generating device according to a second embodiment of the present invention.

FIG. 9 is a diagram illustrating an operation in the second embodiment.

FIG. 10 is a diagram illustrating an operation in the second embodiment.

FIG. 11 is a flowchart illustrating an operation of a switch processing routine according to the second embodiment.

FIG. 12 is a flowchart illustrating an operation of a keyboard processing routine according to the second embodiment.

FIG. 13 is a flowchart for explaining the operation of the interrupt processing routine in the second embodiment.

[Explanation of symbols]

 1 keyboard 2 console panel 3 CPU 4 ROM 5 RAM 6 sound source 7 waveform memory 10 D / A converter 11 amplifier

Claims (5)

[Claims] 1. A storage unit for storing a plurality of voice waveform data, a strength designating unit for designating a sound intensity, a brightness designating unit for designating a brightness of a sound, and the brightness designating unit. Voice waveform data corresponding to a voice having a facial expression according to the brightness of the sound designated by and the strength designated by the strength designating means is selected from the storage means, and based on the selected voice waveform data A sound generation device comprising: a sound generation control unit that specifies a sound generation. 2. The sound generation device according to claim 1, wherein the sound generation control means changes the selected waveform data according to the operation state of the operation element. 3. A storage means for storing a plurality of vowel sound waveform data and a consonant sound waveform data, an operation strength detecting means for detecting an operation strength with respect to an operator instructing sounding, and a brightness of a generated sound. Brightness specifying means for specifying the brightness, and the consonant waveform data stored in the storage means corresponding to the brightness of the sound specified by the brightness specifying means, and the operation strength detecting means selects the consonant waveform data. The vowel sound waveform data stored in the storage means is selected according to the detected strength of the operation on the manipulator, and the sound is generated in response to the operation of the manipulator based on the selected consonant and vowel sound waveform data. A sound generation device, comprising: 4. The voice generating device according to claim 3, wherein the operator is a key. 5. The sound generation control means selects specific consonant sound waveform data from the storage means if the operation time is a first predetermined time at the time of releasing the operation on the operator, and the If the time until the start of the operation is equal to or longer than the second predetermined time, the sounding based on the selected consonant sound waveform data is instructed. 5. The voice generating device according to claim 3, wherein the voice generating device is not performed.