JP6492629B2 - SINGLE GENERATION DEVICE, ELECTRONIC INSTRUMENT, METHOD, AND PROGRAM - Google Patents

Description

  The present invention relates to a singing generation apparatus, an electronic musical instrument, a method, and a program for generating a singing by designating lyrics in a performance operation.

  A person stored in advance based on the pitches determined in advance by operating the performance operator, etc. after inputting the lyrics of the music from the performance operator such as the keyboard, operation panel, keyboard keys, etc. There is known an electronic musical instrument in which waveform data corresponding to lyrics inputted in advance is read out from waveform data of a singing voice and a song is generated (for example, a technique described in JP-A-6-324677). Such an electronic musical instrument makes it possible to utter a human singing voice.

  There is also known an electronic musical instrument that generates a song by inputting pronunciation content (lyrics) with one hand and a pronunciation pitch with the other hand.

JP-A-6-324677

  However, the conventional technique for inputting lyrics and inputting pitches by switching modes has a problem that it is impossible to input lyrics in real time.

  Also, with the conventional technology that uses both hands to input lyrics and pitches, real-time singing input is possible, but both hands are blocked only by singing operations, so a performance operator that is operated by hand, such as a fader, is provided. There was a problem that it was difficult to operate during singing.

  Therefore, an object of the present invention is to enable real-time singing input with one hand.

In an example of the aspect, a specifying unit that specifies an operation state of a plurality of performance operators to which different pitches are assigned , and a state transition that shifts to one of the first and second states based on the operation state And a lyrics determination unit that determines lyrics based on the operation state when the state is shifted to the first state , and a pitch is determined based on the operation state when the state is shifted to the second state. And a sound generation unit for generating sound generation data for generating the determined lyrics at the determined pitch .

  According to the present invention, real-time singing input can be performed with one hand.

It is a figure which shows the hardware structural example of 1st, 2nd, 3rd embodiment of a song production | generation apparatus. It is a flowchart which shows the example of the main process in each embodiment. It is a flowchart which shows the detailed example of the keyboard process in 1st, 2nd embodiment. It is a flowchart which shows the detailed example of the lyric determination process in 1st, 3rd embodiment. It is a flowchart which shows the detailed example of the song production | generation process in 1st, 3rd embodiment. It is a flowchart which shows the detailed example of the lyric determination process in 2nd Embodiment. It is a flowchart which shows the detailed example of the song production | generation process in 2nd Embodiment. It is a flowchart which shows the detailed example of the keyboard process in 3rd Embodiment. It is a figure which shows the hardware structural example of 4th Embodiment of a song production | generation apparatus. It is a flowchart which shows the detailed example of the keyboard process in 4th Embodiment.

  Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings. First, a first embodiment of the present invention will be described. In the song generation process in the first embodiment, the performer performs a plurality of (two or more) keys on the keyboard as performance operators within a short time that can be regarded as less than N milliseconds ("simultaneous"), for example, 4 Press the key in less than a millisecond). By this key combination, lyrics (here, a unit of pronunciation content, for example, a single character such as “A” or “KA” or a phonetic symbol) are designated and stored. This operation is called lyrics determination processing. Thereafter, the performer presses a single key after N milliseconds or more have elapsed since the key was pressed. If no other key is pressed after N milliseconds have elapsed since the key was pressed, the pronunciation of the lyrics is instructed with the pitch assigned to the key. This operation is called song generation processing. If there is another key press within less than N milliseconds after pressing a single key, this will be handled as a simultaneous key press of multiple keys, so the lyrics determination process will be controlled and the stored lyrics Will be overwritten. The above key control process of the keyboard is called a keyboard process.

FIG. 1 is a diagram illustrating a hardware configuration example of a first embodiment (the same applies to second and third embodiments described later) of the song generation device. The singing voice synthesizing apparatus 100 shown in FIG. 1 has a configuration in which a ROM (Read Only Memory) 102, a RAM (Random Access Memory) 103, and a keyboard 104 are connected to a CPU (Central Processing Unit) 101. Prepare. In addition, a singing synthesis unit 105 is connected to the CPU 101, and a sound generation unit 106 is connected to the singing synthesis unit 105. The configuration shown in the figure is an example of a computer that can realize the singing voice synthesizing apparatus 100, and such a computer is not limited to this configuration.

  The ROM 302 is a memory that stores programs including a song synthesis program for controlling the computer. A RAM 303 is a memory that temporarily stores a program or data stored in the ROM 302 when each program is executed. The CPU 101 controls the entire computer by reading each program from the ROM 302 to the RAM 303 and executing it.

  The keyboard 104 is a performance input device that includes a plurality of keys and enables a performer to perform a piano-type keyboard performance.

  The singing synthesis unit 105 selects waveform data corresponding to the lyrics instructed from the CPU 101 from the internal waveform memory, and reads out the waveform data of the lyrics at a reading speed (sampling interval) corresponding to the pitch instructed from the CPU 101. Synthesize and output singing data.

  The sound generation unit 106 converts the digital song data output from the song synthesis unit 105 into an analog sound signal, and amplifies it with an internal amplifier (not shown) and emits the sound from the speaker.

  FIG. 2 is a flowchart showing an example of the main process of the song synthesis process in the first embodiment (the same applies to the second, third, and fourth embodiments). This process is an operation in which the CPU 101 executes a song synthesis processing program stored in the ROM 102.

  First, the CPU 101 initializes the contents of the RAM 103 (step S201), and then performs the keyboard process in step S203 until the performer operates a power switch (not shown) to determine that the performance is finished in step S202. Run repeatedly.

  FIG. 3 is a flowchart showing a detailed example of keyboard processing in the first embodiment (the same applies to the second embodiment).

  First, the CPU 101 scans the keyboard 104 and substitutes a set of keys that are pressed into a key set variable c_keys on the RAM 103 (step S301).

  Next, the CPU 101 calculates a difference set between the current key set variable c_keys and the previous key set variable p_keys of the RAM 103 in which the contents of the key set variable at the previous processing are held, and the result is set to the set variable k on the RAM 103. Substitute (step S302). That is, the newly pressed key from the previous keyboard processing to the current keyboard processing is substituted into the set variable k.

  Next, the CPU 101 determines whether the value of the state variable state indicating the state on the RAM 103 is IDLE (preparation complete state) or ACTIVE (key pressing operation in progress state) (step S303).

  Since the initial value of the state variable state is in the IDLE state in step S201 in FIG. 2, the CPU 101 first advances the process to step S304. Here, the CPU 101 determines the original number of the set variable k.

  When the original number of the set variable k is 0, nothing has changed from the previous time, so the CPU 101 skips the next step S305 and calculates the current key set variable calculated in step S301 for the next keyboard process. After the contents of c_keys are stored in the previous key set variable p_keys (step S313), the current keyboard process in step S203 of FIG. 2 shown in the flowchart of FIG.

  If the original number of the set variable k is 1 or more, the CPU 101 stores the current time in a key press start time variable t_start on the RAM 103 that holds the time at the time of key press start. Further, the CPU 101 stores a state value ACTIVE indicating that the key depression has been started in the state variable state. Further, the CPU 101 stores a set of keys newly pressed at this time in step S302 in a key pressed set variable on_keys that holds information on keys that have been pressed up to the present after the start of key pressing. The contents of variable k are stored (step S205). Thereafter, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S301 for the next keyboard process in the previous key set variable p_keys (step S313), and then is shown in the flowchart of FIG. This keyboard processing in step S203 is terminated.

  The keyboard process in step S203 of FIG. 2 shown in the flowchart of FIG. 3 is repeatedly executed, for example, every 1 millisecond. As a result, in the next keyboard process, the value determination of the state variable state is ACTIVE in step S205, and thus the CPU 101 advances the process from step S303 to step S306. In step S306, the contents of the newly pressed key set variable k are added to the key pressed set variable on_keys.

  Subsequently, the CPU 101 subtracts the value of the key press start time variable t_start from the current time, and determines whether the result is equal to or longer than the above-described N milliseconds or less than N milliseconds (step S307).

  If the determination in step S307 indicates less than N milliseconds, it is regarded as simultaneous key pressing. In this case, after the contents of the current key set variable c_keys calculated in step S301 for the next keyboard process are stored in the previous key set variable p_keys (step S313), they are directly shown in the flowchart of FIG. The current keyboard process in step S203 is terminated. If the time from the key press start time indicated by the key press start time variable t_start to the current time is less than N milliseconds, this process is repeatedly executed, and the key is pressed by the key press completed set variable on_keys until the present after the key press is started. Key information is accumulated.

  Eventually, when time elapses from the key pressing start time and the determination in step S307 indicates N milliseconds or more, the CPU 101 advances the process from step S307 to S308. In step S308, the CPU 101 determines whether the original number of the pressed key set variable on_keys is 1 or 2 or more.

  If the original number of the pressed key set variable on_keys is 2 or more, lyrics determination processing for determining lyrics by a plurality of keys pressed simultaneously is executed (step S309). After this processing, the CPU 101 returns the contents of the state variable state to the IDLE state, and sets the contents of the keyed set variable on_keys to an empty set φ (step S312). Then, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S301 for the next keyboard processing in the previous key set variable p_keys (step S313), and then is shown in the flowchart of FIG. This keyboard processing in step S203 is terminated.

  As described above, the performer performs the lyric determination process by pressing a plurality of keys, and then executes the song generation process by designating the pitch by pressing a single key as described above. In this case, in the key processing of FIG. 3, the determination in step S304 is YES when the single key is pressed first, the current time is stored in the key pressing start time variable t_start in step S305, and the state variable state is set. The value is set to the value ACTIVE indicating the key press, and the contents of the set variable k indicating the contents of the key pressed this time are stored in the key pressed set variable on_keys. After that, while the keyboard process is executed several times, the determination in step 307 is less than N milliseconds, and eventually it becomes N milliseconds or more, and the process proceeds from step S307 to step S308. Is “1”.

  In this case, the CPU 101 stores the content of the key pressed set variable on_keys in the reproduction key set variable play_keys (step S310), and instructs the singing synthesis unit 105 to generate a singing sound (step S311). Thereby, in the singing voice synthesis unit 105, singing data corresponding to the lyrics determined in step S309 is generated and pronounced with the pitch of the newly pressed single key. After the singing generation process, in order to clear the key pressing state by a single key, the variable state indicating the state is returned to the IDLE state indicating the waiting state, and the content of the key pressed set variable on_keys is set to the empty set φ (step S312). ). Then, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S301 for the next keyboard processing in the previous key set variable p_keys (step S313), and then is shown in the flowchart of FIG. This keyboard processing in step S203 is terminated.

  FIG. 4 is a flowchart showing a detailed example of the lyrics determination process in step S309 of FIG. 3 in the first embodiment (the same applies to the third embodiment). Here, the CPU 101 draws (refers to) the lyric data on the keyboard / lyric conversion table stored in the ROM 102 based on the contents of the key pressed aggregate variable on_keys, and stores it in the lyric variable lylic on the RAM 103 ( Step S401). The keyboard / lyric conversion table is data in which, for example, for every two key combinations, which lyrics such as “A”, “I”, “U” are selected by the combination.

FIG. 5 is a flowchart showing a detailed example of the song generation process in step S311 of FIG. 3 in the first embodiment (the same applies to the third embodiment). Here, the CPU 101 determines the singing composition unit 105 based on the contents of the lyrics variable lylic set in step S401 of FIG. 4 and the contents of the reproduction key set variable play_keys set in step S310 of FIG. The singing sound is instructed to be generated, and the singing synthesis unit 105 generates an output waveform W of singing data (step S501).
Specifically, the CPU 101 supplies the singing synthesis unit 105 in FIG. 1 with pronunciation instruction data instructing generation of singing data of the lyrics indicated by the lyric variable lylic and generation at the pitch indicated by the reproduction key set variable play_keys. To do.

  Thereafter, the CPU 101 instructs the singing synthesis unit 105 to pronounce the output waveform W of the singing data. As a result, the singing synthesis unit 105 selects the singing data of the lyrics instructed from the CPU 101 among the plurality of singing data stored in the internal waveform data memory, and the singing data is selected from the sound instructed by the CPU 101. Reading is performed at a reading speed corresponding to high, and interpolation processing of waveform sample data is executed as necessary to generate a final output waveform W, which is output to the sound generation unit 106 of FIG. Thereby, the singing voice of the output waveform W is emitted.

  In the first embodiment described above, the key combination pattern in the lyric determination process is described in order to quickly perform the key press for the song generation process after the lyric designation key press for the lyric determination process. It is preferable that the key combination pattern for designating lyrics and the key of the pitch to be pronounced are close to each other. As an example, it is desirable that the key combination pattern for designating lyrics be within one octave of the keyboard 104 in FIG. 1 so that the same lyrics can be designated by the same operation in any octave of the keyboard 104.

In the first embodiment, for example, the following control process may be adopted when a key is pressed after another key has been pressed for at least N milliseconds.
・ When a single key is released and another single key is pressed after more than N milliseconds have passed since the key was pressed, the lyrics are pronounced with the new key press. (For example, pronounce “ka-ka”).
・ If a single key is not released and another single key is pressed after more than N milliseconds have elapsed since the single key was pressed, the vowel sound is the new key pressed. Sing in a form that connects the two (for example, pronounced “ka-ah”).
The above example may be reversed. In addition, a control operation may be executed in which the lyric is replayed with the current lyrics basically only by keyboard operation, and the vowels are connected only when the sustain pedal is depressed.

  It should be noted that whether the key depression is designated as lyrics designation or song generation designation can be distinguished according to whether the number of simultaneous key depressions at a certain timing is single or plural. So, for example, if you press two keys simultaneously to specify the lyrics and then press another key without releasing the key, the key press is regarded as a pronunciation in the above. A control operation may be executed in which the lyrics stored as the lyrics by the combination of the three keys are overwritten as a part of the designation. In addition, when another key is pressed without releasing the key after sounding, the above sound is pronounced so as to make a transition from the previous sounding to the later sounding, but a new sounding is made while leaving the previous sounding. The control operation may be executed.

  The key release operation is not particularly mentioned in the first embodiment, but may be released naturally at a predetermined time after the key is pressed, or a predetermined control operation may be executed for the player's key release operation. Good.

  Next, a second embodiment of the present invention will be described. FIG. 6 is a flowchart showing a detailed example of the lyrics determination process in the second embodiment. In the second embodiment, the RAM 103 has a lyrics cue variable lylic_queue of a lyrics FIFO (first-in first-out) method. Then, each time the performer continuously executes simultaneous key pressing of a plurality of keys, the CPU 101 repeats the lyrics determination process in step S309 in the keyboard process illustrated by the flowchart of FIG. 3 as in the first embodiment. In this process, as shown in step S601 of FIG. 6, the lyrics data determined by referring to the keyboard / lyric conversion table in the ROM 102 with the content of the key pressed set variable on_keys corresponding to each key pressed. Are overwritten on the lyrics cue variable lylic_queue without overwriting the lyric variable lylic on the RAM 103 shown in the first embodiment.

  FIG. 7 is a flowchart illustrating a detailed example of the song generation process in the second embodiment. After the above-described continuous process of simultaneously pressing a plurality of keys, the CPU 101 repeats the song generation process of step S311 in the keyboard process of FIG. 3 each time the performer continuously executes the pressing of a single key. Run. In this process, the CPU 101 pops (takes out and removes the head) the top lyric data stored in the lyric queue variable lylic_queue, and substitutes the corresponding song data into the lyric variable lylic (step S701). After that, the processing of steps S702 and S703 similar to steps S501 and S502 of FIG. 5 in the first embodiment is executed, and the output waveform W of the song data is generated and pronounced. Each time the performer presses a single key, the song generation process illustrated in the flowchart of FIG. 7 is executed, and the song data in the lyrics cue variable lylic_queue is popped one by one. In this way, each time a pitch by a single key is designated, each song data corresponding to each lyrics designated in a lump can be generated and pronounced sequentially.

  When the lyrics cue variable “lylic_queue” is empty, control operations such as not sounding, sounding with the last lyrics, and sounding with the vowels of the last lyrics may be executed. When the lyrics cue variable lylic_queue overflows, a control operation such as discarding the top lyric data of the cue and adding it to the end and overwriting the last lyric data of the cue may be executed. The first embodiment corresponds to the case where the maximum length of the lyrics cue variable lylic_queue is 1 in the second embodiment.

  Next, a third embodiment of the present invention will be described. In the third embodiment, the lyrics determination process and the song generation process are always executed alternately with respect to the first embodiment described above. In the third embodiment, pitch specification with a plurality of keys is possible, so that the singing is not limited to single singing, and choral performance is possible.

  FIG. 8 is a flowchart showing a detailed example of the keyboard process in step S203 of FIG. 2 in the third embodiment. Each set variable is the same as in the case of the first embodiment described above with reference to FIG.

  First, the CPU 101 scans the keyboard 104 and substitutes a set of keys that are pressed into a key set variable c_keys on the RAM 103 (step S801).

  Next, the CPU 101 determines whether or not the content of the previous key set variable p_keys is not the empty set φ and the content of the current key set variable c_keys is the empty set φ (step S802). If the determination in step 802 is true (hereinafter referred to as “TRUE”), it indicates a state where all of the previous key presses have been released, and if it is false (hereinafter referred to as “FALSE”) This shows a state where the key is pressed.

  If the determination in step S802 is FALSE, the CPU 101 next determines whether the content of the previous key set variable p_keys is different from the content of the current key set variable c_keys (step S803).

  When the content of the previous key set variable p_keys is the same as the content of the current key set variable c_keys (the determination in step S803 is FALSE), the CPU 101 calculates the current key calculated in step S801 for the next keyboard process. After the contents of the set variable c_keys are stored in the previous key set variable p_keys (step S813), the current keyboard process in step S203 of FIG. 2 shown in the flowchart of FIG.

  When the content of the previous key set variable p_keys is different from the content of the current key set variable c_keys (determination in step S803 is TRUE), the CPU 101 is a stage LYLIC for executing the lyrics determination process. The contents of the stage variable phase on the RAM 103 indicating whether or not the stage PLAY is to be executed is checked.

  Since the initial value of the stage variable phase is set to the stage LYLIC in which the lyrics determination process is executed in step S201 of FIG. 2, the CPU 101 first proceeds from step S804 to S805. Here, the CPU 101 adds the contents of the difference set between the current key set variable c_keys and the previous key set variable p_keys to the contents of the pressed key set variable on_keys (step S805), and then in FIG. 3 in the first embodiment. The same lyrics determination process as in step S309 is executed (step S806). Thereafter, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S801 for the next keyboard process in the previous key set variable p_keys (step S813), and is shown in the flowchart of FIG. The current keyboard process in step S203 is terminated.

  Thereafter, when the performer releases the key that was pressed for the lyrics determination process, immediately after that, at the timing when the keyboard process of step S203 of FIG. 2 is executed, the determination of step S802 of FIG. It becomes. In this case, the CPU 101 determines whether the content of the stage variable phase is the stage LYLIC for executing the lyrics determination process or the stage PLAY for executing the song generation process.

  When the lyrics determination process is executed, the content of the stage variable phase is LYLIC (steps S804 → S805), and thus the CPU 101 proceeds from step S809 to S810. Here, the CPU 101 changes the content of the stage variable phase to the stage PLAY for executing the singing pronunciation process. Thereafter, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S801 for the next keyboard process in the previous key set variable p_keys (step S813), and is shown in the flowchart of FIG. The current keyboard process in step S203 is terminated.

  As a result, at the timing when the keyboard processing is executed after the player presses the key next time, the determination in step S802 is FALSE, the determination in step S803 is TRUE, and the determination in step S804 is PLAY. As a result, the CPU 101 executes steps S807 and S808 similar to steps S310 and S311 of FIG. 3 in the first embodiment, and executes a song generation process. Thereafter, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S801 for the next keyboard process in the previous key set variable p_keys (step S813), and is shown in the flowchart of FIG. The current keyboard process in step S203 is terminated.

  Thereafter, when the performer releases the key that was pressed for the song generation process, the determination in step S802 in FIG. 8 is true at the timing immediately after that the keyboard process in step S203 in FIG. 2 is executed. In step S809, the CPU 101 checks the contents of the stage variable phase again.

  Since the content of the stage variable phase is PLAY when the song generation process is executed (step S804 → S807), the CPU 101 proceeds from step S809 to S811. Here, the CPU 101 changes the content of the stage variable phase to the stage LYLIC for executing the lyrics determination process. Further, the CPU 101 sets the contents of the key pressed set variable on_keys to an empty set φ (step S812). Thereafter, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S801 for the next keyboard process in the previous key set variable p_keys (step S813), and is shown in the flowchart of FIG. The current keyboard process in step S203 is terminated.

  As described above, in the third embodiment, each time the performer repeatedly presses and releases the key, the stage LYLIC for executing the lyrics determination process and the stage PLAY for executing the song generation process are toggled alternately. It is. For this reason, when the performer performs a key press to execute the song generation process, it is not always necessary to perform a single key press as in the first embodiment, and a key press designation with a plurality of keys is also performed. And choral singing is possible.

When a plurality of keys are pressed sequentially instead of simultaneously in the singing sound generation process, the sound generation may be controlled corresponding to each key pressed. Again, as mentioned above,
・ If a single key is not released and another single key is pressed after more than N milliseconds have elapsed since the single key was pressed, the vowel sound is the new key pressed. Sing in a form that connects the two (for example, pronounced “ka-ah”).
It is good. Moreover, the pronunciation of the key pressed later may include consonants as well as vowels. In the third embodiment, the timing of returning to the stage LYLIC for executing the lyrics determination process from the stage PLAY for executing the song generation process is the timing when all keys are released, but the key is pressed until immediately before all the keys are released. The lyrics may be determined by the combination of keys that have been used, and the process may be shifted to the stage PLAY where the song generation process is executed.
In the LYLIC stage of lyric determination processing, at the time of the first key press, the lyrics are determined by the combination of keys pressed at the same time, and the lyrics are not changed by the subsequent key press. After the key is released, the process may be shifted to the stage PLAY for executing the singing generation process (actual pronunciation is after the key is pressed again).
Further, in the stage LYLIC for executing the lyrics determination process, when the key is released for the first time, the lyrics may be determined by the combination of the keys released, and then the process may be shifted to the stage PLAY for executing the song generation process. .

  FIG. 9 is a diagram illustrating a hardware configuration example of the fourth embodiment of the singing voice synthesizing apparatus. This configuration differs from the configuration shown in FIG. 1 of the first, second, and third embodiments in that a foot pedal 107 is provided. In the fourth embodiment, the foot pedal 107 switches between the stage of executing the lyrics determination process and the stage of executing the song generation process. For example, when the foot pedal 107 is depressed, the lyric determination process is performed, and when the foot pedal 107 is released, the singing generation process is performed (switching by toggle may be used).

  FIG. 10 is a flowchart showing a detailed example of the keyboard process in step S203 of FIG. 2 in the fourth embodiment.

  First, the CPU 101 scans the keyboard 104 and substitutes a set of keys that are pressed into a key set variable c_keys on the RAM 103 (step S1001).

  Next, the CPU 101 determines whether or not the result of subtracting the contents of the previous key set variable p_keys from the contents of the current key set variable c_keys is an empty set φ, that is, the contents of the previous key set variable p_keys and the current key set variable c_keys. It is determined whether or not the contents are the same (step S1002).

  If the determination in step S1002 is TRUE, the CPU 101 ends the keyboard process in step S203 of FIG. 2 shown in the flowchart of FIG. 10 as it is.

  If the determination in step S1002 is FALSE, the CPU 101 checks the state of the foot pedal 107 in FIG. 9 (step S1003).

  When the foot pedal 107 is stepped on, the CPU 101 executes the lyrics determination process indicated by the processes in steps S1004 and S1005 similar to steps S805 and S806 in the third embodiment. Thereafter, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S1001 for the next keyboard process in the previous key set variable p_keys (step S1008), and is shown in the flowchart of FIG. The current keyboard process in step S203 is terminated.

  When the foot pedal 107 is not stepped on, the CPU 101 executes a song generation process indicated by the processes of steps S1006 and S1007 similar to steps S807 and S808 in the third embodiment. Thereafter, the CPU 101 stores the contents of the current key set variable c_keys calculated in step S1001 for the next keyboard process in the previous key set variable p_keys (step S1008), and is shown in the flowchart of FIG. The current keyboard process in step S203 is terminated.

In the prior art, real-time lyrics cannot be specified, or both hands are blocked only by singing performance. However, according to the first, second, third, and fourth embodiments described above, only one hand is used. Since singing performance becomes possible, it becomes possible to operate other performance controls with the other hand, and the range of performance can be expanded. In addition, for example, the keyboard can be divided into an upper half and a lower half, and separate singer data can be used for each key range so that different lyrics can be sung simultaneously.
In the first to fourth embodiments, a keyboard key is used as a performance operator, but the present invention is not limited to this. For example, any kind of pad can be used as a performance operator, such as a plurality of types of pads for generating percussion instrument sounds, a keyboard used for a PC, and the like.

Regarding the above embodiment, the following additional notes are disclosed.
(Appendix 1)
A state transition unit that determines a performance operation performed by the performance operator and shifts to one of the first and second states based on the determination result;
A lyric determination unit that determines lyrics based on the determined performance operation when transitioning to the first state;
A singing generation unit that generates singing data for singing at a pitch based on the determined performance operation when the state is shifted to the second state;
A pronunciation generation unit for generating pronunciation data to be generated based on the singing data generated by the lyrics generation unit;
A singing generation device comprising:
(Appendix 2)
The lyrics determining unit shifts to the first state when a plurality of the performance operators are operated at the same time within a predetermined period, and shifts to the second state when the single performance operator is operated. The song production | generation apparatus of Additional remark 1 which performs the process to perform.
(Appendix 3)
The lyrics determining unit further stores the determined lyrics in a memory sequentially,
The singing generation unit reads the lyrics stored in the memory in the order stored in advance each time the pitch determination unit determines the pitch, and the read lyrics are determined The song generator according to appendix 1 or 2, which generates song data for singing at a pitch.
(Appendix 4)
The contents of the performance operation performed by the performance operator and the end of the performance operation are discriminated, and each time the end of the performance operator is discriminated, the state shifts from one of the first and second states to the other state. A state transition part;
A lyrics determining unit that determines lyrics based on the content of the determined performance operation when the first state is entered;
A singing generation unit that generates singing data for singing at a pitch based on the content of the determined performance operation when the state is shifted to the second state;
A singing generation unit for generating pronunciation data for generating sound based on the singing data generated by the lyrics generation unit;
A singing generation device comprising:
(Appendix 5)
The performance operator includes a foot pedal, and the state transition unit shifts from one of the first state and the second state to the other state each time the foot pedal is operated. The song generating device according to 1.
(Appendix 6)
A process of determining a performance operation performed by the performance operator, shifting to one of the first and second states based on the determination result, and the determined performance when shifting to the first state A control unit that executes a process for determining lyrics based on an operation and a process for generating song data for singing at a pitch based on the determined performance operation when the second state is entered. When,
A singing generation unit for generating pronunciation data for generating sound based on the singing data generated by the lyrics generation unit;
A singing generation device comprising:
(Appendix 7)
A singing generation method used in a singing generation device including a control unit and a singing generation unit,
The control unit is
The performance operation performed by the performance operator is determined, and the state is shifted to one of the first and second states based on the determination result,
When transitioning to the first state, the lyrics are determined based on the determined performance operation,
When transitioning to the second state, generating singing data for singing at a pitch based on the determined performance operation,
The song generator is
A singing generation method for generating pronunciation data for generating sound based on the generated singing data.
(Appendix 8)
To a computer used as a singing generation device having a singing generation unit for generating pronunciation data for generating sound based on the generated singing data,
Determining a performance operation performed by the performance operator, and shifting to one of the first and second states based on the determination result;
Determining lyrics based on the determined performance operation when transitioning to the first state;
Generating singing data for singing at a pitch based on the determined performance operation when transitioning to the second state;
A program that executes
(Appendix 9)
The song generation device according to attachment 1,
Multiple performance controls,
A pronunciation unit that produces a singing sound based on the pronunciation data from the song generation device;
Electronic musical instrument with

100 song generation apparatus 101 CPU
102 ROM
103 RAM
104 Keyboard 105 Singing synthesis unit 106 Sound generation unit 107 Foot pedal 107

Claims (11)

A specifying unit for specifying an operation state of a plurality of performance operators to which different pitches are assigned;
A state transition unit that transitions to one of the first and second states based on the operation state ;
A lyrics determining unit that determines lyrics based on the operation state when transitioning to the first state;
A pitch determination unit that determines a pitch based on the operation state when the second state is entered ;
A pronunciation generation unit for generating pronunciation data for generating the determined lyrics at the determined pitch ; and
A singing generation device comprising: The operation state includes the combination of a plurality of the performance operators which are operated, the number of engineered the performance operators, the state of one of the of,
The song production | generation apparatus of Claim 1. The lyrics determining unit determines the lyrics based on the operation state used for the determination of the transition when transitioning to the first state,
The pitch determination unit determines a pitch based on the operation state used for the determination of the transition when the transition is made to the second state.
The song production | generation apparatus of Claim 1 or 2. A song generator for generating song data for singing the lyrics already determined by the lyrics determiner at the pitch determined by the pitch determiner when transitioning to the second state; Prepared,
The pronunciation generation unit generates the pronunciation data to be generated based on the song data generated by the song generation unit;
The song production | generation apparatus of Claim 3. The lyrics determining unit further stores the determined lyrics in a memory sequentially,
The singing generator, every time pitch by the pitch determination unit is determined, reads the order previously stored lyrics stored in the memory, which is the determining the read lyric The song production | generation apparatus of Claim 4 which produces | generates the song data made to sing with a pitch. The state transition unit transitions to either the first state or the second state based on the number of the performance operators that have been operated within a predetermined period.
The song production | generation apparatus in any one of Claims 1 thru | or 5. The state transition unit transitions to the first state when a plurality of the performance operators are operated simultaneously within a predetermined period, and operates the single performance operator within the predetermined period. The singing apparatus according to claim 6, wherein the singing state is shifted to the second state. The operation state includes a timing when the operation of the performance operator is finished,
The state transition unit transitions from one of the first and second states to the other state every time the end of the operation is determined.
The song production | generation apparatus in any one of Claims 1 thru | or 5. A singing generation device according to any one of claims 1 to 8 ,
Multiple performance controls,
A pronunciation unit that produces a singing sound based on the pronunciation data from the song generation device;
Electronic musical instrument with The device
Specific processing for specifying the operation state of a plurality of performance operators assigned different pitches;
A state transition process for transitioning to either the first or second state based on the operation state;
Lyrics determination processing for determining lyrics based on the operation state when transitioning to the first state;
A pitch determination process for determining a pitch based on the operation state when transitioning to the second state;
Pronunciation generation processing for generating pronunciation data for generating the determined lyrics at the determined pitch; and
Singing generation method to execute. On the computer,
Specific processing for specifying the operation state of a plurality of performance operators assigned different pitches;
A state transition process for transitioning to either the first or second state based on the operation state;
Lyrics determination processing for determining lyrics based on the operation state when transitioning to the first state;
A pitch determination process for determining a pitch based on the operation state when transitioning to the second state;
Pronunciation generation processing for generating pronunciation data for generating the determined lyrics at the determined pitch; and
A program that executes