JP3405123B2 - Audio data processing device and medium recording data processing program - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice data processing technique, and more particularly to a voice data processing technique for causing a voice producing apparatus to produce a voice.

[0002]

2. Description of the Related Art There is known a voice data processing device capable of processing voice data supplied to a voice producing device. The voice pronunciation device includes a voice synthesis circuit such as a formant sound source. The formant sound source generates a sound signal by synthesizing a formant formed by frequency-analyzing the sound.

A syllable is one of the so-called 50 syllables, for example, "ka". Syllables can be decomposed into phoneme strings on the time axis. For example, the syllable “ka” has a phoneme sequence “CL (7.5 ms) + kha (4 × 7.5 ms) +
aj (infinite length) ”. Phonemes are, for example, “CL”, “kha”, and “aj”. The formant sound source generates a voice signal by inputting a phoneme string.

As a voice data processing device according to the prior art, Japanese Patent Laid-Open Nos. 9-50287 and 9-4417 are available.
No. 9 publication is known. These store melody data and lyrics data in a memory, and read both data sequentially from the memory and supply them to the formant sound source. The formant sound source receives the data and generates a singing voice.

A word processor capable of editing syllables is known as a data processing device. However, since there is no data processing device capable of editing phonemes, the voice data is manually set.

[0006]

[Patent Document 1] Japanese Patent Application Laid-Open No. 9-50287
The data processing device of the publication has to input lyrics data in Roman characters and cannot use kana or kanji, so that inputting or editing the lyrics data is inconvenient.
For example, existing lyrics data in which kana or kanji are mixed cannot be used.

Further, since the lyrics data is assigned to each note, it is necessary to provide a note delimiter (lyric delimiter) symbol in the lyrics data, but the lyrics delimiter is limited to only one symbol (for example, space "_"). ing. Therefore, a line feed mark (a mark that indicates a line feed position on the lyrics display screen) and a sub-line feed mark (not normally used, but the lyrics up to the line feed mark are included in one line due to the display device such as a small display space) It is not possible to use only the line break mark as the lyrics delimiter in the lyrics data for karaoke that contains two types of line break marks (marks indicating the position where the line breaks before the line break mark) However, both the line break mark and the sub line break mark cannot be used as lyrics delimiters.

Further, lyrics data and the like have their own data format, and a dedicated device is required as a lyrics data reproducing device, and cannot be reproduced by a general sequencer or the like.

An object of the present invention is to provide a voice data processing device or a recording medium of a computer program capable of processing text data including kana and the like.

Another object of the present invention is to provide a recording medium for a voice data processing device or a computer program, which can easily edit voice data.

Another object of the present invention is to provide a recording medium for a voice data processing device or a computer program capable of processing voice data having general versatility.

[0012]

According to an aspect of the present invention, a voice data processing device includes a note including text data including non-kana and its reading kana or data in which ruby is paired, and note-based data. And storage means for storing the data, and generating means for discarding the non-kana in the text data and allocating only the reading kana or ruby to the note data in the note data to generate voice data.

[0013]

[0014]

[0015]

[0016]

[0017]

[0018]

[0019]

[0020]

[0021]

1 is a diagram showing a connection between a personal computer 1 and an external sound source device 21. As shown in FIG. The personal computer 1 includes the audio data processing device according to this embodiment. The personal computer 1 may be a sequencer.

First, the configuration of the external sound source device 21 will be described. The detection circuit 33 detects the operation of the switch 34 and generates a switch signal. The switch 34 includes, for example, a switch for setting various parameters.

The bus 22 has a RAM in addition to the detection circuit 33.
24, ROM 25, CPU 26, display circuit 28, MID
The I interface 23, the voice synthesis circuit 32, and the tone waveform synthesis circuit 29 are connected.

The ROM 25 stores formant data for synthesizing voice, other various data, and a computer program. The RAM 24 stores flags, buffers and the like. The computer program is ROM2
Instead of being stored in RAM 5, it may be stored in RAM 24. The CPU 26 performs calculation or control according to a computer program stored in the ROM 25 or the RAM 24.

The CPU 26 generates a musical tone parameter based on the performance data received from the personal computer 1 via the MIDI interface 23 and supplies it to the musical tone waveform synthesizing circuit 29. The musical tone waveform synthesis circuit 29 generates a musical tone signal according to the supplied musical tone parameter and supplies the musical tone signal to the sound system 31. Sound system 31
Includes a D / A converter and a speaker, converts a supplied digital tone signal into an analog format and produces a sound.

The musical tone waveform synthesizer circuit 29 includes a waveform memory system, an FM system, a physical model system, a harmonic synthesis system,
Any method such as a formant synthesis method or a VCO + VCF + VCA analog synthesizer method may be used.

The MIDI interface 23 is an M interface for the MIDI interface 8 of the personal computer 1.
Connected with an IDI cable. The external sound source device 21 and the personal computer 1 can perform MIDI communication.

The CPU 26 receives voice data from the personal computer 1 via the MIDI interface 23 and stores it in the RAM 24. The voice data is
Includes note data and lyrics data.

The CPU 26 reads out the voice data stored in the RAM 24, and supplies the formant data, the note number (pitch), etc. to the voice synthesis circuit 32 based on the formant data stored in the ROM 25. The formant data is, for example, formant center frequency data, formant bandwidth data, formant level data and the like.

The voice synthesizing circuit 32 generates a voice signal according to the supplied formant data and the like. The voice signal has a predetermined pitch and corresponds to a singing voice. The voice synthesis circuit 32 may use a formant synthesis method (formant sound source) or any other method. When the method of the voice synthesizing circuit 32 is other than the formant synthesizing method, the supplied data is not the formant data but the data peculiar to the voice synthesizing method.

A case where the voice synthesis circuit 32 is a formant sound source will be described. To the voice synthesis circuit 32, formant data or the like generated according to voice data is input. The formant data includes formant center frequency data, formant level data, formant bandwidth data, and the like.

The voice synthesis circuit 32 has a voiced sound synthesis group VTG and an unvoiced sound synthesis group UTG. The voiced sound group VTG is the first one according to the formant data and the like.
4 to 4 have voiced sound synthesis units V1, V2, V3, and V4 for forming voiced sound formants, respectively. The unvoiced sound synthesis group UTG, according to the formant data etc.,
It has unvoiced sound synthesis units U1, U2, U3, and U4 for forming first to fourth unvoiced sound formants, respectively.

The first formant generation unit TG1 has a voiced sound synthesis unit V1 and an unvoiced sound synthesis unit U1 for forming the first formant, and the second formant generation unit TG2 has a voiced sound synthesis for forming the second formant. The third formant generation unit TG3 includes a voiced sound synthesis unit V3 and an unvoiced sound synthesis unit U3 for forming a third formant, and the fourth formant generation unit TG4 includes a fourth formant generation unit TG4. A voiced sound synthesizing section V4 and an unvoiced sound synthesizing section U4.

The synthesizers V1 to V4 and U1 to U4 output the audio signals corresponding to the generated formants, respectively. Those audio signals are added up, and the sound system 3
1 is supplied. The sound system 31 converts the supplied digital audio signal into an analog format and produces sound.

A more specific structure of the formant sound source is as follows.
For example, it is shown in FIG. 1 of JP-A-3-200299.

The voice synthesis circuit 32 is not limited to the case where it has four system formant generation units TG1 to TG4. It may have more or less formant generation units.

A special unit may be provided for special sounds such as plosive sounds. Alternatively, a plurality of sound generation channels may be prepared so that a song with two or more voices can be generated. In this case, a plurality of tone generation channels may be formed by using one circuit in a time division manner, or one tone generation channel may be configured by one circuit. .

Speech synthesis circuit 32 and tone waveform synthesis circuit 29
Is not limited to being configured by using dedicated hardware, but may be configured by using a DSP + microprogram, or may be configured by a program of CPU + software.

Next, the configuration of the personal computer 1 will be described. The detection circuit 11 detects input of a key (numerical key, character key, etc.) on the keyboard 12 and generates a key signal. The detection circuit 9 detects a movement operation of the mouse 10 and a switch (right button, left button, etc.) operation, and generates a mouse signal. The operator operates the mouse 10 or the keyboard 12
Can be used to edit the audio data.

The display circuit 7 can display an edit screen for editing voice data and the like. The operator can edit the voice data while referring to the edit screen on the display circuit 7.

The bus 2 has a detection circuit 11, a detection circuit 9 and a display circuit 7, as well as a MIDI interface 8 and RA.
The M3, the ROM 4, the CPU 5, the external storage device 13, and the communication interface 14 are connected.

The ROM 4 stores various parameters and computer programs. The RAM 3 stores flags, buffers, performance data, voice data and the like. Also, R
The AM 3 can also store a computer program, performance data, voice data, etc. supplied from the outside via the external storage device 13 or the communication interface 14. The CPU 5 performs calculation or control for editing or processing of voice data according to a computer program stored in the RAM 3 or the ROM 4.

The timer 6 supplies time information to the CPU 5. The CPU 5 can perform interrupt processing at predetermined time intervals according to the time information.

The MIDI interface 8 is, as described above, the MIDI interface 2 of the external tone generator 21.
3 is connected with a MIDI cable. The personal computer 1 can transmit performance data and voice data to the external sound source device 21 via the MIDI interface 8.

The communication interface 14 is connected to a communication network 41 such as a local area network (LAN), the Internet or a telephone line. A server computer 42 is connected to the communication network 41. The personal computer 1 can receive voice data or a computer program from the server computer 42 via the communication network 41.

The external storage device 13 includes an interface for the external storage device, and is connected to the bus 2 via the interface. The external storage device 13 is, for example, a floppy disk drive (FDD), a hard disk drive (HDD), a magneto-optical disk (MO) drive, C
Examples include D-ROM (compact disc-read only memory) drives, digital multipurpose disc (DVD) devices, and the like. Performance data and voice data are stored in the external storage device 1.
3 or RAM3.

The computer program or the like may be stored in the external storage device 13 (for example, hard disk) without being stored in the ROM 4. R from hard disk
By reading the computer program or the like into the AM 3, the CPU 5 can be caused to perform the same operation as when the computer program or the like is stored in the ROM 4. In this way, by adding the computer program or the like from another external storage medium such as a CD-ROM to the hard disk, it is possible to easily add or upgrade the computer program or the like.

The communication interface 14 is connected to a communication network 41 such as a local area network (LAN), the Internet or a telephone line, and is connected to the server computer 42 via the communication network 41. If a computer program or the like is not stored in the external storage device 13, the server computer 4
2 can download computer programs and the like. The personal computer 1 serving as a client transmits a command requesting a download of a computer program or the like to the server computer 42 via the communication interface 14 and the communication network 41. Upon receiving this command, the server computer 42 delivers the requested computer program and the like to the personal computer 1 via the communication network 41, and the personal computer 1 receives these computer programs and the like via the communication interface 14. The download is completed by accumulating in the external storage device 13.

The personal computer 1 reproduces the performance data and voice data stored in the RAM 3 in accordance with a predetermined tempo, and reproduces the reproduced performance data and voice data via the MIDI interface 8 to the external tone generator 2.
Send to 1. The external sound source device 21 uses the MIDI interface 2 to send the transmitted performance data and voice data.
The received performance data is supplied to the musical tone waveform synthesizing circuit 29 to form an accompaniment tone signal, and the received voice data is supplied to the voice synthesizing circuit 32 to form a singing signal. That is, the personal computer 1 and the external sound source device 21 constitute a system for generating a song accompanied by an accompaniment tone.

Next, the processing performed by the voice data processing device (personal computer 1) will be described.

FIG. 2 is a diagram showing the import processing.
In the import process, the text data (lyric data) 52 is assigned to the performance data 51, and the note plus data 5
This is a process of generating 3. In this specification, performance data 5
1, the text data 52, and the note plus data 53 are all called voice data.

The performance data 51 has, for example, note events N1 to N8 corresponding to eight notes (notes). Each of the note events N1 to N8 includes, for example, note-on timing data, note number (pitch), velocity (volume), and gate time (time from note-on to note-off).

The text data 52 is, for example, "red (red) _i / sunset [yuu! Hi] _ is \". here,
() Indicates reading kana, [] indicates ruby, and “_,
!! ", Etc. indicate lyrics delimiters, and" /, \ "etc. indicate exhalation symbols.

The exhalation symbol is a symbol indicating breathing. When there is an exhalation symbol, the lyrics data before that is pronounced for the gate time, and then becomes silent. When there is no exhalation symbol, one lyric data and the next lyric data are pronounced so as to be smoothly connected. That is, the game time is ignored.

The lyrics delimiter is the lyrics data (first lyrics event) corresponding to one note (note event).
And the lyrics data corresponding to the next note (the second lyrics event). The lyrics data 52 is decomposed into lyrics events according to the lyrics delimiter and assigned to each note event N1 to N8. For example, "N1, red", "N2, i ▽", "N3, Yu", "N4,"
Hi ”,“ N5, is ▽ ”. Here, “∇” is the exhalation data corresponding to the exhalation symbol “/, ¥”, and indicates the breathing and also serves as a lyrics delimiter, as can be seen in the example of “N2, i∇”. A set of the note event and the lyrics (character string) event or the phoneme string event (a phoneme string event will be described later) is called a note plus event (an event in which a lyrics event is added to a note). Note plus data 5
3 is a group of note plus events.

In the note-plus data 53, the kanji is deleted and only the corresponding kana is left in consideration of the reading kana and ruby. When there is a reading kana symbol "()", the kanji "red" and the reading kana symbol "()" before () are deleted and the kana "kana" in () is left. The same applies to the case of the ruby symbol "[]".

The kana in the note plus data 53 is converted into a phoneme string. For example, the kana "ka" is the phoneme sequence "CL
(7.5 ms) + kha (4 × 7.5 ms) + aj (infinite length) ”. A syllable such as "ka" can be decomposed into a phoneme string on the time axis. The phoneme is, for example, "C
L ”,“ kha ”, and“ aj ”.

The personal computer 1 stores a table for converting between kana and phoneme strings.
This table can be used to convert from kana to phoneme sequences. However, it is not possible to convert kanji, symbols, alphanumeric characters, etc. into phoneme strings. This is because there is a possibility that there are multiple ways of reading kanji, symbols, alphanumeric characters, etc.

The kana is converted into a phoneme string, and the kanji and symbols are left as they are. The personal computer 1 stores a dictionary for converting Chinese characters and the like into kana. By using this dictionary, it is possible to convert kanji and the like into kana and then convert the kana into phoneme strings. The personal computer 1 supplies a phoneme string or the like to the external tone generator 21 (FIG. 1). The external sound source device 21 receives a phoneme string or the like and produces a sound.

The personal computer 1 can use text data as lyrics data. For example, existing text data such as text data for karaoke can be used.

Further, the lyrics delimiter is "_ ,!", etc., and the exhalation sign is "/, \" etc., and a plurality of types can be set for each. For example, when using text data for karaoke, it is possible to set both the line feed mark and the sub line feed mark as lyrics delimiters and the page break mark as an exhalation symbol. Since multiple types of lyrics delimiters can be set, both the line feed mark and the sub-line feed mark can be used as the lyrics delimiter.

The line feed mark is a mark indicating a line of lyrics when the lyrics are displayed on the screen. The sub-line feed mark is a mark for line feed used only when displaying on a small screen. That is, on the normal screen, a line feed is made only by the line feed mark, and on the small screen, a line feed is made by both the line feed mark and the sub line feed mark. For example, "_" can be used as a line feed mark and "!" Can be used as a sub line feed mark. Note that the lyrics delimiter and exhalation symbol are not limited to the case where a symbol originally inserted for other purposes such as a line feed mark or a lucky line mark is diverted as the lyrics delimiter or exhalation symbol. A symbol arbitrarily set and inserted may be used as a lyrics delimiter or an exhalation symbol.

FIG. 3 shows the CP of the personal computer 1.
It is a flowchart which shows the text import process which U performs. As described above (FIG. 2), the text import process is a process of allocating the text data 52 to the performance data 51 and generating the note plus data 53.

At step SA1, a text file to be imported is selected. The text file includes the text data 52 (FIG. 2) and is stored in the external storage device 13 (FIG. 1).
Etc. The operator uses the mouse 10 or the like to
One can be selected from one or more text files.

At step SA2, the pointer is set at the position of the first note event N1 in the performance data 51. The performance data 51 is stored in the RAM 4 or the external storage device 13.
(FIG. 1) and the like.

In step SA3, the text data 52
Set the pointer to the position of the first character code in.
For example, the pointer is set at the position of “red” in the text data 52.

In step SA4, it is checked whether the character code matches any of the character codes set as the exhalation symbol. For example, "/, \" is set as the exhalation symbol. If they do not match, the process proceeds to step SA5 following the NO arrow.

In step SA5, it is checked whether the character code matches any of the character codes set as the lyrics delimiter. For example, “_ ,!” is set as the lyrics delimiter. If they do not match, the process proceeds to step SA6 following the NO arrow.

At step SA6, the character code is added to the contents of the character string register. String register is 1
It is a register for storing one or a plurality of character codes (hereinafter referred to as lyrics event) corresponding to a note, and stores nothing at the initial stage.

At step SA7, the pointer is set at the position of the next character code. For example, text data 5
Set the pointer to the position of "(" in 2. Then,
Returning to step SA4, the above processing is repeated.

If the character code at the position of the pointer set in step SA7 is "_" or "!", The process proceeds to step SA5 via step SA4, and the character code "_" or "!" It judges that it is a lyric delimiter and follows the YES arrow to proceed to step SA9.

On the other hand, if the character code at the position of the pointer set in step SA7 is "/" or "\", it is determined in step SA4 that the character code "/" or "\" is an exhalation symbol. , YES to the step SA8. In step SA8, the exhalation mark “∇” is added to the contents of the character string register. afterwards,
Go to step SA9.

In step SA9, phonetic kana and ruby processing is performed. Specifically, when the lyric event stored in the character string register includes a phonetic kana or ruby, the kanji or the like is deleted to leave the kana. Details of this processing will be described later with reference to the flowchart of FIG. Then, it progresses to step SA10.

At step SA10, the contents of the character string register are added to the note event where the pointer is located,
Generate a note plus event. For example, the lyrics event "red" in the character string register is added to the note event N1 to generate a note plus event "N1, red".

In step SA11, the text data 5
Check if there is a continuation of 2. That is, it is checked whether the text data 52 is the end. If it is not the end, follow the YES arrow and follow step SA1.
Go to 2.

At step SA12, the contents of the character string register are cleared to prepare for the storage of the next lyrics event.

At step SA13, the pointer is set at the position of the next note event in the performance data 51.
For example, the pointer is set at the position of the note event N2.

In step SA14, the text data 5
The pointer is set to the position of the next character code in 2.
For example, the pointer is set at the position of the character code “i”. Then, it returns to step SA4 and repeats the above process.

When the processing of the last character code of the text data 52 is completed, the text import processing is completed according to the NO arrow in step SA11. By this processing, note plus data 53 is generated.

In the above flow chart, the case of importing the text data 52 in the text file has been described, but the text included in the predetermined track in the standard MIDI file may be imported.

FIG. 4 is a flow chart showing details of the reading kana and ruby processing shown in step SA9 of FIG.

At Step SB1, it is checked whether or not there is a character code of a kana enclosed by () or [] in the character string register. For example, “(red)” or “[Yuhi]” in the text data 52 corresponds to this.

In the case of "sunset [yu! Hi] _",
This applies to both the character string "Yu" and the character string "Hi". That is, if there is a kana character enclosed in brackets of either () or [], it corresponds to this.

When the character code is present, the YES arrow is followed and the process proceeds to step SB2. When there is no such character code, the reading kana and ruby processing is ended according to the NO arrow.

At step SB2, the non-kana character code in front of () or [] in the character string register is deleted. The non-kana character code includes kanji, symbols, and alphanumeric characters. For example, "red" or "sunset" in the text data 52
To delete.

At step SB3, () or [] in the character string register is deleted. As a result, for example, the character string "Yu" is converted into "Yu" and the character string "Hi""
Is converted to "hi". After that, the reading kana and ruby processing is ended, and the processing returns to the text import processing of FIG.

By the above-described processing, the kanji or the like in the kana or the like in which the reading kana or ruby is attached is deleted and the kana remains. Kanji and the like are character codes unnecessary for pronunciation.

FIG. 5 shows the CP of the personal computer 1.
It is a flowchart which shows the phoneme sequence conversion process which U performs.
The phoneme sequence conversion process is a process of converting the lyrics event into a phoneme sequence after the note plus data 53 (FIG. 2) is generated.

In the display circuit 7, one note plus event is displayed on one line. That is, one row corresponds to one note plus event. In step SC1, the line to be converted is selected from the note plus data 53. The operator can refer to the display content of each line displayed on the display circuit 7 and select one line or a plurality of lines (that is, one or a plurality of note plus events) by using the mouse 10 or the like.

At step SC2, the pointer is set at the position of the first row in the selected row. For example, the pointer is set at the position of the note plus event "N1, red".

At step SC3, it is checked whether or not the character string (lyric event) of the line is only kana. If only kana is used, phoneme string conversion is performed, so the flow advances to step SC4 following the YES arrow. N if not only kana
Follow the arrow of O and perform step SC6 without performing phoneme sequence conversion.
Go to. If the exhalation symbol is included in addition to the kana, it is determined that only the kana is included.

At step SC4, the phoneme string corresponding to the character string is obtained by referring to the kana-phoneme string table.
For example, a phoneme string corresponding to the character string “red” is obtained.

At step SC5, the character string and the phoneme string are replaced. That is, in the note plus data 53, the character string “red” is deleted and the phoneme string corresponding to it is written. Then, it progresses to step SC6.

At step SC6, it is checked whether or not there is a next row in the selection range row. If there is only one selected range row, there is no next row, so the phoneme string conversion processing is ended following the NO arrow. YES if there is a next line
Follow the arrow to move to step SC7.

At step SC7, the pointer is set at the position of the next row. For example, note plus event "N
2. Set the pointer to the position of "2. Then, it returns to step SC3 and repeats the above process. The exhalation symbol "▽" is not converted into a phoneme sequence. Only the kana "i" is converted into a phoneme sequence.

When all the lines have been processed, step S
In C6, the phoneme string conversion process is ended following the NO arrow. In a later process, the converted phoneme string is supplied to the external sound source device 21 (FIG. 1) and is sounded.

After being converted into a phoneme string, the phoneme string is converted into a character string again, and both the phoneme string and the character string are displayed on the screen of the display circuit 7. The operator can know the character string and the corresponding phoneme string. Also, it is possible to know whether or not the conversion of the phoneme sequence has been normally performed.

Next, the blank line insertion processing will be described. As described above, after assigning the lyrics event to the note event, it may be desired to insert new lyrics between the lyrics. For example, a part of the lyrics is missing, and the note event and the lyrics event are associated with each other in a state where the lyrics after the missing part are jammed forward, and as a result, the correspondence between the note event and the lyrics event is incorrect. This is the case when a Note Plus event is generated in. In that case, a blank line may be inserted at a desired position and new lyrics may be filled in the blank line.

FIGS. 6A to 6D are diagrams for explaining the blank line insertion processing. FIG. 6A shows an example of note plus data. The note plus data has, for example, eight rows of note plus events. The note plus event on the first line is note event N1 and lyrics event L
Has 1.

The operator can use the mouse or the like to specify the position PT at which a blank line is to be inserted and the number of lines LL.
For example, the third row can be designated as the insertion position PT, and two rows can be designated as the number of inserted rows LL.

First, the lyrics events L3 to L8 after the insertion position PT are copied to the buffer of FIG. 6 (B). Note that the buffer may include blank lines.

Next, as shown in FIG.
The lyric event is erased by the number of inserted lines LL (two lines) from the insertion position PT (third line) of the note plus data in (A). That is, the lyrics events L3 and L4 are deleted.

Next, as shown in FIG.
The lyrics event in the buffer (FIG. 6B) is copied to the fifth and subsequent lines in the note plus data in (C). However, since there is no note event behind the note event N8, the remaining lyrics events are collectively assigned to the note event N8. That is, the lyrics events L6 to L8 are assigned to the note event N8.

In the note plus data of FIG. 6D, the lyric event on the blank line is assigned to the note events N3 and N4. As a result, two blank lines have been inserted. Then, in the note plus event after the blank line insertion position, the correspondence between the note event and the lyrics event is changed. For blank lines,
Since the lyrics event has been deleted, it becomes a normal note event instead of a note plus event.

FIG. 7 is a flow chart for realizing the above blank line insertion processing. At step SD1, a blank row insertion position PT and an insertion row LL are designated.

At step SD2, the lyrics event (phoneme string / character string) after the insertion position PT is buffered (see FIG. 6).
(B)).

At step SD3, the lyric event for the number of insertion lines LL is deleted from the insertion position PT (FIG. 6).
(C)).

At step SD4, the buffer (see FIG.
The lyric event of (B) is assigned to the note event after the insertion line number LL from the insertion position PT. At this time, if the lyrics event remains in the buffer even though there are no note events to be assigned, all the remaining lyrics events are added to the last note event. This completes the blank line insertion process.

The method of inserting a blank line in the note plus data of FIG. 6A to generate the note plus data of FIG. 6D may be a method other than the above.

FIGS. 8A to 8F are diagrams for explaining the lyrics event insertion processing. FIG. 8A shows an example of note plus data. The operator can use the mouse or the like to specify the range RG of the lyrics event to be inserted. For example, as the range RG, the lyrics event L5
And L6 can be specified.

First, the lyrics events L5 and L6 within the range RG are copied to the clipboard in FIG. 8B.

Next, as shown in FIG. 8C, the operator can specify the insertion position PT using a mouse or the like. For example, the note plus event “N3, L3” on the third line can be designated as the insertion position PT.

Next, the lyrics events L3 to L8 after the insertion position PT are copied to the buffer of FIG.

Next, as shown in FIG.
(C) Note events N3 and N after the insertion position PT
4, the lyrics event L on the clipboard (Fig. 8 (B))
Assign 5 and L6.

Next, as shown in FIG.
The fifth line “N5, L in the note plus data of (E)
5 ”and thereafter, the lyrics events L3 to L8 in the buffer (FIG. 8D) are allocated. However, note event N8
Since there is no note event behind that, the remaining lyrics events L6 to 8 are assigned to the note event N8 all together. As a result, the lyrics events L5 and L6 are inserted at the insertion position PT in the note plus data of FIG. 8 (F).

FIG. 9 is a flow chart for realizing the above-mentioned lyrics event insertion processing.

At step SE1, a lyrics event having an arbitrary number of lines RG is designated (FIG. 8A) and copied to the clipboard (FIG. 8B).

At step SE2, the insertion position PT of the lyrics event is designated (FIG. 8 (C)).

At step SE3, the lyrics event after the insertion position PT is copied to the buffer (FIG. 8D).

At step SE4, the lyrics events on the clipboard (FIG. 8B) are assigned to the note plus data by the number RG of inserted lines from the insertion position PT (FIG. 8).
(E)).

At step SE5, the buffer (see FIG.
The lyrics event (D) is assigned to the note plus data after the insertion line number RG from the insertion position PT (FIG. 8 (F)). This is the end of the lyrics event insertion process.

The method of inserting the desired lyric event into the note plus data of FIG. 8A to generate the note plus data of FIG. 8F may be a method other than the above.

FIGS. 10A to 10E are diagrams for explaining the lyrics event deletion process. FIG. 10A shows an example of note plus data. The operator can use the mouse or the like to specify the range RG of the lyrics event to be deleted. For example, as the range RG, the lyrics event L
3 and L4 can be specified.

First, in the clipboard of FIG. 10 (B),
Copy the lyrics events L3 and L4 within the range RG.
Although the contents of this clipboard are not used in the subsequent processing, they can be used as the clipboard of FIG. 8B if necessary.

Next, the range R is stored in the buffer of FIG.
Copy the lyrics events L5 to L8 behind G.

Next, as shown in FIG.
The lyrics events L3 to L8 after the range RG in (A) are deleted.

Next, as shown in FIG.
In the note events N3 to N8 after the range RG of (D),
Lyrics events L5 to L8 in the buffer (FIG. 10C)
Assign However, since the lyrics events L3 and L4 are deleted, the last note events N7 and N8 are assigned blank line lyrics events. In the note plus data of FIG. 10 (E), since the lyrics events L3 and N4 are deleted, the lyrics event L5 follows the lyrics event L2.

FIG. 11 is a flow chart for realizing the above-mentioned lyrics event deletion processing.

At step SF1, a lyrics event of an arbitrary number of lines RG to be deleted is designated (FIG. 10 (A)) and copied to the clipboard (FIG. 10 (B)).

At step SF2, the lyrics event after the deleted row RG is copied to the buffer (FIG. 10C).

At step SF3, the lyrics data on and after the line RG to be deleted is erased (FIG. 10 (D)).

At step SF4, the buffer (see FIG.
The lyrics event of (C)) is assigned to the note plus data on and after the deleted row RG (FIG. 10 (E)). This is the end of the lyrics event deletion process.

The lyric event deleting process is effective when the number of note events is reduced by arranging performance data, or when the number of lyric events is increased due to an erroneous input of lyric data.

The method of deleting the desired lyric event from the note plus data of FIG. 10 (A) to generate the note plus data of FIG. 10 (E) may be a method other than the above.

FIGS. 12A to 12D are diagrams for explaining the first lyrics automatic allocation processing.

FIG. 12A shows an example of note plus data. The operator can use the mouse or the like to specify the range RG of the note plus event to be automatically assigned. For example, the second to sixth note plus events can be designated as the range RG.

First, the range R is stored in the buffer of FIG.
Copy the lyrics event "Iou Okakiku" in G.

Next, as shown in FIG.
The lyrics event in the range RG of (A) is deleted.

Next, as shown in FIG.
To the note events N2 to N6 in the range RG of (D), the lyric event “Iouokakiku” in the buffer (FIG. 12B) is assigned character by character. However, the remaining note string N6 is assigned to the last note event N6. In the note plus data of FIG. 12D, the range RG
One character is assigned to each of the note events N2 to N5, and three characters “Kakiku” are assigned to the note event N6. Basically, one character is assigned to one note (syllable).

The first lyrics automatic allocation processing is performed in the range RG.
Within this number of lines, that is, without increasing or decreasing the number of lines, a lyrics event is automatically assigned to each line.

FIG. 13 is a flow chart for realizing the above-mentioned first lyrics automatic allocation processing.

In step SG1, a plurality of lines RG to which lyrics are automatically assigned are selected (FIG. 12 (A)).

At step SG2, the selected plural rows R
Buffer the lyrics event included in G (Fig. 12 (B))
To copy. However, when the lyrics event is composed of a phoneme string, the phoneme string in the buffer is converted into character data.

In step SG3, the selected plural rows R
The lyrics event in G is deleted (FIG. 12 (C)).

At step SG4, the buffer (FIG.
The character string data in (B)) is assigned to the note plus data of the selected plural rows RG (FIG. 12D). This is the end of the first lyrics automatic allocation processing.

14A and 14B are views for explaining the blank line automatic insertion processing. FIG. 14A shows an example of note plus data. The operator can select a row RG in which a blank row is to be inserted by using a mouse or the like. For example, the note plus event in the fourth row can be designated as the selected row RG.

First, the number of characters forming the lyrics event in the selected row RG is counted. For example, the number of characters in the lyrics event "Okakiku" is four.

Next, as shown in FIG. 14B, a blank line of "the number of characters-1" is inserted at the position of the selected line RG. For example, 4-1 = 3 blank lines are inserted. A blank line is assigned to the note events N4 to N6 after the selected line RG, and a lyric event "Okakiku" is assigned to the note event N7.

The blank line may be inserted after the lyrics event "Okakiku" without being limited to the case where the blank line is inserted before the lyrics event "Okakiku" of the selected line RG.

After that, the fourth to seventh rows are selected and the lines shown in FIG.
By performing the first lyrics automatic allocation processing shown in 2, it is possible to allocate the lyrics event to the blank line. Specifically, "N4, oh", "N5, ka", "N6, ki",
It can be assigned like "N7, Ku".

FIG. 15 is a flow chart for realizing the above blank line automatic insertion processing.

In step SH1, a row RG in which a blank row is automatically inserted is selected (FIG. 14 (A)).

At step SH2, the number of characters of the lyrics event in the selected row RG is detected. When the lyric event is composed of a phoneme string, the phoneme string is converted into characters and then the number of characters is detected.

In step SH3, blank lines having the number of characters "-1" are inserted at the position of the selected line RG. This insertion can be performed using a buffer as in the blank line insertion described above. This is the end of the blank line automatic insertion process.

FIGS. 16A to 16D are diagrams for explaining the second lyrics automatic allocation processing.

FIG. 16A shows an example of note plus data. The operator can use the mouse or the like to specify the range RG of the note plus event to be automatically assigned. For example, the second to sixth note plus events can be designated as the range RG.

First, the range R is stored in the buffer of FIG.
Copy the lyrics event "Iou Okakiku" in G.

Next, as shown in FIG. 16C, since the number of characters (7 characters) in the buffer is two characters more than the number of lines (5 lines) in the range RG, two blank lines are added to the range RG. Insert in position.

Next, as shown in FIG. 16 (D), the number of note events N2 to N8 from the position of the selected row RG to the number of characters (7 characters) in the buffer (FIG. 16 (B)) are changed in the buffer. Allocate the character string “Iue Okakiku” one by one. A one-character lyrics event is assigned to each of the note events N2 to N8.

The second lyrics automatic allocation processing is processing for ensuring that all the characters included in the selected line RG are allocated one character per one note event.

FIG. 17 is a flow chart for realizing the above-mentioned second lyrics automatic allocation processing.

At step SI1, a note plus event of a plurality of lines RG to which lyrics are automatically assigned is selected (FIG. 16 (A)).

At step SI2, the selected plural rows R
The lyrics event in G is copied to the buffer (FIG. 16 (B)). However, when the lyrics event is composed of a phoneme string, the phoneme string in the buffer is converted into character data.

At step SI3, the number of characters in the buffer is detected. In step SI4, the number of selected rows RG
And the number of characters detected.

At step SI5, it is checked whether the compared numbers of rows are equal, large, or small. When the number of lines is large, the process proceeds to step SI6, the extra number of lines is deleted, and the process proceeds to step SI8. When the number of lines is small, step S
Proceed to I7, insert the missing number of lines (Fig. 16 (C)),
Go to step SI8. When the number of lines is the same, the process proceeds to step SI8 without changing the number of lines.

In step SI8, the character string data in the buffer is deleted from the first row of the selected row RG by the number of selected rows RG, the number of deleted rows, or the number of inserted rows.
It is assigned to a note plus event (FIG. 16 (D)).
This completes the second lyrics automatic allocation processing.

FIGS. 18A to 18D are views for explaining the plural line merging process. FIG. 18A shows an example of note plus data. The note plus data is the note plus event "N1, a", "N2, i", "N3.
U ”,“ N4, E ”, and“ N5, O ”.

The operator can specify the range RG of note plus events to be merged by using a mouse or the like. For example, the second and third note plus events can be designated as the range RG.

First, the range R is stored in the buffer of FIG.
Copy the lyrics event "say" in G.

Next, as shown in FIG. 18C, the range R
Only the lyrics event "i" in the first line in G is left, and the lyrics event "u" in the remaining lines is deleted. The lyric event "E, O" behind the range RG is packed in the previous line. The note plus data is the note plus event "N1,
"A", "N2, I", "N3, E", "N4, O",
It has “N5, _”.

Next, as shown in FIG. 18D, the range R
The character string "say" in the buffer is assigned to the note event N2 in the first row in G. Second note event N2
, The lyrics event “say” within the range RG is merged and assigned.

In the above-mentioned first or second lyrics automatic allocation processing, one note event is allocated to one note event. However, by using this plural line merging processing, one note event is allocated. It can be modified to assign a lyric event of two or more characters.

FIG. 19 is a flow chart for realizing the above-mentioned plural line merging process. In step SJ1,
Select multiple row RGs that merge lyrics events (Fig. 18
(A)).

At step SJ2, the selected plural rows R
The lyrics event in G is copied to the buffer (FIG. 18 (B)). However, when the lyrics event is composed of a phoneme string, the phoneme string in the buffer is converted into character data.

At step SJ3, the selected plural rows R
Of G, one line is left and the lyrics events of the remaining lines are deleted (FIG. 18C). The lyrics event after being deleted is packed in the previous line.

In step SJ4, the character string data in the buffer is assigned to the note event in the note plus data of the remaining lines in the range RG (FIG. 18 (D)). This completes the multi-line merging process.

20 (A) and 20 (B) are diagrams for explaining the lyrics event division processing. FIG. 20A shows an example of note plus data. Note Plus data is
Note plus event "N1, a", "N2, say",
Has "N3, E".

The operator uses the mouse or the like to select the line RG in which the lyrics event is to be divided, and sets the cursor CS to the character position in the line where the division is desired. For example,
As the range RG, the second note plus event “N2,
"I" is selected and the cursor CS is set between "I" and "U". After that, the operator presses the return key (execute key)
The following division processing is performed by operating.

As shown in FIG. 20B, the characters "i" and "u" are divided. The note plus data is the note plus event "N1, a", "N2, i", "N3.
Have "up". The character "U" behind the cursor CS is
It is merged with the lyrics event "E" in the line after the selected line RG.

This is convenient when a large number of lyrics events in blank lines are inserted (FIG. 6) and lyrics are sequentially assigned from the beginning.

FIG. 21 is a flow chart for realizing the above-mentioned lyrics event division processing.

At step SK1, the row RG including the lyrics event to be divided is selected (FIG. 20 (A)).

At step SK2, the cursor CS is set at the division position by operating the mouse or the like (FIG. 20 (A)).

At step SK3, the lyrics event after the division position is added to the lyrics event in the note plus data of the next line in response to the operation of the return key (FIG. 2).
0 (B)). This is the end of the lyrics event division process.

22A to 22C are diagrams for explaining the kana conversion processing. FIG. 22A shows an example of note plus data. The note plus data is the note plus event "N1, red", "N2, i", "N3.
"Evening" and "N4, day".

The operator uses the mouse to move the mouse pointer 71 to the position of the lyric event "evening" for which kana conversion is desired.
Move and click the right mouse button.

As shown in FIG. 22B, the phonetic candidate for the lyrics event "evening" is displayed on the menu 72. In the menu 72, for example, "Yu" and "cough" are displayed. The operator moves the mouse pointer 71 to the position of the proper reading kana character string "Yu" and clicks the left button of the mouse.

As shown in FIG. 22C, the lyric event assigned to the note event N3 is converted from the Chinese character "evening" to the kana "Yu." In addition to kanji, alphanumeric characters and symbols can be converted into kana. If kanji or alphanumeric characters are converted to kana, then the kana can be converted to a phoneme string (FIG. 5), so that the phoneme string can be pronounced.

FIG. 23 is a flow chart for realizing the above-mentioned kana conversion processing. In step SL1, the line of characters to be converted into kana is designated by the mouse pointer 71 (FIG. 22 (A)).

In step SL2, when the right mouse button is clicked, the phonetic alphabet corresponding to the selected character is retrieved from the database and displayed as a phonetic kana candidate (FIG. 22B). The database is a dictionary for converting Kanji or the like into kana, and is an external storage device 13 or RAM.
3 etc. are stored.

At step SL3, one of the reading kana candidates is designated by the mouse pointer 71 (FIG. 22).
(B)).

In step SL4, when the left mouse button is clicked, the designated phonetic kana candidate is assigned to the note event in the note plus data as the character string of the designated line (FIG. 22 (C)). This is the end of the kana conversion processing.

By supplying the note-plus event (character string converted into a phoneme string by the phoneme string conversion processing of FIG. 5) created as described above to the external sound source device 21, the external sound source device 21 Voice synthesis circuit 32
In, a singing voice signal can be generated. The audio data processing device (personal computer 1) described in the present embodiment is a so-called dedicated audio data processing device capable of reproducing the note plus data in the same format. This voice data processor is a note event and a lyrics event (phoneme sequence).
Can be sent in the proper order.

Now, an appropriate sequence for transmitting the note event and the lyrics event will be described. Figure 24
(A), (B) is a figure for demonstrating the transmission order of a note event and a lyrics event. The note plus event is, for example, a set of a note event N1 and a lyrics event L1 (FIG. 6 (A)). The note event N1 and the lyrics event L1 should theoretically be transmitted at the same timing, but since MIDI communication is serial communication, there is a problem which one should be transmitted to the external sound source device 21 (FIG. 1) first. . The note event N1 is converted into a note-on event and transmitted. The lyrics event is converted into a phoneme string (FIG. 5) and transmitted.

FIG. 24A shows the timing of transmission in the order of phoneme sequence and note-on. First, the phoneme sequence “aj” corresponding to the syllable “a” is transmitted. After that, when a note-on is sent, the pronunciation of "a" starts at that timing. Next, the phoneme string “ij” corresponding to the syllable “I” is transmitted. After that, when a note-on is sent, the pronunciation is switched from "a" to "i" at that timing. Next, the phoneme string "uj" corresponding to the syllable "u" is transmitted. When a note-on is sent after that, the pronunciation is switched from "i" to "u" at that timing.

In FIG. 24A, note-on (note event) and phoneme string (lyric event) are associated with each other, and an appropriate song can be sung.

FIG. 24B shows the timing of transmission in the order of note-on and phoneme sequence. First, note-on is transmitted. At this point in time, since the phoneme string is not set, the initially set phoneme string such as "A" starts to sound. After that, the phoneme sequence "a" corresponding to the syllable "a"
j ”is transmitted. At this point, there is no change in pronunciation, and the initially set pronunciation of "A" is maintained. Next, when the note-on is transmitted, the pronunciation of the phoneme string “A” which is initially set at that timing is switched to the pronunciation of the transmitted phoneme string “A”. After that, the phoneme string "ij" corresponding to the syllable "i" is transmitted, but the pronunciation is not affected. Next, when a note-on is transmitted, the pronunciation is switched from "a" to "i" at that timing. After that, the phoneme string "uj" corresponding to the syllable "u" is transmitted, but the pronunciation is not affected.

In FIG. 24B, the note-on (note event) and the phoneme sequence (lyric event) are not associated with each other, and the lyrics are pronounced with a delay of one note. With this, the proper song cannot be sung.

As shown in FIG. 24A, it is necessary to first transmit the phoneme string and then the corresponding note-on to the external tone generator 21. As described above, the voice data processing device according to the present embodiment is a device that can correctly handle note-plus data, and transmits the phoneme sequence first and then the note-on for the one note-plus event. It is supposed to do. However, note-plus events cannot be handled correctly in a sequencer or the like other than this voice data processing device. Therefore, the audio data processing device has a function of converting the note plus data into a more general-purpose standard MIDI file. At this time, if the standard MIDI file is simply converted, the inconvenience described below may occur. Therefore, the conversion is performed in consideration of the transmission order as described above. Hereinafter, a method of converting the note plus data into a standard MIDI file will be described in consideration of the transmission order of the note event and the lyrics event (phoneme string).

FIG. 25 is a diagram for explaining a method of converting the note plus data 75 into a standard MIDI file 76.

The note plus data 75 includes note event N1, lyrics event L1, duration T2, and
It has a note event N2, a lyrics event L2, and a duration T3.

The note event N1 has a note-on event NON1 and a gate time GT1. The note-on event NON1 includes, for example, a note number (pitch) and velocity (volume). The gate time GT1 is the time from note-on to note-off, and is, for example, 450
Is. The lyrics event L1 is composed of a phoneme string. However, the exhalation symbol and the Chinese character are included in the lyrics event L1 as a character string.

The note event N2 is a note event N.
Similar to 1, it has a note-on event NON2 and a gate time GT2. The gate time GT2 is, for example, 22
It is 0. Like the lyrics event L1, the lyrics event L2 is composed of a phoneme string.

The duration T2 is the note-on event NON2 from the start of sounding of the note-on event NON1.
Is the time until the start of the sound generation of, for example, 480. The duration T3 is the time from the start of sounding the note-on event NON2 to the start of sounding the next note-on event, and is, for example, 240. The gate time and the duration value are represented by the number of clocks. One clock is, for example, 1/480 of a quarter note length.

The standard MIDI file 76 is M
It is a file in a general format suitable for the IDI standard. The standard MIDI file 76 is composed of a set of duration and event.

The standard MIDI file 76 includes, in order, the lyrics event L1, the duration TT1, the note-on event NON1, the duration TT2, the note-off event NOFF1, the duration TT3, the lyrics event L2, the duration TT4, the note-on event NON2, the duration TT5, and the note. It has an off event NOFF2 and a duration TT6.

The lyrics event L1 and the note-on event NON1 are the same as the lyrics event L1 and the note-on event NON1 in the note plus data 75,
The order is different. By transmitting the lyrics event L1 before the note-on event NON1,
A normal song can be sung like (A).

The duration TT1 is the lyrics event L
It is the time from the transmission of 1 to the transmission of the note-on event NON1, which is, for example, 5. If the lyric event L1 is located at an address before the note-on event NON1, it can be considered that the duration TT1 may be 0 theoretically.

However, if the duration TT1 is set to 0, the transmission order may differ depending on the personal computer 1 or sequencer that transmits the standard MIDI file. That is, the duration TT1 is set to 0
If set to, lyrics event L1, note-on event NO
The note-on event NON1 and the lyrics event L1 may be transmitted in this order, not necessarily in the order of N1. Further, a similar phenomenon occurs when the duration TT1 is not 0 but 1 or 2.

This is because the sequencer or the like has performed the lyrics event L1.
It is considered that the note-on event NON1 and the note-on event NON1 are compared, the note-on event NON1 is judged to be high in importance, and the note-on event NON1 is transmitted before the lyrics event L1. The lyrics event can be transmitted by a system exclusive message determined by the MIDI standard.

In order to prevent the above adverse effects, the duration TT1 is set to 5. The duration TT1 is preferably 3 or more, but too large a value may adversely affect the previous event. Duration TT1 is 3
-10 is preferable. However, this is the case where the note resolution of the clock is 1/480 of the quarter note length, and when the note resolution is different, the above preferable numerical value takes another value. For example, if the note resolution of the clock is 1/96 of a quarter note, about 1 to 3 is preferable.

Duration TT2 is note-on NO
It is the time from N1 to note-off NOFF1 and corresponds to the gate time GT1 of the note-plus data 75, which is 450, for example.

The duration T2 in the note plus data 75 is from the note-on NON1 to the next note-on NO.
It is the time until N2, and is decomposed into the durations TT2, TT3, and TT4 in the standard MIDI file 76.

The duration TT2 is, as described above,
It is 450, which is the same as the gate time GT1 of the note plus data 75. The duration TT4 is the time from the transmission of the lyrics event L2 to the transmission of the note-on event NON2, and is 5 like the duration TT1. Duration TT3 is the note-off event NO
It is the time from the transmission of FF1 to the transmission of the lyrics event L2, and is represented by TT3 = T2-TT2-TT4. That is, TT3 = 480−450−5 = 25.

As described above, the lyrics event L1 is located at an address before the note-on event NON1,
Moreover, by setting the duration TT1 to 5, it is possible to reliably transmit the lyrics event L1 before the note-on event NON1. At that time, since the timing of the lyrics event L1 is advanced by 5 without changing the timing of the note-on event NON1, the note-on NO
The sound generation timing by N1 does not deviate.

Also, by converting the note plus data 75 into the standard MIDI file 76, the versatility is increased, and it becomes possible to process it with other sequencers and the standard MIDI file 76 is supplied to the user by a floppy disk or the like. It becomes possible to do.

As shown in FIG. 26A, the standard MIDI file (SMF) conversion means 81 can convert the note plus data 75 into SMF data 76. The personal computer 1 has an SMF conversion means 8
In 1, the note plus data 75 is standard M
The IDI file 76 can be converted and saved in the external storage device 13. Standard MID saved
The I file 76 can be used in other personal computers, sequencers and the like. Further, when the personal computer 1 sends the standard MIDI file 76 to the external tone generator device 21, the external tone generator device 21 performs sound generation processing according to the standard MIDI file 76.

Further, as shown in FIG. 26B, the note plus conversion means 82 can convert the SMF data 76 into the note plus data 75. The personal computer 1 can load the standard MIDI file 76 stored in the external storage device 13, convert it into the note plus data 75 in the note plus converting means 82, and store it in the RAM 3. At this time, contrary to the conversion of the note plus data 75 into the standard MIDI file 76, the lyrics event stored independently of the note event is processed into one note plus event. First, search the data in the standard MIDI file sequentially from the beginning,
When the note-on event is found, the corresponding note-off event is searched for and one note-on event including the gate time is generated. Then, the lyrics event existing up to 5 clocks before the note-on event is searched, and if found, the lyrics event is added to the note-on event and the note-on event and the lyrics event are paired. Generate an event. The operator can perform various edits on the note plus data 75, as described above. It should be noted that the same process as the process of converting the Note Plus data 75 into a standard MIDI file may be applied to the case of converting into a relatively versatile data format other than the standard MIDI file.

The case is not limited to the case where the voice synthesis circuit in the external tone generator 21 is caused to generate a phoneme string. A sound source board including a voice synthesis circuit may be inserted into the personal computer 1 to cause the sound source board to generate a phoneme string. In that case, the external sound source device 2 is connected to the personal computer 1.
There is no need to connect one.

The audio data processing device according to this embodiment is not limited to the form of a personal computer and application software, but may be of an electronic musical instrument or a sequencer. The application software may be stored in a storage medium such as a magnetic disk, an optical disk, or a semiconductor memory and supplied to a personal computer, or may be supplied via a network.

The audio data format is 1 at the time of occurrence of a performance event like a standard MIDI file.
Not limited to "event + relative time" expressed as the time from the previous event, "event + absolute time" expressed as the time when a performance event occurred in absolute time within a song or measure,
"Pitch (rest) + note length", which represents performance data by the pitch and note length of a note or rest and rest length, reserves a memory area for each minimum resolution of the performance, and a performance event occurs. A format such as a "solid method" in which performance events are stored in a memory area corresponding to time may be used.

The audio data may have a format in which data of a plurality of channels are mixed, or may have a format in which the data of each channel is divided for each track.

The present invention has been described above with reference to the embodiments.
The present invention is not limited to these. For example, it will be apparent to those skilled in the art that various modifications, improvements, combinations, and the like can be made.

[0224]

As described above, according to the present invention,
Since text data including kana or kanji can be processed, inputting or editing of text data is easy. Existing text data in which kana or kanji are mixed can be used.

[0225]

[0226] Also, the voice data may be converted to the standard M, for example.
Since it can be handled in the IDI file format, versatility is increased and MIDI communication can be performed.

[Brief description of drawings]

FIG. 1 is a diagram showing a hardware configuration of a system in which a personal computer and an external tone generator are connected.

FIG. 2 is a diagram showing a process of assigning a text event to a note event.

FIG. 3 is a flowchart showing a text import process.

FIG. 4 is a flowchart showing details of phonetic kana and ruby processing shown in step SA9 of FIG.

FIG. 5 is a flowchart showing a phoneme string conversion process.

FIG. 6 is a diagram for explaining blank line insertion processing.

FIG. 7 is a flowchart showing blank line insertion processing.

FIG. 8 is a diagram for explaining lyrics event insertion processing.

FIG. 9 is a flowchart showing a lyrics event insertion process.

FIG. 10 is a diagram illustrating a lyrics event deletion process.

FIG. 11 is a flowchart showing a lyrics event deletion process.

FIG. 12 is a diagram for explaining a first lyrics automatic allocation process.

FIG. 13 is a flowchart showing a first lyrics automatic allocation process.

FIG. 14 is a diagram for explaining blank line automatic insertion processing.

FIG. 15 is a flowchart showing a blank line automatic insertion process.

FIG. 16 is a diagram for explaining a second lyrics automatic allocation process.

FIG. 17 is a flowchart showing a second lyrics automatic allocation process.

FIG. 18 is a diagram for explaining a process of merging a plurality of lines.

FIG. 19 is a flowchart showing a process of merging a plurality of lines.

FIG. 20 is a diagram for explaining lyrics event division processing.

FIG. 21 is a flowchart showing lyrics event division processing.

FIG. 22 is a diagram for explaining a kana conversion process.

FIG. 23 is a flowchart showing a kana conversion process.

FIG. 24 (A) is a diagram showing pronunciation when phonemes are transmitted in the order of note-on, and FIG. 24 (B) is a diagram showing pronunciation when notes are transmitted in the order of phoneme sequence. .

FIG. 25: NotePlus data and standard MID
It is a figure which shows conversion with an I file.

FIG. 26 (A) shows a standard MIDI file converting means, and FIG. 26 (B) shows a note plus converting means.

[Explanation of symbols]

1 personal computer, 2 buses, 3
RAM, 4 ROM, 5 CPU, 6 timer, 7 display circuit, 8 MIDI interface, 9, 11 detection circuit, 10 mouse,
12 keyboard, 13 external storage device,
14 communication interface, 21 external sound source device,
22 bus, 23 MIDI interface, 24 RAM, 25 ROM, 26 CP
U, 28 display circuit, 29 musical tone waveform synthesizing circuit,
31 sound system, 32 speech synthesis circuit,
33 detection circuit, 34 switch, 41
Communication network, 42 server computer,
51 performance data, 52 text data,
53 note plus data, 75 note plus data, 76 standard MIDI files, 81
Standard MIDI file conversion means, 82
Note plus conversion means

─────────────────────────────────────────────────── ─── Continuation of the front page (56) Reference JP-A-9-50287 (JP, A) JP-A-9-44179 (JP, A) JP-A-7-146695 (JP, A) JP-A-4- 331990 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) G10H 1/00-7/00 G10L 13 / 00,13 / 04

Claims (2)

(57) [Claims] 1. Storage means for storing text data containing non-kana and its reading kana or ruby paired data and note data containing note-unit data, and destroying non-kana in the text data. And a ruby kana or ruby only for the note data in the note data to generate voice data. 2. A recording medium of a computer program for processing text data including data in which a non-kana and its reading kana or ruby are paired and note data including data in units of notes, wherein the non-character in the text data is A medium recording a program for causing a computer to execute a procedure of discarding a kana and assigning only the reading kana or ruby to note data in the note data to generate voice data.