JP2003050591A - Musical performance output device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize transition to a musical sound synthesis means at a low cost while effectively using an existing musical sound reproduction means. SOLUTION: This musical performance output device comprises a musical sound reproduction means 10 for storing audio signals of a program and reproducing and outputting audio signals of specified pieces on the program, musical sound synthesis means 22, 23 for storing event information on the program and synthesizing and outputting musical sound at sound sources according to the event information on the specified pieces on the program, and a commander part 21 for accepting an instruction for selecting a program. Based on the instruction for selecting the program, the commander part 21 selectively controls whether to reproduce the program by the musical sound reproduction means 10 or to synthesize the musical sound of the program by the musical sound synthesis means 22, 23.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a performance output device including a musical sound reproducing means and a musical sound synthesizing means.

[0002]

2. Description of the Related Art A karaoke device having a function of reproducing a designated musical piece and mixing and outputting voices inputted from a microphone is not limited to restaurants such as snacks, but also karaoke boxes and buses. It is also popular in mobile units. An ordinary karaoke device has a storage system that stores performance information and image information, an acoustic system that reproduces a performance song and mixes the voice of a singer with it, and a video image that displays a background image and lyrics image according to the performance song. There are many types of musical sound reproduction types that include a system and a control system that controls each of these systems. On the other hand, recently, a musical tone synthesizing type karaoke apparatus has appeared in which necessary music data is taken in from a host station via a communication line. This musical sound synthesizing type karaoke apparatus has a built-in sound source for generating musical sounds based on music data.

The musical tone reproducing type karaoke device cannot respond to requests other than the musical composition stored in the storage system, but if the musical tone synthesizing type karaoke device is used.
Since it is possible to have the song data supplied in response to the singer's request from many song data stored in the host station, in an extreme example, even if there is no song data on the karaoke device side, It will be possible to meet all requests from the singer. The complete tone synthesizing karaoke device has such a configuration, but in order to reduce the communication fee and communication time for accessing the host station each time a request is made, a storage device is provided on the karaoke device side, and it is imported from the host station here There is also a semi-tone synthesis type that stores song data and reuses it.

[0004]

In any case, when a new musical tone synthesizing karaoke device is introduced,
Completely stopping the use of the existing musical tone reproduction type karaoke apparatus is economically disadvantageous for both the side for supplying the musical tone synthesizing type and the side for installing the musical tone synthesizing type. The present invention is
It is an object of the present invention to provide a performance output device which improves such points and realizes the transition to the musical sound synthesizing means at low cost while effectively utilizing the existing musical sound reproducing means.

[0005]

In order to achieve the above object, according to the present invention, a musical tone reproducing means for storing an audio signal of a musical piece to be played and reproducing and outputting an audio signal of the designated musical piece, and event information of the musical piece. And a commander unit for accepting a music selection instruction of the performance music, and a commander unit for accepting a music selection instruction of the performance music, and the commander unit also receives the accepted music selection instruction. In addition, it is characterized in that selection control is performed as to whether the musical tone reproducing means reproduces the musical piece to be played or the musical tone synthesizing means synthesizes the musical sound of the musical piece to be played.

In one embodiment of the present invention, in the case where the performance sound corresponding to the received music selection instruction is stored in the musical sound reproducing means, the commander section causes the musical sound reproducing means to play the musical composition. When the musical composition corresponding to the received music selection instruction is not stored in the musical sound reproducing means and is stored in the musical sound synthesizing means, the musical sound synthesizing means It is the one for synthesizing.

[0007]

According to the present invention, the musical sound reproducing means and the musical sound synthesizing means are used together, and either one is selectively controlled. This enables control to preferentially use the musical sound reproduction means when designating a music piece. The musical tone synthesizing means is used only when playing a piece of music which the musical tone reproducing means does not have.
Also at this time, if there is a musical tone synthesis type storage device for storing music data, the data in the storage device is preferentially used, access to the host station is avoided as much as possible, and communication cost and communication time are reduced.

[0008]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is an external view of the entire integrated karaoke system according to the present invention. In this figure, an existing tone reproduction karaoke apparatus 10 includes an LD changer system 11, a monitor (TV) 12, a mixing amplifier 13, a microphone 14,
The speaker 15 is included. The karaoke control device 20 of the present example is a tone synthesis synthesizing type having a built-in sound source unit, and is combined with the monitor 12 and the mixing amplifier 13 to form a tone synthesis karaoke device.

This karaoke control device 20 is called an integrated system commander, and it is a musical sound reproduction type karaoke device 1
By partially connecting to 0, it becomes possible to integrate and control all systems including tone synthesis / tone reproduction. Reference numeral 30 is a host station for supplying the requested song data to the karaoke controller 20 through a communication line 40 (for example, a high-speed ISDN line). Reference numeral 50 denotes a remote controller (remote controller) for giving instructions such as designation of performance music to the system.

The musical tone reproduction type karaoke apparatus 10 is a self-contained system. In the LD changer system 11, a plurality of LD (s) are stored as a storage medium storing a large number of audio signals of a musical composition and their background images and lyrics images. A laser disk) is housed therein, and the LD outputs the audio output AUDIO of the designated music to the mixing amplifier 13 and the video output VIDEO to the monitor 12. The mixing amplifier 13 mixes the singing voice of the singer input from the microphone 14 with the musical piece to be output, and the speaker 15 outputs the mixing signal in a loud voice. At this time, the monitor 12 stores the image information stored in correspondence with the music,
That is, the background image and the lyrics image are displayed to guide the singer's singing.

FIG. 2 is a block diagram showing the internal structure of the karaoke controller 20 in detail. The karaoke control device 20 is roughly classified into a commander unit 21 for tone synthesis / tone playback, a file system unit 22 for tone synthesis, and a performance / display unit 23 having various functions.

The commander section 21 is a key for receiving a command signal from a remote control receiver 21A for receiving a command signal from the remote controller 50 and a key input section (24 in FIG. 1) laid on the front panel of the karaoke control apparatus 20. According to the input from the input interface 21B, the remote control receiver 21A or the key input interface 21B, the music selection / reservation controller 21C for controlling the performance of the music, the display 21D for displaying the content of the music selection / reservation, and the display 21D are driven. A display driver 21E, an AV input / output selector 21F for selecting audio (A) and visual (V) inputs and outputs, and a control output interface 21G for controlling the LD changer 11A in the LD changer system 11. .

The file system unit 22 selects music
A file controller 22A that receives a performance instruction of a selected song from the reservation controller 21C and an ISDN line 4
An ISDN controller 22B (communication control device) that communicates with the host station 30 via 0 and a storage device 22C that stores the music data taken in through the ISDN line 40. When the file controller 22A receives a song selection / reservation instruction from the song selection / reservation controller 21C, it first searches the storage device 22C to determine whether or not the song data of the corresponding song is stored. If it is stored, the corresponding song data is read from the storage device 22C, and if it is not stored, the ISDN controller 22 is read.
Control B and have the host station transfer the corresponding song data. As the storage device 22C, for example, 100 KB HDD (hard disk) capable of storing song data for 1000 songs, with 10 KB per song (B is byte).
To use.

The performance / display unit 23 includes a data buffer 23A for temporarily holding music data from the file controller 22A, and a multimedia sequencer for controlling various events such as musical sounds, images and effects in accordance with event information contained in the music data. (MMS) 23B, a tone generator 23C that forms a musical tone of a performance song under the control of the sequencer 23B, a video processor 23D that generates a background image and lyrics image corresponding to the performance song, and a video processor 23D. Video output to LD changer 11
And a video superimposer 23E which is superimposed on an image (for example, a background image of a moving image) read from the background image LD (BGV-LD) of A.

AV input / output selector 21 of commander section 21
Regarding the audio, F selects either the audio output from the LD changer 11A or the musical sound synthesized by the sound source processor 23C and outputs it as the audio output to the mixing amplifier 13. For images, LD
The image from the changer 11A and the image from the video superimposer 23E are selected to be video output to the monitor 12. If the AV input / output selector 21F and the LD changer 11A are not connected, the karaoke control device 20 is systematically independent of the musical sound reproduction type karaoke device 10. It is the AV input / output selector 21F and the control output interface 21G that put the two in a coupled relationship.

In the integrated karaoke system having the connection relationship as shown in FIGS. 1 and 2, the commander section 21 which receives a song selection instruction from the remote controller 50 or the like first sets the corresponding song to L
It is confirmed whether it is stored in the D changer 11A. That is, the musical tone reproduction type karaoke apparatus 10 is given the first priority and its sources (audio and video) are effectively used. If the LD changer 11A has information on the corresponding song, the karaoke performance is performed using the audio and video outputs.

On the other hand, when the LD changer 11A does not have the data of the corresponding music, the file controller 2
The music number is transferred to 2A, and the music data on the music synthesis side is used. Also in this case, the host station 30 is not accessed immediately, but it is first confirmed whether or not the music data of the corresponding music is stored in the storage device 22C. Here, if the data of the corresponding song is stored in the storage device 22C, it is read and used. That is, the storage device 22
The second priority is given to the song data read from C. When the storage device 22C does not have the music data of the corresponding music, the host station 30 transfers the music data of the corresponding music through the ISDN line. That is, the acquisition of the music piece data from this host station is set as the third priority. In the complete tone synthesis type without the memory device 22C, data acquisition from the host station has the second priority.

FIG. 3 is a conceptual diagram showing the relationship of main parts of the system components shown in FIGS. 1 and 2.
That is, the microcomputer board 25 included in the karaoke control device 20 according to the present invention has a function capable of controlling the musical tone synthesis system and the musical tone reproduction system. This microcomputer board 25 is an HDD for responding to the music number designation from the remote controller 50.
Refer to the index data stored in, to use the tone synthesis system, that is, use the song data from the communication line 40 (the sequence sound source data can be used when the CD-ROM 41 is equipped), or the tone reproduction system, That is, various source LDs mounted on the autochanger 11
Whether or not to use the data contained in the CD / CDG / CDI / CD moving image is determined, and the control is performed. In this embodiment, storage device 22C and video processor 23
D is a standard equipment of a musical sound synthesizing system, and other control parts are integrated in the microcomputer board 25. Since this microcomputer board 25 is compatible with any model, it is connected to the main body by a mechanism that can be replaced. Of course, the main part of the microcomputer board is also standard equipment, and a program ROM that controls a specific model. Only the part (not shown) may be replaceable.

FIG. 4 is a system configuration diagram for explaining the functions of the integrated karaoke system according to the present invention in more detail. The parts not shown in FIGS. 1 to 3 will be described below. A CPU (central processing unit) 61 controls and manages the operation of the entire karaoke system under the program ROM 23B. 62 is the CPU 6
1. Random access memory (RAM) used when controlling and managing the operation of this karaoke system
Is. 63 is a bus (data, address) for connecting the entire system. 64 is MIDI to the outside
(Musical Instrument Digital Interface) An interface for connecting the musical instrument 42. 65 is an LD
This is a changer controller for controlling the changer 11, and has a replaceable configuration so as to be compatible with a plurality of models.

66 is a multimedia sequencer 23B
It is an ADPCM processor that expands ADPCM data output from the device by bit conversion and frequency conversion. 6
A pitch shifter 7 controls the pitch of the output of the ADPCM processor 66 according to the pitch information. Reference numeral 68 is an effector mixer that receives the output of the pitch shifter 67 and the output of the sound source processor 23C. 69 is the microphone 1
4 is a microphone effector that adds acoustic effects such as echo and exciter to the output of FIG. 70 is a microphone effector 69
2 is an audio DSP (sound field generator) that gives a sound field effect to the output of the above and the audio output of the LD changer 11.

Reference numeral 71 denotes an ISDN for the storage device 22C.
It is a portable downloader used when loading a large amount of song data without using the line 40. For example, it should be brought in and loaded by the installer at the initial stage when the tone synthesis system of this system was installed. Thus, the installer side does not have to communicate with the host station 30 through the ISDN line 40 for a long time. Other uses of the download machine 71 are possible.

The tone generator processor 23C is a so-called music synthesizer which is driven by song data and synthesizes tone signals of a performance song. For example, the tone generator processor 23C includes a tone generator unit which synthesizes tone sounds based on MIDI signals, and a control unit which controls the tone generator unit. ing. The video processor 23D processes character information such as lyrics corresponding to a performance song and background image information such as a still image and a moving image to generate an image signal for display. Here, an example is shown in which the video processor 23D1 for the lyrics image and the video processor 23D2 for the background image are used. The LD changer 11 functions when playing a karaoke piece recorded on an optical disk, or when playing only video information, and when used as a musical tone synthesis system, in synchronization with the accompaniment performance synthesized by the sound source 23C, Video information including a still image of a predetermined frame of a predetermined disc or a moving image having a predetermined frame as a starting point is reproduced. The video selector 23E combines these three types of images into one screen and outputs the combined image.

The operation will be described below. (01) Musical sound reproduction / musical sound synthesis control By operating the remote control 24 or the front panel 50,
When the musical piece to be played is designated, the CPU 61 causes the storage means 22C.
First, it is checked whether or not there is data of a music piece corresponding to the LD in the LD changer 11 as the first rank by referring to the index table stored in the table, and if there is, that portion is reproduced. The video signal from this LD is the video selector 23E.
To be displayed on the monitor 12, and the audio signal is input to the audio DSP 70 and output from the speaker 15. The singing voice of the singer is converted into an electric signal by the microphone 14, and the audio signal is transmitted through the microphone effector 69.
Input to SP70. Since the mixer 13 mixes the performance music and the singing voice, the speaker 15 produces a mixed acoustic output.

When the designated song is not in the LD changer 11, the CPU 61 searches the song held in the HDD device 22C as the second rank. If there is a corresponding song in the HDD device 22C, the song data of the played song is read out and R is read.
Loaded to AM62 and sound source 23C according to this song data
The musical sound is synthesized with and a karaoke performance is performed. The performance of such a sound source karaoke is performed under the control of the sequencer 23B. With respect to the musical piece to be played, musical tone information is successively generated in the tone generator processor 23C in accordance with the musical piece data read from the RAM 62, and is reproduced from the speaker 15 via the mixer effector 68, the audio DSP 70, and the mixer 13.

Regarding the image, similarly, based on the music data read from the RAM 62, the video processor 23D1 and 23D2 generate the lyrics image and the background image according to the progress of the music in accordance with the instruction of the sequencer 23B.
It is displayed on the monitor 12 via the video selector 23C. When the background image from the LD changer 11 is also used, these three images are combined and displayed by the video selector 23E. The lyrics image displayed on the monitor 12 is instructed so that the singer can easily understand where to sing because the color of the displayed lyrics is changed to a character of another color according to the progress of the singing. It Therefore, the singer is monitor 1
Sing with the microphone 14 while watching the lyrics image displayed in 2.

When the corresponding song data is not stored in the HDD device 22C, the CPU 61 controls to fetch the data of the relevant song online from the host station 30 as the third rank. That is, the ISDN line 40 is used to call the host station 30, and when it responds, the music number is sent and the corresponding music data is transferred. Since the music data may be reused, it is stored in the HDD device 22C.

(02) Transfer data format from the host station The format of the karaoke piece data used in the sound source karaoke apparatus usually starts from the piece header as shown in FIG. Data for one song is constructed by sequentially storing each track such as a lyrics display track, a sound source performance track, and a digital voice track having information. When data is transferred from the host station in this format, the karaoke apparatus cannot start playing unless the last digital voice track is loaded, so it takes a considerable time from the user's designation of the song to the actual start of playing.

For example, the capacity of the music header is 15 KB to 20
Assuming that the total capacity from KB, lyrics display track to digital audio track is 100 KB, communication time of about 15 seconds is required even if a high speed ISDN line capable of transferring 8 KB per second is used. Actually, since the overhead such as the calling time of the host station or the music selection time at the host station is added to this, it takes about 20 seconds in total. This waiting time is 10 times or more as long as the time (1 to 2 seconds) required to read one song from the hard disk.

Therefore, the format of the transfer data is shown in FIG.
Change as shown in (b). That is, in the format of FIG. 5A, the lyrics display track is time-series data A1, A.
2, ... AN, a sound source performance track is a set of time series data B1, B2, ... BN, and a digital audio track is a time series data C1, C2 ,.
In the format of FIG. 2B, the first data A1, B1, C1 of each track is collected as the first track, and the second data A2, B2, C2 is collected as the second track. , And so on, similarly to the Nth data AN, B
Collect N and CN to form the Nth track.

As an example, when N = 6, the transfer of the first track (about 15 KB) is completed in about 2 seconds. Since it takes about 2 seconds to transfer the song header, it takes only 4 seconds to load the first track. Even if other overhead is added to this, it takes about 8 seconds. Moreover, since the karaoke device receiving this first track has obtained all the information of the lyrics display, sound source performance, and digital voice of the song, although it is a part, the karaoke performance is displayed at the event such as lyrics display. Along with this, it can be started promptly.
According to a trial calculation, for example, about 30 seconds of performance can be performed only by the first track obtained by dividing a song having a playing time of 3 minutes into 6 pieces.

Therefore, when the first track is loaded, the first track
Start playing the tracks, and load the second track onwards during that time. Then, when the performance of the first track is completed, a process of simultaneously loading the remaining tracks while performing the second track is performed in parallel. By doing this, you can start playing quickly while online.
Moreover, you can play to the end without interruption.

The data amount of each of the lyrics display track, the sound source performance track, and the digital voice track is small like MIDI. On the other hand, a large amount of data such as back chorus may be added. In order to reduce the amount of data in such a case, in the example shown in FIG.
ADPCM (Adaptive Delta PC) compressed to 4 bits at a sampling rate of 11K or 22KHz
M) Data is added to the last track. This AD
The data size of the PCM track is expected to reach several hundreds of KB, and it takes much longer to transfer all the data. In such a case, as shown in FIG.
If the properly divided ADPCM data D1, D2, ... Are separately transferred to the first track, the second track, and so on, even a musical piece with a back chorus can be started quickly.

(02-2) Download to HDD There are several methods for downloading to the HDD 22C through the ISDN line 40. The first is a method of immediately loading to the HDD 22C. However, since the HDD is used as a cache, it is necessary to store only song data of frequently used songs. Therefore, here, a method for preventing unnecessary song data from being stored will be described.

That is, as shown in FIG. 6, several buffers are set in the RAM 62, and the data fetched from the ISDN line 40 by the CPU 61 is first stored in the ISDN download buffer 62A and written in the performance buffer 62B. play. Shortly after the performance is started, the singer confirms whether or not the desired music is the desired music, and sometimes cancels the performance of the music that is not desired. In many cases, the canceled piece of music is mistakenly designated as a piece of less frequently used music and does not need to be stored in the HDD 22C, so the data is erased without being stored in the HDD 22C. In contrast,
Even after a while from the beginning of the performance (for example, 30 seconds have passed)
If it is not canceled, it can be determined that the musical piece to be played is desired, and in that case, the data is transferred from the download buffer 62A to the HDD buffer 62C and stored in the HDD 22C.

(02-3) Deleting Data from HDD FIG. 6 shows a method for preventing unnecessary data from being stored in the HDD 22C. However, even canceled song data is marked with a cancel mark, for example, in the HDD 22C. If you make a target that can be erased preferentially when the HDD's free space becomes low, the HD will be stored without accessing the host station when a request for the same song is made.
There is also the possibility of immediate supply from D22C. In this method, in addition to / instead of referring to the data usage frequency and usage time for the data management of the HDD 22C, the priority erasing condition of being canceled is referenced.

(02-4) Arrangement Designation Data Structure The data structure of FIG. 5 is such that only one type of data is assigned to one song, but the data structure shown in FIG. 7 is for one song. On the other hand, a plurality of types of data are associated with each other so that the user can select and use them. For example, the lyrics track 1 contains normal Japanese lyrics data, the lyrics track 2 contains lyrics data of the foreign language, and the lyrics track 3 contains lyrics data of a Japanese alternation song. And so on. The same applies to the performance track and the image track, and the data of the arranged contents are included in each. A part of the song header has a table that can specify a combination of these tracks by a simple arrangement number.

In the illustrated example, the arrangement number. Reference numeral 1 indicates an arrangement content in which a lyrics track 1, a performance track 1 (in many cases, a plurality of tracks are actually used), and an image track 1 are used in combination. Such a data structure can reduce the total data capacity (some data can be shared) compared to the case where independent song data is set for each arrangement, and contributes to reduction of communication time and storage capacity. To do. Also, since multiple types of data relating to the same song are associated, there is an advantage that data management is easy.

Even if the data structure is like this, the user does not need to be aware of the arrangement designation. That is, if the Japanese version and the foreign language version of the same song are displayed with different song numbers, as in a normal karaoke song list (book), the song number is simply specified. This is because the arrangement is done automatically.

(02-5) Lyrics font designation When arranging to change only the font of the same lyrics,
Providing lyrics tracks for each font increases the amount of data. In order to avoid this point, here, FIG.
As shown in, insert a font ID that specifies a font in the song header or the lyrics track,
A method of changing the font according to the designation each time the font ID appears will be described. The lyrics track is a time-series arrangement of character codes for displaying lyrics images. If there is a font ID in the song header,
R is the font block stored in the HDD 22C of FIG.
Expand to AM62, start displaying lyrics using characters corresponding to the character code in the lyrics track, and if a font ID is inserted in the lyrics track in the middle, change to the font of that content (different font RAM blocks
Continue reading the lyrics (read to 62).

Although the font data is usually stored in the ROM, if the font is also stored in the HDD 22C and is expanded in the RAM 62 for use as described above, the host station 30 can change or add the font. Therefore, it is not necessary to add a font ROM to the karaoke device side. For this reason, if you want to change not only the font of the same language but also the song that started with lyrics only in Japanese first, and then change to the lyrics of another foreign language, use this font specification function. Available. That is, while changing the font ID included in the music data from the host station, the corresponding font is changed from the HDD 22C to the RAM 6
It can be used after expanding to 2.

(03) Multimedia Sequencer The multimedia sequencer 23B is based on a MIDI sequencer, and is provided with an OS function capable of simultaneously executing a plurality of tasks, so that events of a plurality of tracks included in one piece of music data can be processed by software. It can be executed in real time in synchronization with. An event is a wide range of events that accompany a karaoke performance, and includes not only performance but also lyrics display, background display, sound effects, external device control, and the like.

The song data read from the RAM 62 by the CPU 61 is input to the sequencer 23B.
This song data is, for example, as shown in FIG. 5, MIDI data from a lyrics track to a digital audio track and ADPCM data added at the end. When the sequencer 23B inputs MIDI data for tone generator performance to the tone generator processor 23C, it is converted into musical tone information there. Further, the sequencer 23B adds the ADPCM data to the AD
When the data is output to the processor 66, it is decompressed into data before being compressed. The sequencer 23B controls the expansion timing of the AD processor 66 according to various event information.

As shown in FIG. 9, the structure of one track of normal MIDI data is as follows: from the top of the track TOT to the end of the track EOT.
Event EVENT and duration (waiting time) Δt
Are alternately arranged, the processing by the sequencer 23B is performed in a circulating manner as shown by arrows in the figure (however, the content of each event EVENT is different).

On the other hand, in the case of the track to which the ADPCM data is added as shown in FIG. 5, the MI as shown in FIG.
The ADPCM event is executed in parallel with the execution of the DI event. Since the timing of executing the ADPCM event is described in the MIDI data, the both are executed in software in synchronization. The contents included in the ADPCM event are 1) ADPCM sound number, 2) pitch shift valid / invalid flag, 3) volume, and 4) pitch shift amount.

The ADPCM data assumed here is waveform data during a musical composition, and even if it is compressed, the data amount thereof is still larger than that of MIDI data. However, as far as the back chorus is concerned, there is also a part in which the same back chorus is simply transposed and repeated in one song. Therefore, a portion that can be shared by a plurality of back choruses is prepared in advance as a plurality of independent ADPCM data, and at the time of reproducing the performance music, the ADPCM data necessary for the back chorus reproduction included in the music is selected and combined. In this way, the amount of data transferred from the host station is reduced and the storage capacity is also small. The ADPCM sound number of 1) is for selecting a desired one from a plurality of such ADPCM data.

When the same ADPCM data is used by changing the pitch, it is necessary to show how much the pitch should be shifted. The pitch shift amount of 4) becomes information indicating this change amount. The pitch shift is performed by the pitch shifter 67 shown in FIG. The pitch shifter 67 is a karaoke processor (K
It is configured by using a DSP called P), and is used not only for the transposition in which the pitch is partially shifted, but also in the case of transposing all one piece of music by the user's operation. In this case, the pitch shift is performed by adding to the transpose.

The ADPCM data is used not only for the back chorus but also for waveform data such as sound effects. The valid / invalid of pitch shift of 2) is to avoid pitch shift of sound effects that become unnatural in ADPCM data even if the user specifies transpose, even if the user specifies transpose. This flag is a flag that prevents the pitch shift from being performed only in the reproduction of. The volume 3) is for automatically setting the volume for each event.

(04) General backing chorus / phrase The back chorus of the song is a) "Wow""La".
It has a simple pronunciation like a vowel that accompanies a consonant such as "ah", and can be repeated many times in the same song, or a chorus that can be shared by multiple songs, such as "wow wow. There is a general-purpose chorus such as "..." or "Wowwawwa ...", and a phrase chorus peculiar to the performance song such as b) "Nagasaki". In the sound source karaoke device, it is possible to treat a general-purpose chorus, which is a combination of simple pronunciations, as one type of musical sound, and to pronounce using a sound source. By doing this, the amount of back chorus data can be reduced, so that the time for online communication can be shortened,
The storage capacity can be reduced.

A specific description will be given with reference to FIG.
1) Regarding the general-purpose chorus data E1, waveforms with pronunciations such as “Wa”, “La”, and “Ah” are sampled in advance, and the waveform data is stored as a library on the sound source side. 2) With respect to the phrase chorus data E2, waveform data of a pronunciation such as "Nagasaki" recorded for each song is stored in the HDD 22C or the like and loaded into the RAM 62 at the same time as the performance data. 3) The phrase chorus data E2 is controlled to sound based on the phrase timing data E3 included in the performance data. The general-purpose chorus data E1 is controlled to sound based on sound source control data E4 such as MIDI. Either of them is selected by the selector 80 (in some cases, both are sounded simultaneously). The function of the selector 80 portion may be OR logic. 4) Both of the chorus data E1 and E2 are mixed with other music data generated by the sound source and output.

(05) ADPCM data compression The back chorus data has a large amount of data. Therefore, consider that the back chorus data is compressed on the host station 30 side and expanded on the karaoke device side. Although the compression method is arbitrary, for example, when ADPCM is adopted, the karaoke apparatus expands the AD processor 66. In this case, general chorus data may be included as the back chorus data, but only the phrase chorus data will be considered here.

On the host station side, the phrase chorus data is digitally sampled and then compressed. In this case, for frequency compression, resampling processing is performed on the original data, and the resampling frequency is described in the header of the file (music data). The compression of the number of bits is performed by using a technique such as ADPCM, and the compression information is also described in the header. To give an example,
Normal compact disk data is 44.1K, 16
It is the bit sampling rate, which can be set to 2
Compress to 2.05K, 4-bit, or 11.025K, 4-bit ADPCM data.

On the karaoke machine side, as shown in FIG. 12, the compressed data e2 input to the selector 66A of the AD processor 66 is expanded according to the data conversion information E6 which describes the compression method and the resampling frequency. That is,
One of the bit number converters 66B is selected according to the conversion information E6, and the conversion frequency of the frequency converter 66C is determined. Phrase data E2 expanded in this way
Are mixed with other pronunciations in the mixer 13.

By using such a compression technique, the amount of data to be transferred online can be reduced, and the capacity to be stored in the HDD 22C can be reduced. The conversion information E6 may be set so that the same information is applied throughout one song. In this case, the conversion information E6 is described in the song header of FIG.
This conversion information E6 can be set for each phrase. In the case of setting for each phrase, a header is set for each divided data in the digital audio track, and it is described in the header. When conversion information is set for each phrase, the compression rate is partially changed in the same song, and for example, important parts are compressed with a higher sampling rate and a larger number of bits than other parts. Freedom is born.

(06) Pitch Control In the configuration shown in FIG. 4, the output of the AD processor 66 is input to the pitch shifter 67, where pitch adjustment (pitch shift) is performed according to the pitch information. Pitch adjustment includes transposition, which changes a part of the pitch in the middle of a song, and transpose, which uniformly shifts the entire song. Here, it is considered that the pitch shifter 67 realizes a back chorus in which the same phrase data E2 is used while changing the pitch, for example, a back chorus in which the pronunciation "Nagasaki" is repeated while continuously increasing the pitch.

As shown in FIG. 13, the pitch information E5 input from the outside corresponds to the pitch shifter 67 and the sound source control E.
4 and the back chorus data is input to the pitch shifter 6
At 7, the musical tone is pitch-controlled by the tone generator control E4. Further, in the case of chorus data having different pitches with the same chorus phrase such as when a transposition occurs in one song, transposition information is sent from the sound source control to the pitch shifter, and the pitch of the chorus data is changed.

(06-2) Pitch shift exclusion of sound effects Inputs to the pitch shifter 67 include not only back chorus data but also sound effects. When the sound effect is pitch-shifted, its acoustic effect is completely impaired, and on the contrary, a sense of discomfort is often generated. Therefore, the pitch shift effective /
With reference to the invalid flag, the same pitch is maintained for this sound effect even if the pitch information E5 is input.

(06-3) Delay Compensation of Chorus Data The chorus data is subjected to the bit conversion as described above from the time when the sounding instruction is received to the time when it is output to the mixer effector 68.
Delay occurs due to passing through frequency conversion and pitch shifter.
For example, by using a pitch shifter that is specified to have a delay of 30 ms ± 30 ms in real time, especially MID.
In order to obtain an acoustic output synchronized with the output of the I sound source, the pronunciation instruction data of the chorus data in the sequencer is stored 30 minutes ahead of the data of the sound source to be simultaneously pronounced and stored, and then read out. ,
A sounding instruction is output to the ADPCM processor 66 in advance so as to cancel the delay.

The sequencer 23B controls the shift amount of the pitch shifter 67, and further controls the microphone effector 69 and the sound field generator 70 in synchronization with the progress of the musical composition. FIG. 14 is an explanatory diagram of this. The effector mixer 68 outputs the sound of the sound source processor 23C (for example, sounding ch1 is a guitar, ch2 is a piano,
ch3 is input with the output of the pitch shifter 67 (expansion data corresponding to the ADPCM track), and the signal of the channel selected based on the specified data output from the sequencer by the programmable effector 68A at the input stage is echoed. , Equalizer, reverb, etc. are applied to the processing unit 68B for providing acoustic effects to receive respective processing.

The output of the effector mixer 68 is input to the sound field generator 70, where it is subjected to sound field control according to the performance by the control signal from the sequencer 23B.
The sound field generator 70 digitally generates a plurality of reflected waves having different delay amounts and levels based on the input signal, and synthesizes them to produce a sound as if being heard in a concert hall or a live venue. It is a processor that gives a wider space. When using the sound field generator 70, the speaker 15
It is more effective to use a plurality of systems. For example, in addition to the front, rear, left, and right 4 speakers, the speakers are arranged in the front center and the center left and right, respectively, to form a 6 to 7 speaker system as a whole.

The output of the microphone effector 69 is also introduced into the sound field generator 70, though not necessarily required. In this case, the microphone effector 69 also performs sound effect processing such as conversion from monaural to stereo, echo, exciter, harmonizer, equalizer, etc. even for input from the microphone 14. The sequencer 23B is a microphone effector 6
A control signal is also sent to 9 to control the sound effect in synchronization with the performance. The control signal from the sequencer 23B is each control signal for sound field control, effect control, image processing, lighting processing, stage control, etc., which is configured as one track in the sequence data read from the RAM 62. Is read by the sequencer 23B in the same manner as the performance data, and is controlled in synchronization with the musical tone performance.

(07) Information Display of Reserved Music In a usual karaoke apparatus, when the reserved operation of the performance music is performed, only the music number of the music is displayed on the monitor. Therefore, it is often the case that the request error is not noticed until the performance is actually started. To improve this point,
As shown in FIG. 15, when the display information dictionary is set in the HDD 22C and the song number is designated, the song number, song name, composer name, lyricist name, reservation order, etc. are displayed on the monitor 12,
The requested song can be confirmed in advance from the displayed song name, and if the song name is the same, the singer name, composer name, and the like. The location to be displayed is selected, for example, in the upper right corner of the screen so as not to hinder the currently playing music.

Such confirmation of the request contents is necessary for avoiding unnecessary access when music data is fetched from the host station as in the case of musical tone synthesis type sound source karaoke. In addition, even in LD karaoke, a mistake in selecting a song is common, and it is useful for avoiding this.

(08) Automatic transpose With regard to transpose, when the user specifies a range in which he can easily sing, the karaoke device automatically determines an optimum transpose value and performs an automatic performance such as playing. There is. Specifically, the karaoke device records the highest-frequency range for each karaoke song in the header portion of the song data, and when the user specifies the median value or range (upper and lower limits) of his or her vocal range, the karaoke device , The transpose value is set so that the specified median value of the vocal range or the median value of the range matches the median value of the musical range of the karaoke piece, and transposition is automatically performed.

As the median value of the range of karaoke songs, for example, as shown in FIG. 16, a distribution curve of the number of pronunciations for each pitch or the product of the number of pronunciations and the pronunciation length is obtained, and the distribution curve is set to an appropriate threshold value. Use the intermediate value of the local range obtained by slicing with. This median value information is included in the header of the music data. There are various methods of automatic transposition that make it easier to sing a karaoke song, but the method of matching the median range of the song with the median range of the singer's voice as in this example has the highest frequency with the song. Based on the idea that if the part that appears is easy to sing, the whole part will be easy to sing.

When the applied karaoke apparatus is a sound source karaoke apparatus, as shown in FIG.
The median value of the musical range is read from 0 (which may be the HDD 22C) and used. Reference numeral 101 is a ROM storing a program executed by the CPU 61. Also, as shown in FIG.
In a karaoke apparatus that uses a digital storage medium such as DG / CDI, range data is written in the device 102 such as CDG / CDI, and the tune number is read and used. Further, as shown in FIG. 19, in a karaoke device such as an LD in which digital data cannot be read, a range data storage device 103 is provided and used. The storage device 103 may be a part of the HDD 22C.

(09) MIDI instrument connection The MIDI instrument 42 is connected to the karaoke system of FIG. 4 so that an external MIDI instrument 42 (for example, percussion) can produce instrument sounds, sound effects, etc. during karaoke performance. Input terminal, MIDI interface 64
Is provided. The external MIDI data input from the MIDI interface 64 is taken in by the CPU 61 and supplied to the sequencer 23B. Then, without providing a sound source dedicated to the external musical instrument, a musical instrument sound or the like is generated using one channel of the karaoke sound source (external musical instrument sounding ch). For this purpose, the sound source processor 23 for karaoke
C is an external MIDI for all channels, for example, 1 channel out of 16 channels (32 channels if there are 2 systems)
It is necessary to create data for karaoke songs in advance so as to leave space for musical instruments. Alternatively, when the CPU outputs the external MIDI data received from the interface 64 to the sound source, the MI that is not used in the performance data at that time.
It is also possible to forcibly switch to the DI channel.

(09-2) MIDI merge MIDI data is 8-bit stream data that starts with a status byte and is followed by a data byte. Therefore, M data for karaoke performance read from the RAM 62.
When the MIDI data for musical instruments, which is asynchronously input from the interface 64, collides with the IDI data, data destruction occurs. In order to avoid this point, performance MI
When DI data is flowing, MI for musical instruments is externally supplied.
When DI data is input, the latter transfer is delayed until the former data byte is completed, and merge is required to avoid collision between the two.

(10) Chord name display In a normal karaoke device, the lyrics of the song being played are displayed in order on the monitor, but the pitch information is not displayed. For the user who connects and plays the MIDI musical instrument 42, the monitor 12
For example, when the chord name is displayed in
It becomes easier to play the guitar). It would be even more convenient if this code name display were made to correspond to the lyrics display. Figure 20
(A) is an example in which the chord name (C, F, G7 ...) Is displayed on the monitor 12 in addition to the lyrics ("Honyara ..."). In the same figure (b), when the same song is transposed,
In this example, the code name is automatically shifted (C #, F #, G7 # ...) In association with the shift amount (+1 in this example).

The following method can be considered for realizing the code name display. First, the following "key" is recorded in the header part of the music data. (C, C #, Db, D, D #, Eb, E, ... A #, Bb, B) + major / minor Used to determine the code name when transposing this (# system or another system? ). The display data of the lyrics in the sound source karaoke is "display event" and "duration (Duratio
n) ”is repeated. The display event is the display coordinate x,
y, an attribute, a display character string, a display time, and a color change timing are shown, and a duration shows a time between display events.

The lyrics are displayed by the character string of the lyrics to be displayed, the display coordinates, and the display attribute (color designation), and the lyrics color is changed in synchronization with the performance in accordance with the "color change timing". On the other hand, when the "display time" has elapsed after displaying the lyrics, the lyrics are deleted. If the display events are described consecutively without sandwiching the duration, they are displayed simultaneously.

The "display character string" includes a character code, ruby character designation, and character pitch designation. E for character code
When the UC code is used, the "display character string" is expressed as follows (0xXX represents a hexadecimal number). Character code 1-byte character: 0xXX..0x7e 2-byte character: 0xa0..0xf4 (2nd byte is 0xa0 to 0xff) Space between characters: 0x01..0x1f Ruby: 0xfe Ruby character string 0x00 Ruby character string 0x00

By expanding the above data structure, it becomes possible to display a chord code having the following data structure. 1st byte: chord chord ID (0xFF) 2nd byte: key bit 0..3: = 0: C 1: D 2: E 3: F 4: G 5: A 6: B bit 4..5: = 0:-1: # 2: b bit 7: = 0: major 1: minor (Example: C major = 0xXX, Ab minor = 0xa5) 3rd byte: Decoration code 0: Major code (no decoration code) 1: 7th 2: 6th 3: Diminish 4: sus4 5: 7th sus4 ...

An example is as follows. C7: = 0xff, 0x00,0x01 Am6: = 0xff, 0xa5,0x02 F # m7: = 0xff, 0x93,0x01 Dsus4: = 0xff, 0x01,0x04

When the transpose is not specified, the above display is performed. When the transpose is specified, the code name to be displayed is determined from the following three values. a) Key value of the second byte of chord chord data b) Transpose value c) Key of the song recorded in the header

An example is as follows. Song key: C major Chord: G Transpose: +1 Display: G # Another example is as follows. Song key: A minor Chord: G Transpose: +1 Display: Ab

(10-2) Minus One Performance A musical instrument name (timbre) is stored for each channel of the sound source used in the performance track. For example, channel 1 is a piano, channel 2 is a guitar, and so on. Then, when the MIDI musical instrument 42 is connected and the tone color is designated,
Do not play the musical tone by not sending the tone color data to the sound source. In this way, the player of the MIDI musical instrument 42 can enjoy playing the musical instrument as if playing a part of the musical piece.

(11) Triple image The image displayed on the imposing monitor 12 can be superimposed on the triple image at maximum. An example will be described with reference to FIG. 21. First, a background image signal (external video signal) video of a moving image output from the LD changer 11 of FIG.
It is 0. Second, the computer screen generation device 302
This is the background image signal video1 of the still image output from (23D1 in FIG. 4). The third is the lyrics image signal video2 output from the computer screen generation device 303 (23D2 in FIG. 4). Two screen generation devices 302,
Reference numeral 303 has a video RAM (VRAM) independently, and it is possible to specify transparency at all points of the image space by the signals YS1 and YS2. .

In a normal karaoke apparatus, 1) a method of superimposing a computer generated lyrics image on a background image of a moving image recorded on an LD or the like, and 2) a background image of a still image generated by a computer. Similarly, there is a method of superimposing computer generated lyrics images. In the method 1), when the recording medium such as an LD is limited, the same background image may be used for different musical pieces, which may be uninteresting. In the method 2), since the background image is a still image, the screen expression becomes more monotonous than the method 1). On the other hand, as in the present invention, if a still image of a small area is overlaid and displayed on the background image of the moving image, and further the lyrics image is overlaid and displayed thereon, a superimposition of three layers at maximum is possible. Imposing is possible, and even if only a small amount of moving image information is used, the screen expression can be varied.

Output vid of the screen generation devices 302 and 303
eo1 and video2 are combined by the switching device 306 using a high-speed switch. A switching signal generation device 307 generates a necessary switching signal in the switching device 306. The output of the switching device 306 is the superimposing device 310 (see FIG.
23E) is further combined with the external video signal video0. Reference numeral 311 is a switching signal generation device that generates a necessary switching signal in the superimposing device 310 using a high-speed switch. This superimposing device 310 has a combination of RGB and NTSC both externally and internally.

For the composition in the superimposing device 310, it is necessary to synchronize the RGB internal video signal video1 / 2 output from the switching device 306 and the external video signal video0 supplied in the NTSC system. Three
Reference numeral 08 denotes an external horizontal sync detection unit for detecting the horizontal sync signal Hsync of the external video signal, 309 denotes an external vertical sync detection unit for detecting the vertical sync signal Vsync of the external video signal, 30
4 is an internal vertical synchronization signal Vsync of the screen generation device 302.
It is an internal vertical synchronization detection unit for detecting.

For synchronization between the outside and the inside, a PLL (phase locked loop) including a VCO (voltage controlled oscillator) 301 and a vertical synchronization device 305 is used. The horizontal synchronization device 305 is a block including this PLL phase detector and loop filter. Internal screen generation device 30
The vertical reset signal Vreset from the internal vertical synchronization detection unit 304 that detects the synchronization signal Csync output from the screen generation device 302 is used for the vertical synchronization and the horizontal synchronization between the two and 303. This reset signal Vre
If set is used, it is possible to synchronize two internal images without an external video signal.

(12) The center file-safe tone synthesis type karaoke system lacks reliability unless the function of the host station is always guaranteed. In order to improve the reliability of the communication system, a normal communication control device (CC
Introducing U) and multiplexing increases the system cost.

The file safe mechanism of the center shown in FIG. 22 has a plurality of communication control units 204 installed between the public network 202 on the karaoke machine side for accessing the center 201 (host station) and the network 203 on the center side. The karaoke side network 202 is divided into groups so that any communication control unit 204 can be accessed with a representative number. Each communication control unit 204 separates the power supply system of the communication control communication server WS (workstation) 205 and TA (terminal adapter) 206. TA206 is an interface I / F for ISDN line and a modem m for analog line
Includes ode, ISDN I / F is Ska G bus SCS
I is connected to the communication server WS205, and the modem is R
It is connected to the communication server WS 205 with an S232C cable.

A DB server WS (monitoring WS) 207 is provided above the communication server WS 205. The DB server WS207 monitors normality / abnormality of each communication server WS205 by polling, for example, and sends information to the power controller 208 when an abnormality is detected.
TA 206 under the faulty communication server WS 204
Turn off the power. At this time, the failed communication server W
The power of S204 is not turned off. This is the communication server WS
This is for avoiding the influence (crash) on the HDD and the like accompanying the 204. In some cases, the TA 206 may fail. Also in this case, the power controller 208
Turns off only the power of TA206.

When the power of the TA 206 of a part of the communication control units 204 among the plurality of communication control units 204 is turned off, even if an attempt is made to make a call to the TA 206 from the public line network 202, there is no response and the call cannot be made. In such a case, in the representative line system, the next line is automatically selected and accessed, so that it becomes possible to access the center 201 through the normally operating communication control unit 204. The advantages of this system are that no special communication control device is required, that the line can be added inexpensively, and that the WS side does not crash because the TA power is simply turned off.

If such a measure is not taken, TA2 of the communication control unit 204 in which the communication server WS205 is out of order
06, in which case there is no response from the communication server WS 205 forever. Therefore, it is considered that the karaoke device side waits wastefully for a long time until it receives a response, or attempts to hang up and call back again. On the other hand, by turning off the power supply of the TA 206 as described above, it is possible to prevent such waste.

(13) Portable download machine For example, in the HDD 22C of the karaoke system shown in FIG.
When downloading song data for 0 songs at a time, IS
If the DN line 40 is used, the line charge will be high. The portalable (carryable) downloader 71 is for realizing such a large amount of data download by musical sound reproduction, and thus at low cost. FIG. 23 shows a state in which the download device 71 is transported to a karaoke device installation store that needs to download song data and is loaded in the storage device 22C (hard disk HDD or mini disk MD or the like). The downloaded downloader 71 is transported to the next store and used for downloading.
This downloader 71 can be used not only when loading new data but also when updating all data at regular intervals.

(14) Updating of data to mobile unit The above-mentioned portable downloader 71 is more effective when used for downloading to a mobile unit to which it is difficult to connect the ISDN as described above. In FIG. 24, a sound source karaoke device 402 is mounted on a bus 401, and passenger types (foreign passengers,
According to elderly customers, female customers, male customers, etc.), the song data to be used for the day before the operation is downloaded from a general-purpose download machine 403 called karaoke loader (corresponding to 71 in FIG. 4) to the storage device of the karaoke apparatus 402. An example is shown. A plurality of types of download machines 403 are prepared, and necessary music data is loaded from the file server 404 on a regular basis. To the file server 404, all necessary music data is supplied from the host station 30 via the ISDN line 40. By doing so, the system of the present invention can be introduced even in a mobile body whose line connection is difficult or whose communication speed is slow. FIG. 25 is a functional block diagram of FIG. 24, and the file server 404 is shown as a sub-host of the bus company.

[0089]

As described above, according to the present invention, it is possible to provide a karaoke control device which realizes the transition to the musical sound synthesizing type karaoke device at low cost while effectively utilizing the existing musical sound reproducing means. .

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of a general karaoke system to which the present invention is applied.

FIG. 2 is a block diagram of a karaoke control device.

FIG. 3 is a block diagram functionally showing the system of FIG.

FIG. 4 is a configuration diagram showing the system of FIG. 1 in detail.

FIG. 5 is a data structure diagram showing a format example of music data.

FIG. 6 is an explanatory diagram showing functions of various buffers.

FIG. 7 is a data structure diagram showing another example of the format of music data.

FIG. 8 is an explanatory diagram of a font ID.

FIG. 9 is an explanatory diagram showing an execution mode of a MIDI event.

FIG. 10 is an explanatory diagram showing an execution mode of an ADPCM event synchronized with a MIDI event.

FIG. 11 is a configuration diagram of a case where a part of chorus data is universally sounded from a sound source.

FIG. 12 is an explanatory diagram of compression / expansion of phrase chorus data that cannot be sounded from a sound source.

FIG. 13 is an explanatory diagram of pitch shift.

FIG. 14 is an explanatory diagram of multiple control by a sequencer.

FIG. 15 is an explanatory diagram of multiple display of performance music information.

FIG. 16 is an explanatory diagram of a method of determining a median value of performance music pieces.

FIG. 17 is a configuration diagram showing a first example of a median value storage method of music data.

FIG. 18 is a configuration diagram showing a second example of a median value storage method of music data.

FIG. 19 is a configuration diagram showing a third example of a median value storage method of music data.

FIG. 20 is an explanatory diagram of a code name display.

FIG. 21 is a device configuration diagram that enables triple image superimposition.

FIG. 22 is a configuration diagram of a communication system in which center files are multiplexed.

FIG. 23 is an explanatory diagram of a musical sound synthesis portable downloader.

FIG. 24 is a system configuration diagram of a download to a moving body equipped with a karaoke device.

FIG. 25 is a block diagram functionally showing FIG. 24.

[Explanation of symbols]

10 ... Music reproduction type karaoke apparatus, 11 ... LD changer system, 12 ... Monitor, 13 ... Mixing amplifier, 1
4 ... Microphone, 15 ... Speaker, 20 ... Karaoke control device, 21 ... Commander part, 21A ... Remote control receiver, 2
1B ... Key input interface, 21C ... Song selection / reservation controller, 21D ... Display, 21E ... Display driver, 21F ... AV input / output selector, 21G ... Control output interface, 22 ... File system unit, 22A ... File controller, 22B ... ISD
N controller, 22C ... Storage device, 23 ... Performance / display unit, 23A ... Data buffer, 23B ... Multimedia sequencer (MMS), 23C ... Sound source processor, 23D ... Video processor, 23E ... Video super imposer, 25 ... Microcomputer board , 30 ... Host station, 40 ... Communication line, 50 ... Remote control, 61 ... CPU,
62 ... RAM, 63 ... ISDN interface, 64
... MIDI interface, 65 ... Changer control, 66 ... ADPCM processor, 67 ... Pitch shifter, 68 ... Effector mixer, 69 ... Microphone effector, 70 ... Sound field generator, 71 ... Portable downloader.

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 5D080 BA01 DA06 FA01 HA01                 5D108 BA04 BB03 BB06 BC12 BG07                       BG08                 5D378 MM52 MM97 QQ01 QQ38 TT08

Claims (2)

[Claims] 1. A musical tone reproducing means for storing an audio signal of a musical piece to be played and reproducing and outputting an audio signal of the designated musical piece, and a sound source according to the event information of the musical piece to be stored, which stores event information of the musical piece to be played. The musical sound synthesizing unit for synthesizing and outputting musical tones, and the commander unit for receiving a music selection instruction of a performance song, the commander unit, based on the received music selection instruction,
A musical sound output device, which selectively controls whether the musical sound reproducing means reproduces a musical piece to be played or the musical sound synthesizing means synthesizes musical sounds of a musical piece to be played. 2. The commander section causes the musical sound reproducing means to reproduce the musical piece when the musical piece corresponding to the received musical selection instruction is stored in the musical sound reproducing means, and the musical piece reproducing means reproduces the musical piece. When the musical composition corresponding to the selected music selection instruction is not stored in the musical sound reproducing means and is stored in the musical sound synthesizing means, the musical sound synthesizing means synthesizes the musical performance. The performance output device according to claim 1.