KR20050018929A - The method and apparatus for creation and playback of sound source - Google Patents

Abstract

PURPOSE: A method and a device for generating/reproducing a sound source to make music, MR(Melody Recite), caption, chorus, and music video with one digital audio file are provided to generate the MR, the music including a voice of a signer, lyric synchronization information, lyric information, image information, image synchronization information, and voice information into one digital audio format. CONSTITUTION: A sound source generator comprises a sound source analyzer/channel divider analyzing/dividing the input sound source into each channel, a sound channel mixer mixing the channels of the input sound source, a synchronization information assigner assigning start synchronization information of caption or an image, and a lyric voice inserting recites the lyrics in advance. A sound source player comprises a channel generator generating the channels in stereo, a channel controller controlling the generated sound source, a synchronization information separator separating the synchronization of the caption/image, and a multimedia player synchronizing/playing the image and audio.

Description

The method and apparatus for creation and playback of sound source capable of general music, accompaniment music, lyrics pre-loading, chorus function, music video production in one file

The present invention provides a method and apparatus for generating a sound source capable of implementing accompaniment music and a singer's voice dubbed in a sound source production process, lyrics, video, chorus function, and pre-loading of lyrics, and caption caption, image synchronization, The present invention relates to a reproducing method and apparatus for reproducing lyrics, accompaniment music, and voice-dubbed music as a single sound source file. More specifically, a sound source generated by the present invention is a simple WAV, MP3, WMA, It is a file with a digital audio bitstream. It inserts synchronization information about subtitles and video, and includes general music, accompaniment music, lyrics loading function, general music and accompaniment music including the singer's voice depending on the playback control method. In addition to natural exchanges, subtitles and images such as song cycles can be processed using the synchronization information contained in the file. It is.

In general, in order to produce a record, a singer voice is mixed with an accompaniment music (MR) produced separately in a studio to produce a record containing a singer voice. Records recorded on records are usually music containing singer voices, and nowadays they are sold in records called MR. However, since the sound source containing the MR track and the singer voice of the same song name exist as two separate tracks, the capacity of the storage medium is large.

In addition, the related art simply includes MR music on the CD to listen to only musical instrument sounds, but does not provide other additional functions. MR music, which consists only of musical notes, maintains better quality than MIDI music in the song cycle or music synthesized in MP3, but does not include the ability to color lyrics as the music progresses. And when creating a file containing images and music, such as a music video, the music video is made to unilaterally watch by inserting the music video's video and the singer's song. Can't use it. In general, in order to use a music video in a karaoke cycle, when the music video is played, the music part is muted and the music video is played in synchronization with MIDI music produced by a karaoke cycle company.

In the conventional art, in the accompaniment music of the song accompaniment type, when there is no subtitle, the accompaniment music alone cannot easily know the lyrics, so that the lyrics call function is added before the song starts to enhance the familiarity of the lyrics. For example, in song workshops, the instructor sings songs while playing the instrument, and pre-songs are pre-sung before the measure begins. I can't. It is possible to create a sound source by adding the song lyrics function to the MR sound source, but it is not easy to perform the task so that the section for accommodating the song lyrics of the accompaniment music does not overlap the singer voice part, and it is expensive and time-consuming. . In addition, if only accompaniment music is required, there is an inconvenience in that a separate music source without the lyrics recall function must be stored.

And in the chorus function with chorus function, accompaniment music and chorus music are stored separately and synchronized while playing chorus music and accompaniment music at the same time. In order to insert the chorus function, chorus is used to create a chorus part separately from the accompaniment music, but it costs a lot of cost, time, equipment, and personnel, and it is not easy to create a chorus part from a single music file.

In addition, no device has been developed in which the accompaniment music is played without the voice music by simply inputting a simple control button when singing along with the music containing the singer's voice. Have a significant difference in sound quality. In addition, even when music is selected from an output device such as a speaker while simultaneously accompaniment music and music containing a singer's voice, it is difficult to completely synchronize the two music, and the sound quality is damaged when the music is switched.

And the lyrics of the sound source purchased by the user are printed in a separate paper form, and even if it is stored as a digital file, the subtitles are not colored or captioned in the desired section like the song cycle, and if the music is printed or saved as a lyrics file If present, there are difficulties in management and storage. In addition, the caption caption is inconvenient for the user to create a caption caption synchronized to the sound source through a separate operation.

In order to solve the above part, multiple files must be managed for the same song instead of one file, many hardware resources are required to reproduce multiple files at the same time, and synchronous reproduction for multiple files is difficult. It also takes a lot of time, money, people and equipment to make other files involved. For example, since chorus music, accompaniment music, music containing singer voice, lyrics, video, etc. must be used in synchronization with one piece of music, synchronous playback and file management are difficult, and storage capacity is large.

An object of the present invention is to solve such a problem in the prior art, the music and accompaniment music containing the singer voice, lyrics sync information and lyrics information, video sync information and video information, lyrics voice information as one digital audio Method and apparatus for generating in a format (PCM, MP3, WMA, OGG, etc.), using the generated sound source to listen to the sound source containing the singer's voice, a chorus cycle with chorus function, song lyrics, song lyrics coloring, The present invention provides a method and apparatus for reproducing a music video in karaoke cycle content.

Generating and reproducing a first to eighth step of generating and reproducing a sound source in order to achieve the above object; A sound source generator formed of a sound source analysis and channel separator, a sound source channel mixer, a sync information allocator, and a lyrics voice inserter, and a playback device formed of a channel generator, a channel controller, a sync information separator, and a multimedia player.

In a conventional sound source generation and reproduction method, after inserting synchronization information for a subtitle or video into one digital audio bitstream file, a sound source is generated and played using the synchronization information to generate general music, accompaniment music, and chorus. Function, enabling the preloading of lyrics.

The audio bitstream file may include a first step of analyzing a sound source input through a sound source analysis and a channel separator, and then separating each sound source for each channel; A second step of mixing channel data through a sound source channel mixer to generate a new audio bitstream (RO, LO); A third step of allocating start synchronization information of the caption and video through the synchronization information allocator; Inserting a pre-singing song lyrics function through the lyrics voice inserter; Generating a mono channel to a stereo channel through a channel generator; A sixth step of controlling a sound source through a channel controller and performing a function of listening to music and pre-song lyrics; A seventh step of separating the synchronization information of the caption and the image through the synchronization information separator; After the video and audio are separated by the multimedia player, the eighth step of synchronous playback is generated and reproduced.

In the above-described fifth step of generating a channel, a phase filter or a spatial filter used in 3D stereo sound is employed to increase the sense of space corresponding to the stereo output.

In the above, the performing steps of the first to fourth steps may be varied, provided that the first step always has a structure that precedes the second step, and the third step always has a structure that precedes the fourth step. .

In the above, the first to fourth steps, which are the generating methods, are performed in the order of the first, third, second, and fourth steps, or the first, third, fourth, and second steps. The process may be performed in the order of the third step, the first step, the second step, the fourth step, or the third step, the fourth step, the first step, or the second step.

In the above, the third step may be omitted when it is not necessary to allocate the synchronization information of the subtitle and the video, and the fourth step may be omitted when the song lyrics preload function does not exist.

A conventional sound source generating and reproducing apparatus, comprising: a generating device for generating a sound source and a reproducing device for reproducing the generated sound source, wherein the generating device comprises: a sound source analysis and channel separator for separating input channels after analyzing an input sound source; A sound source channel mixer for mixing the channel of the input source; A synchronization information allocator for allocating start synchronization information of the caption or video; And a lyrics voice inserter for pre-singing the song lyrics. The playback apparatus includes: a channel generator for generating a stereo channel; A channel controller for controlling the generated sound source; A sync information separator for separating the sync information of the caption and video; And a multimedia player for synchronously playing back video and audio.

Hereinafter, the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the technical idea of the present invention.

First, a generation method and apparatus for generating a sound source in the present invention will be described.

1 is a representative view showing a generator and a playback device of the present invention, Figure 1 (a) is a representation of the structure of the generator of the present invention, the generator is a sound source analysis and channel separator, sound source channel mixer , Sync information allocator, lyrics voice inserter can be composed of various combinations, wherein the input of the generator is lyrics, video, accompaniment music (MR), the sound source containing the singer voice, lyrics voice sound source, the output is synchronized Digital audio bitstream containing sync start information of the lyrics and video.

And, among the component devices constituting the generating device, the sync information allocator 1 and the lyrics voice inserter 2 maintain the preceding structure, and the sound source analysis and channel separator 3 and the sound source channel mixer 4 are also maintained. It has a preceding structure before and after. The generating device may be composed of various combinations (1-2-3-4, 1-3-4-2, 3-1-2-4, 3-4-1-2, etc.) satisfying the above conditions. .

Figure 2 (a) represents the sound source and the MR sound source containing the input mantissa voice input to the sound source analysis and channel separator.

FIG. 2 (b) shows the output bitstream of the sound source analysis and channel separator. The sound source analysis and channel separator analyzes each sound source with respect to the input MR music and the singer voice. Make the ends the same. If the input and the end of each input source do not match, insert or delete ZERO so that the start or end of the sound source is the same so that the beginning and end of the two sources are completely matched. Separate the two sound sources into R / L channels and input them to the sound source channel mixer.

2 (c) represents the weight and the sound source of the sound source assigned to the channel as the output of the sound source channel mixer. The sound source channel mixer divides the channel of the input sound source into MR music (R1, L1) and singer voice (R2, L2) and mixes the channel data so that it can be included in a single bitstream. , LO). If the accompaniment music is to be put into the R channel RO of the new bitstream, the music including the singer voice is inserted into the L channel LO of the new bitstream. It is preferable to allocate RO with different weights rather than assigning R1 that does not include the L1 component or assigning an average of (R1 + L1) / 2. For example, if the channel information of the sound source is completely out of phase, simply adding the audio signal may cause the audio signal itself to become zero. The allocation scheme for the channel mixer is preferably x > y >

RO = xR1 + yL1

L0 = xL2 + yR2

(x + y = 1, (x> y> 0))

FIG. 2 (d) shows that synchronized subtitles A, B, C, and D are allocated when an audio bitstream sounds "A, B, C, D" as an output of the subtitle and video synchronization information allocator. Would express.

The caption and video synchronization information allocator is for processing subtitle coloring and background images like song cycles, and assigns start synchronization information indicating the start of synchronous playback of the synchronized subtitles and the background image. Sync information is allocated using a byte counter to represent on the stream. The source analysis and channel separator performs synchronization while playing the music containing the singer's voice whose start and end match the two sound sources or the synthesized sound source in the sound source channel mixer.

The allocation method of synchronization information using the byte counter is as follows.

3 is a diagram for allocating still images, moving pictures, subtitles, and the like to an audio bitstream in a synchronization information allocator. The synthesized audio bitstream is composed of a byte sequence of b0, b1, b2, b3, b4, b5, b6..b97. The pronunciation of "TEXT1A" starts at the 0th byte b0 and ends at the 6th byte b6. If the synchronization information for the subtitle is allocated to (start byte, end byte) for each character by using the byte count value, the synchronization information of each character is T: (0,1), E: (1,2), X : (2,3), T: (3,4), 1: (4,5), A: (5,6) That is, FIG. 3 shows that the synchronization information for the image is allocated. In the same manner as the synchronization information is assigned to the letters, the synchronization information is allocated to the image as IMAGE1: (0,90) and IMAGE2: (90,98).

 FIG. 2 (e) shows a synthesized speech signal which is inserted in a lyrics pre-loading insertion section as an output result of the lyrics speech inserter and inserted by zero in a section having a lyrics speech signal and a mantissa voice. If there is synchronization information of the subtitles on the audio bitstream with reference to the synchronization of the subtitles of the lyrics in FIG. 2 (d), the lyrics pre-loading cannot be performed to the part having the singer voice. The remaining empty space is a free space for inserting pre-speech voice data into an area having only accompaniment and no voice of the singer. The pre-loaded voice signal is pre-loaded with lyrics. The voice bitstream is mixed by one measure in the voice insertion section, or the voice signal synthesized using the automatic voice synthesizer such as subtitle synchronization information and TTS. Adjust the utterance rate of the voice signal to enter the section or reduce the playback width of the voice signal using the speed variable to pre-load the lyrics. Insert the voice in the voice insertion section, and zero the bitstream of the voice in the non-insertion section. To make a lyric voice signal.

FIG. 4 is a state diagram for explaining a pre-loading voice insertion section. When reproducing a WAVEFORM-type sound source containing a singer's voice, the synchronization information for caption culling is assigned to a subtitle portion synchronized with the singer's voice. For example, the part to which the synchronization information is not allocated may be a lyrics pre-speech voice insertion section and a lyrics voice signal for the next measure may be inserted.

The lyrics voice inserter is introduced to pre-song lyrics, and may be omitted if there is no lyrics pre-loading function.

FIG. 2 (f) shows a final synthesized sound source, weighted lyrics voice signals, is added to the R0 channel to which MR is assigned, and assigned to the R-CH of the final sound source, and the L-CH of the final sound source is a sound source channel mixer. Allocate to the data of L0 output from.

That is, FIG. 2 (f) shows the form of the final output bitstream implemented in the sound source generator. The final audio bitstream is a mixture of the lyrics voice signal and the output of the sound source channel mixer.

Looking at the playback method and apparatus using the sound source generated in the present invention,

FIG. 1 (b) shows a structural diagram of a playback device of the present invention, wherein the playback device is composed of a combination of a channel generator, a channel controller, a sync information separator, and a multimedia player.

First, the channel generator performs a stereo generation function to make a sound source in stereo on the MONO channel. The R-CH of the generated sound source is a MONO sound source combining accompaniment music and lyrics pre-speech voice data, and the L-CH of the generated sound source is It consists of a MONO sound source containing the voice of a singer. If the MONO is playing the music in stereo and outputs the same data to the R / L channel at the same time, the sound is not good and there is a significant difference in sound quality from the original two sources that were input to the generator. To compensate for this weakness, the MONO channel was created with weights so that the sound source channel mixer could slightly reinforce the parts of the two channels, and in order to increase the spatial feeling corresponding to the stereo output in the channel generation, the phase filter or the 3D stereoscopic Adopt a spatial filter used in sound. However, if the filter consumes a lot of computation for real-time playback, it can be simplified to invert the phase by 180 degrees or to output stereo audio to both R / L channels simultaneously.

FIG. 5 illustrates an example of a phase filter for making MONO music into stereo, and generates another channel while shifting a phase according to the frequency of the input MONO music. MONO Assigns music to one channel of stereo and music output from the phase filter to another channel of stereo and outputs it in stereo.

The HRTF (HEAD RELATED TRANSFER FUNCTION) filter, which can generate 3D stereo sound, creates a sound source in stereo so that the listener feels a stereoscopic sense of the sound source as shown in the figure.

6 is a system diagram for obtaining an HRTF used in three-dimensional stereo sound. While playing music with a computer, the sound source output through the speaker propagates the space and measures how it is input to the actual listener's ear, and calculates the input and output relations to calculate the relationship between HRTFR (right transfer function) including stereoscopic and spatial sense, Implement HRTFL (left transfer function). If you input the sound source to the filter with these two transfer functions, you will feel as if there is a speaker in space. Therefore, using the three-dimensional stereo filter can create stereo music.

The sound source generated as described above may be controlled through a channel controller to perform a function of listening to music including a vocal cycle and a singer's voice and pre-loading song lyrics. The control method of the channel controller will be described below.

First, in order to listen to only the sound source containing the singer's voice, the L generator, which is a channel containing the singer's voice in the sound source inputted to the channel generator, is made R / L in the channel generator and outputted. Representing the reproduction order in FIG. 2 (f), data is read in the order of 1-2-3-4-5-6-7-8 and output in stereo. If you want to listen to only MR music, the part where no lyrics are inserted in the L channel and the part where there is no voice corresponding to pre-loading the lyrics in the R channel are read and played alternately. In FIG. 2 (f), the reproduction order is expressed, and data is read in stereo sequence in the order of 1-2'-3-4'-5-6'-7-8 'and output as stereo. If you want to listen to the accompaniment music while the lyrics are pre-loaded, only the R channel is output in stereo from the channel generator. In Fig. 2 (f), the reproduction order is expressed, and data is read in the order of 1'-2'-3'-4'-5'-6'-7'-8 'and reproduced in stereo. To play music containing a singer voice while the lyrics preload function is turned on, the R channel mixed with the lyrics voice signal and the MR is played first, and then the L channel is played in the section including the lyrics. Where the singer's voice does not appear, the MR music and the music containing the singer's voice are the same, so that the music is reproduced in the order of 1'-2-3'-4-5'-6-7'-8 in FIG. .

The control method of the player for operating like a song half cycle with a chorus function is as follows. Regardless of whether the lyrics preload function is on or not, the R-CH that is MR is played, the L-CH that has a singer voice is played for the part requiring chorus, and the R-CH is played again if the chorus is not needed. You can use your voice as a chorus of half-songs. Since the singer's back-chorus voice is usually pre-inserted in the MR, the chorus can function even if only the MR music is played.

If there is no MR music used in the present invention, using the VOICE CUT or devocalizer algorithm can be used to treat the music with reduced or deleted singer voice as MR, if the lyrics preload function is omitted, the voice of the singer is included If the R channel is configured as a temporary MONO channel with a VOICE CUT or devocalizer algorithm while playing music, the above-described operation may be performed in real time.

The caption and video sync information separator is an apparatus for separating caption and video sync information included in or separately stored on an audio bitstream and inputting a synchronous control signal of the caption and video signals to a multimedia player. The synchronization information may be allocated as time information, information related to a frame, or a byte count value on the bitstream. The caption synchronization information provides a synchronization control signal for representing a corresponding caption in an OSD device including an LCD when the audio bitstream is input to the multimedia player to play music. For example, if the caption synchronization information is allocated as the byte count value of the audio bitstream, when playing the audio bitstream while increasing the byte counter value, if the byte count value, which is the set caption synchronization information, and the value of the byte counter match, the multimedia is multimedia. A control signal for expressing a caption is input to the video player to express the caption on the screen.

The image synchronization information provides a synchronization control signal for representing a background image in an OSD device including an LCD when the multimedia player plays music. For example, if the start sync information of the video is allocated as the byte count value of the audio bit stream, when the audio bit stream is played while increasing the byte counter value, the byte count value, which is the set video start sync information, matches the value of the byte counter. If the control information is input to the multimedia image player.

The multimedia player is an A / V player that includes music and video, and the audio bitstream is converted into an audio player (PCM decoder, WAV decoder, MP3 decoder, WMA decoder, OGG decoder, AAC decoder, etc.), and the video information is converted into a video player (MPEG1 decoder). , MPEG2 decoder, DIVX decoder, etc.) and the subtitles are overlaid on the video and output to the screen.

7 shows a compressed A / V bitstream constituting a general music video. The bitstream constituting the music video A / V is divided into video and audio data therein, and the video and audio are compressed. Therefore, if the final generated sound source of the present invention is used here, the music video can be used as one song cycle.

FIG. 8 is a block diagram of an encoder for generating a compressed video bitstream by inputting subtitle synchronization information synchronized with an uncompressed bitstream, an audio uncompressed bitstream, and an audio bitstream of a music image to a multimedia encoder. The input audio bitstream of is the final sound source generated in the present invention, and the input sync information or sync information file is assigned as byte count value for the audio bitstream as sync information for the caption caption. The input video bitstream represents a video uncompressed bitstream in which audio is not included. Multimedia encoders including MPEG, DIVX, and the like generate data in the form of A / V bitstreams in FIG. The multimedia encoder receives data from an input video or audio bitstream buffer and increases the multimedia bit force byte counter value when data is input into the multimedia output buffer. The multimedia encoder reads an image bitstream to generate a video frame, and reads data from an input audio bitstream buffer to generate an audio frame. When the multimedia encoder reads data from the audio bitstream buffer, it increases the audio bitstream byte counter value and, if there is subtitle synchronization information synchronized with the audio bitstream, the byte counter value of the currently output multimedia bitstream. Is assigned as new subtitle synchronization information to have subtitle synchronization information synchronized to the A / V bitstream.

9 shows a block diagram for playing a music video including A / V as an example of the present invention. In order to play the created A / V bitstream, the A / V bitstream is read into the file system and input into the multimedia bitstream buffer. The multimedia bitstream is divided into video and audio. When data is input to the playback device, the byte counter value is increased. If the byte counter value and the byte count value allocated as the subtitle synchronization information are the same, the subtitle is colored.

In order to listen to only the singer's voice, listen to the accompaniment music, and play the lyrics in advance, using the above-described reproduction control method of the present invention, it is possible to easily implement a song cycle including images and subtitles.

Even though the audio bitstream and the video composed of a single file with subtitle sync information and subtitles consist of A / V bitstreams such as MPEG and DIVX, the video and audio are separated in the multimedia player, and the video and audio are played in synchronization. By using the bitstream byte counter, the subtitle data synchronized with the audio bitstream can be colored or expressed on the screen. The playback control method of the present invention allows the song half-period and lyrics to be matched to the image of the music video. The periodic function can be performed.

In the description of the present invention, the R / L channel is fixed to the sound source, and the synchronization information allocation has also been described by a method using a byte counter. However, the part about R / L can be replaced with L / R, and the method of allocating synchronization information can include a method of including time information, frame information of compressed data, a method included in a frame, ID3 TAG information, and the like. .

As described above, the present invention generates the music and accompaniment music, lyrics sync information and lyrics information, video sync information and video information, lyrics voice information including singer voice as a single digital audio file, By using the sound source, you can listen to the sound source containing the singer's voice, sing a song cycle with chorus function, sing a song lyrics, color the lyrics of a song, and use the music video produced by the music label as the song record content. In other words, the music containing the singer's voice, accompaniment music, lyrics information, video information, and song lyrics voice data are integrated into one digital audio format (PCM, MP3, WMA, OGG, etc.). You can listen to the music that contains the music and song lyrics included separately, there is no damage to the sound quality when switching between the three sources, implement the chorus function using the sound source switching, lyrics subtitles can be colored In addition, the present invention provides a method and device that is convenient for integrated management of sound sources, and that when a generated sound source is inserted into a music video, the music video can be used as a song cycle.

1 is a block diagram showing a production device and a playback device of the present invention;

2 is a state diagram showing a process of generating a sound source through the generating device

3 is a diagram illustrating allocating still images, moving images, subtitles, and the like to an audio bitstream through a synchronization information allocator;

4 is a state diagram for explaining the lyrics pre-speech voice insertion interval

5 is a graph showing a phase filter for making MONO music in stereo

6 is a system diagram for obtaining an HRTF used in three-dimensional stereo sound

7 is a diagram illustrating a compressed A / V bitstream constituting a general music video.

8 is an encoder block diagram for generating a compressed video bitstream using subtitle synchronization information.

9 is a block diagram illustrating an example in which the present invention is applied to a music video including A / V.

Claims (10)

In a typical sound source generation and playback method, after inserting synchronization information for a subtitle or video into one digital audio bitstream file, a sound source is generated and played using the synchronization information to generate general music, accompaniment music, and lyrics. How to create and play back music, which can play general music, accompaniment music, pre-loading lyrics, chorus function, and music video as one file. The audio bitstream file of claim 1, further comprising: a first step of analyzing a sound source input through sound source analysis and a channel separator, and then separating each sound source for each channel; A second step of mixing channel data through a sound source channel mixer to generate a new audio bitstream (RO, LO); A third step of allocating start synchronization information of the caption and video through the synchronization information allocator; Inserting a pre-singing song lyrics function through the lyrics voice inserter; Generating a mono channel to a stereo channel through a channel generator; A sixth step of controlling a sound source through a channel controller and performing a function of listening to music and pre-song lyrics; A seventh step of separating the synchronization information of the caption and the image through the synchronization information separator; After separating the video and audio with a multimedia player, the eighth step of the synchronous playback; it is created and played through a single file characterized in that the general music, accompaniment music, pre-loading lyrics, chorus function, music video production How to create and play sound. The general music and accompaniment music as one file according to claim 2, wherein in the fifth step of generating a channel, a spatial filter used in a phase filter or a 3D stereo sound is adopted to increase the spatial feeling corresponding to the stereo output. How to create and play back music, pre-load lyrics, chorus function, and music video production. 3. The method according to claim 2, wherein the first step always has a structure that precedes the second step, and the third step always has a structure that precedes the fourth step, wherein the performing steps of the first to fourth steps are variable. How to create and play back music with general music, accompaniment music, lyrics pre-loading, chorus function, and music video as one file. The method of claim 2, wherein the first to fourth steps of the generating method are performed in the order of the first step, the third step, the second step, and the fourth step. How to create and play back music with pre-loading lyrics, chorus function, and music video production. The method of claim 2, wherein the first to fourth steps are generated in the order of the first step, the third step, the fourth step, and the second step. How to create and play back music with pre-loading lyrics, chorus function, and music video production. The method of claim 2, wherein the first to fourth steps are generated in the order of the third step, the first step, the second step, and the fourth step. How to create and play back music with pre-loading lyrics, chorus function, and music video production. The method as claimed in claim 2, wherein the first to fourth steps are generated in the order of the third, fourth, first, and second steps. How to create and play back music with pre-loading lyrics, chorus function, and music video production. The method as claimed in claim 2, wherein the third step is omitted when it is not necessary to allocate the synchronization information of the subtitles and the video, and the fourth step is omitted when the song lyrics preload function does not exist. How to create and play back music that can play general music, accompaniment, lyrics pre-loading, chorus function, and music video as a file. A conventional sound source generating and reproducing apparatus, comprising: a generating device for generating a sound source and a reproducing device for reproducing the generated sound source, wherein the generating device comprises: a sound source analysis and channel separator for separating input channels after analyzing an input sound source; A sound source channel mixer for mixing the channel of the input source; A synchronization information allocator for allocating start synchronization information of the caption or video; And a lyrics voice inserter for pre-singing the song lyrics. The playback apparatus includes: a channel generator for generating a stereo channel; A channel controller for controlling the generated sound source; A sync information separator for separating the sync information of the caption and video; Multimedia player for synchronous playback of video and audio; Forms a single file, characterized in that the general music, accompaniment music, lyrics pre-loading, chorus function, the sound source generation and playback device capable of producing a music video.