TWI227851B - A method for synchronizing audio and video - Google Patents

Abstract

The present invention is in related to a method for synchronizing audio and video, wherein played audio information and video information being assembled by a plurality of audio information segments and video information segments. Before playing, the audio information segments and the video information segments may be processed and stored in an audio information buffer zone and a video information buffer zone temporarily and individually; a relative location of processed audio information segments is a played location at the moment and is picked up in the audio information segments during the processing before playing. The method for synchronizing audio and video is firstly to record an initial writing time of the audio information segments of the audio information buffer zone; secondly to accumulate a total writing amount of the audio information segments of the audio information buffer zone; thirdly to compile a total rest amount of the audio information segments, not played yet, in the audio information buffer zone. The total rest amount is equal to that the total writing amount subtracts a product of a sampling frequency and a passing time. The sampling frequency is defined as a information amount of played video information segments per time unit, and the passing time is described as a difference of a present time subtracting the initial writing time; a real played location is then a difference of a present played location subtracting the total rest amount; at last, a playing of the video information segments is capable of adjusting according to the real played location.

Description

1227851

[Technical field to which the invention belongs] The present invention relates to a computer-readable recording medium for a method and recording method for synchronizing sound and video, and particularly to a method for adjusting image data by estimating the actual playback position of the figure A method for synchronizing the played sound and video and a computer-readable recording medium recorded with the method. In today's computer applications, multimedia is undoubtedly one of the hottest and most important fields. With rich multimedia image and sound effects, the inner valley that you want to spread can be vividly presented, and it is easy to deeply read and listen to the public. In my mind, when some multimedia materials such as sound and video are present, the sound data in the data buffer, and then Converter], but at this time, the slow synchronization of the apparent synchronization of the sound-induced video output is to solve the data. The mid-range data of the time clip is digitally output through the data buffer. As mentioned above, when playing multimedia materials in an operating environment, synchronization sometimes occurs. This situation is particularly serious when playing with subtitles. The reason for this is that after pre-processing such as decompressing the playback audio data, it is first stored in a slow analog converter (DAC, Digital to Analog 5), and some multimedia player programs are used to calculate how much data is transmitted to the buffer. It may be that there are remaining data fragments that have not been output, resulting in a low estimate of the real car parent multimedia player program, and the output of the guide sound is also fast. Therefore, if there is no sound or video, it is correct to estimate how much sound is actually output. This method is based on the previous system

1227851 V. Description of the invention (2): C: I:: t 自动 Automatically call after the end-function refers to #, I * in the user calculates & two 'to his own program function, and in this program function-returns the length The length of the function: This is the length of the data. The second is the amount of data provided by the system. The number S can be left in the buffer when the call is made. It is difficult to implement it on your own, ^ ^ Brother, under the system development side, improve sound and video J; system program problems. "The situation where the method is pacing is really a question that needs to be resolved [the invention refers to the specific letter sounds and shadows as the method of expression, in which the plurality of pen sounds and sound data are not temporarily stored, which represents the location of the paragraph obtained in the paragraph First of all] The above-mentioned questions ca n’t be used as one of the audio data fragments and sound processing items in the above project. If the situation is synchronized under the environment of the industry, the audio material fragments and the complex image data buffer the sound data. On January 1J, the audio and video data of the playback position and the image are clear. "No need to form Fang Ming to provide material and shadows. Several video segments are placed in the area and the fragments are placed in the same area to provide a systematic method for changing the system. A sound and The image data is a segmented data segment. Before playing, it is necessary to pass an image data. For the preprocessing of sound data, the recorded sound data is first planted in a non-formable image. The provision of recorded audio clips is enough to improve the synchronization of the corresponding party. The above data is divided into domains. Page 10 1227851 V. Description of the invention (3) Writing time, the difference between the amount of written data in one of the data segments in the sound data segment in the area, and the difference between the amount of data taken, and then the current playback Inter-position playback position adjustment ^ In addition, this π has been recorded ~ sound and, the 3 sound data pieces f and clips played ~ ~ sounds are processed by the sound information table is currently known. This sound $ '先Written sound: time, followed by sound data: written in 4 clips; one of the bucket clips total cumulative sound data piece data piece written sound piece one total written data buffer total remaining data amount, divided by one When the sampling frequency and the same frequency are per unit, the elapsed time is calculated by deducting the total video image and the audio data segment of the video and the composite data. The buffered audio data playback bit and the video and audio data are accumulated. An actual synchronization method is provided for the remaining amount of data in the actual playback position. Data and image data. Several image data clips are placed in the playback area and a video segment. One of the audio data segments in the buffer area of the synchronization method is written into the data buffer volume, and the above-mentioned elapsed time in the data area in the segment is to be broadcasted to the current position, and then played back. Brain-readable use of this material system Before segmenting the clip, it is necessary to process the audio data of the video data first. The total written audio volume in the audio data field is first recorded, and then the remaining audio data between the two remaining data is not broadcasted. The difference between the broadcast sound and the best recorded sound is composed of the relative. The pre-processing buffer area material fragments are recorded by the sound data and the sound material fragments are included in the sound data. Then the remaining data is buffered and the sound is calculated. The material volume is the total product. When the data segment starts to be written, the playback position is based on the real media. The synchronization of the image should be the plural of the above sounds and temporarily suspended. Instead, the data segment in the relative position material segment starts the writing buffer. The volume of the sound data of the district calculation is the total write

1227851 V. Description of the invention (4) The difference between the amount of input data and the amount of sampled data, and then the pre-calculated playback position is deducted from the playback position. Because the actual broadcast of this data provides a specific function and the same In the case of the operating environment, the frequency of the sampling is set to the frequency of the playback of the clip and the image after each amount of data in the playback position. The overall image data needs to be changed. It is resolved by a specific background. The time is a current calculation and the actual number of sound placement position adjustments except for the total remaining image data. Therefore, the user-side problem 0 is the product of the two. The deduction of the initial writing time and the actual playback position of the audio data segment is the difference. Finally, the sound is estimated by the actual method, so there is no need for the system side. The limitation of the sound of invention, but applicable to the end does not provide a specific function, the sound is different from the image [Implementation] The following will refer to the accompanying diagram, the second method of synchronization with the image and the computer-readable 4 J preferred embodiment Voice, Wu: The media will be explained with the same reference symbols. It is the same as shown in FIG. 1. According to the present invention, rt ′, which is used to process multimedia data during playback, is a method of synchronizing video: Compression of multimedia data including both sound and shadow damage = synchronization. Format is used to store image negative materials, chanting in η, and sound and image data can be pre-processed such as unpacking and shrinking, corresponding to the amount of data output by the time of multiple pen sounds; for body soap and image data Incision] Feng Xiang _____ 乃 # Also. Page 121227851 after finishing the processing. V. Description of the invention (5) Before the actual clip is played, the data will be stored in a digital analog converter (DAC, ❿): output, punch " ', and then the sound data will be rotated out. Movie f and Anai 0g Converter) ^ film slave and video data clips displayed on the screen. The requirements of the present invention do not stand without curling up-this sound is not a clear synchronization method. The image data is also applicable to this issue. Please refer to Figure 2. The sound of the present invention is left in the audio data output buffer; indirectly, the actual output audio data is indirectly estimated. # 备 贝 料The thickness of the sound data in the buffer is 1 = Beckory 'to avoid the error caused by the amount of data transmitted: the sound data of the international rotation. The sound and video of the present invention are written with the effect of sound and video synchronization The sound data segment (S2〇1) includes: determine whether the beginning of the sound π output buffer is written u: day :: segment. Write the sound into the sound shell output buffer. Parent = write (s2〇3), Determine whether to pause, skip, or replay; 2. Sound data in the gauge that has not yet been played *; | ”A sound data output buffer, calculate the actual playback position (= segment = remaining data amount (㈣)) Play (S20 8), play two: Intermediate adjustment of the unplayed sound f J, the input and output of the sound data roll out to zero (S210), judgment;; ==), write the total Information and playing the written sound data sound material ... S2U) Ye Houchong is a medium and still Unplayed sound data page 131227851 V. Description of the invention (6) Niu Gu, i 2 1 2), the following will explain in detail the steps of the V method of the voice of the present invention. = ， Continued in the process, θ Tong Jin 钚 'earlier said the 枓 枓 fragment, and then step S202. Then ,,, ° Bunben, Gua Cheng, if it is, then continue to step S202, the next position will be sighed μ " The initial writing of the output buffer = time? 22 $ First write the data of the sound data output by the human voice data. The steps of ρ and 〃 彳 are to estimate the actual elapsed time of the above initial writing time ct. Among them, step _3, the time. The total written data of the audio data segment is two, and the value in the ^ 2 material wheel out buffer is zero. With the audio data piece ρ ^, the initial accumulated amount of written data. 'Writing again, the total amount of written data will also follow / thinker, the total amount of written data and the amount of sound data in the sound data segments that are roughly recognized by the sound data are not M = = now in the rotation buffer t: Reduced by the digital analog converter output :: each of the audio data sheet output buffer In addition, Shengli 'writes the same amount of data as the sound information, but even if the sound information is different: the situation is different without affecting the implementation of the present invention. The amount of data here will be judged in step 4; ; No Pause Yes, execute step S2 09 'Play the written sound immediately, and replay, if ί = $ 的 之 音 数据 段, and then execute the puncher and still ... return to zero, in order to temporarily #, jump or Heavy; =, the starting point of writing the total target is

IM

Page 14 1227851 V. Description of the invention (7) Yes, re-accumulate the total amount of written data written in the sound data wheel, and then execute the judgment result in step S204 is no,

At step S205, it will be judged at this time that the line of step S211 is judged at this time, and it is judged whether there is still sound ^, then go back to step S203, and continue to accumulate the total amount of written data when the sound is out of the buffer 7 Play the written sound data buffer == Then end this process. If it is judged in step S205 that 'the following calculation sound I' is performed: 枓 η & 々 & Xiao Yue Bei's step out of the total remaining data amount of the relief film _ = step 82 () 6, will be calculated The total remaining of the i G broadcast sound data clips is the long time that has passed since the calculation of the difference, and the time recorded in step S202 is deducted: is the system time; secondly, "= the actual broadcast of the film in the past time; the sample frequency is every The unit time is ^ 2 and 4; the final deduction of the step within the elapsed time is the writing of the sound data. 仏 山 '际 播 出 枓 The total remaining data in the output buffer crying. In step S2Q7, the Calculate the audio data segments in the actual playback buffer. Step 3201 and subsequent steps. I Continue to step S20 5. If no image is displayed, * No, execute the data segment and wait for rotation. , Then step S21 2 is performed, and the unplayed sound data segment 疋 ′ is then continued to step S 20 6 the unplayed sound data in the puncher, the remaining data of the audio data, and the sound that has passed the time and time elapses, and S20 3 tired sound will be played. And not yet lost The amount of time is calculated by multiplying it by the sampled data fragments. The calculation method is based on the current frequency above the present frequency, that is, the amount of sound data. The position of the total volume of data and audio data, which can be played by the current

Page 15 1227851 V. Description of the invention (8) __ This position is obtained after deducting the total remaining data. It represents the most similar sound data segment between the pre-processed sound data and the sound playback position. Compared with the sound and the current time, the sound is very expensive. Therefore, the error of the sound that is not played at the current playback position is in the sound data output buffer. Poor 'amount of data. ^ Sound Expansion

For example, please refer to Figure 3. Assume that five strokes, B, C, D, E and other five-tone sound buttons include the definition of A, and the current playback position is c, D: sound :::: Because there are still two audio data segments such as B and c temporarily stored in the sound stand, but not yet played in the device, the actual playback, ::, buffer and the actual playback position error. Mamu's playing position is when the player plays multimedia data and jumps or replays. Actually: there is no pause, skip directly to the sampling frequency, but at; 2 can be multiplied by the elapsed time and played. There is a pause Ge ^ 新 播 二 # When the situation is not played from the beginning or obtained according to the above method, 'the image data of the video data segment must still be adjusted according to the position', then the video segment corresponding to the 4 material is added before Set processing and output. After completing this step, more material

1227851 V. Description of the invention (9) S211, to determine whether there is still sound in this process or to execute the above loop. Because this method is pre-written at the time, and when the sound data segment in the streamer is played, it can be broadcast based on the elapsed time. The remaining positions in the sound material output buffer remaining in the sound data piece output buffer are corrected to the actual playback of the bit image data piece. In addition, the invention is a function that should be determined, and the synchronization method that does not need to be changed is not subject to specific Operating environment. So it completely solves the problem that it is difficult for the user to implement it by themselves. Step S203 of the audio data piece. The total volume of data and sampling frequency data accumulated in the audio data process is set to the audio data. Finally, the sound and the program-side system are used to solve the problem. To step S208, the volume of audio data has been written at the beginning of playback, and the data amount is calculated indirectly at the rate of 'calculation'. The amount of data entered can be based on the implementation of the actual broadcast image synchronization. Provide specific sounds and images to determine the process, record data to calculate the sound when the video is displayed, and use it to position the effect. The sound on the system side is used for the asynchronous synchronization of all functions. The end will start. The next buffer will be output. The image data will be real audio data. The audio data will be played before.

1227851 Schematic illustration [Schematic description] Figure 1 shows the schematic diagram of the multimedia data output process; Figure 2 shows the flowchart of the method for synchronizing the sound and video of the present invention; Figure 3 shows the current playback position and Schematic diagram of the actual playback position error. [Symbol description] S201 Determines whether to write the audio data segment S20 2 Records the audio data segment first writes the start time of the audio data output buffer

5203 Accumulated total written data volume of sound data fragments written in the sound data output buffer 5204 Determine whether to pause, skip, or replay S20 5 Determine whether to display the image S20 6 Calculate the data written to the sound data output buffer but not yet Total remaining data volume of the audio data segment being played S20 7 Calculate the actual playback position S208 Adjust the playback of the video data segment according to the actual playback position

S20 9 Play the audio data fragments that have been written to the audio data output buffer but have not yet played 5210 Reset the total written data amount to 5211 Determine whether there are still audio data fragments 5212 Play the audio data buffer that has been written to the audio data buffer but has not yet played Audio clip

Page 18

Claims (1)

1227851 VI. Scope of patent application1. A sound and video data, such as multiple strokes, and other video data. The relative position of the audio data sheet records the sound in the area. The sound in the domain is accumulated. The sound in the domain is calculated. The rate and the time-lapse rate are the size of each order. The same material with the image is obtained from the audio data of the audio data in the buffer area of the broadcast data segment. The audio data is the audio data segment. Wait until the sound data is not played. The remaining data volume is over time. Both time is to be played. The elapsed time is #. 'The plural of them, these sounds are pre-processed by the image data. For these positions, they can be written at the same time as the first write of the image. The sound information obtained after writing the data for the total product is the current time minus the sound information data that is played. And divide the area of the sound resources in front of the sound method; the sound resources; the total remaining amount of sound resources minus the difference, and the material fragments should be stored in the beginning of the sound data material fragments and fragments, and substitute materials The rationale of the clip includes: data buffer data buffer [^ r data buffer data amount, a sampling frequency of the data amount of the sampling frequency writing time meaning is very consistent consistent playback position, where the actual playback position is the playback The difference between the position after deducting the total remaining data amount; adjust the image data fragments according to the actual playback position with a mesh. Heart playback as described in item 1 of the scope of patent application Page 19 1227851 VI. Scope of Patent Application ____〉 The audio data buffer area is audio play buffer. The poor output buffers stolen; the synchronizing side of the sound and the video described by ir * 4 ": ί time is taken from the system time. 4. As described in item 2 of the patent scope, where the current time is taken from the system time: 、 -Image step method 5. As described in item i of the scope of the patent application, it also includes the synchronization method to play these sound data II /, shirt image synchronization buffer area and has not yet played the sound to write the sound data 6. If the scope of the patent application is the fifth item, it is included in the sound data [^ shirt image synchronization method and has not yet played the sound film ;: the sound data is later put in the sound data fragment first yuan After the completion, the voice data buffer of the voice data segment of the domain is returned, and the total written data volume is returned to the gate 6 and the initial writing time 7. If the patent application scope item 丄 heart sound 'more Contains a method of synchronizing the erasing of such image data / already = 8. The two persons in the buffer area who have not yet played the scene image data can be read by the brain = note: = synchronization method, using this sound Audio The sound clips played by the same ν method as the image data are in contrast to the two pieces of image data. The plurality of sound data pieces and these image data pieces are temporarily stored in the-sound data buffer. ::;: = = = 6. The scope of the patent application, and the photo of the audio data segment that is being processed is indicated as to which one is currently playing: the material segment is obtained from the concentric record of the sound and video. The audio data segment is first written into the audio data area. The initial writing time of one of the sound data fragments; buffering and accumulating the total amount of written data of one of the sound data fragments in the sound data buffer domain that has been written into the sound data fragments; οσ Calculating the The total remaining data amount of one of the sound data fragments written into the sound data buffer area but not yet played, wherein the total remaining data amount is obtained by subtracting the product of two sampling frequencies and one from the total written data amount and the time two Difference, and the sampling frequency is the number of the audio data segments to be broadcasted every t-bit time, the time, hour = is the current time minus the initial writing time; calculate the actual playback position, where the The actual playback position is the difference between the current playback position and the total remaining data amount; and adjusting the video data based on the actual playback position The playback period of 9.3: ΐ: 8 of the computer readable recording medium of: square to: a sound data region round punch-out buffers (audio Play buffer). Mountain and J port 10. The computer-readable recording medium as described in item 8, the writing time is taken from the system time. η · A computer-readable recording medium as described in Item 8 on 227851 Page 22