JPH06339129A - Time deviation adjustment device between video signal and audio signal - Google Patents

Abstract

PURPOSE:To automate time deviation adjustment of a video signal and an audio signal in a television broadcast network by detecting a time deviation between the video signal and the audio signal so as to adjust the time deviation between the video signal and the audio signal based on the detection. CONSTITUTION:When an audio signal sent by beeing superimposed on a video signal sent via a video transmission line 9A is delayed by a time difference TD from the audio signal sent via an audio transmission line 2A, time data 23A representing an obtained time deviation TD are outputted from an arithmetic operation circuit 23 of a measured reception section R and inputted to a voice signal delay circuit 25. A voice signal 12A inputted from a voice signal reception circuit 12 is delayed based on time data 23A in a voice signal delay circuit 25 and the delayed voice signal 25A is outputted. The audio signal 25A obtained in this way and the video signal 19A outputted from the video signal reception circuit 19 are outputted as a broadcast television signal from output terminals 26, 27.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a device for automatically adjusting a time difference between a video signal and an audio signal in a television broadcasting network.

[0002]

2. Description of the Related Art Generally, a television broadcasting system includes a "material source" that supplies a television signal including a video signal and an audio signal by picking up an image with a television camera and collecting a sound with a microphone, and a television based on the supplied television signal. It is composed of a "television broadcasting station" that broadcasts. A dedicated transmission line is provided between the source of material and the broadcasting station, and the TV signal is transmitted through it. However, the video signal and the audio signal of the TV signal may be transmitted using separate transmission lines. .

Generally, the synchronization signals of the material source and the television broadcasting station are not the same, but the frequencies and phases are slightly different. Therefore, in the broadcasting station, the video signal sent from the material source is converted into a digital signal by A / D conversion and temporarily recorded in the memory. Then, the recorded digital video signal is read out by using a synchronization signal of a broadcasting station which is defined in advance, and is D / A converted to be converted into a video signal synchronized with the synchronization signal of the broadcasting station. A device for performing the above conversion is known as a frame synchronizer (product name frame synchronizer), and this conversion causes a delay in a video signal. Further, when a digital transmission line is used for voice transmission, a delay may occur due to conversion into a digital signal as is well known. As a result, a time lag occurs between the video signal and the audio signal. If the television signals of the video signal and the audio signal, which have the time lag, are broadcast as they are, the time lag is caused between the video and the voice at the time of reception, and the viewer feels uncomfortable. Therefore, the television broadcasting station monitors the transmitted video signal and audio signal and delays the signal (generally an audio signal) that is ahead in time by a delay circuit so that the time lag between the two is inconspicuous. I am adjusting. This adjustment is performed by the operator by manually changing the delay time of the delay circuit.

[0004]

Generally, the material source is not constant, and the material source may change for each program. When the material source is changed, the transmission path also changes, so the time lag between the video signal and the audio signal also changes. In sports broadcasting and the like, television signals are transmitted from a relay vehicle dispatched to the stadium to a broadcasting station. The frequency of the synchronizing signal in the image signal of the relay vehicle is generally slightly different from that of the broadcasting station, and also varies with time. Therefore, since the time lag between video and audio is not constant but constantly fluctuates, it is extremely complicated to manually adjust the time lag and it is difficult to accurately adjust the time lag. Further, there is no measuring means for measuring the amount of time lag during broadcasting.

[0005]

SUMMARY OF THE INVENTION A time shift adjusting device between a video signal and an audio signal of the present invention converts a part of the audio signal of a television signal having the video signal and the audio signal into a first signal, and A video signal having a superimposing means for superimposing the video signal on a predetermined scanning line is transmitted by the transmitting means, and the video signal transmitted by the transmitting means is received by the reproducing means. , The first signal superimposed on the scanning line is reproduced. Further, the audio signal is transmitted by a transmission means for transmitting the audio signal, a part of the transmitted audio signal is converted into a second signal, and a video signal and an audio signal are generated based on the first signal and the second signal. An adjusting unit is provided for detecting a time lag between the video signals and the audio signal based on the detection by the detecting unit.

[0006]

A first signal produced based on a part of the audio signal of the original television signal inserted and superposed in the blanking period of the scanning line of the video signal is a video transmission system including a video memory together with the video signal. Is transmitted via. On the other hand, the audio signal is transmitted via an audio transmission line different from the video signal. By the second signal created based on a part of the audio signal that is inserted during the blanking period of the video signal and sent through the video transmission system and the audio signal transmission system created based on the original signal The shift between the corresponding content parts of the sent first signal is equal to the time shift between the transmitted video and audio signals.

[0007]

1 and 5 are block diagrams of an embodiment of an apparatus for adjusting a time difference between a video signal and an audio signal according to the present invention. In both figures, this time shift adjusting device is composed of a measurement transmitting unit K provided at the material source and a measurement receiving unit R provided at the broadcasting station. An audio signal is input to the terminal 1 and a video signal is input to the terminal 7 in the measurement transmitting unit K.
The audio signal input to the terminal 1 is sent to the audio transmission line 2A from the terminal 2 and also input to the detection circuit 3. The detection circuit 3 has a cutoff frequency of, for example, 30 Hz.
The low-pass filter LPF of FIG. 2 is provided, and the voice signal is envelope-detected to obtain the detection signal 3A shown in FIG. The detection signal 3A is input to the sampling circuit 4 and is sampled at a sampling frequency of 60 Hz synchronized with the vertical synchronizing signal output from the sampling signal generation circuit 10. That is, sampling is performed once per field. FIG. 3 shows a sample signal 4A composed of a pulse train obtained by sampling. Sample signal 4
A is input to the audio level pulse width conversion circuit 5.

In the audio level pulse width conversion circuit 5,
The level of the sample signal 4A is converted into a pulse signal (first signal) having a time width corresponding to the level (that is, PWM modulation). This pulse signal is referred to as "voice level pulse signal" 5A, and its waveform is shown in FIG. In FIG. 4, the time width T corresponds to the level of the sample signal 4A shown in FIG. 3, and the time width T changes corresponding to the instantaneous value level of the audio signal. The voice level pulse signal 5A is input to the VITS inserter 6.
The VITS inserter 6 is a well-known device that is generally commercially available, and is a device for adding a desired signal to a scanning line during a vertical blanking period of a video signal input from the terminal 7. In this embodiment, the audio level pulse signal 5A is added to the 21st scanning line 21H and the 284th scanning line 284H of the video signal. Scan lines 21H and 2
84H is not visible to the viewer because it is outside the range of the display surface in a home television receiver. Voice level pulse signal 5A
The video signal 6A to which is added passes through the video signal transmission circuit 8 and is converted as a video signal 7A from the terminal 9 to the video transmission line 9A.
Sent to.

In this way, the audio signal transmitted via the transmission line 2A and the video signal transmitted via the transmission line 9A are received by the measurement receiving unit R shown in FIG. Since a transmission system for transmitting a video signal generally includes a video signal memory in a relay station or the like in addition to a transmission line, a considerable delay occurs due to input / output to the normal memory.

In the measurement receiving section R, the voice signal is input to the voice signal receiving circuit 12 and amplified to a predetermined level. The amplified audio signal 12A is input to the detection circuit 13 and also to the audio signal delay circuit 25. The audio signal delay circuit 25 is a known device that is generally commercially available. The configurations and operations of the detection circuit 13, the sampling circuit 14, and the audio level pulse width conversion circuit 15 are the same as those of the detection circuit 3, the sampling circuit 4, and the audio level pulse width conversion circuit 5 shown in FIG. 1, respectively. That is, the voice signal is envelope-detected by the detection circuit 13, and the detection signal is input to the sampling circuit 14.

In the sampling circuit 14, the audio signal is synchronized with the horizontal synchronizing signal and is counted from the position of the scanning line 21H to start scanning the first horizontal scanning line at each position which divides the screen of one field into 16 equal parts. Is sampled at the timing of. This horizontal synchronizing signal is obtained from the video signal input to the video signal receiving circuit 19 from the terminal 20 via the transmission line 9A. The sampling frequency in this case is 16 times the frequency of the sampling signal applied to the sampling circuit 4. The sample signal 14A composed of the pulse train obtained by sampling is converted by the audio level pulse width conversion circuit 15 into a pulse signal (second signal) having a time width corresponding to the level of the sample signal 14A as in FIG. (Ie PW
M modulated). This pulse signal is the voice level pulse signal 15A, which has a waveform similar to that of the voice level signal 5A shown in FIG. 4 and is temporally shifted.

The voice level pulse signal 15A is input to the A / D converter 16 and converted into digital value time width data Ta. The time width data Ta converted into a digital value is stored in the memory 21 of the arithmetic unit 23.

On the other hand, the video signal received by the video signal receiving circuit 19 is input to the audio level pulse signal detecting circuit 18 and the audio level pulse signal 5A shown in FIG. To be done. The audio level pulse signal 5A is input to the A / D converter 17 and converted into digital time width data Tb. The converted time width data Tb is stored in the memory 22 of the arithmetic unit 23. The time width data Ta is data obtained at a sampling cycle of 1 / (60 × 16) seconds, and the time width data T
b is data obtained at a sampling period of 1/60 seconds. And that value corresponds to the level of the audio signal. The time width data Ta and Tb stored in the memory are stored in the CPU 2
4 and the arithmetic processing described below is performed.

Memory 21 controlled by CPU 24
After storing 1600 and 100 continuous time width data Ta and Tb, respectively, 22 and 22 suspend the memory operation and hold the data. The 100 audio level pulse signals correspond to 100 fields of the video signal, and the time is about 1.66 seconds. Time width data T
The values a and Tb are read by the CPU 24 and converted into level data representing levels corresponding to respective time widths.
At this time, the time width data Ta is the scanning lines 21H and 284.
Only the data sampled by the sampling signal synchronized with H is read. That is, 16 consecutive
Of the 00 time width data Ta, 1 out of 16 will be read. The 15 pieces of data that are not read at this time are used in the processing described later.

The level data group LA [i] (i: 1 to 100) for the time width data Ta is LA [1], LA
[2] ,. ． ． , LA [100], the level data group LB [j] for the time width data Tb is represented by LB [1], LB.
[2] ,. ． ． , LB [100]. Level data LA
[1] ,. ． ． , LA [100] and LB [1] ,. ． ． , LB [10
The level of [0] is displayed as an envelope on the same graph as shown in FIG. In the figure, the horizontal axis indicates each level data LA
Field number [i] or [j] of [i], LB [j]
Is represented. A curve 30 is an envelope based on the data of the time width Ta, which represents a temporal change in the level of the audio signal transmitted via the audio transmission line 2A. Further, a curve 31 is an envelope based on the data of the time width Tb, which is superimposed on the scanning line of the image and transmitted by the transmission line 9 of the image.
3 shows a temporal change of a voice level pulse signal transmitted via A. Curves 30 and 31 represent a state in which voice signals having substantially the same instantaneous value shape are temporally shifted due to the difference in the transmission system. In the example shown in FIG. 6, the audio signal (curve 31) transmitted by being superimposed on the video signal transmitted via the video transmission line 9A is transmitted through the audio transmission line 2A (curve 30). It indicates that the time difference TD is delayed.

Next, the processing for obtaining the time difference TD will be described. Experience shows that the time lag between video and audio signals usually does not exceed 0.3 seconds.
The number of data processed in 24 is set. Therefore, the following processing is performed using 60 pieces of data corresponding to 60 fields from each of 100 pieces of level data groups LA [i] and LB [j]. As a method of selecting the 60 data, for example, level data LA [1], LA
[2] ,. ． ． , Level data LA [20] in LA [100],
LA [21] ,. ． ． , LA [80] may be used. As preprocessing of the calculation, a calculation for making the levels of the level data groups LA [i] and LB [j] the same is performed, but since this calculation is well known, its explanation is omitted. The CPU 24 uses the following formula (1)
Perform the operation performed in.

[0017]

[Equation 1]

In the equation (1), [k] is a number in the range of 1 to 40, and the numerical value designates the level data LA [i] at the starting point for starting the calculation of the equation (1). Therefore, the numerical value [k + j] corresponds to the field number [i] of the level data LA [i]. The value LC [k] obtained by this equation is called the sum. This sum represents the sum of the absolute values of the differences between the level data groups LA [k + j] and LB [19 + j] in the range of j = 1 to j = 60. That is, the expression (1) in FIG. 6 finds the level difference between the curves 30 and 31 for each field number within the range of field numbers 20 to 80,
This shows that the sum of the differences of the obtained 60 levels is calculated.

This total LC [k] corresponds to the area of the region Z surrounded by the curves 30 and 31. This area Z is curve 3
This occurs because 0 and 31 are shifted by the time TD, and when the curves 30 and 31 are not time-shifted and both match, the region Z does not occur and the sum LC [k] becomes zero (actually Since there is noise in the transmission line of, curves 30 and 31
The shapes are not exactly the same. Therefore, the sum LC [k] does not become zero but becomes a predetermined minimum value. Also, the deviation time TD
When the number is small, the area of the region Z is also small. Therefore, the sum L
For example, when the curve 30 is shifted rightward in the figure by one field number so that C [k] becomes the minimum value and the total sum LC [k] is calculated each time, the total sum LC [k] becomes minimum. It is possible to determine that the curve 30 matches the curve 31.

The above calculation is performed by the CPU 24, and its process is shown in the flow chart of FIG. Figure 7
In step 50, [k] is initialized to “1”,
In step 51, the sum LC is initialized to "0" and [j] is initialized to "1". At step 52, the level data group LA [k
The difference AK [j] between + j] and LB [19 + j] is calculated. Next, in step 53, it is determined whether the value of the difference AK [j] is positive or negative, and if positive, the process skips to step 55. If it is negative, the constant "-1" is multiplied in step 54 to obtain a positive value (absolute value calculation).

In step 55, the values of the difference AK [j] are sequentially added to obtain the total LC. In the judgment step 56, the number of [j] is counted, the value of [j] is increased in step 57 until [j] reaches 60, and the processing from steps 52 to 55 is repeated. When the value of [j] reaches 60, the process proceeds to step 58, and the sum LC obtained in the above step
Is stored as the sum LC [k] for the value [K].
Next, in step 59, the value of [k] is determined, and the value is sequentially increased in step 60 until the value reaches 40, and the processes in steps 51 to 58 are performed. In step 60, changing the value of [k] means changing the field number of the curve 30 in FIG. 6, thereby changing the field number of the curve 30 corresponding to a certain field number of the curve 31. . That is, the minuend LA in the subtraction operation performed in step 52.
[k + j] changes. This means translating the curve 30 to the right or left in FIG. In step 60, since the value of [k] is increased, the curve 30 is translated in the right direction.

In step 59, the value of [k] is changed from 1 to 4
Although it is changed to 0, this is 100 level data LB [1] ,. ． ． , LB [100] of 60 level data LB [20] ,. ． ． , LB [80] are used as the total sum, the change range of the number of level data LA [i] is LA [1] ,. ． ． , LA [20] range and LA [8
0] ,. ． ． , LA [100], and the number is 40. As a result, in step 58, 40
Total sum LC [1]. ． ． LC [40] is obtained.

Next, in step 61, a sufficiently large numerical value CX (for example, a value of 2 bytes of 30000) is set, and [k] is set to "1". In step 62,
A comparison of the numerical value CX and the sum LC [k] is performed. When the sum LC [k] is smaller than the number CX, step 63
Then, the sum LC [k] is stored as a numerical value CX and the value of [k] is stored as a numerical value X. Next, in steps 64 and 65, the above-described magnitude comparison and the process of storing the smaller value as the numerical value CX are repeated until the value of [k] reaches 40. As a result, the numerical value CX is the sum LC
The value becomes equal to the minimum value of [k], and the value of "X" becomes the numerical value of [k] of the sum LC [k] of the minimum values.

Next, in step 66, the numerical value "20".
And the difference FX between the numerical value "X" and the numerical value "X". The difference FX represents the number of fields corresponding to the time shift TD between the curves 30 and 31 in FIG. The time for one field is (1 /
Since it is 60) seconds, the deviation time of both curves is (1/60)
× FX seconds. The shift time is a value obtained as a result of performing a process of shifting the curve 30 by a time corresponding to one field (hereinafter referred to as a field cycle) so as to match the curve 31. That is, since the curve 30 is displaced by the stepwise value of the field period, it does not always come to the position where it completely overlaps the curve 31. That is, the curves 30 and 31 may have a slight deviation. Therefore step 6
The minimum sum LC [k] obtained in 3 is not always the true minimum value. Therefore, next, a process for obtaining the minimum value of the total sum with higher accuracy is performed.

In the processing described below, only the two fields of the field of the field number X of the minimum sum LC [k] and the field next to the field are processed. The reason for this is that, of the fields before and after the field X for which the total sum LC [k] has become the minimum, a sufficiently high accuracy can be obtained by processing the subsequent fields. In this processing, 16 level data per field obtained in the sampling circuit 15 shown in FIG. 5 are used. This level data is
[m] ”. In this embodiment, as described above, 1600 level data LA [m] for 100 fields are stored in the memory 21, of which 32 level data LA [m] (m for the two fields are stored. : 1 、
2 ,. ． ． , 32) are used. The process of this process is substantially the same as that shown in the flowcharts of FIGS. 7 and 8. That is, in this process, the character "k" in each step of this flowchart is changed to "m". Further, in step 55, “j = 6
“0” is changed to “j = 2”, and “k = 40” in steps 59 and 64 is changed to “m = 32”.

By the above processing, the sum LC [m] can be obtained for every 1.04 milliseconds, which is obtained by dividing 1/60 second which is one field time into 16 times. As a result, the sum L
The minimum value of C [m] is obtained with high accuracy, and the match between the level data LA [m] and the level data LB [j] at the minimum value is detected with higher accuracy. That is, in FIG. 6, the time shift TD between the curves 30 and 31 is obtained with an accuracy of about 1 millisecond.

Time lag TD obtained by the above processing
Is output from the arithmetic circuit 23 of the measurement receiving unit R shown in FIG. 5 and input to the audio signal delay circuit 25. In the audio signal delay circuit 25, the audio signal 12A input from the audio signal receiving circuit 12 is delayed based on the time data 23A, and the delayed audio signal 25A
Is output. The audio signal delay circuit 25 may be, for example, a known circuit that is delayed by converting an audio signal into digital data, recording the digital data in a memory, and reading the recorded digital data at a timing different from that at the time of recording. . The audio signal 25A obtained as described above and the video signal 19A output from the video signal receiving circuit 19 are output as broadcast television signals from the respective output terminals 26 and 27.

Step 62 of the flow chart shown in FIG.
In the processes from 1 to 65, the total sum LC [k] has the minimum value even when the level change of the audio signal is small or the level thereof is extremely low (hereinafter, such a state is referred to as a silent state). However, the minimum value in this case is not obtained by the result of the calculation of steps 52 to 60 of the flowchart of FIG. 7, but is the result inevitably caused by the silent state. Therefore, in this case, step 66
If the audio signal delay circuit 25 is controlled by using the time data 23A based on the difference FX obtained in step 1, correct control will not be performed. Therefore, in this embodiment, the silent state is determined by the processing described below. In the silent state, the level of the audio signal is always constant or substantially zero, so the sum LC
[k] is the field number [k] as shown by the curve 32 in FIG.
It is always the minimum value regardless of. On the other hand, when it is not in the silent state, as shown by the curve 33 in FIG.
[k] has a minimum value in a certain field number [k], and increases gradually before and after the certain field number [k]. This determination processing is carried out by the CPU 24, and when it is determined that there is no sound, the time data 23A is not output. In this embodiment, the audio signal used is that of an actual television signal, so that its level is constantly changing. Therefore, the V-shaped shape of the curve 33 shown in FIG.
For example, it may be an open V-shape as shown by a dotted line 34. Therefore, in the CPU 24, for example, the curve 3
When the gradient of the V-shaped diagonal line portion of 3 is detected and the gradient is less than or equal to a predetermined value indicated by the straight line 35, time data 23A
Is controlled not to output. Such control is
For example, the field number [k] for which the sum LC [k] has the minimum value
After the above is obtained, the value of the field number [k] is increased or decreased by a predetermined value to obtain the value of the total LC [k] in both field numbers.

[0029]

According to the present invention, an audio signal of a television signal having a video signal and an audio signal is sampled and a first pulse signal having a pulse width corresponding to the level is superimposed on a predetermined scanning line of the video signal. Then, it is sent out through the video signal transmission circuit. On the other hand, the audio signal is transmitted through the audio transmission circuit, and the receiving side obtains the second pulse signal similar to the signal superimposed on the previous video signal. Then, by detecting the time lag between the first pulse signal and the second pulse signal, the time lag between the audio signal and the video signal can be automatically obtained, and the audio based on the data indicating the time lag. It can be automatically adjusted so that there is no time lag between the signal and the video signal. The above adjustment can be performed during the broadcast by directly using the sound in the television broadcast.

[Brief description of drawings]

FIG. 1 is a block diagram of a measurement transmission unit according to an embodiment of the present invention.

FIG. 2 is a waveform diagram of an envelope detection signal according to the embodiment of the present invention.

FIG. 3 is a waveform diagram of a sample signal in the embodiment of the present invention.

FIG. 4 is a waveform diagram of a voice level pulse signal in the embodiment of the invention.

FIG. 5 is a block diagram of a measurement receiving unit in the embodiment of the present invention.

FIG. 6 is a waveform diagram of a voice level signal in the embodiment of the invention.

FIG. 7 is a flowchart showing the operation of the embodiment of the present invention.

FIG. 8 is a flowchart showing the operation of the embodiment of the present invention.

FIG. 9 is a graph showing a control operation determination method according to an embodiment of the present invention.

[Explanation of symbols]

 2A transmission line 3 detection circuit 4 sampling circuit 5 voice level pulse width conversion circuit 6 VITS inserter 9A transmission line 13 detection circuit 14 sampling circuit 15 voice level pulse width conversion circuit 18 voice level pulse signal detection circuit 23 arithmetic unit 25 voice signal delay circuit

Claims (2)

[Claims] 1. A conversion means for converting a part of the audio signal of an original television signal having a video signal and an audio signal into a first signal, and superimposing the first signal on a predetermined scanning line of the video signal. Means, transmission means for transmitting the video signal on which the first signal is superimposed, detection means for receiving the video signal transmitted by the transmission means, and detecting the first signal on the scanning line, Transmitting means for transmitting an audio signal, receiving the transmitted audio signal, and transmitting a part of the audio signal to a second
Of the original television signal based on the first signal and the second signal, the detecting means for detecting a time lag between the audio signal of the original television signal and the transmitted audio signal; A time deviation adjusting device between the video signal and the audio signal, comprising an adjusting means for adjusting the time deviation between the video signal transmitted based on the detection and the audio signal transmitted. 2. The detecting means for obtaining a first data set formed by a predetermined number of first signals and a second data set formed by a second signal, Arithmetic means for determining the difference between the first data and the second data while changing the time relation between the first data aggregate and the second data aggregate, and determining the time relation for minimizing the difference. The time shift adjusting device between the video signal and the audio signal according to claim 1, further comprising a judging means.