JPH0993615A - Method for measuring time difference between video image and sound signal - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain an optimum time difference correction amount of a system by measuring quickly and precisely a time difference between a video image and a sound signal at a receiver end. SOLUTION: A transmission end test signal section 1 is made up of a video pattern generator 11 and an audio signal generator 12, and video and sound signal generating timings are strictly arranged. A receiving side test signal generating 2 is a sound signal generator of the same system as the generator 12 and an sound test signal is generated in matching with the timing of the video signal arrived to the receiving end. The sound test signal is a signal generated synchronously with simulation to a video image with a delay resulting from a video signal given to a transmitter end via a video processing circuit and a transmission line 5 and has video arrival time information. Then a time difference of two signals received consecutively is intermittently displayed by detecting/displaying/storing the time difference between both the signals on a monitor 3 from a correlation function of an envelope waveform between the sound test signal sent by the transmitter side, received at a receiving end and the sound test signal generated at the receiving end.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to the delay of a video signal due to video format conversion, insertion of a frame synchronizer, and the transmission of video and audio through separate paths in the production and transmission of television programs. Compensating for the time difference between video and audio caused by:
The present invention relates to a technique necessary for ensuring synchronization between video and audio.

[0002]

2. Description of the Related Art Conventionally, for measuring the time difference between video and audio and determining the amount of correction, tap the hand in front of the camera or say a simple comment, or use the video and audio that are easy to understand when the sound is generated. The recorded VTR tape is played back and sent from the sending end of the transmission system, and at the receiving end, the amount of correction of the time difference (usually until it is judged visually and aurally that the utterance timings of the video and audio are subjectively matched) , The sound is electrically delayed and corrected to match the delayed image). However, this method
If the time difference between video and audio is small, such as within one frame (33.3 ms), the judgment will vary depending on the person, which takes time and is not sufficient in terms of accuracy.

Therefore, some attempts have been made to develop a method for quantitatively detecting a time difference without relying on subjective judgment. For example, the time code value is superimposed on a known video at the audio generation point and sent together with the audio from the sending end, and the VTR is recorded at the receiving end, and then jog playback (fine adjustment of time) is performed on the video. There is a method of reading the time code value between the original voice generation point and the actual voice generation point and detecting the time difference from the difference.
On the other hand, a method has also been proposed in which amplitude information of an audio signal is superimposed and transmitted in a vertical blanking period of a video signal to detect a time difference from the audio main line. (Wada, Yamashita: Video and audio time lag measuring device) Technical Report of the Television Society, Vol. 1
7, No. 31, pp7-12 (1993.5.2)
8)).

[0004]

Among the conventional time difference detection methods, the method of temporarily recording and measuring the test signal at the receiving end by the VTR requires a long time for operation, especially in a short test time before live broadcasting. There are some problems to do in. Also,
The method of superimposing audio information during the vertical blanking period of video has the advantage that time differences can be detected even during broadcasting, but conversion from high-definition to NTSC or NTSC and PAL
There is a problem that it cannot be used in systems including video format conversion and satellite transmission lines such as conversion between / SECAM (because the vertical blanking period of the video cannot be fully utilized), future high-definition and standard television will be integrated It is desired to develop a rational measurement method that can be used in production and international collaboration.

Therefore, an object of the present invention is to solve the above-mentioned problems, to quickly and accurately measure the time difference between video and audio on the receiving end side, and to find the optimum time difference correction amount in this system. It is intended to provide a method for measuring the time difference between video and audio.

[0006]

To achieve this object, according to the method for measuring time difference between video and audio according to the present invention, the video signal of the video and audio signals transmitted simultaneously at the transmitting end is processed for various signal processing. , To determine the amount of delay that the video signal is delayed from the audio signal on the receiving end side, the transmitting end side synchronously generates and transmits the reference video signal and the first audio signal, and the receiving end side receives the reference video image. In synchronization with the signal, a second audio signal that is the same as the first audio signal generated at the transmission end side is generated, and the second audio signal and the first audio signal transmitted from the transmission end side are generated. The time difference between the two, that is, the time difference between the arriving reference video signal and the first audio signal transmitted from the transmitting end side is obtained from the correlation between the two.

[0007]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described in detail based on the embodiments with reference to the accompanying drawings. FIG. 1 shows the basic configuration of an apparatus for carrying out the method of the present invention. This device has a transmission end side test signal generating section 1 for generating a video signal and a first audio test signal to be input to the system under measurement on the transmission end side, and a second signal triggered by a video signal arriving at the reception end side. Receiving end side test signal generating section 2 for generating an audio test signal, and a cross-correlation function of envelope waveforms of the first audio test signal received at the receiving end side and the second audio test signal generated at the receiving end side. Is composed of three parts of a correlation type time difference monitor 3 for detecting, displaying and storing the time difference between both signals.

Furthermore, in the time difference adjusting device for video and audio to which the method of the present invention is applied, the delay amount is controlled by the detected and stored time difference between the two signals, and the audio signal reaching the receiving end side is controlled. A delay circuit 4 for delaying is further provided.

The sending end side test signal generator 1 comprises a video pattern generator 11 and an audio signal generator 12. The image pattern generator 11 is, for example, an electronic circuit that intermittently generates a white punch signal (a rectangular wave signal including the rising and falling edges of the signal) from a black level in a cycle of 5 to 6 seconds, or a voice generation. Repeated playback by a semiconductor memory that records a moving image of about 5 to 6 seconds (for example, a so-called clapperboard image used when shooting a movie) or a VTR that allows the timing to be clearly determined can be used.
The video system is NTSC, PAL / SECAM system, high-definition, computer-based video system, etc.
It shall correspond to the video to be measured.

The audio signal generator 12 is, for example, a tone burst (short sound) generator which is generated in synchronization with the white punch signal which is intermittently generated at the above-mentioned period of 5 to 6 seconds, or the above-mentioned moving image. Repeated reproduction by a semiconductor memory in which an accompanying audio signal is recorded, VTR, or the like can be used. In either case, the generation timing of the audio signal and the generation timing of the video signal are strictly aligned in advance.

The receiving end side test signal generating section 2 is an audio signal generator of exactly the same system as the transmitting end side audio signal generator 12, and the first end audio is synchronized with the timing of generation of the video signal on the transmitting end side. Similar to the case where the signal is generated, the second audio test signal is generated here in synchronism with the generation timing of the video signal arriving at the receiving end side. Here, the second audio test signal has the same waveform as the signal from the audio signal generator 12 on the transmission end side, or in the case of a tone burst (short tone) generator, at least the envelope waveform is the same. And In other words, the second audio test signal is pseudo-synchronized with the video that arrives with a delay after the video signal input at the sending end passes through the video signal processing circuit such as format conversion and frame synchronizer and the transmission path 5. The second audio test signal generated in this way is a signal having information on the arrival time of the video.

As shown in the measurement principle of FIG. 2, the correlation monitor 3 obtains the envelope waveforms of two audio signals by the interface 7, and the CPU 8 calculates the cross-correlation function of the envelope waveforms.
A device for displaying the time difference between both signals on the time difference display 9 (Japanese Patent Publication No. 1-25151, "Control system for information recording / reproducing device"), which is continuously input by application of a personal computer. The time difference between signals is intermittently and continuously displayed in a cycle of about 3 to 5 seconds. By using this measuring method, quick and automatic measurement display is possible, and the video signal arriving for this purpose is once converted into the second audio test signal, and the time difference from the arriving first audio test signal is obtained. Is being detected.

The measurement accuracy of the time difference which can be detected by the method of the present invention is ± 7 ms (0.2 frame) accuracy.
Approximately 2 seconds and ± 0.3 seconds with an accuracy of approximately 1 ms.

Although the method of the present invention has been described above based on the embodiment, the method of the present invention is not limited to this, and various modifications and changes can be made within the scope of the invention described in the claims. It will be obvious to those skilled in the art.

[0015]

The time difference measurement according to the method of the present invention is carried out at the time of a test after the transmission line is connected. However, in addition to the pattern generating circuit as the test signal source of the video / audio, the repeated reproduction by the semiconductor memory and the VTR reproduction are performed. HDTV, NTSC, PA that can be used as the video signal source on the sending end
By preparing one that supports L / SECAM,
It has become possible to quickly measure the time difference between video and audio in a system that includes system conversion, which has been difficult to perform quantitative measurement in real time.

Further, by applying the time difference measurement technique utilizing the cross-correlation between two audio signals, which has already been put to practical use and has a track record of operation, it is possible to obtain the time difference in an extremely short time of about ten seconds after the connection of the measured system. It has become possible to detect and display, and it has become possible to quickly and efficiently correct the time difference even when only a short test time is available.

[Brief description of drawings]

FIG. 1 is a diagram showing a basic configuration of an apparatus for carrying out the method of the present invention.

FIG. 2 is a diagram showing a principle configuration of a correlation type time difference monitor.

[Explanation of symbols]

 1 Sending end side test signal generating part 2 Receiving end side test signal generating part 3 Correlation type time difference monitor 4 Delay circuit 5 Video processing circuit or transmission line 7 Interface 8 CPU 9 Time difference indicator 11 Video pattern generator 12 Audio signal generator

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Ren Tsujimoto 3-43 Babacho, Chuo-ku, Osaka-shi, Osaka Japan Broadcasting Corporation Osaka Broadcasting Station

Claims (1)

[Claims] 1. The video signal of the video and audio signals transmitted at the sending end side is subjected to various signal processing, and therefore the sending end finds the delay amount at which the video signal is delayed from the audio signal at the receiving end side. Side, the reference video signal and the first audio signal are synchronously generated and transmitted, and the receiving end side is synchronized with the arrived reference video signal and the same second audio signal as the first audio signal generated at the transmitting end side is transmitted. The audio signal is generated, and the time difference between the second audio signal and the first audio signal transmitted from the transmitting end side is obtained from the correlation between the two, that is, the received reference video signal and the transmitting end side are transmitted. A method for measuring a time difference between a video and an audio, characterized by obtaining a time difference from a first audio signal.