JPH0898158A - Coupling image-voice synchronizer - Google Patents

Abstract

PURPOSE: To make images and sounds match exactly to each other and to prevent sound signals from deteriorating by maintaining constant the total number of sound samples in a sequence of a field read out from a memory buffer in resynchronization. CONSTITUTION: Data samples of incoming images and sounds are written in a memory buffer 3 which is equipped with a capacity for at least 2 frames, and data stored are read out from the memory buffer 3 at a rate which is frequency and phase modulated with an incoming data and a reference clock signal. A resynchronization means 8 adjusts a sequence of addresses used for reading out or writing of stored data for avoiding over and underflow of stored contents of the memory buffer 3, and executes resynchronization of the data read from the memory buffer 3 by selectively repeating or deleting frames of stored data. The resynchronization means 8 maintains constant the total number of sound samples in the sequence of the field read from the memory buffer 3 when the resynchronization is demanded.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a combined video and audio synchronizing device (hereinafter referred to as "combined video / audio synchronizing device").

[0002]

BACKGROUND OF THE INVENTION Synchronizers are used to phase and frequency synchronize two independent video and audio program sources with audio data inserted in the horizontal and vertical timing periods of the video signal. The independent sources have nominally the same frequency but have an indefinite phase relationship that tends to change continuously. As a result, the sync memory fills faster than it empties, or vice versa, and the information is read to reorganize the source and resync the data from the sync memory by adjusting the read address. It will need to be dropped or repeated. This information is dropped or repeated as one frame, each consisting of two video fields.

Serial Digital Interface (SDI)
In a combined digital video / audio synchronizer having the following, audio data is inserted in the horizontal and vertical timing periods of the video. Devices of this type treat such audio and video information as a single entity, as described above, which is fed to a "data" synchronizer. The features of this device are that the perfect "match" between the video and audio is maintained and that the audio signal is not degraded in the synchronizer during normal operation.

[0004]

A problem arises in a 525 line 59.94 Hz system that performs 48 kHz audio sampling because the number of audio samples in each field of the video signal is not an integer. Therefore, when audio and video samples are combined, the same number of audio samples cannot be interleaved in each field, and instead the total number of audio samples must match a sequence of fields (525 lines with 48 kHz sampling). In a 59.94 MHz system, over 5) it will be necessary to interleave a different number of audio samples with the video data to make up the whole. However, this creates a problem for the synchronizer in that the pattern of the number of audio samples in the field sequence from the output of the synchronizer which is currently in progress when the resynchronization operation is occurring may be incorrect.

It is therefore an object of the present invention that a block of audio sample data is temporally interleaved with samples of successive fields of video sample data and a sequence of a fixed number of audio samples is repeated per field of video sample, the sequence of which is repeated. In a combined video / audio synchronizer of the type such that the number of audio samples in at least one field of the sequence differs from the number of audio samples in the remaining fields of the sequence and the total number of audio samples in each sequence is constant before sending to the synchronizer. During the resynchronization operation, a frame to be dropped or repeated should be selected so that the total number of audio samples in the field sequence does not change.

[0006]

SUMMARY OF THE INVENTION The present invention provides a block of audio sample data that is temporally interleaved with samples of consecutive fields of video sample data to provide a fixed number of audio sample sequences per field of video samples. Repeated, the number of audio samples in at least one field of the sequence is
In a combined video / audio synchronizer of the type such that the total number of audio samples in each sequence is constant before being sent to the synchronizer, unlike the number of audio samples in the remaining fields of the sequence: a) at least video and audio samples A memory buffer 3 having a storage capacity for two frames; and b) writing incoming video and audio data samples into the memory buffer 3 and at a rate that is frequency and phase synchronized with the incoming data and reference clock signal, respectively. And read / write control circuits 4 and 6 for reading the data stored therein, and c) used for reading or writing data to the memory buffer 3 in order to avoid overflow or underflow of the stored contents of the memory buffer 3. Read or write address sequence And resynchronizing means 8 for resynchronizing the data read from the memory buffer 3 by selectively repeating or dropping frames of stored data, d) the resynchronizing means 8 , When resynchronization is requested, the timing of resynchronization operation and read address adjustment, and
The combined video is characterized in that the frames to be repeated or dropped are selected such that the total number of audio samples in the sequence of fields read from the memory buffer 3 does not change when resynchronization is performed. It is a voice synchronizer.

Further, in the present invention, when the resynchronization operation is executed, it is preferable that the read address is adjusted by an amount corresponding to the integer frame data stored in the memory buffer.

Also, in the present invention, when requested to drop a frame, the resynchronization means operates to select and drop a frame having a total sample number equal to the total sample number of the first two fields of the sequence. Then
When requested to repeat a frame, the resynchronization means 8 may be operable to select and repeat a frame having a sample count equal to the sample count of the last two fields of the sequence.

Further, in the present invention, the first field of the sequence has a first predetermined number of samples and the second field of the sequence has a second predetermined number of samples, the last field of the sequence. The field has a second predetermined number of samples, and when the resynchronization means 8 is requested to drop a frame, the frame selected for dropping is:
When it is requested to repeat a frame, the frame selected for repetition is composed of both the first sample number field and the second sample number field, and the frame selected for repetition is the second sample number two field. Good to work with, as it consists of.

Further, in the present invention, the number of fields in the sequence is 5, of which 4 fields have 801 audio sample numbers and 1 field has 800 audio sample. .

[0011]

Embodiments of the present invention will now be described in detail with reference to the drawings. FIG. 1 is a block diagram showing a synchronizer according to an embodiment of the present invention, and 1 indicates the synchronizer as a whole. For example, the signal from the video signal source is provided to the synchronizer 1 in order to synchronize with other signals broadcast from the main transmitter.

In this embodiment, an input signal in the form of a bitstream of serial video / audio data is fed to a serial-to-parallel converter 2 and the resulting parallel video / audio data, eg 27 megasamples / 10-bit video / audio data of the second, write controller and timing generator 4
2 frame memory 3 controlled by.
The write controller and timing generator 4 obtains a write clock signal from the serial-parallel converter 2 and clocks the data write to the 2 frame memory 3.

The 2-frame memory 3 operates as a circular buffer, and the stored data can be read from the memory 3 at a rate different from that at the time of writing.

A reference signal is supplied from the signal pulse generator 5 to the read controller and timing generator 6. The reference signal is a signal that is synchronized with the horizontal and vertical sync signals of the signal with which the input signal is synchronized. The read controller and timing generator 6 supplies a read clock signal to the 2-frame memory 3 to control the rate of data read from the 2-frame memory 3. The data read from this two-frame memory 3 is sent to a parallel-to-serial converter 7 for reconversion into a 270 Mbit / sec serial stream output from the synchronizer.

Further, a system controller (control circuit) 8 is provided, and the system controller 8 receives an external control signal to control the synchronizer and controls other parts of the synchronizer.

As noted above for some of the common functions of combined digital video and audio synchronizers, information must be dropped or repeated to maintain synchronization. This loss or repetition of information is performed because the capacity of the data memory in the synchronizer is limited and continuous changes in the phase relationship between the input and output signals cannot be buffered indefinitely.

In a 625 line 50 Hz video system,
Since the audio sample rate when synchronized with the video reference (signal) is 48 kHz, an integer number of audio samples (1920 samples) are generated per frame. When the synchronizer drops or repeats a frame of information, this frame disruption also maintains an integer number of audio samples.

However, the field rate of the 525 line system is not exactly 60 Hz, but 59.94 H.
Since z, the number of audio samples at 48 kHz in one frame period is 1601.6, not an integer.

Assuming that the audio sample rate is synchronized with the video reference, the audio sample data is time-compressed to match the horizontal timing period of the video signal to the serial digital interface. A sample repeat sequence occurs. 5
The number of 48 kHz audio samples over the 5 fields of a 9.94 Hz video becomes an integer (4004). These audio samples, in a serial digital interface,
One field of 800 samples may be distributed for each audio channel, followed by four fields of 801 samples each. This sequence (repeated sequence) is shown in FIGS.

However, if the synchronizer misses or repeats information, and if the synchronizer does not operate according to the invention, it disturbs this 5-field sequence 4003.
Audio sample sequences can occur.

For example, as shown in FIG. 4, if frame 5 is dropped, frame 6 replaces it, which causes 400 instead of the normal 4004 audio sample.
A block with only three audio samples is created. Similarly, as shown in FIG. 5, when frame 6 is repeated,
A block of 4003 samples results. As mentioned above, this phenomenon causes unacceptable perturbations in the audio signal. More specifically, when the audio data is separated from the serial digital interface and time expanded, as required prior to transmission or recording,
A sequence of 4003 samples results in a single sample of audio perturbation, which risks transmitting errors to downstream equipment.

In the preferred embodiment of the invention shown in FIG.
The system controller 8 is configured to control the read controller and the timing generator 6 so as to maintain a block (sequence) of 5 fields with 4004 samples even if one frame is dropped or repeated.

It will be noted that if frame 6 is missing, the resulting block remains at 4004 samples, as shown in FIG. That is,
In the 4004 sample sequence example, when the synchronizer must drop a frame of information,
By dropping the border frame containing one field of 0 samples and one field of 801 samples, the 5 field sequence of 4004 samples is maintained.

Similarly, as can be seen from FIG. 3, even if frame 5 is repeated, the resulting block remains 4004 samples. So when the synchronizer has to repeat one frame of information,
Choosing a border frame containing two fields of 801 samples each would retain the total number of samples as well.

According to the present invention, when performing the resynchronization operation, the system controller 8 selects which frame is to be dropped or repeated so that the number of output audio samples in each block of each 5-field sequence is constant. Is a number 4
Keep 004. Repeating a frame effectively delays the incoming sequence by two fields, whereas dropping a frame effectively advances the incoming sequence by two fields. Therefore, 40
In order to maintain the value of 04, the frame that has the same number of audio samples in the last two fields of the above sequence is selected as the frame to be repeated by the system controller 8 and the missing frame is Select the frames with the same total number of audio samples in the first two fields of the sequence.

In view of the above points, by specially choosing the border frame to use during resynchronization, the synchronizer
Without disturbing the sequence of 4004 audio samples,
Frames can be dropped or repeated.

Thus, when operating with a 525 line 59.94 Hz video signal, a five field digital audio sample sequence is reliably maintained in the combined digital video and audio synchronizer.

This allows 625 lines 50 Hz and 52 with serial digital interface (SDI).
"Data" in which both video and audio data are combined in a 5 line 59.94 Hz video system
Then, the synchronizing device becomes operable.

The present invention is not limited to interlaced television signals, but can also be applied to non-interlaced television signals. When applying the present invention to non-interlaced television signals, the term "field" as used herein is replaced with the term "frame".

According to the synchronizer of the embodiment, blocks of audio sample data are temporally interleaved with samples of consecutive fields of video sample data and a fixed number of audio sample sequences per field of video samples are repeated, At least one of the sequences
In a combined video / audio synchronizer of the type such that the number of audio samples in one field is different from the number of audio samples in the remaining fields of the sequence and the total number of audio samples in each sequence is constant before sending to the synchronizer. A) a memory buffer 3 having a storage capacity of at least two frames of video and audio samples, and b) writing incoming video and audio data samples into the memory buffer 3, and the incoming data and reference clock signal and their respective frequencies. And write and read control circuits 4, 6 for reading the data stored therein from the memory buffer 3 at a phase-synchronized rate, and c) in order to avoid overflow or underflow of the stored contents of the memory buffer 3. For reading or writing data to 3 Resynchronizing means 8 for resynchronizing the data read from the memory buffer 3 by adjusting the sequence of read or write addresses to selectively repeat or drop frames of stored data. D) The resynchronization means 8 determines the timing of resynchronization operation and read address adjustment when resynchronization is requested, and
The frame to be repeated or dropped is selected so that the total number of audio samples in the sequence of fields read from the memory buffer 3 does not change when resynchronization is performed, so that can get.

That is, blocks of audio sample data are temporally interleaved with samples of consecutive fields of video sample data, a constant number of audio sample sequences per field of video samples are repeated, and at least one field of the sequence. In a combined video / audio synchronizer of the type such that the number of audio samples in the sequence is different from the number of audio samples in the remaining fields of the sequence and the total number of audio samples in each sequence is constant before sending to the synchronizer. When performing synchronization, it is possible to have a frame choose to repeat or drop so that the total number of audio samples in the sequence of fields read from the memory buffer does not change, which allows the video and audio to match exactly. Voice signal Deterioration does not occur.

[0032]

According to the invention described above, blocks of audio sample data are temporally interleaved with samples of consecutive fields of video sample data, and a fixed number of audio sample sequences per field of video samples are repeated. , At least one of its sequences
In a combined video / audio synchronizer of the type such that the number of audio samples in one field is different from the number of audio samples in the remaining fields of the sequence and the total number of audio samples in each sequence is constant before sending to the synchronizer. A) a memory buffer having a storage capacity of at least two frames of video and audio samples, and b) writing incoming video and audio data samples into the memory buffer, and frequency and phase of incoming data and reference clock signal, respectively. A write and read control circuit for reading the data stored therein from the memory buffer at a synchronized rate; and c) for reading or writing data to the memory buffer in order to avoid overflow or underflow of the stored contents of the memory buffer. Read or write to use A resynchronization means for resynchronizing the data read from the memory buffer by adjusting the sequence of addresses only to selectively repeat or drop frames of stored data, and d) resynchronization. Means select the timing of the resync operation and adjustment of the read address, and thus the frame to be repeated or dropped, when resync is requested so that when resync is performed, it is read from the memory buffer. Since the total number of audio samples in the sequence of different fields is kept unchanged, the following effects can be obtained.

That is, blocks of audio sample data are temporally interleaved with samples of successive fields of video sample data, a constant number of audio sample sequences per field of video samples are repeated, and at least one field of the sequence is repeated. In a combined video-audio synchronizer of the type such that the number of audio samples in is different from the number of audio samples in the remaining fields of the sequence and the total number of audio samples in each sequence is constant before sending to the synchronizer,
When performing a resynchronization, it is possible to choose what frames are repeated or dropped so that the total number of audio samples in the sequence of fields read from the memory buffer does not change, which allows the video and audio to be accurate. They match each other, and the voice signal is not deteriorated.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a frame sequence before and after a missing frame according to the embodiment of FIG.

FIG. 3 is an explanatory diagram showing a frame sequence before and after a repeating frame according to the embodiment of FIG.

FIG. 4 is an explanatory diagram showing a frame sequence before and after a missing frame by a conventional device.

FIG. 5 is an explanatory view showing a frame sequence before and after a repeating frame by a conventional device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 synchronizer 2 serial-parallel converter 3 frame memory (memory buffer) 4 write controller and timing generator 5 signal pulse generator 6 read controller and timing generator 7 parallel-serial converter 8 system controller

 ─────────────────────────────────────────────────── ————————————————————————————————————————————————————————————————————————————————————————————————————— Germany— Hedley Wilkinson UK RG26 6UN Hampshire, Tadley, Heathlands, Humble Drive 17

Claims (5)

[Claims] 1. A block of audio sample data is temporally interleaved with samples of consecutive fields of video sample data, and a constant number of audio sample sequences per field of video samples is repeated, at least one field of the sequence. In a combined video-audio synchronizer of the type such that the number of audio samples in the sequence differs from the number of audio samples in the remaining fields of the sequence and the total number of audio samples in each sequence is constant before sending to the synchronizer: ) A memory buffer having a storage capacity for at least two frames of video and audio samples, b) Writing incoming video and audio data samples to the memory buffer, and frequency and phase synchronization with the incoming data and reference clock signal, respectively. With the rate A write and read control circuit for reading the data stored therein from the memory buffer; and c) used for reading or writing data to the memory buffer in order to avoid overflow or underflow of the stored contents of the memory buffer. Resynchronizing means for resynchronizing the data read from the memory buffer by adjusting the sequence of read or write addresses to selectively repeat or drop frames of stored data; d ) When the resynchronization is requested, when the resynchronization is requested, the timing of the resynchronization operation and the adjustment of the read address, and therefore the frame to be repeated or dropped, is selected, and thereby, when the resynchronization is performed. , In the sequence of the fields read from the memory buffer A combined video / audio synchronizer characterized in that the total number of audio samples does not change. 2. The combined video / audio synchronization device according to claim 1, wherein when the resynchronization operation is executed, the read address is
A combined video / audio synchronizing device characterized in that the amount is adjusted by an amount corresponding to the integer frame data stored in the memory buffer. 3. The combined video / audio synchronizer according to claim 1 or 2, wherein the resynchronization means is equal to the total number of samples in the first two fields of the sequence when requested to drop a frame. The resynchronization means operates to select and drop frames with the total number of samples, and when requested to repeat the frames, the resynchronization means selects frames with the total number of samples equal to the total number of samples in the last two fields of the sequence. A video / audio synchronizing device characterized by being selected and operated to be repeated. 4. The combined video and audio synchronizer according to any one of claims 1 to 3, wherein the first field of the sequence has a first predetermined number of samples. The second field has a second predetermined number of samples, the last field of the sequence has the second predetermined number of samples, and the resynchronization means is required to drop a frame. When the frame selected for the drop consists of both the first number of samples field and the second number of samples field, and when the frame is requested to repeat, the repetition A frame selected for use in a video operation characterized in that it is composed of two fields of the second number of samples.
Audio synchronizer. 5. The combined video / audio synchronizing device according to claim 1, wherein the number of fields in the sequence is 5, and 4 fields out of which 801 audio sample numbers are included. A combined video / audio synchronizer characterized in that one field has 800 audio samples.