JPH0832621A - Video and audio coded data transmission device - Google Patents

Abstract

(57) [Abstract] [Purpose] It is an object to provide a video / audio coded data transmission device that does not cause reproduction deviation between the video and audio on the receiver side. [Structure] Image encoder 11 for encoding a moving image signal
An audio encoder 12 that encodes an audio signal, a frame multiplexer 13 that multiplexes image coded data and audio coded data in a fixed length frame, and temporarily stores video and audio multiplexed data. FIFO memory 14 and the FI
A frame demultiplexer 15 that demultiplexes the image / audio multiplexed data output from the FO memory, and an image packetizer 16 that adds time information to the image coded data to generate a packet.
An audio packetizer 17 for adding time information to audio encoded data to generate a packet, and image packet data and audio packet data output from the image packetizer are packet-multiplexed to generate a transmission stream. It has a packet multiplexer 18 for

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video / audio coded data transmission apparatus for compressing and transmitting a moving picture signal and a voice signal by a highly efficient coding process.

[0002]

2. Description of the Related Art In recent years, a moving image coding apparatus has been widely used as a device for converting a moving image into digital data having a smaller required band, such as a communication medium, a storage medium, or a multimedia of a combination thereof. Is being considered.

A conventional video / audio coded data transmission apparatus will be described below. FIG. 3 shows the configuration of a conventional video / audio coded data transmission device. In FIG. 3, 3
1 is an image encoder for encoding a moving image signal, 32 is an audio encoder for encoding an audio signal, 33 is an image FIFO memory for temporarily storing image data, and 34 is for temporarily storing audio data. Audio FIFO memory, 35 is an image packetizer for adding time information to encoded image data to generate packets, 36 is an audio packetizer for adding time information to encoded audio data to generate packets, 37 Is a packet multiplexer that packet-multiplexes image packet data and audio packet data to generate a transmission stream. FIG. 4 shows the processing steps of a conventional video / audio coded data transmission apparatus. In FIG. 4, 41 is an image F
IFO memory storage status, 42 audio FIFO memory storage status, 43 image packet data format,
Reference numeral 44 is a voice packet data format, and 45 is a transmission stream format.

The operation of the video / audio coded data transmission apparatus configured as described above will be described below with reference to FIGS. 3 and 4. First, the image signal is encoded by the image encoder 31 to form variable-length burst image encoded data, which is temporarily stored in the image FIFO memory 33. Similarly, the voice signal is encoded by the voice encoder 32 to form burst-shaped voice coded data of variable length,
It is temporarily stored in the O memory 34. By reading these encoded data from each FIFO memory at a constant rate, burst data is converted into continuous, continuous data.

For example, the image coded data is an image FIFO.
As shown in the accumulation state 41 of the memory, even if the image coded data is injected in bursts, the data is read at a constant rate equivalent to the injection rate, so that a certain amount of the FIFO memory capacity is set as the FIFO operating point. , Its storage capacity is maintained. The same applies to the case of voice coded data. The image coded data stored in the image FIFO memory 33 is read at a constant rate, and header information, time information, and error correction code are added by the image packetizer to become image packet data 43.

Similarly, the voice coded data stored in the voice FIFO memory 34 is read at a constant rate, and header information, time information and error correction code are added to the voice packetizer 44 to give voice packet data 44. Become. The packet data 43 and 44 are output as a single transmission stream 45 by the packet multiplexer 37 according to the respective packet generation rates.

[0007]

However, in the above-mentioned conventional configuration, the FIFO in which the image and the sound are independent from each other is used.
Since it has a memory, in order to validate the time information added in the subsequent processing, the relative value of the operating point of each FIFO memory is made constant before adding the time information,
It is necessary to make the delay amount of the encoded data of each of the image and the sound constant, and the relative value of the operating point of each FIFO memory is deviated due to the interruption of the encoded data generation from the encoder or the disturbance. When this occurs, it is necessary to reset the FIFO memory and synchronize the FIFO memory operation of the image and the audio again. Also, for image coded data having a bit rate higher than that of audio coded data, it is necessary to set the operating point of the FIFO memory to an appropriate value according to the bit rate. Each time the bit rate is changed, the delay amount of the image coded data in the FIFO memory must be calculated to change the FIFO operating point of the audio coded data.

The present invention solves the above-mentioned problems of the prior art by always keeping the delay amount of the encoded data of the image and the voice output from each encoder of the image and the voice constant, and performing the processing in the subsequent stage. An object of the present invention is to provide an image / sound coded data transmission device which holds the accuracy of added time information and does not cause reproduction deviation between an image and a sound on the receiver side.

[0009]

In order to achieve this object, a video / audio coded data transmission apparatus of the present invention comprises a video coding means for coding a video signal and a voice coding for coding a voice signal. Means, a frame multiplexing means for multiplexing the image coded data outputted from the image coding means and the speech coded data outputted from the speech coding means in a fixed length frame, and the frame multiplexing means FIFO for temporarily storing the output image / audio multiplexed data
A memory, a frame separating means for separating the image / audio multiplexed data output from the FIFO memory, and an image packetizing means for adding time information to the image encoded data output from the frame separating means to generate a packet. An audio packetizing means for adding time information to the audio coded data output from the frame separating means to generate a packet, and image packet data output from the image packetizing means and the audio packetizing means. Packet multiplexing means for packet-multiplexing the output voice packet data to generate a transmission stream is provided.

[0010]

With this configuration, it is possible to always keep the delay amount of the coded data of the image and the voice output from each of the image and voice coding means constant. That is, the image coded data and the audio coded data immediately after being output from each coding means are sampled in a burst form and time-division multiplexed in a fixed length frame of a length determined by the frame multiplexing means in a burst form. And generate the video and audio multiplexed data,
It is temporarily stored in one FIFO memory.

When the storage amount of the FIFO memory reaches the operating point, the image / audio multiplexed data is read out from the FIFO memory at a constant rate, separated by the frame separating means into image coded data and audio coded data, and respectively. Since the packetizing means adds the time information to packetize the packet and then the packet multiplexing means generates the transmission stream, the reproduction deviation between the image and the audio due to the deviation of the FIFO operating point, which has been a conventional problem, is eliminated. In addition, since the image and the sound are always mixed and stored in one FIFO memory, the synchronization of the image and the sound at the time of changing the bit rate of the image coded data becomes much easier.

[0012]

【Example】

(Embodiment 1) An embodiment of the present invention will be described below with reference to the drawings.

In FIG. 1, 11 is an image encoder which is an image encoding means for encoding a moving image signal, 12 is a voice encoder which is an audio encoding means for encoding a voice signal, 1
3 is a frame multiplexer which is a frame multiplexing means for multiplexing the image coded data output from the image encoder 11 and the audio coded data output from the audio encoder 12 in a fixed length frame, and 14 is a frame multiplexer. FI for temporarily accumulating the video / audio multiplexed data output from the frame multiplexer 13.
An FO memory, 15 is a frame separator which is a frame separating means for separating the image / audio multiplexed data output from the FIFO memory 14, and 16 is a unit for adding time information to the image coded data output from the frame separator 15. An image packetizer, which is an image packetizer that generates a packet, and 17 is an audio packetizer, which is a voice packetizer that adds time information to the audio encoded data output from the frame separator 15 to generate a packet. Reference numeral 18 denotes a packet multiplexer which is packet multiplexing means for packet-multiplexing the image packet data output from the image packetizer 16 and the audio packet data output from the audio packetizer 17 to generate a transmission stream. is there.

The operation of the moving picture coding apparatus configured as described above will be described below with reference to FIG. Figure 2
In the figure, 21 is image coded data output from the image encoder 11, 22 is audio coded data output from the voice encoder 12, and 23 is a frame multiplexer 13 that multiplexes in fixed length frames. Burst-shaped video / audio multiplexed data, 24 is continuous video / audio multiplexed data read from the FIFO memory 14 at a constant rate, 25 is an image encoded data frame separated and output from the frame separator 15, and 26 is A voice encoded data frame separated and output from the frame separator 15, 27 is a format of image packet data output from the image packetizer 16, 28 is a format of voice packet data output from the voice packetizer 17, 29 Is the format of the transport stream output from the packet multiplexer 18.

First, the image signal is input to the image encoder 11 and is subjected to a process such as high efficiency encoding to generate the image encoded data 21. The encoding process is performed in a certain block unit. As a result, the encoded data that is output has a burst form. Similarly, the voice signal is input to the voice encoder 12, and burst-like voice encoded data 22 is generated. The burst coded data is sampled at a predetermined data length by the frame multiplexer 13 and is time-division multiplexed to generate burst video / audio multiplexed data 23 in which video and audio coded data are mixed. Is generated. After that, the image / audio multiplexed data 23 is stored in the FIFO
The data is transferred to the memory 14 in a burst form and temporarily stored.

When the storage amount of the FIFO memory 14 reaches the set value, reading from the FIFO memory 14 is started at a constant rate, and continuous video / audio multiplexed data 24 is obtained.
Is obtained. At this time, the average value of the data inflow amount to the FIFO memory 14 and the average value of the data outflow amount are controlled to match.

Therefore, the storage amount of the FIFO memory 14 is controlled so as to always be a constant amount centered on a certain operation point. The continuous image / audio multiplexed data 24 is subjected to image and audio separation processing by the frame separator 15 to generate an image encoded data frame 25 and an audio encoded data frame 26. The image coded data frame 25 becomes image packet data 27 by adding header information, time information, error correction code and the like in the image packetizer 16.

Similarly, the voice coded data frame 26 is added with header information, time information, error correction code and the like in the voice packetizer 17 to become voice packet data 28.
These packet data 27 and 28 are output as one transmission stream 29 by the packet multiplexer 18 according to the respective packet generation rates. On the receiving side, after obtaining the time information added to each packet data of image and sound, and after the decoding of the data in the packet is completed,
The system clock held by the receiver is compared with the acquired time information, and if the times match, the reproduction of the image and sound is started.

As described above, according to the present embodiment, the image coded data 21 and the audio coded data 22 are time-division-multiplexed and then temporarily stored in one FIFO memory 14 to perform the transfer rate conversion, and the FIFO. Since it is demultiplexed after being read from the memory 14, the delay amount of the image coded data and the audio coded data can be always kept constant. Therefore, the accuracy of the time information added by the packetizers 16 and 17 in the subsequent stage is maintained.

Although the frame multiplexer 13 and the frame separator 15 are used in this embodiment, if the image coded data and the audio coded data can be multiplexed and demultiplexed by a predetermined amount, the data can be demultiplexed. A simple switch such as a bus switch may be used instead. Further, if the stability of the operating point of the FIFO memory is ensured, the reading of the video / audio multiplexed data from the FIFO memory does not have to be continuous. Further, the FIFO memory may be a dedicated memory or a general-purpose memory as long as the FIFO operation is possible.

[0021]

As described above, the present invention is a frame multiplexer for multiplexing image coded data output from an image coder and audio coded data output from a voice coder in a fixed length frame. And a FIFO for temporarily accumulating video / audio multiplexed data output from the frame multiplexer
A memory, a frame demultiplexer for demultiplexing the video / audio multiplexed data output from the FIFO memory, and an image packetizer for adding time information to the image coded data output from the frame demultiplexer to generate a packet An audio packetizer for adding time information to the audio coded data output from the frame separator to generate a packet, image packet data output from the image packetizer and the audio packetizer By providing a packet multiplexer that packet-multiplexes the output voice packet data to generate a transmission stream, the delay amount of the image coded data and the voice coded data can always be kept constant, and the packetization in the latter stage can be performed. Since the accuracy of the time information added by the devices 16 and 17 is maintained, the receiver side caused by the deviation of the FIFO operating point A is an image and audio reproduction deviation eliminated, further one FIFO always images and sound are mixed
Since the data is stored in the memory, it is possible to realize the video / audio coded data transmission device in which the synchronization of the video and the audio at the time of changing the bit rate of the video / coded data is significantly easier.

[Brief description of drawings]

FIG. 1 is a block diagram of a video / audio coded data transmission device according to an embodiment of the present invention.

FIG. 2 is a diagram showing a process for explaining an operation of the video / audio coded data transmission device in the embodiment.

FIG. 3 is a block diagram of a conventional video / audio coded data transmission device.

FIG. 4 is a diagram showing a process for explaining an operation of a conventional video / audio coded data transmission device.

[Explanation of symbols]

 11 image encoder 12 audio encoder 13 frame multiplexer 14 FIFO memory 15 frame separator 16 image packetizer 17 audio packetizer 18 packet multiplexer 21 image encoded data 22 audio encoded data 23 burst-like Image / audio multiplexed data 24 continuous image / audio multiplexed data 25 image encoded data frame 26 audio encoded data frame 27 image packet data format 28 audio packet data format 29 transmission stream format

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI Technical indication H04N 7/24 H04N 7/13 Z

Claims (1)

[Claims] 1. An image encoding means for encoding a moving image signal, an audio encoding means for encoding an audio signal, image encoded data output from the image encoding means and the audio encoding means. From the FIFO memory, a frame multiplexing unit that multiplexes the audio encoded data that is output with a fixed-length frame, a FIFO memory that temporarily stores the video and audio multiplexed data that is output from the frame multiplexing unit, A frame separating means for separating the output image / audio multiplexed data, an image packetizing means for adding time information to the image encoded data output from the frame separating means to generate a packet, and the frame separating means. An audio packetizing means for adding time information to the audio encoded data to be output to generate a packet, and an output from the image packetizing means. Image and audio encoded data transmission apparatus characterized by comprising the image packet data and voice packet data outputted from the audio packetization means a packet multiplexing means for generating a transmission stream packet multiplexed with that.