JPH09185870A - A/v server - Google Patents

Abstract

PROBLEM TO BE SOLVED: To smoothly reproduce a voice during double-speed reproducing. SOLUTION: A RADIO 2V is provided for video data and a hard disk 2A is provided for audio data. For the unit of the video data of the RAID 2V, data three times as much as the audio data corresponding to the unit is recorded/reproduced. Thus, even when the video data is intermittently reproduced by one frame per three frames during triple-speed reproducing, a series of audio data equivalent to the three frames are obtained. By thinning this reproduced audio data, a voice is smoothly reproduced.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an A / A system for recording / reproducing a digital video signal and a digital audio signal on / from a randomly accessible recording medium.
Regarding V server.

[0002]

2. Description of the Related Art Digital video signals and digital audio signals have been recorded on recording media such as MO (magneto-optical recording) disks and hard disks. Unlike magnetic tapes, these recording media can be randomly accessed, and it has been attempted to use them in place of a magnetic tape device (VTR) in a broadcasting station system, for example. Although not limited to the broadcasting station system, in the case of editing or the like, not only the reproduction at the normal speed but also the reproduction at the double speed, the triple speed or the like is possible in order to grasp the recorded content of the recording medium.

[0003]

Conventionally, double speed, 3
When the double speed reproduction such as the double speed is performed, the image data is read out intermittently, and the audio data also becomes intermittent. For example, in double speed reproduction, only one frame of audio data can be read out every two frames. Such audio data becomes fragmentary as a reproduced sound, which causes a problem that the contents cannot be fully heard. Therefore, in the case of a reproduction speed exceeding 1 × speed, there is only a method of muting the audio or connecting the readable audio data as it is, and outputting a discontinuous sound, so that satisfactory audio reproduction cannot be performed.

The above-mentioned problem at the time of the conventional double speed reproduction is caused by the fact that the processing of the audio signal is performed by the same method as the processing of the video signal. That is, since the image data is configured in units of 1 frame (1/30 second), it is possible to determine the image after thinning out even if it is thinned out in frame units, whereas if the audio data is thinned out at similar intervals, Since the thinning-out unit is too large, necessary audio data is lost, resulting in unnatural sound after thinning out, or inability to hear the contents.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a digital signal recording method capable of reproducing a reproduced voice to such an extent that a reproduced voice can be heard at the time of double speed reproduction in consideration of the above point.

[0006]

According to the invention of claim 1, video data and audio data are externally accessed A
/ V server, a first storage device capable of random access for video data and an audio device capable of random access for audio data and corresponding to a first data unit of video data read from the first storage device. A second storage device adapted to read audio data of a second data unit in an amount N times as large as the amount of data;
In the N-fold speed reproduction operation in which the video data is thinned to 1 / N, and the video data in the data unit and the audio data in the second data unit are controlled so as to be reproduced. It is an A / V server characterized in that it can be reproduced without loss.

Further, according to the present invention, the second storage device is a hard disk, and 4-channel audio data is recorded on the hard disk, and the number of channels is reduced during a double speed reproduction operation exceeding N. .

According to the present invention, separate storage devices are provided for video data and audio data. The N times speed reproduction operation is performed by thinning out the video data to 1 / N and reproducing it. Since the audio data is reproduced with a data unit of N times the data unit of the audio data corresponding to the data unit of the reproduced video data, the loss of the audio data does not occur. Therefore, even if the video data is thinned out at the time of double speed reproduction, audio data without loss can be reproduced. As a result, the quality of sound during double speed reproduction can be improved.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows the configuration of an A / V server according to this embodiment. In this embodiment, the number of channels that can be simultaneously input / output is 6. Input / output data of 6 channels CH1 to CH6 are IO
RAID (Redundant Arrays of Inexpensive Disks) via the processors 1 ₁ , 1 ₂ , 1 ₃ , 1 ₄ , 1 ₅ and 1 _6.
Combined with 2V and hard disk 2A. RAID
Is a combination of a plurality of disk array units each including a plurality of hard disk devices. RA
A plurality of IDs are defined from level 0 depending on the difference in error correction processing. The RAID 2V is for video data, and the hard disk 2A is for audio data. Since audio data has a smaller data amount than video data, a hard disk or a disk array device is used. in this way,
Separate hard disks are used for video data and audio data, and the recording / playback processing is different between them.

Each of the IO processors 1 ₁ to ₁₆ has a buffer memory, and a control signal is supplied to each of the IO processors 1 ₁ to ₁₆ . The input / output video data is, for example, video data compressed by MPEG2. However, the compression is not always necessary. The audio data is digital data sampled at a predetermined sampling frequency. IO processor ₁ 1 to 1 _6, for RAID2V and hard 2A, the system controller 3
The control signal from is supplied. The system controller 3 controls the recording (writing) / reproducing (reading) operation of the RAID 2V and the hard disk 2A, and
It also controls the synchronization relationship between the two.

Data of each channel of CH1 to CH6 is time-division multiplexed in a certain time slot. In the case of video data, the data stream (video data) of each channel is divided into stripes having a length corresponding to a data unit, for example, 1 second (30 frames). FIG. 2 shows video data for one channel of CH1. The time-division multiplexed video data of CH1 to CH6 is written in RAID2V. Video data is also reproduced in this data unit. This data unit is merely an example, and can be set as appropriate in consideration of the configuration of the storage device used.

For example, in the case of handling audio data for four channels, as shown in FIG. 2, the unit of video data (for one second) and the data amount of the corresponding audio data are three times (that is, three seconds). The audio data A1, A2, A3, and A4 of each channel are set as one unit, these data are time-division multiplexed and recorded on the hard disk 2A, and the audio data is reproduced in this unit. The data of each audio channel to be recorded / reproduced is in the order of channel 2 (A2), channel 1 (A1), channel 3 (A3), and channel 4 (A4).

In one embodiment of the present invention, it is assumed that the reproduction operation at about (N = 3) times speed is the maximum speed at which the contents can be heard, and therefore, it is set to three times the access unit of the video data. Set the access unit of audio data. Then, as shown in FIG. 3, with respect to audio data, three times as many audio data as the recording unit number of the video data are recorded in a time slot in which the video data is consecutively obtained at the same speed reproduction. None Two of the three time slots are not used. One time slot in FIG. 3 represents one frame or the unit (1 second) described above. It should be noted that in the normal speed reproduction, the video data is continuously read, in the double speed reproduction, the video data is intermittently read at a rate of one frame every two frames, and in the three times speed reproduction, one frame every three frames. The video data is read intermittently at a rate of.

The process of writing audio data to the hard disk 2A in this embodiment will be described with reference to FIGS. 4 and 5. As shown in FIG. 4, audio data A1 to A4 of the first to fourth channels
3 seconds is written in the buffer memory, compressed along the time axis, and the data array is rearranged. That is,
Start with channel A data A2, then channel 1
Data A1, channel 3 data A3, and channel 4 data A4 are arranged in this order and then written to the hard disk 2A. The amount of 4-channel audio data for 3 seconds is, for example, 1,440 kbytes. This data can be reproduced by the hard disk 2A within the time for reproducing one second of video data.

FIG. 5 shows a state in which the audio data compressed as described above and rearranged in order is written in the hard disk 2A. In FIG. 5, the hard disk 2A is configured by four hard disks HD1 to HD4. As described above, the audio data of all channels can be reproduced from the hard disk 2A during the double speed reproduction up to the triple speed.
When the speed ratio becomes higher and there is no time to spare,
It is necessary to reduce the amount of data read from the hard disk 2A. In actual operation, the amount of data is reduced by reducing 4 channels to 2 channels. More specifically, the main L / R mode using two channels of the first and second channels, the main L / sub L mode using two channels of the first and third channels, the third and fourth Either sub-L / sub-R mode, which uses two channels, is often adopted.

As described above, in the case of reducing the number of channels at the time of reproducing at a speed higher than 3 times speed, the method of recording 4-channel audio data on the hard disk as shown in FIG. 5 enables continuous reading without performing head seek. There is an advantage that the reproduction data amount can be increased by the amount corresponding to the seek time.

FIG. 6 shows a data reproducing operation in one embodiment of the present invention. As shown in FIG. 6A, the video data is reproduced in units of 1 second. At the same speed reproduction, all the video data are read from the RAID 2V. On the other hand, the audio data of four channels is read from the hard disk 2A in the order of A2, A1, A3, and A4 in a time period of 1 second. As shown in FIG. 6B, the audio data of each channel is expanded on the time axis. In the normal speed reproduction operation, the audio data of all the channels read out and expanded in the time axis are used as they are.

On the other hand, in the 2 × speed reproduction operation or the 3 × speed reproduction operation, all the audio data is read out as in the case of the 1 × speed reproduction. When reproduced as it is, the time axis becomes three times as large as the video data, so audio data is thinned out and the time axis is aligned with the video data. Taking the 3 × speed reproduction operation as an example, as shown in FIG. 6C, for example, data for 9 seconds (reading 3 seconds) is thinned to 1/3. In this case, it is preferable to perform re-sampling at a frequency (16 kHz) that is ⅓ of the sampling frequency (for example, 48 kHz) of the audio data, and then thin the re-sampled unit to ⅓. That is, the audio data is resampled at almost phoneme intervals, and two samples are thinned out every three samples. During double speed reproduction, the resampled data is thinned to 1/2. The processing circuit for this thinning is provided inside or outside the IO processor.

A phoneme constitutes a syllable and has a vowel and a consonant. Although it is difficult to perform thinning out in a phoneme unit strictly, audio data is resampled so as to be divided by a phoneme of a general length. In short, in one embodiment of the present invention, when the audio data is restored to an analog signal and made continuous, the audio data is thinned out in units of a length that allows the contents to be discriminated to some extent.

The audio data processing at the time of the double speed reproduction or the triple speed reproduction operation is not limited to the above-described thinning processing, and smooth voice reproduction may be performed using a harmonizer.

If the double speed and the triple speed are set to low speed, the reproduced audio data will be lost in the middle speed operation such as 6 speed and 12 speed. In one embodiment of the invention, 6
In the case of double speed reproduction, only data for two channels, for example, audio data of channels 1 and 2 are reproduced from the hard disk 2A. Continuous audio data can be obtained for these two channels.
During 12 × speed reproduction, audio data of only one of the four channels is reproduced. Thereby 1
Continuous audio data can be obtained for the channels. In this way, smooth sound reproduction is possible even during double speed reproduction exceeding 3 times speed.

A storage device other than the RAID and hard disk may be used. Also, if the video data is MPE
When compressed in G2, double speed reproduction may be performed by selectively reproducing only I pictures.

[0023]

EFFECTS OF THE INVENTION The present invention, in the A / V server,
While the video data is intermittently reproduced in the double speed reproduction operation, the audio data can be reproduced without loss, and the audio can be reproduced smoothly by processing the reproduced audio data. Further, according to the present invention, even if the speed ratio becomes higher, the number of channels of audio data of a plurality of channels is reduced, so that smooth voice reproduction is possible.

[Brief description of the drawings]

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

FIG. 2 is a schematic diagram for explaining a recording process according to an embodiment of the present invention.

FIG. 3 is a schematic diagram showing a relationship between video data and audio data according to an embodiment of the present invention.

FIG. 4 is a schematic diagram showing processing of audio data according to an embodiment of the present invention.

FIG. 5 is a schematic diagram for explaining recording of audio data on a hard disk according to an embodiment of the present invention.

FIG. 6 is a schematic diagram for explaining a reproducing operation in the embodiment of the present invention.

[Explanation of symbols]

1 ₁ to ₁₆ IO processor RAID for 2V video data Hard disk for 2A audio data 3 System controller

Claims (2)

[Claims] 1. An A / V server for externally accessing video data and audio data, wherein a first storage device capable of random access for video data and a first storage device capable of random access for audio data are provided. A second storage device adapted to read audio data of a second data unit in an amount N times the amount of audio data corresponding to the first data unit of the video data read from the device; In the N-times speed reproduction operation in which the video data of one data unit and the audio data of the second data unit are controlled to be reproduced in association with each other, the video data is thinned to 1 / N. , A / V server that can play audio data without loss. 2. The A / V server according to claim 1, wherein the second storage device is a hard disk, and four-channel audio data is recorded on the hard disk, and the double speed exceeding N is achieved. An A / V server characterized by reducing the number of reproduced channels of the audio data during reproduction.