JP2003242717A - Recording and reproducing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the unnaturalness of audio during intermittent reproduction. <P>SOLUTION: In an MPEG encoder 16, the average level value of length equivalent to one audio frame of inputted audio signals is calculated and an identifier indicating the average level value is added to the header of Audio PES prepared from one audio frame. The Audio PES is recorded in the HD of a HDD 20. When reproducing the Audio PES, the identifier added to the Audio PES is compared with a prescribed threshold in a stream processing circuit 22 for the audio, and if the size relation of the identifier and the threshold indicates that the average level value of the audio signals is smaller than a prescribed value, the Audio PES is thinned. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a recording / reproducing apparatus, and more particularly to recording an encoded audio signal obtained by encoding an input audio signal in a packet system on a recording medium.
The present invention relates to a recording / reproducing device that reproduces an encoded audio signal from a recording medium.

[0002]

2. Description of the Related Art When reproducing a program recorded by a video recorder adopting the MPEG system, MPEG-P is used.
The S (PS: Program Stream) is separated into a video PES (Packetized Elementary Stream) and an audio PES by a demultiplexer, and the video PES and the audio PES are respectively MPEG (Moving Picture).
Experts Group) Video decoder and MPEG audio decoder. As a result, the program image is displayed on the display, and the program sound is output from the speaker. Here, when the normal reproduction is switched to the fast-forward reproduction, the decoding of the audio PES is interrupted, and the packets forming the video PES are intermittently decoded.

[0003]

However, if the output of the program sound is interrupted during the fast-forward reproduction, it becomes difficult to grasp the content of the program. On the other hand, if the packets forming the audio PES are decoded intermittently, the level change of the decoded audio signal is not continuous, and the voice becomes unnatural at the discontinuous point of the level change.

Therefore, a main object of the present invention is to provide a recording / reproducing apparatus capable of improving the unnaturalness of sound during intermittent reproduction.

[0005]

The present invention provides a recording / reproducing apparatus for recording an encoded audio signal obtained by encoding an input audio signal in a packet system on a recording medium and reproducing the encoded audio signal from the recording medium. Calculation means for calculating the average level value of the input voice signal for each length corresponding to one packet, and average level information related to the average level value calculated by the calculation means is embedded in each packet forming the encoded voice signal. A recording / reproducing apparatus comprising: an embedding means and an intermittent reproducing means for intermittently reproducing an encoded audio signal based on average level information.

[0006]

According to the present invention, when the voice packet encoded by the packet method is intermittently reproduced, the voice packet is thinned out based on the average level value of the voice signal (input voice signal) included in the voice packet. Decide whether to play.

That is, first, the calculating means calculates the average value level of the input voice signal for each length of the voice signal corresponding to one voice packet. Then, the embedding unit embeds the average level information related to the average level value calculated by the calculating unit in the voice packet formed by encoding the input voice signal. Further, the intermittent reproduction means thins out the packets whose average level value of the audio signal is lower than a predetermined threshold value based on the average level information, and reproduces only the audio packets whose average level value is equal to or higher than the predetermined threshold value.

In the case of a hard disk recorder that performs intermittent reproduction of video packets in parallel with intermittent reproduction of audio packets, the amount by which audio packets are thinned out is adjusted for the video packets thinned out at regular intervals.

Each audio packet and each video packet includes a time stamp, and the audio packet and the video packet are reproduced at the timing according to this time stamp. The video packets are thinned out at regular intervals, while the voice packets determine the thinned voice packets based on the result of comparison between the average level value of the voice signal and the threshold value. Therefore, if no care is taken, there will be a time lag between the reproduced audio packet and the reproduced video packet.

Then, the comparison means compares the time stamp of the audio packet with the time stamp of the video packet,
When the time stamp of the audio packet is larger than the time stamp of the video packet, the number of audio packets is thinned out. Therefore, the threshold value is lowered to make it difficult to perform audio packet thinning. On the contrary, when the time stamp of the voice packet is smaller than the time stamp of the video packet, the threshold value is raised to facilitate the thinning of the voice packet.

[0011]

According to the present invention, it is possible to select and thin packets which have a low level of the voice signal and have a small influence on the sound quality, so that the unnaturalness of the voice during intermittent reproduction can be improved.

The above-mentioned objects, other objects, features and advantages of the present invention will become more apparent from the detailed description of the embodiments below with reference to the drawings.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Referring to FIG. 1, a hard disk recorder 10 as a recording / reproducing apparatus of this embodiment includes an MPEG encoder 16. The MPEG encoder 16 has A /
The video signal digitized by the D converter 12 and the audio signal digitized by the A / D converter 14 are input. That is, the hard disk recorder 10 can record and reproduce video and audio. The video signal and the audio signal input to the MPEG encoder 16 are encoded and output as a video PES (packet) and an audio PES (packet), respectively, and are output from an HD (Hard Disk) (not shown) of an HDD (Hard Disk Drive) 20. Recorded in. When reproducing a video signal (video PES) and an audio signal (audio PES) recorded in HD, the video PES and audio PES read from the HD are stream processing circuit 24 for video and audio, respectively. Of the stream processing circuit 22. The HDD 20 of the hard disk recorder 10 can simultaneously read out the video PES and the audio PES while recording them in the HD.

In the stream processing circuit 24 and the stream processing circuit 22, the video PES and audio PE are connected.
Predetermined stream processing is performed on S (thinning of video PES and audio PES and replacement of time stamps described later). Then, the video PES and the audio PES subjected to the stream processing are output to the MPEG decoder 30, respectively. In the MPEG decoder 30, the time stamp (PT recorded in the PES header is recorded.
S: Decoding (playback output) of video PES and audio PES in the order according to Presentation Time Stamp)
Then, the decoded video signal and audio signal are converted into analog signals by the D / A converter 32 and the D / A converter 34 and output from a monitor or a speaker (not shown).

In the conventional hard disk recorder,
As a function utilizing the random accessibility of the HDD 20, there is a function called catch-up reproduction. This catch-up playback means that when playback is being performed while recording video and audio, when playback is temporarily stopped and playback is restarted,
By reproducing at a speed faster than the recording speed (fast-forward reproduction), the reproduction can be overtaken by the recorded video and audio. In a conventional hard disk recorder, the catch-up reproduction is realized by thinning out video data at regular intervals and stopping output of audio data. However, if the output of the audio data is stopped, it becomes difficult to understand the content, and if the audio data is thinned out at regular intervals, the sound quality will be deteriorated.

Therefore, in the hard disk recorder 10 of this embodiment, the signal level of the input voice is detected and the thinning of the voice data is controlled according to the signal level to reduce the deterioration of the sound quality (unnaturalness of the voice). .

The MPEG encoder 16 is specifically constructed as shown in FIG. MPEG encoder 1
The video signal input to 6 is encoded into a video PES by the MPEG video encoder 40 and output.

On the other hand, the audio signal is input to the MPEG audio encoder 44, and the audio signal corresponding to one audio frame is given to the average value calculation circuit 42. In addition, one audio PES is formed from an audio signal corresponding to one audio frame. The average value calculation circuit 42 calculates an average value of signals for each audio signal corresponding to one audio frame.

The comparator 46 compares the average value calculated by the average value calculation circuit 42 with a plurality of set values recorded in the memory 48, and shows an identifier ("A", ""AB","B") is applied to the MPEG audio encoder 44. MPEG audio encoder 4
4 encodes an audio signal corresponding to one audio frame to form an audio PES, and a comparator 4
The identifier given from 6 is added to the header of the audio PES. The audio PES thus formed is MP
It is output from the EG encoder 16 and recorded in the HD of the HDD 20 together with the video PES.

The operation of the audio system in the MPEG encoder 16 will be described with reference to the flow chart shown in FIG.
First, in step S1, an audio signal corresponding to one audio frame is acquired from the A / D converter 14, and in step S3, an average value "x" of the audio signal corresponding to one audio frame is calculated.

Then, in step S5, an audio signal corresponding to one audio frame is encoded to create an audio PES.

In step S7, the average value "x" is "0".
As described above, it is determined (compared) whether it is smaller than "A".
If "0" or more and less than "A", step S9
The identifier "A" is added to the header of the audio PES. The identifier “A” is a predetermined numerical value. The same applies to the following identifiers “AB” and “B” (A <AB <
B).

On the other hand, if NO in step S7, it is determined in step S11 whether the average value "x" is greater than or equal to "A" and less than "B" (comparison).
To do. If "A" or more and less than "B", the identifier "AB" is assigned to the audio PE in step S13.
Add to S header.

If NO is determined in step S11, the identifier "B" is determined in step S15.
Is added to the header of the audio PES.

In step S17, it is determined whether or not the process of calculating the average value "x" of all the audio signals and creating the PES is completed, that is, whether the input of the audio signals is completed, and all the audio signals are extracted. If the PES has been created, the process ends.

In the above example, the identifiers are "A" and "A".
There are three types, "B" and "B". That is, the average value of the audio signal is evaluated in three stages, but this is for convenience of explanation, and in practice, the evaluation is performed in more stages.

The audio stream processing circuit 22 is specifically constructed as shown in FIG.
Audio PES input to the stream processing circuit 22
Is an output control circuit 50, PTS (Presentation Time Stam)
p) It is given to the detection circuit 52 and the identifier detection circuit 54.

In the PTS detection circuit 52, the audio PE
PTS is detected from S, and the detected PTS is the comparator 2
8 and is recorded in the work area of the memory 60.

In the identifier detection circuit 54, the identifiers ("A", "AB",
"B") is detected. The detected identifier is the comparator 5
6 is compared with the threshold value stored in the memory 58. Then, the result of the comparison is given to the output control circuit 50.

The output control circuit 50 determines whether to output the audio PES or to thin out the audio PES according to the comparison result of the comparator 56. When outputting an audio PES, the PTS of the previously thinned out audio PES
PT recorded in the audio PES in consideration of
Replace S. Since the audio PES is reproduced in the order according to the time stamp indicated by the PTS, when even one audio PES is thinned out, it is necessary to replace the PTSs of the audio PESs after the thinned audio PES and correct the PES.

On the other hand, the video stream processing circuit 24
The video PES is input to the input video PES (see FIG. 3), and the input video PES is output to the output control circuit 70 and the PTS detection circuit 72.
Given to and respectively. In the case of catch-up reproduction, the output control circuit 70 does not output all the video PES, but thins out the video PES at regular intervals and outputs the video PES intermittently. The PTS detection circuit 72 detects the PTS from the video PES and supplies the detected PTS to the comparator 28.

The comparator 28 compares the PTS of the video PES supplied from the PTS detection circuit 72 with the PTS of the audio PES obtained from the memory 60. Then, the comparison result is given to the controller 26. The controller 26 changes the threshold value set in the memory 58 according to the comparison result, and adjusts the amount of audio PES to be thinned out. Audio PES decimated by lowering the threshold
Becomes smaller, and the higher the threshold value, the larger the amount thinned out.

The operation of the audio stream processing circuit 22 will be described with reference to the flow chart shown in FIG.
Although the flow chart of FIG. 5 is described on the premise of catch-up reproduction, it goes without saying that the hard disk recorder 10 can also perform normal reproduction.

First, in step S21, the memory 58
In addition, “A” of “A”, “AB”, and “B” is used as the threshold value.
"B" is set as an initial value. In step S23, the work area "W3" provided in the memory 60 is cleared to zero and initialized. The work area “W3” is a work area for holding the accumulated time of the thinned audio PES. Note that when simply expressed as “W3”, it indicates the value stored in the work area “W3”. The same applies to other work areas below. Further, in step S25, the "thinning flag" is reset. The "thinning flag" is a flag for identifying whether or not thinning has been performed, and when thinning is performed, the flag is set.

In step S27, it is determined whether the "thinning-out flag" is in the set state. If it is not in the set state, one audio PES is read from the HDD 20 in step S29. On the other hand, if the "thinning-out flag" is in the set state, step S29 is skipped and the audio PES is not read.

Then, in step S31, the audio P
The PTS of ES is stored (copied) in the work area “W1”. When the "thinning flag" is in the reset state (when step S29 is executed), the PTS of the audio PES read out in step S29 is stored in the work area "W1", but when the "thinning flag" is in the set state, The PTS of the audio PES read out in step S37 described later is stored in the work area "W1".

In step S33, the audio PE read in step S29 or step S37 is read.
The identifier (any one of “A”, “AB”, and “B”) is extracted from the header of S, and the threshold value (any one of “A”, “AB”, and “B”) stored in the memory 58 is , Which is initialized to "AB" at the beginning). If the identifier is less than or equal to the threshold value, it indicates that the level of the audio signal of the audio PES is lower than the threshold value, and even if the audio PES is thinned out, the influence on the reproduced sound is small, so Audio P
Thin out ES. In step S35, "thinning flag"
To the set state.

When thinning out the audio PES, it is necessary to correct the PTS of the audio PES next to the thinning audio PES so that the stream is continuous.

Therefore, in step S37, the next audio PES is read, and the read audio PE is read.
The PTS of S is set to the work area "W in step S39.
2 ”(copy), and in step S41“ W
2 ”-“ W1 ”is added to the work area“ W3 ”. As a result, the work area “W3” accumulates the time spent when the thinned audio PES is played back. Then, in step S43, the PTS of the audio PES next to the thinned audio PES is set to "W2"-.
Replace with the value of "W3". In other words, this "W2"-
The value of "W3" is the PTS of the audio PES next to the audio PES to be thinned out, and by replacing with this value, the continuity of the stream is maintained in the audio PES before and after the audio PES to be thinned out.

The thinned audio PES is not stored in the buffer (not shown) of the MPEG decoder 30 and is discarded, so that the audio PES is not reproduced. PE stored in the buffer of the MPEG decoder 30
S is decoded and reproduced.

In step S53, the comparator 28 (see FIG. 3) compares the PTS of the video PES obtained from the PTS detection circuit 72 with the PTS of the audio PES obtained from the PTS detection circuit 52. When the PTS of the video PES is smaller than the PTS of the audio PES, the audio PES is thinned out to the video PES.
Since it means that the number is smaller than that of the ES decimation, in step S55, the threshold value set in the memory 58 is lowered by one to make it difficult to decimate the audio PES. In this embodiment in which the threshold is set to three levels, the threshold is changed from "AB" to "A" (A <A
B). Also, the PTS of the video PES is the audio PES.
If it is equal to or higher than the PTS, the audio PES thinning is less than the video PES thinning,
In step S57, the threshold value is increased by one to change from "AB" to "B" to facilitate the decimation of the audio PES (AB <B).

In this embodiment, the identifier and the threshold are set to three levels of "A", "AB", and "B" for convenience of description.
Actually, more detailed control is performed as the identifiers and thresholds in more stages.

In step S59, all audio P
It is determined whether or not the processing for ES is completed, and if not completed, the process returns to step S27. On the other hand, if the processing has been completed for all audio PESs, the operation ends.

If it is determined in step S33 that the identifier is not less than the threshold value, the audio PES is not thinned out. However, when thinning is performed before the audio PES, the PT of the audio PES is
It is necessary to correct S. However, if the PTS is already thinned out, the PTS has already been corrected (step S43), so the correction here (step S47) is not performed.

In step S45, it is determined whether or not the "thinning flag" is in the set state, and if it is not in the set state, the PTS of the audio PES is replaced with the value of "W1"-"W3" in step S47. . On the other hand, when the “thinning flag” is in the set state, step S47 is skipped. Then, in step S49, the "thinning flag" is reset.

In step S51, the audio PE
S is stored in the buffer of the MPEG decoder 30 and decoded. In this way, this audio PES is reproduced. Next, the process proceeds to step S53, but since the operation after step S53 is as described above, the description will be omitted.

As described above, according to the hard disk recorder 10 of this embodiment, when the catch-up reproduction is performed, the average level value of the audio signals included in the audio PES is calculated, and only the audio PES having the low average level value is calculated. To thin out to perform intermittent playback. Therefore, audio P
It is possible to reduce the influence of the thinned ES on the sound, and it is possible to suppress the unnatural sound due to the intermittent reproduction.

[Brief description of drawings]

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

FIG. 2 is a block diagram showing a configuration example of an MPEG encoder.

FIG. 3 is a block diagram illustrating a configuration example of a stream processing circuit.

FIG. 4 is a flowchart showing the operation of the MPEG encoder.

FIG. 5 is a flowchart showing the operation of the stream processing circuit.

FIG. 6 is a flowchart showing the operation of the stream processing circuit.

[Explanation of symbols]

10 ... Hard disk recorder 16 ... MPEG encoder 20 ... HDD (Hard Disk Drive) 22 ... Stream processing circuit 26 ... Controller 30 ... MPEG decoder 42 ... Average value calculation circuit 46 ... Comparator 50 ... Output control circuit 56 ... Comparator

   ─────────────────────────────────────────────────── ─── Continued front page    F-term (reference) 5C053 FA23 GB11 GB37 HA24 HA27                       JA01 JA07 JA12 KA22 KA25                 5D044 AB05 AB07 BC01 CC04 DE02                       DE03 DE12 DE49 DE55 FG09                       FG24 GK08 GK12                 5D045 DA20

Claims (3)

[Claims] 1. A recording / reproducing apparatus for recording an audio packet obtained by encoding an input audio signal in a packet system on a recording medium and reproducing the audio packet from the recording medium, wherein an average level value of the input audio signal is 1 Calculating means for calculating each length corresponding to a voice packet; embedding means for embedding average level information related to the average level value calculated by the calculating means in each voice packet; and the calculating means based on the average level information. A recording / reproducing apparatus comprising a first intermittent reproducing means for intermittently reproducing a voice packet. 2. The first intermittent reproduction means detects the average level information from each of the voice packets,
2. The method according to claim 1, further comprising: comparing means for comparing an average level value indicated by the average level information detected by the detecting means with a threshold value; and decoding means for decoding the voice packet having the average level value exceeding the threshold value. Recording / playback device. 3. A second intermittent reproduction means for intermittently reproducing the video packets obtained by encoding the input video signal in a packet system at regular intervals, and reproducing the audio packets and the video packets. Comparing means for comparing the time stamp of the audio packet with the time stamp of the video packet, and the result of the comparison means comparing the time stamp of the audio packet with the video packet. 3. The recording / reproducing apparatus according to claim 2, further comprising threshold control means for lowering the threshold when the time stamp of the audio packet is smaller than that of the video packet, and raising the threshold when the time stamp of the audio packet is smaller than the time stamp of the video packet. .