JP2000206994A - Voice encoder and decoder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent the generation of an audio omission caused by a sample which is possibly generated when audio materials synchronized to a time code is encoded, and can not constitute any audio frame nor is encoded. SOLUTION: If samples which constitute no voice frames exist in a coding device 10, dummy data equivalent to an insufficient amount are added to conduct a encoding for one extra frame. The information for insufficiency is included in coded bit streams as auxiliary information. In a decoder 20, the number of reading samples from output buffers 23 and 24 is controlled based on the sent auxiliary information to reproduce the samples which are normally omitted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding / decoding apparatus for encoding and decoding a digital audio signal synchronized with a video frame in audio frame units.

[0002]

2. Description of the Related Art Conventionally, when editing an audio material synchronized with a video (video) frame using a time code of a digital VTR or the like, an audio frame corresponding to a video frame frequency is formed into an audio frame. Encoding is performed for each number of samples. However, if the audio sample does not match the integral multiple of the audio frame, some samples cannot be composed of the audio frame, and the sample in that portion will not be encoded. FIG. 7 shows a case in which audio material is encoded based on a conventional time code. When encoding the audio material from the time code t1 to the time code t2, since the audio material is encoded by dividing the audio material into audio frames, there are cases where an audio frame cannot be formed. In this case, since the sample is not coded, audio loss occurs during decoding and reproduction.

[0003]

In the above case, the presence of uncoded samples of the audio material causes a loss of audio (audio gap) when decoded and reproduced. This problem is essential,
For example, this also occurs in material editing on a DVD.
In DVD, multi-story functions include multi-angle playback and parental playback, which allow the viewer to freely select the next scene.

In material editing that enables multi-angle reproduction, there is one audio material, and a plurality of video materials are prepared. In contrast, in parental playback, it is necessary to prepare a plurality of audio and video materials. For this reason, in material editing for multi-angle, it is possible to perform processing to eliminate uncoded samples of audio material, but in material editing for parentalism, this is impossible.

FIG. 3 and FIG. 4 are conceptual diagrams of parental reproduction on a DVD. FIG. 3 shows a case where the viewer returns to the main part B after selecting the part A and the part B from the main part A. PTS1 to PTS4 shown in the figure are presentation time stamps, at which time the decoding device performs display and reproduction.

[0006] In FIG. 3, the reproduction time and the contents are different between part A and part B for both video (video) and audio (audio). For this reason, since it is necessary to perform encoding separately for the main part A, the part A, and the part B, the above-mentioned problem of voice loss occurs in this case.

FIG. 4 shows this state. When audio is missing,
Since nothing is played, an audio gap, which is a silent part, is created. Therefore, an object of the present invention is to improve the encoding device and the decoding device so as to encode all the samples and thereby prevent a loss in reproduced sound of audio.

[0008]

As a means for achieving the above-mentioned object, an encoding apparatus uses a dummy sample corresponding to the number of missing samples for a frame in which a speech frame cannot be formed due to a lack of samples. Encoding is performed by adding data. The number of dummy data added at the time of encoding is included in the encoded bit stream as auxiliary data. Then, in the decoding device, the read buffer switching means controls so as to read out only the number of originally existing samples obtained from the auxiliary data in the coded bit stream, so that a sample which is normally lost is read out. It can be decoded and reproduced.

[0009]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of a speech coding apparatus and a decoding apparatus according to the present invention.
FIG. 1 is a block diagram showing one embodiment of the speech encoding apparatus and the decoding apparatus according to the present invention. The audio encoding device 10 includes a time code reading unit 11, a parameter input unit 12, a time code comparing unit 13, a frame number calculating unit 14, and an encoding unit 15.

On the other hand, the audio decoding device 20 comprises a decoding unit 21, a write buffer switching unit 22,
It comprises two output buffer memories 23 and 24 and read buffer switching means 25.

First, an embodiment of the present invention will be described below with reference to FIG. The time code reading means 11 sequentially reads the time code synchronized with the video. The parameter input means 12 receives time code information indicating the start and end times of encoding set by the user. The time code comparing means 13 compares the set start and end times supplied from the parameter input means 12 with the current time code supplied from the time code reading means 11, and when they match, the start signal , A stop signal is supplied to the encoding means 15 to start and end the encoding.

The number-of-frames calculating means 14 obtains the number Fn of the corresponding audio frames based on the unit frame period (time) set from the start time and the end time when the signal is supplied from the parameter input means 12. If there is a sample that cannot form an audio frame, a value obtained by adding 1 to the obtained number of audio frames (F
n + 1) to the encoding means 15.

Further, information on the number of samples Mn that is insufficient for the number of samples Sn required to form the audio frame corresponding to the set unit frame period is also sent to the encoding means 15. The encoding means 15 supplies the frame number signal, the start signal and the stop signal from the frame number calculating means 14 and the time code comparing means 13 together with the audio input data, and performs encoding.

[0014] As shown in FIG.
Samples that cannot form a frame (Sn-M
Regarding n), a dummy frame is added by the number Mn of missing samples to form an audio frame. The added dummy data is zero data.

Since the end sample of the audio material may not be completely faded out, the encoding is performed after performing the process of fading out the actual data (Sn-Mn) sample. At this time, information on the sample number Mn to which the missing dummy data has been added is incorporated into the coded bit stream as auxiliary data.

FIG. 2 shows an embodiment of the format of the coded bit stream. The format of the encoded bit stream includes synchronization code, side information, main information, auxiliary data into which information on the number of missing samples is inserted,
And a CRC code. For example, Dol
In the case of by-AC3, since Sn = 1536 samples, it is sufficient that the auxiliary data has a capacity to transmit 11-bit information.

Next, an embodiment of the speech decoding apparatus according to the present invention will be described with reference to FIG. Decoding means 2
1 is supplied with an encoded bit stream from the encoding means 15 of the audio encoding device 10, and performs decoding and extraction of auxiliary data. Then, information on the number Mn of insufficient samples taken out by the decoding means 21 is sent to the read buffer switching means 25.

In the write buffer switching means 22,
The buffers are switched so that the decoded data is alternately written to the two output buffer memories 23 and 24 in audio frame units. In the read buffer switching means 25, the output buffer selected by the write buffer switching means 22 is, for example, the output buffer 2
If it is 3, an output buffer 24 different from this output buffer 23 is selected, and samples are read from the output buffer 24.

At this time, the read buffer switching means 2
In 5, when the insufficient sample number Mn sent from the decoding means 21 is not zero, only the sample (Sn-Mn) that could not form a speech frame is read out, and the output of the speech frame is finish. This example is shown in FIG.

By doing so, there are no samples that cannot be encoded due to the problem of audio frames,
All audio material can be encoded. As a result, loss of audio during decoding can be prevented.

[0021]

According to the present invention, encoding is performed by adding dummy data for the number of missing samples to a sample that cannot form an audio frame, and the number of dummy data added at the time of encoding is used as auxiliary data. At the time of decoding, the read buffer switching means reads out only the number of originally existing samples obtained from the auxiliary data in the coded bit stream at the time of decoding, thereby preventing loss of audio during reproduction. I can do it. Further, the present invention is also effective as an audio material editing apparatus for realizing parental reproduction, which is one of DVD special reproductions.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of a speech encoding device and a decoding device according to the present invention.

FIG. 2 is a format example of an encoded bit stream output from the audio encoding device of the present invention.

FIG. 3 is a conceptual diagram of parental reproduction on a DVD.

FIG. 4 is a conceptual diagram of parental reproduction on a DVD.

FIG. 5 is a conceptual diagram of an encoding method in the encoding device of the present invention.

FIG. 6 is a conceptual diagram of an output buffer of the decoding device of the present invention.

FIG. 7 is a diagram illustrating a method of encoding audio material using a time code.

[Explanation of symbols]

 Reference Signs List 10 audio encoding device 11 time code reading means 12 parameter input means 13 time code comparing means 14 frame number calculating means 15 encoding means 20 audio decoding apparatus 21 decoding means 22 writing buffer switching means 23, 24 output buffer memory 25 reading Buffer switching means Fn Number of audio frames Mn Number of missing samples (dummy data) Sn Required number of samples

Claims (5)

[Claims] 1. A time code reading means for reading a time code synchronized with an external video signal; a time code parameter input means for inputting a time code indicating the start and end of encoding; Frame number calculation means for calculating the number of audio frames in the period input to the time code parameter input means based on the number of frames per time; and time code for comparing a current time code with the input time code parameter An audio encoding apparatus, comprising: a comparing unit; and an encoding unit that supplies an output of the frame number calculating unit and the time code comparing unit together with audio input data and performs encoding. 2. The audio encoding apparatus according to claim 1, wherein said frame number calculating means is a unit of a speech frame in which a period calculated from the start and end time codes of said encoding is a unit of audio encoding breaks. A voice encoding device which calculates whether the time is an integral multiple of the time, and outputs the number of audio frames for performing extra coding by one frame if the time is not an integral multiple. 3. A speech encoding apparatus according to claim 1, wherein said frame number calculation means supplies sample number shortage information indicating a lacking sample number to said encoding means when constructing a speech frame. And the encoding means outputs an encoded bit stream including the sample number shortage information. 4. The audio encoding apparatus according to claim 1, wherein said encoding means performs dummy encoding for a number of samples that are insufficient to form an audio frame when encoding one extra frame. A speech encoding apparatus, wherein the speech frame is configured by adding data. 5. A coded bit stream is supplied by the voice coding apparatus according to claim 1 to perform decoding, and sample shortage number information indicating a deficient number of samples is extracted from the coded bit stream. Decoding means for performing decoding, two output buffer memories to which decoded data from the decoding means are supplied, a write buffer switching means for switching writing to the output buffer memory, and sample shortage number information from the decoding means. Is supplied,
An audio decoding device comprising: a read buffer switching unit that controls the number of data read from the output buffer memory to switch according to the sample shortage number information.