JP4511860B2 - ADTS frame audio FS circuit, AAC re-encoder circuit, and audio FS circuit - Google Patents

Description

  The present invention relates to an ADTS frame audio FS circuit, an AAC re-encoder circuit, and an audio FS circuit, and in particular, an ADTS frame audio FS circuit, an AAC re-encoder circuit, and an audio FS that perform an FS operation so as to maintain a unit corresponding to an ADTS frame. Regarding the circuit.

  Conventionally, an FS (Frame Synchronizer) circuit for baseband audio has been a synchronization circuit that synchronizes by the number of samples that does not take into account the unit to be encoded, including every sample.

  In Patent Document 1, a video bit stream obtained by encoding input image information using a video encoder and an audio bit stream obtained by encoding input audio information using an audio encoder are synchronized in a system encoder while multiplexed. In the multiplexed synchronization method to be generated, at least one of the video encoder and the audio encoder sends auxiliary information for synchronization necessary for generating a system bitstream to the system encoder, and the system encoder uses the auxiliary information for synchronization. A multiplexed synchronization method for generating a bitstream is described.

In Patent Document 2, based on the control of the system control unit, an input video / audio signal is compression-encoded by an encoder unit, encoded data that has been compression-encoded is decoded by a decoder unit, and reproduced video / audio decoded by the decoder unit In the re-encoding device in which the signal is generated into a signal that can be displayed on the display means by the video signal generation unit and output to the outside, and the playback video / audio signal is supplied to the encoder unit. Means for multiplexing an encoder unit control command for controlling the encoder unit on the audio signal, and separating the encoder unit control command multiplexed on the input video / audio signal in the encoder unit, and using the encoder unit control command for the input video / audio signal The video signal generation unit masks the encoder unit control command multiplexed on the reproduced video / audio signal. It describes a recoding device including means for outputting the part.
JP 2003-274362 A JP 11-239326 A

  However, in the above-described prior art, in a device that performs re-encoding using the preceding speech coding information, the baseband speech and the preceding speech coding information are out of phase and thus cannot be used, leading to speech degradation. It was.

  Further, even if the number of data to be corrected by the FS operation is set as a speech coding unit, since the number of data to be corrected at a time is large, speech noise is generated.

  Therefore, an object of the present invention is to perform FS so as to maintain a unit corresponding to an ADTS frame at the time of speech encoding, and to correct so that speech noise does not occur when correction is performed by the FS operation. To do.

According to a first aspect of the present invention , in an ADTS frame audio FS circuit that outputs a baseband audio signal generated by decoding a compressed audio signal as an FS baseband audio signal in synchronization with a reference clock signal. the base when the band speech signal has been reference ADTS sync signal and Gaz generated based on the ADTS sync signal and the reference clock signal is a signal indicating the head of a unit when encoding, determines that deviation occurs If the deviation is greater than or equal to a predetermined value, the number of correction samples per time is limited to a unit smaller than the ADTS frame , and the deviation is eliminated by correcting at a plurality of times, and then the FS baseband. An ADTS frame audio FS circuit is provided that outputs an audio signal .
Further, according to the second aspect of the present invention, in an AAC re-encoder circuit which is a circuit for compressing an input signal, an encoding which is connected to the ADTS frame audio FS circuit and is necessary for the compression is performed. An AAC re-encoder circuit is provided that compresses using encoded information output from an AAC decoder circuit as a parameter.
According to a third aspect of the present invention, an audio FS circuit synchronized with an ADTS frame includes an AAC decoder circuit that decodes a compressed signal encoded with MPEG AAC;
An ADTS sync output circuit that outputs an ADTS sync flag indicating a head position for each unit when the baseband audio signal output from the AAC decoder circuit is encoded or decoded; A code output circuit that outputs coding information that is necessary information when performing the coding, a reference ADTS sync generation circuit that generates a reference ADTS sync signal from a reference clock signal, and the baseband audio signal as the reference ADTS An ADTS frame audio FS circuit that outputs an FS baseband audio signal in synchronization with the sync;
An AAC re-encoder circuit that receives and re-encodes the encoded information and the FS baseband audio signal, and the ADTS frame audio FS circuit encodes the baseband audio signal. When the ADTS sync signal, which is a signal indicating the head of the signal, and the reference ADTS sync signal generated based on the reference clock signal are shifted, it is determined that a shift has occurred. An audio FS circuit is provided in which the number of correction samples per unit is limited to a unit smaller than the ADTS frame, and correction is performed at a plurality of times.

  According to the present invention, since the audio FS synchronized with the ADTS frame is performed, the phase of the encoded information extracted by the audio decoder after the FS and the baseband audio signal are matched, so that the compressed signal is decoded. When re-encoding the baseband signal generated in this way, it is possible to re-encode using the encoding information extracted by the decoder circuit as the encoding parameters required for encoding. Can be suppressed.

  According to the present invention, in correction of baseband audio data corresponding to one ADTS, the number of correction samples per time is limited and divided into a plurality of times, so that it is possible to reduce audio noise at the time of FS in ADTS units.

  DESCRIPTION OF THE PREFERRED EMBODIMENTS The best mode for carrying out the present invention will be described below with reference to the accompanying drawings.

  FIG. 1 is a block diagram showing the configuration of the first exemplary embodiment of the present invention.

  As shown in FIG. 1, the audio FS circuit of this embodiment includes an AAC (Advanced Audio Coding) decoder circuit 1 with a code output function, a reference ADTS (Audio Data Transport Stream) sync generation circuit 5, and an ADTS frame audio FS. A circuit 6 and an AAC re-encoder circuit 7 are provided.

  The AAC decoder circuit 1 with a code output function includes an AAC decoder circuit 3, an ADTS sink output circuit 4, and a code output circuit 32.

  The AAC decoder circuit 3 receives and decodes a compressed signal encoded by MPEG (Moving Picture Expert groups) AAC, and outputs a baseband audio signal.

  The ADTS sync output circuit 4 outputs an ADTS sync signal indicating the head of a frame (1024 samples) when the baseband audio signal is encoded or decoded.

  The code output circuit 2 outputs encoded information extracted from the compressed signal.

  The reference ADTS sync generation circuit 5 outputs a reference ADTS sync signal based on the reference clock signal.

  The ADTS frame audio FS circuit 6 receives the baseband audio signal, the ADTS sync signal, and the reference ADTS sync signal, performs an audio FS operation synchronized with the ADTS frame, and an FS that is a baseband audio signal after the ADTS frame audio FS operation. A baseband signal and an FS correction signal are output.

  The AAC re-encoder circuit 7 receives the encoded information, the FS baseband audio signal, and the FS correction signal, performs re-encoding, and outputs a re-compressed signal.

  FIG. 2 is a timing chart showing the waveform of the output signal of the AAC decoder circuit 1 with a code output function.

  The AAC decoder circuit 3 receives the compressed signal encoded by MPEG AAC and outputs the baseband audio signal shown in FIG.

  The ADTS sync output circuit 4 outputs an ADTS sync signal shown in FIG. 2C, which indicates the head of the encoded frame synchronized with the baseband audio signal.

  The code output circuit 42 outputs encoded information shown in FIG. 2B synchronized with the baseband audio signal. This encoded information is in units of 1024 samples.

  FIG. 3 is a timing chart showing the waveform of the reference signal input to the ADTS frame audio FS circuit 6.

  The reference ADTS sync generation circuit 5 outputs the reference ADTS sync signal shown in FIG. 3B based on the reference clock signal shown in FIG.

  FIG. 4 is a timing chart for explaining the correction processing.

  In the following, a correction operation when the drive reference clock is slower (the frequency is lower) than the decode clock and the reference ADTS sync signal is delayed from the ADTS sync signal will be described.

  The ADTS frame audio circuit 6 enters a correction process when the phase difference between the input ADTS sync signal and the reference ADTS sync signal is equal to or greater than a preset threshold value.

  In FIG. 4, time T1 corresponds to the timing of entering the correction process. During the subsequent correction operation, the AAC re-encoder circuit 7 cannot use the encoded information.

  In the correction process, the correction amount per time is set to about m (natural number) samples in which no sound noise is generated, and this is repeated n (natural number) times, and ADTS when the final correction amount m × n is speech-coded. This is performed so that the frame is equivalent to 1024 samples.

  The time T2 in FIG. 4 indicates the time when the correction for 1024 samples is completed, and after that, the phase of the encoded information coincides, so that the AAC re-encoder circuit 7 can use the encoded information again. It becomes possible.

  At the same time, an FS correction signal indicating that correction for reducing data is performed is output to the AAC re-encoder circuit 7.

  On the other hand, when the reference clock is earlier (the frequency is higher) than the decode clock and the reference ADTS sync signal is earlier than the ADTS sync signal, an FS correction signal indicating that correction for increasing data is performed is output.

  When the AAC re-encoder circuit 7 receives the FS correction signal indicating that the correction for reducing the data has been performed, the AAC re-encoder circuit 7 discards the encoded information for one frame.

  Further, when the AAC re-encoder circuit 7 receives the FS correction signal indicating that the correction for increasing the data has been performed, the AAC re-encoder circuit 7 repeatedly uses the encoded information for one frame and re-encodes it.

  In the above embodiment, m × n is performed so as to correspond to 1024 samples. However, the present invention is not limited to this method, and m × n + 1 may be performed so as to correspond to 1024 samples.

  FIG. 5 is a chart for explaining the difference between the prior art and the present invention.

  T3 indicates the time when the phase of the FS writing system and the reading system is shifted by one sample, T1 indicates the time when the phase of the FS writing system and the reading system is shifted by 1024 samples, and time T4 is the time when the FS writing system is And the time when the phase of the readout system is shifted again by one sample, and time T2 indicates the time when the audio FS circuit according to the present invention ends the correction operation.

  As shown in FIG. 5, in the prior art, since one sample is corrected at time T3 shifted by one sample, the encoded information cannot be used until time T1 when a shift of 1024 samples corresponding to 1 ADTS occurs thereafter.

  On the other hand, in the FS circuit proposed in the present application, since correction is performed at a shift of 1024 samples (conceptual), the time period in which encoded information cannot be used is short.

It is a block diagram which shows the structure of the 1st Embodiment of this invention. It is a timing chart which shows the waveform of the output signal of AAC decoder circuit 1 with a code output function. 4 is a timing chart showing a waveform of a reference signal input to an ADTS frame audio FS circuit 6; It is a timing chart for demonstrating a correction process. It is a chart for demonstrating the difference with a prior art and this invention.

Explanation of symbols

1 AAC decoder circuit with code output function 2 Code output circuit 3 AAC decoder circuit 4 ADTS sync output circuit 5 Reference ADTS sync generation circuit 6 ADTS frame audio FS circuit 7 AAC re-encoder circuit

Claims (5)

In an ADTS frame audio FS circuit that synchronizes a baseband audio signal generated by decoding a compressed audio signal with a reference clock signal and outputs it as an FS baseband audio signal,
When an ADTS sync signal that is a signal indicating the head of a unit when the baseband audio signal is encoded and a reference ADTS sync signal generated based on the reference clock signal are shifted, it is determined that a shift has occurred, When the deviation is equal to or greater than a predetermined value, the number of correction samples per time is limited to a unit smaller than the ADTS frame , and the deviation is eliminated by correcting at a plurality of times, and then the FS baseband audio signal An ADTS frame audio FS circuit. 2. The ADTS frame audio FS circuit according to claim 1, wherein the number of correction samples per one time is about a sample at which audio noise does not occur. 3. In an AAC re-encoder circuit that compresses an input signal,
It is connected to the ADTS frame voice FS circuit according to claim 1 or 2, and compresses using encoding information output from an AAC decoder circuit as an encoding parameter which is information necessary for the compression. AAC re-encoder circuit. The audio FS circuit synchronized with the ADTS frame is
An AAC decoder circuit for decoding a compressed signal encoded by MPEG AAC;
An ADTS sync output circuit that outputs an ADTS sync flag indicating a head position for each unit when the baseband audio signal output from the AAC decoder circuit is encoded or decoded;
A code output circuit for outputting encoded information which is information necessary for encoding extracted from the compressed signal in the AAC decoder circuit;
A reference ADTS sync generation circuit for generating a reference ADTS sync signal from the reference clock signal;
An ADTS frame audio FS circuit that outputs the FS baseband audio signal by synchronizing the baseband audio signal with the reference ADTS sync;
An AAC re-encoder circuit that receives and re-encodes the encoded information and the FS baseband audio signal ;
When the ADTS frame audio FS circuit deviates from an ADTS sync signal, which is a signal indicating the head of a unit when the baseband audio signal is encoded, and a reference ADTS sync signal generated based on the reference clock signal The voice FS is characterized in that it is determined that a deviation has occurred, and when the deviation is equal to or greater than a predetermined value, the number of correction samples per time is limited to a unit smaller than the ADTS frame and correction is performed at a plurality of times. circuit. 5. The audio FS circuit according to claim 4, wherein the number of correction samples per time is about a sample in which no audio noise is generated.