JP3606388B2 - Audio data reproducing method and audio data reproducing apparatus - Google Patents

Description

を満足するとき、連続していると判断する。なお（１）式中、ｄｉｒｅｃｔｉｏｎ は前方向（Ｆｏｒｗａｒｄ）のとき１、後方向（Ｒｅｖｅｒｓｅ）のとき−１である。
【００２７】
変速再生時に帯域制限して、耳障りな高域成分を出力しないように制御する方法として、例えば第１に全てのブロツクの帯域を例えば周波数１０〔ＫＨｚ〕以下に制限したり、第２に全てのフレームの境界のブロツク（Ｂｌｏｃｋ０、Ｂｌｏｃｋ５）の帯域を制限したり、第３に不連続点でフレームＦＮＢ（ｘ−１）側の最後のブロツクと、ＦＮＢ（ｘ） 側の最初のブロツクの帯域を制限することが考えられる。
【００２８】
なお帯域制限部３１は例えばＤＳＰ（デイジタル信号処理装置）構成でなり、図２に示すような帯域制限処理のプログラム手順を内蔵し実行する。ここでモード信号ＦＩＬＴＭＤ＝０の時、全てのブロツクの上限値ＢＡＮＤＬＭを越えるバンドのデータを値「０」に置き換える。またモード信号ＦＩＬＴＭＤ＝１の時、全てのフレームの境界のブロツク（Ｂｌｏｃｋ０、Ｂｌｏｃｋ５）で、設定値ＳＨＬＩＭ を越えたバンドのデータを値「０」に置き換える。さらにモード信号ＦＩＬＴＭＤ＝２の時、不連続点の境界のブロツク（Ｂｌｏｃｋ０）で、設定値ＳＨＬＩＭ を越えたバンドのデータを値「０」に置き換える。
【００２９】
さらにまたモード信号ＦＩＬＴＭＤ＝３の時、不連続点の境界のブロツク（Ｂｌｏｃｋ５）で、設定値ＳＨＬＩＭ を越えたバンドのデータを値「０」に置き換える。システムとして、ブロツクの上限値ＢＡＮＤＬＭ、モード信号ＦＩＬＴＭＤが設定される。モード信号ＦＩＬＴＭＤに関しては、不連続点においてフレームＦＮＢ（ｘ−１） ではモード信号ＦＩＬＴＭＤ＝３、ＦＮＢ（ｘ） でモード信号ＦＩＬＴＭＤ＝２になるように制御することで、上述した第３の制御方法が実現される。
【００３０】
以上の構成によれば、再生するサブバンドデータの周波数の上限値を設定し、デコード時に上限値の周波数を越える帯域のサブバンドデータを値「０」に置き換えて、再生音を帯域制限するようにしたことにより、変速再生時に境界ブロツクで生じる異音の発生を有効に回避して再生音の音質を向上し得る。
【００３１】
なお上述の実施例においては、再量子化後のサブバンド成分のデータについて帯域制限処理した場合について述べたが、同様の処理を再量子化部の前で帯域制限処理するようにしても、上述の実施例と同様の効果を実現できる。
【００３２】
また上述の実施例においては、デイジタルオーデイオ信号を周波数分解して符号化する符号化方式の例として、サブバンド符号化方式を用いた場合について述べたが、本発明はこれに限らず、周波数を線形に処理して分割する離散コサイン変換（ＤＣＴ）等の直交変換符号化を含む、他のすべての符号化方式について適応できる。
【００３３】
【発明の効果】
上述のように本発明によれば、オーデイオ信号を周波数分割して符号化し、複数の符号化ブロツクを１記録単位として順次記録したオーデイオデータを、記録時の記録単位の順序と異なる順序で再生する際に、当該再生する符号化データの周波数の上限値を設定し、デコード時に上限値の周波数を越える帯域で記録単位のうち隣接する符号化ブロツクについてのみ、当該符号化ブロツクに含まれる符号化データを所定値に置き換えて、再生音を帯域制限するようにしたことにより、記録単位間に生じる異音の発生を有効に回避して、再生音の音質を向上し得るオーデイオデータ再生方法及びオーデイオデータ再生装置を実現できる。
【図面の簡単な説明】
【図１】本発明によるオーデイオデータ再生方法及びオーデイオデータ再生装置を適用したデイジタルオーデイオ機器を示すブロツク図である。
【図２】図１のデイジタルオーデイオ機器で実現する帯域制限処理を示すプログラムリストである。
【図３】サブバンド符号化復号化方式の原理の説明に供するブロツク図である。
【図４】ビデオフレームとサブバンドブロツクとの関係の説明に供する略線図である。
【図５】従来のサブバンド符号化によるデイジタルオーデイオ機器を示すブロツク図である。
【図６】変速再生におけるブロツクの不連続性の説明に供する略線図である。
【図７】フイルタのタツプ数と隣接ブロツクとの関係の説明に供する略線図である。
【図８】不連続点での影響の範囲の説明に供する略線図である。
【符号の説明】
１……エンコーダ、２……デコーダ、３……分割フイルタ（ＢＰＦ）、４……量子化器（Ｑ）、５……ビツトストリーム、６……逆量子化器（Ｑ^−１）、７……合成フイルタ（ＢＰＦ）、１０、３０……デイジタルオーデイオ機器、１１……分割フイルタ部、１２……ビツト割当部、１３……量子化部、１４……ビツトストリームフオーマツト部、１５……ビツトストリームデコード部、１６……ビツト割当デコード部、１７……逆量子化部、１８……合成フイルタ部、１９……シスコンインターフエース部、２０……タイミング発生部、３１……帯域制限部。[0001]
【table of contents】
The present invention will be described in the following order.
Industrial application fields Conventional technologies (Figs. 3-5)
Problems to be Solved by the Invention (FIGS. 5 to 8)
Means for solving the problem (FIG. 1)
Action (Fig. 1)
Example (FIGS. 1, 2, 4, 6, and 7)
Effect of the Invention
[Industrial application fields]
The present invention is suitable for application in audio data reproduction method and audio data reproduction apparatus, particularly when the digital audio signal is frequency-divided and encoded and is reproduced with variable speed.
[0003]
[Prior art]
Conventionally, in order to reduce the amount of data for a digital audio signal, a subband encoding method, a transform encoding method, and the like are used. These utilize the property that audio signals are unevenly distributed in the frequency axis direction and reduce redundant information for encoding. FIG. 3 shows the principle of a subband encoding / decoding system that divides a digital audio signal into M parts. First, on the encoder 1 side, an input digital audio signal is divided into M frequency domain subband components by M division filters (BPF) 3.
[0004]
Each subband component is assigned a quantization level so as to have a constant data amount as a whole in accordance with the amount of generated data. The data of each subband component is quantized by the quantizer (Q) 4 at the assigned quantization level, and rearranged into the format of the bit stream 5. Bit number information and maximum value information are also recorded on the bit stream 5 as the quantization level of each subband component data.
[0005]
On the decoder 2 side, from the state of the bit stream 5, the data is rearranged into the data string of each subband component, and the original data is dequantized in the inverse quantizer (Q ⁻¹ ) 6 according to the recorded quantization level. Return to. Then, the original digital audio signal is decoded by the M synthesis filters 7 from the sub-quantized subband component data.
[0006]
FIG. 4 shows a case where a digital audio signal sampled at 48 [KHz] corresponding to a PAL (625/50) video frame is recorded by a subband encoding method having a division number of 32 and a code length of 320 samples. (FIG. 4A and FIG. 4B). The data of 32 subband components constitute one subband block with 10 samples (FIG. 4C).
[0007]
In each subband block, the quantization level at the time of encoding, that is, the bit number information (ALLOC) indicating the number of bits of each subband data in 4 bits, and the maximum value of data in each subband block are 6 Maximum value information (SF) indicated by bits is simultaneously recorded (FIG. 4D). The sample data of each sub-band usually varies in length because the quantization level differs depending on the band. Here, the total information amount is constant.
[0008]
Information such as bit number information (ALLOC) and maximum value information (SF) is the same in each subband block. For example, when the video frame is a PAL (625/50) system, the audio data in one frame is 1920 samples, so one frame is composed of 6 subband blocks. Here, as shown in FIG. 4 (E), the frame number FN (0 to 32) is recorded as a block header (ANC) at the head of the subband block.
[0009]
FIG. 5 shows the configuration of the digital audio device 10 using such a conventional subband encoding method. In practice, a 16-bit digital audio signal S1 sampled at, for example, 48 [KHz] sent from an audio processor (not shown) at the time of encoding is sent to 32 divided filter sections (Analysis Filter Bank) 11. The subband component is divided into 32 pieces of digital data S2, and a bit allocation unit (Bit Allocation) 12 assigns a quantization level so as to obtain a constant data amount as a whole according to the amount of generated data.
[0010]
Next, the data S2 of each subband component is respectively quantized at the quantization level assigned by the quantizing unit (Quantize) 13, and the bit stream formatting unit (Bit Stream Formatting) 14 as described above with reference to FIG. The bit stream data S3 is sent to an ECC block (not shown) to which information for the next error correction is added and recorded or transmitted.
[0011]
The bit stream data S4 whose error has been corrected at the time of decoding is distributed to the data of each subband block by a bit stream decoding unit (Decoding of Bit Stream) 15, and a bit allocation decoding unit (Decoding of Bit Allocation) 16 and an inverse quantization unit (Dequantize) 17 is dequantized, and each subband component data S5 is synthesized by 32 synthesis filter sections 18 and sent to an audio processor (not shown) as a digital audio signal S6. Is done.
[0012]
The processing timing of the division filter unit 11, the bit allocation unit 12, the quantization unit 13, the bit stream format unit 14, the bit stream decoding unit 15, the bit allocation decoding unit 16, the inverse quantization unit 17 and the synthesis filter unit 18 is as follows. Controlled by a system controller (not shown) through a syscon interface unit (syscon interface) 19, an operation clock is generated by a timing generator unit (timing generator) 20.
[0013]
[Problems to be solved by the invention]
By the way, in the digital audio apparatus using the above-described subband encoding method, the order of the compressed data may be different from the recorded order in the playback mode other than the variable speed playback, that is, the 1 × speed playback. Here, as a unit constituted by a plurality of blocks, for example, as shown in FIGS. 6 (A) to 6 (D), a case of jumping (JUMP) in frame units is considered. When jumping in units of frames in this way, adjacent blocks at the boundary of the frame are different from the blocks at the time of recording.
[0014]
Here, when decoding is performed using the synthesis filter unit 18 as in the subband encoding method, data for the same number of samples as the number of taps of the synthesis filter unit 18 is used in the calculation. Therefore, for example, as shown in FIG. 7 (A), when the composite filter unit 18 is constituted by a 512-type FIR type filter, the block adjacent to the block boundary is 512, as shown in FIG. 7 (B). Data of 256 samples, which is half of the tap, overlap.
[0015]
However, as described above, when subband blocks are discontinuous during variable speed reproduction, as shown in FIG. 8, samples of data of adjacent subblocks are included in the filter processing. As a result, the data recorded under the influence of adjacent sub-blocks at discontinuous points cannot be reproduced correctly, it can be heard as an annoying high-frequency component, easily detected as an abnormal sound, and the sound quality of the reproduced sound deteriorates. Hot.
[0016]
The present invention has been made in consideration of the above points. When audio data obtained by frequency-division encoding and sequentially recording a plurality of encoding blocks as recording units is reproduced in an order different from that at the time of recording, abnormal noise is generated. The present invention is intended to propose an audio data reproducing method and an audio data reproducing apparatus for preventing sound from occurring and obtaining sound quality that is easy to hear.
[0017]
[Means for Solving the Problems]
In order to solve such a problem, in the present invention, the digital audio signal (S3) is obtained by frequency-dividing and encoding the digital audio signal (S1) and sequentially recording a plurality of encoding blocks each including encoded data as one recording unit. In the audio data reproduction method for reproduction, when audio data (S4) is reproduced in an order different from the order of recording units at the time of recording, an upper limit value of the frequency of the encoded data to be reproduced is set, and the upper limit value is set at the time of decoding. Only in the coding block adjacent to the recording unit in the band exceeding the frequency, the coded data included in the coding block is replaced with a predetermined value to limit the band of the reproduced sound.
[0018]
In the present invention, the digital audio signal (S1) is frequency-divided and encoded, and audio data reproduction for reproducing the audio data (S3) sequentially recorded by using a plurality of encoding blocks each including encoded data as one recording unit. In the apparatus (30), when the audio data (S4) is reproduced in an order different from the order of recording units at the time of recording, an upper limit value of the frequency of the encoded data to be reproduced is set, and the frequency of the upper limit value is set at the time of decoding. Only for the adjacent encoded block in the recording unit in the band exceeding the band, the encoded data included in the encoded block is replaced with a predetermined value, and band limiting means (31) for limiting the band of the reproduced sound is provided.
[0019]
[Action]
When audio data (S3, S4), in which the audio signal (S1) is frequency-divided and encoded and sequentially recorded with a plurality of encoding blocks as one recording unit, is reproduced in an order different from the order of recording units at the time of recording. In the band limiting means (31), the upper limit value of the frequency of the encoded data to be reproduced is set, and only the adjacent encoding block in the recording unit in the band exceeding the upper limit frequency at the time of decoding is set as the encoding block. By replacing the included encoded data with a predetermined value so as to limit the band of the reproduced sound, it is possible to effectively avoid the occurrence of abnormal sound generated between recording units and improve the sound quality of the reproduced sound.
[0020]
【Example】
Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.
[0021]
In FIG. 1, in which parts corresponding to those in FIG. 5 are assigned the same reference numerals, reference numeral 30 denotes a digital audio device using the subband encoding method according to the present invention. In practice, the digital audio signal S1 sent from the audio processor at the time of encoding is divided into subband component digital data S2 by the division filter unit 11, and the bit allocation unit 12 keeps the whole constant according to the amount of generated data. The quantization level is assigned so that the amount of data becomes.
[0022]
Next, the data S2 of each subband component is respectively quantized at the quantization level assigned by the quantization unit 13, and is rearranged by the bit stream format unit 14 as described above with reference to FIG. Then, it is sent to an ECC block (not shown) to which information for error correction is added and recorded or transmitted. The bit stream data S4 that has been error-corrected at the time of decoding is distributed to the data of each subband block by the bit stream decoding unit 15, dequantized through the bit allocation decoding unit 16 and the inverse quantization unit 17, and the band limiting unit (Limiter) ) 31, the sub-band component data S10 is synthesized by the synthesis filter unit 18 and sent as a digital audio signal S11 to an audio processor or the like.
[0023]
In the above configuration, jumping in units of frames as shown in FIG. 6 as the variable speed reproduction is different from the block when the adjacent block is recorded at the boundary of the frame. When the synthesis filter unit 18 is used as in the case of subband coding, as described above with reference to FIG. 7, the same number of samples as the number of taps of the synthesis filter unit 18 are used in the calculation. Then, the recorded data cannot be reproduced because it is affected by the adjacent block, and it is easily detected as an abnormal sound.
[0024]
Therefore, in this embodiment, the upper limit of the subband to be reproduced is set, and the subband data having a frequency equal to or higher than the set value is replaced with a predetermined value of, for example, a value “0” at the time of decoding the code to limit the band of the reproduced sound That makes it easy to hear. On the decoder side, the data is rearranged into the data sequence of each subband from the state of the bit stream and returned to the original data according to the recorded quantization level. Then, the sub-quantized data of each sub-band component is obtained.
[0025]
Here, the band limiting unit 31 of this embodiment replaces the data of the corresponding subband with the value “0” from the upper limit value BANDLM of the subband to be reproduced set by the system controller or the like and the mode signal FILTMD. When the subband component data S10 whose band is limited in this way is decoded into a digital audio signal by the synthesis filter unit 18, a reproduction sound with a limited frequency band is obtained.
[0026]
Further, the discontinuity of the frame can be easily detected from the frame number FNB written in the block header ANC (FIG. 4E) of the subband block. Such timing signals can also be easily generated by a reproduction signal processing circuit or a servo circuit of the reproduction system. The actual discontinuity is detected by extracting the value of the frame number FNB (x) from the block header ANC at the head of the subband block, and the frame number FNB (x) is expressed by the following equation:

When satisfying, it is determined to be continuous. In the equation (1), direction is 1 for the forward direction and -1 for the backward direction.
[0027]
As a method for controlling the band so that no disturbing high-frequency components are output at the time of variable speed reproduction, for example, firstly, the band of all blocks is limited to, for example, a frequency of 10 [KHz] or less, or secondly, The bandwidth of the block at the boundary of the frame (Block 0, Block 5) is limited. Third, the bandwidth of the last block on the frame FNB (x-1) side and the first block on the FNB (x) side at the discontinuity point. It is possible to restrict.
[0028]
The band limiting unit 31 has a DSP (digital signal processing apparatus) configuration, for example, and incorporates and executes a program procedure for band limiting processing as shown in FIG. Here, when the mode signal FILTMD = 0, the data of the band exceeding the upper limit value BANDLM of all the blocks is replaced with the value “0”. When the mode signal FILTMD = 1, the band data exceeding the set value SHLIM is replaced with the value “0” in the block (Block 0, Block 5) of the boundary of all frames. Further, when the mode signal FILTMD = 2, the data of the band exceeding the set value SHLIM is replaced with the value “0” in the block (Block 0) at the boundary of the discontinuous points.
[0029]
Further, when the mode signal FILTMD = 3, the data of the band exceeding the set value SHLIM is replaced with the value “0” in the block (Block 5) of the boundary of the discontinuous points. As a system, a block upper limit value BANDLM and a mode signal FILTMD are set. With respect to the mode signal FILTMD, the above-described third control method is performed by controlling the mode signal FILTMD = 3 in the frame FNB (x−1) and the mode signal FILTMD = 2 in FNB (x) at the discontinuity point. Is realized.
[0030]
According to the above configuration, the upper limit value of the frequency of the subband data to be reproduced is set, and the subband data in the band exceeding the upper limit frequency at the time of decoding is replaced with the value “0” to limit the band of the reproduced sound. By doing so, it is possible to effectively avoid the generation of abnormal noise that occurs in the boundary block during variable speed reproduction and improve the sound quality of the reproduced sound.
[0031]
In the above-described embodiment, the case where the band limitation process is performed on the subband component data after the requantization has been described. However, the same process may be performed before the requantization unit. The same effect as the embodiment can be realized.
[0032]
In the above-described embodiment, the case where the subband coding method is used as an example of the coding method for frequency-decomposing and coding the digital audio signal is described. However, the present invention is not limited to this, and the frequency is not limited. It can be applied to all other coding schemes including orthogonal transform coding such as discrete cosine transform (DCT) that is linearly processed and divided.
[0033]
【The invention's effect】
As described above, according to the present invention, the audio signal is frequency-divided and encoded, and the audio data sequentially recorded with a plurality of encoding blocks as one recording unit is reproduced in an order different from the order of the recording units at the time of recording. At this time, the upper limit value of the frequency of the encoded data to be reproduced is set, and the encoded data included in the encoded block is included only in the adjacent encoded block in the recording unit in the band exceeding the upper limit frequency at the time of decoding. Is replaced with a predetermined value to limit the band of the playback sound, effectively avoiding the generation of abnormal noise between recording units and improving the sound quality of the playback sound and audio data A playback device can be realized.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a digital audio apparatus to which an audio data reproducing method and an audio data reproducing apparatus according to the present invention are applied.
FIG. 2 is a program list showing a bandwidth limitation process realized by the digital audio device of FIG. 1;
FIG. 3 is a block diagram for explaining the principle of a subband coding / decoding system;
FIG. 4 is a schematic diagram for explaining a relationship between a video frame and a subband block;
FIG. 5 is a block diagram showing a digital audio device using conventional subband coding.
FIG. 6 is a schematic diagram for explaining block discontinuity in variable speed reproduction;
FIG. 7 is a schematic diagram for explaining the relationship between the number of taps of a filter and adjacent blocks.
FIG. 8 is a schematic diagram for explaining a range of influence at a discontinuous point;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Encoder, 2 ... Decoder, 3 ... Split filter (BPF), 4 ... Quantizer (Q), 5 ... Bit stream, 6 ... Inverse quantizer (Q ^<-1> ), 7 ... ... Synthetic filter (BPF), 10, 30 ... Digital audio equipment, 11 ... Split filter section, 12 ... Bit allocation section, 13 ... Quantization section, 14 ... Bitstream format section, 15 ... Bit Stream decoding unit, 16... Bit allocation decoding unit, 17 .. inverse quantization unit, 18... Synthesis filter unit, 19 .. syscon interface unit, 20 .. timing generation unit, 31.

Claims (12)

In an audio data reproducing method for reproducing audio data in which a digital audio signal is frequency-divided and encoded, and a plurality of encoded blocks each including encoded data are sequentially recorded as one recording unit,
When reproducing in an order different from the order of the recording unit at the time of recording the audio data, the recording in a band that sets the upper limit of the frequency of the coded data to the reproduction exceeds the frequency of the upper limit value when decoding An audio data reproducing method characterized in that, for only the adjacent encoded block in a unit, the encoded data included in the encoded block is replaced with a predetermined value to limit the band of the reproduced sound. 2. The audio data reproducing method according to claim 1, wherein the predetermined value for replacing the encoded data is a value “0”. It said plurality of said one recording unit of the encoding block is audio data reproducing method according to claim 1, characterized in that it is selected according to a predetermined recording unit of the video data. In an audio data reproducing method for reproducing audio data in which a digital audio signal is frequency-divided and encoded, and a plurality of encoded blocks each including encoded data are sequentially recorded as one recording unit,
When reproducing in the order different from the order of the recording unit at the time of recording the audio data, and sets the upper limit of the frequency of the encoded data to the reproduction bandwidth of the code beyond the frequency of the upper limit value when decoding Only for the encoded blocks adjacent to the discontinuous recording unit in the encoded data, the encoded data included in the encoded block is replaced with a predetermined value to limit the band of the reproduced sound. and to Luo Deiodeta reproduction method. The audio data reproducing method according to claim 4 , wherein the predetermined value for replacing the encoded data is a value "0" . The audio data reproducing method according to claim 4 , wherein the one recording unit of the plurality of encoding blocks is selected according to a predetermined recording unit of video data . In an audio data reproducing apparatus for reproducing digital data in which a digital audio signal is frequency-divided and encoded, and a plurality of encoded blocks each including encoded data are sequentially recorded as one recording unit,
When reproducing in an order different from the order of the recording unit at the time of recording the audio data, the recording in a band that sets the upper limit of the frequency of the coded data to the reproduction exceeds the frequency of the upper limit value when decoding Audio data reproduction characterized by comprising band limiting means for band limiting the reproduced sound by replacing the encoded data contained in the encoded block with a predetermined value only for the adjacent encoded block in the unit apparatus. 8. The audio data reproducing apparatus according to claim 7, wherein the predetermined value for replacing the encoded data by the band limiting unit is a value “0”. It said plurality of said one recording unit of the encoding block is audio data reproducing apparatus according to claim 7, characterized in that it is selected according to a predetermined recording unit of the video data. In an audio data reproducing apparatus for reproducing digital data in which a digital audio signal is frequency-divided and encoded, and a plurality of encoded blocks each including encoded data are sequentially recorded as one recording unit,
When playing on the Symbol audio data in an order different from the order of the recording unit at the time of recording, to set the upper limit of the frequency of the coded data to the reproduction, the above band exceeding the frequency of the upper limit value when decoding Band limiting means for limiting the playback sound band by replacing the encoded data included in the encoded block with a predetermined value only for the adjacent encoded block of the discontinuous recording unit in the encoded data
Features and to Luo Deiodeta reproducing apparatus that comprises a. The audio data reproducing apparatus according to claim 10 , wherein the predetermined value that replaces the encoded data by the band limiting unit is a value "0" . The audio data reproducing apparatus according to claim 10 , wherein the one recording unit of the plurality of encoding blocks is selected according to a predetermined recording unit of video data .

Images