MPEG 1.0/2.0/2.5 对声音的压缩分三层：Layer I、Layer II和Layer III。相比于Layer3，Layer2的解码就简单得多了。
1.变量初始化 申明的成员变量在构造方法内初始化，这些常量值是从解码规范的文档里直接COPY过来的。
2.帧的数据结构 以Layer2编码的文件按帧存放，解码时也是逐帧解码。每一帧内依次存放位分配信息、增益因子选择信息、增益因子、主数据。采用标准立体声编码的Layer2一帧内的主数据有12个粒度组（granule），一个粒度组内最多可有32个子带（subband），每个子带两个声道。每个声道内3组样本值连续存放，每一组32个值。
3.解码过程 Layer2编码不采用心理声学模型，解码端不用IMDCT；Layer2编码端没有哈夫曼编码。逆量化没有采用解码Layer3那样的查表法而是直接运算，Layer2的逆量化公式也比Layer3的简单得多。解码一帧Layer2只需6步，这6个步骤在decodeFrame方法内注解出来，每一步的细节一看就懂，不再赘述。解码Layer2的源码如下：
/*
* Layer2.java -- MPEG 1.0/2.0/2.5 Audio Layer II Decoder
* Copyright (C) 2010
*
* This program is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program. If not, see <http://www.gnu.org/licenses/>.
*
* If you would like to negotiate alternate licensing terms, you may do
* so by contacting the author: <http://jmp123.sf.net/>
*/
package decoder;
public final class Layer2 implements ILayer123 {
private static Header header;
private static BitStream bs;
private static Synthesis filter;
private static int nch, aidx, sblimit;
private static byte[][] allocation; //[2][32]
private static byte[][] nbal; //[5][]
private static byte[][] sbquant_offset; //[5][]
private static byte[][] scfsi; //[2][32]
private static byte[][][] scalefactor; //[2][32][3]
private static int[] cq_steps; //[17]
private static float[] cq_C; //[17]
private static float[] cq_D; //[17]
private static byte[] cq_bits; //[17]
private static byte[] bitalloc_offset; //[8]
private static byte[][] offset_table; //[6][15]
private static byte[] group; //[17]
private static int[] samplecode; //[3]
private static float[][][] syin; //[2][3][32]
// Layer1也用到factor[]
public static float[] factor; //[63]
public Layer2(BitStream bitstream,Header h, Synthesis filter, int wch) {
header = h;
bs = bitstream;
this.filter = filter;
nbal = new byte[5][];
// ISO/IEC 11172-3 Table 3-B.2a
nbal[0] = new byte[] { 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 3, 3, 3, 3, 3,
3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2 };
// ISO/IEC 11172-3 Table 3-B.2b
nbal[1] = new byte[] { 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 4, 3, 3, 3, 3, 3,
3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2, 2, 2 };
// ISO/IEC 11172-3 Table 3-B.2c
nbal[2] = new byte[] { 4, 4, 3, 3, 3, 3, 3, 3 };
// ISO/IEC 11172-3 Table 3-B.2d
nbal[3] = new byte[] { 4, 4, 3, 3, 3, 3, 3, 3, 3, 3, 3, 3 };
// ISO/IEC 13818-3 Table B.1
nbal[4] = new byte[] { 4, 4, 4, 4, 3, 3, 3, 3, 3, 3, 3, 2, 2, 2, 2, 2,
2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2, 2 };
// ISO/IEC 11172-3 Table 3-B.1
factor = new float[] {2.00000000000000f, 1.58740105196820f,
1.25992104989487f, 1.00000000000000f, 0.79370052598410f,
0.62996052494744f, 0.50000000000000f, 0.39685026299205f,
0.31498026247372f, 0.25000000000000f, 0.19842513149602f,
0.15749013123686f, 0.12500000000000f, 0.09921256574801f,
0.07874506561843f, 0.06250000000000f, 0.04960628287401f,
0.03937253280921f, 0.03125000000000f, 0.02480314143700f,
0.01968626640461f, 0.01562500000000f, 0.01240157071850f,
0.00984313320230f, 0.00781250000000f, 0.00620078535925f,
0.00492156660115f, 0.00390625000000f, 0.00310039267963f,
0.00246078330058f, 0.00195312500000f, 0.00155019633981f,
0.00123039165029f, 0.00097656250000f, 0.00077509816991f,
0.00061519582514f, 0.00048828125000f, 0.00038754908495f,
0.00030759791257f, 0.00024414062500f, 0.00019377454248f,
0.00015379895629f, 0.00012207031250f, 0.00009688727124f,
0.00007689947814f, 0.00006103515625f, 0.00004844363562f,
0.00003844973907f, 0.00003051757813f, 0.00002422181781f,
0.00001922486954f, 0.00001525878906f, 0.00001211090890f,
0.00000961243477f, 0.00000762939453f, 0.00000605545445f,
0.00000480621738f, 0.00000381469727f, 0.00000302772723f,
0.00000240310869f, 0.00000190734863f, 0.00000151386361f,
0.00000120155435f };
// cq_xxx: Layer II classes of quantization, ISO/IEC 11172-3 Table 3-B.4
cq_steps = new int[] { 3, 5, 7, 9, 15, 31, 63, 127, 255, 511, 1023,
2047, 4095, 8191, 16383, 32767, 65535 };
cq_C = new float[] { 1.3333333f, 1.6f, 1.1428571f, 1.77777778f,
1.0666667f, 1.0322581f, 1.015873f, 1.007874f, 1.0039216f,
1.0019569f, 1.0009775f, 1.0004885f, 1.0002442f, 1.000122f,
1.000061f, 1.0000305f, 1.00001525902f };
cq_D = new float[] { 0.5f, 0.5f, 0.25f, 0.5f, 0.125f, 0.0625f,
0.03125f, 0.015625f, 0.0078125f, 0.00390625f, 0.001953125f,
0.0009765625f, 0.00048828125f, 0.00024414063f, 0.00012207031f,
0.00006103516f, 0.00003051758f };
cq_bits = new byte[] {5,7,3,10,4,5,6,7,8,9,10,11,12,13,14,15,16};
sbquant_offset = new byte[][] {
// ISO/IEC 11172-3 Table 3-B.2a
{7,7,7,6,6,6,6,6,6,6,6,3,3,3,3,3,3,3,3,3, 3,3,3,0,0,0,0},
// ISO/IEC 11172-3 Table 3-B.2b
{7,7,7,6,6,6,6,6,6,6,6,3,3,3,3,3,3,3,3,3, 3,3,3,0,0,0,0,0,0,0},
// ISO/IEC 11172-3 Table 3-B.2c
{5,5,2,2,2,2,2,2},
// ISO/IEC 11172-3 Table 3-B.2d
{5,5,2,2,2,2,2,2,2,2,2,2},
// ISO/IEC 13818-3 Table B.1
{4,4,4,4,2,2,2,2,2,2,2,1,1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,1,1} };
bitalloc_offset = new byte[] { 0, 3, 3, 1, 2, 3, 4, 5 };
offset_table = new byte[][] { { 0, 1, 16 }, { 0, 1, 2, 3, 4, 5, 16 },
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 },
{ 0, 1, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 16 },
{ 0, 2, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16 } };
group = new byte[] { 2, 3, 0, 4, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };
allocation = new byte[2][32];
scfsi = new byte[2][32];
scalefactor = new byte[2][32][3];
samplecode = new int[3];
syin = new float[2][3][32];
//nch,aidx,sblimit...
nch = header.getChannels();
int bitrate = header.getBitrate() / nch;
if(header.getVersion() == Header.MPEG1)
if(bitrate <= 48)
aidx = (bitrate == 32) ? 3 : 2;
else if(bitrate <= 80)
aidx = 0;
else
aidx = (bitrate == 48) ? 0 : 1;
else //mpeg 2.0/2.5
aidx = 4;
byte[] limit = {27, 30, 8, 12, 30};
sblimit = limit[aidx];
}
private void requantization(int index, int gr, int ch, int sb) {
int nb, s, c;
int nlevels = cq_steps[index];
if ((nb = group[index]) != 0) { // degrouping
c = bs.getBits17(cq_bits[index]);
for (s = 0; s < 3; s++) {
samplecode[s] = c % nlevels;
c /= nlevels;
}
nlevels = (1 << nb) - 1; //用于计算fractional
}
else {
nb = cq_bits[index];
for (s = 0; s < 3; s++)
samplecode[s] = bs.getBits17(nb);
}
for (s = 0; s < 3; s++) {
float fractional = 2.0f * samplecode[s] / (nlevels + 1) - 1.0f;
// s'' = C * (s''' + D)
syin[ch][s][sb] = cq_C[index] * (fractional + cq_D[index]);
// s' = factor * s''
syin[ch][s][sb] *= factor[scalefactor[ch][sb][gr >> 2]];
}
}
private void stereo(int index, int gr, int sb) {
int nb, s, c;
int nlevels = cq_steps[index];
if ((nb = group[index]) != 0) {
c = bs.getBits17(cq_bits[index]);
for (s = 0; s < 3; s++) {
samplecode[s] = c % nlevels;
c /= nlevels;
}
nlevels = (1 << nb) - 1;
}
else {
nb = cq_bits[index];
for (s = 0; s < 3; s++)
samplecode[s] = bs.getBits17(nb);
}
for (s = 0; s < 3; s++) {
float fractional = 2.0f * samplecode[s] / (nlevels + 1) - 1.0f;
// s'' = C * (s''' + D)
syin[0][s][sb] = syin[1][s][sb] = cq_C[index] * (fractional + cq_D[index]);
// s' = factor * s''
syin[0][s][sb] *= factor[scalefactor[0][sb][gr >> 2]];
syin[1][s][sb] *= factor[scalefactor[1][sb][gr >> 2]];
}
}
public void decodeFrame() throws Exception {
int maindata_begin ,bound, sb, ch;
int slots = header.getMainDataSlots();
bs.append(slots);
maindata_begin = bs.getBytePos();
bound = (header.getMode() == 1) ? ((header.getModeExtension() + 1) * 4) : 32;
if(bound > sblimit)
bound = sblimit;
/*
* 1. Bit allocation decoding
*/
for (sb = 0; sb < bound; sb++)
for (ch = 0; ch < nch; ch++)
allocation[ch][sb] = (byte)bs.getBits9(nbal[aidx][sb]); // 2..4 bits
for (sb = bound; sb < sblimit; sb++)
allocation[1][sb] = allocation[0][sb] = (byte)bs.getBits9(nbal[aidx][sb]);
/*
* 2. Scalefactor selection information decoding
*/
for (sb = 0; sb < sblimit; sb++)
for (ch = 0; ch < nch; ch++)
if (allocation[ch][sb] != 0)
scfsi[ch][sb] = (byte)bs.getBits9(2);
else
scfsi[ch][sb] = 0;
/*
* 3. Scalefactor decoding
*/
for (sb = 0; sb < sblimit; ++sb)
for (ch = 0; ch < nch; ++ch)
if (allocation[ch][sb] != 0) {
scalefactor[ch][sb][0] = (byte)bs.getBits9(6);
switch (scfsi[ch][sb]) {
case 2:
scalefactor[ch][sb][2] = scalefactor[ch][sb][1] = scalefactor[ch][sb][0];
break;
case 0:
scalefactor[ch][sb][1] = (byte)bs.getBits9(6);
case 1:
case 3:
scalefactor[ch][sb][2] = (byte)bs.getBits9(6);
}
if ((scfsi[ch][sb] & 1) == 1)
scalefactor[ch][sb][1] = scalefactor[ch][sb][scfsi[ch][sb] - 1];
}
int gr, index, s;
for (gr = 0; gr < 12; gr++) {
/*
* 4. Requantization of subband samples
*/
for (sb = 0; sb < bound; sb++)
for (ch = 0; ch < nch; ch++)
if ((index = allocation[ch][sb]) != 0) {
index = offset_table[bitalloc_offset[sbquant_offset[aidx][sb]]][index - 1];
requantization(index, gr, ch, sb);
} else
syin[ch][0][sb] = syin[ch][1][sb] = syin[ch][2][sb] = 0;
//mode=1(Joint Stereo)
for (sb = bound; sb < sblimit; sb++)
if ((index = allocation[0][sb]) != 0) {
index = offset_table[bitalloc_offset[sbquant_offset[aidx][sb]]][index - 1];
stereo(index, gr, sb);
} else
for (ch = 0; ch < nch; ch++)
syin[ch][0][sb] = syin[ch][1][sb] = syin[ch][2][sb] = 0;
for (ch = 0; ch < nch; ch++)
for (s = 0; s < 3; s++)
for (sb = sblimit; sb < 32; sb++)
syin[ch][s][sb] = 0;
/*
* 5. Synthesis subband filter
*/
for (ch = 0; ch < nch; ch++)
for (s = 0; s < 3; s++)
filter.synthesisSubBand(syin[ch][s], ch);
} // for gr=0 to 12
/*
* 6. Ancillary bits
*/
int discard = slots + maindata_begin - bs.getBytePos();
bs.skipBytes(discard);
}
}
下一篇：（十七）用JAVA编写MP3解码器——解码Layer1
相关推荐
纯java实现的MP3解码器，装逼使用，面试使用，很好的实现。
java的mp3解码器(非JMF、控制台播放音乐) 用于在你自己的程序中引入,播放MP3文件 运行前先编译---build.bat/build.sh
基于通用可编程GPU的视频编解码器——架构、算法与实现
纯java写的解码器 不含任何jar包
jmp123是一个解码速度很快的纯JAVA的开源的MP3解码器。在2核2.9G CPU上跑，解码速度达到每秒15000帧以上；在4核 CPU上会跑得更快。
C++实现的Mp3解码器
MP3是 ISO/MPEG标准的一部分，ISO/MPEG标准描述了使用高性能感知编码方案的音频压缩，此标准一直在不断更新以满足“质高量小”的追求，现已形成 MPEG Layer 1、Layer 2、Layer 3三个音频编码解码方案。MPEG Layer 3...
好不容易找到的，虽然使用matlab实现，但对理解MP3格式很有帮助！
1、好用的mp3播放器，可以试试 2、mp3解码器
H.264的三大编解码器之一JM最新版，为VC下C工程源码，可直接使用或二次开发。 文件中包含说明文档
许多刚接触mina的朋友，对于mina的编解码器的编写很迷惑.希望这个文档可以帮助朋友们少走弯路。 资源中是一个比较典型的编解码器写法。生成了可执行文件。并对编解码器的代码有详细注释。
Huffman编解码问题——讲解.docx
JCodec 是一个实现一组流行的视频和音频编解码器的库。目前 JCodec 支持 MP4 视频格式。JCodec 是在 FreeBSD 许可证下分发的自由软件。支持的功能：从视频文件中解码特定的图像帧。性能/质量注意事项 由于 JCodec 是...
专用 asf /avi/ai/wpf 解码器 专用 asf /avi/ai/wpf 解码器 专用 asf /avi/ai/wpf 解码器 专用 asf /avi/ai/wpf 解码器 专用 asf /avi/ai/wpf 解码器 专用 asf /avi/ai/wpf 解码器 专用 asf /avi/ai/wpf 解码器 专用...
mp3解码库,千千静听,kugou等mp3播放器使用的解码库
silk2mp3解码器，用于将微信中收藏的silk格式文件解码
用vc6.0编写的MP3播放器和MP3解码器。 单独将MP3解码器编译成静态链接库lib文件， 然后MP3播放器中再调用lib。 很有参考价值。。。
jmp123 JAVA MP3解码,jmp123 JAVA MP3解码
该程序是一个非常好的学习MP3 解码器的说明文档,并且他还介绍了MP3编码的原理对于我们理解MP3解码有很好的帮助,是一个不可多得的好文档