最近有个双声道音频实时转译的需求 要求先把μ-law压缩的双声道wav文件分离声道并解压成常规pcm流调用TTS 网上的方法零零散散 这里整理记录一下解压和分离的过程
声道分离
双声道分离首先要先了解wav文件的构成 这篇文章写的很详细 参考链接: wav文件头格式详解
可以看出先要去掉44位有信息才是音频流 如下
BufferedInputStream wholefile = new BufferedInputStream(new FileInputStream(audioFile));
//去退掉头信息
wholefile.skip(44);
之后是声道分离 参考链接:
如何在Java中将Wav文件分割为多个通道?
这里每3200长度双声道流处理成两段1600长度单声道流
int len;
short s;
byte[] ts = new byte[3200];
while ((len = wholefile.read(ts)) > 0) {
byte[] leftPart = new byte[1600];
byte[] rightPart = new byte[1600];
for(int i=0;i<len;i+=2){
leftPart[i/2] = ts[i];
rightPart[i/2] = ts[i+1];
}
//TODO你要对音频流的处理
}
注意:这个每组音频流的长度和位深度有关 本例为16位深度(2byte)
μ-law解压
对于音频编码的处理可以参考这篇文章: PCM A律μ律编码
本例使用查表法 直接在原本代码上进行修改
private static short muLawDecompressTable[] = new short[]{
-32124, -31100, -30076, -29052, -28028, -27004, -25980, -24956,
-23932, -22908, -21884, -20860, -19836, -18812, -17788, -16764,
-15996, -15484, -14972, -14460, -13948, -13436, -12924, -12412,
-11900, -11388, -10876, -10364, -9852, -9340, -8828, -8316,
-7932, -7676, -7420, -7164, -6908, -6652, -6396, -6140,
-5884, -5628, -5372, -5116, -4860, -4604, -4348, -4092,
-3900, -3772, -3644, -3516, -3388, -3260, -3132, -3004,
-2876, -2748, -2620, -2492, -2364, -2236, -2108, -1980,
-1884, -1820, -1756, -1692, -1628, -1564, -1500, -1436,
-1372, -1308, -1244, -1180, -1116, -1052, -988, -924,
-876, -844, -812, -780, -748, -716, -684, -652,
-620, -588, -556, -524, -492, -460, -428, -396,
-372, -356, -340, -324, -308, -292, -276, -260,
-244, -228, -212, -196, -180, -164, -148, -132,
-120, -112, -104, -96, -88, -80, -72, -64,
-56, -48, -40, -32, -24, -16, -8, 0,
32124, 31100, 30076, 29052, 28028, 27004, 25980, 24956,
23932, 22908, 21884, 20860, 19836, 18812, 17788, 16764,
15996, 15484, 14972, 14460, 13948, 13436, 12924, 12412,
11900, 11388, 10876, 10364, 9852, 9340, 8828, 8316,
7932, 7676, 7420, 7164, 6908, 6652, 6396, 6140,
5884, 5628, 5372, 5116, 4860, 4604, 4348, 4092,
3900, 3772, 3644, 3516, 3388, 3260, 3132, 3004,
2876, 2748, 2620, 2492, 2364, 2236, 2108, 1980,
1884, 1820, 1756, 1692, 1628, 1564, 1500, 1436,
1372, 1308, 1244, 1180, 1116, 1052, 988, 924,
876, 844, 812, 780, 748, 716, 684, 652,
620, 588, 556, 524, 492, 460, 428, 396,
372, 356, 340, 324, 308, 292, 276, 260,
244, 228, 212, 196, 180, 164, 148, 132,
120, 112, 104, 96, 88, 80, 72, 64,
56, 48, 40, 32, 24, 16, 8, 0
};
public static void main(String args[]){
BufferedInputStream wholefile = null;
try {
wholefile = new BufferedInputStream(new FileInputStream(audioFile));
//去退掉头信息
wholefile.skip(44);
int len;
short s;
byte[] ts = new byte[3200];
while ((len = wholefile.read(ts)) > 0) {
byte[] leftPart = new byte[3200];
byte[] rightPart = new byte[3200];
for(int i=0;i<len;i+=2){
s = muLawDecompressTable[ts[i] & 0xff];
leftPart[i] = (byte)s;
leftPart[i+1] = (byte)(s>>8);
s = muLawDecompressTable[ts[i+1] & 0xff];
rightPart[i] = (byte)s;
rightPart[i+1] = (byte)(s>>8);
}
//TODO你要对音频流的处理
}
wholefile.close();
} catch (Exception e) {
System.err.println(e.getMessage());
}
}
注意:由于解压后音频流占位增加 由3200长度双声道解压为两个单声道3200长度的音频流