你分别装入.wav-文件PCM-数据？如果是这样，那么你就可以轻松阅读每个通道的每个样本，它们叠加和渠道的数量除以他们得到一个单声道信号。

Do you load .wav- files respectively PCM- data? If so then you could easily read each sample of each channel, superpose them and divide them by the amount of channels to get a mono signal.

如果您存储在交错的签署短裤形式的立体声信号时，code，计算所得的单声道信号可能是这样的：

If you store your stereo signal in form of interleaved signed shorts, the code to calculate the resulting mono signal might look like this:

    short[] stereoSamples;//get them from somewhere

    //output array, which will contain the mono signal
    short[] monoSamples= new short[stereoSamples.length/2];
    //length of the .wav-file header-> 44 bytes
    final int HEADER_LENGTH=22;

    //additional counter
    int k=0;


    for(int i=0; i< monoSamples.length;i++){
        //skip the header andsuperpose the samples of the left and right channel
        if(k>HEADER_LENGTH){
        monoSamples[i]= (short) ((stereoSamples[i*2]+ stereoSamples[(i*2)+1])/2);
        }
        k++;
    }

我希望，我能帮助你。

I hope, I was able to help you.

***的问候， G_J

Best regards, G_J