How can I mix two PCM audio files

Question

I did test mix two PCM audio file. but don't get true audio file.

I used this example So, my code:

  private void mixSound() throws IOException {

byte[] music1 = null;
music1 = new byte[in1.available()];
music1 = convertStreamToByteArray(in1);
in1.close();


byte[] music2 = null;
music2 = new byte[in2.available()];
music2 = convertStreamToByteArray(in2);
in2.close();

byte[] output = new byte[music1.length];

for (int i = 0; i < output.length; i++) {

samplef1 = music1[i] / 128.0f; 
samplef2 = music2[i] / 128.0f;

float mixed = samplef1 + samplef2;
// reduce the volume a bit:
mixed *= 0.8;
// hard clipping
if (mixed > 1.0f) mixed = 1.0f;

if (mixed < -1.0f) mixed = -1.0f;

byte outputSample = (byte) (mixed * 128.0f);
output[i] = outputSample;

} //for loop

save = openFileOutput(filename, Context.MODE_PRIVATE);
save.write(output);
save.flush();
save.close();
}

public byte[] convertStreamToByteArray(InputStream is) throws IOException {

ByteArrayOutputStream baos = new ByteArrayOutputStream();
byte[] buff = new byte[8000];
int i;
while ((i = is.read(buff, 0, buff.length)) > 0) {
baos.write(buff, 0, i);
}

return baos.toByteArray(); // be sure to close InputStream in calling function

}

2 audio files with bit rate 64000 & sampling rate 16000 GH & sterio

in1 = getResources().openRawResource(R.raw.a_2);
in2 = getResources().openRawResource(R.raw.a_diz_2);

Also try to convert bytes array to short array -> then calculate-> then convert short to byte using converts methods like bytes2Shorts(byte[] buf) and shorts2Bytes(short[] s). But steel have a fail result.

Someone can say me Where is my wrong?

Answer 1

There are a number of issues here and I'll try to address some of them

First, using byte[] suggests that your PCM wave data format is AudioFormat.ENCODING_PCM_8BIT (or it should be this format if it already isn't). This format uses 8-bit (1 byte) unsigned , which means that the sound samples are stored in the [0, 255] range (not in the [-127, +128] or [-128,+127] range).

This means that the negative values are in the [0, 127] range and the positive samples are in the [128,255] range.

When mixing values, it's best to prevent clipping right from the start so I'd use

byte mixed = (music1[i] + music2[i])/2; //this ensures that mixed remains within the `correct range` for your PCM format

You can also divide your samples by 128 (if you want to convert them to floating point values)

float samplef1 = (((float)music1[i]-127)/128 ; //converting samples to [-1, +1] range -- -1 corresponds a sample value of 0 and +1 to 255

float samplef2 = (((float)music2[i]-127)/128;

float mixed = (samplef1+samplef2)/2;

Note that you now have 2 options to play data(samples) generated in this way. Either, convert floats back to bytes or use the AudioFormat.ENCODING_PCM_FLOAT format.

audio files with bit rate 64000 & sampling rate 16000 GH & sterio

This can't be correct. Typical sampling rates are 4000Hz, 8000Hz, 11000Hz, 16000Hz, 22050Hz or 44100Hz . For bit depths, audio usually uses 8 bits, 16 bits or 32 bits .

For instance, CD quality audio uses 44100Hz, 16bit, stereo format.