I have noticed a possible bug/anomaly with the various AudioFormatReaders. The docs say that if you try to read the source into a single buffer it will copy a merged version of the stereo source. This doesn’t appear to be the behaviour implemented though.
Shouldn’t line 326 in juce_WavAudioFormat.cpp read something like:
else if (right == 0)
{
for (int i = numThisTime; --i >= 0;)
{
int mixedSample = 0;
mixedSample += (int) swapIfBigEndian ((unsigned short) *src++) << 16;
mixedSample += (int) swapIfBigEndian ((unsigned short) *src++) << 16;
*left++ = mixedSample * 0.5;
}
}
And a similar thing in AiffAudioFormat?