I2S Audio -- how does the Audio library identify which channel is being read or written?

I’ve been watching Bill's ESP32 Sound - Working with I2S,

video and have a couple of questions.

In the chapter, at 10:18 - Using an I2S Microphone, the example is all mono, left channel.  How would I do both channels for stereo?  Obviously I would hook up a 2^nd microphone strapped for the right channel, and change the channel_format in the i2s_config structure.  In Serial Monitor, and other places, I’d need to know the data for each channel separately.  So, in the Audio library, in loop(), where the i2s_read() is performed, how can the code determine from which channel the data is coming when in stereo mode?  Is that encoded in buffer data, or is there some corresponding metadata?  I also want programmatically generate sound data to send to output amplifiers, and would like to know how the write function discriminates between channels.

It seems probable that the channels are interleaved in the read buffer, but I can't find where that is specified.  Nor is it clear (assuming the above assumption is correct) which channel is first in the buffer.

Thanks for any thoughts.

- John